UBC Faculty Research and Publications

“Family-Friendly” Jobs and Motherhood Pay Penalties : The Impact of Flexible Work Arrangements Across… Fuller, Sylvia; Hirsh, C. Elizabeth 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-Fuller_S_et_al_Family_Friendly.pdf [ 264.83kB ]
Metadata
JSON: 52383-1.0366253.json
JSON-LD: 52383-1.0366253-ld.json
RDF/XML (Pretty): 52383-1.0366253-rdf.xml
RDF/JSON: 52383-1.0366253-rdf.json
Turtle: 52383-1.0366253-turtle.txt
N-Triples: 52383-1.0366253-rdf-ntriples.txt
Original Record: 52383-1.0366253-source.json
Full Text
52383-1.0366253-fulltext.txt
Citation
52383-1.0366253.ris

Full Text

  “Family-Friendly” Jobs and Motherhood Pay Penalties: The Impact of Flexible Work Arrangements Across the Educational Spectrum   Sylvia Fuller* C. Elizabeth Hirsh*  Postprint version of:  Fuller, Sylvia and C. Elizabeth Hirsh (2018) “‘Family Friendly’ Jobs and Motherhood Career Penalties The Impact of Flexible Work Arrangements Across the Educational Spectrum?” Work and Occupations http://journals.sagepub.com/doi/abs/10.1177/0730888418771116?journalCode=woxb    *Department of Sociology, the University of British Columbia sylvia.fuller@ubc.ca  ABSTRACT This article focuses on how flexible work arrangements affect motherhood wage penalties for differently situated women. While theories of work–life facilitation suggest that flexible work should ease motherhood penalties, the use of flexibility policies may also invite stigma and bias against mothers. Analyses using Canadian-linked workplace–employee data test these competing perspectives by examining how temporal and spatial flexibility moderate motherhood wage penalties and how this varies by women’s education. Results show that flexible work hours typically reduce mothers’ disadvantage, especially for the university educated, and that working from home also reduces wage gaps for most educational groups. The positive effect of flexibility operates chiefly by reducing barriers to mothers’ employment in higher waged establishments, although wage gaps within establishments are also diminished in some cases. While there is relatively little evidence of a flexibility stigma, the most educated do face stronger wage penalties within establishments when they substitute paid work from home for face time at the workplace as do the least educated when they bring additional unpaid work home. Overall, results are most consistent with the work–life facilitation model. However, variability in the pattern of effects underscores the importance of looking at the intersection of mothers’ education and workplace arrangements.   INTRODUCTION For many families, integrating work and care remains a challenge. Both academic research and popular media often claim that “flexibility” in work arrangements is a critical resource that helps parents manage competing demands across domains. At the same time, some qualitative and experimental research suggests that flexible work may carry a price (Munsch, Ridgeway & Williams 2014, Williams, Blair-Loy, & Berdahl 2013). Arguments about a “flexibility stigma” posit that rearranging one’s work to accommodate the demands of care violates deeply held assumptions about what it means to be an “ideal worker”. Gendered norms surrounding parenting encourage mothers in particular to adapt their employment to family demands. Consequently, understanding whether and how such adaptations exacerbate or ameliorate earnings penalties associated with motherhood is important. Of course, not all mothers have the same access to workplace flexibility. Intersectional feminist scholarship has reinforced the point that experiences of employment are far from uniform among women. Notably, differences in social class and educational attainment (along with other social relations, such as those of race and ethnicity) affect access to the kinds of jobs that might have flexibility as an option (Clawson & Gerstel, 2014; Duxbury & Smart, 2011; Kossek & Lautsch 2017). Those in professional and managerial jobs are more likely to enjoy various forms of flexible work arrangements as compared to those in shift or frontline service work (Estes & Glass, 1996; Glass & Noonan, 2016; Johnson & Provan, 1996).  Moreover, it is not simply access to different kinds of work that may vary by class and education, but also their consequences. Different jobs are associated with variation in the intersection of employment norms, power relations, cultural expectations, and household resources. For example, employment norms in professional jobs that presuppose an “ideal   worker” devoted to their job above all else especially clash with upper-middle-class expectations for intensive mothering (Blair-Loy 2003). At the same time, the higher remuneration available to those with more education can allow access to more reliable child-care, reducing the odds that care demands will conflict with work obligations.  The small body of research addressing the consequences of workplace flexibility for mothers’ careers has tended to focus on relatively homogeneous samples of women (typical of qualitative and experimental research), or on aggregate impacts for mothers as a whole. Studies explicitly comparing the impact of differing workplace conditions for motherhood pay gaps across groups of women are currently lacking. Yet, as we argue below, the familiar theoretical frameworks marshaled to understand the motherhood pay gap, including work-life conflict, work effort, compensating differentials, and discrimination, all imply that workplace conditions should matter, but differently for women of varying social positions. Indeed, in a recent study, Yu and Kuo (2017) show that mothers’ wage penalties vary with their occupational conditions, including attributes such as job autonomy, schedule regularity, competitiveness, and team work.  Drawing on data from Statistics Canada’s Workplace and Employee Survey (WES), we identify the workplace circumstances under which motherhood pay gaps are widened or narrowed for women with varying levels of education. We focus on two work conditions that affect the ease with which individuals can reconcile employment with caregiving: control over start and stop times (temporal flexibility), and performing some or all work duties from home (spatial flexibility). Recognizing that both access to flexible work arrangements and the capacity to combine work and caregiving are experienced differently by women of varying social positions, we examine how the effect of temporal and spatial flexibility on wages varies by women’s educational level. Finally, taking advantage of the linked employee-workplace nature   of the WES, we go beyond past research that has relied solely on individual-level data to assess the extent to which the impact of flexible arrangements on pay gaps is generated within or across establishments.  PARENTING PAY GAPS AND WORK CONDITIONS: THEORY AND EVIDENCE Scholars have investigated a number of potential mechanisms underlying parenting pay gaps. Individual-level perspectives focus on underlying differences between women who become parents and those who do not, such as differences in career ambition (Lundberg & Rose, 2000, 2002) or work effort in the face of caregiving responsibilities (Becker, 1993; Budig & England, 2001; Kmec, 2011). Human capital theory highlights the role of employment breaks after birth, which reduce mothers’ experience and therefore, wages (Aisenbrey, Evertsson, & Grunow, 2009; Budig & England, 2001; Phipps, Burton & Lethbridge 2001). If breaks are longer than statutory protected leaves, they also force women to find new employers when they re-enter the labor force, reducing seniority and firm-specific capital, and leading to larger wage losses (Baum, 2002; Bishop, 1997; Phipps et al., 2001). While such individual-level explanations help unravel women’s different career pathways and experiences, they are silent on how women actually negotiate their work lives in the context of organizational options and constraints. Moving beyond an individual perspective, the recent work of Yu and Kuo (2017) examines how the structural conditions of mothers’ occupations, including levels of schedule regularity, autonomy, on-the-job training, competitiveness, teamwork, and exposure to hazardous conditions – can moderate wage penalties. Building on and extending this research, our approach focuses on how particular employment arrangements exacerbate or diminish motherhood penalties by altering how women organize their work both temporally and spatially. As our data includes linked employee-employer information, we go   beyond previous studies in that we observe women in their actual workplace settings and directly measure the use of flexible work arrangements rather than simply its availability or as an occupational average (i.e., Yu and Kuo 2017).  In the discussion that follows, we outline how flexible work arrangements may impact motherhood penalties and how these in turn may vary in their effects by mothers’ education. On the one hand, perspectives based on theories of work-life facilitation suggest that flexible work arrangements should make it easier for mothers to combine earning and caring by allowing women to accommodate care demands without disrupting productivity and performance at work. This has the potential to mitigate any performance differences between mothers and childless women, as well as associated employer biases. On the other hand, status characteristics approaches to motherhood and theories of compensating differences warn that flexible work arrangements may undermine women’s advancement by activating stereotypes and requiring workers to make tradeoffs between family-friendly arrangements and career progress. For both perspectives, the salience of particular flexible work arrangements for motherhood penalties may also vary with education insofar as it shapes access to various employment options and the resources available for care. While considerable research investigates variation in the motherhood wage disadvantage, we have little direct quantitative evidence about how alternative work arrangements may differentially impact mothers of with varying educational attainment. Thus we consider not only how flexibility affects motherhood wage gaps but also its impact across different education-status groups. Finally, exploiting our employee- and establishment-level data, we explore how flexibility affects mothers’ pay penalties both within and across establishments, allowing us to consider the potential benefits and perils of flexibility for mothers’   pay relative to childless women in their current workplace and in terms of mothers’ access to employment in desirable, well-paying firms.  Flexibility and Work-Life Facilitation for Mothers The expectation that flexible work arrangements should enhance mothers’ career outcomes comes from theories of work-life facilitation, which point to so-called “family friendly” policies as protecting mothers from work-family conflict and enhancing mothers’ career progress (Johnson & Provan, 1996). In contrast, policies that rest on ideal worker norms and expectations of total career devotion will undermine mothers’ advancement and wages.  Flexibility and job performance. Arguably the most important work-life facilitation policy is that of scheduling flexibility, which provides workers with greater autonomy and control over when and where their work occurs. Temporal flexibility – or flexibility in when work occurs – helps workers accommodate family demands and deal with unpredictable family circumstances, such as school closures and sick children or caregivers without incurring penalties for absenteeism (Anderson, Binder, & Krause, 2003; Boushey, 2008; Damaske, Ecklund, Lincoln, & White, 2014). Spatial flexibility – or flexibility in the location of work, such as the option to work from home or off-site – facilitates combining employment with mothering by reducing commute times and more efficiently integrating paid work and care. To the extent that both minimize work-family strain1, temporal and spatial flexibility may reduce fatigue and distraction that could harm parents’ productivity (Glass, 2004) and allow workers to maintain productivity in the face of unpredictable family circumstances. While potentially beneficial to all workers, flexibility should matter more for mothers given their disproportionate responsibility for care-work. By reducing work performance differences between mothers and childless women, flexibility, then, should level the playing field within workplaces.     