Written Testimony of Janese Murray Vice President, Diversity & Inclusion Exelon Corporation

Hearing of March 16, 2016 - Public Input into the Proposed Revisions to the EEO-1 Report

Introduction

Chair Yang and distinguished Commissioners, thank you for the opportunity to provide my assessment of the proposed changes to the EEO-1 form to collect pay data. I am Janese Murray, Vice President of Diversity and Inclusion for Exelon Corporation. Exelon is a Fortune 150 energy company doing business in 48 states, D.C., and Canada and employing approximately 30,000 people nationwide.

At Exelon, I am responsible for leading a team in the development and execution of effective strategies to increase workforce diversity and create an inclusive environment. In my more than 30 years' experience in the field of HR, I have worked in talent acquisition, employee relations, HR operations, and executive onboarding and support. I am also a personal and professional coach. I appear before you today on behalf of the Society for Human Resource Management, where I have been a member for over 20 years.

SHRM strongly supports nondiscrimination in compensation policies and practices and believes that compensation decisions should be based on an individual's qualifications and ability to perform a job, not on characteristics that have no bearing on job performance. To assist HR professionals, SHRM provides a variety of educational resources for its members on issues related to compensation and nondiscrimination including seminars, articles, toolkits, webinars and conference programming. Compensation and nondiscrimination are topics addressed at nearly all SHRM conferences, including the role of compensation as a talent management strategy.

SHRM plans to submit detailed written comments on the Commission's proposed changes to the EEO-1 form. In my testimony today, I will not go into as great of detail as our written comments; instead, I will focus on the real-life experience of the HR professional when creating compensation structures and determining employee pay. I will also address the limitations of the information sought by the Commission and the challenges HR will face collecting and reporting the data.

After reviewing the proposed data collection, I believe the Commission should look critically at whether the cost to employers of this significant increase in data reporting can be justified when its utility in reaching the Commission's anti-discrimination objectives is far from clear.

The Role of Compensation in an Organization's Talent Management Strategy

One of the key focuses for HR overall and certainly for me, as the head of diversity and inclusion, is managing talent in order to create a world-class work environment. One aspect of talent management includes creating an effective total rewards strategy to recruit and retain employees that is made up of compensation, benefits, personal growth opportunities, and increasingly, workplace flexibility options. I'd like to spend a few minutes describing how organizations approach the development of a total rewards strategy because I think it helps provide context for understanding the role compensation plays in the workplace.

In developing a total rewards strategy, HR seeks to provide the employer and employees with an approach for compensating employees that is compatible with the organization's mission and strategy and culture, appropriate for the specific workforce, and also internally and externally equitable.

The degree of market competition, the level of product demand, and industry characteristics all have an influence on compensation and benefits philosophy. In order to effectively recruit new employees and retain existing ones, an organization must have internal equity, where employees feel that performance or job differences result in corresponding differences in rewards that are fair. Organizations also must ensure external equity where an organization's compensation levels and benefits are competitive with organizations in the same labor market that compete for the same employees. An organization is likely to use a combination of strategies in approaching pay. For example, for critical jobs and competencies, the organization may decide to lead the competition in compensation whereas in other areas, the organization may match what its competitors are paying employees.

Once an organization has defined its compensation philosophy, HR typically creates a pay system which consists of grouping jobs into pay grades and creating a pay range that sets the upper and lower limits of compensation for employees in each grade. The midpoint is often considered the market rate paid to an experienced employee meeting performance expectations.

A well-designed pay system not only helps attract employees but also plays an important role in motivating and retaining employees. Base pay is, therefore, affected by other types of pay. For example, merit pay or pay for performance ties subsequent wage increases to performance and the degree to which job mastery is attained. Other common differentiators include productivity-based pay determined by the employee's output, such as a piece-rate system, as well as person-based pay which ties pay to desired employee characteristics such as knowledge, which includes certifications and other education credentials; skills; and competencies that an individual employee may possess, such as experience directing or training others.

