2016-16692. Agency Information Collection Activities; Notice of Submission for OMB Review, Final Comment Request: Revision of the Employer Information Report (EEO-1)

AGENCY:

Equal Employment Opportunity Commission.

ACTION:

Notice.

SUMMARY:

In accordance with the Paperwork Reduction Act of 1995 (PRA), the Equal Employment Opportunity Commission (EEOC or Commission) announces that it is submitting to the Office of Management and Budget (OMB) a request for a three-year PRA approval of a revised Employer Information Report (EEO-1) data collection. Employers have submitted the EEO-1 report for over fifty years. The Commission is responsible for PRA compliance for the EEO-1, although it is a joint data collection to meet the statistical needs of both the EEOC and the U.S. Department of Labor's Office of Federal Contract Compliance Programs (OFCCP). This PRA submission has two components. Component 1 describes the data now collected by the currently approved EEO-1, which is data about employees' ethnicity, race, and sex by job category (demographic data). Component 2 describes the W-2 (Box 1) and hours-worked data that will be added to the EEO-1 with OMB's approval under this PRA request (pay data). EEO-1 respondents must comply with the 2016 filing requirement for the currently approved EEO-1.

DATES:

Submit comments on or before August 15, 2016.

ADDRESSES:

Comments on this notice must be submitted to Joseph B. Nye, Policy Analyst, Office of Information and Regulatory Affairs, Office of Management and Budget, 725 17th Street NW., Washington, DC 20503, email [e-mail]. Commenters are also encouraged to send comments to the EEOC online at http://www.regulations.gov,, which is the Federal eRulemaking Portal. Follow the instructions on the Web site for submitting comments. In addition, the EEOC's Executive Secretariat will accept comments in hard copy by delivery by COB on August 15, 2016. Hard copy comments should be sent to Bernadette Wilson, Acting Executive Officer, EEOC, 131 M Street NE., Washington, DC 20507. Finally, the Executive Secretariat will accept comments totaling six or fewer pages by facsimile (“fax”) machine before the same deadline at [phone]. (This is not a toll-free number.) Receipt of fax transmittals will not be acknowledged, except that the sender may request confirmation of receipt by calling the Executive Secretariat staff at [phone] (voice) or [phone] (TTY). (These are not toll-free telephone numbers.) The EEOC will post online at http://www.regulations.gov all comments submitted via this Web site, in hard copy, or by fax to the Executive Secretariat. These comments will be posted without change, including any personal information you provide. However, the EEOC reserves the right to refrain from posting libelous or otherwise inappropriate comments including those that contain obscene, indecent, or profane language; that contain threats or defamatory statements; that contain hate speech directed at race, color, sex, national origin, age, religion, disability, or genetic information; or that promote or endorse services or products. All comments received, including any personal information provided, also will be available for public inspection during normal business hours by appointment only at the EEOC Headquarters' Library, 131 M Street NE., Washington, DC 20507. Upon request, individuals who require assistance viewing comments will be provided appropriate aids such as readers or print magnifiers. To schedule an appointment, contact EEOC Library staff at [phone] (voice) or [phone] (TTY). (These are not toll-free numbers.)

FOR FURTHER INFORMATION CONTACT:

Ronald Edwards, Director, Program Research and Surveys Division, Equal Employment Opportunity Commission, 131 M Street NE., Room 4SW30F, Washington, DC 20507; [phone] (voice) or [phone] (TTY). Requests for this notice in an alternative format should be made to the Office of Communications and Legislative Affairs at [phone] (voice) or [phone] (TTY).

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Background

II. The EEOC's Legal Authority To Propose This EEO-1 Report

A. Title VII of the Civil Rights Act of 1964

B. The Paperwork Reduction Act of 1995

III. Revisions to the EEO-1 Report Are Necessary for the Enforcement of Title VII, the EPA, and Executive Order 11246

IV. Who Will Report Pay Data on the Revised EEO-1

A. Employers That Currently File the EEO-1

B. 60-Day Notice: Which Employers Would File Pay Data

C. Public Comments

D. 30-Day Notice: Employers With 100 or More Employees Will File Components 1 & 2

V. When To File: Filing Deadline and Workforce Snapshot Period

A. 60-Day Notice

B. Public Comments

C. 30-Day Notice

1. Deadline for Filing the EEO-1

2. “Workforce Snapshot” Period

VI. What Pay Data To Report: Measure of Pay for the EEO-1

A. 60-Day Notice: Options for Measuring Pay

B. Public Comments

1. Supporting the Use of W-2 Income

2. Opposing the Use of W-2 Income

C. 30-Day Notice: W-2 (Box 1) Income Is the Measure of Pay

1. W-2 Income and Employee Choice

2. Supplemental Income Is Important and May Be Linked to Discrimination

3. Bridging HRIS and Payroll

VII. What Data To Report: Hours Worked

A. 60-Day Notice

B. Public Comments

C. 30-Day Notice

1. The Importance of Collecting Hours Worked

2. Defining “Hours Worked”

3. Reporting Hours Worked for Nonexempt Employees

4. Reporting Hours Worked for Exempt Employees

VIII. How To Report Data in Component 2: Pay Bands and Job Categories

A. 60-Day Notice

B. Public Comments

C. 30-Day Notice

IX. How the EEOC Will Use W-2 and Hours-Worked Data

A. 60-Day Notice

B. Public Comments

C. 30-Day Notice

1. Early Assessment of Charges of Discrimination

2. EEOC Publications Analyzing Aggregate EEO-1 Data

3. EEOC Training on the Pay Data Collection

X. Confidentiality of EEO-1 Data

A. 60-Day Notice

B. Public Comments

C. 30-Day Notice

1. Legal Confidentiality

a. EEOC

b. OFCCP

2. Data Protection and Security

XI. Paperwork Reduction Act Burden Estimates

A. Background

B. 60-Day Notice

C. 30-Day Notice

1. Annual Burden Hours

2. Hourly Wage Rates

XII. Formal Paperwork Reduction Act Statement

A. Overview of Information Collection

1. 2016 Overview of Information Collection—Component 1

2. 2017 and 2018 Overview of Information Collection—Components 1 and 2

a. Component 1 (Demographic and Job Category Data)

b. Components 1 and 2 (Demographic and Job Category Data Plus W-2 and Hours Worked Data)

B. 30-Day Notice PRA Burden Statement

I. Background

This final proposal to supplement the longstanding EEO-1 employer information report (currently approved by OMB under Control Number 3046-0007) is intended to support the EEOC's pay discrimination investigations by collecting employer- and gender-, race-, and ethnicity-specific pay data to identify pay disparities that may result from discriminatory practices or policies. This Notice provides stakeholders with their second opportunity to comment on this proposal.

The EEOC published the first notice of this proposed revision in the Federal Register on February 1, 2016, for a 60-day comment period (the “60-Day Notice”).^[1] It announced which employers would be required to file pay data, what data would be collected, when the due date would be, how the data would be analyzed, and how the proposed collection and analysis would protect confidentiality and privacy. As required, the 60-Day Notice estimated the cost to employers of completing the current EEO-1 (Component 1) and the proposed revision of the EEO-1 (Components 1 and 2).

The EEOC received 322 timely public comments in response to the 60-Day Notice. The comments were submitted by individual members of the public, employers, employer associations, Members of Congress, civil rights groups, women's organizations, labor unions, industry groups, law firms, and human resources organizations. Over 120 of the 322 comments were part of mass mail campaigns mostly supporting the proposal, although one mass mail campaign opposed the proposal. The mass mail campaigns included submissions from organizations that collected up to thousands of signatures from their members or supporters.

The Commission also held a public hearing on March 16, 2016, and heard from 15 witnesses representing a range of stakeholders including employers, employees, and academics. The Commission reviewed their detailed written submissions, heard them discuss their different perspectives on the proposal, and asked them questions.^[2]

Pursuant to the required procedures under the PRA, the Commission now publishes its final proposal to supplement the EEO-1 for a second round of public comments, to last 30 days (hence the “30-Day Notice”). The EEOC also is formally submitting the proposed EEO-1 revisions to OMB for consideration and decision.

This 30-Day Notice summarizes the 60-Day Notice, describes the public comments, and explains the Commission's decisions. In making these decisions, the Commission took into account all of the hearing testimony and public comments. The Commission also assessed government data regarding components of compensation in United States workplaces, relevant academic literature on compensation practices and on discrimination, and the conclusions of two studies commissioned by the EEOC to examine how and whether to implement a pay data collection.^[3] This 30-Day Notice sets forth the EEOC's conclusions about the ways the proposed pay data collection will be used to enhance and increase the efficiency of enforcement efforts while facilitating employer self-evaluation and voluntary compliance.

II. The EEOC's Legal Authority To Propose This EEO-1 Report

In written comments in response to the 60-Day Notice, several interested parties questioned whether the EEOC has legal authority to collect pay data and whether the agency should have conducted a formal rulemaking to impose a pay data reporting requirement. As explained in more detail below, the EEOC has the legal authority to collect pay data under Title VII of the Civil Rights Act of 1964, as amended (Title VII),^[4] without conducting a formal rulemaking because the EEOC is responsible for enforcing federal laws that prohibit wage discrimination on the basis of sex, race and national origin, and Title VII grants the EEOC broad authority to collect data from employers regarding compliance with federal anti-discrimination laws. The EEOC has exercised this statutory authority by implementing a regulation to establish the EEO-1 reporting requirement, and now administers the EEO-1 report pursuant to the PRA.

A. Title VII of the Civil Rights Act of 1964

The EEOC is responsible for enforcing Title VII, which prohibits all employment discrimination, including pay discrimination, based on race, color, religion, national origin, or sex.^[5] The EEOC also enforces other federal laws prohibiting employment discrimination, including the Equal Pay Act of 1963 (EPA), which prohibits certain gender-based pay discrimination.^[6]

The EEOC's authority to promulgate the EEO-1 report is found in section 709(c) of Title VII, which requires employers covered by Title VII to make and keep records relevant to whether unlawful employment practices have been or are being committed, to preserve such records, and to produce reports as the Commission prescribes by regulation or order, after public hearing, “as reasonable, necessary, or appropriate for the enforcement of this subchapter or the regulations . . . thereunder.” ^[7] The Commission prescribes the EEO-1 report by regulation at 29 CFR part 1602, subpart B, which requires private employers with 100 or more employees to “file [annually] with the Commission or its delegate executed copies of [the] . . . EEO-1 [report] in conformity with the directions set forth in the form and accompanying instructions.” The EEOC administers the EEO-1 jointly with OFCCP, which enforces the employment discrimination prohibitions of Executive Order 11246, as amended, for federal contractors and subcontractors (contractors), including specific provisions regarding pay discrimination and transparency.^[8] OFCCP's regulations require contractors to submit “complete and accurate reports on Standard Form 100 (EEO-1) . . . or such form as may hereafter be promulgated in its place.” ^[9] The Joint Reporting Committee, composed of the EEOC and OFCCP and located at the EEOC, administers the EEO-1 as a single data collection to meet the statistical needs of both agencies while avoiding duplication.

B. The Paperwork Reduction Act of 1995

Since 1995, the EEO-1 report also has been governed by the Paperwork Reduction Act of 1995 (PRA), which provides standards for federal data collections and requires periodic Office of Management and Budget (OMB) review and renewal.^[10] The EEOC is responsible for maintaining PRA approval of the EEO-1.

The EEOC, like other federal agencies subject to the PRA, generally follows a multi-step process for maintaining OMB approval of an information collection, which culminates in OMB deciding if the proposed collection “strikes a balance between collecting information necessary to fulfill [the agency's] statutory mission[ ] and guarding against unnecessary or duplicative information that imposes unjustified costs on the American public.” ^[11] The first step is for the agency to publish a proposed information collection for a 60-day public comment period, which ran from February 1 to April 1, 2016 for this EEO-1 revision.^[12] Then, in light of the public comments and its statutory mission, the agency formulates a final data collection, which it publishes in the Federal Register and submits to OMB for approval, subject to a 30-day public comment period.^[13] The current document, which has been approved by a majority of the Commission, is the EEOC's 30-Day Notice for the revised EEO-1.

The EEOC has consistently used the PRA renewal process to change the EEO-1. Most recently, in 2006, the PRA process was used to significantly revise the EEO-1 by adding a new race category, requiring employers to ask employees to self-identify by race and ethnicity, and requiring employers to ask about ethnicity (Hispanic or Latino) in a separate question.^[14] The 2006 EEO-1 revision also added a new job category.^[15]

III. Revisions to the EEO-1 Report Are Necessary for the Enforcement of Title VII, the EPA, and Executive Order 11246

Some public comments opposing the EEOC's proposal in the 60-Day Notice questioned whether there are still pay disparities that are caused by discrimination linked to gender, race, or ethnicity and, accordingly, whether there is actually a need for more effective enforcement of the prohibitions on pay discrimination in Title VII, the EPA, and E.O. 11246.

Based on federal data and a robust body of research, the Commission concludes that: (1) Persistent pay gaps continue to exist in the U.S. workforce correlated with sex, race, and ethnicity; (2) workplace discrimination is an important contributing factor to these pay disparities; and (3) implementing the proposed EEO-1 pay data collection will improve the EEOC's ability to efficiently and effectively structure its investigation of pay discrimination charges.

