Career, Family, and the Well-being of College-educated Women Pfd Free

J Marriage Fam. Author manuscript; available in PMC 2015 Feb ane.

Published in final edited class as:

PMCID: PMC4041155

NIHMSID: NIHMS532420

The Motherhood Penalty at Midlife: Long-Term Effects of Children on Women's Careers

Javier García-Manglano

^*Department of Folklore, Nuffield College, University of Oxford, Manor Road Building, Oxford OX1 3UQ, United Kingdom

Suzanne M. Bianchi

^**Department of Sociology, Box 951551, 264 Haines Hall, 375 Portola Plaza, Los Angeles, CA 90095-1551

Abstruse

The authors build on prior research on the motherhood wage penalty to examine whether the career penalties faced past mothers alter over the life course. They broaden the focus across wages to also consider labor force participation and occupational status and employ information from the National Longitudinal Survey of Immature Women to model the changing impact of maternity as women historic period from their 20s to their 50s (n = 4,730). They plant that motherhood is "costly" to women'south careers, but the effects on all 3 labor force outcomes attenuate at older ages. Children reduce women's labor force participation, but this issue is strongest when women are younger, and is eliminated by the 40s and 50s. Mothers also seem able to regain basis in terms of occupational status. The wage penalty for having children varies by parity, persisting across the life course simply for women who have 3 or more children.

Keywords: families and work, stock-still effects, longitudinal, midlife, motherhood, women's employment

A growing torso of research has shown that mothers pay a pregnant wage penalty for having children (Avellar & Smock, 2003; Budig & England, 2001; Budig & Hodges, 2010;Waldfogel, 1995, 1997). The main argument is that having and raising children interferes with the aggregating of homo capital and hence the level of productivity, which and then translates into lower wages. Women who, as a effect of having or planning to accept children, either cutting brusk their education, drib out of the labor forcefulness for an extended period, cut back to part-time employment, choose occupations that are more than family friendly, devote less endeavor on the job, or laissez passer upwardly promotions considering of time or locational constraints, finish upwardly achieving less than childless women who stay on track with total-time employment and take advantage of opportunities for grooming and career advocacy (Aisenbrey, Evertsson, & Grunow, 2009; Anderson, Binder, & Krause, 2003; Baum, 2002; Gangl & Ziefle, 2009; Jacobsen & Levin, 1995). Those who become mothers at younger ages and go on to have more children are more probable to make these kinds of accommodations in their piece of work lives and therefore suffer greater career penalties than practise women who await longer and have fewer children (Blackburn, Bloom, & Neumark, 1993; Chandler, Kamo, & Werbel, 1994; Miller, 2011; Taniguchi, 1999). Some researchers also argue that mothers may face workplace discrimination because some employers believe that mothers are less competent or committed to their jobs than childless women (Budig & England, 2001; Correll, Benard, & Paik, 2007). Considering discrimination is so difficult to measure empirically, evidence of information technology is typically inferred from balance wage differences that remain after decision-making for human uppercase and other relevant characteristics (see Correll et al., 2007, for a notable exception).

Most research on the motherhood penalization has focused on short-term wage penalties among women who are still raising relatively young children, typically when they are in their 20s and 30s. Past focusing on the tiptop child-rearing ages, nonetheless, these studies exercise non consider the longer term effects on women'due south career paths as mothers empty the nest and launch their children from the parental domicile. Do the careers of mothers eventually take hold of upwardly to those of childless women, or do mothers fall further and further behind every bit they age? It seems unrealistic to assume that the burdens of motherhood remain fixed over time, simply it is unclear from prior research whether career penalties ease or accumulate over the life course. This question is important, because persistent penalties, especially in terms of job tenure and earnings, may accumulate over time to leave mothers with less access to pensions or retirement income in later life.

In this study, nosotros use information from the National Longitudinal Study of Young Women (NLS-YW; see http://www.bls.gov/nls/nlsorig.htm) to model the maternity penalization over the grade of women'due south careers every bit they age through their 40s and into their early 50s—a time when nigh all women have finished begetting children and most accept seen their children either leave the home or at to the lowest degree enter adolescence. Moreover, we expand the focus beyond wages to besides consider maternity penalties associated with labor force participation and occupational condition; in this way, we provide a more comprehensive view of how career outcomes of mothers and childless women change over the life course.

Background

Studies accept generally found average wage penalties ranging from 5% to 10% per kid amid women in their 20s and 30s (Anderson et al., 2003; Budig & England, 2001; Waldfogel, 1997). Much, but not all, of the maternity wage penalization has been explained by differences in productivity as measured past homo capital indicators such every bit education and accumulated work experience (Budig & England, 2001; Gangl & Ziefle, 2009). In much of the literature, the maternity penalty has also come to ascertain the (unexplained) lower wages of mothers compared with childless women, fifty-fifty subsequently productivity related factors are controlled. The unexplained differences between mothers and nonmothers could still reflect unmeasured productivity differences among women if, for instance, motherhood diverts one's energy and commitment away from the chore (Evertsson & Breen, 2008), but it could too be due to differential treatment or discrimination, as suggested in the work of Correll et al. (2007).

Studies take shown that the penalty mothers pay also may vary beyond income levels, with those at the bottom of the income distribution paying a larger penalty than those at the top (Budig & Hodges, 2010). Others take institute that the penalties are greatest for highly skilled working mothers (Wilde, Batchelder, & Ellwood, 2010). No prior research has considered whether the penalty changes as women grow older and the demands of childrearing refuse, however.

There are skillful reasons to believe that motherhood penalties may change over time. On the one hand, as women gain more experience balancing work and family, and as their children abound older and more independent, mothers may exist able to refocus on their work lives and as a result eventually narrow the wage gap with childless women (Anderson et al., 2003). On the other hand, mothers may suffer a growing disadvantage over fourth dimension if their lack of early on investment in man capital and aperture in piece of work experience keep them out of higher paying occupations and deny them opportunities for pregnant wage growth and occupational mobility subsequently in life. In this case, ane might expect a widening of the motherhood punishment as women age into midlife (Blackburn et al., 1993; Loughran & Zissimopoulos, 2008).

Several studies using cantankerous-exclusive data to simulate women's lifetime earnings losses associated with motherhood have estimated a gradual narrowing with historic period, but nevertheless a persistent maternity gap by the mid-40s (Davies, Joshi, & Peronaci, 2000; Sigle-Rushton & Waldfogel, 2007). Simulations presented by Sigle-Rushton and Waldfogel (2007) suggest that, in the Usa, the annual motherhood earnings gap narrows from about $7,500 amid women at age 27 to about $2,500 past age 45, with a complete closing of the gap between mothers with i or two children. These findings are quite suggestive but, as simulations for hypothetical individuals (i.e., women with average education and either goose egg, 1, or two births), based on cross-sectional data with few controls for important factors such every bit work experience, they are not representative of the lived experiences of actual cohorts of women. In the present analysis, we utilise detailed longitudinal data for a large, representative cohort to compare women'southward employment outcomes over much of their working lives. These information allow united states to assess the extent to which the careers of mothers (of all parities) and childless women either converge or diverge over the life course.

