Keywords

Concerns about the prevalence of underemployment have grown with rising educational attainments and economic slowdowns in most industrialized countries. However, women have been facing underemployment for some time. Familiar terms abound to describe the experiences unique to women in the paid labor market (e.g., glass ceiling, sticky floor, old boys club, pay equity, occupational ghetto, pink collar, double day, and second shift). These terms suggest the reasons for women being more at risk of underemployment. Systemic discrimination, occupational and job segregation, wage inequality, the sexual division of unpaid labor, and more limited returns on education and experience all challenge a woman’s ability to achieve labor market equity with her male counterparts because they prevent the full usage and recognition of her knowledge, skills, and abilities (KSAs).

These issues have received considerable attention in the popular press and the academic literature over the past five decades. However, they have yet to be discussed under the unifying theme of underemployment. This chapter presents a complete picture of the underemployment risk faced by women through an assessment of their experiences in the formal, paid labor market. In particular, based on the conceptualizations of Feldman (1996) and Livingstone (2009), I discuss three dimensions of underemployment: pay-based, time-based, and skill-based.

The chapter begins with an overview of several important trends that act as a backdrop to women’s underemployment experiences. Next, the literature on the three dimensions of underemployment is reviewed. Ultimately, I conclude that the impact of domestic and reproductive work constrains women in their paid work experiences due to power relations in the home and work, ‘family friendly’ government and employer policies, gender ideology, and individual choice. The chapter closes with a discussion of challenges that still exist in documenting the underemployment experiences of women and highlights the fact that definitions of underemployment to date capture only paid work in the formal labor market and formal learning experiences. This negates the lived experience of most women in which unpaid work and informal learning experiences play a critically important role.

Labor Force and Educational Trends

An examination of the prevalence of underemployment among women is particularly relevant given the trend toward increased female labor force participation and increased attainment of secondary (i.e., high school) and tertiary (i.e., college) levels of formal schooling (Krahn & Lowe, 2002). Together, these trends put more women in the workforce with advanced qualifications and place them at greater risk for underemployment.

The participation of women in the formal sectors of the paid labor force has increased dramatically in the last 60 years. In Canada and the USA during that time, the labor force participation rate (LFPR) for women doubled to two-thirds of the population (Fullerton, 1999; Bureau of Labor Statistics, 2010; Statistics Canada, 2010). Women account for 40% of the global workforce (Catalyst, 2010a), and in recent decades the labor force growth has been higher for women than men in all regions of the world except Africa. These increases are most notable among women aged 25–35, who tend to be married with young children.

Explanations for the increase of women’s LFPRs across industrialized countries include declining fertility rates and increased female control over fertility (Hakim, 2004; Bergmann, 2005), equal opportunity legislation and other government and employer policies, the erosion of the family wage and subsequent need for dual-income families (Duffy & Pupo, 1992), the rise of the service sector and more ‘female friendly’ work settings (Hakim, 2004), and changing social norms around maternal roles and women’s career aspirations. Much of the increase in female LFPR has been in part-time or other nonstandard employment (Cranford, Vosko, & Zukewich, 2003). This work has been consistently described as more precarious than traditional full-time work.

Coupled with the rising participation of women in the labor force is their increasing educational attainment. Though both men and women of Generations X and Y have more formal education than the generations preceding them, the rise has been sharper for women. Among OECD countries, 80% of women aged 25–34 years have attained upper secondary education compared to 63% of women aged 45–54 years. Seeing similar gains, the proportion of women aged 25–34 with tertiary-level qualifications has surpassed that of men (37% vs. 30%; OECD, 2009).

Also relevant to the discussion of underemployment is the gendered division of domestic labor. Women still perform the majority of the unpaid reproductive and caring labor of the home (Hochschild, 2003; Fuwa & Cohen, 2007). Theories of power relations link this sexual division of domestic work to the financial resources of the spouses (Ross, 1987), where men’s share of female-type tasks is inversely related to the ratio of men’s earnings to total family income (Estes, Noonan, & Maume, 2007). The sharing of domestic work becomes more equal as the income of the secondary earner (typically the female spouse in heterosexual partnerships) increases (Anxo, 2003; Estes et al., 2007). Men also perform a greater degree of routine housework when their female spouses have longer employment histories, though relative income and hours of work are stronger predictors (Cunningham, 2007).

Situations of a completely egalitarian division of household labor, though, are rare regardless of the woman’s income level. Women who enter full-time paid employment spend the least amount of hours on domestic tasks when compared to women working part-time and full-time homemakers, but the work that is dropped by these women is not completely compensated for by the rise in men’s contributions (Bianchi, Milkie, Sayer, & Robinson, 2000). As well, women retain more complex tasks such as the organization and planning of household schedules (i.e., appointments, external activities, and daily routines) while men pick up specific and isolated tasks, often only when asked (Daly, 2002). The lack of movement toward gender equity in the home is important to the following discussions of underemployment because domestic, caring, and reproductive work – the double day or second shift (Hochschild, 2003) – is often put forth as a barrier to women’s equity in the labor market.