Flexibility and reducing barriers to hire. Flexible work arrangements may also moderate wage gaps by reducing employer biases against mothers that can pose barriers to employment at more desirable firms. A convincing body of experimental and qualitative research shows that employers hold biased assumptions about mothers’ job commitment and competency (Bornstein, 2013; Correll et al., 2007; Cuddy, Fiske, & Glick, 2004; Demaiter & Adams, 2008; Ridgeway & Correll, 2004). Employers are especially likely to rely on such stereotypes when they have less information about employees, such as at the point of hire. Stereotypes about mothers that characterize them as less committed and productive may prove a particular barrier to employment at high-wage firms.2 Where work can be organized flexibly, employers may be less likely to see the demands of their jobs as incompatible with motherhood, thus alleviating concerns about mothers’ productivity, work effort, and job commitment. Indeed, recent research shows that mothers’ relegation to lower-paying firms is a key mechanism underlying motherhood pay penalties, and that organizational characteristics that should reduce opportunities for discrimination at the hiring stage (unionization, HR departments, non-profit) also substantially reduce or eliminate motherhood wage gaps in Canada (Fuller 2017). Flexibility may similarly serve as a brake on employers’ bias at the point of hire, providing mothers equal access to desirable firms and jobs, and thus reducing between-establishment wage penalties that might arise due to mothers’ overconcentration in lower-paying establishments. To date, evidence of the positive impact of temporal and spatial flexibility on wage penalties for mothers is limited. Glauber (2012) finds work schedule flexibility only decreases motherhood penalties for women in female dominated occupations, and those with one child in male-dominated or integrated occupations. However, she is only able to measure access to flexibility not use. Yu and Kuo (2017) find that wage penalties for motherhood are smaller in   occupations with high levels of autonomy, yet find no evidence that schedule regularity reduces mothers’ wage penalties. Because both studies use individual-level survey data, neither is able to isolate whether these effects occur by minimizing wage gaps within workplaces or by reducing barriers to hire that sort mothers into lower-paying establishments. Flexibility, facilitation, and education effects. From a facilitation perspective, there are a number of reasons to expect that flexibility in the place and time of work will vary in its effect for more or less educated women, though the nature of this relationship is unclear. On the one hand, less educated women experience more rigid work schedules and less access to flexibility, as compared to women in higher status, professional jobs (Glass and Noonan 2016; Johnson and Provan 1996). Both tardiness and absenteeism due to caregiving demands can lead to job loss, and low-wage workers who signal a need for flexibility are routinely let go (Williams, 2006). Thus less educated women arguably stand to benefit more from temporal and spatial flexibility as compared to higher educated women who face less rigid schedules and discipline systems. Less educated mothers are also more vulnerable to disruptions caused by unpredictable care demands and hours. Given the high cost of licensed centre-based childcare in Canada, less educated workers more commonly rely on informal care, which is, by its nature, less reliable (Strazdins, Clements, Korda, Broom, & D’Souza, 2006). Low income is also associated with more health problems for children (and mothers), which can interfere with mothers’ employment (Pulkingham & Fuller, 2012). Thus, less educated workers may experience more interruptions to work and lack the resources to provide critical buffers in the absence of job flexibility. This reasoning implies that flexibility may be more advantageous in ameliorating motherhood penalties for less educated women as compared to more highly-educated women, both with respect to reducing actual performance differences (hence minimizing wage differences within   establishments) and alleviating employer concerns (thus making it less likely mothers will be sorted into less desirable workplaces).  On the other hand, flexible hours and the ability to work from home may be particularly critical for mothers with higher education as they face greater demands for productivity. The “career devotion schema,” which dictates that work should be an all-encompassing commitment (Blair-Loy, 2003; Stone & Hernandez, 2013), is especially common in professional and managerial jobs, and expectations of long hours are deeply embedded in organizational cultures (Cha & Weeden, 2014). Such demands are easier to meet while still fulfilling care obligations when some work may be accomplished at flexible times or at home when children are co-present (Ridgeway & Correll, 2004). This suggests that highly educated mothers in particular would face smaller within-establishment wage gaps and also be less often sorted into lower-paying establishments than their childless counterparts when work arrangements are flexible.  The Perils of Flexibility: Stigma, Discrimination, and Trade-offs While the worklife-facilitation approach suggests that indicators of workplace flexibility will mitigate motherhood wage penalties, other theoretical and empirical work draws attention to the perils of so-called “family friendly” or flexible work arrangements for women and mothers. According to this logic, the use of such policies may activate gendered stereotypes, call into question one’s commitment to work, invite discrimination, and come at the expense of career progress and wages. Discrimination, stigma, and motherhood status. Explanations of the motherhood wage gap focusing on discrimination presume that motherhood is a status characteristic that, when triggered, can impact workplace decision makers’ assessments of women’s competence, commitment, and value to the organization (Correll, Benard, & Paik, 2007; Ridgeway & Correll,   2004). Although the facilitation perspective suggests that flexibility may ease employers’ concerns about mothers’ work commitment and competence, approaches rooted in discrimination argue that flexible work arrangements would invite bias against workers who take advantage of them, insofar as they clash with ideal worker norms (Blair-Loy & Wharton, 2002; Cech & Blair-Loy, 2014; Glass, 2004; Munsch, Ridgeway, & Williams, 2014; Wharton, Chivers, & Blair-Loy, 2008). For mothers, they may be particularly stigmatizing insofar as they are interpreted as accommodations to care-demands, render motherhood status salient, and hence trigger negative assessments of mothers’ job commitment, competency, and performance. Consistent with this, in an experimental study, Cuddy et al. (2004) note that several participants told the experimenter that they had discriminated against the mother because she telecommutes, not because she recently had a child. Glass and Noonan (2016) also find that mothers pay a substantial penalty for reducing their hours on-site below the traditional 40 hours per week, and Kmec et al. (2014) find that mothers who violate ideal working norms (by working anything but full time), perceive more unfair treatment. The “flexibility stigma” should operate within establishments, insofar as it speaks to a violation of organizational work norms. In one of the rare studies to isolate within-establishment wage impacts of flexible scheduling, Winder (2009) finds that the generally positive wage returns associated with flexible hours are less pronounced for mothers, although only for those with older children.  Flexibility stigma and education effects. Of course, as Ridgeway and Correll (2004) note, motherhood may be more or less salient in particular contexts. They argue that jobs that place strict and time intensive demands on workers conflict more with presumptions about mothers’ proper role than jobs that are less demanding. In professional and managerial jobs, employees are   subject to expectations for near-constant availability (Cahusac & Kanji, 2013; Duxbury & Smart, 2011; Schieman, Milkie, & Glavin, 2009) and time on the job is often equated to commitment and productivity (Epstein, Seron, Oglensky, & Saute, 2014; Fried, 1998; Williams, 2010). This implies that taking advantage of flexibility options that reduce the visibility of work-effort, such as flexible hours or working from home instead of in the office, will be particularly stigmatizing for highly-educated mothers. Consistent with this, Glass (2004) reports that months working from home retard wage growth most for managers and professionals.  Compensating differentials. The theory of compensating differentials, which posits that workers trade wages for desirable job features, also points to negative career effects for those who take advantage of flexible work arrangements, but for different reasons. Job characteristics that reduce conflicts between employment and care by increasing flexibility should be especially attractive to mothers (Becker, 1993; Glauber, 2012; Johnson & Provan, 1996). Conversely, jobs that are difficult to reconcile with mothering may be worth avoiding even if they offer high wages. While men as well as workers without children may also value flexibility and request flexible work arrangements (McCrate, 2005), the key assumption of compensating differentials perspectives is that these job features are more valuable to mothers, who are thus willing to accept larger wage trade-offs to have them.  Arguments about both compensating differentials and flexibility stigma posit that flexible work arrangements may exacerbate motherhood wage gaps, but the level at which they operate likely differs. As noted previously, flexibility stigma should operate within establishments, as mothers who take advantage of flexible arrangements are negatively assessed relative to their childless co-workers. Compensating differentials, on the other hand, while also potentially relevant within establishments, should have a greater impact on which establishment one   ultimately works in. The availability of particular work arrangements is often conditional on one’s workplace (Heywood, Siebert, & Wei, 2007; Pocock & Charlesworth, 2017; Sweet et al., 2014). Shifting to a work arrangement more accommodating to caregiving demands may require changing employers and have its wage impact by virtue of leading to segregation in lower-paying establishments. Compensating differentials and education effects. The kinds of trade-offs workers make to access or avoid certain kinds of work arrangements will also vary by education. While the theory of compensating differentials presumes choices are available between jobs with different amenities, empirically, choices are constrained by job and education level.3 More educated workers tend to enjoy more autonomy in their work, making it more amenable to temporal and spatial flexibility (Gerstel & Clawson, 2014; Winder, 2009). Working-class employees, however, are often faced with close supervision, rigid schedules, and little control over their work (Dodson, 2013; Gerstel & Clawson, 2014). Compensating differentials arguments thus imply that more highly educated women will have greater access to temporal and spatial flexibility and, as a result, will be more likely to have opportunities to make wage trade-offs for them.  