Abstract
This article documents macro-level trends regarding gender equity for women managers in paid workplaces and examines the importance of factors related to equitable promotion. Primary evidence is drawn from 1982, 2004 and 2010 surveys of work and learning activities of the employed Canadian labor force. These surveys provide unique national-level data on the managerial levels, qualifications, sex of supervisor and divisions of paid and unpaid labor among male and female managers which could provide benchmarks for further international surveys. Women’s representation in top-level jobs remains very restricted, and most women still manage only women. Greater employment experience and higher educational qualifications are now generally significant factors for promotion of women as well as men. But glass ceilings maintained by men and women’s own primary responsibility for household work remain the major obstacles to equitable promotion. Women managers’ increasing economic power remains contingent on facing up to these interrelated barriers.
Introduction
Women have experienced numerous inequities in paid labor. They have been most severely underrepresented in the most powerful managerial positions. If this inequity is to be overcome, the major barriers to women’s promotion opportunities should be clearly recognized and addressed. This inquiry offers new insight into the extent of inequity in the managerial hierarchy and the relative importance of different factors assumed to be related to women’s promotion prospects. We expect that the sexual division of paid and unpaid labor and the occupational segregation of women managers remain fundamental barriers to gender equity in the upper managerial hierarchy. This article is based on unique Canadian national survey data that document both changes in managerial relations and the distribution of paid and unpaid work. While there are substantial international general differences in the extent of both women’s participation in paid employment (e.g. Pettit and Hook, 2005) and men’s participation in unpaid household work (e.g. Hook, 2006), we suggest that similar patterns would be found regarding managerial promotion and household divisions of labor by comparable surveys in other advanced capitalist economies. In the following sections, our theoretical perspective is outlined, and then the survey data sources are described. The inequitable distribution of household and child care work and their relation to the paid work time of women and men managers and other employees is summarized. Then, the increasing role of managers and the rise of women managers are traced. Next, the representation of women in upper management and the extent of segregation of women managers are estimated. Finally, the relative importance of factors related to women managers’ promotion is analyzed and conclusions are drawn.
Theoretical Perspective
Various theoretical approaches have been used to try to explain sex differences in managerial representation, including role theory or expectation states theory (see Weyer, 2007) and human capital theory (e.g. Metz and Tharenou, 2001). The general theoretical perspective that informs this investigation of women managers’ representation in management focuses first on the relevance of both paid and unpaid labor to the constituting of class positions including management positions, and secondly on the resistance of those in positions of privilege to sharing power with others.
Unpaid labor remains largely invisible in the analysis of class positions in advanced capitalist economies. The organization of time in such societies has remained structured on inherently patriarchal grounds (Sirianni and Negrey, 2000). As the commodification of our lives proceeds, more women whose labor had been primarily devoted to unpaid work in the household are drawn more fully into circuits of goods and services commodity production. While it is conceivable that even childbirth could be fully commodified, both paid labor in commodity circuits and unpaid labor to reproduce labor power remain necessary for the foreseeable future of capitalism. As Armstrong and Armstrong (1983: 39) observed:
because capitalism is premised on free wage labor – on the separation of most aspects of workers’ reproduction from the production process – women’s reproductive capacities separate them out of the production process for childbearing work. This establishes the basis for the elaboration of sex differences, a sexual division of labor which subordinates women and pervades all levels of human activity under capitalism.
Although economic class analysis has predominantly focused on individual relations in the production process – particularly ownership of means of production versus hired labor – both paid and unpaid divisions of labor should be taken into account in identifying household classes (e.g. Livingstone and Asner, 1996). Understanding how household classes are constituted in terms of divisions of both paid and unpaid labor becomes increasingly pertinent as most women seek to be in the paid labor force and availability for paid work increasingly involves renegotiation of the division of unpaid labor. In order to comprehend opportunities for and limits on women’s entry into paid labor, and particularly here in an investigation of women’s upward promotion into management positions, women’s and men’s roles in the division of unpaid labor should be explicitly considered. Women may continue to be drawn into the paid labor force to reach virtual parity in numbers with men, but as long as women have to do most of the household and child care work, relatively few women living with men are likely to have time and opportunity to ascend to the top of the managerial ladder.
There is growing incidence of arguments that intellectual skills are eclipsing physical skills as key ingredients in economic success, and that ‘knowledge economies’ that utilize talents of women as well as men will excel. A global Organization for Economic Cooperation and Development study (Jutting et al., 2006) has found that the greater the economic and political power of women, the greater the country’s economic success. However, these are still capitalist economies, in which the central imperative is conversion by private enterprises of everything conceivable – including water, air and diverse personal services – into a vendible commodity, with relative indifference to gender equity. Inter-firm competition drives enterprises to produce profitable commodities cheaper than others by adopting more efficient technologies and minimizing labor costs. In order to live and qualify for jobs in such a dynamically changing ‘knowledge economy’, female and male workers of all sorts are impelled to seek greater access (via public education systems and such forums as the internet) to knowledge previously restricted for commodity production by private firms, as well as to diverse sources of knowledge for everyday life. This demand for knowledge generates unemployed reserves of qualified labor, while funds of underemployed knowledge among the employed become more systemic problems in advanced capitalist economies (Livingstone, 2009). Historically, the prevalence of male breadwinner norms and associated job selection practices generated male occupational preserves, such as corporate elites and organized industrial workers, which served to exclude or devalue the actual skill competencies of women (e.g. Livingstone and Luxton, 1989). As birth control measures, increased access to formal education and needs for additional family income all encouraged more women to enter paid labor markets, women have been enabled to contend more and more widely with men for good jobs. But those in established positions of power have rarely relinquished their privileges easily and there is little reason to expect that men who retain economic power in competitive capitalist workplaces will now react much differently. We expect to find preserves of power reproduced by exclusionary networks of men at the highest levels of otherwise changing knowledge economies.
