Written Testimony of Elizabeth Hirsh Associate Professor Sociology and Canada Research Chair in Inequality and Law University of British Columbia

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

I would like to begin by thanking you for the privilege to speak before the EEOC Commissioners. My testimony reflects my experience as a sociologist and expert in the fields of inequality, employment discrimination, work organizations, and legal interventions. Much of my research on these topics over the last decade has involved the use of the EEO-1 private sector employer reports and EEOC charge and litigation data. I have obtained access to these data sources through an Intergovernmental Personnel Act appointment to the EEOC. As well, I served as an unpaid intern at a district EEOC office, working with intake officers, charge investigators, and lawyers to obtain the firsthand experience necessary for my research on charge processing. Finally, I served on the EEOC-commissioned National Academy of Science (NAS) panel charged with assessing the utility of collecting compensation data.

My remarks today will focus on : (1) how the collection of pay and hours data could be used to enhance evidence-based enforcement efforts on the part of EEOC district and field offices, including charge intake and investigations, (2) how such efforts are well-supported by the academic literature on contemporary employment discrimination, and (3) how the academic research community might make use of EEO-1 pay data to better understand pay disparities and discrimination more broadly. I will conclude with a discussion of how the proposed pay data collection strategy reflects the suggestions put forth by the National Academy of Science panel on collecting compensation data.

An Evidence-Based Model of EEO Enforcement

A large body of social scientific and legal research has demonstrated that contemporary discrimination is more subtle than the overt sex and race biases characteristic of the civil rights era. Bias is often structural in nature, veiled in policies, practices, assessments, and relations that may appear neutral at face-value. For instance, gender discrimination may not involve the outright exclusion of female applicants from managerial positions, but employers' underestimation women's competencies and worth due to assumed caregiving responsibilities.¹ Race discrimination may not involve explicit antipathy toward minority groups, but assessments of how racial minority workers might "fit in" with the current work group or team.²

The implication of these forms of contemporary discrimination for the EEOC's enforcement model is that much discrimination is not traceable to a discreet event or even immediately identifiable to potential victims and their employers.³ Thus enforcement efforts based exclusively on individual level reporting via complaints is at odds with the ways in which social scientists observe discrimination in present-day workplaces.

A more appropriate enforcement model would require the EEOC to make use of additional sources of evidence - in addition to complaints - to focus enforcement efforts and target specific industrial sectors, localities, firms, and/or establishments. EEOC charge investigators already use the EEO-1 data to examine composition asymmetries at establishments that are charged. The proposed plan to collect compensation data would enable intake teams and investigators to identify within-job pay disparities at targeted establishments, firms, and/or industries. This would enhance the investigative toolkit by providing an additional empirical indicator of the equity profile of workplaces and move the Commission further toward an evidence-based enforcement plan.

At EEOC field and district offices, EEOC intake units could use the pay data to investigate how establishments subject to charge(s) may vary in their pay practices as compared to a local or national industry standard. Investigators could set up contingency tables comparing the within-occupation earnings profile of a protected group or groups against dominant group members at the respondent establishment versus the local (or national) industry average. This technique could be made available to intake teams through a simple statistical application via a desktop interface and/or integrated into the information management system. This would help investigators establish a prima facie case for discrimination - or at the very least, suggest that an establishment's pay profile is (or is not) out of synch with industry averages. A similar approach could be taken with Commissioner's charges, where investigators could use pay data and comparisons to identify and build cases of systemic bias.

Of course, such calculations would not constitute sufficient evidence to determine reasonable cause, and there may be perfectly good reasons unrelated to discrimination that could account for earnings differential. Thorough investigations would flush these out. Still, such a data-guided enforcement approach could help intake teams and investigators more effectively implement the Priority Charge Handling System, where a respondent's pay profile relative to the industry standard could be used as an additional piece of evidence in classification and investigation processes. Given large charge volumes and the need to classify charges before receiving a position statement from the respondent, such additional evidence would assist intake teams in making more efficient and accurate categorization decisions.