Flexibility, Motherhood Penalties, and Organizational Segregation In sum, theories of work-family balance and facilitation would predict a negative relationship between workplace flexibility and motherhood wage penalties, such that mothers with more control over the time and place of work would be shielded, at least in part, from wage penalties. This may occur by leveling the playing field within establishments or by reducing barriers to mothers’ hire in those that pay higher wages. If we find evidence for the latter, this also would suggest that work-life facilitation is not simply about enhancing individual mothers’ productivity, but about shifting understandings of what is necessary for “success” in   organizations more broadly. The protective effect of flexibility on wages may vary by mothers’ education and job status, though the direction of this effect is unclear.  Conversely, perspectives of discrimination based on status characteristics and theories of compensating differentials would predict that the uptake of such “family friendly” arrangements would activate biases and lead to the devaluation of mothers’ wages. This would chiefly operate within establishments in the case of the flexibility stigma, while also impacting wages by virtue of how mothers are sorted across establishments in the case of compensating differentials. Because the time intensive demands of high status, professional work conflicts more with assumptions of motherhood and caregiving, the negative wage effects of stigma and trade-offs should be stronger for more highly educated women.  DATA AND METHODS Data come from Statistics Canada’s Workplace and Employee Survey (WES), a linked employer-employee dataset fielded from 1999 to 2005. Employees were followed for two years, and the employer sample was longitudinal with the sample refreshed every second year to maintain its representativeness. The target population is all employers in Canadian provinces that have paid employees, with the exception of public administration (the broader public sector, including health and education is included); crop production and animal production; fishing, hunting and trapping; private households; and religious organizations. The WES sample is drawn from the Business Register, which is a monthly list of all businesses in Canada maintained by Statistics Canada. The sampling frame is stratified by industry, region, and size. In each odd year (1999, 2001, 2003, and 2005), up to 24 employees (either at work or on leave) were randomly sampled from each employer.4 The number of employees sampled varied by employer size; in establishments with fewer than four employees all employees were selected.   We pool data from the first year of each of these waves to maximize sample size. For large establishments, selected individuals are unlikely to overlap from one wave to the next, although this is possible in the smallest organizations. In total, 28% of the individuals in the sample are observed more than once, although this drops to less than 8% for those in the smallest educational groups (those without a high school diploma or with a postgraduate degree). The analytic sample is restricted to workers between the ages of 24 and 44. The WES only reports if workers are currently living with children. Beyond 44, the group without resident children would likely include numerous women who had children at younger ages, and whose children have now left home (we also exclude women 25-44 who have children older than 19 living with them).5 An unavoidable consequence of this restriction is that mothers of older children are underrepresented. While such mothers may have less need for and hence use of flexible work arrangements, insofar as wage penalties accrue over the life course, our estimates of motherhood wage penalties are likely conservative. We truncate the sample at 24 because it is impossible to tell whether individuals are currently enrolled in school, which is problematic when variation by educational attainment is a key concern. This results in a final sample of 20,879 observations, of which 58% are of mothers, employed at 5,805 unique establishments.  Measures  The dependent variable is logged total hourly wage, which accounts not only for base wages, but also overtime premiums, bonuses, profit sharing, tips etc. WES respondents report different bases of pay (i.e. hourly, weekly, yearly), which Statistics Canada converts to an hourly wage rate if necessary. The key independent variable uses a binary mother/childless distinction that reflects whether the person is living with a child under the age of 19. While it is more common to estimate parenthood effects for children under 18, delayed adulthood means a   substantial portion of children still live at home after high-school graduation. We extend the age range by one year to capture the period of dependency while children are transitioning from high school (setting the cut-off to 18 has no substantive impact on estimates). The education measure distinguishes between those with a high school degree or less; those who have completed a non-university postsecondary certificate; undergraduate degree holders; and those with a postgraduate degree.  The key job characteristic variables include two measures of flexibility. Temporal flexibility is assessed with a question that asks “Do you work flexible hours? (this means you may work a certain number of core hours, but you can vary your start and stop times as long as you work the equivalent of a full week)”. The measure thus indexes employee-centred temporal flexibility, which is most relevant as a tool for managing intersections of work and care, not variable scheduling dictated by employer needs. Spatial flexibility is measured with a variable reflecting whether the individual ever works from home and the nature of such work: work that is paid and part of regularly scheduled hours, work that is paid and in addition to regularly scheduled hours, and work that is unpaid and in addition to regularly scheduled hours. The first and last categories are most theoretically relevant, and are the focus of our investigation. Work that is paid and part of regularly scheduled hours represents a substitution model most closely corresponding to the type of worker-centered flexibility that should ease work-family conflicts. Work that is unpaid and in addition to regularly scheduled hours represents work-to-family spillover that may be valuable in helping workers meet intensive work demands, but that is oriented chiefly to employer needs, and which may increase rather than decrease work-family conflict. Work that is paid and in addition to regularly scheduled hours constitutes “paid   overtime” that is used by workers largely to supplement wages rather as an opportunity to ease work-family conflict. Thus, we do not focus on this category. Indicators of flexibility capture actual use, not formal availability. This is important insofar as companies may make flexible arrangements available in theory at the same time as informal norms or supervisor objections make it difficult for workers to actually use them. In such a case, the formal availability of flexibility policies has little bearing on their actual relationship to work-life facilitation or stigma (insofar as the compensating differentials argument speaks to worker selection of employers the availability of flexibility may be germane, although use should still matter more). Further, workers may gain access to flexible arrangements that have implications for motherhood penalties outside of formal policy, either through informal arrangements or because the nature of work and prevailing norms facilitate it (bringing unpaid work home, for example, is unlikely to reflect either formal policy or negotiated informal arrangements per se).  Demographic control variables consist of age and its square, an age by education interaction to capture stronger wage growth among the more educated, and an indicator of racialization and immigration status (white Canadian-born; white immigrant; non-white Canadian-born; non-white immigrant). We also include controls related to household dynamics and resources. Parents who expend less time on domestic work can devote more energy and time to their jobs. Having a partner can be helpful both because they can share domestic work and because additional earnings provide resources for outsourcing it. Because more educated parents are more likely to be partnered and have partners with substantial earnings, these household dynamics should reduce motherhood pay gaps. Household labour is also divided more equally among Canadians with higher levels of education (Marshall, 2009). We therefore control for   living with a spouse (married or common-law), and other family income. Time away from employment and lower tenure is also relevant to earnings. The possibility of longer employment breaks for caregiving may be less available for those with the least and most amount of education - for the former because household resources do not permit prolonged loss of mothers’ earnings, and for the latter because human and social capital depreciation is more damaging (Anderson, Binder, & Krause, 2002).  We control for characteristics that may be correlated both with flexibility and wages: a trichotomous measure of usual/average weekly work hours (including paid and unpaid overtime) to capture distinctions between part time (<35 hours), full-time (35-44), and long hours full time (>44)6 and unionization. We do not control for occupation as we expect that compensating differentials tied to flexibility may operate partly through occupational choice/downgrading. Mothers and childless women across the educational spectrum may differ in their orientation to different types of work conditions, implying differential selectivity into them. The WES provides a rich set of variables that measure individual motivations related to decisions about work intensity, which we include as controls to capture differences in orientations towards employment and relevant barriers. The WES asks employees if they would prefer to work more hours for more pay, the same number of hours for the same pay, or fewer hours for less pay (given their current rate of pay). Those who answer that they would prefer to work more hours are asked why they did not work these additional hours. We construct a set of dummy variables indicating if the individual cited the following reasons7: childcare unavailable; going to school; employer did not offer additional hours; payment not sufficient; personal and family responsibilities (with those who did not want extra hours and those who did but did not cite a given reason coded as zeros). We also include a set of dummies indexing whether individuals   indicated that they would prefer to work fewer hours for various reasons: family responsibilities; work-related stress; health reasons; leisure. To help capture differences in otherwise unmeasured attributes that could be correlated both with motherhood and one’s competitiveness on the job market, we include a measure of the number of months the individual spent unemployed in the last five years. One of the challenges in assessing the wage consequences of different work conditions is that employers may make them differentially available based on the perceived productivity of workers. While we cannot perfectly control for this, we include measures of whether the individual participated in paid or unpaid career-related training in the past year to capture an individual’s standing with their employer (assuming that higher performers are more likely to be offered opportunities for paid training) and career-ambition. Finally, we include a control for disability to capture conditions that may impact both fertility and employment. Models Teasing out dynamics underlying possible relationships between motherhood wage gaps and flexibility requires considering not only the overall direction of wage effects, but the level (within establishments versus across them) at which they operate. To do so, our analysis employs OLS and establishment fixed-effects regression models. We estimate the impact of different flexible work arrangements on motherhood wage gaps by specifying three-way interactions between motherhood, education, and the flexibility measures. First, we use OLS models to assess the aggregate impact of flexible work arrangements on motherhood pay gaps.   = + ∗  ∗  +     (1) In equation 1 is the natural log of hourly wage for individual i in establishment j,  indexes observed individual and job characteristics that affect wages,  ∗  ∗  is a vector capturing the three way interaction between motherhood, education, and measures of   flexibility as well as the two-way interactions between each of these terms and the main effects, and  is a stochastic mean-zero error term. OLS estimates reflect the combined effect of wage differences arising both within-establishments as well as differences due to how women are sorted into higher or lower-paying establishments.  Next, we use models with establishment fixed effects to net out any sorting effects and focus only on within-establishment wage gaps. In these models, any motherhood pay gap is solely a function of wage differences within a given establishment between mothers and childless women. The difference between the two then indicates wage differences tied to patterns of segregation in establishments that, net of the characteristics of their workers, pay higher or lower average wages. Petersen, Penner, and Høgsnes (2011); (2014) use this approach to assess the contribution of establishment segregation to group wage gaps, and Pendakur and Woodcock (2010) provide a formal proof and associated tests of significance for assessing differences in estimates.  Because the number of women in any given establishment in the WES in our age range is relatively small (mean = 3.9), we adapt Canay’s (2011) two-step strategy for estimating fixed effects with sparse within-group observations (in his case used for quantile regressions), which Javdani (2012) also uses to estimate establishment fixed-effects in the WES. Following Javdani, in the first step, we use the entire WES sample to estimate the establishment effects, regressing log-hourly wages on the individual control variables along with further controls for gender and each establishment. This increases the average number of employees per establishment to 17 (up to a maximum of 79).  =  + + ψ +                 (2)   In equation 2,  is a vector of indicators for each firm and ψ is a vector of establishment effects measuring establishment-specific average wages conditional on worker characteristics. The establishment effects are saved and then subtracted from log-hourly wages to create a new, transformed dependent variable,  which is the individual’s log hourly wage purged of the impact of workplace-specific characteristics. This transformed dependent variable for the sample of women 25-45 is then used to estimate the within-establishment motherhood wage gap by education and work conditions in the second step:  =  +  ∗  ∗  +     (3) The difference between the OLS and establishment fixed effects estimates reveals the degree to which any wage differential is due to mothers working in lower (if the difference is negative) or higher-waged establishments than their childless counterparts. A Hausman test is used to test the significance of this sorting effect (Pendakur and Woodcock 2010). The two-step approach removes the fixed effects and gives consistent estimates of slope parameters so long as the fixed effects are the same within a given establishment for our sample of women and the full WES sample. In other words, we assume that establishments tend to be high or low-wage for all workers within them. Lane, Salmon and Spletzer (2007) show that this is generally a reasonable assumption as establishment wage differentials are highly correlated across occupations in the United States. In our case, we also find a strong correlation between establishment wage effects calculated using all workers and those using only women between 25 and 44 (.88). Sensitivity tests repeating key analyses estimating establishment-specific effects only for the main analytical sample (women between 25 and 44) confirm no substantive difference in results.    All models use replicate weights provided by Statistics Canada to correct standard errors for the clustering of individuals within organizations, for the possibility of multiple observations on the same individual across waves, as well as other aspects of the complex multistage survey design. Limitations The above modeling strategy rests on assumptions for making causal arguments that may not hold. Notably, although we have an unusually rich set of controls, we are unable to estimate individual fixed effects that could fully account for stable unobserved differences between women who become mothers and those who do not. However, past Canadian research with longitudinal data finds motherhood penalties differ little between cross-sectional and person-fixed effects models (Zhang, 2009). Of course even fixed effects models do not entirely address the challenges in identifying the causal effect of motherhood on career outcomes. Reverse causality is an issue, as work conditions and perceived career prospects may shape fertility decisions. While some research has used instrumental variables for motherhood to address this, it is difficult to find good instruments, and the WES provides no likely candidates.  Another potential source of bias stems from the fact that we can only observe those who are currently employed. Studies of the motherhood penalty using Heckman selection models find little evidence that selection bias impacts results (e.g. Gash, 2009; Glauber, 2008, 2012), although it is admittedly difficult to find good predictors for the selection equation. In the time period of the study, maternal employment was the norm in Canada, with rates of maternal and female employment for individuals in the prime child rearing years (25-55) differing by a relatively small five percentage points in 2000 and four percentage points in 2007 (OECD, 2015). The norm of maternal employment suggests that selection bias will be relatively   unimportant in Canada. Even so, given the mutual constitution of decisions around fertility, education, and work arrangements, the analysis should be viewed as presenting a rich analytical exercise rather than a strong causal argument. RESULTS Descriptive Results Before turning to the multivariate results, Table 1 presents descriptive statistics by motherhood status and education. Workers with higher education are more likely to have flexible hours, with the difference particularly pronounced for those with postgraduate degrees. However, mothers are not more likely to have flexible hours across the board. In fact, women without children are more likely to have flexible schedules among those with a non-university post-secondary credential or Bachelor’s degree (although the difference is very small in the former case). While the least educated mothers do have a higher prevalence of flexible hours, the difference is small. It is only among women with postgraduate degrees that we see a much greater incidence of flexible schedules for mothers (53% vs 41%). The prevalence of working some of one’s regularly scheduled hours at home (substitution work at home) rises with education and is more common for mothers, with the difference once again particularly pronounced among women with postgraduate degrees. Mothers do not consistently have greater access to additional work hours spent at home that are paid by the employer. Indeed, these are more common for childless women in all educational groups save the lowest. There is little difference in the prevalence of taking additional work home without extra pay by motherhood status for women without university degrees. However, childless women are much more likely to do this among those with Bachelor degrees (21% vs 13%), while mothers are slightly more likely to follow this pattern among postgraduates (36% vs 32%).    Regression Results: Motherhood Wage Gaps by Flexible Work Arrangements and Education We now turn to our main concern – the consequences of flexible work arrangements for wage gaps by motherhood. To facilitate interpretation, model results are presented graphically as marginal effects (i.e. estimates include all components of interaction terms with appropriate adjustment for statistical significance) with points representing the percentage difference in hourly wages between mothers and women without children (derived by subtracting one from the exponentiated marginal effect and multiplying by 100). Points below zero represent a parenting pay penalty while those above 0 indicate a premium. Bars indicate 95% confidence intervals. Full models are presented in Appendix A.1.  Figure 1 shows how overall net parenting pay gaps among women vary by flexible work conditions. The panel on the left presents OLS estimates. These tell us about the aggregate effect of flexible work arrangements on motherhood penalties without consideration of the level (within or across establishments) at which they occur. The panel on the right presents estimates adjusted for establishment fixed effects, indicating the effect of flexible work arrangements within establishments.  Because we are chiefly interested in the effects of flexible work arrangements on the size of the motherhood wage gap, we focus on the vertical distance between estimates for women with and without a particular flexible arrangement. The larger the distance, the greater the effect of a flexible work arrangement on motherhood wage penalties. Comparing across the two panels reveals the extent to which the differential in the motherhood wage gap by flexibility is a product of within-establishment dynamics or sorting between establishments. If the vertical distance between points is identical (or nearly so) across panels, the effect of flexibility on the   motherhood wage gap is driven by within-establishment dynamics. The more the distance diverges, the greater the contribution of sorting between establishments (similarly, a greater difference in the location of a particular point across panels indicates a stronger sorting component in the motherhood wage differential for that particular group). Looking across panels also allows us to evaluate the applicability of various theoretical arguments. If flexibility has a more positive/less negative impact on mothers’ relative wages in the right panel (with establishment fixed effects), this implies a role for compensating differentials whereby mothers – more-so than childless women – choose employers with lower wages in exchange for flexible work. Conversely, if flexibility has a more positive/less negative impact on mothers’ wages in the left panel (OLS estimates), this suggests barriers to mothers’ employment in higher-waged firms are lowered when workers engage in flexible work.  Looking first at the aggregate (OLS) effects in the left panel of Figure 1 reveals that flexible hours and substituting some paid work hours at the office for hours at home reduce motherhood pay gaps (by 4.8 percentage points and 3.7 percentage points respectively), and bringing extra work home eliminates it entirely. Thus, the aggregate picture is clearly consistent with the facilitation argument, not stigma or compensating differentials. Full model results (first data column in Appendix Table A.1) indicate that flexible hours are associated with lower wages for women overall, but that the better outcomes for mothers fully offset this. Unpaid additional and substitution of paid work hours at home raise wages for all women, but more so for mothers.  