Frontline pay gaps in Canada, the UK, and the US can't be ignored

ES
LG
catalyst-iwd-home-feature_1080x1080

By  Ellie Smith, PhD ,  Lyangela Gutierrez, PhD  &  Tara Van Bommel, PhD

10 min read

| Published on
catalyst-report-hero-general

Executive summary

In Canada, the United Kingdom, and the United States, national gender pay gaps remain striking: on average, women make 16.1% less than men in Canada, 13.5% less in the UK, and 16.4% less in the US.1 But these numbers tell only part of the story. Our data show that frontline employees2 — often the essential backbone of industries — face significant and under-addressed pay disparities. This analysis highlights inequities experienced by frontline employees3 that are shaped by gender, childcaring responsibilities, and region, and provides actions HR professionals can take to address them.

Gender pay gaps persist across regions and frontline roles

Critical insights

  • Canada: Women are equally likely as men to be paid hourly,4 but more likely to earn under $CAD20/hour.5 The gap appears to narrow when employees are paid an annual salary.6
  • UK: Women are equally likely as men to be salaried,7 but significantly more likely to earn under £20/hour8 and under £40,000/year.9
  • US: 62% of women earn under $20/hour,10 and 24% earn less than $40,000 annually11 — both higher than men.

Despite ongoing awareness and advocacy, women consistently earn less than men across all three countries.12 While some structural factors such as type of pay (hourly vs. salaried) are gender-neutral in Canada and the UK, income disparities remain significant. In the US, women are more concentrated in hourly roles13 and earn less on both hourly and annual measures. Some research also points to age-related disparities, showing that young women in unstable, low-paid roles often earn less than their male counterparts14 — a pattern that reinforces early-career inequality and compounds over time.

catalyst-frontline-pay-gaps-2025_chart-1a
catalyst-frontline-pay-gaps-2025_chart-1b

Take action

  • Immediate action: Run a simple pay equity check using existing payroll data. Filter by gender, pay type (hourly/salaried), and income bands (e.g., under £/$/CAN 20/hour, under £/$/CAN 40,000/year). Flag roles with the widest gaps.
  • Longer-term action: Build a pay equity dashboard that updates quarterly. Include filters for gender, geography, job level, and pay type. Use it to track progress and inform compensation decisions.

In many cases, gender-based pay differences widen with parenthood

Critical insights

  • Women consistently earn less than men regardless of parental status.
  • In almost all cases, men make more than women, and parents make more than non-parents.
  • Patterns in the data suggest that men who are not parents tend to earn more than women who are parents.

The interplay between gender and childcare responsibilities15 reveals important nuances. Across Canada,16 the UK,17 and the US,18 when we look at hourly employees, we see that the pay gap between working mothers and fathers is larger than the pay gap between women and men who are not parents. This evidence supports prior research pointing to the existence of a “fatherhood bonus” and/or a “motherhood penalty”19 that exacerbates inequities.

Frontline employees with childcare responsibilities are more likely than those without childcare responsibilities to be in salaried roles, a trend driven largely by women.20 This suggests that having childcare responsibilities may be linked to a preference for structured pay and predictable schedules.21 When we look at salaried employees, we find a motherhood penalty in Canada.22 We don’t see it in the UK23 or the US,24 although women are still consistently earning less than men in those countries — even working mothers are paid less than non-parent men.

catalyst-frontline-pay-gaps-country-2025_chart-2a
catalyst-frontline-pay-gaps-country-2025_chart-2b

Take action

  • Immediate action: Ensure frontline employees have access to the benefits available to them. Examples could include parental leave or retirement plans,25 as well as flexible scheduling.
  • Longer-term action: Develop a section for caregivers in your organization’s pay equity policy. Include guidance on flexible scheduling, promotion pathways, and salary review triggers for employees with childcare responsibilities.