Of course, a variety of pay adjustments also affect take-home pay: cost-of-living adjustments; general pay increases based on local competitive markets; seniority increases; lump-sum and performance bonuses; as well as differential pay based on the type of work. Differential pay includes additional pay for less desirable shifts; emergency shifts; premium pay for working holidays or extra hours; hazard pay; on-call pay; reporting pay; travel pay; and overtime pay. Geographic differential pay includes accommodating cost-of-living in different locations; attracting workers to certain locations; or foreign pay. Incentive pay for meeting organizational goals in productivity or sales is also common.

As you can see, employers design their pay structures to reflect the characteristics of their jobs, interest qualified applicants and retain qualified employees who are attracted to the mix of work and rewards offered by that employer.

Pay structures must be reevaluated over time to ensure the ranges remain both internally equitable and externally competitive. In fact, an essential part of maintaining equity and fairness in the workplace is regular evaluation of the organization's total reward strategy-including pay, benefits, performance appraisals, professional development and other career opportunities. It is also important that organizations share their compensation philosophy throughout the organization and are transparent with regard to how pay decisions are made.

Now that we have explored how companies set pay, I think it's important to acknowledge the raw wage gap that exists between men and women. There have been numerous studies analyzing the pay gap, yet disagreements exists as to the size of the gap.  Furthermore, a complete explanation of the reasons for the pay gap has remained elusive.  Some of the most recent work in the area of determining the factors that influence pay differences between men and women, points to more nuanced factors. Claudia Goldin, in her research, describes the cost of "temporal flexibility" and Anne-Marie Slaughter similarly refers to the "motherhood" or "care" penalty that leads many women to pursue jobs that prioritize flexibility over salary. One powerful way to decrease that gap, it is argued, is to increase the demand for and value of workplace flexibility.  SHRM, with our partner, the Families and Work Institute (FWI), has championed the creation of flexible workplaces for ALL employees- men and women alike- by providing training materials to help enhance flexibility in all types of workplaces and by honoring employers through our When Works Work Award that are achieving results in this area. The reality is that an employee's own chosen career path is an important driver -- previous jobs, departments, experience, education, and geographic locations all affect pay. Similarly, levels of responsibility, such as the number and type of direct reports, oversight responsibilities for budgets or customer accounts, and performance history affect individual compensation.

From the HR perspective, these differences in knowledge, skill, ability, proficiency, responsibility, and geographic location, provide a legitimate basis for differences in pay among employees doing similar work. The key, however, is figuring out where illegal discrimination may be occurring, if at all. Unfortunately, the data the EEOC wants to collect from employers does not help root out those employers with illegal and discriminatory practices.

Collecting data by EEO-1 category

The EEOC's proposal to collect compensation data at the level of EEO-1 job category is unlikely to shed any light whatsoever on whether an employer's pay practices are discriminatory. This is because each EEO-1 job category includes a wide range of job titles, for which vastly different rates of pay are provided based on a variety of legitimate, nondiscriminatory factors. In Exelon's case, for example, the Company reported just under 10,000 employees in the Professionals EEO-1 category on its 2015 EEO-1 reports. These individuals include many of Exelon's engineers, lawyers, financial analysts, and HR and compliance professionals, along with our only pilot, a handful of meteorologists and many other unique positions. It goes without saying that the pay for each of these positions, even though they are within the same broad EEO-1 job category, varies significantly because these individuals perform very different work. However, there is no way for the EEOC to understand this legitimate reason for pay differences within this job category under the proposed data collection.

In addition, reporting compensation data by EEO-1 establishment compounds this problem. For example, another employer in the same industry as Exelon may have a single establishment where most IT professionals work. Yet Exelon has its IT professionals scattered among many different establishments, such that the compensation data of our IT professionals will be reported with that of other employees in the Professionals EEO-1 job category within each establishment. Even assuming Exelon and its competitor employ the same number of male and female IT professionals and pay them at the same rate, the EEO-1 reports for each employer will look vastly different because the relevant data is aggregated by the artificial parameters of EEO-1 job category and establishment location.

Collecting aggregated W-2 gross income

Similarly, the agency's proposed collection of W-2 gross income information is misplaced. As the EEOC recognizes, W-2 gross income includes other non-discriminatory variables that may impact pay, including shift differentials, bonuses, commissions, and overtime compensation. As a result, while this data may provide the agency a broader view of pay practices, collecting this data will not allow the EEOC to evaluate comparable compensation data points.

