Harassment, Workplace Culture, and the Power and Limits of Law

Harassment, Workplace Culture, and the Power and Limits of Law

70 Am. U. L. Rev. 419 (2020).

* Herbert and Doris Wechsler Clinical Professor of Law, Columbia Law School. Many thanks for helpful comments and conversation to Lisa Bernstein, Mary Anne Case, Liz Emens, Julie Goldscheid, Todd Henderson, Vicki Jackson, Olati Johnson, Lou Kelley, Suzanne Kim, Gillian Lester, Lance Liebman, Henry Monaghan, Chuck Sabel, Bob Scott, Susan Sturm, Kendall Thomas, and colleagues in faculty workshops at Columbia Law School, The Ohio State University Moritz College of Law, and the Law and Society Association Annual Meeting as well as students and faculty in the University of Chicago Workshop on Regulation of Family, Sex, and Gender. Thanks also to Natalie Cha, Brett Christensen, Arielle Feldshon, Idun Klakegg, Natasha Korgaonkar, Sneha Pandya and Anna Qiran Xie for excellent research assistance. Although I draw on my experience as a university executive vice president and as a reporter for an American Law Institute (ALI) project on procedural frameworks for campus sexual misconduct cases, this article expresses my views in my personal capacity only and does not comment or express views on behalf of Columbia University, ALI, or any other institution.


This article asks why it remains so difficult for employers to prevent and respond effectively to harassment, especially sexual harassment, and identifies promising points for legal intervention. It is sobering to consider social-science evidence of the myriad barriers to reporting sexual harassment—from the individual-level and interpersonal to those rooted in society at large. Most of these are out of reach for an employer but workplace culture stands out as a significant arena where employers have influence on whether harassment and other discriminatory behaviors are likely to thrive. Yet employers typically make choices in this area with attention to legal accountability rather than cultural contribution. My central claim is that these judgment calls—about policy, procedures, training, and operations—shape workplace culture and that it is a mistake to view them only through a compliance lens. With this insight, it becomes clear that each of these will be more effective in shaping culture when the employee user-experience is a focal point, and this article suggests many ways to achieve this result.

By seeing harassment prevention and response as an opportunity for culture creation in addition to being a compliance obligation, it also becomes clear that harassing behavior may negatively affect the targeted employee and the broader workplace even when there is no risk of liability. This includes “low-grade harassment,” a category I use to describe behaviors that are intentionally harassing but not severe or pervasive enough to meet doctrinal thresholds. Also relevant are microaggressions and interactions that reflect implicit bias, as these are unlikely to expose a firm to liability because they lack the discriminatory intent required by legal doctrine but nonetheless can create significant challenges for employees and organizations. This is not to suggest that employers should respond in an identical way to all of these occurrences. Rather, the point is that inattention to experiences that go beyond legal-accountability requirements is likely to spill over into the broader workplace culture and diminish the effectiveness of other harassment prevention and response efforts.

The good news is that there are specific steps an employer can take to have harassment prevention and response become part of the workplace culture rather than being sidelined as compliance. Thoughtfully crafted legislative and policy interventions, along with litigation settlements, also can bridge this gap and create a more seamless set of cultural expectations for how employees interact with each other at work and what they can expect from their employer when challenges arise.


In discussions prompted by the #MeToo movement, we have seen myriad stories of employer failures to respond in a meaningful way to sexual harassment allegations. Although there has not been a correspondingly prominent social movement regarding racial harassment at work, case law likewise reveals a similar phenomenon.1Cf. Angela Onwuachi-Willig, What About #UsToo?: The Invisibility of Race in the #MeToo Movement, 128 Yale L.J.F. 105, 111 (2018) (discussing criticism of the #MeToo movement for “ignoring the unique forms of harassment and the heightened vulnerability that women of color frequently face in the workplace”). With the Black Lives Matter movement generating renewed national attention to systemic racism, many major U.S. corporations and industry leaders have announced plans to address racial inequity in the workplace, but discussion thus far has focused more on hiring, promotion, and community investment than on harassment and other aspects of workplace climate. See Richard Feloni & Yusuf George, These Are the Corporate Responses to the George Floyd Protests that Stand out, Just Cap. (June 30, 2020), https://justcapital.com/news/notable-corporate-responses-to-the-george-floyd-protests [https://perma.cc/9764-MLU8] (collecting corporate statements made in response to the death of George Floyd, many addressing hiring and promotions but few addressing workplace culture). For an extended analysis of earlier racial-harassment cases, see generally Pat K. Chew & Robert E. Kelley, Unwrapping Racial Harassment Law, 27 Berkeley J. Emp. & Lab. L. 49 (2006).