National pay structures shape frontline realities

Critical insights

  • UK frontline employees earn less overall: Salaried roles are more prevalent in the UK,26 but only 36% of frontline employees earn more than £40,000 annually,27 and just 20% earn more than £20/hour.28
  • Canada and the US report higher average earnings, but hourly pay dominates,29 especially for women:
    • In Canada, 64% of frontline employees earn more than $CAD20/hour,30 and 89% earn more than $CAD40,000/year.31
    • In the US, 48% of frontline employees earn more than $20/hour,32 and 82% earn more than $40,000/year.33

Varied regulation and uneven compliance structures help explain why frontline pay disparities vary so widely across countries, shaping everything from benefit access to income predictability; this disproportionately impacts women and younger employees.34

Although the UK introduced pay gap reporting in 2017,35 a lack of legal enforcement has limited its impact.36 In contrast, the EU’s forthcoming pay transparency directive37 is expected to introduce more robust accountability measures, which will reverberate well beyond Europe, particularly for multinational organizations. Meanwhile, in Canada and the United States, where regulation is lighter,38 disparities remain deeply embedded, especially for women in hourly roles. These findings provide critical insight for employers and policymakers seeking to close pay gaps at the front line, where the risk of inequity is often greatest but least visible.

catalyst-frontline-pay-gaps-2025_chart-3

Take action

  • Immediate action: Map your frontline roles by country and pay type. Identify where hourly roles dominate and where pay falls below national benchmarks (e.g., £20/hour in the UK).
  • Longer-term action: Align internal pay transparency practices with the EU directive. Start by publishing anonymized pay bands for frontline roles and setting up a review process for outliers.

Job rank drives pay, but not equally across pay types

Critical insights

  • Hourly pay rises with rank: 59% of first-line managers and 47% of supervisors earn more than £/$/CAN 20/hour, compared to 33% of employees with no supervisory responsibilities.39
  • Annual salary patterns are less linear: 66% of first-line managers and 66% of supervisors earn over £/$/CAN 40,000 annually, compared to 48% of non-managers. The identical proportion of supervisors and managers in the higher salary band suggests a plateau rather than a step-wise increase.40

As expected, frontline employees in more senior roles (such as first-line managers and supervisors) are more likely to earn higher hourly wages. However, this association is weaker for salaried employees, suggesting that role-based pay progression is not always consistent.

Take action

  • Immediate action: Review compensation for supervisors and first-line managers. If they’re earning the same despite different responsibilities, flag for adjustment in the next pay cycle.
  • Longer-term action: Redesign career progression pathways for frontline roles. Ensure each step up in responsibility comes with a clear and proportional pay increase.

Embed equity into everyday decisions

Make frontline pay equity visible and actionable.

Pay disparities at the frontline are not just reflections of broader economic trends — they are symptoms of deeply embedded structural inequities shaped by job type, gender, childcare responsibilities, geography, and regulation. While national pay gap figures offer a high-level view, they miss the granular, complex realities of the frontline environment.

Employers and policymakers must push beyond compliance to uncover where inequities persist within their frontline workforce. That means moving beyond averages, disaggregating data, and understanding how pay interacts with role, region, and identity.

For organizations committed to equity, this is a call to act — to audit compensation structures, invest in transparent promotion pathways, and ensure that pay equity is not just a corporate value but a lived experience for every employee, especially those whose work is essential yet too often undervalued.

