Moving Up or Falling Behind? Gender, Promotions, and Wages in Canada

• Date: 01 April 2019
• Published date: 01 April 2019
• DOI: http://doi.org/10.1111/irel.12231
• Author: Mohsen Javdani,Andrew McGee

Moving Up or Falling Behind? Gender,

Promotions, and Wages in Canada*

MOHSEN JAVDANI and ANDREW MCGee

We estimate that Canadian women working full time are 1.8 percentage points

less likely to be promoted, receive fewer promotions, and experience 2.8 percent

less wage growth following promotions than similar men. Significant “family

gaps”exist among women. Women without children are less likely to have been

promoted than similar men but experience similar wage growth following promo-

tions, while women with children are as likely to have been promoted but experi-

ence less wage growth following promotions. Weekly hours and overtime hours

explain significant fractions of these gender gaps. Though not precisely estimated,

gender gaps in promotions also exist among part-time workers.

Introduction

Women struggle to reach the top of the career ladder overall and in particu-

lar occupations (e.g., Long, Allison, and McGinnis 1993; Bertrand and Hal-

lock 2001; Matsa and Miller 2011; Goldin and Katz 2012; Smith, Smith, and

Verner 2013; Azmat and Ferrer 2017). Likewise, gender differences in earn-

ings within establishments appear to grow over the course of the career (Barth,

Kerr, and Olivetti 2017; Goldin et al. 2017). Taken together, these findings

signal the potential importance of gender differences in upward mobility via

promotions to gender differences in career outcomes.

In this study, we use a well-defined measure of promotion to provide the

first estimates of gender gaps in promotion experiences for a nationally repre-

sentative sample of Canadian workers between ages 18 and 65 from 2000 to

2004 using data from the Workplace and Employee Survey (WES), a linked

employer–employee survey. Women working full time were an estimated 1.8

percentage points less likely to have been promoted between interviews,

JEL: J16, J31, J62, J7.

*The authors’affiliations are, respectively, University of British Columbia—Okanagan, Kelowna, British

Columbia, Canada. E-mail: mohsen.javdani@ubc.ca; and University of Alberta, Edmonton, Alberta, Canada.

E-mail: mcgee1@ualberta.ca. The authors thank conference participants at the Western Economics Associa-

tion Meetings and the Canadian Economics Association Meetings.

INDUSTRIAL RELATIONS, DOI: 10.1111/irel.12231. Vol. 58, No. 2 (April 2019). ©2019 The Regents of

the Univers ity of Califo rnia Published by Wiley Periodicals, Inc., 350 Main Street, Malden, MA 02148,

USA, and 9600 Garsington Road, Oxford, OX4 2DQ, UK.

189

received 0.16 fewer promotions with their current employers, and experienced

2.8 percent less wage growth in the year of a promotion than similar full-time

men controlling for worker characteristics, job characteristics, and industry.

Thus, we join a large body of research finding that women are less likely to

be promoted and enjoy smaller wage returns to promotions than similar men.

1

In a novel contribution to the literature on gender gaps in promotions, we

also report estimates of the gender gaps in promotion experiences among part-

time workers. Part-time women were an estimated 4.2 percentage points less

likely to have been promoted, received 0.05 fewer promotions with their cur-

rent employers, and experienced 8.9 percent less wage growth following pro-

motions than similar men. Our sample of part-time workers, however, is one-

tenth the size of the full-time sample, and as a result these gender gaps are not

precisely estimated. As such, most of our discussion focuses on full-time

workers—although we report estimates for both groups of workers and discuss

differences where appropriate.