The reduction in the motherhood wage differential for women with flexible hours is much less pronounced but still significant when accounting for establishment fixed effects (right panel). Flexible hours only reduce the motherhood wage gap within establishments by 0.7 percentage points, but reduce the wage gap associated with sorting by 4.1 percentage points.   Flexible hours may modestly improve mothers’ positions within establishments by equalizing performance, but their chief impact appears to be via reducing barriers to employment in better paying establishments. Mothers who substitute paid hours for time at the workplace fare better than their childless counterparts by a similar margin in both the aggregate and within establishment models (by 3.6 and 3.7 percentage points respectively).8 In contrast to the finding for flexible hours, this suggests that the advantage of this form of spatial flexibility is primarily a product of within-establishment dynamics. Finally, bringing unpaid additional work home reduces the motherhood wage gap both within and, especially, across establishments by 1.9 and 4.5 percentage points respectively. In jobs with intensive demands to work beyond the regular work-day, the ability to take extra work home likely reduces employers’ concerns about mothers’ ability to handle the job, reducing hiring bias against them. While both forms of spatial flexibility thus reduce motherhood penalties, their chief impact occurs via different pathways. We next turn to the education results. Figure 2 shows net parenting pay gaps by temporal flexibility (coefficients reported in Appendix Table A.2). In the aggregate (left panel), flexible hours reduce the motherhood wage gap for all groups save the least educated, with the impact growing with higher levels of education. For women with postgraduate degrees, flexible hours make a particularly dramatic difference - in their absence, mothers earn 7% less than childless women, but among those working flexible hours, mothers enjoy a 12% wage premium.9  Thus while the overall effect of flexible hours on motherhood wage gaps supports the facilitation argument, the facilitation effect is stronger for more rather than less educated mothers, arguably due to the greater productivity demands and norms of career devotion that managerial and professional women face.    Although effects on sorting still account for the bulk of the flexible hours advantage for university-educated mothers, stronger positive effects within establishments are now apparent. The right panel reveals that mothers with Bachelor’s degree who work rigid hours face a -2.6% wage penalty within establishments, but there is no significant wage gap among those who work flexible hours, and the neutral effect for postgraduate mothers without flexibility turns into an 11.3% advantage. Moreover, flexible hours eliminate the -2.1% motherhood wage gap within establishments among the least educated. However, in marked contrast to the other groups, flexible hours magnify motherhood wage gaps within establishments for women with postsecondary diplomas. This potential flexibility stigma is obscured in the OLS estimates where it is offset by the positive effect of flexible hours on these mothers’ sorting across establishments. We also see offsetting effects for the least educated. Although flexible hours eliminate significant motherhood wage gaps within establishments, they sort them into lower-wage establishments.   The educational results thus reveal not only unequal benefits across the educational spectrum, but differences in underlying dynamics. While facilitation is still the dominant pattern, the degree to which this occurs via productivity or reducing barriers to hire varies. Interestingly, while we predicted that stigma and compensating differentials would be most pronounced for highly educated women, it is only among the moderately educated that there is evidence of the former and among the least educated the latter.  Educational patterns for working at home are also nuanced. Figure 3 contrasts those who substitute work at home for regular paid working hours to those who do not work at home. While the overall effect across educational groups is positive, this obscures substantial non-linear variation by education. Substituting hours at home improves the situation for some (mothers with   a high school degree or less or a Bachelor’s degree), makes no significant difference to others (those with postsecondary education10), and dramatically increases motherhood wage gaps for the most educated.  Comparing these results to the within-establishment estimates in the second panel reveals that the greater disadvantage faced by the most educated mothers when substituting work at home is largely a product of within-establishment differentials. This type of flexibility increases the motherhood wage gap within establishments by 17 percentage points (by contrast sorting increases the wage gap by 6 percentage points). This is consistent with our assumption that it is the stigmatizing impact of reduced face-time that matters most. Only the most educated mothers experience this stigma. Although mothers with Bachelor’s degrees also appear to accept employment in lower-paying workplaces in exchange for the opportunity to substitute some work at home (consistent with the assumption that this trade-off would only be possible for more educated women), this is more than offset by the advantage they gain within establishments. Rather than leading to stigma, substituting work at home appears to facilitate high performance for these mothers, reversing the -3.6% within-establishment wage gap experienced by those who never work at home to a 9.5% premium. The least educated women also benefit within establishments when they can perform some of their regular work hours at home (the wage gap is -2.1% among those with rigid hours but small and not significant among those with flexible hours). However, reductions in hiring barriers appear more important insofar as sorting into higher paying workplaces plays a somewhat larger role, reducing the motherhood wage gap for this group by 2.9%.  Figure 4 depicts motherhood wage gaps by performing additional unpaid work at home. The OLS estimates in the first panel reveal that this practice minimizes pay penalties for mothers   with more than a high school degree (although the difference is not significant for the most educated women11), decreasing the wage gap from -4.6% to -2.4% for women with postsecondary credentials and reversing it entirely for women with Bachelor’s degrees (from -9.7% to 9.2%). However, for the least educated women, bringing extra work home unpaid is associated with a significantly larger motherhood penalty (-7.9% vs -4.9%), contradicting our expectation that this would facilitate the integration of work and family obligations in a way that would ease motherhood disadvantages.  As shown in the second panel, looking only within establishments eliminates the aggregate 2.3 percentage point work at home advantage for mothers with postsecondary credentials and reduces it by more than half for those with Bachelor degrees (from 19 to 7.7 percentage points). In these cases, reducing barriers to hire drives the positive impact of bringing work home, although for mothers’ with Bachelor degrees within-establishment dynamics also matter. In the within establishment estimates, the disadvantage of bringing extra unpaid work home for the least educated mothers vis-à-vis their childless counterparts is reduced by about a third (from 3 to 2.1 percentage points). For this group alone, there is some evidence consistent with stigma, and, to a lesser degree, compensating differentials. DISCUSSION  Financial, familial, and workplace contexts intersect to impact the strategies people can use to reconcile earning and caring, and how difficult it is to do so (Crompton & Lyonette, 2010; Debacker, 2008; Duffy & Pupo, 2011). Of particular concern in this paper was how the temporal and spatial organization of jobs – particularly with regard to flexibility – relate to motherhood pay gaps and how their effects might vary across the educational spectrum. The jobs available to women with more and less education often differ in their amenability to adjustments along   dimensions of working time and space, as well as in the possibilities for employee control over such. While a broad literature addresses how aspects of flexibility impact psychological aspects of well-being and perceptions of work-life conflict (e.g. Duxbury & Smart, 2011; Kelly, Moen, & Tranby, 2011; Schieman et al., 2009), their implication for career prospects and parenting pay gaps across the educational spectrum is unclear. Do flexible work arrangements harm mothers’ careers by stigmatizing them (particularly in the most demanding jobs), or requiring wage trade-offs (for those fortunate enough to be able to make them)?  Or by allowing care-demands to be more readily accommodated without disrupting productivity, will increased flexibility ultimately benefit mothers’ careers? Do mothers in the most demanding jobs benefit most from flexibility, or is it most advantageous for those who face rigid schedules, and more precarious work and care arrangements? Table 2 summarizes support for the different theoretical mechanisms across types of flexibility and education. Bolded X’s indicate where support is strongest (across rows). Overall, our results are clearly most consistent with the work-life facilitation argument. Out of the twelve combinations of education and flexibility, there are only four instances where patterns support a compensating differentials scenario: the least educated mothers are sorted into lower paying workplaces more often than their childless counterparts when they have flexible hours and when they bring additional unpaid work home, as are the university educated when they substitute paid work at home for time in the workplace. In only one case, however, does this negative sorting effect drive the overall pattern underlying differences in wage gaps by work arrangement. Flexibility stigma only reduces mothers’ relative wages in three instances: women with postsecondary education who work flexible hours, women with postgraduate degrees who substitute work at home, and women with a high school education or less who bring additional   unpaid work home, but it is the dominant explanation in only the two last cases. In all other instances, flexibility improved mothers’ wages overall relative to childless women. Arguments about flexibility equalizing performance within workplaces and reducing barriers to hire each received support six times, with the explanation is dominant once with respect to equalizing performance within workplaces and all six times with respect to reducing barriers to hire.  While the overall pattern of results thus suggests that flexible work arrangements typically reduce motherhood wage penalties, and that they most strongly do so by reducing barriers to hire, there is also substantial variation across both types of flexibility and education. Support for the facilitation perspective is clearest with respect to flexible work hours. Importantly, we find that flexible hours play the largest role in mitigating motherhood wage penalties among the most educated. This suggests that it is the opportunity to better accommodate a more demanding paid work load, characteristic of managerial and professional jobs, that matters most. For women with lower levels of education, temporal flexibility still offsets motherhood wage penalties, but to a lesser extent.  