Inequalities in the Service Sector

Add worker quote

Understanding Inequality

Women and people of color are over-represented in the service sector, occupying a wide array of roles, from retail and hospitality to food service. Many of these positions are understood as “bad jobs”, as defined by job quality measures such as job insecurity, limited benefits, and low wages. One important part of the story of inequality in America is the segregation of women and people of color into this sector of the economy.

As the Shift Project has delved deeper into this sector, we have discovered that inequalities exist not only in the sector as compared to others, but also within the sector and even among co-workers at the very same firms. Dissecting the data to understand the scope and shape of these inequities, we have examined disparities between different groups across scheduling, wages, Paid Sick Leave (PSL), Paid Family and Medical Leave (PFML), and breaks, among other job quality measures.

Examples of Inequality in Shift Data:

Racial/Ethnic Inequality in Job Quality

Workers of color, and especially women of color, are between 3 and 5 percentage points more likely to face schedule unpredictability across five different measures, including X, Y, and Z.

Gender Gap in Paid Sick Leave Access

Paid sick leave (link to that page)_provides crucial support for workers who need to stay home to recover from illness or to care for family members. But, despite bearing much of the load on such care work, women hourly workers in the service sector are significantly less likely to have access to paid sick leave than their male co-workers.  

The Role of Firm Segregation

Inequalities in job quality exist within the service sector and one key driver of these inequalities is that women, mothers, and people of color are segregated into employment at the firms in the service sector that have the most precarious jobs.  

It works like this:

  • Job quality actually varies a lot between firms in the service sector, even firms in the very same sub-sector. 
  • Workers from marginalized groups – like mothers and workers of color – are disproportionately represented at these lower-road businesses – segregated. 
  • The result is that these groups have more precarious jobs and that this segregation accounts for an important part of the inequality between groups.

Firms Segregation and Gaps in Work Schedule Quality

Shift researchers[1] found that workers of color, especially women of color, had significantly more precarious work schedules – things like on-call shifts and limited advance notice. While some of these gaps were explained by occupation and education, we discovered that the segregation of workers of color into the specific firms with the most precarious scheduling practices explained an important part of the gap.  

Firm Segregation and the Gender Wage Gap

There is a large gender wage gap in the service sector, about $X/hr[SD1] .  Just like with racial/ethnic gaps in scheduling, Shift researchers found that the segregation of women into lower-paying firms explained an important part of that gap.  Another way to say that is that hourly wages declined with the percentage of the firm’s workforce who were identified as female, as shown below.

[1] ASR on gender/race inequalities in scheduling

 [SD1]Please fill in from the paper with Carmen Brick

We have found that this between-firm variation also contributes to inequality by gender between men and women[1][2] and between cisgender and transgender/nonbinary people[3], the intersection of race and gender[4][5], and motherhood[6].

[1] W&O on gender inequalities in wages

[2] Shift brief on gender and PSL (

[3] Working paper on TNB inequality and job quality

[4] ASR on gender/race inequalities in scheduling

[5] Breaks paper

[6] Working paper on motherhood inequalities in wages

Durable Inequality

Across all papers addressing inequality at the Shift Project, we find that inequality between groups persists across models. It remains after controlling and adjusting for individual differences, demographics, and labor market conditions, and after adjusting for segregation into lower-quality jobs.  This unexplained inequality points to discrimination. This is not to say that some of these other factors are not also discrimination. Human capital factors such as educational inequality and sorting and segregation into lower-quality jobs are two clear examples of upstream discrimination that impact service sector workers. However, it is important to know that these upstream inequalities cannot fully explain service sector disparities. Unexplained discrimination in our data often occurs at the managerial level, with items that allow for managerial discretion such as scheduling often seeing larger disparities between groups, as opposed to more standardized items such as wages and benefits.

Policy and Labor Standards

What helps to close these gaps?

Shift Project research shows public policy has a key role to play.  By passing and implementing labor standards that “raise the floor” on job quality, government can also “close the gap.” 

Key findings:

  • Labor standards serve to “raise the floor” by establishing basic rights and protections for all workers
  • These polices can also “narrow the gap” by mitigating the concentration effects in certain service sector jobs and by reducing inequalities within the sector itself.

Raising the Floor on Breaks

Shift researchers’[1] finding that when the floor was raised for breaks policies, women and people of color disproportionately benefitted overall due to their concentration in affected jobs.

[1] Breaks paper

Narrowing the Gap in PSL

Shift researchers[1] found stark gender inequality in access to paid sick leave among service sector workers, with only part of this gap explained by sorting and position type. However, in states and localities with mandatory paid sick leave laws, workers were much more likely to report access to paid sick leave and the gender gap in access disappeared.

[1] Health affairs gender PSL

Want to learn more?

To read more on this work, see: