Disparate Impact and Employment Actions During COVID-19

In response to the COVID-19 pandemic, employers faced and continue to face unprecedented employment decisions across many industries throughout the United States. These decisions include whether to lay off or furlough employees, whether to change employees’ pay or hours, as well as how to staff for partial or full re-openings. Besides concerns over profitability and the health of their employees, employers also should be mindful of the potential for employment decisions to have disparate impact on employees and job applicants. The Equal Employment Opportunity Commission (“EEOC”) and Office of Federal Contract Compliance Programs (“OFCCP”) are continuing to enforce and ensure compliance with employment non-discrimination laws during the pandemic.

Disparate impact occurs when an employer’s policies or practices are facially neutral but have a disproportionately negative impact on individuals based on their race, color, religion, sex including sexual orientation and gender identity, national origin, age, disability status, or veteran status (if the employer is a federal contractor). Numerous employment practices or decisions in response to the COVID-19 pandemic could lead to disparate impact, even though they appear to be facially neutral. For example, employers may seek to prioritize retention of more experienced workers due to their firm-specific knowledge or skills. However, if an employer had recently expanded its diversity outreach efforts and hired a more racially diverse workforce leading up to the onset of the COVID-19 pandemic, its retention efforts may have disparate impact on one or more racial groups. Additionally, shutdowns and rehiring based on geography or company division may lead to disparate impact if targeted locations or divisions are disproportionately staffed with an older or more diverse workforce. Further, laying off a disproportionate number of part-time employees may adversely impact women, who according to 2019 data from the Bureau of Labor Statistics (“BLS”) are twice as likely as men to be working in or seeking part-time employment.

Determining whether there is disparate impact typically requires measuring the selection rate for each group within a particular category, such as the number of hires by race. The selection rate for each group is then compared to the selection rate for the most-favored group. Among positive employment outcomes such as hiring, the most-favored group has the highest selection rate (for example, the group with the most hires per 100 applicants). Among negative employment outcomes such as layoffs or reduced hours, the most-favored group has the lowest selection rate. For example, given women’s greater propensity to work part-time, the most-favored group associated with an employer’s layoff of part-time workers is likely to be men. Such a layoff is likely to be subject to greater scrutiny or possible legal challenge if the selection rate for a given group (e.g. women) is statistically significantly higher than that of the most-favored group (e.g. men). Another guideline utilized to indicate disparate impact – useful in some but not all situations, and not dispositive in and of itself – is whether any given group’s selection rate is less than 80% or four-fifths of the selection rate of the most-favored group.

However, accurate determination of whether a facially neutral selection process has disparate impact on a particular group requires more than just a simple comparison of selection rates. Additional statistical considerations also should be incorporated to the extent possible, such as limiting comparisons to similarly situated employees, capturing the process or levels at which decision making occurs, and using techniques such as Fisher’s Exact test or regression analysis. For example, using a regression analysis, an employer may compare hiring rates across groups by looking within each organizational unit or job title and controlling for relevant factors such as performance ratings, tenure, and location. A Fisher’s Exact test, which compares frequencies, may be more suitable than regression analysis or other statistical tests when sample sizes are small or one or more selection rates are low. Employers, as well as employees seeking to challenge an employment practice, should consider the practical significance as well as the statistical significance of potential disparate impact. Practical significance can be measured by comparing the difference in selection rates across groups and assessing the shortfall of a group’s actual selection rate relative to its expected selection rate.

Importantly, even if an employer’s selection procedure results in disparate impact, it may still be permissible if the employer can show that the procedure is job-related and consistent with business necessity and that there are no less discriminatory alternatives available. For example, a café re-opening for breakfast and lunch service may hire only individuals that are available to work on-site between the hours of 7 a.m. and 2 p.m. This may disparately impact certain gender, race or national origin groups if those groups are more likely to need to work remotely or flexibly due to child care, distance learning or other familial responsibilities during the pandemic. However, the café may be able to successfully defend its hours as a business necessity for which there is no less-discriminatory alternative.

The COVID-19 pandemic also may increase challenges faced by businesses such as restaurants as they re-open indoor and outdoor service and now confront a changed labor pool. Such businesses may confront greater labor supply from the large number of unemployed workers and recent college graduates seeking employment but reduced labor supply among former workers who are older or have underlying health conditions and opt to drop out of the labor force or seek employment in an industry perceived as posing less COVID-19-related risk. Hiring from a younger-on-average external applicant pool rather than rehiring former employees could lead to claims of disparate impact or even disparate treatment (i.e., intentional discrimination).

The effects of COVID-19 will continue to impact employers and employees in myriad ways for the foreseeable future. Many industries are likely to experience additional business re-openings, new shutdowns, hiring, and layoffs as the pandemic continues to have an uneven impact across states. Careful consideration and statistical analysis of COVID-19- related employment decisions can assist in preventing disparate impact on any particular group of employees.

EI Vice President Erica E. Greulich is an empirical microeconomist who consults on employment-related issues including matters of alleged discrimination due to disparate impact.