Most studies of the motherhood penalisation have used wages as the only measure of economic achievement—see Aisenbrey et al.'s (2009) study of occupational penalties for a notable exception. Although wages are certainly an important economic indicator, they are merely one attribute of economical success, reflecting the extent to which local labor markets advantage workers. When because the long-term costs of motherhood to women'south careers, it is also of import to consider the continuity of women's employment over time equally well as their occupational status. We recognize that employment differences by motherhood status are not "penalties" in the same manner as wage or occupational differences, yet we include them in our investigation because they are a cadre component of women's careers, and they have often been overlooked in previous studies of the motherhood penalty. Moreover, by focusing only on the experiences of working women, past studies have ignored the selectivity into employment and do non consider how motherhood may influence employment decisions, especially at older ages. Whereas past research has focused on breaks in employment effectually the time of childbirth (Desai & Waite, 1991; Joshi & Hinde, 1993; Joshi, Macran, & Dex, 1996; Klerman & Leibowitz, 1999), less is known about longer term patterns: How does mothers' labor force attachment modify as their children grow up and leave home? Practise mothers remain less fastened to the labor force than nonmothers at older ages, or do they eventually take hold of up to, or fifty-fifty surpass, nonmothers? Few studies have used longitudinal data to consider women'southward employment patterns throughout the adult life course every bit their children grow upwardly and exit the home.

From a life form perspective, labor force participation and continuity is of involvement in its own right. Many mothers go on to take some time out of the labor strength when their children are immature, and studies have shown that there tends to exist considerable labor force churning for mothers surrounding the first and second births (Joshi & Hinde, 1993; Joshi et al., 1996; Klerman & Liebowitz, 1999). Interruptions that are short and early on in a woman's career may non be all that costly over the long term considering women are non out of the labor force long plenty for skills to erode. In fact, some inquiry has found that one fashion women combat the motherhood penalisation is to drop out for a brusque period after the birth of a child and then go to work for a new employer. Estes and Drinking glass (1996) showed that wages were higher for mothers who changed jobs than for those who did not. Staff and Mortimer (2012) found that how a female parent used her time when she was not in the labor force was also of import; years spent gaining additional teaching and training did not increase the penalty, whereas years when a mother was neither enrolled in school nor working for pay increased the wage gap betwixt mothers and childless women. Finally, mothers may sometimes be able to minimize the effects of time out of the labor force by choosing occupations with lower levels of skill depreciation and fewer penalties for time out, thus reducing the effect on wages (Mincer & Ofek, 1982; Mincer & Polachek, 1974; Okamoto & England, 1999). Some occupational skills can go on to be honed in volunteer activities during fourth dimension out of the labor strength, increasing the continuity of labor strength careers.

On one manus, the more than children a adult female has, the more hard it may be to exit the labor force for only a short period, complete additional schooling, or spend big amounts of time on volunteer activities. On the other hand, more children create greater force per unit area for labor force reentry and a 2nd income later in life as those children age into boyhood, graduate from high school, and attend college. An examination of differences in labor force trajectories beyond the life form for mothers with dissimilar numbers of children is thus likely to be informative.

Labor force continuity may potentially be even more than important for occupational location than wages. Occupational condition scores show how well an occupation is valued in society, at least in terms of prestige; they likewise reflect the preparation required (education) every bit well as the remuneration level (earnings) for people in that occupation in general, thereby providing a broader view of the incumbent'due south relative success in the work world. One's occupational achievement may capture something more enduring than wages, and loftier levels of occupational attainment may be more than or less easily recaptured than wages later years of absence from the workforce or reduced hours in the workforce. How interesting one'southward work is, whether one has supervisory authority over others, and how much autonomy and command 1 has over ane's work may accrue over a career, making it difficult for mothers who take time out of the labor strength or cutting back on work hours to rear children to exist besides positioned occupationally later in life as those who practice not brand these types of adjustments. Comparative piece of work has shown that in countries, such as Sweden, where parental leaves are generous, occupational gender segregation was really higher than in countries, such every bit the United states of america, where paid maternal leaves are relatively brusk and less widely bachelor than in Sweden (Mandel & Semyonov, 2005, 2006). Other research has plant a significant corporeality of occupational downgrading immediately after childbirth (Aisenbrey et al., 2009; Joshi & Hinde, 1993), but these studies were unable to decide whether, in the long run, women recovered from these curt-term labor–supply accommodations.

Labor force trajectories, occupational attainment, and wages are influenced past a host of characteristics that vary beyond individuals and that are typically controlled in studies of the maternity wage penalty. These include cardinal aspects of human being capital accumulation, such as early investments in educational attainment, years of work feel, whether work is function time or full time, and utilization of on-the-job training, all of which can touch further labor force attachment, occupational advancement, and wages. Some have argued that women develop strong preferences for work versus family unit early on in life and that this affects later life choices and outcomes (Hakim, 2002; Shaw & Shapiro, 1987). Women who are more committed to having a career may devote greater effort on the task and brand choices that open opportunities for advancement compared with women who are more dwelling house oriented. Although prior studies of the motherhood wage penalty accept not incorporated straight measures of work–family preferences, contempo work past García-Manglano (2012a) showed that preferences are an of import mechanism linking human being capital letter to labor force outcomes.

Finally, women are differentially positioned on other social statuses that tin can bear on employment outcomes. Those who are married and have admission to a husband's sizable income may face up different incentives for labor force participation and career advocacy than do women who are single or who accept depression-earning husbands. White women may have different opportunity structures than non-White women that can touch decisions about education and employment that in turn influence life form trajectories of occupational advocacy and earnings attainment. By research on the motherhood penalty has typically included well-nigh of these factors in estimating the effects of children on wages.

In summary, this article builds on prior enquiry on the maternity wage penalisation by considering the long-term clan between motherhood and careers beyond the adult life course, to determine whether the maternity penalisation grows larger or smaller as women historic period from young adulthood to their early 50s. Different previous studies, which have focused well-nigh exclusively on wage penalties, we augment the focus to consider 3 interrelated career outcomes: (a) labor forcefulness participation, (b) wages, and (c) occupational status measured throughout the 35 years of the NLS-YW panel. Our key focus throughout the analysis is on changes over the life course in the relationship between maternity and women's career outcomes; nosotros pay item attention to how much of the penalty is explained by human capital accumulated across respondents' lives and how much remains unexplained. Our goal is to depict how labor forcefulness outcomes play out over a mother'southward life time as women make choices well-nigh their family unit lives and thus self-select (either intentionally or inadvertently) into different career trajectories.