Endnotes

  1. Women in work: Productivity and gender. (2025). PwC.
  2. This report focuses on employees in frontline roles directly involved in production, processing, and service delivery in non-office settings that typically do not require higher-education qualifications. In these jobs, employees are required to work in person at a specific physical location (e.g., factory, hotel, restaurant, store) during shifts that may be set or variable and often include hours outside the current Monday to Friday, 9-to-5 paradigm of office work. For a full definition of who constitutes a frontline employee, please see Smith, E. & Van Bommel, T. (2025, May 20). Four drivers of frontline employee satisfaction and business results. Catalyst.
  3. We surveyed 8,990 frontline employees and managers in the United States (n = 4,028, 45%), United Kingdom (n = 2,932, 33%), and Canada (n = 2,028, 23%) in a range of industries including retail (n = 2,584, 29%), wholesale (n = 368, 4%), hospitality (n = 1,230, 14%), manufacturing (n = 1,367, 15%), construction (n = 1,176, 13%), finance and insurance (n = 1,200, 13%), mining, quarrying, oil and gas extraction (n = 80, 1%), utilities (n = 724, 8%), and transportation and warehousing (n = 261, 3%). Our sample was majority cisgender women (50%, n = 4,469), with just less than half men (49%, n = 4,362) and some representation of other genders (<2%, n = 144). More than half of the respondents were White (63%, n = 5,631) and our sample included representation from other racial and ethnic identities as well: Asian (9%, n = 789); Black (15%, n = 1,310); Latine (4%, n = 324); Indigenous (1%, n =90); MENA (<1%, n =62); multiracial (7.5%, n = 669); Arab, Iranian, Turkish, Romani, or a combination of these (1%, n = 17); and others who self-described their race (<1%, n = 47). Most respondents identified as heterosexual/straight (85%, n = 7,639), and our sample represented other sexual identities as well (e.g., asexual, bisexual, gay, lesbian, pansexual, or queer employees (13%, n = 1,137). The average participant age was 40 years old, and ages ranged from 18 to 75. Almost half of the respondents were Millennials (44%, n = 3,922), 20% were Gen Z (n = 1,798), around a quarter were Gen X (26.5%, n = 2,378), and the rest were Baby Boomers (10%, n = 892). Roughly a third of the respondents were in non-managerial roles with no supervisory responsibilities (36%, n = 3,190), a quarter were in non-managerial roles with supervisory responsibilities (27%, n = 2,448), and over a third were in first line or frontline managerial roles (37%, n = 3,352). Note that participants could skip demographic questions, so totals may not equal 100% or total sample size.
  4. Survey respondents were asked whether they were paid “hourly” or by “annual salary.” This was used as their “pay type.” A chi-square test of independence found no significant association between gender and pay type in Canada, χ²(3) = 4.78, p = .188. Women (67%, z = -0.1) and men (67%, z = -0.1) were equally likely to be paid hourly. Respondents identifying as another gender were somewhat more likely to be paid hourly (83%, z = 1.1), but this difference was not statistically meaningful (z < 2). Similarly, no significant gender differences were observed among salaried employees.

  5. Respondents were asked their hourly wage, and responses were dichotomized into “below £/$/CAN 20/hour” and “above £/$/CAN 20/hour.” A chi-square test of independence revealed a significant association between gender and earning below or above $CAD20/hour in Canada, χ²(3) = 98.68, p < .001. A significantly higher proportion of women earned under $CAD20/hour (47%, z = 4.8) than men (23.5%, z = -5.4). Men were significantly more likely to earn $CAD20/hour or above (77%, z = 4.0) than women (53%, z = -3.6). Respondents identifying as another gender were also more likely to earn below $CAD20/hour (72%, z = 3.3) and less likely to earn $CAD20/hour or above (28%, z = -2.5), suggesting notable disparities — though small sample sizes warrant interpretive caution.