Three Categories of Underemployment

Pay-Based Underemployment

According to human capital theory, the overall male–female earnings differential is due to differences in the educational attainment and labor market experience of men and women. Therefore, the gap should be closing over time as women achieve higher levels of secondary and tertiary schooling (Krahn & Lowe, 2002) and experience fewer and shorter labor market interruptions. A growing number of women with young children return to the paid labor force rather than staying at home (Hochschild, 2003). More women also have access to parental leaves that give job protection and reduces their need to quit and drop out of the labor force when their children are born. Similarly, women have greater access to and ability to pay for the growing expense for child care for their young children (CBC, 2005; Baccari, 2010) so they can return to work (Duffy, 2007). Finally, strides have been made to eliminate intentional discrimination, and pay and employment equity programs have been created to address systemic discrimination.

However, women still experience pay-based underemployment. According to Feldman (1996), workers face pay-based underemployment if they earn 20% less than a previous job or, for new graduates, 20% less than the average of their graduating cohort. If this definition is extended to a comparison of the male and female wages in the same occupations (therefore with the same human capital requirements), women experience pay-based underemployment across most occupational groups. In 1985, Tipps and Gordon found that women were twice as likely to receive inequitable pay vis-à-vis males with similar human capital characteristics. The data in Table 6.1 from the 2006 Canadian Census present a similar picture. It presents the yearly salary by sex for full-time workers in the five highest- and five lowest-paying occupations in Canada and shows women’s wages to be less than 20% of men in many cases (see also Krahn & Lowe, 2002 and for US data, Bergmann, 2005).

Table 6.1 Median full-time earnings for selected occupations, by sex
Full size table

Bergmann (2005) and other dual market theorists (Krahn & Lowe, 2002; Phillips & Phillips, 1993) attribute the gender pay gap to labor market segregation. Specifically, industrial or horizontal segregation confines women to particular sectors of the labor market and occupational or vertical segregation confines women to the lower ranks of particular jobs within an occupation (Evans, 2002; Scott-Dixon, 2004). With respect to the former, women are over-represented in clerical, administrative, and service- and health-related positions (Bergmann, 2005; Legault, 2009). As such, in some ways men and women operate in separate labor markets that dictate different wages. Due to restricted size and scope and generally lower recognized skill requirements and productivity levels (Acker, 2006; Armstrong & Armstrong, 1994), the female labor market becomes crowded and wages become depressed (Bergmann, 2005).

Legault (2009) notes that the gender pay gap is particularly persistent in jobs that require the lowest educational levels and are the most highly gender segregated. There was a narrowing of the gap among the university graduates in her sample (24 to 20% from 1997 to 2008), while the rate was slightly slower for those with high school diplomas (30 to 27%) and practically stagnant for those with no high school diploma (39.4 to 39.2%). Legault also documents how equal access programs such as pay and employment equity are the most ineffective in low-skill, low-pay jobs because they tend to be nonunionized and in the private sector where such policies are not enforced.

Though the gender pay gap has been narrowing for highly educated women in male-dominated industries, it is still large. For example, female professors earned 82% of their male counterparts in 2005 (Canadian Association of University Teachers, 2010). And despite gains, women are still under-represented in SET (science, engineering, and technology) industries as a whole and particularly in the higher-status, higher-pay jobs within traditionally male-dominated industries (Wynarczyk & Renner, 2006). The liberal feminist strategy of obtaining the necessary human capital has not reduced the vertical and horizontal segregation of these occupations (Define, 1992) and this places women at risk for both pay-based underemployment and skill-based underemployment (discussed in the section below). Women made up almost half of the law students in the USA in 2007–2008, but in 2009 they represented only 32% of lawyers and 19% of law firm partners (Catalyst, 2010b). Women earned about 35% of MBAs in 2008–2009 in the USA and Canada and hold half of managerial, professional, and related positions, but only 13.5% of Fortune 500 top executives (Catalyst, 2010b) and 15% of corporate board members were women (Catalyst, 2010a; see Sealy, Vinnicombe, & Singh, 2009 for UK data). Women managers are most often middle or line managers and tend to manage women in female-dominated occupations (Coyle, 1995; Krahn & Lowe, 2002; Livingstone & Pollock, 2009), whereas most men are managed by other men.