Method

Information and Measures

The data for the assay come up from the NLS-YW cohort. The original NLS-YW cohort is based on a national sample of v,159 women who were ages xiv through 24 in 1968. These women were born between 1944 and 1954 and were the leading half of the Baby Smash generation. They are an peculiarly interesting cohort because they came of historic period right at the fourth dimension that women'due south piece of work and family roles were being redefined by the civil rights and women's movements of the 1960s and early 1970s. The NLS-YW cohort was reinterviewed either every twelvemonth, or every other year, until the last interview in 2003, when they were ages 49 through 59. The NLS-YW is well suited for the nowadays analysis because it follows the cohort long by the intensive child-rearing phase of life and therefore permits the assessment of longer term outcomes than was possible in previous studies. It besides includes detailed employment and family information collected repeatedly throughout the developed lives of the respondents. Although women in afterward cohorts may differ from the NLS-YW in terms of labor force attachment and fertility beliefs, some research suggests that the motherhood penalties by the early on 40s had not changed essentially between the NLS-YW and the National Longitudinal Survey of Youth 1979 (NLSY79) cohorts (Avellar & Smock, 2003; García-Manglano, 2012b). Of grade, it is as well soon to predict whether more contempo cohorts will continue to follow the aforementioned patterns into their 50s as the NLS-YW cohort, but the current analysis provides a useful baseline against which to compare the experiences of future cohorts.

Our ultimate goal is to model cumulative alter throughout the adult life form, and thus we define our initial sample quite broadly as women who participated in at to the lowest degree 2 interviews between the initial wave in 1968 and the final wave in 2003. Almost respondents contributed far more than two waves of information: Xc-one percent had at least v interviews, and 74% had at least 10 interviews. Considering only one-half the sample at the last interview was older than age 54 (i.due east., the accomplice was ages 49–59 at the final interview), we set up the upper limit of our period of observation every bit age 54, thereby assuasive u.s.a. to observe all women as they anile into their early on 50s. By focusing on women's experiences prior to age 55, we endeavour to limit the potential bias due to early retirement that occurs increasingly by the late 50s. Sensitivity analyses (available on request) bear witness that this historic period brake does not modify our results in a substantive way.

Following prior studies of the motherhood wage penalty (e.yard., Budig & England, 2001), nosotros utilise fixed-furnishings methods to estimate the impact of maternity and parity (measured equally the cumulative number of children reported at each interview, i.eastward., nil, one, ii, three or more) on labor strength participation, wages, and occupational status. Fixed-furnishings estimates show the boilerplate impact of motherhood on employment outcomes, across the woman's career, controlling for the effects of stock-still, unobservable (and observable) characteristics that may be influencing both fertility and employment outcomes and thereby producing spurious motherhood effects. Although prior studies have found meaning boilerplate wage penalties for women in their 20s and 30s, few take studied the effects at older ages, and none has considered whether the penalties might modify every bit women (and their children) grow older.

Using the NLS-YW data, we are able to examine the motherhood penalisation as women age from their 20s to their 50s and, past testing for Age × Parity interactions, we assess the extent to which these penalties grow larger or smaller at older ages. We translate the age interactions not as a reflection of the crumbling or maturation process for women merely rather as a proxy for the declining burden of childrearing that accompanies the aging of a mother's children. Nosotros tried several culling specifications that attempted to directly tap the changing bear upon over time of both the timing and numbers of births also as the ages of the youngest child, merely we came to the same conclusions that we draw based on parity.

The stock-still-effects model requires that our data be organized into person-year records, one for each year in which the adult female was interviewed (with a minimum of 2 observations per woman), and then pooled across years and beyond women. Our overall analytic sample consists of 4,730 women who contributed two or more person years to one or more of our models. Sample sizes vary, depending on the outcome under consideration. Nosotros first by modeling labor force participation across all person-year records for all women, and and so nosotros model wages and occupational condition for working women, based simply on the person-yr records in which women were employed. For the labor force participation analysis, 4,006 women contributed 60,376 person-years of observation. Because the fixed-effects model requires that women experience at least 1 change in the outcome variable beyond the years of ascertainment, nosotros dropped 724 women from our analytic sample considering they were either employed every twelvemonth or not employed in whatsoever twelvemonth and therefore were never observed to change employment status. For the wage models, iv,351 women contributed 35,272 employed person-years of observation; for the occupational prestige analysis, iv,476 respondents contributed 39,569 employed person-years of observation.

Our dependent variables reflect three employment outcomes measured each year: (a) labor force participation, (b) hourly wages, and (c) occupational status. A woman was considered to exist currently employed if she answered "working" to the question "What where you doing most of concluding week—working, going to schoolhouse, or something else?" Hourly wages are obtained from women's direct reports of their hourly charge per unit of pay, consistent with previous research on the motherhood wage penalty (cf. Avellar & Smock, 2003; Budig & England, 2001; Budig & Hodges, 2010). All wages are expressed in 1990 dollars, and we use the natural log of wages in the regressions. We measured occupational status using the Hauser–Warren Socioeconomic Index, which incorporates 1990 demography occupational codes and occupational prestige ratings equally reported in the 1989 General Social Survey (Hauser & Warren, 1997). The calibration is a blended measure created by regressing occupational prestige ratings onto occupational earnings and education and then using the results to generate socioeconomic scores for all of the 1990 detailed occupation categories. Values range from 0 to eighty.five.

Key independent variables include the number of children e'er born (or adopted), every bit reported at each interview. This time-varying measure out was coded categorically, thereby allowing the comparison of childless women with those who have ane, two, or 3 or more children. Information technology should exist noted that as women brand the transition to motherhood at older ages, the limerick of the comparison group of childless women changes with age, with many of the all-time educated women switching from being childless when they get mothers at older ages. This is a problem inherent to near longitudinal fixed-effects studies of the motherhood punishment that crave time-varying measures of fertility. In preliminary analyses, without stock-still effects, nosotros experimented with alternative specifications, including using a fixed mensurate of completed (final) fertility instead of the time-varying mensurate of cumulative fertility, and so that the (always) childless comparison group would be stock-still over time. Perhaps because most mothers in the NLS-YW had their first birth by their early 30s, the selection of fixed versus time-varying measures of fertility made trivial difference to the results at older ages; in other words, the childless category in this accomplice did not change much after the early 30s.