First, persistent pay gaps exist in the U.S. workforce correlated with sex, race, and ethnicity. As of 2014, for women of all races and ethnicities, the median annual pay for a woman who held a full-time, year-round job was $39,621, while the median annual pay for a man who held a full-time, year-round job was $50,383.^[16]

African American and Hispanic or Latina women nationwide now experience the largest pay disparities. As of 2014, African American women were paid almost 40% less than white, non-Hispanic, men and approximately 20% less than white, non-Hispanic women.^[17] At a national level, African American women were paid 18% less than African American men.^[18]

Similarly, Latina women were paid approximately 44% less than white, non-Hispanic men, and 27% less than white, non-Hispanic, women in 2014.^[19] The result of the wage gap is that the average Hispanic or Latina woman would be paid approximately $1,007,000 less than the average white, non-Hispanic, male over a 40-year period.^[20]

A similar pattern exists for Native Hawaiian and Pacific Islander women and Native American women who were paid approximately 38% and 41% less than white, non-Hispanic men, respectively.^[21] Asian American women were paid 10% less than white, non-Hispanic men.^[22]

Wage disparities also exist for men of color. In 2014, African American men who worked full time in wage and salary jobs had median weekly earnings of $680, which represented approximately 76% of white men's median weekly earnings ($897).^[23] Hispanic men earned $616, or approximately 69%, of white men's median weekly earnings.^[24]

Employment discrimination may play both direct and indirect roles in creating these pay disparities. Economists Francine Blau and Lawrence Khan found that 64.6% of the wage gap between men and women can be explained by three factors: Experience (14.1%), industry (17.6%), and occupation (32.9%).^[25] Men are more likely to work in blue collar jobs that are higher paying, including construction, production, or transportation occupations, whereas women are more concentrated in lower paying professions, such as office and administrative support positions.^[26] Most of the remaining 35.4% of the gender gap cannot be explained by differences in education, experience, industry, or occupation.^[27] Blau and Khan argue that discrimination—intentional or unintentional, systematic or at the individual level—plays a role in explaining the gap.^[28]

Gender bias may become more obvious when occupations have a greater proportion of women. One study found that, in an occupation dominated by men, pay declines when women enter that occupation in large numbers, even after controlling for factors such as education and work experience.^[29] The opposite effect occurred when a larger proportion of men entered a profession previously dominated by women, i.e., pay increased.^[30]

One way that gender discrimination may influence pay is through implicit or unconscious bias during hiring, promotion decisions, or job assignments.^[31] A study by McKinsey & Company found that women are almost three times more likely than men to have missed out on an assignment, promotion, or increase in wages because of their gender.^[32] Another study shows that women who engage in pay negotiations are more likely than men to face backlash due to gender stereotypes.^[33]

Similar to gender discrimination, racial discrimination may influence pay through implicit or unconscious bias. A series of studies by MIT Sloan found racial bias in salary negotiations even after controlling for the applicants' objective qualifications.^[34] Research by Roland Fryer, Devah Pager, and Jörg L. Spenkuch found that discrimination accounts for at least one-third of the black-white wage gap.^[35] The authors concluded that, compared to whites with comparable resumes, black job seekers were offered lower compensation by potential new employees and were more likely to accept the lower compensation. The researchers found that, although the wage gaps narrow over time as black workers stay at the same job, an unexplained gap nonetheless persists.^[36]

Voluntary compliance is an important part of the effort to prevent discrimination and improve pay equity, and many employers are taking steps to ensure equal pay for equal work. For example, more than 25 companies have signed a White House Equal Pay Pledge to take action to reduce wage disparities in the workplace.^[37] These employers committed to conducting an annual company-wide gender pay analysis across occupations, reviewing hiring and promotion processes and procedures to reduce unconscious bias and structural barriers, and embedding equal pay efforts into broader enterprise-wide equity initiatives.^[38]

There is also evidence that pay equity is good for business. For example, a McKinsey & Company study found that gender parity in the United States could lead to $4.3 trillion of additional GDP by 2025, which is 19% higher than if current trends in pay inequity continue.^[39] Another recent study found that, on average, companies with greater gender diversity outperformed their peers with less diversity over the previous five years, and had a higher return on equity.^[40] The study measured gender diversity according to the following factors: (1) Equality in pay; (2) empowerment (defined as number of women at the highest levels of the corporation and on key committees); (3) representation of women at different levels (including as members of the board of directors, senior executives, and regular employees); (4) work life balance programs; and (5) diversity policies. Pay parity and empowerment were weighted more than the other factors.^[41]

Despite voluntary compliance and the strong business case for fair pay, pay discrimination persists as a serious problem that EEOC and OFCCP are statutorily required to address. The EEOC's mission is to stop and remedy unlawful employment discrimination. The OFCCP's purpose is to enforce, for the benefit of job seekers and wage earners, the contractual promise of affirmative action and equal employment opportunity required of those who do business with the federal government. To fulfill these goals, the EEOC and OFCCP need to be as effective and efficient as possible in their investigations of alleged discrimination. They now lack the employer- and establishment-specific pay data that, prior to issuing a detailed request for information or a subpoena, would be extremely useful in helping enforcement staff to investigate potential pay discrimination. Balancing utility and burden, the EEOC has concluded that the proposed EEO-1 pay data collection would be an effective and appropriate tool for this purpose, for all of the reasons explained below.^[42]

IV. Who Will Report Pay Data on the Revised EEO-1

A. Employers That Currently File the EEO-1

All private employers that are covered by Title VII and have 100 or more employees now file an EEO-1 report about the sex, race, and ethnicity of their employees, which is designated here as Component 1 (demographic data).^[43] Federal contractors with 50 or more employees also file the EEO-1 if they are not exempt as provided for by 41 CFR 60-1.5. Single establishment employers file one EEO-1, and multi-establishment employers file EEO-1 reports or data for each establishment.^[44] Federal contractors with 1 to 49 employees and other private employers with 1 to 99 employees do not file EEO-1 reports.

B. 60-Day Notice: Which Employers Would File Pay Data

In the 60-Day Notice, the EEOC proposed that EEO-1 private employers and federal contractors with 100 or more employees would submit the EEO-1 with pay and hours-worked data (Component 2) in addition to Component 1 data. The 60-Day Notice also stated that federal contractors with between 50 and 99 employees would continue to submit Component 1 data but would not submit Component 2 data.

C. Public Comments

The EEOC received comments urging it to remove employers with fewer than 200, or fewer than 500, employees from the requirement to report pay and hours-worked data on the EEO-1 (Component 2), in order to avoid imposing a burden on them. Some comments also encouraged the EEOC to eliminate the requirement to provide establishment-level pay data for establishments with fewer than 50 or 100 employees. These comments also expressed concern that reporting pay data for small employers, or small employer establishments, could reveal employee-level pay information. Conversely, other comments urged the EEOC to collect data from smaller employers by lowering the reporting threshold for pay data to 50 or more employees for federal contractors.

D. 30-Day Notice: Employers With 100 or More Employees Will File Components 1 and 2

The Commission has considered the arguments for increasing the size of those employers subject to Components 1 and 2 and has decided to retain the same employee thresholds as in the 60-Day Notice. Exempting employers with fewer than 500 employees, or even fewer than 250, from Component 2 would result in losing data for a large number of employers who employ millions of workers, and thus would significantly reduce the utility of the pay data collection. In addition, the EEOC and OFCCP have decided not to exempt federal contractors with 50-99 employees from filing Component 1 of the EEO-1. The Commission's proposal reduces employer burden by changing other aspects of the EEO-1, such as the reporting deadline. See section V.

In sum, all employers with 100 or more employees will be subject to Components 1 and 2 of the EEO-1 starting with reporting year 2017. Federal contractors with 50-99 employees will not experience a change in their EEO-1 reporting requirements as a result of this proposal; they will not file Component 2 and will continue to file only Component 1. Consistent with current practice, federal contractors with 1 to 49 employees and other private employers with 1 to 99 employees will be exempt from filing the EEO-1; they will file neither Component 1 nor Component 2.

V. When To File: Filing Deadline and Workforce Snapshot Period

This 30-Day Notice proposes to change the EEO-1 filing deadline to March 31st^, of the year that follows the reporting year. This Notice also proposes to change the “workforce snapshot” to a pay period between October 1st and December 31st of the reporting year, starting with the EEO-1 report for 2017.

Note that the reporting schedule for 2016 data remains unchanged; EEO-1 respondents must comply with the September 30, 2016, filing requirement for the currently-approved EEO-1, and must continue to use the July 1st through September 30th workforce snapshot period for that report. Under the proposed changes to the reporting schedule, EEO-1 reports for 2017 data would be due on March 31, 2018.

A. 60-Day Notice

In the 60-Day Notice, the EEOC proposed to retain the current September 30th EEO-1 filing deadline. The EEOC explained that, starting in 2017, employers with 100 or more employees would document their employees' W-2 earnings for a 12-month period starting October 1st and ending the next September 30th. The 60-Day Notice reasoned that W-2 earnings are generally recorded in 3-month periods (calendar year quarters) and that, because the third quarter ends on September 30th, employers could calculate the 12-month W-2 wages without significant difficulty.^[45] The 60-Day Notice also retained the current “workforce snapshot” approach of allowing each employer to choose a pay period between July 1st and September 30th during which it would count its employees to be reported on the EEO-1.^[46] The employees counted during this pay period would be the ones reported on the EEO-1.

B. Public Comments

Employers and other groups objected vigorously to the burden of reporting non-calendar year W-2 data (i.e., October 1st to September 30th). These parties argued that the EEOC, by choosing to impose this unique 12-month reporting period, would significantly increase their costs by compelling them to recalculate W-2 earnings for the sole purpose of completing the EEO-1.

On a related point, employers reliant on human resource information systems (HRIS) ^[47] and payroll software said that they would have insufficient time to budget, develop, and implement new reporting systems if the 2017 EEO-1 report were to be due on September 30, 2017. Employers lacking HRIS and payroll software said they would have a variety of implementation challenges, depending on how they organized their records.

Many commenters suggested changing the 12-month EEO-1 reporting period to be the same as the W-2 reporting period (a calendar year) and moving the EEO-1 filing deadline into the subsequent year, preferably after W-2s are due. A few stakeholders suggested that the EEOC conduct the pay data collection every two years.

C. 30-Day Notice

1. Deadline for Filing the EEO-1

For the upcoming 2016 EEO-1 report, the filing deadline will remain September 30, 2016. However, beginning with the 2017 report, the reporting deadline for all EEO-1 filers will be March 31st of the year following the EEO-1 report year. Thus, the 2017 EEO-1 report will be due on March 31, 2018. Changing the filing deadline will give employers subject to Component 2 six more months to prepare their recordkeeping systems for the 2017 report, and it will give them 1.5 years without filing an EEO-1 report (September 30, 2016 to March 31, 2018). At the same time, this change will align the EEO-1 with federal obligations to calculate and report W-2 earnings as of December 31st; the EEOC will not require a special W-2 calculation for the EEO-1.^[48] These changes will reduce the burden on employers of gathering Component 2 data.

The Commission declines to adopt an alternate-year schedule for filing the EEO-1 report. If collected only in alternate years, the utility of EEO-1 data would be diminished because it would become stale before the new data became available.

2. “Workforce Snapshot” Period

The “workforce snapshot” period refers to the pay period when employers count the total number of employees for that year's EEO-1 report. The EEO-1 has always used this “workforce snapshot” approach, which gives employers a choice but freezes EEO-1 employment numbers as of the chosen pay period. Some employers criticized the “workforce snapshot” approach because it would not reflect same-year promotions that have the effect of moving the employee into a different EEO-1 job category or pay band after the “snapshot” was taken. The Commission addresses this concern in part by moving the “workforce snapshot” period to the fourth quarter, October 1st to December 31st, so that there are fewer opportunities for unreported changes after the “snapshot.” This will preserve employer choice as to the “workforce snapshot,” while at the same time accommodating the established federal schedule for preparing W-2's. In sum, while employers will count their employees during a pay period between October 1st and December 31st, they will report W-2 income and hours-worked data for these employees for the entire year ending December 31st.^[49]

This change will not affect the 2016 EEO-1, for which the July 1st to September 30th “workforce snapshot” period remains effective.

VI. What Pay Data To Report: Measure of Pay for the EEO-1

This 30-Day Notice proposes that employers use Box 1 of Form W-2 (hereafter “W-2 income”) as the measure of pay for Component 2 of the EEO-1.^[50] By definition, W-2, Box 1 includes income that is received between January 1st and December 31st of the relevant calendar year. In reaching this decision, the Commission considered government studies that analyze compensation in U.S. workplaces, relevant academic literature on compensation practices, the public comments and public testimony, and the analyses reflected in the EEOC's NAS study ^[51] and its own Pilot Study.^[52]

A. 60-Day Notice: Options for Measuring Pay

The EEOC's 60-Day Notice described five different measures of individual compensation that are used by the federal government.^[53] After narrowing its consideration to two of these—the Bureau of Labor Statistics' Occupational Employment Statistics (OES) measure of pay ^[54] and the Internal Revenue Service's W-2 definition ^[55] —the EEOC proposed to use W-2 income because it is already calculated by employers, therefore limiting burden, and because it is a comprehensive measure of pay that would be more likely to capture the effect of employment discrimination on different kinds of compensation.^[56] In the 60-Day Notice, the EEOC did not specify which box on the W-2 it would use, but the Commission now specifies that employers will report on income provided in Box 1 of the W-2 form.

B. Public Comments

1. Supporting the Use of W-2 Income

Comments in support of using W-2 income emphasized that it is a comprehensive measure of pay that encompasses overtime, shift differentials, and production and non-production bonuses, which are increasingly important elements of pay. These parties stated that employment discrimination can be manifested when employers decide which employees get opportunities to earn shift differentials or overtime pay, or get large bonuses or awards. Using a measure of pay that excludes so much pay that could be influenced by discrimination would radically reduce the utility of this data collection for the EEOC and OFCCP.

2. Opposing the Use of W-2 Income

Comments in opposition to using W-2 income fell into three categories.

Objection 1: W-2 Income Reflects Employee Choice and Is Not a Reliable Measure of Employer Discrimination

The most widely articulated objection to using W-2 income was that it was not indicative of discrimination because it may reflect employee choice more than employer discretion and that the EEOC cannot differentiate the two in an aggregate pay data collection. Commenters making this argument identified elective participation in overtime, working shifts that provide pay differentials, and working faster or better than another employee (e.g., payments for piecework, commissions, or production), as governed by employee choice. Some of these comments argued that using W-2 income will in fact cause the EEOC to find “false-positives” indicating discrimination because the agency will assume that pay disparities are caused by discrimination rather than employee choice.

Some of these parties urged the EEOC to use “base pay” rather than W-2 income because “base pay” is controlled entirely by employers and therefore is better suited to documenting potential discrimination. Another advantage to using “base pay,” they maintained, is that it would be significantly less expensive and easier for them to report on the EEO-1 because their HRIS now include records of base pay but not W-2 income. These stakeholders did not define “base pay,” apart from noting that it does not include supplemental pay such as overtime, shift differentials, and bonuses, and that it can be stated as an hourly rate or as an annual salary.

Objection 2: Collection of W-2 Data Burdens Employers by Requiring the Integration of HRIS and Payroll Systems

Employers argued that reporting W-2 income would impose an inordinate burden and expense because they store W-2 income data in computerized payroll systems that are entirely separate from the HRIS where they maintain EEO-1 demographic data. They asserted that procuring or developing new software to bridge these two systems would be time-consuming and extremely costly.

Objection 3: Collection of W-2 Income Data for October 1st to September 30th Is Burdensome

Finally, employers argued that reporting W-2 income for October 1st to September 30th of every year would be burdensome because employers' payroll systems collect and report W-2 income on a calendar-year basis for tax purposes. By proposing to change the filing date for the revised EEO-1 from September 30th to March 31st, the EEOC has addressed this objection.

C. 30-Day Notice: W-2 (Box 1) Income Is the Measure of Pay

1. W-2 Income and Employee Choice

The Commission is not persuaded by the argument that W-2 income is an unsuitable measure for a pay data collection by an agency that enforces anti-discrimination laws because it may reflect employee choice as well as employer policy or decisions. As the White House Council of Economic Advisers notes, “In many situations, the delineations between discrimination and preferences are ambiguous.” ^[57] For example, higher commission income may, as some public comments noted, reflect an employee's higher performance, but it may also reflect an employer's discriminatory assignment of more lucrative sales opportunities to employees based on race, ethnicity, and/or sex. As another example, a statistically significant difference in overtime pay between men and women in the same job may result from an employer's gender-biased assumptions that lead to more overtime opportunities being offered to men than to women, whom they may assume have competing family responsibilities. Pay discrimination is complex, and it would be an oversimplification to conclude that only those measures of pay that are shown to be exclusively dependent on an employer's decision or policy can be relevant to assessing allegations of pay discrimination.