This Article considers the power and limits of law to affect workplace harassment and other abusive behavior2By harassment, I mean remarks, physical contact, and other behaviors that are unwelcome and inappropriate in the workplace, in keeping with the United States Equal Employment Opportunity Commission (EEOC) definition of harassment as: “[U]nwelcome conduct that is based on race, color, religion, sex (including pregnancy), national origin, age (40 or older), disability or genetic information. Harassment becomes unlawful where 1) enduring the offensive conduct becomes a condition of continued employment, or 2) the conduct is severe or pervasive enough to create a work environment that a reasonable person would consider intimidating, hostile, or abusive.” Harassment, U.S. Equal Emp. Opportunity Commission, https://www.eeoc.gov/​laws/types/harassment.cfm [https://perma.cc/MNG8-7478]. The EEOC adds: “Petty slights, annoyances, and isolated incidents (unless extremely serious) will not rise to the level of illegality. To be unlawful, the conduct must create a work environment that would be intimidating, hostile, or offensive to reasonable people.” Id. Prohibited conduct, per the EEOC, “may include, but is not limited to, offensive jokes, slurs, epithets or name calling, physical assaults or threats, intimidation, ridicule or mockery, insults or put-downs, offensive objects or pictures, and interference with work performance.” Id. California law provides a helpful definition of abusive workplace behavior: “[C]onduct of an employer or employee in the workplace, with malice, that a reasonable person would find hostile, offensive, and unrelated to an employer’s legitimate business interests. Abusive conduct may include repeated infliction of verbal abuse, such as the use of derogatory remarks, insults, and epithets, verbal or physical conduct that a reasonable person would find threatening, intimidating, or humiliating, or the gratuitous sabotage or undermining of a person’s work performance. A single act shall not constitute abusive conduct, unless especially severe and egregious.” Cal. Gov’t Code § 12950.1(g)(2) (West 2020). For in-depth discussion of bullying and other abusive workplace behavior not specifically prohibited by antidiscrimination laws, see, for example, David C. Yamada, Workplace Bullying and American Employment Law: A Ten-Year Progress Report and Assessment, 32 Comp. Lab. L. & Pol’y J. 251 (2010); David C. Yamada, The Phenomenon of “Workplace Bullying” and the Need for Status-Blind Hostile Work Environment Protection, 88 Geo. L.J. 475 (2000). through the lens of workplace culture. I focus in particular on the way that employers’ choices about harassment policies, trainings, and systems for handling complaints shape that culture, often invisibly and unwittingly. I focus as well on the untapped potential of legal interventions to affect these choices and make a dent in abuses of workplace power.