In an account of women in the information technology industry, Scott-Dixon (2004) noted that women have made inroads only because of the feminization of many aspects of IT work. Women are not found in ‘hard’ technical jobs, but in work that requires ‘soft’ and helping skills stereotypically associated with women: user-testing, usability and design, call centers and help desks, Web design, and technical writing (see also Crump, Logan, & McIlroy, 2007). These feminized IT jobs are more likely to be contingent (though billed as flexible), lower-waged, benefit-free, and devalued as compared to ‘core’ IT work where men still dominate (Scott-Dixon, 2004). Work requiring high emotional labor is often underpaid vis-à-vis commensurate jobs (England, Budig, & Folbre, 2002). Even emerging forms of work such as new media are highly gendered despite work characteristics that might lend to easier work–life balance for women (Perrons, 2003), though Chasserio and Legault (2010) note that the approaches to project management that are prevalent in the IT sector are particularly disadvantageous to women.

An increasing body of literature explains the gender pay gap and industrial and occupational segregation by the gendered division of domestic labor. Powers (2003) found that domestic labor tasks account for 19% of the gap between male and female earnings in professional, managerial, and technical occupations. Motherhood has also been associated with a wage penalty for certain racial groups (Glauber, 2007). Women in most industrialized countries are delaying marriage and families to achieve high levels of formal education and establish their careers (Cobb-Clarke & Dunlop, 1999; Drolet, 2001). However, reproductive and domestic work undermines their efforts as it serves as a barrier to promotion (Cobb-Clarke & Dunlop, 1999) and to achieving high-level jobs within their occupations and organizations (Krahn & Lowe, 2002; Livingstone & Pollock, 2009). Livingstone and Pollock (2009) reported that female managers spend twice as long per week as male managers in unpaid household tasks, while male managers spend an average of five additional hours per week in their paid jobs than their female counterparts. It is illustrative that recent popular articles that report higher wage rates for women than men refer to the young, single, and childless status of those women (Roberts, 2007; Dougherty, 2010; Intini, 2010).

Pratt and Hanson (1991) discuss the role of domestic labor in creating occupational segregation. They argued that because of their domestic responsibilities, women choose jobs with particular characteristics: that is, proximity to home, having complementary hours (to their spouse’s job and other time-based needs), and being flexible or part-time. These jobs tend to be in female-dominated occupations and exhibit greater pay inequality. Therefore, women face a narrower range of job choices due to the structure of the labor market and their domestic time constraints and this places them at greater risk for pay-based underemployment.

Time-Based Underemployment

Time-based underemployment occurs when individuals are working fewer hours than they desire. This issue is particularly relevant to women due to the rise in part-time and nonstandard work, the low quality of that work in general, and women’s predominance in it (Armstrong & Armstrong, 1994; Cranford et al., 2003; Duffy & Pupo, 1992, 2000; McGovern, Smeaton, & Hill, 2004; OECD, 2008).

Tipps and Gordon (1985) used two measures to conceptualize underemployment through inadequate hours. They found that women are more likely to be involuntarily part-time than men of the same racial background. Women were not found more likely to have intermittent employment in their study. However, mothers who are making a transition from welfare to work seem to face high degrees of underemployment through intermittent employment and inadequate hours (Lein, Benjamin, McManus, & Roy, 2005). Single mothers and families with children are more likely to be underemployed and among the working poor because very few mothers leaving welfare can obtain a full-time, high-wage job. Lein et al. (2005) reported that only nine out of the sample of 99 had near full-time employment and that this was gained through multiple jobs; 34 faced consistent underemployment in the informal or formal sector and 24 had intermittent employment.

Yu (2002) found similar challenges for mothers at midlife who attempt to return to work after their children are older. Family lifecycle and family–work conflict theories contend that women enter part-time work at particular stages in their lives due to specific ideologies around domestic and reproductive responsibilities (see Blair-Loy, 2003). Malenfant, LaRue, and Vezina (2007) interviewed women who said they ‘chose’ to give priority to their relationship or family life and as a result took on more precarious, but arguably more flexible, employment. Hakim (2004) also noted that women stay in part-time work because they prioritize nonmarket activities. However, Yu (2002) argued that these perspectives imply that part-time work is transitional and that women can move between full-time and part-time at no cost whenever they choose. However, as will be discussed below, women face penalties for taking part-time work and these jobs often become a trap.

Women’s time off for child rearing puts them at a strong disadvantage in the ‘permanent employment model’ of most advanced market economies. Women who have left the labor market manage to re-enter the part-time ranks much more readily than the full-time ranks (Yu, 2002; Lein et al., 2005). Yu (2002) noted that it is much easier and more common for women to transition from full-time to part-time than the other way around. Also, women who remain in full-time jobs while they bear and rear their young children (due to their access to adequate maternity leave) are more likely to be found in full-time employment than those who must leave the market (Yu, 2002). Similarly, Gash (2008) found that women in Denmark and France are less constrained to part-time work than women in the UK because of its more supportive policies around maternal employment.

The issue of choice presents a particular challenge when determining time-based underemployment and complicates any normative judgments that may be directed against women in different situations. For example, VandenHeuvel (1999) reported that 78% of women in his sample chose to work part-time and did not want to work more hours. Liberal feminists would argue that this self-reported ‘choice’ must be interpreted within a complex web of individual, family, societal, and institutional constraints that reinforce sexual divisions of labor and keep women tied to the reproductive work of the home (Barker, 2005; Bergmann, 2005; Blair-Loy, 2003; Spain & Bianchi, 1996; Weisberg & Buckler, 1994).