  6. Respondents were asked their annual salary, and responses were dichotomized into “below £/$/CAN 40,000/year” and “above £/$/CAN 40,000/year.” A chi-square test of independence showed no significant association between gender and earning below or above $CAD40,000/year in Canada, χ²(2) = 3.18, p = .204. While women were somewhat more likely to earn below $CAD40,000 (13%, z = 1.1) than men (9%, z = -1.0), the standardized residuals fell below the ±2 threshold, indicating that the differences were not statistically meaningful. Similarly, no significant differences were observed for respondents identifying as another gender.
  7. A chi-square test of independence revealed a statistically significant overall association between gender and pay type in the UK, χ²(3) = 16.55, p < .001. Women were slightly more likely to be paid hourly than men (54% vs. 48%), while men were slightly more likely to be salaried (52% vs. 46%). Respondents identifying as another gender were more likely to be paid hourly (72%) than salaried (28%). However, nearly all standardized residuals were below ±2, indicating that these differences were not statistically significant at the group level. The only standardized residual that reached significance corresponded to the lower likelihood for people of other genders to be salaried (28%, z = -2.0); still, the small sample size warrants caution.
  8. A chi-square test of independence showed a significant association between gender and earning below or above £20 per hour in the UK, χ²(3) = 37.46, p < .001. A higher proportion of women earned below £20/hour (86%, z = 1.8) compared to men (74%, z = -1.7). Men were significantly more likely to earn £20/hour or above (26%, z = 3.5) than women (14%, z = -3.7), indicating a meaningful deviation from expected counts. Additionally, respondents identifying as another gender were more likely to earn £20/hour or above (35.5%, z = 2.0), although the small sample size warrants caution.
  9. A chi-square test of independence showed a significant association between gender and earning below or above £40,000 annually in the UK, χ²(3) = 57.96, p < .001. A higher proportion of women earned below £40,000 (74%, z = 3.1) compared to men (56%, z = -2.6). Men were significantly more likely to earn above £40,000 (44%, z = 3.5) than women (26%, z = -4.1), indicating meaningful deviations from expected counts. Respondents identifying as another gender were more likely to earn above £40,000 (83%, z = 2.7), although the small sample size warrants caution.
  10. A chi-square test of independence showed a significant association between gender and earning below or above $20 per hour in the US, χ²(3) = 173.34, p < .001. A significantly higher proportion of women earned below $20/hour (62%, z = 5.9) compared to men (38%, z = -6.9). Men were significantly more likely to earn $20/hour or above (62%, z = 7.1) than women (38%, z = -6.1). Respondents identifying as another gender did not show significant differences, with residuals below ±2, though the trend resembled women’s.
  11. A chi-square test of independence showed a significant association between gender and earning below or above $40,000 annually in the US, χ² (2) = 25.19, p < .001. Women were significantly more likely to earn below $40,000 (24.5%, z = 3.4) than expected, while men were significantly less likely to earn below $40,000 (13%, z = -2.9). No significant deviations were observed for earnings above $40,000 or for respondents identifying as another gender, with standardized residuals all within ±2.
  12. A chi-square test of independence revealed a significant association between country and hourly wage bands (χ²(2) = 600.65, p < .001), with patterns differing by gender. UK women and men were significantly more likely to earn below £20/hour (86%, z = 7.3; 74%, z = 12.1), while Canadian women and men were more likely to earn above that threshold (53%, z = 7.9; 76.5%, z = 7.1). US employees of both genders showed more balanced distributions (z = ±1.3–2.8), with 38% of US women and 62% of men earning more than $20/hour. Among respondents identifying as another gender, differences were not statistically significant (χ²(2) = 0.63, p = .729). Annual income patterns followed a similar trend (χ²(2) = 801.11, p < .001). UK women and men were significantly more likely to earn below £40,000/year (74%, z = 11.0; 56%, z = 12.2), while US and Canadian employees were more likely to earn above that threshold — 76% of US women (z = 5.7) and 87% of Canadian women (z = 7.8), versus 87% of US men (z = 5.6) and 91% of Canadian men (z = 4.9). Again, income differences for another gender were not significant (χ²(2) = 2.32, p = .313).
  13. A chi-square test of independence revealed a statistically significant association between gender and pay type among US employees, χ² (3) = 73.90, p < .001). The standardized residuals highlight notable deviations from expected values: women were overrepresented in hourly roles (79.5%, z = 2.8) and underrepresented in salaried roles (20.5%, z = –4.7), while men showed the opposite pattern: they were underrepresented in hourly roles (68%, z = –3.1) and overrepresented in salaried roles (32%, z = 5.3).