The starting premise is that workplace culture at both the organizational and “local” levels plays a substantial role in shaping how employees experience the workplace,3Individuals who are not categorized formally as employees may not be entitled to certain legal protections or subjected to employee-training requirements. See Seth C. Oranburg, Unbundling Employment: Flexible Benefits for the Gig Economy, 11 Drexel L. Rev. 1, 24–25 (2018) (discussing differences in benefits and training based on employment status); Ryan Vacca, Uncertainty in Employee Status Across Federal Law, 92 Temp. L. Rev. 121, 122–23 (2019) (describing variations in legal protections for employees and independent contractors); Kara Hertzog, Opinion, How Do Independent Contractors Fit into Sexual Harassment Training Rules?, HR Dive (Dec. 20, 2019), https://www.hrdive.com/news/how-do-independent-contractors-fit-into-sexual-​harassment-training-rules/569085 [https://perma.cc/2VPQ-83UB] (pointing out the uncertainty of how mandatory sexual harassment training laws affect independent contractors). Still, they may experience the culture and people in a workplace in ways similar to employees. The additional challenges associated with harassment in co-working spaces that may not have common organizational oversight are beyond the scope here but worthy of further consideration. On interactions in co-working spaces generally, see, for example, Ricarda B. Bouncken & Andreas J. Reuschl, Coworking-Spaces: How a Phenomenon of the Sharing Economy Builds a Novel Trend for the Workplace and for Entrepreneurship, 12 Rev. Managerial Sci. 317 (2018). including their reactions to harassment and harassment-prevention efforts. Defined as the beliefs, understandings, values, and behaviors shared widely within an organization, workplace culture has multiple sources, both internal and external to the organization.4Stephen A. Linstead, Organizational Culture, in International Encyclopedia of the Social Sciences 10930 (2001). See infra Part II for extended discussion of the sources of workplace culture. While this culture is most often associated with a tone set by management, anyone who has worked in a large organization knows that the local culture also matters a great deal to an individual employee’s work experience, including the setting and the people with whom one works most directly.5Cf. Cynthia L. Estlund, Working Together: The Workplace, Civil Society, and the Law, 89 Geo. L.J. 1, 3–4 (2000) (describing the workplace as “the single most important site of cooperative interaction and sociability among adult citizens outside the family” and observing that “[t]he workplace and the relationships and conversations that it spawns may in fact be more important for the formation and interchange of political and social views among the majority of adult citizens than the voluntary civic and political organizations that make up much of civil society as it is conventionally defined”). Because organizations vary in the degree of in-person contact among employees, I refer interchangeably to workplace culture and organizational culture to underscore that this shared set of beliefs and behaviors is not limited to the physical workspace.6While the COVID-19 pandemic prompted a substantial increase in remote work, telecommuting is hardly a new phenomenon. See Adam Hickman & Jennifer Robison, Is Working Remotely Effective? Gallup Research Says Yes, Gallup (Jan. 24, 2020), https://www.gallup.com/workplace/283985/working-remotely-effective-gallup-research-says-yes.aspx [https://perma.cc/BM4Y-43PK] (describing changes in data from 2012 to 2016 and noting that the Society for Human Resource Management “found a threefold increase in the number of companies offering remote work options between 1996 and 2016”); see also Michelle A. Travis, Equality in the Virtual Workplace, 24 Berkeley J. Emp. & Lab. L. 283, 285, 302–18 (2003) (reviewing research showing “that telecommuting actually is increasing gender inequality both in the workplace and in the home”). There has been limited attention to sexual harassment in remote work environments. See, e.g., Miriam A. Cherry, A Taxonomy of Virtual Work, 45 Ga. L. Rev. 951, 979 (2011) (considering the “potential for sexual harassment, as well as other types of harassment, [to] abound in virtual work”); see also Scarlet Hannington, Yes, You Can Get Sexually Harassed While Working at Home Too, Vice (Aug. 7, 2020, 4:15 AM), https://www.vice.com/en/article/v7gxxy/sexual-harassment-working-from-home [https://perma.cc/L5CW-Q8XA]; Elisa Martinuzzi, As Work Has Moved Home, so Has Harassment, Wash. Post (June 17, 2020, 2:18 PM), https://www.washingtonpost.com/business/as-work-has-moved-home-so-has-harassment/​2020/06/17/59bf3a26-b071-11ea-98b5-279a6479a1e4_story.html [https://perma.cc/X8PP-5LY6]. Sexual harassment in other online environments has been documented more extensively. See, e.g., John T. Holden et al., The #E-Too Movement: Fighting Back Against Sexual Harassment in Electronic Sports, 52 Ariz. St. L.J. 1, 11–14 (2020) (reviewing reports of sexual harassment in the electronic gaming community). Courts also have recognized that sexual harassment can take place in virtual as well as in-person settings. See, e.g., Mary Anne Franks, Sexual Harassment 2.0, 71 Md. L. Rev. 655, 667 (2012) (discussing Blakey v. Continental Airlines, Inc., 751 A.2d 538 (N.J. 2000), which found the airline liable for sexual harassment on a pilots’ electronic bulletin board).

Law, as such, is often thought to be a minor player in day-to-day workplace interactions.7Many have taken the point further, arguing the #MeToo movement arose because of law’s failures in addressing sexual harassment and violence. See, e.g., Leigh Goodmark, #MeToo and the Failure of Law, U. Minn.: Gender Pol’y Rep. (May 22, 2018), https://genderpolicyreport.umn.edu/MeToo-and-the-failure-of-law [https://perma.cc/ABF2-BEC8] (arguing that “[r]elying on the law to change a culture in which [discriminatory] behavior flourishes is like putting a band-aid on a gaping wound,” and noting that the “after-the-fact remedy” offered by the legal system “provides justice in a very limited sense and sometimes inflicts further damage on those seeking relief”). Law and society literature explores in depth the ways in which law is mediated through other sources of authority and influences on human behavior. See, e.g., Lawrence M. Friedman, The Law and Society Movement, 38 Stan. L. Rev. 763, 763 (1986) (describing the “scholarly enterprise that examines the relationship between two types of social phenomena: those conventionally classified as ‘legal’ and those that are classified as nonlegal”). While statutes, regulations, and case law can impose liability and create pressure on employers to revise policy, set up systems, and instruct managers, they do not apply directly to individual employees.8See Goodmark, supra note 7. Further, these sources of law provide general parameters by design rather than specific guidance for the complex situations presented by real life. Formal legal intervention to resolve conflicts is also rare within a workplace.9For a discussion of alternate dispute resolution approaches to conflicts that arise in federal workplaces, see Inter-Agency Alt. Dispute Resolution Workplace Conflict Mgmt. Section, Federal Workplace Conflict Management Desk Reference: A Compilation of Alternative Dispute Resolution (ADR) Processes, Partners, and Resources 7 (Oct. 2013), https://www.adr.gov/pdf/desk-reference-handbook-2013.pdf [https://perma.cc/3LVW-3DJN], outlining extralegal options and tools for resolving workplace conflicts.

As a result, employers must make a series of judgment calls regarding complaints and conflicts that interact with these sources of law, and each of these in turn influences the workplace culture. There are, for example, the organizational policy choices that set behavioral expectations in the workplace.10For a discussion of alternate dispute resolution approaches to conflicts that arise in federal workplaces, see Inter-Agency Alt. Dispute Resolution Workplace Conflict Mgmt. Section, Federal Workplace Conflict Management Desk Reference: A Compilation of Alternative Dispute Resolution (ADR) Processes, Partners, and Resources 7 (Oct. 2013), https://www.adr.gov/pdf/desk-reference-handbook-2013.pdf [https://perma.cc/3LVW-3DJN], outlining extralegal options and tools for resolving workplace conflicts. Similarly important are the procedural and organizational-design decisions about how and where complaints will be handled.11For a discussion of alternate dispute resolution approaches to conflicts that arise in federal workplaces, see Inter-Agency Alt. Dispute Resolution Workplace Conflict Mgmt. Section, Federal Workplace Conflict Management Desk Reference: A Compilation of Alternative Dispute Resolution (ADR) Processes, Partners, and Resources 7 (Oct. 2013), https://www.adr.gov/pdf/desk-reference-handbook-2013.pdf [https://perma.cc/3LVW-3DJN], outlining extralegal options and tools for resolving workplace conflicts. There are also the operational decisions about staffing, training, and communicating these structures and policies to employees that are critical to the effective functioning of any workplace process. And there are the “as-applied” judgments to resolve discrete conflicts between employees.

These four types of judgment calls—policy, procedures, operations, and application—may not be top of mind as contributors to organizational culture but a central claim of this Article is that they should be. These choices affect far more than liability risk, and their influence on workplace culture is disregarded at an employer’s peril. An organization’s approach to its legal-compliance obligations regarding harassment and other discrimination expresses values and sets expectations that influence employees’ interactions with each other and with clients, customers, or other stakeholders. Put simply, legal-accountability culture is part of the broader workplace culture, and each culture has the potential to reinforce or undermine the other.