Survey Data Sources
The main data analyses presented in this article are based on national surveys of the Canadian adult population conducted in 1982, 2004 and 2010, primarily focused on economic class relations. These are the only surveys conducted in Canada to date that permit identification of managers who exercise different levels of authority over other workers and also provide identification by all respondents of the gender identity of those who supervise them. In addition, the 2004 and 2010 surveys are the only known national surveys anywhere that provide data on both the paid and unpaid (household and child care) work of these managers with different levels of authority. As such, these surveys offer benchmarks for further international studies of managerial relations and paid and unpaid work.
The Canadian Class Structure Survey (CCS) was conducted in 1982. The CCS survey covered the entire adult population of Canada (N=2577). Our analysis focuses on the employed labor force over 18 years of age (N=1759). The CCS was a detailed survey of the economic class structure of the labor force which included extensive demographic data (sex, age, ethnicity, etc.) and permitted analysis of the gender composition of all class positions in the labor force, including managerial positions, as well as gender of one’s own supervisor. Further information on the CCS survey can be found in the main project publication by Clement and Myles (1994). The 2004 Work and Lifelong Learning (WALL) survey drew on the CCS survey for a number of questions. It also covered the entire adult population of Canada (N=9063). The current analysis focuses on the employed labor force over 18 years of age (N=5733). The WALL survey addressed paid employment conditions as well as unpaid housework, child care and voluntary work, and generated the first systematic, empirical assessments of paid and unpaid work conditions in relation to an array of adult learning practices: formal schooling, further education courses, informal training, and self-directed informal learning. The WALL survey also provided demographic data and detailed occupational information to identify managerial employees and other economic class positions. 1 The 2010 WALL survey included a slightly reduced set of questions, with a smaller national sample of the entire adult population (N=2028) and employed labor force (N=1256). Further information on both WALL surveys may be found at the research network website (www.wallnetwork.ca). Several population surveys by Statistics Canada and corporate executive surveys by Catalyst are also drawn on for secondary analyses.
This study has used descriptive, bivariate and multivariate statistical techniques. Logistic regression is applied for bivariate and multivariate analysis which includes raw and adjusted odds ratios (Hosmer and Lemeshow, 2000). The multivariate logistic regressions include controls for employment experience, educational attainment, sex of supervisor and divisions of household labor. Both types of logistic regression include lower and upper 95 percent confidence intervals (CI). Statistically significant results are often indicated in the following way: p<0.05=*, p<0.01=**, and p<0.001=***, except for chi-square values (a < 0.001).
Managers and Divisions of Labor
Family responsibilities have consistently been among the most important promotion barriers commonly identified by women in senior management themselves (e.g. Broadbridge, 2008). There is a burgeoning literature on women in management dealing with career patterns, job stresses and promotion barriers, but it has been largely divorced from any analysis of the domestic division of labor, the central indicators of family responsibilities. The importance of the paid work-family nexus and the desirability of balance between them are increasingly alluded to by both researchers and policy-makers, but with little empirical attention to the full division of labor involved. A basic issue is the management of time.
Virtually all in-depth studies of household relations have found that the domestic duties of women in heterosexual couples take significantly more time and energy than the domestic tasks assumed by men. While women who delay childbearing have made impressive gains in advanced education, absorption in family responsibilities appears to have continued to serve as a major barrier to job promotion (Cobb-Clark and Dunlop, 1999). Time series studies in Canada and the USA confirm that as women’s paid work hours have increased, their housework hours have declined slightly, whereas men have decreased their paid work hours slightly and increased their time ‘helping out’ in household work but continue to spend much less time at it than women, still spending more time instead on paid work and in free-time activities (Marshall, 2006; Parker and Wang, 2013; Sayer, 2005). In addition to differences in hours, women still bear the major responsibility for coordinating and performing most work inside the home that is closely related to children, and work that is performed regularly with little discretion, such as cooking, cleaning and daily child care (Breen and Cooke, 2005). This inequity is particularly intense for women who attempt to combine full-time jobs and child care.
In any estimates of unpaid work, substantial portions remain hidden in both the household and community because people continue to regard many of these activities as intrinsic parts of everyday life rather than ‘work’ (Macredie and Sewell, 1998: 8). In comparison to paid employment, general housework and child care tasks are more diffuse labors that respond to different rhythms than paid clock-time and may often overlap. It is also important to recognize that men have typically overestimated their domestic labor time, while women underestimate theirs (Berk and Berk, 1979). 2
The 2004 and 2010 surveys provide the first national-level estimates of the time devoted to paid employment, general housework, child care and total work time by male and female employees who report they have managerial roles and those who do not. Table 1 summarizes the basic findings for all managers in comparison with other employees.
Gender, average work time and primary household work responsibility, managerial and non-managerial employees, Canada, 2004–10.
Sources: 2004: WALL I (2005): N=4526; 2010: WALL II (2011): N=991.
= p<.05, **= p<.01, ***= p<.001.
Indicates Chi-Square significant at p<.001.