  14. Turner, B., Oyedoyin, B., Kihm, C., O’Sullivan, E., Oakley, E., Perry, D., Stanley, O., Baldoza, S., Chowdhury, S., & Hughes, S. (2025). Living precariously: young women’s experiences of insecure work. Young Women’s Trust; A world not designed for us: Annual survey 2024. (2024). Young Women’s Trust; Moyser, M. (2019). Measuring and analyzing the gender pay gap; A conceptual and methodological overview. Statistics Canada; Girls on the job: Realities in Canada. (2019). Girl Guides; Fry, R. (2022, March 28). Young women are out-earning young men in several U.S. cities. Pew Research Center.
  15. In the survey, we asked respondents whether they had any childcare responsibilities. Answers were collected as “yes” or “no.” If respondents answered “yes,” then they formed the “employees with childcare responsibilities” group. In this research, we use "working parents/mother/father" and "employees with childcare responsibilities" interchangeably. For comparison groups, we use “employees without childcare responsibilities” and “employees who are not parents/non-parent women/non-parent men” interchangeably.
  16. Hourly income was measured using sequential pay bands, with the first category rising by $CAD 9.99 (i.e., $CAD 0 - $CAD 9.99) and subsequent categories increasing by increments of $CAD 4.99 (e.g., $CAD 10 - $CAD 14.99, $CAD 15 - $CAD 19.99). A one-way ANOVA was conducted to investigate the impact of gender and parental status on hourly income in Canada. In our model, we combined gender and parental status into one variable (i.e., “gender/parental status”) with four categories: working mothers, working fathers, non-parent women, and non-parent men. In addition, we included age and job rank as control variables and investigated the potential impact of an age by gender/parental status interaction. After accounting for employees’ age and job rank, the results indicated no significant effect of gender/parental status, F(3, 1,289) = 1.82, p = .14, and no significant interaction between age and gender/parental status, F(3, 1,289) = 0.74, p = .53. Bonferroni pairwise comparisons provide more nuance to the patterns, showing that non-parent women earned less hourly than non-parent men (MΔ = -0.59, p < .001) and working fathers (MΔ = -0.91, p < .001); working mothers also earned less hourly than non-parent men (MΔ = -0.50, p < .001) and working fathers (MΔ = -0.82, p < .001). There was no significant difference in hourly earnings between working mothers and non-parent women (MΔ = 0.09, p = 1.0). Similarly, there was no significant difference between working fathers and non-parent men (MΔ = 0.32, p = .08).
  17. Hourly income was measured using sequential pay bands, with the first category rising by £9.99 (i.e., £0 - £9.99) and subsequent categories increasing by increments of £4.99 (e.g., £10 - £14.99, £15 - £19.99). A one-way ANOVA was conducted to examine the effect of gender and parental status on hourly income in the UK. In our model, we combined gender and parental status into one variable (i.e., “gender/parental status”) with four categories: working mothers, working fathers, non-parent women, and non-parent men. Furthermore, we added age and job rank as control variables and investigated the potential impact of an interaction between age and gender/parental status. Even when controlling for employees’ age and job rank, the results indicated a significant main effect of gender/parental status, F(3, 1,434) = 6.90, p < .001, but the interaction between age and gender/parental status was not significant, F(3, 1,434) = 1.27, p = .284. Bonferroni pairwise comparisons detail the patterns in the data, demonstrating that non-parent women earned less than non-parent men (MΔ = -0.46, p < .001) and working fathers (MΔ = -0.63, p < .001). Similarly, we find that working mothers earned less than non-parent men (MΔ = -0.38, p = .004) and working fathers (MΔ =-0.55, p < .001). There was no significant difference in hourly pay between working mothers and non-parent women (MΔ = -0.08, p = 1.0), and no significant difference either between working fathers and non-parent men (MΔ = 0.17, p = .99).
  18. Hourly income was measured using sequential pay bands, with the first category rising by $9.99 (i.e., $0 - $9.99) and subsequent categories increasing by increments of $4.99 (e.g., $10 - $14.99, $15 - $19.99). A one-way ANOVA was performed to examine the impact of gender and parental status on hourly income in the US. In our model, we combined gender and parental status into one variable (i.e., “gender/parental status”) with four categories: working mothers, working fathers, non-parent women, and non-parent men. We also included age and job rank as control variables and investigated the potential impact of an interaction between age and gender/parental status. After accounting for employees’ age and job rank, the results showed a significant main effect of gender/parental status, F(3, 2,855) = 5.88, p < .001, but no significant interaction between age and gender/parental status, F(3, 2,855) = 1.35, p = .26. Bonferroni pairwise comparisons provide more nuance to the patterns, showing that non-parent women earned less than non-parent men (MΔ = -0.61, p < .001) and working fathers (MΔ = -1.05, p < .001). Akin to this, working mothers also earned less than non-parent men (MΔ = -0.48, p < .001) and working fathers (MΔ =-0.92, p < .001). There was no significant difference in hourly pay between working mothers and non-parent women (MΔ = 0.13, p = .57). However, working fathers earned more hourly than non-parent men (MΔ = 0.44, p < .001).