Pay Data Reporting as Encouraging Self-Monitoring

Although the EEOC receives close to 90,000 charges per year, the Commission investigates allegations of discrimination at just a small fraction of all work establishments covered by Title VII and related antidiscrimination laws.⁴ No matter how effective, targeted enforcement efforts cannot reach all corners of workplace. However, research has shown that charges (and related enforcement mechanisms) can generate a compliance effect even among those establishments that are not charged.⁵

Likewise, the requirement to report pay data to the EEOC could create a self-monitoring effect among all firms required to report compensation data in the proposed revision to the EEO-1 form. In addition to much organizational research showing that accountability mechanisms enhance efforts toward equity, a recent field study on pay decisions shows that in a large private firm, the adoption of accountability and transparency mechanisms in pay decisions led to declines in pay gaps by gender, race, and foreign nationality.⁶ To the extent that requiring firms to record, report, and (presumably) scrutinize pay differences across pay bands by sex, race, and ethnicity in the proposed EEO-1 form makes pay disparities visible and increases accountability for pay decisions, we might expect similar equity-enhancing effects among reporting firms. In other words, the available research evidence suggests that EEO-1 pay data reporting may provide an opportunity for self-monitoring and self-reflection, especially among firms subject to additional EEO pressures, prior charges/litigation, or voluntarily-adopted diversity initiatives.

Research Possibilities Using EEO-1 Pay Data

In recent years, a small group of academic researchers, working closely with EEOC researchers, has gained access to EEO-1 composition data through confidentiality agreements. These collaborations have extended our social scientific knowledge of discrimination and enforcement efforts in at least three important directions. As I discuss below, each of these lines of research could be extended and enriched with the addition of compensation data.

First, careful research using EEO-1 data since the 1960s has shown considerable variation in the efficacy of enforcement efforts across time, political era, and industrial sector.⁷ EEO enforcement has proved most effective in improving the representation of protected groups in the workforce and managerial jobs when it is accompanied by strong political and social pressures. In the current era, inequality is best understood as locally-produced, such that some industries and locales are doing much better than others in the employment of women and racial-minority groups.

To exemplify this, Don Tomaskovic-Devey has prepared reports ranking industries (and some large firms) on their employment of women and racial-minorities based on comparisons to labor market standards.⁸ These reports and rankings are based on composition data. However, similar rankings could be produced based on industry and firm pay profiles. Such research could aid the EEOC in directing enforcement efforts, but also direct researchers to conduct more in-depth studies of the industries (and even firms) that suffer the most from pay disparities. In addition, focusing on the "local environment" of pay gaps might lead researchers to identify how pay gaps vary by establishment and firm size, industrial sector, and state and local policy environments. A focus on local policy environments would be especially interesting given that many states have more generous pay equity policies as compared to federal law, and researchers have found state policies to be consequential for levels of gender equity in management.⁹

A second line of research using EEO-1 compositional data involves matching EEO-1 composition information to survey data on establishments' diversity-oriented human resource (HR) practices.¹⁰ This approach has allowed researchers to assess the efficacy of various popular diversity practices in expanding employment opportunities for protecting groups, most commonly, in management. Researchers have found, for instance, that policies that stress accountability and assign direct responsibility for diversity goals are much more effective than policies aimed at training or educating employees or managers on diversity issues. We know very little, however, on how such practices may affect pay equity in work organizations. The proposed data collection initiative would make such investigations possible, and enable researchers to identify best practices for pay equity as well as general employment or managerial opportunities.

Finally, researchers have investigated if and how discrimination charges filed with the EEOC and litigation - both EEOC and private - as well as OFCCP compliance reviews have led to gains in employment opportunities for women and racial minorities.¹¹ This work has focused on the representation of women and African-Americans in management as well as the distribution of women and racial minorities across the occupational hierarchy. But again, with the collection of pay data, researchers could examine whether discrimination charges and lawsuits - including those involving pay discrimination - reduce gender and racial pay disparities following charges or lawsuits. For examine, Sheryl Skaggs' research on litigation in the supermarket industry could examine how gender lawsuits affected gender pay gaps at the sales, managerial, or executive level. My own research on high-profile litigation (pursued by the EEOC and privately) could examine how various terms and conditions of rulings and settlements affect pay disparities between targeted groups.

Owing to collaboration between the EEOC and academic researchers over the last decade, we have learned much about how political pressures, EEOC and OFCCP enforcement efforts, private litigation, and employers' voluntary adoption of HR policies affect compositional dynamics in work organizations. With compensation data, the academic research community could extend this research to understanding pay disparities and their remediation.

NAS Recommendations and EEO-1 Pay Data Proposal

As a member of the NAS panel commissioned by the EEOC to explore methods of collecting and analyzing pay data, I'd like to conclude by stressing that the pay data initiative under consideration is in keeping with the recommendations of the NAS report. As our NAS panel recommended, the EEOC undertook a thorough, independent pilot study, directed by Kidan Furtulus at UMass-Amherst, to evaluate options for data collection and analysis. The final decision to collect data across pay bands by occupational and sex-race ethnicity is consistent with the results of this pilot study. The EEOC has also been clear in specifying a plan for data sharing and coordination, particularly with the OFCCP, has taken concerns of reporting burden seriously by calculating reporting costs to employers for the various data collection strategies under consideration, and has outlined a clear confidentiality and disclosure plan - all of which were urged by the NAS panel.

