Ask and Ye Shall Receive? The Dynamics of Employer-Provided Flexible Work Options and the Need for Public Policy
Correspondence concerning this article should be addressed to Victoria L. Brescoll, Assistant Professor of Organizational Behavior, Yale School of Management, 135 Prospect Street, New Haven, CT 06511 [e-mail: firstname.lastname@example.org].
This article addresses two fundamental questions about flexible scheduling: Do managers use ascriptive information in deciding which requests for flexible work scheduling to grant among employees? And, do employees comprehend this managerial bias in deciding whether to ask for flexible work arrangements? Study 1 found that managers were most likely to grant flextime to high-status men seeking flexible schedules in order to advance their careers. In contrast, flexible scheduling requests from women were unlikely to be granted irrespective of their job status or reason. In Study 2, we found that employees were unaware of these managerial biases: women assigned high-status jobs and requests for career advancement reasons were the most likely to think their requests would be granted, while men in the same scenarios were least likely to believe this. Organizational and policy implications are discussed.
In contrast to most developed nations, the United States stands nearly alone in its reliance on the free market to set work hours and work schedules among employees in both the public and private sectors. While the United Kingdom and European Union (EU) have enacted legislation granting employees the “right to ask” for a variety of work schedule accommodations (Hegewisch, 2005) and most EU countries have either maximum work hours legislation and/or provisions for shorter work hours for parents of young children (Gornick & Meyers, 2003), the United States continues to rely predominantly on a voluntary approach. This approach is typically characterized by individual employees asking their supervisors or employers for these types of schedule accommodations and managers deciding on a case-by-case basis whether to approve or deny their employees’ requests. Most work organizations, however, either have no flexible scheduling policies or craft policies that specifically protect managerial discretion in decisions regarding who will be allowed flexible work schedules (Kelly & Kalev, 2006).
Within the relatively weak regulatory framework of the Fair Labor Standards Act of 1938, negotiations over the hours or times of work are left to the individual discretion of supervisors and employers. This leaves most U.S. employees without any entitlement to alterations in their days or times of work. Thus, most American workers who want schedule flexibility will have to ask permission of their employers at some point for such modifications.
This system of managerial discretion requires that employees be willing to engage in bargaining and negotiation with their immediate supervisor if rates of flexibility in employment are to improve. But employees’ willingness to do so must surely reflect their estimates of the costs and benefits to be derived from the attempt to bargain. The system of managerial discretion also requires employers who are sometimes willing to approve requests for flexibility, if rates of flexible employment in the labor force are to grow. According to surveys of employers, growing percentages of employers report that they have employees in alternative work arrangements (Beers, 2000; BLS, 2009; Bond, Galinsky, Kim, & Brownfield, 2005), suggesting reasons for cautious optimism that voluntary cooperation might yet yield significant improvements in schedule flexibility among workers (Kelly, Moen, & Tranby, 2011). Yet there remain grounds for doubt that managerial discretion alone can produce flexibility for workers proportionate to their need to provide family care.
Our goal is to better understand the social dynamics of both the employee “ask” for accommodations and employer acceptance of flexibility requests. Without understanding the contours of this process, which directly contrasts with the statutory provision of employee rights to flexibility found in public policies in Europe, scholars and public policy analysts have little evidence that the current free market system either fails to efficiently provide flexibility to employees who want it, or operates in a discriminatory fashion against certain classes of employees. These are key pieces of evidence necessary to convince policymakers that the problem of inadequate flexible work options is both pervasive and not amenable to private employer solutions.
Variability in organizational acceptance of flexible work practices has been widely documented and its patterning has been the subject of empirical research. Organizational scholars have shown that the availability of work–family policies (like other fringe benefits) is associated with high-performance organizations that employ large proportions of professional and technical workers (Deitch & Huffman, 2001; Osterman, 1995). Extending flexible work options to workers in high-performance workplaces where autonomy and schedule control increase worker productivity and promote retention makes sense, especially when labor markets are tight and firm-specific skills are important.
But it does not necessarily follow from these dynamics of diffusion that female-dominated workplaces would be more likely to offer flexible schedules or that female workers within firms would be more likely to work flexible schedules, despite women's disproportionate responsibility for family care. Female-dominated jobs in retail and clerical settings, and even many professional and semiprofessional jobs such as teacher or social worker, are often structured as support positions or client-focused positions that make schedule flexibility more challenging to implement. Empirical research demonstrates little association between the proportion of female workers in an occupation and the proportion working a flexible schedule; when associations are found, they are often negative, revealing more flexibility in male-dominated occupations (Glass & Camarigg, 1992; Golden, 2001).
Within workplaces that allow flexible work options, access to flexible scheduling is by no means universal across job categories. Both Lambert (2009) and Deitch and Huffman (2001) note the increasingly bifurcated nature of the contemporary U.S. labor market—on the one hand are educated and highly skilled workers whose hours are long but whose daily activities are increasingly less monitored and controlled. On the other hand are disposable workers in low-skilled jobs fighting for more hours and permanent schedules rather than the increasingly contingent and unstable work shifts offered to them in highly competitive labor markets. The long-term shift in employment from unionized manufacturing enterprises to service industries that use part-time and temporary workers has left millions of less skilled workers willing to accept inferior working conditions in order to attain a permanent full-time job (Kalleberg, Reskin, & Hudson, 2000). The bulk of these employees are readily replaceable, and so have little bargaining power with which to gain schedule modifications to accommodate family care.