Many of these arguments presuppose that the most optimal route to gender equality is for women to operate in the labor market similarly to men – that is, to have the same access to human capital development and full-time work. In this vein, even voluntary part-time or reduced hours may be seen as a form of skill-based underemployment in the sense that the latent abilities of these women are being underutilized in domestic work. Socialist feminists, on the other hand, argue that women do not need to become like men in the labor market; rather their unpaid work should be revalued (Barker, 2005). As Warren (2008) found great variation in the working-time contentment of female part-timers, the subjectivity of underemployment, the heterogeneity of women, and the rationalizing ‘myths’ they form around their choices must be acknowledged (Hakim, 2004; Hochschild, 2003).

That said, even voluntary part-time work can lead to other dimensions of underemployment. MacDermid, Lee, Williams, Buck, and O’Sullivan (2002) presented several cases of female professionals and managers who cut back their work hours to cope with the ‘double day’ of paid and unpaid work, and to increase their flexibility to achieve work–life balance. These women had the power and privilege to obtain reduced schedules that were on average 17 h less than their regular work hours. The success rate of these arrangements was relatively high; however, those who were less satisfied with their arrangements attributed their feelings of dissatisfaction to underemployment. Many reported that their upward mobility was on hold until they returned to full-time work; that they were given less interesting, visible, or stimulating assignments; that they could not take on certain projects or professional development opportunities; and that they were missing out on networking opportunities. Women in unsuccessful reduced load arrangements reported subjective underemployment due to feelings of isolation and to unsupportive bosses and coworkers who did not assign them work, did not take them seriously as committed colleagues, and did not respect their schedules. Such perceptions that part-time workers are less committed to their careers are related to glass ceiling effects (Wenger, 2001).

The greater domestic burden that is placed on women seems central to perceptions of time-based underemployment. Notably, Bender and Skatun (2009) found more time-based underemployment among the men in their sample (70% versus 47% for women). However, this is likely because of the fact that men remain the primary earners and therefore desire and require full-time hours (Hakim, 2004). Factors such as being married, having children at home, and having a higher family income were all related to a desire for fewer hours among women. Marriage and children at home did not affect men’s desires regarding number of work hours (Bender & Skatun, 2009). The probability of time-based underemployment increases with the age of dependent children (Kjeldstad & Nymoen, 2010). As children get older, women want to work more hours and any constraints, such as those outlined above, translate to involuntary reduced hours.

Industrial and occupational segregation is also closely linked to time-based underemployment. Kjeldstad and Nymoen (2010) found that 3 out of 4 women are employed part-time, that being employed in female-dominated jobs increased underemployment for both sexes, and that women in general are four times as likely to experience time-based underemployment than men. The authors concluded that these findings were not due to individual differences in gender roles or work preferences, but to systematic differences in the participation of men and women in various segments of the labor market (see also Barrett & Doiron, 2001). Part-time work is concentrated in a very small segment of the labor market that is dominated by women. As Wenger (2001) noted, half of the women who work part-time do so in 4% of the industries in the labor market as a whole. Only one-third of female employees work full-time work in those industries. These jobs are the lowest paid and the lowest skilled; therefore, it does not seem likely that this concentration is due to an individual preference for these specific jobs. Gender differences in time-based underemployment are significantly reduced when men and women are working in the same labor market and same type of job (Kjeldstad & Nymoen, 2010).

Time-based underemployment is also associated with pay- and skill-based underemployment. For example, hourly wages for part-time women workers are 20% less on average than full-time women workers when controlling for education, experience, and family structure (Hudson, 2000). Part-time work also offers limited human capital advantages in terms of labor market experience. Concoran, Duncan, and Ponza (1984) showed that staying in the labor force as a part-time worker gave no more advantage to wage growth than dropping out completely and returning at a later date. Evidence also suggests that part-time workers see lower wage returns on their qualifications than full-time workers (Kalleberg, 2000). These studies highlight that part-time work offers relatively little in the way of advancement opportunities, responsibility, decision-making authority, and challenge, and cannot be counted on to lead to wage increases, upward mobility, or re-entry into full-time work (see also Yu, 2002; Krahn & Lowe, 2002). Indeed it may act as a signal to future employers and further cement workers on their current ‘career’ paths (Rainbird, 2007).