At its best, a thoughtful approach to the interaction between legal accountability and other aspects of workplace culture can make clear that harassment and other abuses of power are outlier behaviors within an organization. But choices related to legal accountability can also diminish or doom harassment prevention-and-response efforts.12See infra Parts III–IV. Consequently, illuminating the culture-shaping implications of these choices can teach us a great deal, both theoretically and practically, about whether employer efforts and legislative innovations designed to reduce the incidence of harassment and other discrimination are likely to have their intended effect.13There are analogous challenges and opportunities in higher-education environments. See generally Suzanne B. Goldberg, Is There Really a Sex Bureaucracy?, 7 Calif. L. Rev. Online 107 (2016) (discussing factors that shape educational institutions’ efforts to prevent and respond to sexual harassment and other sexual misconduct).

None of these choices takes place in the abstract, so this Article proceeds by digging into the factors that make workplace harassment so seemingly intractable, starting with the dynamics that individuals bring into the workplace, then moving to organizational and external factors, and ultimately to the ways in which legislative interventions might prompt greater attention to legal-accountability cultures within organizations.

Part I offers an extensive map of individual, organizational, and societal factors that feed into an organization’s culture, drawing on the socio-ecological model used in public health to assess complex environments.

Part II looks in greater depth at how variations in an employer’s attention to workplace culture affect an employee’s “local” environment.

Part III then turns to employers’ legal-accountability decisions, examining the sources of law that make up the compliance landscape. This Part gives special attention to one of the more difficult choices employers face: how to handle behaviors that do not give rise to legal liability but may look and feel like harassment, especially to the targeted employee. These include what I call “low-grade harassment,” a category of behaviors that are clearly harassing but not “severe or pervasive” enough to meet doctrinal thresholds.14See infra notes 174–84 and accompanying text. In particular, see note 175 for discussion of the “severe or pervasive” legal standard applied by federal courts to assess sexual harassment claims under Title VII, 42 U.S.C. § 2000e-2 (2018), and of state jurisdictions that do not use the “severe and pervasive” formulation. The analysis here carries over to other forms of discrimination as well, with comments and actions that are similar in kind but not degree to acts that “count” for legal liability. The discussion also considers the effects of implicit biases and microaggressions, including indirect, subtle, or unintentional comments and behaviors that can have a significant negative effect on an employee’s ability to participate fully in the workplace.15See infra note 166–69 and accompanying text.

Part IV then considers several points at which legislative or doctrinal interventions might do more to shape organizational culture, looking first at the potential for influencing individual employees and the external environment. Primarily, though, this Part focuses on organization-level interventions related to policy, procedure, and training and considers whether any of these can produce a shift in the legal-accountability choices that shape the overarching culture. Part V concludes.

* * *

In essence, then, this Article aims to deepen our understanding of why it continues to be so difficult for employers to prevent and respond effectively to sexual harassment; to clarify law’s relevance to our thinking about workplace culture; and to identify points of intervention for legislatures, employers, and advocates that have potential to support constructive change. While I focus on sexual harassment, the analysis is intended to carry over to other forms of harassment and discrimination as well.