These organizational dynamics get reflected in the availability and use of flexible work practices among different categories of workers. Mirroring the distribution across firms, flexible schedules are more commonly available to workers in high-status positions of authority and in managerial and professional occupations (Glass & Fujimoto, 1995; Golden, 2001). Rather than being distributed according to need, then, fathers have more access to flexible schedules than mothers, especially low-income single mothers (Glass & Camarigg, 1992; Lambert, 2009; McCrate, 2002). Much of the difference between fathers’ and mothers’ access to flexibility can be explained by mothers’ tendency to cluster in lower-status occupations with significantly less authority.
The availability of flexible scheduling should not be confused with employee take-up or use of flexible work schedules, however. Evidence suggests that women, particularly mothers, are more likely to ask for and receive schedule flexibility than men when it is possible to do so (Fried, 1998; Kelly, Ammons, Chermack, & Moen, 2010; Powell, 1997). The combination of lower availability among mothers but greater uptake when available leads to rough gender parity between women and men in the use of flexible scheduling on a regular basis (BLS, 2009).
Why should there be significant reluctance to ask for flexible scheduling, or take advantage of advertised work life programs within firms, particularly among men? In contemporary American workplaces where salaried workers are expected to show devotion to their organization through nearly continuous availability (Jacobs & Gerson, 2004; Schieman, Milkie, & Glavin, 2009), asking for a work life accommodation itself could alter the perception of that employee's commitment to the firm and subsequent chances for advancement (Blair-Loy, 2003). Given these risks, accurately assessing one's chances for successfully negotiating an altered work schedule becomes crucial. Wasted requests might not just result in a negative decision from management; they might jeopardize an employee's future with that firm.
Since women often couch their requests for flexible work schedules in terms of their family care obligations, their caregiving status becomes particularly salient and may lower employers’ willingness to comply, believing it is not in the work unit's best interest to do so. According to qualitative accounts, fathers are much less likely than mothers to admit that their need for workplace flexibility is family related (Gerson, 1995; Powell, 1997), thereby masking men's family status, which could increase their odds of receiving employer approval. Because women are, on average, more likely than men to signal family needs as reasons for their flexible scheduling requests, women may experience more resistance when trying to negotiate flexible work schedules even when they are evasive or give nonfamily reasons for wanting flexibility.
Many scholars go further to predict that even employees who have successfully negotiated a flexible work arrangement end up paying a steep price in foregone earnings and promotions as employers differentiate between traditional workers and those using flexible work options. Narrative accounts of workers afraid to use their employer's flexibility policies because they believe their work careers would suffer as a result are plentiful (Crittenden, 2001; Hochschild, 1997; Williams, 2001). In these qualitative accounts, both men and women fear negative career consequences from any deviation from a standard long-hour schedule. Respondents often report that their use of flexibility policies would make their status as caregivers more visible and salient in the workplace, much to their disadvantage. It is not completely clear in these accounts whether flexibility employed for other purposes, like continuing education for career advancement, would be penalized as well.
Specifically, family care is often conceptualized by scholars and employers as a competing interest that absorbs workers’ time and energy (Budig & England, 2001). As such, workers with significant responsibilities for family care may be viewed by employers as exploiting flexible work practices in order to shirk their work responsibilities. Indeed, even the expectation of lower productivity can create bias in managers’ performance evaluations and pay decisions (Bailyn, Ridgeway, & Correll, 2004). If parenthood becomes salient, as it might if an employee works an irregular schedule, managers and co-workers might activate stereotypes in their appraisal of that colleague's contribution to the work unit. Motherhood in particular, acts as a status characteristic that invokes negative stereotypes about competency, productivity, and commitment in the workplace (Correll, Benard, & Paik, 2007).
This review of the literature suggests three salient dimensions affecting employer willingness to grant flexible scheduling requests: (1) the gender of the employee, (2) the status and authority of the job held by the employee, and (3) the reason or justification given for the request. In surveys and qualitative research, these dimensions co-vary in ways that make it difficult to isolate the separate contributions of each factor to the outcome of flexibility requests. Thus, our goal was to disentangle these factors by using an experimental paradigm to examine the specific impact of each on managers’ willingness to grant employees’ flexible work scheduling requests.
Here we present the results of two empirical studies. Our first study investigated how managers react to different types of requests for flexible work arrangements and whether their reactions vary depending on the gender of the employee, the status of the job held by the employee, and the reasoning given by the person making the request. We randomly assigned adults who were currently in a supervisory role to various scenarios depicting a fictitious employee requesting a flexible work arrangement. The target employee was either male or female and in either a relatively high-status managerial/professional occupation or a lower-status hourly wage occupation. Further, these fictive employees presented one of two reasons for their request for a flexible work schedule—either they had family responsibilities (i.e., childcare) or they wanted to engage in an activity that would ultimately increase their work productivity (e.g., taking professional development classes). We chose these two types of requests for flexible work arrangements because they have been found to be the two most common categories of reasons for flextime requests and, not coincidentally, are either much more typical of women (family responsibilities) or men (increasing productivity; Travis, 2003). Our study participants (all managers) estimated their likelihood of granting the employee's request then indicated how much they respected and admired the target.
