Mobilizing Equal Employment Rights: The Social and Political Determinants of Discrimination Complaints (2009–2018)

ABSTRACT

This article explores the regional and national determinants of workplace discrimination complaints across the US states from 2009–2018. Drawing on the EEOC charge data supplemented with a number of additional data sources, the authors examine the extent to which socioeconomic, demographic, and political environments explain variation in the rate of total, race, and sex-based employment discrimination charges. Building on the neoinstitutional and power resource theories, the authors examine the role of social-structural factors as important determinants of workplace discrimination charges across US states. In fixed-effects regressions, the authors find evidence that union density and collective bargaining, democratic partisanship in legislatures, and demographic composition at the state level and contentious politics and economic inequalities at the national level are important determinants of workplace discrimination claims.

KEYWORDS:

• Employment discrimination
• EEOC
• power resource theory
• economic sociology
• political sociology

Disclosure Statement

No potential conflict of interest was reported by the authors.

Notes

1. Data on state employment are drawn from the Bureau of Labor Statistics (BLS 2018).

2. Tomaskovic-Devey and McCann’s paper aggregates the charges from 2012–2016 but does not conduct time-series analysis the way we do in this paper.

3. It should be noted that our data does not cover the most recent waves of Black Lives Matter protests given that the time span of this study is from 2009 to 2018.

4. The lowest level by which the EEOC charges data are available are at the state, and a careful state-level analysis of the EEOC data can be particularly informative as to how social-structural factors explain the observed variation in the rate of employment discrimination across US states.

5. Despite a general decline of unions in the United States, there is still considerable variation within and between states, see the scatter plot (Figure 2).

6. Our analysis includes charges filed in Washington, D.C., in addition to the other 50 states.

7. We had also included measures of racial (Black vs White) and gender wage gaps, but due to multicollinearity with other independent variables, we removed these measures from our analysis.

8. As an additional robustness check, we also controlled for “female labor force participation” in our sex-specific model. As Tomaskovic-Devey and McCann (2021) paper shows, there is considerable variation in charge rates. We included the results controlling for labor force participation in the appendix of this paper.

9. We verified this via multiple correspondence with EEOC.

10. In auxiliary analysis, we also explored the between-state effects of the independent variables on the dependent variables. We conducted the random-effects models with an autoregressive AR (1) disturbance and the results are largely consistent with the fixed effects model. We would like to note again that the random-effects models are significantly limited given that the publicly available EEOC data does not include those of the FEPA.

11. We also added two-way fixed effects and the results are largely similar, though the effect for protest was marginally significant (α = 0.053). Due to lack of variation of some key national-level independent in later years (2016–2018), we faced multicollinearity problem for those years. In an effort to preserve the independent variables, we opted to keep the linear measure of year rather than including year fixed effects. The results are largely similar in both cases.

12. To investigate whether changes in the leadership of the EEOC bears on discrimination charges, we also included a dummy variable for the political affiliation of the Chair of the Equal Employment Opportunity Commission. The results, however, remained consistent.

13. We could not label the states in the figure given that we included them all. However, we generated small line plots for each state in Figure 3.

Additional information

Notes on contributors

Masoud Movahed

Masoud Movahed is a doctoral candidate in Sociology at the University of Wisconsin-Madison. His work lies at the nexus of economic sociology and social stratification. Specifically, he examines the institutional determinants of social and economic inequalities (email: [email protected]).

Elizabeth Hirsh

Elizabeth Hirsh is a professor of Sociology at the University of British-Columbia. Her research examines how law and policy can most effectively be used to reduce discrimination and promote equity. Her research topics include work and organizations, gender and race employment discrimination, as well as law and society (email: [email protected]).