The discussion seeks to complement legal scholarship that takes other approaches to sexual harassment law, such as identifying and correcting doctrinal defects, conceptualizing sexual harassment in relation to other forms of discrimination, and critiquing non-disclosure agreements and other legal arrangements that have enabled harassment to continue without penalty.16There is already copious scholarship addressing these issues in the wake of the #MeToo movement, building on earlier work in this area. See Vicki Schultz, Open Statement on Sexual Harassment from Employment Discrimination Law Scholars, 71 Stan. L. Rev. Online 17 (2018) [hereinafter Schultz, Open Statement] (offering principles and proposals to reform sexual harassment law, and citing earlier foundational sources); see also, e.g., Deborah L. Brake, Coworker Retaliation in the #MeToo Era, 49 U. Balt. L. Rev. 1 (2019) (evaluating the interaction of retaliation law with #MeToo claims); Claudia Flores, Beyond the Bad Apple—Transforming the American Workplace for Women After #MeToo, 2019 U. Chi. Legal F. 85 (2019) (considering international and comparative law sources to examine the role of dignity and equality in sexual harassment law); L. Camille Hébert, Is “MeToo” Only a Social Movement or a Legal Movement Too?, 22 Emp. Rts. & Emp. Pol’y J. 321 (2018) (discussing doctrinal changes that might flow from the #MeToo movement); Ramit Mizrahi, Sexual Harassment Law After #MeToo: Looking to California as a Model, 128 Yale L.J.F. 121, 150 (2018) (examining state-law reforms and arguing that “measures targeting retaliation can have the most positive effect in reducing sexual harassment” because of “the chilling effect [retaliation] has on some employees’ willingness to come forward with their own stories”); Jean R. Sternlight, Mandatory Arbitration Stymies Progress Towards Justice in Employment Law: Where to, #MeToo?, 54 Harv. C.R.-C.L. L. Rev. 155, 156–57 (2019) (arguing that “continued progress towards justice is currently in jeopardy due to companies’ imposition of mandatory arbitration on their employees” (footnotes omitted) and that “[b]y denying their employees access to court, companies are causing employment law to stultify”). For earlier work, see, for example, Kathryn Abrams, Gender Discrimination and the Transformation of Workplace Norms, 42 Vand. L. Rev. 1183, 1202–09 (1989), advocating an approach that takes into account that sexual harassment is often experienced differently by women than by men; Katherine M. Franke, What’s Wrong with Sexual Harassment?, 49 Stan. L. Rev. 691 (1997), arguing that sexual harassment is a technology of sexism; Vicki Schultz, Reconceptualizing Sexual Harassment, 107 Yale L.J. 1683 (1998), linking sexual harassment to gender hierarchies in workplaces. For related discussions of the limitations of discrimination law more generally, see, for example, Samuel R. Bagenstos, The Structural Turn and the Limits of Antidiscrimination Law, 94 Calif. L. Rev. 1, 22–26 (2006), arguing that judicial opposition to second-guessing employer decisions stymies structural approaches to discrimination claims; Richard Thompson Ford, Bias in the Air: Rethinking Employment Discrimination Law, 66 Stan. L. Rev. 1381, 1384 (2014), arguing that “law should replace the conceptually elusive goal of eliminating discrimination with the more concrete goal of requiring employers, government officials, and other powerful actors to meet a duty of care to avoid unnecessarily perpetuating social segregation or hierarchy”; Linda Hamilton Krieger, The Content of Our Categories: A Cognitive Bias Approach to Discrimination and Equal Employment Opportunity, 47 Stan. L. Rev. 1161 (1995), examining the role of cognitive bias in employment discrimination; Michael Selmi, Why Are Employment Discrimination Cases so Hard to Win?, 61 La. L. Rev. 555, 561 (2001), arguing that “courts approach cases from a particular perspective that reflects a bias against the claims” and that this ideological bias colors how courts adjudicate discrimination claims; Sandra F. Sperino & Suja A. Thomas, Unequal: How America’s Courts Undermine Discrimination Law (2017), showing through extensive discussion of caselaw how courts have constrained the capacity of antidiscrimination law to protect workers. I have also written in this area. See, e.g., Suzanne B. Goldberg, Discrimination by Comparison, 120 Yale L.J. 728 (2011) (addressing the ways that comparators function as a barrier to complex discrimination claims). For social-science-based approaches to similar questions, see Catherine Albiston, Institutional Inequality, 2009 Wis. L. Rev. 1093, 1094 (2009), arguing from a sociological perspective that in applying antidiscrimination law, courts fail to take sufficient account of the structural conditions and institutional processes that give rise to discrimination; Lauren B. Edelman & Jessica Cabrera, Sex-Based Harassment and Symbolic Compliance, 16 Ann. Rev. L. & Soc. Sci. 361 (2020), reviewing social science and other scholarship on sexual harassment and arguing that organizational policies tend to be of limited effect, in part as a result of judicial doctrine that gives undue weight to symbolic compliance. It also adds a legal lens to social scientific investigations of harassment, discrimination, and employers’ capacity to effect change in workplace dynamics.17See infra Parts I–II.