We hypothesized that participants would be less likely to grant men flextime when they cited a family responsibility reason compared to women who made the same request and compared with men who provide a productivity reason for their request. We base these hypotheses in the backlash literature (Brescoll, 2011; Brescoll, Dawson, & Uhlmann, 2010; Brescoll, Uhlmann, Moss-Racusin, & Sarnell, 2011; Coltrane, Miller, DeHaan, & Stewart, 2013; Rudman, 1998; Rudman & Mescher, 2013; Vandello, Hettinger, Bosson, & Siddiqi, 2013), which shows that individuals who engage in gender atypical behaviors tend to experience social and economic penalties. Men who cite family responsibilities as the reason why they want flexible work arrangements are engaging in counternormative behavior (Brescoll & Uhlmann, 2005) and may therefore be less likely to have their request granted.
We based our second set of hypotheses addressing the interactions between all three factors (employee gender, status, and reason for the flextime request) on system justification theory (SJT; Jost & Banaji, 1994). SJT proposes that in addition to motives to see the self and one's in-groups positively, people have a fundamental desire to view their social system as just, fair, and good and will thereby engage in a number of strategies to rationalize the status quo (Jost & Banaji, 1994). In some instances, the status quo in social and political arrangements is nonhierarchical and egalitarian but in many instances in contemporary society, social/political arrangements are hierarchical and even exploitative of some social groups. One example germane to the present research is that men have higher status and more power than women (Sidanius & Pratto, 1999).
One key way that individuals can perpetuate the status quo is to give further benefits to those already in positions of privilege. Thus, we hypothesized that high-status men who requested flextime for career advancement reasons would be more likely to have their request granted than equally high-status women making the request for the same reason. Granting a flexible work schedule only to high-status men serves to perpetuate the status quo as only their careers will advance whereas women's careers will not. Furthermore, we hypothesized that managers will be more likely to grant flextime to high-status men than high-status women who are both asking for career development reasons because managers accord greater respect to men in these positions than women. High-status men, in general, are more respected than high-status women (Eagly & Steffan, 2000). But, SJT would further predict that high-status men looking to advance their careers would be especially likely to garner respect and admiration compared to high-status women doing the same, as this behavior further reinforces a gender hierarchy in which men not only have more initial status and power than women, but continue to accrue more status and power than women via educational credentialing and skill development.
Building off the findings from our first study, our second study examined employee expectations of success if they asked for a flexible work arrangement. In this study, we presented male and female adult participants with the same workplace scenarios from Study 1 in which it would be reasonable to ask for flextime. Respondents were then asked to imagine themselves in that scenario and estimate the likelihood that they would receive flexible work accommodations if they requested it. We hypothesized that male participants would be less likely than female participants to think their request would be granted if they asked for family (as opposed to career) reasons because they might experience backlash if they did so.
Study 1: Methods
Seventy-six managers (31 female) participated in an online study in exchange for a 1/40 chance of winning $40. On average, participants were 39.87 (SD = 10.92) years old. The majority of our managers were White (N = 60), four were African American, three were Hispanic, seven were Asian, and two identified themselves as “Other.” Our managers worked in a variety of occupations but all had at least one year of managerial/supervisory experience, which makes them more likely than persons without managerial experience to be in a position to address employee requests for flexible work schedules (such as college student participants). Managers were recruited from an online database maintained by an East Coast University. They had participated in previous studies with this online database and in these studies had indicated that they were managers. Thus, in our study, we were able to specifically recruit these individuals without them knowing we were targeting them for their supervisory experience.
Participants were randomly assigned to read one of eight vignettes about an employee requesting a compressed work schedule. The employee was either a male (John) or a female (Jennifer) who was either in a low- or high-status occupation (pharmacy clerk or chief pharmacist). In the vignette, the employee was requesting a compressed work schedule either to take professional development classes or to take care of his or her children.
To examine managers’ reactions to requests for a flexible work schedule, we chose to focus on a request for a compressed work schedule, where an employee would work a full work week (40 hours) but would be allowed to rearrange his or her schedule to have two afternoons off per week, because this request holds constant the number of work hours and location, thereby eliminating concerns about possible reductions in salary, change in benefits, or other confounding sources of variation, which are more typically associated with a reduced-hours schedule or telecommuting arrangements. Moreover, a pilot test with a separate sample of 30 participants (14 female) revealed that ratings of the likelihood of having the request for a compressed work schedule granted by the company (M = 3.87; SD = 1.80) and ratings of how costly it would be for the company to grant this request (M = 3.57; SD = 1.14) fell roughly in the middle of a 7-point scale ranging from not at all to extremely. Whether or not to grant an employee's request for a compressed work schedule is not an unequivocally easy decision for managers. But, in choosing a compressed work schedule for our vignette, we could (to whatever extent possible) eliminate managers’ concerns about possible salary reductions, changes in benefits, or other factors and thereby allow us to test our hypotheses about the influence of employee gender, job status, and reason on managers’ decision to grant the employees’ flextime request. Furthermore, this particular option, because it requires no cutback in hours, is probably less likely to trigger turndown or bias than other flextime options (e.g., shifting to a part-time schedule) and is therefore a considerably more conservative test of our hypotheses than other flexible scheduling arrangements.