2. Supplemental Income Is Important and May Be Linked to Discrimination

Based on its consideration of public comments and government and private sector research, the Commission concludes that supplemental pay is a critical component of compensation and it can be influenced by discrimination, so any measure of income for purposes of enforcing the pay discrimination laws should include supplemental pay. W-2 income incorporates different kinds of supplemental pay that would not be available for analysis if the EEOC were to collect only “base pay” or another basic measure of pay that ignored major sources of compensation.^[58] For employers, W-2 income is a well-defined, familiar, and universally-available measure of pay; for the EEOC and OFCCP, it is useful data for exploring potential pay discrimination.

Supplemental pay is becoming more and more prevalent in the United States. As noted by the Bureau of Labor Statistics, Department of Labor (BLS), “For many occupations in the U.S. labor market supplemental pay—including overtime, bonuses, and shift differentials—is an important component of overall cash compensation. Overtime pay is especially important in production occupations and other blue-collar jobs; bonus pay is mostly a feature of high-wage managerial and sales occupations; and shift differentials play a prominent role in . . . healthcare [] and technical occupations.” ^[59] This pattern also is apparent in some of America's highest paying professions. In the legal profession, for example, bonuses at law firms can account for a significant portion of an associate's total compensation, beyond base salary.^[60]

The human resources consulting firm Aon Hewitt's 2014 U.S. Salary Increase Survey of 1,064 organizations found that variable pay (such as performance-based bonuses) for exempt employees comprised 12.7% of payroll that year.^[61] This represented the highest ratio companies have paid out of their budgets toward bonuses since the consulting firm started keeping records 35 years ago and is an increase from 2008 when 10.8% of their total compensation budgets were devoted to variable pay for exempt employees.^[62] Ken Abosch, leader of Aon Hewitt's compensation practice, stated that companies prefer to give performance-based pay because this practice “keeps employees focused on good performance rather than just showing up, and it allows companies to reward and retain their really valuable employees.” ^[63] In addition, Abosch noted that performance-based pay allows companies to keep their base salaries lower and that companies will only allocate bonuses “if [the company] has good or great results.” ^[64]

In some industries, shift differentials ^[65] and overtime pay ^[66] are important aspects of income. Eighty-three percent of manufacturing and production companies, 59% of customer service and support entities, and 51% of transportation and distribution companies surveyed in 2010 offered shift differentials.^[67] Hospitals and health care service organizations also pay shift differentials for holiday and weekend shifts more than other industries.^[68] Overtime is particularly important in production, transportation, and material moving industries, with workers earning 2% of their income in overtime pay in December 2015.^[69] Employers can control who gets the opportunity for assignments to lucrative shifts that pay premium wages or overtime pay, and withholding such assignments because of a protected basis such as race, ethnicity, or sex would violate Title VII.

Incentive pay for top executives also may be subject to discrimination. For example, at the five highest executive level positions (chief executive officer, vice chair, president, chief financial officer, and chief operating officer), research based on data from 1992-2005 shows that women received a lower share of incentive pay (including bonuses and stock option grants) than their male counterparts, accounting for 93% of the gender pay gap at that level.^[70] This difference remained even after taking into account differences of age, tenure, and titles.^[71]

3. Bridging HRIS and Payroll

In light of employers' argument that bridging employers' HRIS and payroll software for the new EEO-1 will be so burdensome that it outweighs the utility of W-2 income, the EEOC examined three of the HRIS tools that it sees most often in systemic investigations: ADP Enterprise, PeopleSoft, and UltiPro. All three HRIS allow for the collection of EEO-1 demographic data, and all three offer the capacity to record year-to-date gross and paid earnings.^[72] The EEOC recognizes that many employers may not choose to use this capacity, but its existence suggests that creating software solutions for the EEO-1, Components 1 and 2, may not be as complex or novel as some comments suggested.

The EEOC intends to support employers and HRIS vendors as appropriate to accommodate Component 2 of the proposed EEO-1. For example, the EEO-1 Joint Reporting Committee plans to post online its new Data File Specifications for Components 1 and 2 of the modified EEO-1 as soon as OMB approves the information collection. The EEO-1 data file specifications will be for data uploads (submitting EEO-1 data in one digital file), but they also will describe the formatting of data for direct data entry onto the firm's secure EEO-1 account with the Joint Reporting Committee. For reference, the current EEO-1 data file specifications can be found at https://www.eeoc.gov/​employers/​eeo1survey/​ee1_​datafile_​2013.cfm.

VII. What Data To Report: Hours Worked

A. 60-Day Notice

The Commission proposed collecting the number of “hours worked” for non-exempt employees by job category, subdivided into pay band cells, to account for periods when employees were not employed or were engaged in part-time work. With regard to exempt employees, the EEOC suggested that “[o]ne approach would be for employers to use an estimate of 40 hours per week for full-time salaried workers. The EEOC [was] not proposing to require an employer to begin collecting additional data on actual hours worked for salaried workers, to the extent that the employer does not currently maintain such information.” ^[73]

B. Public Comments

Public comments from many employers objected to collecting hours worked data due to the cost of creating new systems to collate and report data about hours worked with W-2 income, and EEO-1 Component 1 data. Some employers inquired how the EEOC would define “hours worked,” so they would know what to report. These employers focused on two alternatives: (1) The FLSA definition of hours worked; and (2) the Affordable Care Act (ACA) approach.

The question of how to count hours worked for employees exempt from overtime received a lot of attention, especially the EEOC's proposal to count 40 hours per week for full time, exempt workers. Supporters of the revised EEO-1 said it was reasonable to use a proxy of 40 hours per week for full-time exempt employees. Those who objected to using the 40-hours per week proxy observed that it simply would not reflect the reality of the hours worked by many full-time exempt employees, who may work substantially more than 40 hours in any given week and may work less than 40 hours in another week. Some comments argued that, since the 40-hour estimate would be incorrect in many instances, reporting 40 hours per week would require them to submit and certify inaccurate information to the federal government.

C. 30-Day Notice

1. The Importance of Collecting Hours Worked

Collecting hours worked is of central importance because this data will enable the EEOC and OFCCP to account for part-time and partial-year work and to assess potential pay disparities in the context of this information. The importance of “hours worked” data can be illustrated by example. If two men and two women in the same job category are paid comparable wage rates, but the men are employed full-time and the women are employed part-time, it would initially appear on Component 2 of the EEO-1—without any data on their hours worked—that the employer was paying the women significantly less than the men (the women would be counted in a lower pay band). On the other hand, if it was known that the men worked 40 hours per week and the women worked 20 hours per week, then their different hours would provide a potential explanation of what initially appears to be a gender-based pay disparity. Of course, explaining a pay disparity in this way would not rule out the possibility that it was also caused by a discriminatory practice or policy that may be identified through further investigation.

In addition to helping to assess pay disparities, hours-worked data may be useful in its own right. The EEOC receives charges of discrimination alleging that an employer gave the charging party fewer hours than other employees, or denied overtime or premium pay hours based on race, ethnicity, sex, or another statutorily-protected basis. Collecting “hours worked” data on the EEO-1 would be useful in the initial stages of such an investigation, as the EEOC seeks to assess how the employer assigns work hours.

2. Defining “Hours Worked”

The Commission adopts the FLSA definition for “hours worked” because it is familiar to employers, designed in conjunction with pay, and applies to all employers subject to the EEO-1.^[74] By contrast, the ACA approach to “service hours” gives employers a range of choices about how to count hours, which would not provide clarity for the EEO-1.^[75]

Under the FLSA, the term “hours worked” includes “all time an employee must be on duty, or on the employer's premises or at any other prescribed place of work, from the beginning of the first principal activity of the workday to the end of the last principal activity of the workday.” ^[76] Numerous court decisions have also helped shape this definition. The FLSA and its regulations require employers to maintain certain records for nonexempt employees, including hours the employee worked each day and the total hours the employee worked each workweek.^[77] Payroll records are to be preserved for at least three years and records upon which wage computations were made (e.g., time cards) should be maintained for at least two years.^[78]

Federal contractors that file the EEO-1 also are subject to the 2014 Fair Pay and Safe Workplaces Executive Order, which, once implemented by regulation, will require them to supply employees with a document each pay period showing the employee's hours worked, overtime hours, pay, and any additions made to, or deductions made from, pay as recorded for purposes of the FLSA.^[79] Adopting the FLSA definition of “hours worked” for the EEO-1 promotes consistency for contractors subject to both requirements.

3. Reporting Hours Worked for Nonexempt Employees

The Commission will require private employers and contractors to report the “hours worked” as recorded for FLSA purposes for nonexempt employees in Component 2 of the proposed EEO-1. “Hours worked” will be reported for the total number of employees in each pay band by ethnicity, race, and gender, for the entire calendar year. For example, assume an employer reports on the EEO-1 that it employs four African American women as administrative support workers in the sixth pay band. The employer would report their total “hours worked” for the entire year in the appropriate pay band cell under “Hours Worked” (for example, 8,160 hours). If one of the workers resigned after the employer took its “workforce snapshot” but before December 31st, the employer would report only the total number of hours she actually worked that year prior to her resignation, which would account for her partial-year employment (for example, rather than 2,040 hours, it might report 1,900 hours).

4. Reporting Hours Worked for Exempt Employees

Although the Commission seeks to minimize employer burden, the importance of hours-worked data necessitates its collection on the EEO-1. The EEO-1 Instructions will give employers the option to: (1) Report a proxy of 40 hours per week for full-time exempt employees, and 20 hours per week for part-time exempt employees, multiplied by the number of weeks the individuals were employed during the EEO-1 reporting year; or (2) provide actual hours of work by exempt employees during the EEO-1 reporting year if the employer already maintains accurate records of this information.

With this approach, the company official who certifies the firm's EEO-1 report would certify that the reports are “accurate and . . . . prepared in accordance with the instructions.” Since the new EEO-1 instructions will give employers the option to record 40 hours per week for full-time exempt employees and 20 hours per week for part-time exempt employees, or to report actual hours-worked data for exempt employees, employers using the proxies can certify with confidence that they completed their EEO-1 reports accurately and in accordance with the instructions.

VIII. How To Report Data in Component 2: Pay Bands and Job Categories

This 30-Day Notice does not change the proposal to collect W-2 income and hours-worked data in the twelve pay bands used by the Department of Labor's Bureau of Labor Statistics (BLS) Occupational Employment Statistics (OES),^[80] for each of the 10 EEO-1 job categories. Such data will support the EEOC's ability to discern significant pay disparities in the early stages of its investigations and, in conjunction with other information, to make more efficient decisions about how to plan the investigations going forward.

A. 60-Day Notice

The 60-Day Notice proposed that Component 2 of the EEO-1 report would collect W-2 income and hours-worked data within twelve distinct pay bands for each job category. These pay bands were based on the twelve wage intervals used by the BLS for the OES survey, which is a semi-annual survey designed to measure employment and wage estimates ^[81] for over 800 occupations.^[82] These OES pay bands are different from the pay bands used on the EEO-4 report now completed by state and local government employers.

Table 1—EEO-4 Pay Bands

Pay bands	Pay bands label
1	$100-$15,999.
2	$16,000-$19,999.
3	$20,000-$24,999.
4	$25,000-$32,999.
5	$33,000-$42,999.
6	$43,000-$54,999.
7	$55,000-$69,999.
8	$70,000 and over.

Table 2—Proposed EEO-1 Pay Bands

Pay bands	Pay bands label
1	$19,239 and under.
2	$19,240-$24,439.
3	$24,440-$30,679.
4	$30,680-$38,999.
5	$39,000-$49,919.
6	$49,920-$62,919.
7	$62,920-$80,079.
8	$80,080-$101,919.
9	$101,920-$128,959.
10	$128,960-$163,799.
11	$163,800-$207,999.
12	$208,000 and over.

B. Public Comments

Many stakeholders argued that the twelve OES pay bands are overly broad, particularly for the highest pay band ($208,000 and over) and also for the lower or middle income pay bands ($30,000 to $80,000). Opponents of the proposal argued that broad pay bands would not produce reliable data because the employees within each pay band may have different levels of experience or hold different jobs within an organization. Some comments advocated for additional and narrower pay bands to better capture pay disparities.

C. 30-Day Notice

Collecting W-2 income and hours-worked data in the twelve OES pay bands will enable the EEOC to gather pay data about most employees and EEO-1 filers, as the majority of wages in the United States are well below the highest OES pay band ($208,000 and over), even after including some types of supplemental income. According to the U.S. Census Bureau, the estimated median earnings for full-time, year round civilian workers 16 years of age and over were $43,545 in 2014. For management occupations, the median earnings were $71,112.^[83]

In Component 2 of the EEO-1, employers will report the number of employees whose annual W-2 income falls in each of the job category's twelve pay bands. For example, an employer may report that it has twelve employees in pay band 3 for Professionals, and that four are white men, four are Asian men, and four are white women.

The EEOC is not convinced that using twelve pay bands in conjunction with the EEO-1 job categories will undermine the utility of W-2 income and hours-worked data. The EEOC does not intend or expect that this data will identify specific, similarly situated comparators or that it will establish pay discrimination as a legal matter. Therefore, it is not critical that each EEO-1 pay band include only the same or similar occupations. The data will be useful for identifying patterns or correlations that can inform the early stages of the investigative process, as explained in more detail in section IX.

In addition, many EEO-1 firms and establishments do not report widely divergent occupations in each EEO-1 job category. It also is likely that similar firms and establishments in the same geographic area will have similar distributions of occupations within the job groups and pay bands, thus making statistical comparisons between EEO-1 reports a reasonable approach to using this data.

IX. How the EEOC Will Use W-2 and Hours-Worked Data

A. 60-Day Notice

As explained in the 60-Day Notice, Component 2 data would support EEOC data analysis at the early stages of an investigation, using statistical tests to identify significant disparities in reported pay. EEOC enforcement staff who conduct these analyses would use them, in the larger context of other available economic data and information, to evaluate whether and how to investigate the allegations of discrimination in more depth. Moreover, the 60-Day Notice also explained how employers would be able to use the summary pay data that the EEOC intends to publish to generally assess their own pay practices.