For example, two production workers could have different W-2 gross wages if one was excused from working overtime hours as a reasonable accommodation under the Americans with Disabilities Act ("ADA"), while another not only worked continuously throughout the year but also worked all overtime hours offered to her. Providing hours worked by both employees does not adequately account for the differences in pay because there is no way to account for the fact that some of the hours of one employee were paid at a premium rate, while the other employee asked to be excused from all overtime hours for a legitimate, nondiscriminatory reason.

Likewise, two employees in the same job title may report different W-2 gross wage information in a calendar year if one of the employees receives a $25,000 signing or retention bonus that year, and the other did not. This is the case even if the other employee received the same $25,000 signing or retention bonus when he or she began employment in a different EEO-1 reporting year. If the two employees are of different races or genders, aggregating the W-2 wage information of these two employees will make it appear as if there is a potential pay discrimination issue. Again, reporting total hours worked for these two employees would not account for the legitimate, nondiscriminatory reason for the difference in pay.

Collecting total hours worked

EEOC's proposal would require that employers report actual hours worked by race/ethnicity and gender in each EEO-1 job category. Most SHRM members, Exelon included, do not collect actual hours worked data for exempt employees. Exempt employees are compensated for their performance and accomplishing organizational goals, not simply for hours worked. The burden associated with collecting actual hours worked for exempt employees, and the impact it has on other compliance obligations as well as overall company culture, should not be underestimated. Even without further study, it is clear that using a standard hours figure, such as 2080 hours for all exempt employees, would not be a reliable approximation of the number of actual hours worked by these employees, making the collection of such data for most exempt employees utterly useless for the EEOC's stated purposes.

The reality of data collection and reporting under the current EEO-1 report

Underlying this data collection proposal is the flawed assumption that accurate and fully verified EEO-1 reports can be automatically generated using technology. This is simply not the case currently and certainly will not be under the proposed changes which will require two different systems - HRIS and Payroll - to talk to one another. This is particularly troublesome for smaller employers that may lack the funds to change their data management systems or to hire an outside consultant.

As anyone who has called their IT department helpdesk knows, technology is only as good as the human being who is using it. Studies on topics from network security to the accuracy of hospital medication lists, show that human factors play a more important role in data accuracy than technology factors. This is even more complicated for large employers like Exelon, where many people at different locations may be entering employee data into the Company's HRIS system. For this reason, even employers that are able to generate reports from an HRIS system do not submit them without first checking their accuracy. Companies like us use the annual EEO-1 report filing process to ensure that the data has been entered correctly and consistently across our many establishments.

I will use my company's experience as an example. Exelon uses an outside company, Berkshire Associates, to help us file our annual EEO-1 reports. In 2015, we filed 415 such reports using the batch upload method. In order to file these reports, Exelon generated a roster of our active employees, a job code table which lists job codes/titles and their corresponding EEO-1 categories from our HRIS system, and an establishment ID table listing Exelon's 1,517 locations. Berkshire then reviews and verifies the underlying data-such as making sure that all employees are assigned a valid race and sex, and that the EEO-1 categories for each job code/title are appropriate.

If data is missing, Exelon's HR, often with the help of supervisors, must supply the information. Our HRIS system cannot answer these types of verification questions with the push of a button - it requires manual review by one or more individuals on the Exelon team. After additional back and forth and comparing the 2015 data against the 2014 reports to check for errors or anomalies, Berkshire tests the draft report to ensure that it will be uploaded to EEOC's system appropriately and then emails the file to EEOC.

Despite the rigorous data checking prior to submission, when notified by EEOC a month after submission that the report had been received, 20% of the individual reports were still marked as incomplete and had to be manually reviewed by our vendor. Most locations were marked incomplete due to the fact that the locations were now closed. Although our vendor had sent the EEOC a closed location file, the agency failed to mark these locations as closed when it uploaded the file. As a result, our vendor had to manually review and then 'close' 69 locations. I share this with you to point out that human error occurs on both side of the process, and can significantly add to the amount of time it takes to file a company's EEO-1 reports.