Measures of human capital include educational attainment, reflecting the highest level attained by a given year. Women were categorized as having completed less than high school, a high schoolhouse caste, some higher, or at least a 4-twelvemonth college caste. Nosotros include measures of years of full-time work experience and function-fourth dimension work experience and the square of each of these terms. Because part-time jobs are often lower paying and of lower condition than full-time jobs, nosotros also include (in the wage and occupational status models) an indicator for whether the respondent was currently employed part time at the time of the interview. Part-time status was divers as working less than 35 hours per week in the week prior to the interview. A final measure of homo capital reflects on-the-job training received over time and was calculated equally the cumulative number of weeks of chore training completed every bit of each interview. A squared term for this indicator is as well included in the multivariate models, consistent with prior enquiry.

In add-on to measures of fertility and man capital, we also include a series of time-varying demographic characteristics, but because the fixed-effects model requires that covariates change over time, we do not include whatsoever fixed characteristics, such equally race or parents' socioeconomic status. Our key life course indicator is electric current age (at each interview), coded continuously and so grouped into decade categories reflecting the 20s, 30s, 40s, and early on 50s. We tried unlike age specifications, including age and age squared, and found similar results, but nosotros chose to present the chiselled historic period specification because it is easier to interpret than the quadratic specification, especially when interacted with other variables. In improver to age, we control for marital status at each interview, coded every bit married versus not currently married (including never- and previously married), and husband's income (measured in thousands of 1990 dollars and coded equally 0 for unmarried women). We include hubby's income to alphabetize the fiscal needs of the family unit in the absence of the female parent'due south earnings. Finally, we include a adult female'southward work–family orientation, using her responses to the question "What would you like to be doing when y'all are 35 years old?", which was asked at each interview until the woman reached age 34. Possible responses include "working (at a different or the aforementioned chore)," "married, keeping house, raising [a] family," or "other/don't know." Specifically, we define work expectations cumulatively equally falling on a continuum from 0 to ten, representing the cumulative percentage of interviews (divided by 10) in which a woman indicated that she expected to be doing something other than working at age 35. This measure of home-oriented work–family expectations is time varying up to age 34 (the last historic period when the women were asked the question) and and then remains fixed, whereby those older than age 34 are assigned the final value for this variable.

Analytic Strategy

Nosotros first nowadays summary characteristics for the NLS-YW accomplice at ages 25, 35, 45, and 52. Throughout the multivariate analysis, we use the total longitudinal data collected up through age 54 from women who were interviewed at least twice betwixt 1968 and 2003, including those who dropped out prior to the end of the written report. Nosotros estimate fixed-furnishings models for a binary issue for the labor force analysis and for continuous outcomes in the case of wages and occupational prestige. All models are run in 3 steps. First, parity and historic period, measured in historic period decades of the life course (i.e., 20s, 30s, 40s, and early 50s), are used to predict the effect of interest, cyberspace of demographic controls and early life expectations about paid work at age 35. The 2nd model adds human majuscule variables, thus allowing us to determine how much of the motherhood penalty can be explained by teaching and work feel. A third set of models includes interactions between parity and the maternal age dummies in lodge to test whether the impact of motherhood varies by age. All results are weighted to adjust for sample attrition, and standard errors accept been corrected for the correlation between observations from the same individual.

Results

Table 1 presents means and distributions on the career outcomes and human capital variables for the NLS-YW cohort at around ages 25, 35, 45, and 52, separately by parity, thereby assuasive a comparison of childless women with women who had one, two, or iii or more children. (Distributions on all variables used in the analysis of each of the three outcomes are included in the Appendix) Childless women at age 25 were better educated, more likely to be employed, earned college wages, and worked in occupations of higher prestige than women who became mothers early, in particular those who had more than i kid before age 25. Whereas 76% of childless women at age 25 were in the labor force, merely 28% of mothers with three children were working for pay. At age 25, those who already had three (or more) children had completed simply 10.6 years of schooling, compared with an average of 13.viii years for those who were childless at age 25. The gap between childless women and mothers in labor force participation, occupational attainment and, to a lesser extent, wages, appeared to attenuate at older ages. By age 52, mothers at all parities were more likely than childless women to be employed, and they had narrowed the gap in wages and occupational status compared with younger ages. It is unclear from these data whether the narrowing gap reflects improvements over fourth dimension in the position of mothers, or the changing composition of childless women occurring as late child-bearers joined the group of mothers, bringing with them their relatively college levels of homo capital, particularly educational attainment. Nevertheless, Tabular array 1 shows that the boilerplate level of education of childless women did non pass up at older ages and in fact continued to increase slightly with historic period.

Table 1

Characteristics of Women at Selected Ages, by Electric current Parity: National Longitudinal Survey of Young Women, 1968–2003

Variable–	All Women all ages	Age 25				Age 35
		Childless women	1 child	two children	three+ children	Childless women	1 child	ii children	3+ children
N still in study	4,730	1,592	ane,056	842	455	578	651	1,185	1,091
% in each parity grouping (of North still in study)	100	40.iv	26.8	21.3	11.5	16.5	18.6	33.eight	31.one
Career outcomes
Labor strength participation (% employed)	61.2	76.0	44.five	33.1	27.9	79.two	69.6	61.two	l.five
Hourly wages (if employed)	ix.half-dozen	9.two	vii.9	seven.2	half-dozen.2	11.4	10.1	viii.vi	7.9
Occupational prestige (HWSEI alphabetize, if employed)	35.9	39.four	32.5	28.8	25.4	forty.ix	36.1	35.6	31.4
Human majuscule accumulation
Completed years of schooling	13.iii	thirteen.8	12.5	eleven.7	ten.6	14.4	thirteen.5	xiii.2	12.iv
Years of full-time work feel (>35 hours/week)	9.4	iii.three	3.2	2.four	2.3	ix.0	8.2	6.5	v.iii
Years of part-time work feel (<35 hours/week)	2.9	i.4	1.0	1.1	0.ix	2.7	ii.five	2.7	2.4
		Age 45[KAS1]				Age 52
		Childless women	i kid	two children	three+ children	Childless women	ane child	ii children	3+ children
N still in written report		446	543	1,080	1,107	469	556	i,099	one,114
% in each parity group (of North notwithstanding in written report)		fourteen.0	17.1	34.0	34.9	14.5	17.2	33.9	34.four
Career outcomes
Labor force participation (% employed)		74.7	75.0	72.vii	67.6	62.iv	64.0	66.9	63.1
Hourly wages (if employed)		xiii.2	xi.iii	10.8	8.eight	xiii.1	12.3	11.2	10.i
Occupational prestige (HWSEI index, if employed)		42.7	38.ii	38.half-dozen	34.1	41.7	38.four	39.iv	35.8
Homo capital accumulation
Completed years of schooling		fourteen.5	13.viii	13.6	12.8	14.7	thirteen.8	13.seven	12.9
Years of full-time work experience (>35 hours/calendar week)		15.ii	14.5	11.seven	nine.9	18.vi	18.2	xv.2	13.2
Years of function-time work feel (<35 hours/week)		4.one	4.0	four.7	4.four	4.ix	4.viii	5.nine	5.6