Skill-Based Underemployment

Skill-based underemployment actually comprises a set of distinct dimensions. Here, I discuss six gaps between employee and work situation: talent use, general knowledge, credential, performance, relevance, and subjective gaps. The talent use gap reflects “educational discrimination against youths from poorer economic class backgrounds, as well as those with subordinated race, gender, or other ascriptive characteristics, in terms of their chances to attain qualifications before entering the job market” (Livingstone & Pankhurst, 2009). The general knowledge gap refers to the KSAs that workers acquire through their myriad job experiences, which often exceeds the requirements of any specific job (Livingstone & Pankhurst, 2009; Weststar, 2009a). The credential gap is the match between education attained and education required for entry into the job (Feldman, 1996; Livingstone, 2009). The performance gap is the match between the education attained and that needed to perform the job, whereas the relevance gap refers to the degree to which the focus of one’s education is related to the content of the job. Finally, the subjective gap is an omnibus measure of individuals’ perceptions of the match between their qualifications and job requirements (Livingstone, 2009).

Talent Use and Relevance Gaps

For women in industrialized countries, the rates of educational achievement have never been higher (Krahn and Lowe, 2002). That said, the talent use gap is still manifest in two ways. First, industrial segregation has roots in education segregation (Valentova, Smidova, & Katrnak, 2007 as cited in Livingstone et al., 2008). A lack of women in educational programs related to specific industries and occupations will naturally preclude them from gaining employment in those areas, particularly in a labor market that increasingly values formal credentials. A case study of engineering and science fields in the UK points to the gendered nature of educational choices that lead to those fields (Define, 1992).

The arguments explaining gendered educational choices can be summarized in two camps. Those that emphasize agency theory focus on the choices of girls and women. This approach takes as a starting point the early socialization of girls and women that steers them away from SET jobs toward careers deemed acceptable for women (Glover & Fieding, 1999; Radsma, 2009). Children’s clothing and toys provide ample anecdotal evidence. Women may also actively turn away from jobs in which they know they will be out of place and alienated (Cockburn, 1987, as cited in Glover & Fielding, 1999). Interventionist solutions have abounded to try to change girls’ and women’s perceptions of SET fields and frame them in a light more appealing to females. The second argument is most often taken up in feminist literature; it places the blame on the stereotypically masculine agenda of the SET fields that reinforce patriarchy, male dominance, and the exclusion of women (Glover & Fielding, 1999). Characteristics of the jobs themselves – long work cultures, competitiveness, hostile coworkers and managers, and expectations, norms, and rewards for careerism and uninterrupted work patterns – may discourage women from even enrolling in educational programs in these areas (Chasserio & Legault, 2010; Define, 1992; Scott-Dixon, 2004).

Glover and Fielding (1999) distinguished between ‘getting in’ to a particular field and ‘getting on’ in that field. In recent years, more women have gained access to the previously male-dominated fields of law, medicine, and management. But the numerical parity in these occupations masks the vertical sex segregation that remains. In medicine, the decision to specialize is often gendered, with women emphasizing patient interaction and fewer hours (i.e., general practice and internal medicine over surgery) and consequently being less expectant than men of high salaries (Bergquist et al., 1985; McFarland & Rhoades, 1998). Women are reported to be five times more likely to practice medicine part-time than men, arguably due to their family-based priorities (McFarland & Rhoades, 1998). In science fields, both getting in and getting on have been slow to reach gender parity. Glover and Fielding (1999) reported that 25% of women got into science, engineering, and technology programs in the UK in 1994 as compared to 46% of men. As discussed, women in these programs were more likely to take jobs of lesser status and pay and that arguably utilized fewer of their educational qualifications than men. They were less likely to hold professional jobs in their fields, more likely to be teachers (and at the elementary rather than secondary level), and less likely to be managers compared to male graduates.

Women with advanced degrees who take jobs in less specialized and less technical areas will also experience a relevance gap (and, incidentally, credential, performance, and pay gaps). Conversely, women who obtain more general or liberal education without specific careers in mind (Hakim, 2004) are also more likely to report that their education then bears little relation to their jobs. Such is also the case with jobs that require more diffuse skills such as those in the services or administration that are more populated by women. Pratt and Hanson (1991) found that women who ‘choose’ jobs in either female-dominated occupations or the more female-friendly jobs that have become available within male-dominated industries de-prioritize the fit of the job with their education or qualifications and focus instead on job characteristics that help them overcome domestic scheduling problems. Men, on the other hand, tend to choose jobs based on the opportunity for promotion and the match with their skill set.

Women who leave the labor market for child rearing may also find a greater relevance gap when they return if the value of their education has expired (i.e., in rapidly changing fields) and they must find other work. Indeed, many women workers in fast-paced jobs are at risk of decreased relevance of their education, training, and prior work experience even with short-term leaves. Without some effort at maintenance over the course of their leave, they could return to work without the necessary skills and knowledge, which would negatively impact their performance. To date there has been limited research on these specific topics.