B. Public Comments

Employers opposing the proposal expressed concern that the EEOC would make unfounded inferences of discrimination based on its statistical analysis of the EEO-1 Component 2 pay data which, in turn, would result in unwarranted and burdensome EEOC investigations. Some interested parties criticized the particular statistical analyses that the EEOC described in the 60-Day Notice, arguing that these tests would not yield meaningful results when applied to data reported in pay bands and broad EEO-1 job categories. These commenters also raised concerns about the dangers of Type I or Type II errors in analyzing Component 2 data: In statistics, “Type I” errors are referred to as “false positives” and “Type II” errors are “false negatives.” ^[84]

Finally, employers expressed skepticism that the EEOC's reports based on aggregated EEO-1 pay data would be useful for evaluating their own pay practices and promoting voluntary compliance. Several employers explained that they do not use W-2 data to analyze their own compensation practices, but rather rely on more complete compensation data that they have at their disposal.

C. 30-Day Notice

This 30-Day Notice expands on the discussion in the 60-Day Notice and explains in more detail how the data collected with this information collection will support enforcement of, and compliance with, Title VII, the EPA, and E.O. 11246.

1. Early Assessment of Charges of Discrimination

Currently, the EEOC enforcement staff can retrieve a respondent's EEO-1 report using existing EEO-1 analytics software to assess the distribution of different demographics (sex, race, and ethnicity) in an employer's job groups. When W-2 income and hours-worked data is added to the EEO-1 report, the EEOC's EEO-1 analytic software tool will be expanded to allow for the examination of pay disparities based on job category, pay bands, and gender, ethnicity, or race. For example, if a charging party alleges that she was paid less than her male colleagues in a similar job, the EEOC's enforcement staff might use the expanded EEO-1 analytics tool to generate a report comparing the distribution of the pay of women to that of men in the same EEO-1 job category.^[85] They also might use statistical tools to determine generally whether there are significant disparities in reported pay in job groups based on race, gender, or ethnicity.

EEOC enforcement staff could then examine how the employer compares to similar employers in its labor market ^[86] by using a statistical test to compare the distribution of women's pay in the respondent's EEO-1 report to the distribution of women's pay among the respondent's competitors in the same labor market. With the proposed addition of hours-worked data to the EEO-1, statistical tests could be used to determine whether pay disparities remain among relevant groups such as men and women, controlling for hours worked. More specifically, statistical tests could determine whether factors such as race, ethnicity, gender, and hours worked impact the distribution of individuals in pay bands. The EEOC envisions that any statistical test would be accompanied by an indication of the practical significance of pay differences.

After considering the results of several statistical analyses in conjunction with allegations in the charge, and sometimes also assessing how the EEO-1 pay data compares to statistics for comparable workers using Census data, EEOC enforcement staff would decide how to focus the investigation and what information to request from the employer. When EEOC enforcement staff requests information from an employer, the employer has the opportunity to explain its practices, provide additional data, and explain the non-discriminatory reasons for its pay practices and decisions. Only after considering all of this information, and possibly additional information, would the EEOC reach a conclusion about whether discrimination was the likely cause of the pay disparities.

The EEOC has tested whether statistical tests, and the EEO-1 pay data, would be useful tools in the investigation of charges of discrimination and has found them to be effective.^[87] The EEOC used two databases to test the utility of the planned analyses. The first was the EEO-4 database that the EEOC currently uses to collect and analyze pay data from state and local governments. Since the EEO-4 has fewer and different pay bands than the EEOC proposes for the EEO-1 pay data collection, the EEOC also used a synthetic database. The term “synthetic” does not mean that the data was not real. Rather, the EEOC created a large confidential database from HRIS data obtained in actual EEOC investigations that contained certain variables of interest, in particular pay rate history and job titles for all employees, and the statistical tests referenced above were run. Other important variables such as “race,” “gender,” and “EEO-1” job codes were randomly generated for databases that lacked this information. The results supported the EEOC's conclusion that these statistical tests provide insights that are useful in developing a request for information or deciding whether an investigation of a charge should have a more limited scope.^[88]

As noted above, some critics disputed the EEOC's choice of statistical tests, arguing that they would not be useful for data reported in broad pay bands and job categories. The EEOC's Pilot Study reported on a 2007 study finding that, even if collecting income data in bands results in a loss of information, that loss would likely be small and of little concern to many researchers, and would be balanced by reduced cost and burden.^[89] Other researchers have identified the value of banded pay data even to the point of being useful in estimating mean incomes within an accuracy of 1-3 percent.^[90] This research suggests that critics who argue that one cannot detect mean differences that are smaller than the pay bands, or bins, are incorrect.^[91]

In addition, the EEOC is confident that the risk of Type I (false positive) or Type II (false negative) errors will not undermine its statistical analyses of Component 2 data. The chances of incurring Type I errors (false positives) are related to the probability level used in the statistical significance test. The EEOC follows judicially recognized statistical standards for identifying meaningful discrepancies,^[92] and therefore is confident that the probability level it uses is effective at minimizing the risk of Type I (false positive) errors. By contrast, the risk of Type II (false negative) error is inversely related to the sample size: The smaller the sample size, the more likely a Type II error. If a sample size is so small that the EEOC enforcement staff is concerned about Type II errors, it will consider analyzing a differently configured, larger sample. Even if it forgoes such analysis due to an elevated risk of Type II errors, enforcement staff will study the EEO-1 for other relevant information and analyze additional data from other sources. In fact, EEOC enforcement staff expects to analyze data from other sources regardless of the risk of error.

2. EEOC Publications Analyzing Aggregate EEO-1 Data

Using aggregated EEO-1 data, Census data, and potentially other data sources, the EEOC expects to periodically publish reports on pay disparities by race, sex, industry, occupational groupings, and Metropolitan Statistical Area (MSA). Particularly after a few years of data collection, these reports will provide useful comparative data. For smaller employers and others that do not hire consultants to analyze their compensation structures, these reports will be especially informative in light of the business case for equal pay and the need to comply with state equal pay laws.

The EEOC's publication of aggregated pay data, in conjunction with the employer's preparation of the EEO-1 report itself, may be useful tools for employers to engage in voluntary self-assessment of pay practices. For contractors, such self-assessment is encouraged by the OFCCP Rule on Discrimination on the Basis of Sex.^[93] OFCCP states that “[e]ach contractor may continue to choose the assessment method that best fits with its workforce and compensation practices.” ^[94] Although the OFCCP rule does not create new obligations with respect to a covered contractor's self-assessment of its compensation practices, it does provide additional guidance about the kinds of compensation practices the contractors should evaluate to ensure their compliance with E.O. 11246.

3. EEOC Training on the Pay Data Collection

The EEOC will ensure its internal capacity to use the EEO-1 pay data effectively by supplementing existing training for EEOC statisticians, investigators, and attorneys about how EEO-1 pay data and the updated EEO-1 analytics tool can be used to improve the agency's enforcement work. EEOC enforcement staff will receive periodic training on how to use the expanded EEO-1 analytics software tool to examine pay data and identify any disparities. EEOC personnel who conduct intake also would receive periodic training to help them “issue spot” potential pay discrimination and ask appropriate questions to collect relevant anecdotal evidence of possible discrimination and information about employer policies and practices. Further, the agency would provide specialized training to its lead systemic investigators. Finally, as discussed more fully below, the EEOC would continue to ensure that staff is trained with regard to confidentiality obligations with respect to pay data.

The EEOC also would provide enhanced technical assistance and support to employers with seminars or webinars, training, and outreach and education materials. Such materials may include best practice guides and self-assessment tools to promote voluntary compliance and assist employers in identifying and correcting discriminatory pay policies and practices. They may also identify practices that could lead to pay discrimination, such as subjective pay decision-making practices, establishing salary by relying heavily on prior salary, and setting salary based in large part on negotiations.

Finally, the EEOC would conduct outreach to other stakeholders, including employees and their advocates, and academic researchers. Outreach to employees and their advocates would focus on “know your rights” trainings with respect to equal pay for equal work and also include training about how to use the EEOC's planned aggregated pay data reports for research and informational purposes.

X. Confidentiality of EEO-1 Data

This 30-Day Notice expands on the discussion in the 60-Day Notice regarding the privacy and confidentiality protections for Component 2 data. The EEOC has successfully protected the confidentiality of EEO-1 data for over 50 years, since this data was first collected. Recognizing that employers are concerned both about the confidentiality of their business data and the privacy of employees' pay information, the EEOC and OFCCP have committed to vigorously guarding its privacy and confidentiality, as explained below.

A. 60-Day Notice

The 60-Day Notice emphasized that Title VII subjects the EEOC to strict confidentiality requirements, subject to criminal penalties; that OFCCP defers to the EEOC on disclosure of all non-contractor data; and that the OFCCP ensures the confidentiality of contractor data to the maximum extent permissible by law. In the 60-Day Notice, the EEOC explained that EEO-1 Component 2 data would not include any employee personally identifiable information and, since EEO-1 pay and hours-worked data would be anonymous and aggregated, personally identifying information would not be readily apparent.

B. Public Comments

Employers expressed concern that the addition of sensitive pay data to the EEO-1 would make it more valuable to their competitors and that any breach in confidentiality would be significantly more costly than with the current EEO-1. They also expressed concern about the privacy of the data, because an individual's pay could be disclosed if, for example, the employee was one of only a few employees matching a particular race/ethnicity background and gender in a cell on the EEO-1 and the EEO-1 report were disclosed. Some employers expressed concern that federal and state agencies may not be bound by Title VII's confidentiality requirements, and some employers urged the EEOC to prevail on Congress to amend Title VII to expressly extend the statute's confidentiality provisions to other federal and state agencies that might get EEO-1 data.

C. 30-Day Notice

1. Legal Confidentiality

a. EEOC

As recognized by employers and explained in the 60-Day Notice, Title VII forbids the EEOC or any EEOC officer or employee from making public any information, including EEO-1 data, before a Title VII proceeding is instituted that involves that information.^[95] EEOC staff who violate this prohibition are guilty of a criminal misdemeanor and can be imprisoned.

The EEOC directly imposes this Title VII confidentiality requirement on all of its contractors, including contract workers and contractor companies, as a condition of their contracts. With respect to other federal agencies with a legitimate law enforcement purpose, the EEOC gives access to information collected under Title VII only if the agencies agree, by letter or memorandum of understanding, to comply with the confidentiality provisions of Title VII.

Finally, the text of Title VII itself states that the EEOC may only give state and local fair employment practices agencies (FEPAs) information (including EEO-1 data) about employers in their jurisdiction on the condition that they not make it public.^[96]

For the EEOC, its agents and contractors, and the FEPAs, Title VII only permits disclosure of information after suit is filed on the issues that were investigated at the administrative level.

b. OFCCP

Even though OFCCP obtains EEO-1 reports for federal contractors and subcontractors (contractors) through the Joint Reporting Committee with the EEOC, OFCCP obtains this information pursuant to its own legal authority under E.O. 11246 and its implementing regulations.^[97]

OFCCP will notify contractors of any FOIA request for their EEO-1 pay and hours-worked data. If a contractor objects to disclosure, OFCCP will not disclose the data if OFCCP determines that the contractor's objection is valid. FOIA Exemptions 3 and 4 recognize the value of this data and provide, in combination with the Trade Secrets Act, the necessary tools to appropriately protect it from public disclosure. OFCCP will protect the confidentiality of EEO-1 pay and hours-worked data to the maximum extent possible consistent with FOIA.

With respect to companies that are not federal contractors or subcontractors under OFCCP's jurisdiction, the confidentiality provision of Section 709(e) applies. OFCCP will refer all such FOIA requests for EEO-1 data to the EEOC for a response. The EEOC, in turn, is subject to Title VII confidentiality and cannot disclose any of its EEO-1 data to the public, except in an aggregated format that protects the confidentiality of each employer's information. Any FOIA request by a member of the public for such disaggregated EEO-1 data will be denied by the EEOC under Exemption 3 of the FOIA.

2. Data Protection and Security

The EEOC takes extensive measures to protect the confidentiality and integrity of EEO-1 data in its possession. First, all EEOC and FEPA staff ^[98] receive annual training in data protection and security. The EEOC maintains a robust cyber security and privacy program, in compliance with the Federal Information Security Modernization Act of 2014.^[99]

The EEOC also complies with a comprehensive set of security and privacy controls to protect organizational operations and information system assets against a diverse set of threats, including hostile cyber-attacks, natural disasters, structural failures, and human errors. The EEOC's systems are monitored on an ongoing basis to assure compliance with an extensive set of security and privacy requirements derived from legislation, Executive Orders, policies, directives, and standards.^[100] Agency information technology systems are subjected to weekly security scans by the Department of Homeland Security, annual internal audits performed by the EEOC's Office of Inspector General, and expert third-party audits for best practices and compliance with cyber-security standards. Current protections include regular internal and external vulnerability scanning and penetration testing, comprehensive real-time anti-virus scanning and protection on all desktops and servers, Internet and email filtering for malware and spam, strong firewall protections and intrusion detection systems, compliance with security benchmark configuration settings, deep discovery advanced network security analysis and monitoring, secure domain name server configurations, automatic server/firewall monitoring and logging, security awareness training, and comprehensive disaster recovery planning and testing.

The online EEO-1 portal of the Joint Reporting Committee allows firms that currently upload EEO-1 data files to encrypt their data or even create a file transfer site for EEOC to download the data. After collecting and reconciling EEO-1 data through a process that may involve input from the employer or contractor, the Joint Reporting Committee at the EEOC provides the database to OFCCP on an encrypted storage device.

XI. Paperwork Reduction Act Burden Estimates

A. Background

The revised EEO-1 data collection has two components. The first component (Component 1) will collect information identical to that collected by the currently approved EEO-1, through which employers report data on employees' ethnicity, race, and sex by job category. The second component (Component 2) will collect data on employees' W-2 (Box 1) income and hours worked. Because of the complexity of this PRA burden calculation, the EEOC is providing the following background information to explain the rationale behind its methodologies for calculating the annual and one-time burden of filing EEO-1 reports.

The OMB's PRA guidance prescribes the factors for agencies to consider in calculating annual reporting and one-time implementation costs. The prescribed PRA calculation is focused on the time it takes filers to complete the tasks required for the proposed information collection and the hourly rates of the employees who spend that time. For this reason, the following discussion of the costs of transitioning and annually filing Components 1 and 2 of the EEO-1 must be formulated through the PRA analysis of hours spent and hourly rates.

OMB's PRA regulations also require consideration of how to reduce the burden of a data collection through the use of technology and automation.^[101] This consideration is particularly relevant to EEO-1 reporting. In the years since the EEOC first estimated the PRA burden of the EEO-1 based only on the time to fill in the cells on a paper EEO-1 report, there have been major advances in technology both for employers and the Joint Reporting Committee. Many employers now rely on HRIS and automated payroll systems.^[102] The Joint Reporting Committee now utilizes an online EEO-1 portal for the confidential filing of EEO-1 reports, either by digital upload or by data entry onto a password-protected, partially pre-populated digital EEO-1.