Our goal was to select a job for our targets that would be perceived as gender neutral, yet could also be either high or low in status. Furthermore, the low- and high-status versions of this job also needed to be perceived as gender neutral so that we could independently examine the effects of target gender and job status on managers’ willingness to grant the flextime request. Thus, we gathered a list of occupations from the Bureau of Labor Statistics (2009) that had roughly equal percentages of male and female employees and conducted an extensive pilot test of four pairs of jobs (office clerk/partner in a consulting firm; pharmacy clerk/chief pharmacist; design clerk/senior product designer; editorial clerk/editor-in-chief at a publishing house). Pilot testing revealed that only the jobs of chief pharmacist and pharmacy clerk emerged as significantly different in perceived status (Mchief pharmacist = 5.88 and Mpharmacy clerk = 3.56 on a 7-point scale ranging from low to high status), t(33) = 7.97, p < .001, but not in perceived masculinity/femininity (Mchief pharmacist = 3.97 and Mpharmacy clerk = 4.00 on a 7-point scale ranging from feminine to masculine), t(37) = .10, p = .92. Furthermore, these two jobs did not significantly differ on participant ratings of “how crucial is it for a person to be present in the office during regular work hours (9 am–5 pm)?” t(32) = .56, p = .58 (Mchief pharmacist = 5.55 and Mpharmacy clerk = 5.36 on a 7-point scale ranging from not at all to extremely).
Finally, in terms of choosing a reason for our targets’ flextime request, our goal was to choose gender-stereotypical reasons: one identified with family and women (caring for children) and one with career advancement and men (taking professional development classes). Yet, at the same time, we wanted to insure that these two reasons for requesting a flextime schedule would be perceived as equally legitimate reasons. Indeed, our pretesting revealed that both reasons provided an equally legitimate basis for making the request (Mtake care of children = 5.37 and Mattend professional development classes = 5.30 on a 7-point scale from not at all to extremely), t(29) = .24, p = .82.
To ensure that managers carefully read the vignettes containing the manipulations of employee gender, job status, and reason for the request, they were asked the following three questions: “What is Jennifer's [John's] position?” “Jennifer [John] would like to rearrange her [his] work schedule by working a full work week but taking _____ afternoons a week off.” “Why does Jennifer [John] need to rearrange her [his] schedule?” Only those individuals who correctly answered all three questions (85%) were included.
Willingness to grant a request
After reading one of the vignettes and answering the manipulation check questions, participants responded to three items: “If you were John's [Jennifer's] manager, would you grant the request?” “How reasonable is this request?” and “How much does John [Jennifer] deserve to have his [her] request granted?” These were on a 7-point scale ranging from not at all to extremely. The three items loaded onto a single factor, and we averaged the responses on these questions to create a composite measure of the willingness to grant a request (Cronbach's α = .92).
Respect and admiration
Participants next indicated how much respect and admiration they felt at that moment toward the target, rating each item on a 7-point scale ranging from not at all to very much. The two items loaded onto a single factor and were averaged together into a composite score (Cronbach's α = .87).
At the end of the survey, participants completed a demographic questionnaire, providing information about their gender, age, ethnicity, and their occupation.
To assess managers’ willingness to grant an employee's request for a flexible work schedule as a function of employee's gender, job status, and reason, we conducted a 2 (employee gender: male vs. female) × 2 (job status: high vs. low) × 2 (reason: taking care of children vs. taking professional development classes) between-subjects analysis of covariance with willingness to grant the request as the dependent variable and participants’ age as a covariate (Table 1).
Table 1. Managers’ Willingness to Grant Flextime Request (Study 1)
We first assessed any main effects of gender, job status, and reason on managers’ willingness to grant the flexibility request. Combined across job status and reason, male employees were no more likely than female employees to have their request granted, F(1, 58) = .12, p = .73, Cohen's d = .09. Similarly, there were no main effects for reason, F(1, 58) = .04, p = .98, Cohen's d = .05, or job status, F(1, 58) = 1.63 p = .21, Cohen's d = .34, on manager's willingness to grant the flextime request.
The results revealed a marginally significant three-way interaction between employee gender, job status, and reason for making the request, F(1, 58) = 3.16, p = .08 (Table 1). For male employees, job status and reason interacted to influence managers’ willingness to grant the request, F(1, 58) = 9.37, p = .003, Cohen's d = .80. Specifically, managers were more likely to grant the flextime request to high-status male employees for the career development reason than for the childcare reason, F(1, 58) = 7.11, p = .01, Cohen's d = .70, whereas this pattern was reversed for the low-status male employees, F(1, 58) = 3.64, p = .06, Cohen's d = .50. In contrast to these findings, there was no interaction between job status and reason for female employees, p = .98.
Furthermore, consistent with our hypotheses, a high-status male employee making a flextime request for career development purposes (M = 6.31; SE = .24) was more likely to have his request granted than a female employee in the same job making the request for the exact same reason (M = 5.66; SE = .42), F(1, 58) = 2.48, p = .06, Cohen's d = .41. This difference was fully explained by the greater respect and admiration managers felt toward high-status male employees (M = 5.69; SD = 1.01) compared to high-status female employees (M = 4.70; SD = 1.51) making the request for career development, Cohen's d = .77. Mediation analyses (modeled after Newheiser, LaFrance, & Dovidio, 2010) revealed that this difference in respect and admiration was significant, F(1, 58) = 11.75, p = .001, and that the difference in the willingness to grant the request was no longer significant when the degree of respect and admiration was included in the model as a covariate, p = .15.