But, it's not just human error that adds time and expense to filing an EEO-1 report, it is also changes to the filing system. For example, 2015 was the first year that EEOC required a FEIN for each location. This requirement was not reflected in EEOC's file until the end of August, which meant any report pulled from our HRIS did not contain this information and therefore it had to be gathered manually with very little notice. Moreover, the EEOC's requirements file and HRIS reporting systems are not always consistent, leading to an additional round of manual edits before an EEO-1 report can be filed. For example, part of our corrections to our 2015 report required reviewing city names which allegedly did not match the zip code. In many cases, this was because EEOC's batch upload requirements did not match the abbreviation preference used in our HRIS system - for example, St. Louis rather than Saint Louis.

Given Exelon's experience, I can only imagine the frustrations that exist for smaller organizations that file the reports themselves. Based on Exelon's experience, EEOC's estimate that it takes each filer - regardless of company size or number of reports filed - 3.4 hours to file the current report is way too low. Berkshire spent 22.93 hours filing and certifying our 2015 reports-even though we had time efficiencies because Berkshire had already verified much of our data when they prepared our affirmative action plan updates.

Adding salary data and hours worked information into the mix will add time and cost, well beyond an additional few hours, particularly since many employers do not store EEO-1 data and payroll data in the same system. Moreover, EEOC's estimate seems to mistakenly assume that it is just as easy to verify compensation data as it is to verify race and gender information. This is, unfortunately, not the case, particularly where the compensation data being collected will cross two calendar years for every employer who is required to file the revised EEO-1 report. By mixing time frames between two taxable years, the picture provided may not be accurate depending on the timing of employer raises, bonuses or other compensation-related variables.

SHRM is conducting an informal survey of members looking at how they currently file EEO-1 reports. Preliminary responses of 250 HR professionals indicate that 85% do not use an outside vendor to help file the reports and 33% either do not have an HRIS system at all or use different vendors for their HRIS and payroll systems. Given these realities, it is not surprising that over 90% of HR professionals either don't know how long it will take to file the new form or estimate that it will take longer than EEOC's estimate of 6.6 hours. Although informal, this survey clearly shows the challenges employers anticipate with this proposal.

Concerns about confidentiality

SHRM and its members are very concerned about the confidentiality of the compensation data the EEOC proposes to collect. The EEOC's proposal would gather very specific compensation information by specific establishments, including very small establishments, using a web-based format. For many small employers, and even larger employers with small establishments such as Exelon, reporting data in this manner will result in the reporting of individual, employee-level data. Our concerns are not just focused on protecting our companies, but also on protecting our employees, many of whom would not be happy if their personal pay information was widely disclosed as a result of a data breach of the EEO-1 reporting system.

Furthermore, large employers like Exelon currently email their EEO-1 Reports to EEOC for batch uploading. It goes without saying that this is obviously not a secure way to transmit large amounts of confidential information, yet EEOC's proposal makes no mention of how the agency plans to revise its own protocols to ensure that employers can safely report their compensation information to the government. SHRM and its members urge the EEOC to not move forward with the implementation of any compensation data collection tool until appropriate data security safeguards are developed, tested, and perfected to ensure protection of our employees' pay data.

Conclusion

In closing, I would like to commend the EEOC on the important role it plays in protecting the civil rights of individuals in the workplace. As a diversity and inclusion professional, I am focused on creating systems and a workplace culture that promote equity and fairness so that my organization can manage its talent in a way that results in organizational success. I have seen great strides in this area.

Our company has an enterprise-wide goal for organizational success around D&I and our CEO is holding each of his direct reports accountable for reaching this goal. Each of the CEO's direct reports has a diversity and inclusion plan for their function and is responsible for reporting progress on a quarterly basis. We track hiring, retention and promotion rates for diverse talent and each of our operating companies periodically analyze their pay data for any gaps in pay equity and adjust as necessary. We also include training focused on preventing workplace discrimination and harassment in several of our leadership development modules and as part of our code of business conduct certification process.

The current National discussion on pay equity is important and one that many of us relate to on a personal level. Unfortunately, collecting pay data in the highly aggregated manner proposed will not prove useful to achieving the objective of rooting out unlawful compensation discrimination. Compounding this issue is the fact that the burden on employers of collecting and reporting the information required is far greater than what the agency has estimated. For these reasons, I believe the EEOC should not expand the EEO-1 form to collect this data.

Thank you for your time. I appreciate the opportunity to share my perspective with you today and would be happy to answer any questions.