Throughout the Joint Reporting Committee's transition to this new system, the EEOC continued to calculate the PRA burden based on its original method of counting all the cells on a paper report and calculating the time needed to enter data into each of them. However, with the 60-Day Notice, the EEOC concluded that both digital recordkeeping and digital filing were sufficiently well-established to transition to a new PRA methodology more suited to the new technology and the time-savings it generated.^[103] The EEOC's new PRA methodology—necessarily expressed in the PRA's terms of hours and hourly labor rates—focuses on the time needed by the employer's staff to complete tasks such as reading the EEO-1 instructions, collecting, verifying, validating, certifying, and submitting the report. Therefore, in the 60-Day Notice, the EEOC considered for the first time the time savings generated by this task-based approach stemming from technology.^[104] This is the reason that the burden of filing the EEO-1 actually declined with the PRA calculations in 60-Day Notice, relative to the paper-based calculation method previously used.

In the 60-Day Notice, the EEOC concluded that most employers would be filing the EEO-1 with a digital file upload by the time they file their EEO-1 reports for 2017 and 2018. Therefore, in the 60-Day Notice, the EEOC reasoned that “each additional report filed [would have] just a marginal additional cost.” ^[105] Accordingly, the burden calculation in the 60-Day Notice was based on the number of firms filing one or more EEO-1 reports, not on the number of reports submitted or the number of separate establishments submitting reports. The EEOC's PRA burden calculations also assumed that all employees working on the EEO-1 would be administrative staff paid an hourly rate of $24.23 per hour.

The EEOC's intent in calculating respondent burden for the 60-Day Notice was to recognize the cost and time savings associated with the accelerating trend toward greater automation. However, employers' public comments indicated that the EEOC's estimates reflected a level of automation that was unlikely to be attained imminently. Some of these comments included estimates about the annual time and costs of completing the EEO-1. While some firms stated that they spent less time each year on the EEO-1 than the EEOC estimated in the 60-Day Notice, many firms reported that they spent more time and used more varied professional staff. These same commenters observed that they used data uploads less frequently than the EEOC had projected.

The EEOC carefully considered employers' input, yet, their comments as a whole reflected widely discrepant estimates of the time needed, jobs involved, and HRIS and software costs associated with digital EEO-1 reporting. Although the EEOC recognizes that the EEO-1 may involve more time than it estimated in the 60-Day Notice, the EEOC also concludes that the amount of time a filer spends each year completing this report varies, because each employer is different in terms of number of establishments, number of employees involved in producing the report, time spent by those employees and their rates of pay, and sophistication of HRIS. Due to the wide range of estimates provided about annual reporting costs, the EEOC also relied on its own experience collecting the EEO-1 reports and working with EEO-1 stakeholders over the years.

In conclusion, the EEOC adjusted its methodology for calculating PRA annual burden in this 30-Day Notice. First, the EEOC took into account the time and pay rates for a range of employees at both the firm- and establishment- levels who are responsible for preparing and filing the EEO-1. The EEOC now accounts for time to be spent annually on EEO-1 reporting by everyone from the executive who certifies it, to the lawyer who reviews it and the human resource professionals who prepare it with the support of information technology professionals and clericals.

Second, the EEOC no longer assumes that all the EEO-1 reports for 2017 and 2018 will be submitted by one data upload filed by the firm on behalf of all the establishments. While still reflecting that the bulk of the tasks performed in completing the EEO-1 report will be completed at the firm level due to the centrality of automation, the EEOC's 30-Day Notice recognizes that there are certain tasks that will be performed at the establishment level for employers who enter their EEO-1 data directly onto the Joint Reporting Committee's secure portal. Therefore, the 30-Day Notice burden calculations are based on the number of hours needed to complete the tasks at the firm level and also at the establishment level for the proportion of EEO-1 filers who do not now use centralized, secure data uploads. To make these calculations, the EEOC distinguished the time spent at the firm and establishment levels on the different types of EEO-1 reports, such as single-establishment Type 1 reports, Type 2 consolidated reports for employers with multiple establishments, and Type 6 or 8 reports for small establishments (under 50 employees).

For those employers who have staff enter EEO-1 data online, which is closest digital equivalent to completing a paper form by hand, the Joint Reporting Committee's password-protected, individualized portal prompts the employer with pre-populated EEO-1 forms that already include identifying information and the prior year totals. Moreover, the Joint Reporting Committee's online portal does not compel these employers to enter “zeros” in the cells for which they do not submit data. No EEO-1 filers enter data in every cell, so basing the annual PRA burden on the total number of cells on the EEO-1 form would be inaccurate.

Therefore, as explained in detail below, the total estimated annual burden hour cost in 2017 and 2018 for those contractors that will complete and submit only Component 1 (contractors with 50-99 employees) will be $1,872,792.41. The total estimated annual burden hour cost in 2017 and 2018 for employers and contractors that will complete both Components 1 and 2 will be $53,546,359.08.

The EEOC estimates that for these filers submitting both Component 1 and 2 data in 2017 and 2018, the addition of pay data will increase the estimated annual burden hour costs by a total of $25,364,064.80 or an average of $416.58 per EEO-1 filer each year, using the 30-Day PRA analysis. This is an average estimate per filer, and actual costs will vary, as explained in this Notice.

B. 60-Day Notice

In the 60-Day Notice, the EEOC estimated burden based on centralized electronic, rather than paper, filing of the EEO-1. Costs were calculated assuming that all tasks were performed at the firm level.

Burden Statement—2016: For reporting year 2016, when all filers will continue to submit only Component 1 demographic data, the EEOC estimated the total annual burden hours required to complete the EEO-1 as 228,296.4 hours, with an associated total annual burden hour cost of $5,531,621.77.

Burden Statement—Component 1 Only: The 60-Day Notice stated that starting in 2017, the estimated number of annual respondents (contractor filers) who will submit Component 1 only would be 6,260.^[106] The 60-Day Notice estimated the burden in 2017 on contractor filers with 50 to 99 employees as follows:

• Annual Burden Calculation: The total annual burden hours required to complete Component 1 of the EEO-1 data collection in 2017 and 2018 was estimated to be 21,284 hours each year, with an associated total annual burden hour cost of $515,711.32. This figure used an average wage rate of $24.23 for employees working on the EEO-1, based on the conclusion that administrative support staff would perform the work in completing an EEO-1 report.

Burden Statement—Components 1 and 2: The 60-Day Notice estimated the number of annual respondents that would submit both Components 1 and 2 starting with the 2017 reporting cycle at 60,886 private industry and contractor filers. Filers required to complete both Components 1 and 2 were estimated to incur 401,847.6 burden hours annually or 6.6 hours per filer.

• Annual Burden Calculation: The estimated total annual burden hours needed for filers to report demographic and W-2 income and hours-worked data via Components 1 and 2 of the revised EEO-1 was estimated at 401,847.6, with an associated total annual burden hour cost of $9,736,767.35. This burden estimate includes reading instructions and collecting, merging, validating, and reporting the data electronically.^[107]

• One-Time Implementation Burden: The estimated one-time implementation burden hour cost for submitting the information required by Component 2 of the revised EEO-1 Report was estimated as $23,000,295.^[108] This calculation was based on the one-time cost for developing queries related to Component 2 in an existing human resources information system, which was estimated to take 8 hours per filer at a wage rate of $47.22 per hour.

The 60-Day Notice also estimated that the addition of W-2 income data to the EEO-1 would result in the EEOC incurring $318,000 in one-time costs and would raise the EEOC's recurring internal staffing cost by $290,478 due to the increased staff time needed to process the additional data.

C. 30-Day Notice

In response to concerns raised in the public comments to the 60-Day Notice, this 30-Day Notice reflects an increased burden estimate by: (1) Reflecting varying labor costs for the different types of staff involved with preparing the EEO-1, (2) adding labor costs for report-level functions, and (3) increasing the total number of burden hours a firm would need to read the EEO-1 instructions and to collect, verify, and enter EEO-1 data on the EEO-1 online portal. This methodology increases the total number of hours spent annually, even though the 30-Day Notice reduced overall burden by no longer requiring employers to make special W-2 income calculations for the EEO-1. This reflects employers' feedback about the annual EEO-1 reporting burden.

1. Annual Burden Hours

The 30-Day Notice revises the annual burden hour estimates to add the estimated time spent on firm-level functions by several different types of employees. These estimates are informed by the comments on the 60-Day Notice, based on the EEOC's experiences in providing technical assistance to employers, and within the range of time suggested by public comments.

To submit a report containing EEO-1 Component 1 data, the EEOC now assumes that, at the firm level, computer specialists would need to spend 4 hours, senior human resource managers, corporate legal counsel, and chief executive officers would each spend 1 hour, and data entry clerks and clerical staff would each spend 0.5 hours, for a total of 8 hours to complete firm-level functions.

Based on information received during the comment period, the addition of Component 2 data would increase the total time spent by each of these employees by a factor of 1.9. Therefore, the EEOC estimates that beginning with the 2017 EEO-1, each firm reporting both Component 1 and Component 2 data would require 7.6 hours by computer specialists, 1.9 hours each by senior human resource managers, corporate legal counsel, and chief executive officers, and 0.95 hours each by data entry clerks and clerical staff, for a total of 15.2 hours per firm for firm-level functions.

In order to analyze annual reporting burden as accurately as possible, the EEOC now also considers the time and effort associated with completing the different types of EEO-1 reports. There are six types of EEO-1 reports, as detailed in the footnote.^[109] All reports except the Type 6 report include the requested EEO-1 workforce data; the Type 6 report includes only the employer's name, address, and the number of employees in each establishment with fewer than 50 employees. An employer having establishments with fewer than 50 employees chooses between filing one Type 6 report or multiple Type 8 reports (a full EEO-1 report for the establishment). If it chooses to file separate Type 8 reports for each establishment with fewer than 50 employees, the Joint Reporting Committee does not require it to complete a consolidated EEO-1 for the entire firm; rather, the Joint Reporting Committee's software generates a Type 2 report for the employer. However, if the employer chooses to submit a Type 6 report, it must also complete a full consolidated report. Accordingly, firms that have establishments with fewer than 50 employees either submit Type 8 reports (one for each establishment) or a Type 6 report (a list covering all establishments) plus a Type 2 report.

Finally, based on the EEOC's experience, most firms complete all the tasks associated with filing EEO-1 Type 1, 2, and 6 reports at the firm level. By contrast, for Type 3, 4 and 8 reports, some of the tasks are performed at the firm level, but others are performed at the establishment level. The EEOC's 30-Day Notice annual burden estimates therefore reflect time spent on establishment-level tasks associated with Type 3, 4, and 8 reports, while time spent on tasks associated with Type 1, 2, and 6 reports (and the firm-level functions associated with Types 3, 4, and 8) are included in the firm-level estimates.^[110]

The EEOC assumes that human resource specialists and data entry clerks will perform all establishment-level functions. For firms filing only Component 1 of the EEO-1, the EEOC estimates that for each establishment report submitted, a human resource specialist and a data entry clerk would each spend 0.5 hours on establishment-level functions, for a total of 1 hour per report. Beginning in 2017, firms filing both Component 1 and Component 2 of the EEO-1 would require 0.95 hours each from the human resource specialist and the data entry clerk on establishment-level functions, for a total of 1.9 hours per report.

In 2014, 1,449 firms submitted their EEO-1 reports via data upload, but they submitted 329,944 Type 3, 4, and 8 reports.^[111] The EEOC estimates that firms using data upload will need to spend less time at the establishment level than firms submitting their reports by data entry. For firms using data upload, the EEOC estimates that data entry clerks will not need to perform any establishment-level tasks.

2. Hourly Wage Rates

Using figures reflecting median pay obtained from the Bureau of Labor Statistics,^[112] the EEOC's 30-Day Notice uses hourly wage rates as follows: Computer specialist $24.75, senior human resource manager $50.21, corporate legal counsel $55.69, chief executive officer $49.37, data entry clerk $13.69, clerical staff $15.41, and human resource specialist $28.06. See Table 3 for an illustration of the jobs, hours, and wage rates described in this Notice. Based on the EEOC's experience, the bulk of the work is now performed by computer specialists and senior human resource managers. At the establishment level, the EEOC concluded that EEO-1 reporting work is more likely to be performed by data entry clerks and human resource specialists, resulting in a lower average wage rate for establishment-level functions.

Table 3—EEO-1 Jobs, Hours, and Wages

Job title	Hours spent on EEO-1 Component 1 only	Hours spent on EEO-1 Components 1 & 2	Hourly wage rates
Firm-Level Functions
Computer Specialist	4	7.6	$24.75
Senior Human Resource Manager	1	1.9	50.21
Corporate Legal Counsel	1	1.9	55.69
Chief Executive Officer	1	1.9	49.37
Data Entry Clerk	0.5	0.95	13.69
Clerical Staff	0.5	0.95	15.41
Report-Level Functions
Human Resource Specialist	0.5	0.95	28.06
Data Entry Clerk	0.5	0.95	13.69

XII. Formal Paperwork Reduction Act Statement

A. Overview of Information Collection

The EEOC has submitted to OMB a request for a three-year PRA approval of a revised EEO-1. The revised EEO-1 data collection has two components. The first component (Component 1) will collect information identical to that collected by the currently approved EEO-1. The second component (Component 2) will collect data on employees' W-2 pay and hours worked. Component 1 can be found at http://www.eeoc.gov/​employers/​eeo1survey/​upload/​eeo1-2.pdf. An illustration of the data to be collected by both Components 1 and 2 can be found at http://10.5.0.211/​employers/​eeo1survey/​2016_​new_​survey.cfm.

For the 2016 reporting cycle, there will be no change to the EEO-1 reporting requirement. All EEO-1 filers will continue to submit the data on race, ethnicity, sex, and job category that is currently collected by the EEO-1 report. The EEOC refers to this demographic and job category data as Component 1 data. Beginning with the 2017 reporting cycle, the EEOC proposes to require EEO-1 filers with 100 or more employees to submit data on pay and hours worked (Component 2 data) in addition to Component 1 data. However, federal contractor filers with 50 to 99 employees will only submit Component 1 data.

1. 2016 Overview of Information Collection—Component 1

Collection Title: Employer Information Report (EEO-1).

OMB Control Number: 3046-0007.

Frequency of Report: Annual.

Description of Affected Public: Private industry filers with 100 or more employees and federal government contractor filers with 50 or more employees.

Number of Respondents: 67,146 firms filing 683,275 establishment reports.

Reporting Hours: 1,055,471.

Respondent Burden Hour Cost: $30,055,086.62.

Federal Cost: $1,330,821.

Number of Forms: 1.

Form Number: EEOC Form 100.

2. 2017 and 2018 Overview of Information Collection—Components 1 and 2

Collection Title: Employer Information Report (EEO-1).

OMB Control Number: 3046-0007.

Frequency of Report: Annual.

Number of Forms: 1.

Form Number: EEOC Form 100.

Federal Cost: $318,000 for one-time costs and $1,621,300 ^[113] for recurring staffing costs.

a. Component 1 (Demographic and Job Category Data)

Description of Affected Public: In 2017 and 2018, contractor filers with 50 to 99 employees will submit only the demographic and job category data collected by Component 1.