Barriers to informal and formal training opportunities is the second manifestation of the talent use gap for women (Cohen, 2003; Rainbird, 2007). Cooke, Zeytinoglu, and Chowhan (2009) found that low-wage, less educated, and nonunion workers are least likely to receive employer-sponsored training (see also Rainbird, 2007). Further, in each of these groups, women are less likely than men to receive training. They also found that women in nonpermanent jobs are half as likely to receive training as those in permanent jobs and that women in professions are twice as likely to receive training as women in nonprofessional jobs. These distinctions do not hold true for men. However, across their data as a whole, there are no significant training differences for women and men. Cooke et al. (2009) therefore suggest that the gender training barrier may be tied not to explicit discrimination, but rather to the over-representation of women in the groups that are denied employer-sponsored training (i.e., labor market segmentation). Where training is accessed, it is often highly discounted in pay and promotion considerations (Rainbird, 2007; Scherer, 2004).

Georgellis and Lange (2007) reported similar findings. They drew on discrepancy theory, equity theory, and social exchange theory to explain that training is perceived as a broken psychological contract between women and their employers because women do not receive the same career development and occupational mobility benefits from training as do men. Wonacott (2000) reported that women are under-represented in apprenticeship programs and gain less benefit from them. It is important to note here that unions tend to increase the participation and success of women in apprenticeships (Berik & Bilginsoy, 2006) and are also associated with higher incidences of education and training for workers overall (Livingstone & Raykov, 2005). Declining union density rates, therefore, could have even stronger detrimental impacts on women’s access to training and their rates of underemployment.

General Knowledge Gap

Workers acquire knowledge, skills, and abilities (KSAs) through various learning mechanisms throughout their work experience. Therefore, it follows that workers will often have reservoirs of working knowledge that exceed the requirements of any given job (Livingstone & Pankhurst, 2009; Weststar, 2009a). While this is true of all workers, women may face some additional challenges in accruing any benefits from their reserves of general knowledge. In a review of call centers, Hunt (2004) noted that women were less likely than men to use the experience gained at the call center to move into managerial ranks, lateral jobs with higher status, or jobs external to the call center such as with the parent or client companies. Scherer (2004) examined this phenomenon with respect to the impact of first jobs on later careers. She refuted the proposition of human capital theory (as well as career mobility theory and internal labor markets) that workers accept unchallenging jobs, low-pay jobs, or jobs below their qualifications or abilities in exchange for training and experience that will lead to advancement (Sicherman, 1991). Rather, Scherer (2004) used labor market segmentation theory and signaling theory to explain how difficult it is for workers to break the pattern of marginalization set by these early job choices.

The above noted research is classified under the general knowledge gap because each case highlights how a failure to win recognition of the broad working knowledge acquired through previous work experience can trap workers in their jobs, occupations, or industries. This is particularly relevant in the absence of other markers of expertise such as additional formal credentials, continuous advancement through employer-sponsored training, promotions, and career continuity. As noted, women are more likely to have mixed histories with respect to these markers. This argument is supported by the work of Corcoran et al. (1984), who found no wage benefit for part-time work, and Yu (2002), who argued that part-time jobs are often not transitions to full-time jobs. It is possible that in each of these cases, the general knowledge obtained through part-time work is discounted in the full-time labor market. The same discounting of the applicability of knowledge is made in regard to unpaid domestic and reproductive work carried out predominately by women. It has been argued that better recognition of the complexity of this work and corresponding credit for it in educational and work spheres would greatly benefit most women (Caplan & Schooler, 2006; Eichler, 2005).

Credential, Performance, and Subjective Gaps

There is very little research that examines gender- or sex-based differences specifically in terms of credential, performance, or subjective underemployment. Most research that explicitly included sex or gender as a variable of interest used widely different conceptualizations of underemployment or discussed skill-based underemployment more broadly. Some of these studies can shed light on these issues nonetheless.

Tipps and Gordon (1985) measured inadequate skill utilization in terms of overeducation and marginal jobs. Their overeducation was an objective measure of the credential gap that compared educational attainment with General Educational Development (GED) scores for the respondent’s job. Tipps and Gordon found that women are less likely to be credentially underemployed than men across all four racial groups studied. Marginal jobs were defined as those with skill requirements of 3 months or less according to the Directory of Occupational Titles (DOT) and were theorized to represent those typically found in the secondary labor market. As can be expected, women did not fare as well on this measure. Across all racial groups, women were considerably more likely to be in marginalized jobs than men. White females fared the best at 13.9%, as opposed to 21.7% for black females and 21.9% for Mexican American females. White men had the lowest rate at 5.3%. Conversely, Brown and Pintaldi (2006) concluded that women are almost twice as likely to be underemployed than men (3.2% versus 1.7%), using a skill-based underemployment measure constructed from a credential gap assessment and their respondents’ reported desire to leave their jobs.