There are several caveats and contextual notes needed for the arguments that follow. First is a familiar but important point that sexual harassment involves the exercise of power and is best understood in the context of broader inequities related to sex and gender in the workplace and surrounding society.18See, e.g., Schultz, Open Statement, supra note 16, at 19 (“[T]he bottom line is that harassment is more about upholding gendered status and identity than it is about expressing sexual desire or sexuality.”). Similarly, racial harassment cannot be understood without reference to racialized dynamics of power in workplaces and their surroundings. See, e.g., Chew & Kelley, supra note 1, at 73–75 (describing incidents of racial harassment and workplace power dynamics). This has been recognized in doctrine, which does not require misconduct to be “motivated by sexual desire to support an inference of discrimination on the basis of sex.”19Oncale v. Sundowner Offshore Servs., Inc., 523 U.S. 75, 80 (1998). The Equal Employment Opportunity Commission has likewise affirmed that harassment can include “unwelcome sexual advances, requests for sexual favors, and other verbal or physical harassment of a sexual nature” but “does not have to be of a sexual nature . . . and can include offensive remarks about a person’s sex.”20Sex-Based Discrimination, U.S. Equal Emp. Opportunity Commission, https://www.eeoc.gov/sex-based-discrimination [https://perma.cc/J4Y5-6TNA]. Scholars, too, have made this point in numerous disciplines, observing and analyzing the use of power to demean women and others who are perceived as weak, vulnerable, or otherwise not fitting within the environment for reasons related to sex, sexual orientation, gender identity, or gender expression.21See supra note 16 (discussing earlier works of Abrams, Franke, and Schultz); see also infra Part I.

Second, the analysis here is not offered as a comprehensive solution to the problem of workplace harassment, much less as a complete response to the need for more fully inclusive workplaces.22Susan Sturm’s work on the architecture of inclusion is an excellent resource for thinking about these broader questions. See Susan Sturm, The Architecture of Inclusion: Advancing Workplace Equity in Higher Education, 29 Harv. J.L. & Gender 247 (2006) [hereinafter Sturm, The Architecture of Inclusion]. My hope, though, is for the discussion to make clear that employers’ obligations to address harassment and other discrimination are about more than protecting against liability.23For this reason, although there is much to say about theories of discrimination and the extent to which law has the capacity to provide meaningful redress, see supra note 7, I will leave those debates to one side here and keep my focus on the framework and lessons just described. Fundamentally, these obligations are, or can be, a site of culture-creation and a bridge to strengthening a more inclusive, less discriminatory workplace culture. For employers that do not already devote resources to these larger efforts, understanding their response to legal obligations as a tool for building culture may be a prompt to further action.24Laura Beth Nielsen, Robert Nelson, and others have argued that legal intervention will do nothing more than play into an existing compliance-oriented culture in which employers do the minimum required to avoid liability and courts endorse those choices, thereby rendering permissible all but the very worst behaviors. See, e.g., Ellen Berry, Laura Beth Nielsen & Robert L. Nelson, Rights on Trial: How Workplace Discrimination Law Perpetuates Inequality 39–40 (2017); Laura Beth Nielsen, Robert L. Nelson & Ryon Lancaster, Individual Justice or Collective Legal Mobilization? Employment Discrimination Litigation in the Post Civil Rights United States, 7 J. Empirical Legal Stud. 175, 176 (2010). More specifically, they argue that it is not possible for law to promote equality in workplaces given the incentives created by legal-compliance regimes. See Berry, Nielsen & Nelson, supra, at 39–40 (arguing that discrimination law disincentivizes employers from providing meaningful protection to employees); Nielsen, Nelson & Lancaster, supra, at 176 (reviewing an extensive set of employment discrimination case filings and concluding that plaintiffs “receive cursory attention in legal process and a limited remedy”). While my view is that law can be effective in reducing barriers to equality, their arguments bear noting for the cautions they offer regarding the analysis here. The same goes for organizations with great-looking but ineffective policies.