Finally, we did not find support for our hypothesis that men would be less likely to be granted flextime than women when they cited childcare as their reason for needing flextime, F(1, 58) < 1, ns.
In Study 1, we found partial support for our hypotheses. First, we found support for our hypothesis that high-status men looking for flexible work schedules for career development reasons would be more likely to obtain these schedules from managers than high-status women and that greater respect for men in these positions accounts for managers’ decisions. These hypotheses were grounded in SJT (Jost & Banaji, 1994), which proposes that individuals have a fundamental psychological motive to defend and perpetuate the status quo. This supervisory decision results in perpetuating the gender status hierarchy by providing a pathway for high-status men to gain even more status than women via career development.
Second, we expected that men seeking flextime for childcare reasons would be discriminated against relative to women seeking flextime for the same reasons since past research demonstrates a bias against women and men engaging in gender-stereotype incongruent behaviors (Brescoll & Uhlmann, 2005), but our manager participants did not show this bias. However, unlike past research, our study not only manipulated the gender of the employee but also the employee's status (thereby disentangling the effect of gender and status). It may be that, for men at least, their status better accounts for decision making on the part of employers than their gender, per se, as our data indicated that low-status men asking for flextime for childcare reasons were more likely to have their request granted than high-status men asking for the same reason whereas high-status men requesting flextime for career advancement reasons were more likely to have their request granted than low-status men in the same positions. For female employees, however, neither their status, nor their reasons for requesting flextime impacted managers’ decisions to grant their request. Although these results are inconsistent with past work that shows a bias against individuals who engage in gender-stereotype incongruent behaviors, they are consistent with Brescoll and Uhlmann's (2008) work which separately manipulated gender and the status of the target and found that for male targets, their status was a better predictor of others’ reactions than whether they engaged in gender atypical behaviors.
It is worth noting that our study design necessarily yielded relatively conservative tests of our hypotheses. For one, the request itself—for a shift in work hours rather than a reduction in work hours—is probably less likely to trigger turndown or bias than other flextime options (e.g., shifting to a part-time schedule) because it requires no cutback in hours. Second, we purposely choose a gender-neutral occupation. Had we choose gender-stereotypical occupations, it is likely that managerial biases would be even stronger than what we found here. Future research should explore using other types of occupations and flexible work arrangements, since most workers work in segregated, gender-stereotypical jobs rather than the gender-neutral ones we choose.
Building off the findings from our first study, our second study examined the parallel issue of who is willing to ask for flexible work arrangements and what situational factors promote willingness to ask for flexible work arrangements. Our primary goal was to assess whether employees would be accurate in their assessment of managers’ likelihood to grant flextime requests. Second, we wanted to examine whether employees’ perception of potential backlash played a role in their assessments. In this study, we presented male and female adult participants with the same workplace scenarios from Study 1 in which it would be reasonable to ask for flextime. Respondents were then asked to imagine themselves in that scenario and estimate the likelihood that they would receive flexible work accommodations if they requested it. Thus, we used naturally occurring variation in respondents’ gender to detect gender differences in expectations of success in securing this flexible schedule. In these scenarios we systematically varied the reason participants would ask for flextime—either family responsibilities (e.g., childcare) or interest in pursuing an outside accreditation or degree. Again, as in our first study, we varied the status of occupation that the participants imagine themselves to be in—either a high-status, managerial job or a lower-status, hourly wage occupation.
Consistent with Linda Babcock's work (Babcock & Laschever, 2003) that shows that women are less likely than men to ask for raises and other workplace rewards than men, we expected that women would be less likely than men to imagine themselves asking for flexible work arrangements than men. However, we also expected that men would be less likely than women to ask for flextime when family responsibilities are the reason for their request given men may feel pressure to conform to prescriptive gender stereotypes. We further predicted that men would be less likely to ask for flextime because they anticipated that this type of request could be stigmatizing and damage their future in the organization (i.e., that they would encounter backlash; Brescoll, 2011; Moss-Racusin & Rudman, 2010).
Study 2: Method
Participants were 159 women and 84 men taking part in an online study in exchange for a 1/40 chance of winning $40. Participants were recruited from an online database kept by a large East Coast University. Unlike Study 1, participants in this study did not have to be supervisors or managers themselves. On average, participants were 40.51 (SD = 16.85) years old, and all had workplace experience. As in Study 1, the majority of our participants were White (82%), while 9% identified as Asian, 2.3% were Black, 2.3% were Hispanic, and 3.5% identified themselves as “Other.”
Each participant was presented with one of the vignettes from Study 1 except that we asked participants to imagine themselves in that scenario and estimate the likelihood that they would receive flexible work accommodations if they requested it. To facilitate this process, we omitted the protagonist's gender by replacing the name (i.e., John or Jennifer) with “you,” which has been extensively pretested in prior research (Brescoll & Uhlamnn, 2005; Moss-Racusin, Dovidio, Brescoll, Graham, & Handlesman, 2012). As in Study 1, we systematically varied the reason participants would ask for flextime—either they had childcare responsibilities or they were interested in pursuing an outside accreditation or degree. Again, as in Study 1, we varied the status of occupation that the participants imagined themselves in—either a high-status managerial job or a lower-status hourly wage occupation.