The survey findings suggest that in the 2004–10 period in the non-managerial labor force, employed women have averaged over five hours less than men per week in paid work, reflective of more women doing part-time jobs. Employed women continue to do most of the housework and child care, still spending nearly twice as much time at these tasks as men. A slightly different pattern is found among managers. Nearly all female managers hold full-time jobs, but they also spend an average of about five hours per week less than male managers at their jobs. Cumulatively, this means that male managers are gaining more than 10 percent more paid work experience than women managers (compare Bertrand et al., 2010). Conversely, women managers – like other women employees – generally spend at least five hours more than men on unpaid household work. In addition, among those with children, women managers spend upwards of 10 hours per week more than male managers on child care. Overall, women managers have been spending more total time working but less time in paid work, although this gap may be narrowing recently (70 total hours versus 65 hours for male managers in 2010). The gap between male and female managers appears to be even greater in terms of responsibility for household work. While many men and women express support for the value of equal sharing of household work, male managers remain much more likely (42%) than female managers (9%) to leave primary responsibility for organizing and performing such work to others. 3 In terms of both time constraints of and responsibilities for unpaid work, male managers are thereby enabled to use their longer employment hours and discretionary time for more extensive attention to job assignments and social networking, while women managers’ greater work hours and responsibility tend to be hidden and consumed in the household.
The general gender differences obscure greater ones within different household forms. The larger 2004 WALL survey, with more information on households, permitted comparisons between male and female managers living in different household forms, including households in which both partners were employed full time, households in which both partners were employed full time with children under 18, and households where both partners were employed and without children. Comparisons among these types of households for managers are summarized in Table 2.
Gender, household form, average work time and primary household work responsibility, full-time managerial employees, Canada, 2004.
Source: WALL I (2005) (Employed F/T with F/T spouse, N=427; Employed F/T with F/T spouse and children under 18, N=230; Employed F/T with F/T spouse and no children, N=197).
= p<.05, **= p<.01, ***= p<.001.
Indicates Chi-Square significant at p<.001.
When comparison is limited to managerial households in which both partners are employed full-time, the same pattern of greater paid work hours for men and greater unpaid housework hours for women still appears, in spite of the stronger case that women employed full-time may have for equal division of domestic labor. But, as suggested by a rare case study of a U.S. manufacturing company (Moore et al., 2007), this general comparison for all managers with partners employed full time masks the strong time gap in child care. If one looks only at those households with children under 18 years of age in which both managers and their partners are employed full time, the time devoted to child care by women in these managerial households is as great as the time spent at their paid jobs, around 40 hours per week. Men in such households also devote substantial time to child care, an average of about 24 hours per week, but much less than women, and they spend as much time at paid work as other male managers. In these managerial households with children, women managers indicate their total work time, including paid work, housework and child care, is around 95 hours per week while men estimate theirs at around 80 hours per week. On average, these women managers spend seven hours less per week in paid work and 23 hours more in unpaid work than male managers do. Aside from the likelihood that male managers’ child care continues to be more passive and less demanding than that done by women managers, the difference of over 15 hours in both child care and total work hours suggests a very large barrier to women’s equitable participation in paid employment and career advancement.
Furthermore, Table 2 shows that, in managerial households with both partners employed full time but with no children under 18, the total work time for both men and women is very similar. Men may spend a bit more time in paid work and women spend a bit more time in housework but, without primary child care, average total work time is under 60 hours per week for both men and women. Comparison of the entire labor force in terms of different age cohorts of partners without children in the 2004 survey finds that the sex differences in paid work hours are somewhat less among younger cohorts because women’s paid hours are increased. Further, the survey finds that differences in household work hours are much less among younger cohorts because women have reduced their household work hours substantially to cope with greater paid work hours. For men, paid work hours are similar for older and younger cohorts while household work may have increased slightly. Further analysis of the work time of men and women employed full time and living alone finds their distributions of total paid and unpaid work time are nearly identical, around 55 hours per week. For women managers, as for other women living with male partners, increasing employment hours have required reducing housework hours given the more limited changes in their male partners’ paid and unpaid work time. However, even among managers without children, the primary responsibility for household work is much more likely to be held by women. Shared responsibility may be more common in such households but men are much more likely (34%) than women (10%) to leave primary responsibility to others. The bottom line is that women managers continue to face major systemic barriers to job promotion that are rooted in the inequitable domestic division of labor and located especially in the relations between men and women over the care of children.
Wajcman’s (1998) pioneering British case study – one of very few to directly examine division of paid and unpaid labor among female and male managers – found that women who made it into senior management positions were much more likely than their male colleagues to have become divorced, separated or remained single. The WALL 2004 survey also found that, among managers, women were more likely to be unmarried, separated or divorced than male managers. As one of Wajcman’s senior woman managers asserted: ‘There is a constant tension between the demands of a management job and the demands of children, particularly when the job requires working outside conventional office hours’ (1998: 146–147).
Young women generally are delaying family formation and marriage to pursue their careers (Fortin and Schirle, 2006; Mills et al., 2011). Given the inequitable burden of unpaid household work, the few women who have made it to top corporate positions have undoubtedly had to make use of nannies and maids to significantly diminish their own household duties. There has been some critical discussion within the corporate world of dealing with a ‘long hours culture’ indifferent to family issues (Chatzitheochari and Arber, 2009; Liff and Ward, 2001), but the practical focus has been largely on flex-time to enable women juggling more paid work with child care rather than on documenting or addressing women’s much longer unpaid hours per se (Baker, 2009).
Women managers’ longer total work time, greater household responsibilities and more limited discretionary time leave them with relatively little time and energy to engage in career development activities. When asked about their preferred job hours in these surveys, the majority of male managers expressed preference for their current hours whereas most women managers wanted to work less hours. Rather than indicating women’s lower career aspirations or lesser job commitment, this may be an expression of a wish to resolve the inequitable binds of family responsibility.