Appendix

Person-Year Ways and Proportions for Variables Used in the Assay of Each Market place Outcome, National Longitudinal Survey of Young Women, 1968–2003

Variable	LFP	Wages	HWSEI
Person-years	60,376	35,272	39,569
Career outcomes
Labor strength participation (% employed)	61.2	100.0	100.0
Hourly wages (if employed)	9.5	9.6	ix.6
Occupational prestige (HWSEI alphabetize, if employed)	36.i	35.7	35.9
Current fertility
Children ever born (electric current)	1.seven	1.4	1.5
Human majuscule
Completed years of schooling	xiii.i	thirteen.two	13.iii
Years of full-time work feel (>35 hours/week)	7.6	9.2	ix.iv
Years of part-time piece of work experience (<35 hours/week)	2.viii	ii.6	2.nine
Currently working role time	24.4	15.viii	17.1
Years of job training	0.5	0.vi	0.6
Sociodemographic controls
Marital status (% married)	65.2	58.3	59.three
Husband'southward income (in thousands of 1990 dollars)	25.seven	24.eight	24.seven
Work expectationsa (0:work − 10:dwelling)	4.2	3.9	3.9

Multivariate Analysis

Estimates of labor strength participation from stock-still-effects models are shown in Table ii. The stock-still-furnishings models are nested, with the first model including the number of children and the woman'south age (measured in decades), the second column adding man uppercase variables, and the third model adding Age × Parity interactions. All models include controls for a woman's marital status, her hubby's income, and whether her expectations were to be at abode versus work at age 35.

Table ii

Fixed-Effects Logistic Coefficients Predicting Women's Participation in Paid Work: National Longitudinal Survey of Young Women, 1968–2003

Variable	Gross effect	+ Human upper-case letter	+ Decade interaction
N	iv,006	four,006	4,006
Person-year observations	60,376	60,376	60,376
Children ever built-in (CEB; ref.: childless)
1 child	−1.521***	−1.641***	−one.837***
2 children	−one.821***	−i.869***	−2.314***
Three or more than children	−1.895***	−1.843***	−two.419***
Age decade (ref.: 20s)a
30s	−0.152**	−0.127**	−0.799***
40s	−0.433***	−0.278**	−1.941***
50s	−0.986***	−0.482***	−2.152***
CEB–age decade interaction (ref.: 20s × childless)
30s × 1 child			0.813***
30s × 2 children			0.922***
30s × iii or more children			0.762***
40s × 1 kid			1.559***
40s × 2 children			ane.940***
40s × 3 or more than children			one.937***
50s × 1 child			i.387***
50s × 2 children			1.829***
50s × iii or more children			ane.901***
Highest educational caste (ref.: <high school)
High schoolhouse graduate		0.014	−0.071
Some higher		0.002	−0.128
College grad and beyond		one.285***	1.286***
Piece of work status, cumulative experience, and grooming
Cumulative years of full-time work experience		0.246***	0.244***
Years of full-time work experience squared		−0.007***	−0.007***
Cumulative years of part-fourth dimension piece of work feel		0.252***	0.239***
Years of part-time work experience squared		−0.006***	−0.005***
Years in chore training		0.066	0.103*
Years in job training squared		−0.016*	−0.019**
Sociodemographic controls
Marital condition at interview (married = 1, other = 0)	−0.330***	−0.422***	−0.350***
Married man's income (in thousands of 1990 dollars)	−0.007***	−0.009***	−0.009***
Home-oriented expectations at 35 (% of interviews)	−0.080***	−0.054***	−0.060***

The first model in Table 2 shows that the number of children and age are both negatively associated with labor force participation. When human capital letter measures are added, the negative association of children remains largely unchanged, just the negative event of age is diminished, specially for women in their 50s. In all likelihood, this reflects the high correlation betwixt historic period and accumulated work experience. In the final model, which adds interactions of parity with age, interactions are all positive and highly significant. Labor force rates tend to rebound later in the life course, counterbalancing some of the earlier negative effects of children and of aging. This pattern is clearly visible in Figure one which shows the net effects of motherhood (calculated every bit the sum of the principal effect of parity plus the interaction of parity and age) for mothers relative to childless women for women in their 20s, 30s, 40s, and 50s. It is evident throughout the life form that college parity is associated with lower employment rates, but the touch at all parities declines with age then that by the 40s, there is very trivial difference between mothers and childless women.

Notation: Net motherhood effects on employment are calculated from Table two as the sum of the coefficients for children e'er born and the interaction betwixt children e'er born and age decade. Asterisks denote significant differences in the chief effects (vs. childless women in their 20s), and daggers denote significant differences in the interactions (vs. women of the same parity in their 20s).

***/†††p < .001.

Results from stock-still-effects models predicting the wages of employed women are presented in Table 3. Every bit expected on the basis of prior studies, children are negatively associated with women's wages, and controlling for homo capital explains much, just not all, of the motherhood penalty (Budig & England, 2001). Whereas the first model suggests a very big gross motherhood penalty whereby having two or more children reduces the growth of wages by 12% to 17%, this result drops to only 3% once we control for human upper-case letter differences. When we add Age × Parity interactions in the tertiary model, the main furnishings attenuate and lose significance, suggesting that mothers in their 20s (the omitted historic period category) face a minimal wage penalisation at whatever parity. The negative and meaning interactions for mothers in their 30s with two or more children suggest a significantly larger punishment in the 30s than in the 20s. In both the 40s and early 50s, even so, we meet significantly larger penalties only for loftier-parity women with 3 or more than children, compared with the effects for women in their 20s. The penalties for mothers over age 40 who have simply one or two children are not significantly different than the small and nonsignificant penalties in the 20s.

Table 3

Fixed-Furnishings Coefficients Predicting Women'south (ln) Hourly Wages: National Longitudinal Survey of Young Women, 1968–2003