In asking whether women are at particular risk for overqualification, Renes and Ridder (1995) showed that sex discrimination exists in hiring practices; women required about 6 months more experience to be hired for the same job as men. Renes and Ridder argued that women represent a greater cost to employers because of their increased probability of leaving the labor market. Since employers can no longer account for that risk through lesser wages (due to pay equity and antidiscrimination legislation and norms), they penalize women through increased hiring standards. Women are therefore underemployed because they have 6 months of additional work experience (and the corresponding KSAs) that the job does not necessarily require. This is a slightly expanded conceptualization of the performance gap which is typically only computed as the difference in formal schooling attained and needed to perform the job. Renes and Ridders’ findings point to extra ‘education’ that must be gained through work experience which is consistent with the general knowledge gap.

In a theoretical article about the antecedents and consequences of underemployment, Feldman (1996) hypothesized that women will experience more underemployment than men. He cited several articles in support. For example, Weitzman (1985) found that divorced women who had been out of the labor market raising their children have a greater risk of credential underemployment upon re-entry because they lack work experience. They are penalized for the gap in or lack of work experience by taking a job that is lower than their educational qualifications warrant. Newman (1988) reported greater downward mobility among women, and both Marshall (1984) and Nowak and Snyder (1983) found that women who have been laid off have more difficulty finding suitable employment than men.

Rosen (1987) noted that women may be more likely than men to accept jobs with lower wages as they balance work and their family obligations. Jobs with lower wages do not necessarily mean they are jobs that require lesser qualifications to access or to perform; however, the general assumption under human capital theory is that wages increase with attributed credentials and educational attainment. Thus, women who take jobs with lower wages are likely to be taking jobs that offer less responsibility, autonomy, creativity, and status. As Bergmann (2005, p. 94) noted: “A secretary in a job with low wages and low-priority duties has low productivity, not because she has low potential, but because she has been penned into the secretarial pool by exclusion from other jobs she might have filled.” Given rising educational attainment rates, women may be at increased risk of credential and performance underemployment if they are not given access to commensurate jobs.

One clear example of the performance gap is provided by women who experience sequential segmentation (Pratt & Hanson, 1991). To accommodate the working schedule of their spouse and the care needs of their children, these women often work the second or third (i.e., night) shift. This essentially removes these women from the nexus of organizational activity in the workplace, which occurs in more standard working hours. Women reported that their jobs ask less of them than the same jobs on the day shift. It also has implications for the pay gap due to the glass ceiling effect. Women who cannot work daytime hours are overlooked for managerial or administrative promotions and have more limited internal mobility.

There has been some discussion about how credential and performance gaps translate into subjective underemployment. Where there is often close overlap between credential and performance gap numbers, self-reports of subjective underemployment are often lower. Though some explanations that account for both sexes have been proposed (see Green & McIntosh, 2007; Weststar, 2009b); women may be more likely to experience a form of cognitive dissonance (Festinger, 1957) and rationalize their choices. On the other hand, the structuralist perspective suggests that women who experience credential or performance gaps because they have actively chosen a particular job that allows them to gain fulfillment or balance in other life areas may report less subjective underemployment.

Duffy and Pupo (1992) have noted the substantial benefits that women gain from working as opposed to being full-time homemakers. The part-time women workers they interviewed reported greater outside stimulation and support, greater self-esteem, greater economic independence, and also improved health because the job is seen to be time for themselves. Rainbird (2007) also noted that women who work in low-wage, low-skill jobs in the public service find great satisfaction in the role their jobs play in society. The structuralist perspective, however, is that choice in employment is a luxury denied to most women. Instead, women are forced into nonstandard work or jobs that are lesser than their credentials or abilities due to systemic discrimination and institutional barriers in the labor market, the uneven sexual division of paid and unpaid labor, and the consistent undervaluing of ‘women’s work.’ Women who interpret their choice in this way may therefore be more likely to report subjective underemployment. More research is needed in this area.

The Canadian Work and Lifelong Learning (WALL) survey conducted with a random sample of over 9,000 individuals in 2004 (see http://www.wallnetwork.ca) provides data on the credential, performance, and subjective gaps; however, no systematic analysis has been conducted concerning potential sex differences. Based strictly on frequencies of the raw data, there does not appear to be much difference between men and women. In both groups, roughly 35% of the sample are credentially underemployed, 30% face performance underemployment, and 27% report subjective underemployment.

A slightly different picture is presented when the employment trends of women are considered. Given the dominance of women in part-time and temporary work and their segregation into specific occupations, it is important to look at these frequencies as well. These data highlight that people who work part-time, who are in temporary work, and who work in sales and service occupations face higher credential, performance, and subjective underemployment (see Table 6.2). Brown and Pintaldi (2006) also reported that service sector workers are the most likely to have skill-based underemployment.

Table 6.2 Credential, performance, and subjective gaps by hours worked, work status, and sector
Full size table

Most part-time and temporary jobs, as well as jobs in the highly feminized sales and services sectors, tend to be more proscribed in terms of managerial control and routinized tasks. These characteristics lead to greater performance underemployment because workers have less opportunity and control to draw on their reservoir of KSAs and fit their abilities to their work (Livingstone, 2009; Weststar, 2009b). They also lead to more credential underemployment because only general qualifications are required for entry and those qualifications tend to be lower. Professional and technical occupations, on the other hand, tend to have lower levels of credential underemployment due to specific and advanced educational requirements for the job (Livingstone, 2009).