The Rise of Women Managers
Canadian census-based analyses have found that all managerial employees made up about 3 percent of the employed labor force in 1971, that managerial employees doubled to over 6 percent of the labor force in 1981 and increased to around 9 percent in 1991 (Lavoie and Roy, 1998). The most recent available census found that about 10 percent of the entire Canadian labor force over 15 years of age were in official managerial occupations in 2006 (Statistics Canada, 2006).
Data from the 1982, 2004 and 2010 surveys used in this article permit more detailed and recent tracking of Canadian managerial employees. Thus, we are able to track the following groups: those who reported that they held a managerial role; those who held a census-based official managerial occupational title; those who held a census-based managerial occupation and also exercised actual managerial authority over others; 4 and finally the managerial level (top; upper; middle; lower) of those who exercised managerial authority. The 1982 and 2004 survey figures for official managers as a proportion of the entire labor force are very close to the 1981 and 2006 census figures; 2011 census figures are not yet available. Table 3 summarizes the basic patterns. The proportion of employees in the labor force reporting they had any designated managerial role doubled in the period from 1982 to 2004, to over 20 percent. The proportion of employees that had official managerial occupations rose from 5 percent in 1982 to 13 percent in 2004. The proportion of managers who had official managerial jobs and actually exercised managerial authority over others was around 2 percent of the labor force in 1982, rising to about 7 percent in 2004. Finally, the proportion of managers who had top or upper managerial jobs with direct authority was about 1 percent in 1982 and 3 percent in 2004. There appears to have been little significant increase between 2004 and 2010.
Managers as proportion of employees, Canada, 1982–2010.
Sources: 1982: Clement and Myles (1994); 2004: WALL I (2005); 2010: WALL II (2011).
Note: Z-scores comparing differences from prior survey: *= p<.05, **= p<.01, ***= p<.001.
Those self-identified as managers according to census job descriptions continue to constitute around half of all who report actual managerial responsibilities. Others with managerial occupational job titles indicate that they do not exercise line authority, performing administrative staff functions for managers who do. Our further analysis will focus on those who report definite managerial roles, with some attention to those with official managerial titles who exercise higher levels of authority over other employees. 5 In any case, we can conclude that the proportion of managerial jobs in each of these terms increased significantly during the past generation but not in the past few years.
Before looking specifically at women’s participation in management, we should note that part-time employment has increased considerably in the past generation, especially for women who now make up about two-thirds of the part-time labor force (Ferrao, 2010). But managers with direct authority for the operational division of labor have remained largely full-time employees; according to our most recent survey, virtually all male managers with direct authority and around 90 percent of female managers were full-time employees in 2010. Secondly, as the managerial hierarchy has expanded, more managers themselves are managed. In 1982, around a third of all managers had no one above them to whom they were required to report; in 2010, the proportion had declined to around 5 percent. Similarly, the proportion of managers who reported that they had discretionary control of the design of their own work declined from nearly 90 percent to about two-thirds. These changes may be reflective of both a more collective form of labor process with greater consultative decision-making and a general inflation of the number of intermediate managers subordinated to other managers.
The most substantial general change in the labor forces of advanced capitalist economies in recent generations has been the increasing participation of women in paid labor. The suddenness and velocity of the shift from homemaker to lifelong dedication to paid employment has been characterized as “little short of revolutionary” (Esping-Anderson, 2009: 7). Women made up a small minority of the paid labor force in Canada from the 1890s to the 1920s, about 13 percent (Statistics Canada, 2004a). These were mostly young, single women who left paid employment at marriage. During the 1939-45 war economy, more married women re-entered the labor force. With the expanding post-war economy, female participation steadily increased to about 27 percent of the labor force in 1961 (Statistics Canada, 2004a) and 31 percent in 1971 (Statistics Canada, 2004b). In 2011, women constituted nearly half (47%) of the employed labor force (Statistics Canada, 2012a).
Growing dependence of families on purchased commodities to make ends meet, women’s rising educational attainments and career aspirations, and the demands of the women’s movement all probably contributed to this sustained growth of the numbers of women in paid employment. Since the mid-1970s, the major growth has been among mothers with children. For example, between 1976 and 2009, the labor force participation rate for women with children under three years more than doubled, from 27 percent to 64 percent (Statistics Canada, 2011). Women’s general participation rate reached almost 72 percent by 2009; in comparison, men’s participation rate peaked at 78 percent in 1981, and has fluctuated in the mid-70 percent range since then (Statistics Canada, 2004b, 2011). Recent survey results show that the overall participation rate in paid employment has risen to around two-thirds of the entire working-age population (Statistics Canada, 2012a). Women, whatever their marital status, have become almost as likely as men to be in the employed Canadian labor force.
Research on women’s paid employment in general has found that women continue to be concentrated in less prestigious, lower paid jobs with less control over their work than men (Reskin and Bielby, 2005; Williams et al., 2012). Armstrong and Armstrong (1994: 67) observed: ‘the dramatic increase in female labor force participation has not freed women from being concentrated in a limited number of sex-typed jobs that frequently parallel those they perform in the home.’ The increase of women in the labor force has not translated into the equitable presence of women throughout different occupations or within different levels within occupations. Women and men have still tended to be segregated into different types of jobs and occupations (Evans, 2002) and, when in the same occupation as men, women have still found themselves at subordinate levels within the same organization (Lyness and Schrader, 2006).
Women have increased their representation in several professional fields recently, for example becoming majorities among doctors and dentists (Ferrao, 2010). Compared to a generation ago, young women today are much more likely to participate in the labor force, to work more paid hours, and to pursue higher prestige occupations with higher qualifications (Chung, 2006; Evans, 2002). In addition, the ‘gender pay’ gap has slowly been narrowing in Canada. In 1997, women full-time paid workers earned about 77 percent of men’s earnings; by 2011, this figure had climbed to almost 83 percent (Catalyst, 2012a). Cumulative lifetime earnings differences remain greater (Zeytinoglu and Cook, 2008). But as female university graduation rates continue to surpass men’s and the economy shifts increasingly from manufacturing to service jobs, job segregation and the cumulative pay gap do appear to be slowly narrowing (Chung, 2006).