Predictor	Gross event	+ Human capital	+ Decade interaction
N	4,351	4,351	4,351
Person-twelvemonth observations	35,272	35,272	35,272
Children ever born (CEB; ref.: childless)
Ane kid	−0.052***	−0.011	−0.011
2 children	−0.122***	−0.031**	−0.020^
Three or more children	−0.172***	−0.033*	0.011
Age decade (ref.: 20s)a
30s	0.020*	0.003	0.023*
40s	−0.033*	−0.025^	−0.004
50s	−0.106***	−0.048*	−0.038
CEB–age decade interaction (ref.: 20s × childless)
30s × ane child			−0.008
30s × ii children			−0.039**
30s × 3 or more children			−0.053**
40s × one child			−0.018
40s × ii children			−0.019
40s × iii or more children			−0.068***
50s × one kid			−0.017
50s × 2 children			−0.005
50s × 3 or more than children			−0.051*
Highest educational degree (ref.: <high school)
High schoolhouse graduate		0.021	0.027^
Some college		0.073***	0.079**
College grad and beyond		0.219***	0.222***
Work status, cumulative experience, and preparation
Cumulative years of full-time work feel		0.052***	0.052***
Years of full-time work experience squared		−0.001***	−0.001***
Cumulative years of office-time work feel		0.023***	0.024***
Years of office-fourth dimension work experience squared		0.000	0.000
Currently employed function time		−0.039***	−0.040***
Years in job grooming		0.053***	0.052***
Years in task training squared		−0.004***	−0.004***
Sociodemographic controls
Marital status at interview (married = 1, other = 0)	−0.023***	−0.029***	−0.030***
Hubby's income (in thousands of 1990 dollars)	0.001***	0.001***	0.001***
Habitation-oriented expectations at 35 (% of interviews)	−0.006***	−0.002	−0.002

These patterns are axiomatic in Figure ii in which we accept plotted the net furnishings of maternity from these regressions: The wage penalties at parities above one child are significantly larger in the 30s than in the 20s, merely past the 40s and 50s only the penalties for the highest parity mothers (with three or more children) remain statistically pregnant, and fifty-fifty they abound smaller by the 50s (4% penalty in the 50s vs. half-dozen% in the 40s). In sum, these results advise very unlike wage penalties co-ordinate to parity: Whereas having one child never appears to significantly hurt mothers' wages, the touch on of having more children increases significantly by the 30s, merely so continues past age forty only for high-parity mothers with at least three children; by the time they reach their 40s, mothers with two children no longer suffer a pregnant wage punishment.

Note: Internet motherhood effects on hourly wages are calculated from Table 3 as the sum of the coefficients for children e'er born and the interaction between children e'er born and age decade. Asterisks and carats denote significant differences in the master effects (vs. childless women in their 20s), and daggers denote pregnant differences in the interactions (vs. women of the same parity in their 20s).

^p < .x. †/*p < .05. ††/**p < .01. †††/***p < .001.

The results for occupational prestige in Table four appear to be more similar to the labor force participation results than the wage results. The first model in Table 4 shows that having children is negatively associated with occupational attainment and, net of parity, occupational attainment is everyman for women in their 50s (compared with women in their 20s). The event of age is largely eliminated in models controlling for human capital letter, and the size of the negative effects of children are diminished, although they remain statistically significant in the 2d model. In the model with Age × Parity interactions, the interactions are positive and statistically significant for women with two children in their 30s and 40s and for all mothers in their 50s, suggesting that the occupational penalties for mothers decline significantly at older ages. This pattern is evident in Effigy 3 in which are plotted the cyberspace furnishings of motherhood on occupational status. The negative furnishings of motherhood decline substantially betwixt the 30s and 40s, and by the fourth dimension employed mothers reach age 50, their occupational attainment is, if anything, higher than that for childless women, suggesting an occupational premium for older employed mothers. Nosotros return to this finding in the Discussion section. The trajectory of improvement in mothers' occupational attainment over the life course (relative to childless women) is very consistent with the rebound in labor force participation seen in Table 2 and Figure 1 The occupation results are also consistent, though fifty-fifty more clear cutting, than the relative improvement in wages for all but the highest parity mothers.

Note: Cyberspace motherhood furnishings on occupational prestige (Hauser–Warren Socioeconomic Alphabetize [HWSEI]) are calculated from Table four as the sum of the coefficients for children always built-in and the interaction between children ever born and age decade. Asterisks denote significant differences in the master furnishings (vs. childless women in their 20s), and daggers denote significant differences in the interactions (vs. women of the same parity in their 20s).

*/†p < .05. **/††p < .01. ***/†††p < .001.

Table 4

Fixed-Effects Coefficients Predicting Women'southward Occupational Prestige (Hauser–Warren Socioeconomic Index) Scores: National Longitudinal Survey of Immature Women, 1968–2003

Predictor	Gross effect	+ Human upper-case letter	+ Decade interaction
N	4,476	four,476	4,476
Person-year observations	39,569	39,569	39,569
Children e'er built-in (ref.: childless)
One child	−1.561***	−ane.030***	−ane.134***
Two children	−2.044***	−0.998***	−ane.573***
Three or more children	−2.875***	−one.250***	−one.424**
Age decade (ref.: 20s)a
30s	0.154	0.264	0.091
40s	−0.237	0.124	−0.537
50s	−1.375**	−0.647	−2.345***
CEB–age decade interaction (ref.: 20s × childless)
30s × 1 kid			0.180
30s × 2 children			0.726*
30s × 3 or more children			−0.179
40s × 1 kid			0.597
40s × 2 children			ane.167**
40s × 3 or more children			0.617
50s × 1 child			i.917**
50s × 2 children			2.229***
50s × 3 or more children			1.706**
Highest educational degree (ref.:<high school)
High school graduate		ane.579***	1.500***
Some college		v.427***	5.302***
College grad and beyond		ten.300***	10.225***
Piece of work status, cumulative experience, and training
Cumulative years of full-time work experience		0.235***	0.226***
Years of full-time work experience squared		0.000	0.000
Cumulative years of part-time work experience		0.115^	0.102^
Years of office-time work feel squared		0.001	0.001
Currently employed part fourth dimension		−i.533***	−1.504***
Years in chore grooming		0.905***	0.932***
Years in job training squared		−0.048**	−0.050**
Sociodemographic controls
Race (non-Hispanic White=1, other = 0)
Marital status at interview (married = 1, other = 0)	−0.028	−0.003	0.038
Hubby's income (in thousands of 1990 dollars)	0.007*	0.006^	0.006^
Home-oriented expectations at 35 (% of interviews)	−0.240***	−0.186***	−0.190***

Discussion

Building on prior studies of the maternity wage penalty, we have looked at the long-term association between motherhood and multiple aspects of women's careers in society to assess whether career penalties ease or accumulate over the life course. Exercise the careers of mothers eventually grab upwards to those of childless women, or practise mothers fall farther backside every bit they age? Nosotros used fixed-effects methods with panel data from the NLS-YW to model the irresolute impact of motherhood on women'south employment, wages, and occupational status as women age from their 20s to their early 50s.

Motherhood may accept a long attain throughout women's lives, but this analysis has shown that its impact on women's careers attenuates over the life form. Children reduce labor strength participation, but this effect is strongest when women are younger, in their 20s and 30s, when their children are younger likewise. Later in life, equally children age, there may exist counter pressures for mothers to increment their labor supply to meet the financial needs of older children. Our employment results showing the narrowing motherhood gap after the 30s gives the states more confidence in interpreting our other results. Whereas prior inquiry has generally ignored selection into the labor force and simply estimated wage penalties amongst working women, we specifically tested for a motherhood gap in employment. Our results confirm that once women are in their 40s, selection into the labor force should not be biasing our wage and occupational prestige estimates. And, in fact, in additional sensitivity analyses in which nosotros assigned imputed wages (and occupational status scores) to nonworking women, nosotros came to like conclusions for the full sample every bit we present here for working women (results available on asking). This provides further support that selection into employment does not alter our results.

