This project seeks to understand how couples optimize their place of employment, occupation, and location and how the optimization changes at different stages of the family life cycle. In particular, we are interested in whether women's earnings are differentially impacted than men's from family moves and from shifts among employers. We are also interested in understanding the impact of these decisions as children go through various developmental stages: from infant, to toddler, to school entry, higher grades, and eventually exited from the household. Unlike most earlier studies that study the impact of family formation, we are particularly focused in understanding what happens to gender career and earnings gaps in the later stages of the family life-cycle, as children grow and family responsibilities become less binding, and how these adaptations are impacted by the type of occupation and firm.
We utilize two sources of data: the NLSY-1979 and restricted access Census data consisting of the Longitudinal Employer Household Dynamics (LEHD) database matched with the 2000 Decennial Census and the American Community Survey 2001-2014. Our analysis time period included 1979-2018 for the NLSY and 1991-2014 for the LEHD. We study men and women, by educational attainment (college graduation versus high-school graduation with no college degree) from birth cohorts 1957 to 1964.
These two revealing longitudinal datasets allow us to gauge detailed information about individuals?their labor force participation, job changes, earnings?and their family formation. The NLSY provides the exact occupation as well as the time intensity of the job in terms of weeks and hours worked. That has enabled us to link job traits from the BLS O*NET database to better group and characterize the kind of employment they are engaged in. On the other hand, the LEHD provides detailed information about the employee?s firm, including the location, sector, and industry of the establishment, the size of the firm as well as a series of metrics that describe the employment composition and level and distribution pay at the establishment.
Our key findings can be summarized as follows. In the NLSY, we estimate the "motherhood penalty", which is the difference between the earnings and hours of women who are currently mothers and those who are not; as well as the "parental gender gap", which is the difference in outcomes between mothers and fathers. Women with young children work far fewer hours per week than do others. But as the children grow up and eventually leave home, women begin to work more hours and transition to higher-earning positions. The motherhood penalty is greatly reduced, and by their fifties college-graduate women with children earn just 7 log points less than those without children, given hours and work experience. On the other hand, fathers manage to maintain their relative gains and do monumentally better than mothers, women without children and men without children. Fathers earn almost 7 log points more per child regardless of the children's age, whereas mothers lose 7 log points per child, holding hours of work and work experience constant. Results for the non-college graduate group are similar but the fatherhood premium is much reduced.
In our study on the firm provision of paid family leave, we find that while there are large differences across industries in the availability of paid leave, three central facts emerge. First, there has been an increase over the past two decades in the provision of paid parental leave. Second, large firms are more likely to offer such leave. Third, firms have been offering more parental leave to their male workers as well. Growth in the provision of paid parental leave has been greatest among large firms in the professional service and technical sectors. Businesses are more likely to provide paid leave the higher the share of workers they expect to remain with them. In fact, the greater the fraction of workers who have skills that are specific to their employer, the greater the likelihood that the employer will provide paid leave. This factor explains why lower wage workers are less likely to be offered paid parental leave. Because they are less attached to their employer?they have fewer skills that are specific to the employer?they are less likely to return to the job and thus lower the value of paid leave to the employer. In addition, their lower wages make them less able and willing to pay for the benefit in the form of even lower wages.
The LEHD based findings on job mobility, dynamics of the within-couple career gaps, and the role of firm characteristics are being prepared for Census disclosure review, and we look forward to sharing them in the form of a research study once they have been disclosed for dissemination.