Profiles of Women Managers
Women’s growing participation in paid employment and the expansion of managerial positions have surely brought more women into management. We will document the growth of women as managers in general and at different managerial levels in the Canadian labor force, and then assess the extent of gender segregation among managers.
Until the 1930s, women must have constituted a tiny fraction of managers, since women made up less than 15 percent of the paid labor force and most women workers were young and not encouraged to pursue managerial careers. At least until fairly recently, the growth in women’s share of management jobs has not matched their increasing share of the labor force, and, within management categories, they have been concentrated in middle management, small companies or lower levels (Armstrong and Armstrong, 1994).
However, the large gain in women’s general labor force participation rate since the 1960s has also led to a significant increase in their participation in managerial roles. Table 4 provides 1982, 2004 and 2010 survey estimates. By the early 1980s, women took up about a third of all self-reported managerial jobs. They held about a quarter of official managerial jobs with direct authority over other workers but only 10 percent of upper managerial jobs. Compared to women’s 37 percent share of the employed labor force, they were still clearly underrepresented in positions of higher managerial power. By 2004, there had been significant increases at all of these managerial levels, approaching parity with their numbers in the general labor force, except at the higher levels. Both census data (Drolet, 2011) and our surveys find that women still only made up around a third or less of senior (or top/upper) managers. Both census analyses (Ferrao, 2010) and our survey data suggest only marginal changes in more recent times. 6 The proportion of women managers with direct authority remains lower in larger organizations. In organizations with over 50 employees, two-thirds of such managers are men; in organizations with fewer than 10 employees, two-thirds of these managers are women.
Women as proportion of managerial employees, Canada, 1982–2010.
Sources: 1982: Clement and Myles (1994); 2004: WALL I (2005); 2010: WALL II (2011).
Note: Z-scores comparing differences from prior survey: * = p<.05, ** = p<.01, ***= p<.001.
The general management participation ratios also disguise the still very substantial sex segregation that persists. Women have continued to be overrepresented in jobs and industry sectors with lower recognized skill requirements and productivity levels (Conference Board of Canada, 2011). According to our 2004 survey, in goods producing (i.e. extractive, manufacturing and construction) industries, men comprised over three-quarters of employees and comparable proportions of managers. In the service sectors (i.e. education, health, social and personal services) where female employment was dominant (75% or greater), men still held nearly half (47%) of the top management posts. 7
Furthermore, as Table 5 shows, men are still very unlikely to be managed or supervised by women. 8 This is the case both among non-managerial employees generally and among managers themselves. In 1982, only about 8 percent of all male employees had a female manager or supervisor, and very few male managers did. By 2004, the proportion of male workers in general who had female managers or supervisors had increased significantly but remained less than 20 percent. The proportion of male managers who had female supervision also increased as female managers became more prevalent, but few had reached the top managerial levels so that less than 10 percent of top male managers had female bosses. The 2010 survey results suggest no significant further increase in women managing men since 2004 (cf. Cohen et al., 2009). Around 80 percent or more of male managers and male workers in general continue to be managed by other men.
Gender structure of managerial relations among employees, Canada, 1982–2010.
Sources: 1982: Clement and Myles (1994); 2004: WALL I (2005); 2010: WALL II (2011).
Note: Z-scores comparing differences by sex of respondent from prior survey: *= p<.05, **= p<.01, ***= p<.001, n.s. not significant.
Among women, more substantial changes have occurred. In 1982, only about 40 percent of women with any self-reported managerial role had a female manager or supervisor; by 2010, about 60 percent did. More strikingly, the vast majority of women’s gain in managerial or supervisory authority has been exercised over other women. Less than 20 percent of women with official managerial titles were supervised by other women in 1982, increasing to nearly half in 2010. Among women managers with direct authority over others, the increases were from virtually universal supervision by men in 1982 to possible majority supervision by other women in 2010. The numbers of top women managers in the survey remain too small for statistical reliability but also suggest marked increase in their supervision by other women.
Male dominance of the overall managerial hierarchy appears to have declined significantly over the 1982–2010 period as more women have assumed managerial roles. Some might argue that there is now a ‘critical mass’ of women managers in the paid workforce who will pave the way for promotion of women into top management. In fact, in 2009 women held about 30 percent of all senior management positions (including executives, presidents and chief executive officers), an increase from about 20 percent in 1987 (Ferrao, 2010). However, as Table 5 suggests, the vast majority of women in senior management are still managing other women and remain very unlikely to be managing men. In the male-dominated goods producing sectors, virtually all top male and female managers who have superiors report to men. In female-dominated service industries, about three-quarters of women top managers do report to other women, but 80 percent of male top managers report to other men. Even in female-dominated sectors, male managers are still very unlikely to be subordinate to women managers.
The top managers with direct authority in the 1982, 2004 and 2010 national surveys are still representative of positions below the pinnacle of managerial power in large organizations. In private industry, corporate executive officers hold the ultimate positions of economic power, exercising the most direct control of the largest economic organizations. These ‘captains of industry’ constitute much less than 1 percent of the labor force and their numbers are too small to access through general population surveys. Selection for these positions has been based largely on presumed capacity to enhance profitability, with relative indifference to employment equity measures. One might expect that they would be a last bastion of patriarchal capitalism.