Number of Respondents: 6,260 firms filing 9,129 establishment reports.

Reporting Hours: 59,166.

Respondent Burden Hour Cost: $1,872,792.41.

b. Components 1 and 2 (Demographic and Job Category Data Plus W-2 and Hours Worked Data)

Description of Affected Public: In 2017 and 2018, EEO-1 filers with 100 or more employees will submit pay and hours worked data under Component 2 in addition to demographic and job category data under Component 1.

Number of Respondents: 60,886 firms filing 674,146 establishment reports.

Reporting Hours: 1,892,979.5.

Respondent Burden Hour Cost: $53,546,359.08.

B. 30-Day Notice PRA Burden Statement

2016: Component 1

Burden Statement: In 2016, all EEO-1 filers will submit Component 1, which only includes the data collected by the currently approved EEO-1. No filer will be required to submit the Component 2 data during the 2016 reporting cycle. The estimated number of respondents required to submit the annual EEO-1 report is 67,146.^[114] This data collection is estimated to impose 1,055,471 burden hours in 2016 or 8 hours per filer for firm-level functions plus an additional one hour per report for establishment-level functions.^[115] The associated burden hour cost for the 2016 reporting cycle is $30,055,086.62.^[116] This estimate assumes electronic filing through the EEO-1 online portal either by data entry or data upload, and accounts for time and cost savings now associated with submission of the EEO-1 via data upload.

2017 and 2018: Components 1 and 2

With respect to the EEO-1 reporting cycles for 2017 and 2018, this Notice will discuss the burden estimates associated with two distinct groups of filers. The first group consists of contractor filers with 50 to 99 employees. This group of filers will continue to submit only the Component 1 data, just as they have done in previous years. The second group of filers includes all EEO-1 filers with 100 or more employees, whether private industry or contractor filers. This larger group will continue to submit Component 1 data as they have always done, but will also submit the newly-added W-2 and hours-worked data of Component 2.

Burden Statement—Component 1 Only: Starting in 2017, the estimated number of annual respondents who are contractor filers with 50 to 99 employees is 6,260.^[117] Again, this calculation assumes 8 hours per filer for firm-level functions plus an additional one hour per individual report for report-level functions. The burden on these contractor filers is estimated as follows:

• Annual Burden Calculation: The estimated total annual burden hours required to complete Component 1 of the EEO-1 data collection in 2017 and 2018 is 59,166,^[118] with an associated total annual burden hour cost of $1,872,792.41.^[119]

Burden Statement—Components 1 and 2: Starting in 2017, the estimated number of annual respondents that will submit Components 1 and 2 is 60,886 private industry and contractor filers. Filers required to complete both Components 1 and 2 are estimated annually to incur a total of 15.2 hours per filer for firm-level functions plus an additional 1.9 hours per individual report for establishment-level functions. The estimated burden is based on electronic filing.

The burden imposed on all private industry employer filers and contractor filers with 100 or more employees as a result of the proposed collection of Component 1 and 2 data is estimated as follows:

• Annual Burden Calculation: The estimated total annual burden hours needed for all filers required to report Components 1 and 2 data is 1,892,979.5 hours,^[120] with an associated total annual burden hour cost of $53,546,359.08.^[121] The EEOC estimates that for these filers submitting both Component 1 and 2 data in 2017 and 2018, the addition of pay data will increase the estimated annual burden hour costs by a total of $25,364,064.80 or an average of $416.58 per EEO-1 filer each year. This burden estimate includes reading instructions and collecting, merging, validating, and reporting the data electronically.

• One-Time Implementation Burden: The 60-Day Notice estimated the one-time implementation burden hour cost associated with submitting the information required by Component 2 of the revised EEO-1 Report to be $23,000,295. This was based on the one-time cost for developing queries related to Component 2 in an existing HRIS, which was estimated to take 8 hours per filer at a wage rate of $47.22 per hour.

Employers filing public comments stated that bridging pay and HRIS systems, or purchasing software updates from vendors, would be extremely expensive. Some of these employers estimated the one-time implementation cost of bridging HRIS and payroll records to report Component 2 data estimated costs could range from $5,000 per firm to $20,000, $30,000, or $40,000 per firm. Although the estimates did not provide details explaining how they were calculated, the EEOC has considered this feedback and increased the one-time implementation burden. It has done so by reflecting that specialized computer software experts with a higher wage rate will be required to do the work necessary to implement the one-time changes required for this proposal.

Using an hourly wage rate for a computer programmer of $55.81, the EEOC now estimates one-time burden hour cost of $27,184,381.28.^[122]

Dated: July 11, 2016.

For the Commission.

Jenny R. Yang,

Chair.

Footnotes

1.  81 FR 5113 (Feb. 1, 2016).

2.  The press release on the hearing is available at EEOC, EEOC Hears Wide Range of Views at Public Hearing on Proposed Changes to EEO-1 Form (Mar. 16, 2016), https://www.eeoc.gov/​eeoc/​newsroom/​release/​3-16-16.cfm. The statements and biographies of the witnesses are available at EEOC, Hearing of March 16, 2016—Public Input into the Proposed Revisions to the EEO-1 Report, http://www.eeoc.gov/​eeoc/​meetings/​3-16-16/​.

3.  The first EEOC-commissioned study, resulting in a 2012 report from the National Research Council, National Academy of Sciences (NAS Report), outlined the potential value for EEOC enforcement of collecting pay data from employers by sex, race, and national origin through a report such as the EEO-1. National Research Council, 2012. Collecting Compensation Data from Employers. Washington, DC: National Academies Press, http://www.nap.edu/​read/​13496/​chapter/​1#ii. The second study, reported by an EEOC contractor in 2015, provided detailed analysis of different approaches to implementing the report and included assessments of different statistical analyses for employer data. Sage Computing, EEOC Pay Pilot Study (September, 2015), http://www.eeoc.gov/​employers/​eeo1survey/​pay-pilot-study.pdf.

4.  42 U.S.C. 2000e, et seq.

5.  Id.

6.  29 U.S.C. 206(d).

7.  42 U.S.C. 2000e-8(c).

8.  E.O. 11246, as amended, 30 FR 12319, 41 CFR 60-1.7(a). Executive Order 13665 amends E.O. 11246 to promote pay transparency for federal contractors, protect employees and job applicants, and make it possible for employees and job applicants to share information about their pay without fear of discrimination. E.O. 13665, 79 FR 20749, available at: https://www.gpo.gov/​fdsys/​pkg/​DCPD-201400250/​pdf/​DCPD-201400250.pdf. OFCCP's recently adopted final rule on sex discrimination (OFCCP Rule on Discrimination on the Basis of Sex) addresses a number of sex-based barriers to equal employment and fair pay. The rule requires contractors to provide equal opportunities “without regard to sex.” 41 CFR part 60-20. See also 81 FR 39108, 39125-39129 (June 15, 2016).

9.  41 CFR 60-1.7(a).

10.  According to the OMB, “collection of information” may include: (1) Requests for information to be sent to the government, such as forms (e.g., the IRS 1040), written reports (e.g., grantee performance reports), and surveys (e.g., the Census); (2) recordkeeping requirements (e.g., OSHA requirements that employers maintain records of workplace accidents); and third-party or public disclosures (e.g., nutrition labeling requirements for food).

Office of Information and Regulatory Affairs, OMB, Memorandum for the Heads of Executive Departments and Agencies and Independent Regulatory Agencies, Information Collection under the Paperwork Reduction Act (Apr. 7, 2010), https://www.whitehouse.gov/​sites/​default/​files/​omb/​assets/​inforeg/​PRAPrimer_​04072010.pdf;​; See also 5 CFR 1320.3(c).

11.  Id.

12.  81 FR 5113 (Feb. 1, 2016).

13.  44 U.S.C. 3507(a)(1).

14.  EEOC, EEOC Implements Finals Revisions to EEO-1 Report (Jan. 27, 2006), https://www.eeoc.gov/​eeoc/​newsroom/​release/​archive/​1-27-06.html;​; See also 70 FR 71294 (Nov. 28, 2005); OMB approved these changes on January 25, 2006, Office of Information and Regulatory Affairs, http://www.reginfo.gov/​public/​do/​PRAViewICR?​ref_​nbr=​200511-3046-001#.

15.  Id.

16.  Carmen DeNavas-Walt and Bernadette Proctor, U.S. Census Bureau, Income and Poverty in the United States: 2014, Current Population, 6 (2015), Table 1: Income and Earnings Summary Measures by Selected Characteristics: 2013 and 2014, https://www.census.gov/​content/​dam/​Census/​library/​publications/​2015/​demo/​p60-252.pdf.

17.  Joan Farrelly-Harrigan, U.S. Dep't. of Labor, Women's Bureau, Black Women in the Labor Force (Feb. 2016), https://www.dol.gov/​wb/​media/​Black_​Women_​in_​the_​Labor_​Force.pdf (reporting that African American women's median annual earnings in 2014 was $33,533, $41,822 for white, non-Hispanic women, and $55,470 for white, non-Hispanic men).

18.  Id.

19.  Michelle Vaca, U.S. Dep't. of Labor Blog, Celebrating Hispanic Women in the Labor Force (Oct. 6, 2015), http://blog.dol.gov/​2015/​10/​06/​celebrating-hispanic-women-in-the-labor-force/​ (reporting that the 2013 median annual earnings for Latinas was $30,209).

20.  Joint Economic Committee, United States Congress, Gender Pay Inequality, 3 (April 2016) http://www.jec.senate.gov/​public/​_cache/​files/​0779dc2f-4a4e-4386-b847-9ae919735acc/​gender-pay-inequality----us-congress-joint-economic-committee.pdf.

21.  American Association of University Women, The Simple Truth About the Gender Pay Gap, 10 (Spring 2016), http://www.aauw.org/​files/​2016/​02/​SimpleTruth_​Spring2016.pdf (reporting that the median annual earnings for Native Hawaiian and Pacific Islander women was $32,893 and $31,191 for Native American women).

22.  Id. (reporting that Asian American women's median earnings in 2014 was $47,776).

23.  U.S. Dept. of Labor, Bureau of Labor Statistics, Women in the labor force; a databook, BLS Reports, 60-61 (Dec. 2015), Table 16: Median usual weekly earnings of full-time wage and salary workers, in current dollars, by race, Hispanic, or Latino ethnicity, and gender, 1979-2014 annual averages, http://www.bls.gov/​opub/​reports/​womens-databook/​archive/​women-in-the-labor-force-a-databook-2015.pdf.

24.  Id.

25.  Francine Blau and Lawrence Kahn, The Gender Wage Gap: Extent, Trends, and Explanations, Institute for the Study of Labor, 73 (Jan. 2016), Table 4: Decomposition of Gender Wage Gap, 1980 and 2010 (PSID), http://ftp.iza.org/​dp9656.pdf (the authors reported that the gender wage gap for purposes of the study was approximately 79 cents on the dollar in 2010).

26.  DeNavas-Walt and Proctor, supra note 16 at 5; see also PayScale, Inside the Gender Pay Gap, (2016), http://www.payscale.com/​data-packages/​gender-pay-gap (reporting that across the United States women are more likely to be overrepresented in lower paying jobs (jobs that pay less than $60,000 per year) and underrepresented in higher paying jobs compared to men. In addition, female pay levels off at $49,000 between the ages of 35-40 whereas men's pay levels off at $75,000 for the ages of 50-55).

27.  Blau & Kahn, supra note 25 at 73, Table 4.

28.  Id. A smaller portion of the gap (approximately 5%) can be attributed to geographic region (0.3%) and race (4.3%). The authors do not provide an explanation about why only 4% of the pay gap is attributed to race despite federal data suggesting that the wage gap between and within minorities is much larger. However, women's gains in education helped to narrow the gender wage gap by almost 6% as women now exceed men in educational attainment.

29.  Asaf Levanon, Paula England, Paul Allison, Occupational Feminization and Pay: Assessing Casual Dynamics Using 1950-2000 U.S. Census Data, Social Forces 88(2) (Dec. 2009), http://statisticalhorizons.com/​wp-content/​uploads/​2012/​01/​88.2.levanon.pdf.

30.  Claire Cain Miller, As Women Take Over a Male Dominated Field, the Pay Drops, NY Times (Mar. 18, 2016), http://www.nytimes.com/​2016/​03/​20/​upshot/​as-women-take-over-a-male-dominated-field-the-pay-drops.html?​_​r=​0 (reporting that when more women became designers, for example, wages fell by 34 percentage points. When male computer programmers outnumbered women computer programmers, the job began to pay more and earned more prestige).

31.  Nancy Lockwood, The Glass Ceiling: Domestic and International Perspectives, 3 Society for Human Resource Management Quarterly 2004, https://www.shrm.org/​Research/​Articles/​Articles/​Documents/​040329Quaterly.pdf (reporting that signs of the glass ceiling in the workplace can be based on gender-based barriers that may be invisible, covert, and overt).

32.  Lean In & McKinsey & Company, Women in the Workplace 2015, 13 (2015), http://womenintheworkplace.com/​ui/​pdfs/​Women_​in_​the_​Workplace_​2015.pdf?​v=​5.

33.  Hannah Riley Bowles & Linda Babcock, How Can Women Escape the Compensation Negotiation Dilemma? Relational Accounts Are One Answer, Psychology of Women Quarterly, 37.1, 81 (2013), http://pwq.sagepub.com/​content/​37/​1/​80.full.pdf+​html (finding that “[n]egotiating for higher compensation is socially costly for women because it violates prescriptive gender stereotypes derived from the gendered division of labor . . ., and its resulting social hierarchy of men in charge and women in caregiving and support roles”).

34.  Moreal Hernandez and Derek R. Avery, Getting the Short End of the Stick: Racial Bias in Salary Negotiations, MIT Sloan Management Review (June 15, 2016), http://sloanreview.mit.edu/​article/​getting-the-short-end-of-the-stick-racial-bias-in-salary-negotiations/​ (MIT conducted three studies focused on racial bias in salary negotiations. In the first study, evaluators reviewed resumes from white and black job applicants. The evaluators were asked to evaluate each job applicant and rate the likelihood that the job applicant would negotiate their salary if offered the job. After controlling for each job applicant's objective qualifications, the evaluators identified the black job applicants as less likely to negotiate compared to the white job applicants. The second study tested whether the evaluators had a racially-biased mindset, which was defined as a person who believes one or a few races were superior to others. The study found that the evaluators had different role expectations of the black applicants compared to the white applicants and they also identified the black job applicants as less likely to negotiate. For the third study, the evaluators and job applicants were required to simulate a job negotiation. Although the black job applicants reported that they negotiated comparably (in terms of the number of offers and counteroffers made) to their white counterparts, their evaluators reported that the black job applicants had negotiated more than the white job applicants. The MIT professors concluded that because the evaluators expected the black job applicants to negotiate less, they had an exaggerated view of their behavior during the job negotiation. In addition, the professors found that the black job applicants received lower starting salaries based on the evaluators perception that the black job applicants were more aggressive).