We again assessed the accuracy of participants’ responses to the manipulation check questions adapted from Study 1. As in Study 1, only individuals who accurately answered all three questions (89%) were included in the study.
Perceived willingness to grant the request
Following the vignettes and the manipulation check, participants were presented with three questions: “How likely do you think your request is to be granted?” “How reasonable do you think your supervisor would find your request?” “Would your supervisor think you deserve to have your request granted?” Participants responded on a 7-point scale from not at all to extremely. These items loaded onto a single factor and were averaged (Chronbach's α = .90).
Participants’ backlash concerns were assessed with the five items adapted from Brescoll (2011) on a 11-point scale ranging from not at all to very much so (“Would you be concerned that you might be disliked by others at work as a result of making this request?,” “If you make this request, will you be concerned about whether someone at work would think that you were acting ‘out of line’?,” “Would you be concerned that people would judge you for making this request?,” “Do you think that making this request would cause you to lose power?,” and “Do you think that making this request would make you seem less effective?”). The five items loaded onto a single factor and were averaged into a composite score (Chronbach's α = .92).
Results and Discussion
We again began by assessing any main effects of respondents’ gender, their assigned job status, and assigned reason on their perceptions that their flexibility request would be granted. Contrary to our hypotheses, we found that female employees, overall, were more likely to perceive that their flextime request would be granted than male employees, F(1, 229) = 6.83, p = .01, Cohen's d = .34 (M = 4.80, SD = 1.22; M = 4.38, SD = 1.25, respectively). In addition, we did not find that respondents assigned to high-status jobs were more likely to believe their requests would be granted, F(1, 229) = .05, p = .81, or that respondents given a career reason for their request universally believed their chances of getting their request approved were higher, F(1, 229) = 1.02, p = .41. Instead, we found a more complicated pattern in which gender significantly shaped the reasons and statuses that respondents believed would yield successful “asks” for schedule accommodations.
In a 2 (participant gender: male vs. female) × 2 (job status: high vs. low) × 2 (reason: taking care of children vs. taking professional development classes) between-subjects ANOVA we found a marginally significant three-way interaction between participant gender, job status, and reason for request, F(1, 229) = 2.85, p = .09 (Table 2). Specifically, male participants rated the highest likelihood of having their request granted when they imagined themselves in a low-status position and asking for a career development reason, whereas female participants rated the highest likelihood of having their request granted when they imagined themselves in a high-status position and asking for a career development reason. In addition, female participants who imagined themselves in a high-status job and making a flexibility request for a career development reason (M = 5.16; SD = 1.16) rated the likelihood of having their request approved significantly higher than did male participants imagining themselves in the same high-status job and the request for the exact same reason (M = 4.15; SD = 1.35), F(1, 229) = 7.34, p = .004, Cohen's d = .80.
Table 2. Employees’ Perceptions That Their Flextime Request Will Be Granted (Study 2)
We also found, consistent with our hypotheses, that regardless of their job status, male participants believed they were less likely to have their flextime request granted if it was to care for children (M = 4.33; SD = 1.26), than female participants who imagined making the same request for the same reason (M = 4.59; SD = 1.15), F(1, 229) = 3.09, p = .02, Cohen's d = .22. Importantly, the reports of anticipated backlash fully explained this difference. As we predicted, male participants who imagined making a flextime request for a childcare reason (M = 4.83; SD = 2.84) anticipated greater backlash at work than did female employees who imagined themselves making a flextime request for childcare reasons (M = 4.22; SD = 2.64), F(1, 229) = 13.45, p < .001, Cohen's d = .23. The observed gender difference in the perceived likelihood of having a childcare request granted was no longer significant when anticipated backlash was included in the model, p = .12. Thus, men's perceptions that they would encounter backlash when asking for flextime in order to care for their children appeared to explain why they were less optimistic about having their request granted than women who were asking for flextime for the same reason (childcare).
The purpose of this research was twofold: (1) to examine how managers use ascriptive information in deciding which requests for flexible work scheduling to grant among employees, and (2) to uncover whether employees understand this managerial logic in deciding whether to ask for flexible work arrangements. In Study 1, we found that managers were more likely to grant flexible work schedules to high-status men asking for these schedules to advance their careers than high-status women asking for flexible schedules for the same reason. We further found that the fact that managers respect high-status men more than high-status women accounts for their gender-biased decision. These findings are consistent with SJT (Jost & Banaji, 1994), which proposes that people have a fundamental psychological motive to defend and perpetuate the status quo. In this case, by being more likely to grant high-status men flextime than high-status women when they were both making the request to advance their careers, our manager participants were not just reinforcing the gender status hierarchy wherein men have more status and power than women, but also perpetuating it.
Past work has shown that men tend to advance in their careers faster than women (Ibarra, Carter & Silva, 2010). Multiple reasons for this have been hypothesized but the most commonly cited reason centers on women's childcare responsibilities. Because women are more likely than men to take time off from work for child rearing, they are more likely to get off career ladders and thus advance more slowly overall than men. However, our study offers an additional explanation—men may be more likely to be granted leeway in their work schedules to advance their careers, particularly if they are already in a nonhourly wage, high-status position. This advantage granted to men already in high-status positions may contribute, in part, to their more rapid career advancement relative to women. This pattern could also be related to the more common observation that women advance more slowly because of family care interruptions. In this case, the signaling association between female gender and family care may be so strong that even high-status women seeking accommodations to advance their careers might be suspected of either dissembling in their reason for the accommodation, or less deserving of the chance to gain further training because they will withdraw or lower their work effort at some point in the future. Even women who have “proven” themselves by achieving a high-status occupation and asking for further career training cannot overcome this actuarial mistrust of women workers.
