Unions and Gender Pay Equity in Academe: A Study of U.S. Institutions

This paper uses 1994-95 faculty salary data from over 1,100 four-year U.S. academic institutions, about one-fourth of them with collective bargaining agreements, to ask if faculty unions make a difference to gender pay equity. Average gender salary differences are negative at every rank and at every category of U.S. institution with or without collective bargaining agreement. Unions may improve gender salary differentials somewhat, particularly at the assistant professor level. There is no evidence that this gain will be lost at higher levels, and mixed evidence that further gains occur for women at the full professor level. The most pervasive and robust consequence of unions is to increase the positive impact that higher proportions of women at senior faculty ranks make on relative salaries at the assistant professor level. However, the influence of these higher-ranked women on gender salary inequalities at the associate and full levels is lower in union schools than nonunion schools. By reducing the flexibility of existing salary structures, collective bargaining apparently reduces the influence of senior women faculty on the salaries of current women faculty members while increasing their attention and influence at entry levels.     

Race,Gender and Regional Labor Market Inequalities in Brazil

This study investigates the relationship between unequal regional development and racial and gender wage inequality in Brazil. Using sample data from the 1991 Brazilian census, I estimated monthly wages for a white, brown and black women and men working in the states of S"o Paulo and Bahia. The findings suggest that while women and Afro-Brazilians in Brazil's most developed region of S"o Paulo had the advantages of higher levels of state sponsored work benefits and more equitable occupational and wage distribution, they nevertheless experienced the greatest discrimination. In contrast, the less developed state of Bahia where racial and gender gaps in education, occupation and wages were the most severe, wage discrimination was lowest.     

Revisiting Occupational Crowding in the United States: A Preliminary Study

In her study of occupational segregation in the United States using the 1960 Census, Barbara R. Bergmann found black males with low levels of education more concentrated in low-skill service and laborer occupations than white males and virtually excluded from higher status occupations. Utilizing a crowding index which, similar to Bergmann's, controls for the education level of the worker, this paper presents an analysis of the employment patterns of black males and females in fifty-nine occupations in Wayne County (Detroit, Michigan) and Allegheny County (Pittsburgh, Pennsylvania) in 1990. Within blue-collar and service employment, males are under-represented in the craft occupations and concentrated in low-skill operative, laborer, and service occupations. Females are under-represented in both craft and operative occupations and concentrated in low-skill service occupations. Within white-collar employment, both males and females are largely excluded from high-skill private sector managerial occupations. Black representation in public sector managerial and private sector professional occupations is better in Detroit than Pittsburgh. The decline in manufacturing employment in both counties has left black males with fewer occupational options and black females over-represented in low status clerical and service occupations.     

Wage protection systems, segregation and gender pay inequalities: West Germany, the Netherlands and Great Britain

This paper provides an empirical test of Rubery and Fagan's(1995) hypothesis that gender inequalities are influenced primarilyby the comprehensiveness of the overall wage protection systemin a country and by the extent of gender segregation. Genderdiscrimination in earnings is compared in West Germany, theNetherlands and Great Britain using 1989 ISSP data. Human capitalearnings functions for married males and married females areestimated. Discrimination is measured using standard decompositiontechniques. Earnings discrimination against females in the morecomprehensive systems was 37% in West Germany and 39% in theNetherlands, much less than the 61% found for the less comprehensivelyregulated Great Britain, the latter figure being higher thanprevious estimates using earlier data. Gender segregation isdemonstrated to have contributed to the relative magnitude ofdiscrimination in Great Britain.     

Gender, piece rates and wages: evidence from matched employer-employee data

Why are women disproportionately on piece rates? We investigatethree possible explanations: (1) Women are more likely to beon piece rates, because they have a shorter expected tenurethan men. (2) A greater demand for flexibility between workand home attracts women to work place technologies suitablefor variable pay based on individual performance. (3) Womenprefer piece rates, because they are subject to less wage discriminationwhen objective performance measures are available. The weightof our empirical evidence supports the third hypothesis. Theunexplained gender wage gap is substantially smaller in thepiece-rate regime than in the time-wage regime.     

Gender earnings inequality in the service and manufacturing industries in the U.S.

This paper compares earnings inequality between women and men in a growing sector of the U.S. economy – the service sector, and a shrinking sector – manufacturing. We examine the hypothesis that deindustrialization will reduce inequality, and find that the absolute magnitude of the gender earnings gap is, in fact, smaller in the service sector. Decomposition analysis is used to partition the gender earnings gap into three parts: (1) earnings differences due to differences in mean characteristics – such as education and experience; (2) earnings differences due to preferential treatment of men; and (3) earnings differences due to disadvantageous treatment of women. The latter two constitute estimates of gender discrimination. The results of this study suggest that, ceteris paribus, deirndustrialization will likely reduce the gender gap in hourly earnings. However, this will come at the cost of lower earnings for both males and females, with the drop in earnings being particularly large for males. While deindustrialization is predicted to reduce the absolute magnitude of male-female earnings inequality, evidence suggests that gender discrimination will persist – discrimination explains about 60 percent of the gender wage gap in both the service and the manufacturing sectors.     

Differences in Work Hours and Hours Preferences by Race in the U.S.

Significant differences exist in actual and preferred work hours by race. Specifically, black males work 20 percent fewer annual hours than white males. The differences between black and white women are small. Black workers are significantly more likely than white workers to prefer additional work and fewer are satisfied with their current hours of work. I use the hours-inequality hypothesis of Bell and Freeman (1995,1997) to evaluate the extent to which race differences in work hours and hours preferences are related to race differences in incentives. I demonstrate that whereas white workers work longer hours in response to overall wage variation in their relevant labor market cell, black workers react to the wage variation among black workers but not to the variation overall. The fact that labor market incentives are different for otherwise similar black and white workers is difficult to reconcile with standard competitive theory.