Our primary contribution is to test hypotheses concerning sources of the

gender gaps in promotion outcomes using rich information about workers and

their employment conditions. First, employers may statistically discriminate

against women when making promotion decisions in anticipation of women

quitting for family reasons (Becker 1985; Lazear and Rosen 1990). Second,

gender differences in hours worked might result in gender differences in pro-

motion rates and the returns to promotions if promotion-track jobs require

more hours and human capital acquired through work hours is rewarded more

highly following promotions (Gicheva 2013). Third, women may be less likely

to be promoted or compensated less generously upon promotion than men if

women opt for family-friendly work arrangements as documented in Goldin

and Katz (2012) and Goldin (2014) and employers value workers who provide

work on-demand and during the employers’preferred hours. Fourth, women

may be less likely than men to enter or win promotion contests if women are

less competitive than men (Niederle and Vesterlund 2011).

1

Studies finding that women are less likely to have been promoted include Blau and DeVaro (2007),

Cabral, Ferber, and Green (1981), Cannings (1988), Cobb-Clark (2001), Cobb-Clark and Dunlop (1999),

Johnston and Lee (2012), Kunze and Miller (2017), McCue (1996), Olson and Becker (1983), Pekkarinen

and Vartiainen (2006), and Ransom and Oaxaca (2005). A handful of studies find that women are either

more or equally likely to have been promoted relative to similar men (Addison, Ozturk, and Wang 2014a,

2014b; Booth, Francesconi, and Frank 2003; Gerhart and Milkovich 1989; Hersch and Viscusi 1996). Stud-

ies finding that women experience smaller wage returns to promotion include Booth, Francesconi, and Frank

(2003); Hersch and Viscusi (1996); Johnston and Lee (2012); and van der Klaauw and de Silva (2011).

Again, a handful of studies find either that women experience larger or comparable wage returns to promo-

tion relative to men (Addison, Ozturk, and Wang 2014b; Cobb-Clark 2001; Blau and DeVaro 2007; Gerhart

and Milkovich 1989; Olson and Becker 1983).

190 / MOHSEN JAVDANI AND ANDREW MCGEE

Statistical discrimination models predict that in competitive markets, firms

statistically discriminate in promotion decisions but not in the wages paid to

promoted workers (e.g., Lazear and Rosen 1990; Bjerk 2008). Employers,

however, have less need to statistically discriminate against women who

already have children as questions surrounding their labor-force participation

conditional on the arrival of children have been answered. Consistent with

employers statistically discriminating against women without children in pro-

motion decisions in anticipation of potential quits should they have children,

women without children working full time were an estimated 2.7 percentage

points less likely to have been promoted than similar men but experienced

wage growth following promotions statistically indistinguishable from that of

similar promoted men. By contrast, women with children working full time

were just as likely to have been promoted but experienced 4.3 percent less

wage growth following promotions than similar men—afinding that cannot be

rationalized by statistical discrimination. We observe similar family gaps

among women working part time.

Turning to the role of hours, gender differences in weekly hours explain as

much as a quarter of the gender gap in the probability of promotion and a

third of the gender gap in the wage returns to promotion among full-time

workers. These findings are consistent with workers who experience higher

disutility from work hours (as women might if they value time with family

more than men) opting into careers with fewer promotions but also requiring

fewer hours. Promoted women who have worked fewer hours than similar pro-

moted men then experience slower wage growth than promoted men if higher-

level jobs pay a larger premium for human capital acquired through on-the-job

learning (Gicheva 2013). Hours, however, explain none of the gaps among

part-time workers as part-time women actually work more hours on average

than part-time men.

We find limited evidence that work flexibility contributes to the gender gaps

in promotion experiences. Full-time men are actually more likely to work flex-

ible schedules in our sample. Likewise, men and women in our full-time sam-

ple are equally likely to work hours that change from week to week, which

presumably would be undesirable to women with families. As a result, work-

ing flexible or varying hours explains none of the gender gaps in promotion

experiences among both full- and part-time workers. At the same time, full-

time women in our sample work significantly fewer paid and unpaid overtime

hours, which might reflect constraints on women’s ability to supply labor at

times of the employer’s choosing. Controlling for weekly paid and unpaid

overtime hours reduces the gender gap in the probability of promotion among

full-time workers by nearly a third but explains little of the gender gap in the

wage returns to promotion. Firms may consider a worker’s ability to supply

Moving Up or Falling Behind? / 191