Academic journal article
By Gariety, Bonnie Sue; Shaffer, Sherrill
Monthly Labor Review , Vol. 124, No. 3
Analysis of the Current Population Survey indicates positive wage differentials overall for women on flextime in 1989 and for both men and women in 1997; significant differentials emerge for selected motivations, industries, and occupations
This article presents an empirical test of wage differentials associated with flextime, by gender, stated motivation for using flextime, industry, and major occupation. The test implicitly compares the relative strengths of two opposing effects: a negative compensating wage differential resulting from workers' preferences for flextime and a positive wage differential associated with higher productivity of workers on flextime attributed to what economists call the "efficiency wage hypothesis." Although previous studies have found evidence that flextime increases both productivity(1) and workers' satisfaction,(2) scant evidence has emerged thus far regarding the net quantitative or qualitative impact of these factors on equilibrium wages.
One exception is an article by Nancy Johnson and Keith Provan,(3) who applied a similar test to a much smaller data set and found flextime to be positively associated with wages for professional women, negatively associated with wages for nonprofessional women, and not significantly associated with wages for men. Johnson and Provan's sample totaled 258, obtained by survey from within a single State. The study reported in the current article, by contrast, uses nationwide samples of more than 5,000 workers, obtained from the U.S. Current Population Survey (CPS) supplement, "Multiple Job Holding, Flexitime, and Volunteer Work," for 1989 and 1997. In addition to estimating aggregate wage effects by gender in each year, the article estimates the flextime wage differential associated with specific reasons each worker reportedly preferred flextime in 1989. (Reasons for choosing flextime were not reported in 1997, preventing a comparison with that year.) Also estimated is the flextime wage differential associated with specific industries and specific major occupations for 1997. (Again, in 1989, the number of workers on flextime in particular occupations and industries was too small to draw a meaningful comparison with the later year.)
Results of the study indicate that flextime is associated with significantly higher wages overall. The size of the flextime wage differential for women is stable across the years 1989 and 1997 and is similar to the 1997 estimate for men. However, the 1989 flextime wage differential for men is much smaller than in 1997 and is not significantly different from zero. This finding suggests that the pattern of compensation has evolved in a similar direction for both male and female workers, but it evolved later for men.
The more detailed regressions for 1989 find that the only stated reason for desiring flextime associated with a significant wage differential among women is transportation. Among men, flextime taken for personal reasons is associated with a positive wage differential at the 0.01 level. Only a small number of industries exhibit significant flextime wage differentials for either men or women in 1997, and all of those differentials are positive. Two major industries (automotive and repair services; and social services, other professional services, and forestry and fisheries--grouped collectively as "professional" industries (see p. 5)--exhibit significant wage differentials for both men and women. Significantly positive flextime wage differentials emerge for men in all major occupations except operators, movers, and handlers, while women exhibit significantly positive flextime wage differentials only for sales and administrative occupations.
The article continues by presenting a brief overview of the history of flextime, describing the empirical and conceptual framework of the analysis, and characterizing the sample data. The article concludes with a discussion of the results and some suggestions for future research. …