Radsma’s (2009) study of clerical workers highlights the case of a specific group of mostly female workers in a highly feminized occupation, but it is informative of the situation of many female workers. On all dimensions of credential, performance, and subjective underemployment, service workers are found to be more underemployed than other workers and clerical workers register even higher levels than the average service worker (Radsma, 2009). Most clerical jobs do not need a formalized set of qualifications, but creeping credentialism and high competition for a declining number of ‘good’ jobs means that clerical workers must continually increase their credentials in whatever way they can to stay on top of the applicant pile. This serves to increase the underemployment of these workers as the jobs are not being upskilled to keep pace – rather, most are being continually deskilled or eliminated altogether by new technologies (Radsma, 2009).

Conclusion

Labor market inequalities should decrease with rising participation of women in the labor market and rising educational attainment by women. Pay and employment equity programs are working slowly to reduce systemic discrimination in hiring and wage setting in some segments of the labor market. Similarly, work by governments, employers, and unions to create supportive policies for pregnancy and parental leave and family-friendly work schedules keep more women in the paid labor force and more upwardly mobile. In terms of the credential and performance gaps, women seem to face no more underemployment than men when they occupy the same jobs with the same entry credentials.

That said, this good news does not apply to all women. Pay equity and employment equity programs do not reach all workers, typically missing those in the lower ranks of the private sector. These workplaces are also the least likely to have family-friendly employment policies and union protection. Poor parental leave supports and inadequate child care force women out of the paid, formal workforce to care for their young children. This gap is penalized upon re-entry. Furthermore, occupational and job segregation are still very apparent in the labor market and will continue to shunt women into jobs that are part-time, lower waged, and nonunionized, and offer less in terms of responsibility, autonomy, and opportunity to expand KSAs. Job segregation and the structure of male-dominated jobs also push women into subspecialties or career paths that seem to be a better fit with their domestic and reproductive responsibilities. Regardless of changing societal and occupational norms around gender roles, the reality is that women carry a larger burden of domestic and reproductive labor and they face a large cost for this in the paid labor market.

The issue of rational choice often dominates the debate on women’s equality. It would be fruitful to seek solutions that can sidestep the quagmire of academic debate about whether women choose jobs with lower wages, lower skill requirements, lower responsibility, and fewer hours to fit their desired lifestyle and ideologies or whether societal, familial, and institutional dictates leave these jobs as the only option for many women. Regardless of how or why women get to these jobs, once in them they are more at risk to experience forms of underemployment and the corresponding negative consequences on their mental and physical health, their future job prospects, and their ability to make a contribution.

It is unlikely that women actively choose to be marginalized and underutilized; therefore the focus should be on how to ‘fix’ nonstandard jobs to better utilize talent and to devise new economic and social solutions for the current market system. The liberal feminist perspective that women should enter and participate in the labor market like men under the system of human capital seems not to be working. Socialist feminists would argue that an alternative is to revalue unpaid and informal labor and informal learning in the domestic sphere vis-à-vis the formal marketplace (Barker, 2005; Eichler, 2005).

This raises a fundamental problem with the conceptualization of underemployment in terms of market-based work and formal learning only. When women are full-time homemakers or working part-time it is easy to say they are underemployed because of the assumption that only full-time paid work can fully utilize their KSAs or that the ultimate goal is to match education with work in the paid labor force. But the reality is that many women (and also men) are engaged in informal work and learning activities that are fulfilling to the individual, the family, the community, society, and the economy. These activities are not included in the current understanding of underemployment and not recognized nor valued in the modern market economy (Eichler, 2005).

Future research in this area, then, would do well to explore definitions of ‘under-work’ as opposed to underemployment that can capture the experiences of all women from full-time homemakers to full-time workers, including the self-employed (Arai, 2000). Similarly, more attention is needed on the true impact of policy interventions to assist women in overcoming underemployment. There is evidence that current work–life balance programs and equity programs are not removing the stresses and barriers that women face. Though positive on the surface, it has been noted that these policies may actually reinforce gender stereotypes and gendered divisions of labor in the household. They place women rather than men in the home, caring for the children and carrying out domestic duties (Fuwa & Cohen, 2007). Work–life balance programs that include flexible work schedules just put more responsibility for organizing the work on the shoulders of the employee and they do not address the core problem of long work hours (see Livingstone & Pollock, 2004). A final concern is that mechanisms that allow women to pass the burden of domestic care to others so that they can engage in the ‘real work’ of the paid marketplace simply shift problems of underemployment to a new set of female workers. Appelbaum, Bailey, Berg, and Kalleberg (2002) suggest a new shared work–valued care model that could address some of these issues.