Thank you for the opportunity to speak on behalf of the proposal to amend the EEO-1 form to collect compensation and hours data.

Footnotes

¹ See Correll, S.J. and Benard, S., 2007. Getting a job: is there a motherhood penalty? 1. American journal of sociology, 112(5), pp.1297-1339.

² Lauren Rivera's study of minority recruitment in professional jobs provides an example: Rivera, L.A., 2012. Diversity within Reach Recruitment versus Hiring in Elite Firms. The ANNALS of the American Academy of Political and Social Science, 639(1), pp.71-90.

³ I make this point directly in a study of plaintiffs' accounts of their workplace experiences in high-profile discrimination litigation: Elizabeth Hirsh, 2013. Beyond Treatment and Impact: A Context-Oriented Approach to Employment Discrimination. American Behavioral Scientist, p.0002764213503328.

⁴ Based on a national sample of EEO-1 reporting establishments matched to EEOC charge data, I estimate that less than 5% of private establishments are charged per year (Hirsh, C.E. and Kornrich, S., 2008. The Context of Discrimination: Workplace Conditions, Institutional Environments, and Sex and Race Discrimination Charges. American journal of sociology, 113(5), pp.1394-1432.)

⁵ C. Elizabeth Hirsh. 2009. "The strength of weak Enforcement: The impact of discrimination charges on sex and race segregation in the workplace. American Sociological Review 74(2):245-271.

⁶ Emilio Castilla. 2015. "Accounting for the Gap: A Firm Study Manipulating Accountability and Transparency in Pay Decisions," Organization Science 26(2): 311-333.

⁷ A comprehensive account of the history of EEO enforcement can be found in: Kevin Stainback and Donald Tomaskovic-Devey. 2012. Documenting Desegregation: Racial and Gender Segregation in Private Sector Employment since the Civil Rights Act. NY: Russell Sage Foundation.

⁸ Industry Gender Disparities: A Preliminary Report on Evidence of Systemic Bias in the Private Sector. Prepared for the U.S. Equal Employment Opportunity Commission. Donald Tomaskovic-Devey, Karen Brummond, Skylar Davidson, University of Massachusetts, Amherst.

⁹ A recent study shows that state equal pay laws, especially those adopted before federal equal pay legislation, family responsibility discrimination protections, and past ERA ratification, are positively associated with women's lower-level managerial presence. See Kmec, Julie A. and Skaggs, Sheryl L., 2014. The "State" of Equal Employment Opportunity Law and Managerial Gender Diversity. Social Problems, 61(4), pp.530-558.

¹⁰ The now landmark study in the field is Kalev, Alexandra, Dobbin, Frank and Kelly, Erin. 2006. "Best Practices or Best Guesses? Assessing the Efficacy of Corporate Affirmative Action and Diversity Policies." American Sociological Review 71: 589-617.

¹¹ For work on discrimination charges, see C. Elizabeth Hirsh. 2009. "The strength of weak Enforcement: The impact of discrimination charges on sex and race segregation in the workplace. American Sociological Review 74(2):245-271; and for work on litigation in the grocery industry, see Sheryl Skaggs, 2010. "Legal-Political Pressures and African American Access to Managerial Jobs." American Sociological Review. 74:225-244 and 2008a. "Producing Change or Bagging Opportunity? The Effects of Discrimination Litigation on Women in Supermarket Management." American Journal of Sociology 113:1148-82. Elizabeth Hirsh and Youngjoo Cha ("For Law or Markets? Discrimination Lawsuits, Market Performance, and Managerial Diversity," working paper University of British Columbia) show litigation effects in high profile lawsuits. Alexandra Kalev and Frank Dobbin (Kalev, A. and Dobbin, F., 2006. Enforcement of civil rights law in private workplaces: The effects of compliance reviews and lawsuits over time. Law & Social Inquiry, 31(4), pp.855-903.) demonstrate the effects of OFCCP compliance reviews on managerial composition. Robert Nelson and George Bridges (Nelson, R.L. and Bridges, W.P., 1999. Legalizing gender inequality: Courts, markets and unequal pay for women in America (Vol. 16). Cambridge University Press.) provide a comprehensive analysis of key pay equity lawsuits and court decisions.