  19. Van der Straaten, K., Pisani, N., & Kolk, A. (2024, June 21). Multinationals could help close parenthood wage gaps. This is how. World Economic Forum.
  20. A chi-square test of independence revealed a statistically significant association between parental status and pay type across Canada, the UK, and the US, χ² (1) = 85.02, p < .001, with notable gendered patterns. Employees with childcare responsibilities were significantly more likely to be in salaried roles (42%, z = 6.2), while those without childcare responsibilities were more likely to be in hourly roles (68%, z = 3.1). This trend was especially pronounced among women, with 40% of working mothers in salaried roles (40%, z = 5.5), compared to just 28% of women who were not parents (z = –4.0).
  21. Amire, R. (2023, March 8). The top 3 ways to retain and support working parents. Great Place to Work; Heinrich, C. J. (2014). Parents’ employment and children’s well-being. The Future of Children, 24(1), 121-146.
  22. Salary was measured using sequential pay bands, with the first category capturing annual salary below $CAD 20,000 and subsequent categories increasing by increments of $CAD 4,000 (e.g., $CAD 21,000 - $CAD 25,000, $CAD 26,000 - $CAD 30,000). A one-way ANOVA was conducted to investigate the impact of gender and parental status on salaried income in Canada. In our model, we combined gender and parental status into one variable (i.e., “gender/parental status”) with four categories: working mothers, working fathers, non-parent women, and non-parent men. In addition, we included age and job rank as control variables and investigated the potential impact of an age by gender/parental status interaction. After accounting for employees’ age and job rank, the results revealed no significant effect of gender/parental status, F(3, 579) = 0.07, p = .98, and no significant interaction between age and gender/parental status, F(3, 579) = 0.19, p = .90. Bonferroni pairwise comparisons provide more detail to the patterns, showing that non-parent women earned less than working fathers (MΔ = -0.99, p = .02) and working mothers earned less than working fathers (MΔ =-1.21, p = .01). There was no significant difference in salaried income between non-parent women and non-parent men (MΔ = -0.22, p = 1.0); likewise, there was no significant difference between working mothers and non-parent men (MΔ = -0.44, p = .98). There was also no significant difference in salaried income between working mothers and non-parent women (MΔ = -0.22, p = 1.0), along with no significant difference between working fathers and non-parent men (MΔ = 0.77, p = .12).
  23. Salaried income was measured using sequential pay bands, with the first category capturing annual salary below £20,000 and subsequent categories increasing by increments of £4,000 (e.g., £21,000 - £25,000, £26,000 - £30,000). A one-way ANOVA was conducted to examine the effect of gender and parental status on hourly income in the UK. In our model, we combined gender and parental status into one variable (i.e., “gender/parental status”) with four categories: working mothers, working fathers, non-parent women, and non-parent men. Furthermore, we added age and job rank as control variables and investigated the potential impact of an interaction between age and gender/parental status. Even when controlling for employees’ age and job rank, the results revealed a significant main effect of gender/parental status, F(3, 1,333) = 5.55, p < .001, but no significant interaction between age and gender/parental status, F(3, 1,333) = 1.47, p = .22. Bonferroni pairwise comparisons detail the patterns in the data, demonstrating that non-parent women earned less than non-parent men (MΔ = -1.07, p < .001) and working fathers (MΔ = -1.88, p < .001). Similarly, working mothers earned less than non-parent men (MΔ = -0.72, p = .01) and working fathers (MΔ =-1.54, p < .001). There was no significant difference in salaried income between working mothers and non-parent women (MΔ = 0.35, p = .84). However, working fathers earned more annually than non-parent men (MΔ = 0.82, p = .003).