35.  Roland Fryer, Devah Pager, and Jörg L. Spenkuch, Racial Disparities in Job Findings and Offered Wages, Journal of Law and Economics, University of Chicago Press, vol. 56(3), 22-23, (Sept. 2011), http://scholar.harvard.edu/​files/​fryer/​files/​racial_​disparities_​in_​job_​finding_​and_​offered_​wages.pdf.

36.  Id.

37.  The White House, White House Equal Pay Pledge, https://www.whitehouse.gov/​webform/​white-house-equal-pay-pledge. See also, Natalia Merluzzi, These Businesses are Taking the Equal Pay Pledge, White House Blog (June 14, 2016), https://www.whitehouse.gov/​blog/​2016/​06/​14/​businesses-taking-equal-pay-pledge.

38.  Id.

39.  McKinsey & Company, The Power of Parity: Advancing Women's Equality in the United States, (April 2016) http://www.mckinsey.com/​global-themes/​employment-and-growth/​the-power-of-parity-advancing-womens-equality-in-the-united-states.

40.  Morgan Stanley, Gender Diversity is a Competitive Advantage (May 12, 2016), http://www.morganstanley.com/​blog/​women/​gender-diversity-work;​; See also Morgan Stanley, Why it Pays to Invest in Gender Diversity (May 11, 2016), http://www.morganstanley.com/​ideas/​gender-diversity-investment-framework.html.

41.  Id.

42.  States also are addressing gender pay inequities, including proposing to establish pay transparency, prohibit retaliation against workers who discuss their wages, and request state agencies to examine their pay practices and develop best practices. For a summary of state equal pay laws, see National Conference of State Legislatures, State Equal Pay Laws—July 2015, http://www.ncsl.org/​research/​labor-and-employment/​equal-pay-laws.aspx. For a summary of state equal pay legislation, see Kate Nielsen, American Association of University Women, 2015 State Equal Pay Legislation by the Numbers (August 20, 2015), http://www.aauw.org/​2015/​08/​20/​equal-pay-by-state/​.

43.  Private employers also must file the EEO-1 if they have fewer than 100 employees but are owned or affiliated with another company or have centralized ownership, control or management so that the group legally constitutes a single enterprise and the entire enterprise employs a total of 100 or more employees. EEOC, EEO-1: Who Must File, https://www.eeoc.gov/​employers/​eeo1survey/​whomustfile.cfm.

44.  Employers and contractors file different types of EEO-1 reports depending on whether they are single-establishment or multi-establishment filers. Single-establishment filers only file one report, the Type 1 report. Multi-establishment filers submit several reports. These are: The Type 2—Consolidated Report, which must include data on all employees of the company; the Type 3—Headquarters Report, which must include the employees working at the main office site of the company and those who work from home and report to the corporate office; the Type 4—Establishment Report, for each physical location with 50 or more employees, which provides full employment data categorized by race, gender and job category. For sites with fewer than 50 employees, filers submit either: Type 6—Establishment List, which provides only the establishment name, complete address and total number of employees; or Type 8—Establishment Report, which is a full report for each establishment employing fewer than 50 employees.

45.  81 FR 5113 (Feb. 1, 2016).

46.  EEOC, EEO-1: When to File, https://www.eeoc.gov/​employers/​eeo1survey/​whentofile.cfm.

47.  These systems are also sometimes called “human resource management systems” or HRMS.

48.  Employers must send the W-2 to the Social Security Administration by the last day of February, although special due dates apply if the employer terminated its business or is filing electronically. Employers must furnish the W-2 to employees by February 1. IRS, Topic 752—Filing Forms W-2 and W-3 (Dec. 30, 2015), https://www.irs.gov/​taxtopics/​tc752.html.

49.  By changing the EEO-1 “workforce snapshot” to the last quarter of each calendar year, EEO-1 contractor filers that also file annual employee reports under the Vietnam Era Veterans' Readjustment Assistance Act of 1974, as amended (VEVRAA), 38 U.S.C. 4212(d), will be in a position to align their VEVRAA data collections with the new EEO-1. Under regulations implementing VEVRAA, certain federal contractors must report annually on form VETS-4212 the number of employees and new hires protected under VEVRAA. 41 CFR 61-300.10(d)(1). Form VETS-4212 collects information for veterans protected by VEVRAA using the EEO-1's 10 job categories. For each reporting year, the federal contractor must report covered employees for the 12-month period preceding a date it selects between July 1st and August 31st that falls at the end of a payroll period. Significantly, the regulations allow contractors to select December 31st as the basis for reporting the number of employees and as the ending date of the twelve-month covered period, if the federal contractor has “previous written approval from the Equal Employment Opportunity Commission to do so for purposes of submitting the Employer Information Report EEO-1, Standard Form 100 (EEO-1 Report).” 41 CFR 61-300.10(d)(2). The implementation notice for the revised EEO-1 will serve as “previous written approval” from the EEOC pursuant to this Department of Labor VEVRAA rule.

50.  The IRS instructions for Form W-2 list the following categories of Box 1 taxable income: “(1) Total wages, bonuses (including signing bonuses), prizes, and awards paid to employees during the year; (2) Total noncash payments, including certain fringe benefits; (3) Total tips reported by the employee to the employer; (4) Certain employee business expense reimbursements; (5) The cost of accident and health insurance premiums for 2%-or-more shareholder-employees paid by an S corporation: (6) Taxable benefits from a section 125 (cafeteria) plan if the employee chooses cash; (7) Employee contributions to an Archer MSA (medical savings account); (8) Employer contributions to an Archer MSA if includible in the income of the employee; (9) Employer contributions for qualified long-term care services to the extent that such coverage is provided through a flexible spending or similar arrangement; (10) Taxable cost of group-term life insurance in excess of $50,000; (11) Unless excludable under Educational assistance programs, payments for non-job-related education expenses or for payments under a nonaccountable plan; (12) The amount includible as wages because you paid your employee's share of social security and Medicare taxes (or railroad retirement taxes, if applicable). If employer also paid the employee's income tax withholding, the employer treats the grossed-up amount of that withholding as supplemental wages and reports those wages in boxes 1, 3, 5, and 7. (Employer uses box 14 if railroad retirement taxes apply.) No exceptions to this treatment apply to household or agricultural wages; (13) Designated Roth contributions made under a section 401(k) plan, a section 403(b) salary reduction agreement, or a governmental section 457(b) plan; (14) Distributions to an employee or former employee from an NQDC plan (including a rabbi trust) or a nongovernmental section 457(b) plan; (15) Amounts includible in income under section 457(f) because the amounts are no longer subject to a substantial risk of forfeiture; (16) Payments to statutory employees who are subject to social security and Medicare taxes but not subject to federal income tax withholding must be shown in box 1 as other compensation; (17) Cost of current insurance protection under a compensatory split-dollar life insurance arrangement; (18) Employee contributions to a health savings account (HSA); (19) Employer contributions to an HSA if includible in the income of the employee; (20) Amounts includible in income under an NQDC plan because of section 409A; (21) Payments made to former employees while they are on active duty in the Armed Forces or other uniformed services; and (22) All other compensation, including certain scholarship and fellowship grants.” IRS, 2016 General Instructions for Forms W-2 and W-3, (Jan. 5, 2016), https://www.irs.gov/​pub/​irs-pdf/​iw2w3.pdf.

51.  NAS Report, supra note 3.

52.  Sage Computing, supra note 3. This EEOC Pilot Study compared the OES definition of compensation to the W-2 and concluded that “[t]he W-2 definition of income . . . offers a more comprehensive picture of earnings data and therefore is more appropriate for identifying discriminatory practices.” In contrast to the OES definition of pay, the W-2 definition includes all the elements of compensation that are captured by the OES definition, but also includes forms of compensation such as overtime wages, shift differentials, fees, commissions, fringe benefits, and bonuses. Box 1 on the W-2 excludes certain elective deferrals or pre-tax deductions such as employer-sponsored retirement plan (401(k) or 403(b)) contributions, flexible spending account contributions for health and dependent care, and medical contributions.

53.  NAS Report, supra notes 3 and 51 at 32-34, 41-45, http://www.nap.edu/​read/​13496/​chapter/​4#32.

54.  The Occupation Employment Statistics (OES) survey defines earnings to include base rate pay, cost-of-living allowances, guaranteed pay, hazardous-duty pay, incentive pay such as commissions and production bonuses, tips, and on-call pay. The OES measure excludes back pay, jury duty pay, overtime pay, severance pay, shift differentials, nonproduction bonuses, employer costs for supplementary benefits, and tuition reimbursements. U.S. Dept. of Labor, Bureau of Labor Statistics, Occupation Employment Statistics, http://www.bls.gov/​oes/​current/​oes_​tec.htm. OES survey uses twelve wage intervals. U.S. Dept. of Labor, Bureau of Labor Statistics, Survey Methods and Reliability Statement for the 2015 Occupational Employment Statistics Survey, 4, http://www.bls.gov/​oes/​current/​methods_​statement.pdf,,

55.  81 FR 5113, 5116 (Feb. 1, 2016). The EEOC initially considered five measures of pay. Three of those measures are used by the U.S. Bureau of Labor and Statistics (BLS) when it reports national employment data: the Occupation Employment Statistics (OES); the National Compensation Survey (NCS); and the Current Employment Statistics (CES) survey programs. One measure was from the Social Security Administration (SSA) and the final measure was from the Internal Revenue Service (IRS) (W-2).

56.  Sage Computing, supra notes 3 and 52.

57.  Council of Economic Advisers Issue Brief, The Gender Pay Gap on the Anniversary of the Lilly Ledbetter Fair Pay Act (Jan. 2016), https://www.whitehouse.gov/​sites/​default/​files/​page/​files/​20160128_​cea_​gender_​pay_​gap_​issue_​brief.pdf.

58.  For example, although the FLSA requires employers to maintain pay rates, those pay rates do not include important sources of supplemental income that the EEOC has determined is important to collect in order to identify potential sources of pay discrimination.

59.  John L. Bishow, U.S. Dept. of Labor, Bureau of Labor Statistics, A Look at Supplemental Pay: Overtime Pay, Bonuses, and Shift Differentials (March 25, 2009), http://www.bls.gov/​opub/​mlr/​cwc/​a-look-at-supplemental-pay-overtime-pay-bonuses-and-shift-differentials.pdf.

60.  National Association of Law Placement (NALP), 2014 Associate Salary Survey, NALP, 67-77 (September, 2014), Associate Bonuses.

61.  Aon Hewitt, New Aon Hewitt Survey Shows 2014 Variable Pay Spending Spikes to Record-High Level (Aug. 27, 2014), http://aon.mediaroom.com/​New-Aon-Hewitt-Survey-Shows-2014-Variable-Pay-Spending-Spikes-to-Record-High-Level.

62.  Id.

63.  Jenna McGregor, Bonuses are making up a bigger and bigger percentage of companies' payrolls, Washington Post, (Aug. 27, 2014), https://www.washingtonpost.com/​news/​on-leadership/​wp/​2014/​08/​27/​bonuses-are-making-up-a-bigger-and-bigger-percentage-of-companies-payrolls/​.

64.  Id.

65.  Shift differentials are paid to compensate employees for working shifts other than regular weekday hours.

66.  Employees who are nonexempt under the Fair Labor Standards Act are entitled to receive overtime pay for hours worked over 40 in a workweek. 29 CFR 778.10. The overtime rate is not less than time and one-half their regular pay rate. U.S. Dept. of Labor, Wage and Hour Division, Overtime Pay, https://www.dol.gov/​whd/​overtime_​pay.htm. See also U.S. Dept. of Labor, Wage and Hour Division, Final Rule: Overtime, https://www.dol.gov/​whd/​overtime/​final2016/​,, and, U.S. Dept. of Labor, Wage and Hour Division, Fact Sheet: Final Rule to Update the Regulations Defining and Delimiting the Exemption for Executive, Administrative, and Professional Employees (May 2016), https://www.dol.gov/​whd/​overtime/​final2016/​overtime-factsheet.htm.

67.  SHRM, Shift Differentials: Compensation for Working Undesirable Hours (Dec. 3, 2010), https://www.shrm.org/​hrdisciplines/​compensation/​articles/​pages/​shiftdifferentials.aspx.

68.  Id.

69.  U.S. Dept. of Labor, Bureau of Labor Statistics, News Release-Employer Costs for Employee Compensation (June 9, 2016), http://www.bls.gov/​news.release/​pdf/​ecec.pdf.

70.  Stefania Albanesi, Claudia Olivetti, Maria José Prados, Liberty Street Economics: Incentive Pay and Gender Compensation Gaps for Top Executives, Federal Reserve Bank of New York, (Aug. 25, 2015), http://libertystreeteconomics.newyorkfed.org/​2015/​08/​incentive-pay-and-gender-compensation-gaps-for-top-executives.html#.VzovwP5JlR0.

71.  Stefania Albanesi, How performance pay schemes make the gender gap worse, World Economic Forum, (Dec.23, 2015), https://www.weforum.org/​agenda/​2015/​12/​how-performance-pay-schemes-make-the-gender-gap-worse/​.

72.  The ADP HRIS software allows for the collection of year-to-date gross pay and pay earnings. It includes paycheck year-to-date totals and provides fields for year-to-date tax amount, overtime hourly earnings, overtime hours, total overtime earnings, and total overtime hours. Further, it appears to provide fields for year-to-date taxable income, taxable gross income year-to-date, and year-to-date taxable amounts. Ultipro allows collection of weekly pay rate, hourly pay rate, and year-to-date taxable gross income, in addition to other measure of pay, hours, and bonus. Finally, PeopleSoft allows collection of hourly rate, minimum hourly rate, maximum hourly rate, and Last 26 Pay Period gross income.

73.  81 FR 5113, 5117 (Feb. 1, 2016).

74.  Under the Fair Labor Standards Act, employers must keep certain records for employees who are subject to the minimum wage provisions alone, or to both the minimum wage and overtime provisions, including records of hours worked each workday and total hours worked each workweek. 29 CFR 516.2(a)(7). Employers are not required to maintain hours worked records for employees who are exempt from minimum wage or minimum wage and overtime requirements. 29 CFR 516.3. “Hours worked” under the FLSA includes “(a) [a]ll time during which an employee is required to be on duty or on the employer's premises or at a prescribed workplace and (b) all time during which an employee is suffered or permitted to work whether or not he is required to do so.” 29 CFR 778.223. Unlike the ACA definition, it does not include paid days off.