Considerable prior empirical research has been devoted to tracking the gender composition of corporate boards of directors both in Canada and internationally. 9 In the early 1990s, Burke (1994) estimated that women made up around 5 percent of corporate board members. Table 6 summarizes the basic findings of the Catalyst Census of Women Boards of Directors of Canada since 2001. In 2001, women held 10 percent of all corporate board seats. By 2011, this proportion had increased to about 15 percent.
Women in corporate management, Canada. 2001–11.
Note: *Data for 2002; **Data for 2012.
The Catalyst surveys have also found that the proportion of corporate boards with any women members increased from 48 percent to 60 percent between 2001 and 2011. The vast majority of board seats are still held by men who continue to control channels of upward corporate promotion. The proportion of corporate officers who are women increased from 10 percent in 2001 to about 18 percent in 2011. Who has corporate line authority is a critical component for board promotion. The proportion of women in corporate officer positions – in the ‘executive pipeline’ – is now very similar to the proportion of women who supervise men in the labor force generally (i.e. around 20 percent). The door to the executive suite may still be largely closed to women, but the trickle appears to be increasing. The proportion of women CEOs in the largest Canadian corporations has tripled from 2 percent to 6 percent within the past decade.
In sum, the recent survey evidence suggests that women managers generally are approaching parity with their nearly equal representation in the entire Canadian labor force, that women’s representation in senior management positions remains lower at around 30 percent and that even women in upper management are unlikely to be managing men. The recent increases in corporate board seats held by women and numbers of female chairs of boards, while still very small proportions, appear to be indicative of a progressive trend. But the sex segregation within the managerial authority structure generally and the persistent underrepresentation of women in senior and top management confirm that the path to the top of the managerial hierarchy still retains systemic barriers or glass ceilings.
Primary Factors Influencing Women Managers’ Promotion
So what are the significant factors related to women managers’ promotion up the managerial hierarchy, and what is the relative strength of their influence? We will select several possible factors suggested in prior studies and then use our survey data to estimate relative effects. We will consider both factors presumed to be generally positively related to higher managerial promotion as well as factors posited as barriers to women’s promotion.
Numerous organizational and motivational features have been pointed to in the literature on managerial promotion, but the factors that have most generally been found to be positively related to higher managerial promotion include educational attainment or formal qualifications as well as tenure in organization or employment experience (e.g. Sharabi et al., 2012). In credential-oriented economies, higher educational attainment has become increasingly recognized as a criterion for managerial promotion. The majority of managerial posts now require a post-secondary education and higher posts are likely to require more advanced credentials. Secondly, the more time one has spent in a kind of employment, the more job-related learning one is likely to have done and the more practical qualification and seniority one is likely to have for promotion (Livingstone, 2009). In terms of our survey variables, employees with greater formal educational qualifications and more job experience are posited as most likely to be promoted to higher management levels.
There is now a large literature attempting to identify barriers to women’s promotion into higher management. A recent review by Hoobler et al. (2011) summarizes major factors as: lack of genetic predisposition to take risks required in high stakes management; insufficient numbers of qualified women in the ‘pipeline’; discriminatory stereotyping of effective leadership as associated with male attributes; and incompatibility between job structures and needs to raise a family with consequent family-work conflict bias against women. 10
Genetic assumptions about women’s more submissive, nurturing nature become more difficult to defend as the career aspirations of women who achieve middle management are sustained (Catalyst, 2004; Wentling, 2003), social networking of female and male managers becomes more similar (Waldstrom and Madsen, 2007), and few individual differences in actual managerial behaviour are found between female and male managers (Anderson and Hansson, 2011).
Claims about insufficient numbers of qualified women in the managerial pipeline are also very doubtful. In Canada, women generally have increased their educational qualifications in recent decades to levels exceeding men (Statistics Canada, 2012b). Our 2004 and 2010 surveys provide estimates of the qualifications of managers of both sexes. These surveys have found that the general distribution of educational attainment among male and female managers is now quite similar, with around two thirds of both sexes having completed a post-secondary degree. There is now little plausible evidence to indicate a lack of qualified women for promotion into senior management.
However, the evidence presented in Table 6 is certainly consistent with persistence of the original ‘glass ceilings’ argument, that in spite of equal qualifications women are being excluded from higher levels of the managerial pipeline (see Frenkel, 1984). Surveys of women in management still identify male-dominated organizational cultures and gender stereotyping as significant barriers that continue to negatively influence women’s own perceptions of their prospects for managerial advancement (e.g. Catalyst, 2007; Haveman and Beresford, 2012; Litzky and Greenhaus, 2007). If gender discrimination were eliminated in managerial hiring, then the sex of one’s manager or supervisor would have no significant effect on promotion. Women with woman supervisors and women with male supervisors would have equal chances for promotion. Since we know the sex of the respondent’s supervisor in our surveys, we can at least make limited estimates about the extent of this effect.
Beyond paid workplaces, women’s greater family responsibilities are the most commonly mentioned barrier to their managerial promotion. But there has been relatively little empirical research on the topic. General studies of employment and family time-use have found that women lose bargaining power relative to men through opting for part-time employment and continuing to bear a greater burden of domestic labor, irrespective of employment status or educational level (e.g. Breen and Cooke, 2005). A recent survey of call centres has found that even for women with similar career aspirations and qualifications to men, domestic responsibilities create obstacles before they reach the glass ceiling for managerial roles, and contribute thereafter to reinforcing their concentration in more intensive, lower status work (Scholarios and Taylor, 2011). A large-scale Danish survey of promotion into top management jobs examined individual educational qualifications and work experience as well as family profiles (Bell et al., 2008), and found that education variables have significant effects for both women and men, while having a partner and children have positive effects for men but not for women. But neither this study nor most other recent studies of promotion into top management have considered division of household labor directly. As documented in the section on divisions of labor, women managers have a significantly heavier time commitment to unpaid household work and greater likelihood of primary responsibility for this work than men. Both of these factors are included in the following analysis.
