At What Level of Labor-Market Intermittency Are Women Penalized?

Abstract

A common explanation offered for the observed wage differential between men and women is that women are less attached to the labor market; they exhibit a greater degree of labor-market intermittency than do men. There are several theories that explain this link between intermittency and lower wages, including differences in human-capital attainment, atrophy of skills during absences, and preferences of employers (see e.g., Solomon W. Polachek and W. Stanley Siebert, 1993; Joyce P. Jacobsen and Laurence M. Levin, 1995; James W. Albrecht et al., 2000). The goal of this paper is to explore in greater depth the role past labormarket intermittency plays in the determination of a woman’s current wage and at what level of intermittent activity women can expect to have that activity affect her wage. Previous methods employed to measure the penalty associated with intermittent activity have either classified workers as intermittent if they have at least one spell of absence from the labor market (Jacobsen and Levin, 1995) or have relied on the percentage of time out of the labor force to classify intermittent workers (Elaine J. Sorenson, 1993). However, if employers perceive intermittent behavior as a signal, then both the frequency of intermittent spells and the duration of the spells should be taken into account. We contribute to this literature by creating an intermittency index that captures both of these factors. We also statistically determine at what level of intermittency a woman will incur a penalty for absence from the labor force. This index is used to determine the magnitude of the penalty associated with intermittent participation in the labor force. The analysis is limited to women as intermittent behavior is more prevalent for women and to avoid potential confounding factors associated with gender discrimination.