  24. Salaried income was measured using sequential pay bands, with the first category capturing annual salary below $20,000 and subsequent categories increasing by increments of $4,000 (e.g., $21,000 - $25,000, $CAD 26,000 - $CAD 30,000). A one-way ANOVA was performed to examine the impact of gender and parental status on hourly income in the US. In our model, we combined gender and parental status into one variable (i.e., “gender/parental status”) with four categories: working mothers, working fathers, non-parent women, and non-parent men. We also included age and job rank as control variables and investigated the potential impact of an interaction between age and gender/parental status. After accounting for employees’ age and job rank, the results showed a significant main effect of gender/parental status, F(3, 927) = 4.67, p = .003, but no significant interaction between age and gender/parental status, F(3, 927) = 1.38, p = .25. Bonferroni pairwise comparisons detail the patterns in the data, showing that non-parent women earned less than non-parent men (MΔ = -1.42, p < .001), working fathers (MΔ = -2.08, p < .001), and working mothers (MΔ = -1.11, p = .001). Furthermore, working mothers earned less than working fathers (MΔ =-0.98, p = .01), but there was no significant difference in salaried income between working mothers and non-parent men (MΔ = -0.32, p = 1.0). Finally, there was no significant difference in salaried income between working fathers and non-parent men (MΔ = 0.66, p = .05).
  25. National Employee Benefits Day: Focus on frontline employees. (2025, April 2). Catalyst.
  26. A chi-square test of independence was conducted to examine the distribution of salaried frontline employees across Canada, the UK, and the US. The analysis revealed a significant association, χ² (2) = 401.89, p <.001, where 49% of UK frontline employees are paid by annual salary, versus 26% of US and 32% of Canadian frontline employees. Standardized residuals indicate that more UK employees were salaried than expected by chance (z = 12.8), while fewer US employees were salaried than expected (z = –9.7). The difference for Canada was not statistically significant (z = –1.8).
  27. A chi-square test of independence revealed a statistically significant association between working in the UK and annual earnings, χ² (2) = 801.11, p < .001, indicating that frontline employees in the UK are more likely to earn less than £40,000/year. The standardized residuals show substantial deviations: UK employees were significantly overrepresented among those earning below £40,000 (64%, z = 16.4) and underrepresented among those earning above £40,000 (36%, z = –12.7). In contrast, US and Canadian employees were significantly more likely to earn above £40,000, with residuals of 8.1 and 8.6, respectively.
  28. A chi-square test of independence revealed a statistically significant association between country and hourly pay level, χ² (2) = 600.65, p < .001, with the UK showing the most pronounced disparity. UK employees were significantly overrepresented among those earning below £20/hour (80%, z = +12.9) and underrepresented among those earning £20/hour or more (20%, z = –14.4). These findings suggest that frontline employees in the UK are disproportionately concentrated in lower-paid hourly roles, highlighting a structural pay disadvantage relative to their US and Canadian counterparts.
  29. A chi-square test of independence was conducted to examine the distribution of frontline employees paid hourly across Canada, the UK, and the US. The analysis revealed a significant association, χ² (2) = 401.89, p <.001, where 51% of UK frontline employees are paid on an hourly basis compared to 74% in the US and 68% in Canada. Standardized residuals indicated that fewer UK employees were paid hourly than expected by chance (z = –9.4), while the US had significantly more hourly employees than expected (z = 7.1). The residual for Canada was not statistically significant (z = 1.3), suggesting no meaningful deviation from expected values.
  30. A chi-square test of independence revealed a statistically significant association between country and hourly pay level, χ² (2) = 600.65, p < .001, with Canada showing a strong upward earnings trend. Canadian employees were significantly overrepresented among those earning $CAD20/hour or more (64%, z = 10.8) and underrepresented among those earning below that threshold (36%, z = –9.7). These results indicate that Canada offers a more favorable hourly pay structure for frontline employees, with a higher proportion reaching the upper wage bracket compared to the UK and US.
  31. A chi-square test of independence revealed a statistically significant relationship between country and annual earnings, χ² (2) = 801.11, p < .001, with Canada showing a distinct pattern. Canadian employees were significantly overrepresented among those earning above $CAD40,000 (89%, z = +8.6) and underrepresented among those earning below $CAD40,000 (11%, z = –11.1). These findings suggest that relative to the UK and US, Canada offers a more favorable pay structure for frontline employees, with a higher proportion reaching the upper income bracket.
  32. A chi-square test of independence confirmed a statistically significant relationship between country and hourly pay level, χ² (2) = 600.65, p < .001, with the US showing a more balanced distribution. US employees were slightly underrepresented among those earning below $20/hour (52%, z = –2.6) and slightly overrepresented among those earning $20/hour or more (48%, z = 2.9). While these residuals are modest, they suggest that US frontline workers are more evenly distributed across pay brackets, with a slight tilt toward higher hourly earnings compared to the UK.