In Study 1, we expected that employers would discriminate against men by being less likely to offer them flextime than women when the request was made in order to accommodate their childcare responsibilities. We based this hypothesis in the literature showing that others discriminate against men and women who engage in gender-stereotype incongruent behaviors (Brescoll & Uhlmann, 2008; Rudman & Glick, 1999). In this case, men who ask for flexible work schedules because they need to care for children would be engaging in such behavior. However, we did not find that managers discriminated against men in this case; they were no more likely overall to grant flextime to women who asked for childcare reasons than men who asked for the same reasons. Somewhat curiously, low-status men who asked for a childcare accommodation were particularly likely to gain approval from the managers in this study, as if their low status made these men's requests for more family involvement laudatory in ways that high-status men could not tap.
Although this finding is puzzling in light of past work, our study was notably different from most research in that we not only manipulated the gender of the employee but also the employee's status (thereby disentangling the effect of gender and status) and carefully ensured that we choose a gender-neutral occupation for the employee. In our study, we found that men's status better predicted employer decision making than their gender, in that low-status men asking for flextime for childcare reasons were more likely to have their request granted than high-status men asking for the same reason while high-status men requesting flextime for career advancement reasons were more likely to have their request granted than low-status men asking for career reasons. Thus, future experimental work examining bias against men who violate gender stereotypes should consider varying the status of the men and women engaging in gender stereotype-incongruent behavior, as status may be a particularly important factor in understanding bias against men.
For female employees, however, neither their status, nor their reasons for requesting flextime significantly impacted managers’ decisions to grant their request. This pattern suggests that motherhood status is so strong in employment decision making that neither high-job status nor career justification for a flexibility request can modify it (Bailyn et al., 2004).
In Study 2, our goal was to examine whether individuals would be accurate in their assessment of mangers’ likelihood of granting flextime requests. Overall, we found that employees were incorrect in their predictions. For one, both sexes believed their requests are less likely to be granted than the managers in Study 1 actually granted. Second, we found that high-status women asking for flextime for career development reasons were the most likely group to believe that their request would be granted, but in Study 1, we found that actual managers were, in fact, unlikely to grant this request relative to other groups (particularly high-status men asking for flextime for career development reasons). Further, we found that high-status men asking for flextime for career development reasons were the least likely to believe that their request would be granted. Thus, it appears that high-status women overestimated the likelihood that their flextime would be granted whereas high-status men underestimated the likelihood of approval of their request.
In Study 2, we also found that men, regardless of their status, believed that they would be less likely to have their request granted than women if their reason for asking for flextime centered on childcare concerns. In this case, the fact that men believed they would encounter backlash if they engaged in this gender stereotype-incongruent behavior accounted for the reason why they believed their request would be less likely to be granted than women's request. This finding is consistent with past research showing that fear of potential backlash accounts for why individuals are unlikely to engage in gender stereotype-incongruent behaviors (Brescoll, 2011; Moss-Racusin & Rudman, 2010). Thus far, this work has centered on the reasons why women are unlikely to engage in masculine behaviors (such as dominating a meeting or self-promoting). Our study is one of the first to show that men also may be subject to such fears (see also Coltrane et al., 2013; Rudman & Mescher, 2013; Vandello et al., 2013). Importantly, however, our work shows that in this case, men were inaccurate in their prediction that if they made a request for flextime for childcare reasons, they would be less likely than women to have it granted. Our first study shows that managers overall do not exhibit this pattern of discrimination.
Our study contributes to the literature on flexible work bias in two important ways: First, we demonstrate that managers proactively favor schedule flexibility requests from high-status men, specifically when those requests are coupled with reasoning consistent with firm objectives (career development) rather than irrelevant or arguably inconsistent ones (family care). Women, irrespective of their reason for requesting flextime and their status, were less likely than high-status men to have their request granted. This is vitally important for public policy—women were not able to use either their status within the firm or the reasoning behind their request to enhance their probability of successfully getting a schedule change. Second, we reveal that employees do not understand the managerial logic undergirding successful requests very well, nor the level of stigma attached to asking. Therefore, the “match” between requesting and receiving schedule flexibility in this dynamic system is less than optimal or efficient. In particular, both sexes underestimated their probabilities of successfully obtaining flextime, though men more so than women. This suggests that perhaps flexible scheduling in the United States has been depressed in part because of employee pessimism about their ability to get their request granted and fears about potential backlash from asking for alternative schedules. Relative to men, however, women may be underestimating the negative consequences of asking and overestimating their true probability of success. Men, especially high-status men, significantly overestimate the negative consequences of asking and underestimate their probability of success.
Like other analyses (McCrate, 2002), this study demonstrates the paradoxes of free market bargaining between employers and employees when it comes to family care. Those workers most in need of flexible scheduling—low-status mothers with childcare needs—were among the least likely to get such an accommodation approved by managers. But all women workers, regardless of status or reason for the request, faced a gendered wall of resistance to their requests for schedule flexibility. Men, on the other hand, faced status-specific resistance. While high-status men received the most approval for career development requests for flexible scheduling, low-status men received the most approval for childcare reasons.