Overall temporal flexibility advantage is a product of both within12- and across-establishment wage differentials. This underscores the importance of considering the impact of flexible work both on women’s experiences on the job and their sorting across more and less desirable firms. Within establishments, having flexible work hours likely reduces differences in mothers and childless women’s ability to manage work demands in light of other obligations. In addition, our results suggest that flexible hours can reduce barriers to mothers’ employment in higher-wage firms. Mothers with more than a high school education are less likely to be segregated in lower paying firms when they have flexible jobs, a dynamic that is the biggest contributor to the flexibility advantage. This highlights the importance of examining the impact   of flexibility not simply in terms of workers’ individual-level circumstances, but also in how it intersects with employers’ perceptions and sorting strategies.  Spatial flexibility – measured here as the ability to work from home – is commonly perceived as smoothing the work-family interface. In the aggregate, we find that it also reduces mothers’ wage disadvantage, and that both worker-oriented flexibility (the substitution-model of working at home) and employer-oriented flexibility (taking extra work home) are similarly useful. The fact that both types of working at home have similar aggregate impacts is interesting in light of their differing relationship to ideal worker norms, with the substitution model challenging these norms and the taking extra work home approach reinforcing them.  At the same time, our analysis uncovered considerable but not entirely uniform impacts of working from home across the educational spectrum. The aggregate picture obscures much variation by education. While most women fare better within establishments when able to substitute at least some hours working at home, the most educated women, for whom ideal worker and work devotion norms are most pronounced, fare much worse. This is particularly interesting in light of the fact that childless women in this educational group are equally likely to take advantage of this type of flexibility, which should arguably minimize its stigmatizing association with work-family accommodation. The fact that we find penalties for this type of working at home for postgraduate mothers suggests that face time remains a key indicator – whether real or perceived – of productivity in high status jobs. As mothers with lower and middle levels of education do not experience the work devotion schema as intensely, penalties for minimizing face time by working at home do not come on as strong, if at all, for lower and middle tiered educational groups.    Bringing extra unpaid work home was neutral for the most educated women, likely because of its normativity for this group. This underscores the fact that “working from home” can refer to quite different types of arrangements, with different implications for inequalities. We expected that any negative effects of flexibility would be most pronounced for the most educated women, but it is actually only among those with a high school degree or less where bringing extra unpaid work home is associated with larger motherhood wage gaps, with this operating both within and across establishments. Whether taking on such extra work particularly reflects heightened vulnerability among the poorest compensated and least educated mothers is an important question for future work. CONCLUSION Overall, our findings that both temporal and spatial flexibility most often facilitate the combination of paid work and caregiving are consistent with Yu and Kuo’s (2017) study that finds smaller motherhood pay penalties for women in occupations that have attributes thought to reduce work-life strain, such as team work and autonomy. In this way, our results contribute to a growing literature that points to the promise of flexible work arrangements in offsetting motherhood penalties and improving work-life facilitation for families (Kelly et al. 2011; Moen et al. 2016). That said, our findings remain at odds with much of the literature on the durability of the flexibility stigma for mothers, and even the Yu and Kuo (2017) study did not find schedule regularity to significantly offset motherhood wage penalties. However, much previous research relies on qualitative or experimental designs (i.e., Blair-Loy 2003; Cech and Blair-Loy 2014; Cuddy et al. 2004; Cohen and Single 2001; Dodson 2013; Munsch 2016; Stone and Hernandez 2013), which tend to focus on mothers in specific jobs, workplaces, or industries, and thus, may not capture the average impact of flexibility policies on motherhood wage gaps within and across   establishments, as we do here. In addition, our data are from Canada, where the federal government mandates a paid maternity and parental leave program for new parents, and parenthood is a legally protected status under antidiscrimination laws. This more family-friendly national policy landscape may buffer Canadian mothers who make use of flexible work arrangements from some of the stigma experienced by their U.S. counterparts. Future work may consider how external policy conditions – such as local and national policies – mediate the impact of flexible work arrangements on motherhood disadvantages.  Future work could also usefully expand the scope of flexible arrangements considered. As Williams, Blair-Loy, and Berdahl (2013) and Jacobs and Padavic (2014) discuss, other forms of flexibility and schedule control, such as the opportunity to work predictable shift patterns, leave for brief periods in the middle of a shift, and receive schedules well in advance, may be particularly important for women in lower and middle tier jobs in areas such as retail, hospitality, and health care. Considering interrelationships among work conditions may also shed additional light on some of the patterns uncovered here. For example, it is possible that being able to control start and stop times goes along with other, employer-centred, forms of temporal flexibility (such as “just in time” scheduling or “gig economy” work) that might offset its positive impact for less advantaged women. Finally, it is important to acknowledge that mothers may be reluctant to take advantage of flexible work arrangements, even if formally available, where workplace norms suggest that to do so would be stigmatizing. Because our measures of flexibility focus on use rather than availability, they may disproportionately capture effects in contexts where working flexibly is normatively acceptable and among workers for whom it has the most value in reconciling work   and family demand. Thus, future research should examine how access to and use of flexible work arrangements might differently condition the effect of motherhood status on pay.  In sum, our work takes up the call by Williams et al. (2013) to unravel the contours of the flexibility stigma or boost across social positioning, class, and workplace context. While our results suggest that flexible work arrangements can facilitate the combination of work and family life for mothers, we find that the triggers and consequences of specific flexible work arrangements vary both by type of policy and by mothers’ educational level. As we demonstrate, postgraduate women benefit more than any other educational group from temporal flexibility but draw few gains and even incur penalties when working at home, as this violates the work devotion schema and long hours of face-time expected in managerial and professional jobs. Women in other educational groups are typically aided by temporal flexibility, though to a lesser extent than their highly educated counterparts, but can usually take advantage of work at home strategies without penalty. These findings highlight the importance of remaining attentive to intersections in women’s positioning, and not presuming that particular “family friendly” work arrangements will have consistent implications for all women. In keeping with identifying the class-based and contextual nature of flexibility effects, future research should explore how alternative forms of flexibility and schedule control, such as predictable shifts, advance schedule notice, and progressive discipline systems for tardiness and absenteeism, may be especially helpful for lower-tier workers. For those in managerial and professional positions, dislodging the cultural schemas of total career devotion and the ideal worker appears critical in ensuring that flexible work arrangements facilitate work and family integration. We see our work as a step in identifying the conditions under which flexibility can minimize or exacerbate pay penalties for   mothers, and the findings presented here urge future attention to the intersection of flexibility policies and workers’ social class positioning.      Funding Research was supported by funding from the Social Sciences and Humanities Research Council of Canada (SSHRC, grant No. 430-2014-00140).  Acknowledgements We thank Christina Treleaven, Natasha Stecy-Hildebrandt and Hazel Hollingdale for initial research assistance. The analyses presented in this paper were conducted at the Simon Fraser University (SFU) branch of the BCIRDC, which is part of the Canadian Research Data Centre Network (CRDCN). The services and activities provided by the BCIRDC are made possible by the financial or in-kind support of SSHRC, the CIHR, the CFI, Statistics Canada, and SFU. The views expressed in this paper do not necessarily represent the CRDCN or its partners.     Notes 1 Flexibility can also exacerbate work-family strain by increasing employer expectations for worker availability and productivity (Bishop, 1997; Duxbury, Higgins, Smart, & Stevenson, 2013; Schieman & Glavin, 2016; Sweet, Besen, Pitt-Catsouphes, & McNamara, 2014). 2 Indeed, research shows that workplaces vary substantially in their average wage rates net of the characteristics of their workers (Abowd, Kramarz, and Margolis 1999; Lane, Salmon, and Spletzer 2007), and motivating greater commitment and effort on the part of their workers is often an important reason to pay above market wages (Shapiro & Stiglitz, 1984). 3 Employment standards legislation in Canada provides no rights for parents around accessing reduced-hour schedules, flexible hours or performing job duties at home. Mothers are reliant upon what their employers are willing to provide and what is generally available in the labor market for workers with their skills and qualifications. 4 Different establishments of the same employer may be selected and would be treated as separate employers. 5 Even with our restricted sample, we may include some “empty-nesters” who had children relatively young in the childless group. This likely reduces aggregate estimates of motherhood penalties. However, this should not unduly bias estimates of interest insofar as the key issue is the relationship between flexibility and motherhood penalties and theoretical arguments are predicated on dynamics only relevant for those with resident children. Sensitivity analyses restricting models to women less than 39 result in substantively similar estimates. 6 In the American context, long hours work has been defined as more than 50 hours per week (Cha & Weeden, 2014). However, average work hours are shorter in Canada, and working 50 or   more hours per week was rare in our sample. An examination of work hour distributions suggested 45 hours as a more relevant cut-point for Canadian women. 7 Other reasons are available in the WES but were cited too infrequently to be meaningfully included. 8 It reverses to a 1.6% premium for mothers who substitute some work at home. 9 As a sensitivity test, we re-ran all analyses restricting the focus to mothers with young children (6 and under). This eliminated the strong educational differential among women with more than a high school degree, meaning that the extra flexibility advantage for the most highly educated mothers emerges over time (largely via within-firm advantages that are not apparent for mothers of younger children). 10 Restricting the sample to mothers of young children revealed a positive impact of this form of spatial flexibility. This suggests that while substituting work hours at home facilitates combining work and family for mothers of younger children (whose care demands are presumably most pressing), mothers of older children who reduce their face time may be stigmatized (wage effects occurred within firms). 