The rewards of mothers' careers, both in terms of wages and occupational status, also appear to regain ground as women age into their 40s and 50s, though with some differences by parity. Whereas mothers with three or more children go along to suffer pregnant wage penalties of at least 4% per kid well into their 40s and 50s, lower parity mothers have generally narrowed the wage gap with childless women by their 40s. In fact, our results suggest that having merely i child never significantly hurts a mother'south wages. The persistent wage penalty for older high-parity mothers is an important finding, because this is the menses of the life form when mothers confront both the financial needs of older children too as their own need to salvage for retirement. Wage penalties make financing children's needs and one'south own future income security more difficult. Yet information technology appears that, for the majority of mothers who take fewer than iii children, their wages approach those of childless women by the time they reach their 40s. If the number of high-parity women continues to decline, this could point a further reduction of the maternity penalty in the hereafter (Kirmeyer & Hamilton, 2011).

Our findings with regard to the wage penalty are consistent with by research, though they reveal the much greater complexity of patterns past both age and parity. Our results for younger women are consistent with findings of studies such as Budig and England (2001), who institute a significant wage penalisation (of about 3%–v%) for women in their 20s and 30s. Nevertheless, by focusing on differences across the life course, our analysis shows that the wage penalty is really relatively low in the 20s and peaks in the 30s for women with two children and in the 40s for women with three or more children. Our wage results are too consistent with the imitation earnings gaps estimated past Sigle-Rushton and Waldfogel (2007), which showed a substantial narrowing of the earnings gap by age 45. Merely past simulating the patterns but for mothers with either one or two children, that study missed the persistent wage gap that we found for higher parity mothers. Our employ of longitudinal data that reflect the aggregating of work and family experiences over the life grade provides a much richer and more accurate cess of the long-term consequences of motherhood for women's careers than is possible using simulations based on cantankerous-exclusive data.

In terms of occupational status, the negative effects of motherhood decline essentially between the 30s and 40s at all parities, and so that by the time employed mothers accomplish historic period 50 their occupational attainment is, if anything, higher than that for childless women. This credible occupational "premium" for older mothers, net of man capital, is intriguing and worthy of further investigation. One possible explanation for this surprising finding is the changing selection into childlessness at older ages. Childless women in their 40s or 50s are an interesting combination of those who remained childless voluntarily (positively selected for having chosen a career or other pursuits instead of maternity), and those who concluded up childless confronting their own will (negatively selected either because of infertility, poor health, the inability to find a suitable partner, or family demands such as caring for aging or disabled relatives, all of which might also affect their market performance). As shown earlier, however, nosotros did not detect that childless women at older ages are less positively selected than younger childless women, at least in terms of their average level of education: Childless women in their 50s had the highest levels of educational activity in our sample (run into Table i). To come across whether the changing selectivity into childlessness was somehow distorting our results, we ran sensitivity analyses using a stock-still measure of completed (final) fertility instead of the time-varying mensurate of cumulative fertility and then that the childless comparison group was fixed over time; we found that the choice of a stock-still versus time-varying measure of fertility made little difference to the results at older ages (results available on request).

As with whatsoever report, this one has several limitations. Start, the NLS-YW is an older accomplice, and much has changed in the work and family lives of subsequent cohorts, peculiarly in terms of their higher educational attainment, delays in spousal relationship and childbearing, and their lifetime attachment to the labor force. Some studies have shown that motherhood penalties are smaller for mothers who delay childbearing (Chandler et al., 1994; Miller, 2011), but nosotros practice not know whether these penalties persist over the long term. As noted earlier, several studies (e.yard., Avellar & Smock, 2003; García-Manglano, 2012b) have compared the motherhood wage penalty for the NLS-YW and NLSY79 cohorts and institute no change over fourth dimension, at least for outcomes measured during the 20s and 30s. The NLSY79 cohort is now reaching their 50s, then it should soon be possible to replicate the electric current analysis for the more recent cohort.

A second limitation of the nowadays study relates to the causal interpretation of our fixed-effects results. Although fixed-effects models accept into business relationship unobservable characteristics that remain fixed over time, just unchanging unobserved characteristics can be considered "controlled" in this type of specification. Attitudes and preferences virtually work and family may not remain fixed over the life grade, peculiarly as women have children, discover both the costs and benefits of mothering, and come across the difficulties inherent in combining paid work with childrearing (Gerson, 1986; Hakim, 2002; Shaw & Shapiro, 1987). In this report, we attempted to command for changes in preferences with our fourth dimension-varying indicator of what women expected to exist doing at age 35, and we institute that not only did preferences change over time, but they were also significant predictors of employment and occupational achievement, net of other factors. We recognize, however, that there may be other unobserved characteristics that affect outcomes, that change over time, and that are non captured in our fixed furnishings modeling approach.

Our findings highlight the multidimensional examination of afterward life outcomes that is needed to provide a full picture of any life form "motherhood penalisation." Not only wages, but also occupational standing and employment levels need to be considered to assess where women with dissimilar child-bearing trajectories cease up. In further work, it would also exist important to consider women's accumulated avails and pension benefits and, related, their marital history, which also connects them to afterward life benefits and well-existence. Finally, more attention should be paid to the selectivity of employment, especially at older ages. In by enquiry that has evaluated the motherhood penalty among younger women in their 20s and 30s, all women were observed at ages when labor force zipper was strong for women and retirement was non yet an option. When because later life stages and accumulated careers, the differential timing of retirement that is afforded those with potent attachment to "good jobs" with (early) retirement options becomes a more important function of the story. In futurity work, this as well needs to be better conceptualized and explored so that the trajectories and economical outcomes of mothers with differing work and family careers can exist adequately compared.

Acknowledgments

This research was supported in part by a grant from the Russell Sage Foundation besides as by funds provided to the Maryland Population Research Centre (Grant R24-HD041041) and the California Center for Population Research (Grant R24-HD041022) from the Eunice Kennedy Shriver National Center for Child Health and Homo Development. We gratefully admit the helpful comments from Paula England, Michelle Budig, and Larry Kahn. An before version of this commodity was presented at the 2010 annual meeting of the Population Clan of America, Dallas, TX.