The experimental results here provide evidence that status characteristics like gender and job level affect managerial decisions about who can receive schedule accommodations. The pattern of evidence suggests market forces cannot adequately provide flexibility to those who require it for childcare reasons, particularly low-status mothers when compared to low-status fathers. Similarly, high-status women appear less likely to receive approval for their requests than high-status men, even when it is justified in business terms.
What are the likely consequences of the labor negotiation system used to distribute schedule flexibility in the United States? Since our evidence suggests workers are unduly pessimistic about their chances of success and often fear retaliation for asking, requests for flexibility are probably depressed relative to employee preferences. As long as “the ask” involves risk and uncertainty, employees will suppress their inclinations and preferences for greater schedule flexibility and control over work scheduling and market change will be slower than optimal. As long as managerial approval for flexibility requests is disproportionately given to high-status men, who are themselves least likely to ask for flexibility, market change will be slower than optimal. As long as women are more likely than men to tolerate the risks of asking (perhaps because of their stronger preferences for schedule flexibility), employees in general will see more rejection of requests than would be true in a more representative sample of employees. This reinforces uncertainty about the wisdom of asking for accommodations, creating a self-perpetuating cycle that suppresses the diffusion of flexible scheduling in the United States labor force.
Is this suppression costly for the American labor force, and for American employers and firms? To the extent that lack of flexibility forces some mothers out of the labor force, and creates stressful childcare and home management for other workers, suppression of flexible scheduling probably contributes to lower standards of living for children and more stress-related decrements to parents’ mental and physical health. Evidence already suggests that American parents are uniquely disadvantaged by the financial, social, and physical challenges of childrearing compared to their counterparts in other developed countries (Glass & Simon, 2012). Combined with children's increasing reliance on their mothers’ earnings over time (Bianchi, Milkie, & Robinson, 2006), the inability of American women to craft schedules that accommodate family care has direct implications for the persistence of child poverty in the United States as well. Our work suggests that crafting public policy that both supports families and encourages paid work will require specific attention to the unmet need for schedule flexibility faced by American workers, and the forces discouraging free market provision of flexibility to those who seek it.
American firms are also negatively affected by the dynamics of labor negotiation. Fewer workers ask for accommodations than desire them, leading perhaps to lower productivity and unnecessary labor turnover. Lower standards of living among children as a result of either labor force withdrawal or frequent job changes among their caregivers negatively affects the human capital of the next generation, and increases the search costs of firms seeking qualified workers. Moreover, the fact that managers are actually more amenable to schedule flexibility than most employees believe suggests these losses are avoidable to a large extent.
If one goal of public policy is to ensure that workers caring for dependents are able to both maximize their earnings to the benefit of their families and minimize the distress involved in providing care while working for pay, these results suggest an appropriate role for public policy in overcoming the resistance among managers to flexible scheduling for their women workers and for care purposes among men. The “right to ask” laws in Great Britain and some EU nations require employers to justify their reasons for denying flexible scheduling requests. While even such requirements may not be strong enough to overcome the resistance to schedule flexibility for women uncovered here, at minimum they force some reflection on the part of managers who may otherwise respond with status and gender-biased beliefs. Other remedies might take the form of enhancement and greater enforcement of equal employment opportunity laws that prohibit gender and racial discrimination in access to work amenities such as flexible scheduling, telecommuting, and work hour reductions. At minimum, protection from retaliatory treatment for workers who request schedule accommodations might substantially increase the number of workers in need who feel free to ask for flexibility, though our results suggest it would have little effect on managerial biases in approval of such requests. These, and other possible remedies for the suppression of schedule flexibility uncovered here, deserve careful evaluation if we are to improve the well-being of working families.
VICTORIA L. BRESCOLL is an Assistant Professor Organizational Behavior in the Yale School of Management. She received her MS, MPhil, and PhD in social psychology from Yale University where she was supported by a graduate research fellowship from the National Science Foundation. Her research focuses on the impact of stereotypes on individuals’ status within organizations, particularly the status of individuals who violate gender stereotypes. Her research has been widely reported on in the popular press including the New York Times, Business Week, the Wall Street Journal, the Washington Post, National Public Radio, and U.S. News & World Report. Her other interests include the cultural origins of stereotypes (e.g., the media), corporate social responsibility, and framing messages to improve health policy. In 2004, Professor Brescoll worked in the office of Senator Hillary Rodham Clinton under a Congressional Fellowship.
JENNIFER GLASS is the Barbara Bush Professor of Liberal Arts in the Department of Sociology and Research Associate of the Population Research Center at the University of Texas, Austin. She has published over 50 articles and books on work and family issues, gender stratification in the labor force, and mother's employment and mental health, with funding from the National Science Foundation, the National Institutes of Health, and the Alfred P. Sloan Foundation. She is currently serving as Vice-President of the American Sociological Association, and sits on the board of the Council on Contemporary Families. Her most recent projects explore the wage effects of flexible work practices, how telecommuting facilitates longer work hours for employees, and whether governmental work–family policies undermine or improve parents’ mental and physical health.
ALEXANDRA SEDLOVSKAYA is currently a Senior Researcher at Harvard Business School. She obtained her doctoral degree in Social Psychology from Yale University in 2011. Alexandra's research focuses on issues surrounding diversity in academic and corporate environments.