So, positive individual factors to be included in this analysis are educational attainment and employment experience. Probable significant barriers for women are male sex of supervisor, unpaid household work time and family responsibility. The data base used is the large 2004 population sample of Canadian employees (N=4526). Statistical analyses include correlation coefficients and logistic regression odds ratios. The main dichotomized dependent variable for the regressions is attainment of a top or upper management position. The findings for main factors related to attaining these positions for women and for men are summarized in Tables 7A and 7B, respectively.
Odds ratios for effects of job experience, educational attainment, sex of supervisor and family responsibility on promotion to top or upper manager, female employees, Canada, 2004.
Source: WALL I (2005).
Odds ratios for effects of job experience, educational attainment, sex of supervisor and family responsibility on promotion to top or upper manager, male employees, Canada, 2004.
Source: WALL I (2005).
The crude odds ratios in both tables indicate that, as in many prior managerial promotion studies, employment experience and educational attainment have significant direct effects on attaining upper managerial positions for both men and women. Seniority is very important, reflective of the time needed to climb managerial ladders. Women with more than 17 years of job tenure are about four times as likely to attain upper managerial posts as those with less than four years; men are about three times as likely. Men with bachelors or graduate degrees are about three and a half times more likely than those without post-secondary qualifications to reach upper posts. But for women, only professional graduate degrees provide a significant advantage for promotion to upper management. In terms of barriers, having a male supervisor is about twice as likely to be related to promotion for both men and women, reflective of male dominance of upper management. However, it should be noted that no significant difference has been found between male or female supervision as a factor in promotion in the female-dominated service sectors. With regard to housework, hours of housework per se do not appear to have a significant effect on promotion to upper management. As summarized in Table 1, there is a difference of at least six hours more per week in the housework time of women. This is consistent across all levels of management. Men in upper management do not claim to do much less housework than other men; nor do upper management women claim to do much less than other women. The zero-order correlation between housework hours and management level for women employees is very low and not statistically significant (r=.003, p>.885). However, findings for division of family responsibility are much different. There is no significant effect of greater family responsibility on men’s managerial promotion. But for women, those who usually rely on others to do household work or share this work equally are twice as likely to reach upper management posts.
The adjusted odds ratios, also presented in Tables 7A and 7B, indicate that the effects of all four of these factors remain consistent when the effects of the others are considered. Women with more than 17 years of job tenure are most likely to be promoted to upper management (adjusted odds ratio=4.758, p=.001); the pattern also remains similar for men (adjusted odds ratio= 2.955, p=.001). In terms of educational qualifications, the effects on promotion are even stronger for men when the other factors are considered, with those having a bachelor’s degree being at least five times as likely to be promoted (adjusted odds ratio=5.889, p=.001). For women, professional graduate degrees continue to be significant factors (adjusted odds ratio=2.622, p=.050). But these adjusted odds ratios suggest that staying in the job and accumulating seniority may be a more distinguishing factor than educational qualifications for most women, whereas post-secondary qualifications may be more important than seniority for many men’s promotion.
Adjusted odds ratios for the effect of sex of supervisor on promotion indicate that the dominance of males in upper management continues to be a significant influence on promotion for both men (adjusted odds ratio=2.720, p=.005) and women (adjusted odds ratio=1.763, p=.005). The finding that the adjusted odds are still greater for men than for women in spite of women’s now equivalent educational qualifications confirms the persistence of gender discrimination against women in the promotion process, in spite of progress in female-dominated service industries.
Adjusted odds ratios for division of family responsibility continue to find that women who report that someone else always or usually performs household work are twice as likely to be in upper managerial positions (adjusted odds ratio = 2.379, p=.011) than those who usually take it on themselves. Women claiming that others share responsibilities equally also benefit (adjusted odds ratio=1.906, p=.003). Having primary responsibility for organizing household care is a very important barrier to managerial promotion of women. Once more, division of family responsibility is not a significant factor for men, except for the tiny number who usually bear the primary burden. Again, further analyses including unpaid household work time do not find significant time effects on managerial promotion for either men or women. Some research indicates that women may choose different paid work-unpaid work time distributions depending on their family situation (Reynolds and Aletraris, 2007). But it is the burden of primary household responsibility more than the greater unpaid hours per se that appears to be the strongest barrier to promotion for those who are compelled to perform these tasks mainly on their own. Women managers may become consummate time jugglers but family responsibility is still held against them at higher managerial levels.
Further studies with comparable data over time and in other countries are needed to confirm these patterns and inferred trends. But on the basis of available evidence we suggest that the rapid increase in women’s sustained participation in paid employment and increasing seniority, combined with their increasingly similar educational attainments to men, have enabled them increasingly to attain lower management positions. Women employees have become adept at juggling the time binds of greater unpaid housework in climbing managerial ladders. Women are still predominantly managing other women and they are still largely excluded from the highest rungs of these ladders. The primary obstacles at this point appear to be the combined weight of glass ceilings in their paid workplaces maintained by male senior managers and, at least as importantly, the binds created by family responsibility without equitable support from partners and others in their households.