References

• Aisenbrey Due south, Evertsson M, Grunow D. Is there a career punishment for mothers' time out? Germany, Sweden and the US compared. Social Forces. 2009;88:573–605. [Google Scholar]
• Anderson DJ, Binder 1000, Krause Grand. The motherhood wage penalty revisited: Experience, heterogeneity, piece of work effort, and work-schedule flexibility. Industrial and Labor Relations Review. 2003;56:273–294. [Google Scholar]
• Avellar S, Smock PJ. Has the price of motherhood declined over time? A cross-cohort comparing of the motherhood wage penalty. Journal of Marriage and Family. 2003;65:597–607. [Google Scholar]
• Baum CL. The effect of work interruptions on women's wages. Labour. 2002;16:i–36. [Google Scholar]
• Blackburn ML, Bloom DE, Neumark D. Fertility timing, wages, and man capital. Journal of Population Economics. 1993;vi:ane–30. [PubMed] [Google Scholar]
• Budig MJ, England P. The wage penalty for motherhood. American Sociological Review. 2001;66:204–225. [Google Scholar]
• Budig MJ, Hodges MJ. Differences in disadvantage. American Sociological Review. 2010;75:705–728. [Google Scholar]
• Chandler TD, Kamo Y, Werbel JD. Do delays in matrimony and childbirth affect earnings? Social Science Quarterly. 1994;75:838–853. [Google Scholar]
• Correll SJ, Benard S, Paik I. Getting a job: Is at that place a motherhood penalty? American Journal of Folklore. 2007;112:1297–1338. [Google Scholar]
• Davies H, Joshi H, Peronaci R. Forgone income and maternity: What do recent British data tell us? Population Studies. 2000;54:293–305. [PubMed] [Google Scholar]
• Desai Southward, Waite LJ. Women'due south employment during pregnancy and after the commencement birth: Occupational characteristics and work delivery. American Sociological Review. 1991;56:551–566. [Google Scholar]
• Estes SB, Glass JL. Job changes following childbirth. Piece of work and Occupations. 1996;23:405–436. [Google Scholar]
• Evertsson M, Breen R. The importance of work. irresolute work commitment following the transition to parenthood; Boston. Newspaper presented at the annual meeting of the American Sociological Association.2008. Aug, [Google Scholar]
• Gangl M, Ziefle A. Motherhood, labor force behavior, and women's careers: An empirical cess of the wage penalty for motherhood in Britain, Germany, and the United States. Demography. 2009;46:341–369. [PMC costless commodity] [PubMed] [Google Scholar]
• García-Manglano J. The human capital letter argument revisited: Women'due south piece of work expectations, human upper-case letter aggregating and market outcomes at midlife; San Francisco. Poster presented at the annual meeting of the Population Association of America.2012a. May, [Google Scholar]
• García-Manglano J. A tale of two cohorts: The market consequences of depression work expectations for early and belatedly Infant Boomers in the The states; London, Uk. Paper presented at the annual conference of the Heart for Longitudinal Studies.2012b. Nov, [Google Scholar]
• Gerson K. Hard choices: How women make up one's mind about piece of work, career, and maternity. Berkeley: University of California Press; 1986. [Google Scholar]
• Hakim C. Lifestyle preferences equally determinants of women'due south differentiated labor market careers. Work and Occupations. 2002;29:428–459. [Google Scholar]
• Hauser RM, Warren JR. Socioeconomic indexes for occupations: A review, update, and critique. Sociological Methodology. 1997;27:177–298. [Google Scholar]
• Jacobsen JP, Levin LM. Effects of intermittent labor strength attachment on women'south earnings. Monthly Labor Review. 1995;15:14–19. [Google Scholar]
• Joshi H, Hinde PRA. Employment after childbearing in mail service-war Britain: Cohort study evidence on contrasts within and across generations. European Sociological Review. 1993;nine:203–227. [PubMed] [Google Scholar]
• Joshi H, Macran S, Dex Due south. Employment afterward childbearing and women's subsequent labour force participation: Bear witness from the British 1958 birth cohort. Journal of Population Economic science. 1996;ix:325–348. [PubMed] [Google Scholar]
• Kirmeyer SE, Hamilton BE. Childbearing differences among three generations of U.South. women. Hyattsville, Dr.: National Center for Health Statistics; 2011. Data Brief No 68. [PubMed] [Google Scholar]
• Klerman JA, Leibowitz A. Task continuity among new mothers. Census. 1999;36:145–155. [PubMed] [Google Scholar]
• Loughran D, Zissimopoulos JM. Why wait? The event of marriage and childbearing on the wage growth of men and women. Periodical of Human Resources. 2008;44:326–349. [PMC gratis article] [PubMed] [Google Scholar]
• Mandel H, Semyonov 1000. Family policies, wage structures, and gender gaps: Sources of earnings inequality in 20 countries. American Sociological Review. 2005;70:949–967. [Google Scholar]
• Mandel H, Semyonov M. A welfare country paradox: Country interventions and women's employment opportunities in 22 countries. American Journal of Sociology. 2006;111:1910–1949. [Google Scholar]
• Miller A. The effects of maternity timing on career path. Journal of Population Economics. 2011;24:1071–1100. [Google Scholar]
• Mincer J, Ofek H. Interrupted work careers: Depreciation and restoration of human capital. Journal of Man Resources. 1982;17:3–24. [Google Scholar]
• Mincer J, Polachek Due south. Family investments in man uppercase: Earnings of women. In Marriage, family unit, human capital, and fertility. Cambridge, MA: National Agency of Economic Enquiry; 1974. pp. 76–110. [Google Scholar]
• Okamoto D, England P. Is there a supply side to occupational sex segregation? Sociological Perspectives. 1999;42:557–582. [Google Scholar]
• Shaw LB, Shapiro D. Women's piece of work plans: Contrasting expectations and actual piece of work experience. Monthly Labor Review. 1987 Nov;110:7–xiii. [Google Scholar]
• Sigle-Rushton W, Waldfogel J. Motherhood and women'southward earnings in Anglo-American, continental European and Nordic countries. Feminist Economic science. 2007;13:55–91. [Google Scholar]
• Staff J, Mortimer JT. Explaining the maternity penalty during the early occupational career. Demography. 2012;49:ane–21. [PMC free commodity] [PubMed] [Google Scholar]
• Taniguchi H. The timing of childbearing and women's wages. Journal of Marriage and Family unit. 1999;6:1008–1019. [Google Scholar]
• Waldfogel J. The cost of motherhood: Family status and women's pay in a young British cohort. Oxford Economic Papers. 1995;47:584–610. [Google Scholar]
• Waldfogel J. The effect of children on women'southward wages. American Sociological Review. 1997;62:209–217. [Google Scholar]
• Wilde ET, Batchelder L, Ellwood D. The mommy track divides: The touch on of childbearing on wages of women of differing skill levels. Cambridge, MA: National Agency of Economic Research; 2010. Working Paper No. 16582. [Google Scholar]

rubinsubly1975.blogspot.com

Source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4041155/

Share this post