  33. A chi-square test of independence confirmed a statistically significant association between country and annual earnings, χ² (2) = 801.11, p < .001, with the US displaying a strong upward earnings trend. US employees were significantly more likely to earn above $40,000 (82%, z = 8.1) and less likely to earn below that threshold (18%. z = –10.4). These results indicate that frontline employees in the US are more likely to access higher annual earnings compared to their UK counterparts.

  34. Turner et al. (2025); A world not designed for us: Annual survey 2024. (2024). Young Women’s Trust; Moyser (2019); Girls on the job: Realities in Canada. (2019). Girl Guides; Fry (2022).
  35. Pay transparency – no more secrets? (n.d.). KPMG; Gender Pay Gap Information Regulations 2017: Summary of reported data for 2017/18. (2018). Government Equalities Office.
  36. Halls, K. & Aranzana, J. G. (2025). Proposals to improve pay transparency in the UK: A lesson from the EU? Linklaters.
  37. Grissom, A. (2024, April 17). Time is ticking for the EU Pay Transparency Directive. Catalyst; EU pay transparency progress: A year of action and growth. (2025, May 5). Catalyst.
  38. Bernard, B. (2024, March 21). Salary transparency increasingly the norm in Canadian job postings. Hiring Lab: Economic Research by Indeed; Starner, T. (2023, November 10). Ontario set to join pay transparency movement in Canada. World at work; Damante, B., Hoffman, L., & Khattar, R. (2023, March 9). Quick facts about state salary range transparency laws. Center for American Progress.
  39. A chi-square test of independence showed a significant association between job rank and earning below or above £/$/CAN 20/hour across Canada, the UK, and the US, χ²(2) = 286.63, p < .001. Non-managers without supervisory responsibilities were significantly more likely to earn below £/$/CAN 20/hour (67%, z = 7.7), while first-level managers were significantly more likely to earn £/$/CAN 20/hour or above (59%, z = 9.2). Supervisors were close to expected distributions (z = ±1.3), indicating a more balanced wage spread. The strong deviations from expected counts suggest a clear upward wage shift as employees transition from non-management to managerial roles.
  40. A chi-square test of independence showed a significant association between job rank and earning below or above £/$/CAN 40,000/year across Canada, the UK, and the US, χ²(2) = 70.60, p < .001. Non-managers without supervisory responsibilities were significantly more likely to earn below £/$/CAN 40,000/year (52%, z = 5.9).  Supervisors and first-level managers were similarly likely to earn £/$/CAN 40,000/year or above (supervisors: 66%, z = 1.3; first-level managers: 66.5%, z = 1.9); however, the standardized residuals for these were below the threshold for statistical significance. This trend suggests that the key shift in earnings occurs when employees move into roles with supervisory responsibilities. While supervisors and first-level managers do not significantly differ from one another in terms of higher salaries, both groups are more likely than non-managers to earn above £/$/CAN 40,000 per year. In other words, the wage increases appear to be tied less to the level of management and more to the transition from non-supervisory to supervisory roles.