Conclusion
There is nothing inherent in the logic of capitalist production per se that ordains men shall retain dominance of the corporate boardrooms at the top of the managerial hierarchy. Indeed, given the overriding preoccupation with the profit imperative, if women managers prove more profitable, their numbers as top executives are likely to continue to increase and the remaining major barrier of inequitable household work and child care responsibility could finally be significantly adjusted to maximize profits, however resistant men remain to assuming fairer shares. In times before women were represented in substantial numbers in managerial positions, any number of claims were made about their lack of capability or motivation to perform effective managerial roles. At this stage, differential effects of women managers’ leadership on the general performance of their organizations can be assessed directly. A few studies have found that large corporations with greater representation of women in senior management have higher levels of organizational performance on a variety of indicators, including operating margins and growth in earnings (Catalyst, 2012c; Desvaux et al., 2008; Welbourne et al., 2007). A US study (Frink et al., 2003) that examined compositional effects of higher proportions of women generally in organizations found a relationship between an increasing proportion of women and organizational productivity and profitability up to a point; but a more balanced gender composition, rather than proportional dominance by either gender, was associated with higher organization performance. Beyond such exploratory correlations, little theoretical or empirical study has yet been done on the combined individual and compositional effects of women managers on various aspects of organizational performance. In any case, if all else were equal, the accumulation of findings on the association of higher proportions of women in senior management with profitability in itself would be an incentive for increasing these proportions up to parity.
The popular demand for access to higher education and decent jobs has now generated an ample supply of qualified women as well as men at virtually all levels of the job market. Both persistent underemployment generally and the gross underrepresentation of women in top management can now be appreciated as substantial wastes of talent (Livingstone, 2009; McKee-Ryan and Harvey, 2011). Policy focus on further educational reforms is likely to encourage a renewed educational arms race toward greater overqualification for the current job structure and still greater waste of talent. Affirmative action measures focused in paid workplaces can facilitate marginally fuller labor force participation by women at most levels (e.g. Clarke, 2011). Unfortunately, most such measures continue to ignore the actual facts of the inequitable division of family responsibility, as well as the extent of the glass ceiling in senior management.
Optimists may suggest we are witnessing a real gender-based managerial revolution through the growing participation of women in organizational management. The most progressive organizational programs have begun to address the constraints of women’s unpaid work time somewhat more directly by offering flexible maternity leave and return-to-work provisions, child care services, flexible work hours and family assistance programs (e.g. Baker, 2009; Catalyst, 2010). In terms of macro-organizational factors, some countries that have enacted more equitable childbirth policies and extensive child care provisions have enabled greater representation of women up the managerial hierarchy (Catalyst, 2010; Wrede et al., 2008). A growing number of countries are establishing senior management quotas (Terjesen and Singh, 2008). Norway is the current leading case in this respect, with women making up over 40 percent of corporate board members in response to equality act amendments since the mid-1990s. In anticipation of the emergence of a ‘knowledge-based economy’, governments and business firms frequently declare intentions to more fully tap into the rich experiences, skills and abilities of previously excluded social groups such as women, particularly for management roles (Dalton and Dalton, 2010). Again, it is arguable that sustained profitability with more talented women managers will serve to break the glass ceiling to top corporate jobs.
However, at this point, women still make up less than 10 percent of CEOs even in Norway (Grant Thornton Business Report, 2012). The persistent exclusion of women from the most senior positions suggests that senior male corporate leaders who still dominate corporate boards and senior managements are continuing to react, as most male leaders have historically, in terms of their own immediate vested interest in protecting patriarchal control of the central levers of globalizing economic power, and remain indifferent to the inequity of unpaid ‘women’s work’ (Acker, 2004).
Recent research suggests that greater proportions of women in management may be associated with less sex segregation generally in organizations (e.g. Huffman, 2013; Stainback and Kwon, 2012) But sex segregation of occupational sectors and upper managerial authority structures remains very extensive; the vast majority of gains continue to be in women managing other women in female-dominated service industries and in smaller organizations. Both our data in Tables 4 and 5 and other surveys (Cohen et al., 2009) suggest ‘stalled progress’ in recent times in North America. Neither organization-centered affirmative action measures nor well-intentioned societal policies are likely to produce systemic change toward gender equity if upper management positions are merely ‘feminized’ in terms of marginally greater numbers.
Actual employment practices can still effectively ignore the inequity in necessary unpaid housework and child care responsibilities, and merely enable women managers to work longer total paid and unpaid hours or forego marriage and childbearing. More sustainable gains in equitable engagement throughout managerial hierarchies will likely require further mobilization by and on behalf of women to recognize and reward the essential labors of child care and housework in relation to the sphere of paid work. Reforms of the job structure (e.g. work-time redistribution, workplace democratization), alternative work-time arrangements for women and men (Kalev, 2009) and fostering gender equality at home offer greater prospects for long-term resolution of talent waste (see Livingstone, 2009; Sirianni and Negrey, 2000). To paraphrase Barbara Bergmann (1986), gender equity in the boardroom requires gender equity in the household. While Sheryl Sandberg (2013), CEO of Facebook, now exhorts women to work harder and aim higher, she at least recognizes that a truly equal world would be one in which men run half the homes – and she apparently has a husband willing to meet her halfway. Women’s trickle into the boardroom otherwise represents persistent inequity outside it in terms of both the many women who shoulder too much family responsibility and the many men who, in the short-term, benefit from this inequity.
Footnotes
Acknowledgements
We are grateful to Margrit Eichler and John Myles, as well as three anonymous reviewers, for their suggestions on prior versions of this article. Michael Mindzak and Julia Lowe offered assistance with the review of literature. The Centre for the Study of Education and Work at OISE/UT provided the auspices for most of the research work.
Funding
The Social Science and Humanities Research Council of Canada funded the research network that conducted the 2004 survey. The Canada Research Chairs Program provided funding for the 2010 survey.