11 When considering only mothers of young children, bringing extra work home lowers mothers’ wages vis-à-vis childless women. This implies offsetting effects for mothers of younger and older children for this group – effects are negative when family demands are greatest, but positive when children are more independent. 12 The one exception is women with postsecondary degrees for whom flexible hours magnify mothers’ disadvantage within establishments.    References Abowd, J. M., & Kramarz, F. (1999). The analysis of labor markets using matched employer-employee data. Pp. 2629-2710 in Ashenfelter, O.C., & Card, D. (Eds.), Handbook of Labor Economics. Amsterdam: North-Holland.  Aisenbrey, S., Evertsson, M., & Grunow, D. (2009). Is There a Career Penalty for Mothers' Time Out? A Comparison of Germany, Sweden and the United States. Social Forces, 88(2), 573-605.  Anderson, D. J., Binder, M., & Krause, K. (2002). The Motherhood Wage Penalty: Which Mothers Pay It and Why? American Economic Review, 92(2), 354-358.  Anderson, D. J., Binder, M., & Krause, K. (2003). The Motherhood Wage Penalty Revisited: Experience, Heterogeneity, Work Effort, and Work-Schedule Flexibility. Industrial & Labor Relations Review, 56(2), 273.  Baum, C. L. (2002). The Efffect of Work Interruptions on Women’s Wages. Labour, 16(1), 1-37.  Becker, G. S. (1993). A Treatise on the Family (Rev. Ed.). Cambridge, MA: Harvard University Press. Bishop, J. H. (1997). What We Know About Employer-Provided Training: A Review of Literature CAHRS Working Paper Series (Vol. 96-09). Ithaca: Cornell University, School of Industrial and Labor Relations, Center for Advances Human Resource Studies. Blair-Loy, M. (2003). Competing Devotions: Career and Family among Women Executives. Cambridge: Harvard University Press. Blair-Loy, M., & Wharton, A. S. (2002). Employees' Use of Work-Family Policies and the Workplace Social Context. Social Forces, 80(3), 813-845.    Bornstein, S. (2013). The Legal and Policy Implications of the “Flexibility Stigma”. Journal of Social Issues, 69(2), 389-405.  Boushey, H. (2008). Family Friendly Policies: Helping Mothers Make Ends Meet. Review of Social Economy, 66(1), 51-70.  Budig, M. J., & England, P. (2001). The Wage Penalty for Motherhood. American Sociological Review, 66(2), 204-225.  Cahusac, E., & Kanji, S. (2013). Giving Up: How Gendered Organizational Cultures Push Mothers Out. Gender, Work & Organization, 21(1), 57- 70.  Canay, I. A. (2011). A simple approach to quantile regression for panel data. The Econometrics Journal, 14(3), 368-386. Cech, E. A., & Blair-Loy, M. (2014). Consequences of Flexibility Stigma among Academic Scientists and Engineers. Work and Occupations, 41(1), 86-110.  Cha, Y., & Weeden, K. A. (2014). Overwork and the Slow Convergence in the Gender Gap in Wages. American Sociological Review, 79(3), 457-484.  Clawson, D., & Gerstel, N. (2014). Unequal Time: Gender, Class, and Family in Employment Schedules. New York: Russell Sage Foundation. Cohen, J. R., & Single, L. E. (2001). An examination of the perceived impact of flexible work arrangements on professional opportunities in public accounting. Journal of Business Ethics, 32(4), 317-328. Correll, S. J., Benard, S., & Paik, I. (2007). Getting a Job: Is There a Motherhood Penalty? . American Journal of Sociology, 112(5), 1297-1339.  Crompton, R., & Lyonette, C. (2010). Family, Class and Gender “Strategies” in Mothers’ Employment and Childcare. In J. Scott, R. Crompton & C. Lyonette (Eds.), Gender   Inequalities in the 21st Century: New Barriers and Continuing Constraints. Cheltenham: Edward Elgar (pp. 174-192). Northampton: Edward Elgar. Cuddy, A. J., Fiske, S. T., & Glick, P. (2004). When Professionals Become Mothers, Warmth Doesn't Cut the Ice. Journal of Social Issues, 60(4), 701-718.  Damaske, S., Ecklund, E. H., Lincoln, A. E., & White, V. J. (2014). Male Scientists’ Competing Devotions to Work and Family: Changing Norms in a Male-Dominated Profession. Work and Occupations, 41(4), 477-507.  Debacker, M. (2008). Care Strategies among High- and Low-Skilled Mothers: A World of Difference? Work Employment Society, 22(3), 527-545.  Demaiter, E. I., & Adams, T. L. (2008). “I Really Didn’t Have Any Problems with the Male-Female Thing until…”: Successful Women’s Experiences in It Organizations. Canadian Journal of Sociology, 34(1), 31-54.  Dodson, L. (2013). Stereotyping Low-Wage Mothers Who Have Work and Family Conflicts. Journal of Social Issues, 69(2), 257-278.  Duffy, A., & Pupo, N. (2011). Employment in the New Economy and the Impact on Canadian Families. In C. Krull & J. Sempruch (Eds.), A Life in Balance?: Reopening the Family-Work Debate. Vancouver: UBC Press. Duxbury, L., Higgins, C., Smart, R., & Stevenson, M. (2013). Mobile Technology and Boundary Permeability. British Journal of Management.  Duxbury, L., & Smart, R. (2011). The “Myth of Separate Worlds”: An Exploration of How Mobile Technology Has Redefined Work-Life Balance. In S. Kaiser, M. J. Ringlstetter, D. R. Eikhof & M. Pina e Cunha (Eds.), Creating Balance? (pp. 269-284). Berlin: Springer    Epstein, C. F., Seron, C., Oglensky, B., & Saute, R. (2014). The Part-Time Paradox: Time Norms, Professional Life, Family and Gender. New York: Routledge. Estes, S., & Glass, J. L. (1996). Job Changes Following Childbirth: Are Women Trading Compensation for Family-Responsive Work Conditions? Work and Occupations, 23(4), 405-436.  Fried, M. (1998). Taking Time: Parental Leave Policy and Corporate Culture. Philadelphia: Temple University Press. Fuller, S. (2017) Segregation across Workplaces and the Motherhood Wage Gap: Why do Mothers Work in Low-Wage Establishments? Social Forces https://doi.org/10.1093/sf/sox087 Gash, V. (2009). Sacrificing Their Careers for Their Families? An Analysis of the Penalty to Motherhood in Europe. Social Indicators Research, 93(3), 569-586.  Gerstel, N., & Clawson, D. (2014). Class Advantage and the Gender Divide: Flexibility on the Job and at Home. American Journal of Sociology, 120(2), 395-431.  Glass, J. (2004). Blessing or Curse? Work-Family Policies and Mother’s Wage Growth over Time. Work and Occupations, 31(3), 367-394.  Glass, J., & Noonan, M. (2016). Telecommuting and Earnings Trajectories among American Women and Men 1989–2008. Social Forces, 95(1), 217-250.  Glauber, R. (2008). Race and Gender in Families and at Work the Fatherhood Wage Premium. Gender & Society, 22(1), 8-30.  Glauber, R. (2012). Women’s Work and Working Conditions: Are Mothers Compensated for Lost Wages? Work and Occupations, 39(2), 115-138.    Heywood, J. S., Siebert, W. S., & Wei, X. (2007). The Implicit Wage Costs of Family Friendly Work Practices. Oxford Economic Papers, 59(2), 275-300.  Jacobs, A. W. and Padavic, I. (2015), Hours, Scheduling and Flexibility for Women in the US Low-Wage Labour Force. Gender, Work and Organization, 22(1), 67–86 Javdani, M. (2012). Glass Ceilings or Glass Doors? The Role of Firms in Male-Female Wage Disparities. Canadian Journal of Economics/Revue canadienne d'économique, 48(2), 529-560.  Johnson, N. B., & Provan, K. G. (1996). The Relationship between Work/Family Benefits and Earnings: A Test of Competing Predictions. The Journal of Socio-Economics, 24(4), 571-584.  Kelly, E. L., Moen, P., & Tranby, E. (2011). Changing Workplaces to Reduce Work-Family Conflict. American Sociological Review, 76(2), 265-290.  Kmec, J. A. (2011). Are Motherhood Penalties and Fatherhood Bonuses Warranted? Comparing Pro-Work Behaviors and Conditions of Mothers, Fathers, and Non-Parents. Social Science Research, 40(2), 444-459.  Kmec, J.A. O’Connor, L. and Schieman, S. (2014). Not Ideal: The Association Between Working Anything but Full Time and Unfair Treatment. Work and Occupations, 41(1), 63-85. Kossek, E and Lautsch, B. (2018). Work-Life Flexibility for Whom? Occupational Status and Work-Life Inequality in Upper, Middle, and Lower-level Jobs. Academy of Management Annals, 12(1), 5-36 Lane, J. I., Salmon, L. A., & Spletzer, J. R. (2007). Establishment wage differentials. Monthly Labor Review, 130, 3-17.   Lundberg, S., & Rose, E. (2000). Parenthood and the Earnings of Married Men and Women. Labour Economics, 7(6), 689-710.  Lundberg, S., & Rose, E. (2002). The Effects of Sons and Daughters on Men's Labor Supply and Wages. Review of Economics and Statistics, 84(2), 251-268.  Marshall, K. (2009). The Family Work Week. Perspectives on Labour and Income, 21(2), 21-29.  McCrate, E. (2005). Flexible Hours, Workplace Authority, and Compensating Wage Differentials in the U.S. Feminist Economics, 11(1), 11-39.  Munsch, C. (2016). Flexible Work, Flexible Penalties: The Effect of Gender, Childcare, and Type of Request on the Flexibility Bias. Social Forces 2016; 94 (4), 1567-1591 Munsch, C., Ridgeway, C., & Williams, J. (2014). Pluralistic Ignorance and the Flexibility Bias: Understanding and Mitigating Flextime and Flexplace Bias at Work. Work and Occupations, 41(1), 40-62.  OECD. (2015). Oecd Family Database (Www.Oecd.Org/Social/Family/Database.Htm)  Pendakur, K., & Woodcock, S. (2010). Glass Ceilings or Glass Doors? Wage Disparity within and between Firms. Journal of Business & Economic Statistics, 28(1), 181-189.  Petersen, T., Penner, A. M., & Høgsnes, G. (2011). The Male Marital Wage Premium: Sorting Vs. Differential Pay. Industrial & Labor Relations Review, 64(2), 283-304.  Petersen, T., Penner, A. M., & Høgsnes, G. (2014). From Motherhood Penalties to Husband Premia: The New Challenge for Gender Equality and Family Policy, Lessons from Norway. American Journal of Sociology, 119(5), 1434-1472.  Phipps, S., Burton, P., & Lethbridge, L. (2001). In and out of the Labour Market: Long-Term Income Consequences of Child-Related Interruptions to Women's Paid Work. Canadian Journal of Economics/Revue Canadienne d'Economique, 34(2), 411-429.    Pocock, B., & Charlesworth, S. (2017). Multilevel Work–Family Interventions. Work and Occupations, 44(1), 23-46.  Pulkingham, J., & Fuller, S. (2012). From Parent to Patient: The Medicalization of Lone Motherhood through Welfare Reform. Social Politics: International Studies in Gender, State & Society, 19(2), 243-268.  Ridgeway, C. L., & Correll, S. J. (2004). Motherhood as a Status Characteristic. Journal of Social Issues, 60(4), 683-700.  Shapiro, C., & Stiglitz, J. E. (1984). Equilibrium Unemployment as a Worker Discipline Device. The American Economic Review, 74(3), 433-444. Schieman, S., & Glavin, P. (2016). The Pressure-Status Nexus and Blurred Work–Family Boundaries. Work and Occupations, 43(1), 3-37.  Schieman, S., Milkie, M. A., & Glavin, P. (2009). When Work Interferes with Life: Work-Nonwork Interference and the Influence of Work-Related Demands and Resources. American Sociological Review, 74(6), 966-988.  Shapiro, C., & Stiglitz, J. E. (1984). Equilibrium Unemployment as a Worker Discipline Device. The American Economic Review, 74(3), 433-444.  Stone, P., & Hernandez, L. A. (2013). The All-or-Nothing Workplace: Flexibility Stigma and “Opting out” among Professional-Managerial Women. Journal of Social Issues, 69(2), 235-256.  Strazdins, L., Clements, M. S., Korda, R. J., Broom, D. H., & D’Souza, R. M. (2006). Unsociable Work? Nonstandard Work Schedules, Family Relationships, and Children’s Well-Being. Journal of Marriage and Family, 68(2), 394-410.    Sweet, S., Besen, E., Pitt-Catsouphes, M., & McNamara, T. K. (2014). Do Options for Job Flexibility Diminish in Times of Economic Uncertainty? Work, Employment & Society, 28(6), 882-903.  Sweet, S., Pitt-Catsouphes, M., Besen, E., & Golden, L. (2014). Explaining Organizational Variation in Flexible Work Arrangements: Why the Pattern and Scale of Availability Matter. Community, Work & Family, 17(2), 115-141.  Wharton, A. S., Chivers, S., & Blair-Loy, M. (2008). Use of Formal and Informal Work–Family Policies on the Digital Assembly Line. Work and Occupations, 35(3).  