75.  Under the Affordable Care Act (ACA), all employers with 50 or more full-time employees or equivalents are considered applicable large employers (ALEs) subject to ACA's shared responsibility provisions for providing health insurance. For this purpose, a full-time employee is, for a calendar month, an employee employed on average at least 30 hours of service per week, or 130 hours of service per month. The ACA provides employers the flexibility to use different measurements of hours worked, or “service hours,” for different categories of exempt employees, provided the measures are reasonable and consistently applied. 26 CFR 54.4980H-3(b)(3)(i).

76.  U.S. Dept. of Labor, Wage and Hour Division, Handy Reference Guide to the Fair Labor Standards Act (November, 2014), https://www.dol.gov/​whd/​regs/​compliance/​hrg.htm.

77.  Additional FLSA recordkeeping requirements include (1) the employee's sex and occupation, (2) time and day of the week when employer's workweek begins, (3) basis on which employee's wages are paid, (4) employee's regular hourly rate, (5) employee's total daily or weekly straight-time earnings, (6) employee's total overtime earnings for the workweek, (7) employee's total wages each pay period, (8) date of payment to employee and pay period covered by payment, and much more. 29 CFR 516. See also United States Department of Labor, Wage and Hour Division, Fact Sheet #21: Recordkeeping Requirements under the Fair Labor Standards Act (FLSA) (July, 2008), https://www.dol.gov/​whd/​regs/​compliance/​whdfs21.htm.

78.  Id.

79.  E.O. 13673, section 5, 79 FR 45309 (Aug. 5, 2014). The Paycheck Transparency provision of the Executive Order on Fair Pay Safe Workplaces provides: “(a) Agencies shall ensure that, for contracts subject to section 2 of this order, provisions in solicitations and clauses in contracts shall provide that, in each pay period, contractors provide all individuals performing work under the contract for whom they are required to maintain wage records under the Fair Labor Standards Act; 40 U.S.C. chapter 31, subchapter IV (also known as the Davis-Bacon Act); 41 U.S.C. chapter 67 (also known as the Service Contract Act); or equivalent State laws, with a document with information concerning that individual's hours worked, overtime hours, pay, and any additions made to or deductions made from pay. Agencies shall also require that contractors incorporate this same requirement into subcontracts covered by section 2 of this order. The document provided to individuals exempt from the overtime compensation requirements of the Fair Labor Standards Act need not include a record of hours worked if the contractor informs the individuals of their overtime exempt status. These requirements shall be deemed to be fulfilled if the contractor is complying with State or local requirements that the Secretary of Labor has determined are substantially similar to those required by this subsection.”

80.  U.S. Dept. of Labor, Bureau of Labor Statistics, Survey Methods and Reliability Statement for the May 2015 Occupational Employment Statistics Survey, supra note 54 at 3, (stating that “employment refers to the number of workers who can be classified as full-or-part-time employees, including workers on paid vacations or other types of paid leave; exempt officers, executives, and staff members of incorporated firms; employees temporarily assigned to other units; and noncontract employees for whom the reporting unit is their permanent duty station regardless of whether that unit prepares their paychecks.”)

81.  U.S. Dept. of Labor, Bureau of Labor Statistics, Occupational Employment Statistics Frequently Asked Questions, http://www.bls.gov/​oes/​oes_​ques.htm.

82.  Id. The OES survey produces estimates of wages or salary paid to employees in non-farm occupations in the United States, in a particular State, or in a particular industry. The occupational wage estimates can be estimates of mean wages or percentiles, such as the median wage.

83.  U.S. Census Bureau, Table Packages, Full-Time, Year-Round Workers and Median Earnings in the Past 12 Months by Sex and Detailed Occupation: 2014, http://www.census.gov/​people/​io/​publications/​table_​packages.html.

84.  Type I errors represent the possibility of rejecting a null hypothesis when it is correct. For example, a null hypothesis might be that the earnings of African Americans and whites are the same and a Type I error would be rejecting it as false when it is true. Type II errors represent the opposite: The possibility of accepting the null hypothesis (for example, that the earnings of African Americans and whites are the same) as true when in fact it is false. Type I errors in this context could suggest a need for an investigation where it may not be needed; Type II errors in this context could result in victims of pay discrimination not receiving relief for discrimination.

85.  Enforcement staff could choose to compare men and women in one particular EEO-1 job category, for multiple job categories, or even all job categories.

86.  EEO-1 reports are identified by location and by each establishment's 5-dight NAICS industry codes. The U.S. Census Bureau maintains only one NAICS code for each establishment based on its primary business activity. The Census Bureau states: “[i]deally, the primary business activity of an establishment is determined by relative share of production costs and/or capital investment. In practice, other variables, such as revenue, value of shipments, or employment, are used as proxies. The Census Bureau generally uses revenue or value of shipments to determine an establishment's primary business activity.” U.S. Census Bureau, “North American Industry Classification System—Frequently Asked Questions,” https://www.census.gov/​eos/​www/​naics/​faqs/​faqs.html.

87.  Sage Computing, supra notes 3, 52, and 56.

88.  Id.

89.  Id. citing Micklewright, John and Schnepf, Sylke V., How Reliable are Income Data Collected with a Single Question? (Nov., 2007), http://papers.ssrn.com/​sol3/​papers.cfm?​abstract_​id=​1047981.

90.  Paul T. von Hippel, Samuel V. Scarpino and Igor Holas, Robust estimation of inequality from binned incomes, Sociological Methodology (Jun. 6, 2016), http://arxiv.org/​abs/​1402.4061.

91.  Id.

92.  Hazelwood Sch. Dist. v. United States, 433 U.S. 299, 311 n.17 (1977) (explaining that “a fluctuation of more than two or three standard deviations would undercut the hypothesis that decisions were being made randomly with respect to [a protected trait]”); Wright v. Stern, 450 F.Supp.2d 335, 363 (S.D.N.Y. 2006) (denying motion for summary judgment in case alleging discrimination against African-American and Hispanic employees in promotions and compensation, the court noted that, “[t]hough not dispositive, statistics demonstrating a disparity of two standard deviations outside of the norm are generally considered statistically significant.”).

93.  41 CFR 60-20. See also 81 FR 39109 (June 15, 2016).

94.  Id.

95.  42 U.S.C. 2000e-8(e).

96.  42 U.S.C. 2000e-8(d). See also EEOC, EEO-1 Survey System Privacy Impact Assessment, https://www.eeoc.gov/​employers/​eeo1survey/​privacyimpact.cfm.

97.  41 CFR 60-1.7(a)(1).

98.  As noted in text above, all FEPAs sign a contractual agreement with the EEOC that requires them to follow the confidentiality provisions set forth in Title VII.

99.  44 U.S.C. 3551; see also relevant provision 44 U.S.C. 3554 discussing federal agency responsibilities for protecting federal information and information systems.

100.  40 U.S.C. 1401 et seq., Information Technology Management Reform Act, identifying standards and guidelines developed by the National Institute of Standards and Technology (NIST) for federal computing systems. NIST, NIST Special Publication 800-53, Rev 4, Security and Privacy Controls for Federal Information Systems and Organizations (April 2013), http://nvlpubs.nist.gov/​nistpubs/​SpecialPublications/​NIST.SP.800-53r4.pdf (explaining specific security controls required by the Federal Information Security Management Act of 2002 and thereafter the Federal Information Security Modernization Act of 2014).

101.  Agencies must “evaluat[e] . . . whether (and if so, to what extent) the burden on respondents can be reduced by use of automated, electronic, mechanical, or other technological collection techniques or other forms of information technology, e.g., permitting electronic submission of responses.” 5 CFR 1320.8(a)(5).

102.  International Public Management Association for Human Resources, Public Personnel Management, Volume 39, No. 3, Fall 2010, http://ipma-hr.org/​files/​pdf/​ppm/​ppmfall2010.pdf (reporting that 90% of human resources departments used some form of HRIS).

103.  81 FR 5113, 5120 (Feb. 1, 2016).

104.  Id.

105.  Id.

106.  81 FR 5113 (Feb. 1, 2016). Of the 67,146 firms that filed EEO-1 reports in 2014, 6,260 were federal contractor filers with fewer than 100 employees.

107.  81 FR 5113 (Feb. 1, 2016). This estimate was calculated as follows: 6.6 hours per respondent × 60,886 respondents = 401,847.6 hours × $24.23 per hour = $9,736,767.35. See also U.S. Dept. of Labor Bureau of Labor Statistics, Employer Costs for Employee Compensation—December 2013 (March 2014), http://www.bls.gov/​news.release/​archives/​ecec_​03122014.htm (listing total compensation for administrative support as $24.23 per hour).

108.  81 FR 5113 (Feb. 1, 2016). This estimate was calculated as follows: 8 hours per respondent × 60,886 employers = 487,088 × $47.22 per hour = $23,000,295. See also U.S. Dept. of Labor, Bureau of Labor Statistics, Employer Costs for Employee Compensation—December 2013, supra note 108 (listing total compensation for a professional as $47.22 per hour).

109.  Type 1 (single establishment firm); Type 2 (consolidated report for headquarters and multi-establishment firm); Type 3 (headquarters report); Type 4 (report for establishments with over 50 employees); Type 6 (list of establishments with under 50 employees); and Type 8 (detailed report for establishments with under 50 employees).

110.  Because of this, the EEOC's burden estimates for firm-level tasks are inflated for those firms electing to file Type 8 reports, because the firm-level estimates include time spent completing a Type 2 and a Type 6 report, even though firms that opt to complete Type 8 reports do not also submit a Type 2 or Type 6 report.

111.  In 2014, contractor filers with 50-99 employees submitted 86 Type 3, 4, and 8 reports via data upload.

112.  U.S. Dept. of Labor, Bureau of Labor Statistics, Occupational Outlook Handbook, http://www.bls.gov/​ooh/​.

113.  The addition of W-2 pay data to the EEO-1 is expected to increase EEOC's internal staffing costs by approximately $290,478. The annual federal cost figure of $1,621,300 includes both the increase in contract costs resulting from the addition of the pay data collection and the estimated internal staffing costs. It reflects an increase of more than $290,478 compared to the estimated federal costs provided in previously published Federal Register notices seeking PRA approval of this information collection because past estimates reflected the cost of the contract with the vendor whose services the EEOC procures to assist with administration and processing of the EEO-1 but did not include EEOC's internal staffing costs associated with processing the EEO-1.

114.  In 2014, 67,146 firms filed EEO-1 reports.

115.  This estimate calculates total time spent by firms assuming no data upload, then subtracts the estimated time saved by firms using data upload, as follows: 8 hours per firm for firm-level functions × 67,146 firms = 537,168 hours; 1 hour per report for establishment-level functions × 683,275 reports = 683,275 hours; 537,168 + 683,275 = 1,220,443 total hours; 0.5 hours per report of data entry clerk time saved by data upload × 329,944 reports filed by data upload = 164,972; 1,220,443−164,972 = 1,055,471.

116.  To reach this estimate, the EEOC multiplied the hourly wage rates for each job by the estimated hours spent by each job in completing the EEO-1 to arrive at a per-firm cost for firm-level functions of $268.82 and a per-report cost for establishment-level functions of approximately $20.88 (rounded). The total burden hour cost for firm-level functions is $18,050,187.7 and the total burden hour cost for establishment-level functions is $14,263,365.6. Firms using data upload are estimated to save $2,258,466.68 (data entry clerk hourly wage rate of $13.69 × 0.5 hours × 329,944 reports filed by data upload). Total firm-level burden hour cost of $18,050,187.7 + total establishment-level burden hour cost of $14,263,365.6−cost savings from data upload of $2,258,466.68 = a total annual burden hour cost of $30,055,086.62.

117.  Of the 67,146 firms that filed EEO-1 reports in 2014, 6,260 were federal contractor filers with fewer than 100 employees.

118.  This estimate calculates total time spent by firms assuming no data upload, then subtracts the estimated time saved by firms using data upload, as follows: 8 hours per firm for firm-level functions × 6,260 firms = 50,080 hours; 1 hour per report for establishment-level functions × 9,129 reports = 9,129 hours; 50,080 + 9,129 = 59,209 total hours; 0.5 hours per report of data entry clerk time saved by data upload × 86 reports filed by data upload = 43; 59,209−43 = 59,166.

119.  To reach this estimate, the EEOC multiplied the adjusted hourly rates for each job by the estimated hours spent by each job in completing the report to arrive at a per-firm cost for firm-level functions of $268.82 and a per-report cost for establishment-level functions of approximately $20.88 (rounded). The burden hour cost for firm-level functions is $1,682,813.2 and the burden hour cost for establishment-level functions is $190,567.875. Firms using data upload are estimated to save $588.67 (data entry clerk hourly wage rate of $13.69 × 0.5 hours × 86 reports filed by data upload). Total firm-level burden hour cost of $1,682,813.2 + total establishment-level burden hour cost of $190,567.875−cost savings from data upload of $588.67 = a total annual burden hour cost of $1,872,792.41.

120.  This estimate calculates total time spent by firms assuming no data upload, then subtracts the estimated time saved by firms using data upload, as follows: 15.2 hours per firm for firm-level functions × 60,886 firms = 925,467.2 hours; 1.9 hours per report for establishment-level functions × 674,146 reports = 1,280,877.4 hours; 925,467.2 + 1,280,877.4 = 2,206,344.6 total hours; 0.95 hours per report of data entry clerk time saved by data upload × 329,858 reports filed by data upload = 313,365.1; 2,206,344.6−313,365.1 = 1,892,979.5.

121.  To reach this estimate, the EEOC multiplied the adjusted hourly rates for each job by the estimated hours spent by each job in completing the report to arrive at a per-firm cost for firm-level functions of approximately $510.76 and a per-report cost for establishment-level functions of approximately $39.66 (these figures are rounded). The burden hour cost for firm-level functions is $31,098,011.6 and the burden hour cost for establishment-level functions is $26,738,315.7. Firms using data upload are estimated to save $4,289,968.22 (data entry clerk hourly wage rate of $13.69 × 0.95 hours × 329,858 reports filed by data upload). Total firm-level burden hour cost of $31,098,011.6 + total establishment-level burden hour cost of $26,738,315.7−cost savings from data upload of $4,289,968.22 = a total annual burden hour cost of $53,546,359.08.

122.  This estimate is calculated as follows: 8 hours per respondent × 60,886 employers = 487,088 × $55.81 per hour = $27,184,381.28. The higher one-time implementation burden estimate in this Notice as compared to the one-time implementation burden estimate in the 60-Day Notice is due to the higher wage rate for the computer programmer, multiplied by 1.46, which is the employer contribution for “management, professional, related.” U.S. Dept. of Labor, Bureau of Labor Statistics, Occupational Outlook Handbook: Computer Programmers, http://www.bls.gov/​ooh/​computer-and-information-technology/​computer-programmers.htm;​; see also U.S. Dept. of Labor, Bureau of Labor Statistics, Employer Costs for Employee Compensation—Dec. 2015 (Mar. 2016), http://www.bls.gov/​news.release/​archives/​ecec_​03102016.htm (computing the rate of employer contribution by dividing total compensation by total salary).