Williams, J. (2010). Reshaping the Work-Family Debate. Cambridge: Harvard University Press. Williams, J., Blair-Loy, M., & Berdahl, J. (2013). Cultural Schemas, Social Class, and the Flexibility Stigma. Journal of Social Issues, 69(2), 209-234.  Williams, J. (2006). One Sick Child Away from Being Fired: When 'Opting out' Is Not an Option. Retrieved from  doi: http://dx.doi.org/10.2139/ssrn.2126303 Winder, K. L. (2009). Flexible Scheduling and the Gender Wage Gap. The BE Journal of Economic Analysis & Policy, 9(1), 1935-1682.  Yu, W. H., & Kuo, J. C. L. (2017). The Motherhood Wage Penalty by Work Conditions: How Do Occupational Characteristics Hinder or Empower Mothers?.American Sociological Review, 82(4), 744-769. Zhang, X. (2009). Earnings of Women with and without Children. Perspectives on Labour and Income, 10(3), 5-13.        Author Bios  Sylvia Fuller is Associate Professor of Sociology at the University of British Columbia. Her research centers on understanding how labour market inequalities develop and erode and the implications of changing employment relations and social policy for people’s economic security and mobility.   Elizabeth Hirsh is Associate Professor of Sociology and Canada Research Chair in Inequality and Law at the University of British Columbia.  Her research focuses on gender, race, and ethnic inequality at work, organizational policy, and the impact of anti-discrimination law on workplace inequality.      Figure 1 Marginal Estimates of Aggregate and Within Establishment Motherhood Wage Gap by Work Arrangements  -10-505AggregateDifference in Hourly Wage by Motherhood (%)-10-505Within EstablishmentData: Workplace and Employee SurveyEstimates are net of controlsRigid Hours No Work from HomeFlexible Hours Paid SubstituteUnpaid Additional  Figure 2 Marginal Estimates of Aggregate and Within Establishment Motherhood Wage Gap by Flexible Hours and Education -1001020<=High school Postsec Bachelor PostgradAggregateDifference in Hourly Wage by Motherhood (%)-1001020<=High school Postsec Bachelor PostgradWithin EstablishmentData: Workplace and Employee SurveyEstimates are net of controlsRigid Hours Flexible Hours  Figure 3 Marginal Estimates of Aggregate and Within Establishment Motherhood Wage Gap by Substituting Paid Work Hours at Home and Education  -30-20-1001020<=High school Postsec Bachelor PostgradAggregateDifference in Hourly Wage by Motherhood (%)-30-20-1001020<=High school Postsec Bachelor PostgradWithin EstablishmentData: Workplace and Employee SurveyEstimates are net of controlsNo Work from Home Paid Substitute  Figure 4 Marginal Estimates of Aggregate and Within Establishment Motherhood Wage Gap by Unpaid Work at Home and Education  -10-50510<=High school Postsec Bachelor PostgradAggregateDifference in Hourly Wages by Motherhood (%)-10-50510<=High school Postsec Bachelor PostgradWithin EstablishmentData: Workplace and Employee SurveyEstimates are net of controlsNo Work from Home Unpaid Additional  Table 1 Means (continuous variables) and Proportions (categorical variables) by Motherhood and Education      High school or less   Postsecondary   Bachelors   Postgraduate   Childless Mother   Childless Mother   Childless Mother   Childless Mother Flexible hours 0.331 0.345  0.340 0.325  0.379 0.345  0.414 0.525 Work from home (none) 0.797 0.747  0.798 0.771  0.704 0.773  0.460 0.390 Unpaid, part of regular hours 0.029 0.056  0.031 0.056  0.055 0.065  0.129 0.219 Unpaid, addition to regular hours 0.138 0.136  0.139 0.146  0.211 0.135  0.322 0.356 Paid, addition to regular hours 0.036 0.061  0.032 0.026  0.030 0.028  0.089 0.035 Survey year (1999)            2001 0.229 0.283  0.219 0.256  0.230 0.263  0.244 0.239 2003 0.256 0.262  0.284 0.264  0.256 0.247  0.257 0.228 2005 0.216 0.196  0.256 0.250  0.273 0.238  0.288 0.299 Age 33.756 36.399  33.198 36.428  33.411 37.194  33.461 38.113 Experience 12.028 13.867  11.452 13.051  10.666 13.036  8.479 12.630 Seniority 5.858 6.412  5.587 6.802  5.645 7.016  4.948 6.304 Race/Immigration (white Cdn-born) 0.836 0.800  0.812 0.808  0.762 0.766  0.625 0.666 White immigrant 0.057 0.065  0.076 0.070  0.080 0.087  0.114 0.163 Visible minority Canadian born 0.058 0.051  0.046 0.035  0.062 0.035  0.066 0.008 Visible minority immigrant 0.049 0.084  0.066 0.088  0.097 0.111  0.195 0.163 Spouse 0.493 0.761  0.565 0.828  0.531 0.851  0.447 0.893 Other Income 2.696 3.321  2.784 3.668  2.814 3.855  2.949 5.563 Union 0.190 0.182  0.207 0.270  0.221 0.243  0.321 0.414 Weekly hours            Part-time 0.215 0.312  0.191 0.320  0.176 0.317  0.101 0.221 Full time 0.583 0.521  0.570 0.508  0.552 0.481  0.450 0.413 Long hours 0.202 0.168  0.239 0.172  0.272 0.202  0.449 0.366 Reason unable to work preferred extra hours           Childcare unavailable 0.000 0.006  0.000 0.013  0.000 0.005  X X Personal/family responsibilities 0.008 0.017  0.003 0.018  0.006 0.016  X X Going to school 0.008 0.002  0.004 0.003  0.010 X  X X Employer doesn't offer 0.209 0.130  0.183 0.135  0.173 0.138  0.106 0.093 Pay insufficient 0.033 0.017  0.014 0.009  0.028 0.005  0.032  Why prefer less hours            Family responsibilities 0.011 0.050  0.020 0.056  0.020 0.070  0.023 0.155 Stress 0.016 0.011  0.025 0.012  0.013 0.014  0.045 0.025 Health 0.005 0.001  0.007 0.004  0.005 0.002  0.017  Leisure 0.040 0.015  0.031 0.017  0.042 0.021  0.063 0.011   Employer-paid training 0.054 0.031  0.047 0.049  0.040 0.031  0.114 0.089 Unpaid training 0.091 0.081  0.123 0.101  0.115 0.059  0.212 0.210 Disabled 0.057 0.061  0.052 0.055  0.063 0.043  0.059 0.047 Months unemployed 1.315 2.016  1.312 1.341  1.203 1.583  0.994 1.173 N (rounded to nearest 5) 1495 2340   3195 4790   3640 4635   445 345 X Small cell size makes estimate unreliable             Table 2: Support for Theoretical Predictions   Flexibility reduces motherhood wage penalty  Flexibility increases motherhood wage penalty  Work life facilitation       Equalizing performance (within establishment) Barriers to hire (sorting)  Compensating differentials (sorting) Stigma  (within establishment) Flexible hours      <= High school X   X  Postsecondary  X   X Bachelor's  X X    Postgraduate   degree X X          Substituting work hours at home     <= High school X X    Postsecondary      Bachelor's  X   X  Postgraduate degree    X X       Unpaid additional work at home     <= High school    X X Postsecondary  X    Bachelor's  X X    Postgraduate degree              Table A.1 OLS and Establishment-fixed effects Estimates of Motherhood Wage Gaps by Work Arrangements                          Aggregate Within-establishment Mother -0.085*** -0.023***                          (0.002) (0.002) Flexible hours -0.033*** 0.015***                          (0.002) (0.002) Mother * Flexible hours 0.051*** 0.007**                          (0.003) (0.003) Work at home (none)   Paid substitution 0.229*** 0.114***                          (0.006) (0.004) Paid additional 0.256*** 0.152***                          (0.007) (0.005) Unpaid additional 0.209*** 0.164***                          (0.003) (0.002) Mother*Work at home (none)   Mother * Unpaid substitution 0.039*** 0.037***                          (0.009) (0.007) Mother* Paid additional -0.018* 0.004                          (0.009) (0.006) Mother* Unpaid additional 0.062*** 0.019***                          (0.004) (0.003) R-squared                0.300 0.244 Standard errors in parentheses   * p<0.05,  ** p<0.01,  ***p<0.001   Models control for age and its square, tenure and its square, experience and its square (centred), interactions between education and experience terms, Race/immigration, spouse, other household income, union, hours worked, employer-paid training, unpaid training, disability, months unemployed in past 5 years,  prefer more or less work hours and why.     Table A.2 OLS and Establishment-fixed effects Estimates of Motherhood Wage Gaps by Work Arrangements and Education                          Aggregate Within-establishment Mother -0.044*** -0.024***                          (0.005) (0.004) Education (High school or less)   Postsecondary (non-university) 0.230*** 0.147***                          (0.006) (0.005) Bachelor's 0.103*** 0.077***                          (0.006) (0.005) Postgraduate 0.306*** 0.166***                          (0.012) (0.012) Mother*Education (High school or less)   *Postsecondary (non-university) -0.020*** 0.021***                          (0.006) (0.004) *Bachelor's -0.086*** -0.023***  (0.006) (0.005) *Postgraduate -0.023* 0.021*                          (0.011) (0.008) Flexible hours -0.020*** -0.000                          (0.005) (0.004) Mother*Flexible hours -0.016* 0.014**                          (0.007) (0.005) Flexible hours*Education ( High school or less)   *Postsecondary (non-university) -0.009 0.033***                          (0.006) (0.004) *Bachelor's -0.008 0.016**  (0.007) (0.005) *Postgraduate -0.068*** -0.034***                          (0.012) (0.010) Mother*Flexible hours*Education (High school or less)   *Postsecondary (non-university) 0.063*** -0.038***                          (0.009) (0.006) *Bachelor's 0.093*** 0.013*  (0.008) (0.006) *Postgraduate 0.232*** 0.093***                          (0.017) (0.013) Work from home (none)   Paid substitution 0.154*** 0.101***                          (0.013) (0.005) Paid additional -0.016* 0.026**                          (0.007) (0.008) Unpaid additional 0.221*** 0.177***                          (0.007) (0.005) Mother*Work from home (none)   *Paid substitution 0.172*** 0.055***                          (0.018) (0.013) *Paid additional 0.201*** 0.115***                          (0.014) (0.009) *Unpaid additional -0.032*** -0.021**                          (0.009) (0.007) Postsecondary*Work from home (none)   *Paid substitution 0.007 -0.007                          (0.017) (0.010) *Paid additional 0.266*** 0.124***                            (0.011) (0.012) *Unpaid additional -0.104*** -0.078***                          (0.009) (0.007) Bachelor’s*Work from home (none)   *Paid substitution 0.104*** 0.011                          (0.017) (0.007) *Paid additional 0.333*** 0.129***                          (0.010) (0.011) *Unpaid additional 0.037*** 0.024***                          (0.008) (0.006) Postgraduate*Work from home (none)   *Paid substitution 0.194*** 0.115***                          (0.020) (0.016) *Paid additional 0.479*** 0.305***                          (0.030) (0.018) *Unpaid additional 0.035** 0.002  (0.011) (0.011) Mother*Postsecondary*Work from home (none)   *Paid substitution -0.170*** -0.073***                          (0.023) (0.018) *Paid additional -0.271*** -0.165***                          (0.019) (0.014) *Unpaid additional 0.055*** 0.023**                          (0.011) (0.008) Mother*Bachelor’s*Work from home (none)   *Paid substitution -0.063** 0.072***                          (0.022) (0.014) *Paid additional -0.216*** -0.079***                          (0.018) (0.012) *Unpaid additional 0.222*** 0.098***                          (0.012) (0.008) Mother*Postgraduate*Work from home (none)   *Paid substitution -0.436*** -0.239***                          (0.028) (0.022) *Paid additional -0.053 0.047                          (0.047) (0.035) *Unpaid additional 0.050** 0.064***                          (0.018) (0.018) Constant                 2.223*** 1.974***                          (0.005) (0.004) R-squared                0.311 0.253 Standard errors in parentheses   * p<0.05,  ** p<0.01,  ***p<0.001   Models control for age and its square, tenure and its square, experience and its square (centred), interactions between education and experience terms, Race/immigration, spouse, other household income, union, hours worked, employer-paid training, unpaid training, disability, months unemployed in past 5 years,  prefer more or less work hours and why.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0366253/manifest

Comment

Related Items