It also bears noting that workplace ecologies fluctuate in response to changes in organizational leadership.25See, e.g., Gary W. Evans et al., Capturing the Ecology of Workplace Stress with Cumulative Risk Assessment, 44 Env’t & Behav. 136 (2012) (examining the cumulative effects of multiple factors that contribute to employee stress); Crystal M. Harold & Brian C. Holtz, The Effects of Passive Leadership on Workplace Incivility, 36 J. Organizational Behav. 16 (2015) (finding that passive leadership has a significant effect on behavioral incivility and employees’ experience of incivility); Junghyun Lee & Jaclyn M. Jensen, The Effects of Active Constructive and Passive Corrective Leadership on Workplace Incivility and the Mediating Role of Fairness Perceptions, 39 Group & Org. Mgmt. 416 (2014) (finding a relationship between active constructive leadership and decrease in workplace incivility). See infra Part I for discussion of other influences. Dynamics at the local level—we might call them microclimates—will also produce variation within organizations and mediate efforts to create change.26C.f. Gregory A. Aarons & Angelina C. Sawitzky, Organizational Climate Partially Mediates the Effect of Culture on Work Attitudes and Staff Turnover in Mental Health Services, 33.3 Admin. & Pol’y Mental Health & Mental Health Servs. Res. 289, 290 (2006) (observing that “[w]hile culture reflects behaviors, norms, and expectations, organizational climate reflects workers’ perceptions of and emotional responses to the characteristics of the work environment”). In other words, this is an area where one size definitely does not fit all, and the effectiveness of interventions may vary over time. My aim is thus not to identify what will work in every setting for every moment. It is, instead, to consolidate our understanding of the factors in play and how employer choices about legal accountability affect an organization’s culture and the ways people interact at work.

I. The Landscape: Individual Employees and Their Surroundings

The array of factors bearing on whether harassment is likely to occur and persist in a given workplace is vast. Yet we must contend with this vast array if we are to understand the conditions that affect whether harassment prevention and response efforts are likely to succeed and to identify entry points for law and barriers to the effectiveness of legal interventions.

The socio-ecological model, developed in public health as a method for evaluating complex influences on behavior, is especially useful for this effort. The model insists that we look at the interactions among multiple ecosystems: individuals, their relationships to each other, the community in which those relationships take place, and the institutional and societal factors that bear on the environment.27C.f. Gregory A. Aarons & Angelina C. Sawitzky, Organizational Climate Partially Mediates the Effect of Culture on Work Attitudes and Staff Turnover in Mental Health Services, 33.3 Admin. & Pol’y Mental Health & Mental Health Servs. Res. 289, 290 (2006) (observing that “[w]hile culture reflects behaviors, norms, and expectations, organizational climate reflects workers’ perceptions of and emotional responses to the characteristics of the work environment”). It helpfully resists reductionist thinking that one act or one type of intervention will make all of the difference.

This Part sets out an extensive inventory using a simplified version of the socio-ecological model’s typology, looking primarily at individuals but also briefly at organizational and societal influences. To be sure, no discussion could possibly capture every variation in individual and workplace dynamics and even among the elements included here, there are as many variations as there are people and environments.28C.f. Gregory A. Aarons & Angelina C. Sawitzky, Organizational Climate Partially Mediates the Effect of Culture on Work Attitudes and Staff Turnover in Mental Health Services, 33.3 Admin. & Pol’y Mental Health & Mental Health Servs. Res. 289, 290 (2006) (observing that “[w]hile culture reflects behaviors, norms, and expectations, organizational climate reflects workers’ perceptions of and emotional responses to the characteristics of the work environment”). For this reason, the Sections below collect archetypal issues, understanding that each may vary in appearance and significance depending on the environment.

A. The Individual

Harassment occurs between and among individuals and there are, not surprisingly, many factors at the level of the individual that bear on harassment prevention and response in any workplace.29By harassment, I mean to include both unwelcome physical contact and other behaviors that can create a hostile environment. See supra note 2. As noted at the outset, although this Article focuses primarily on harassment, my arguments aim to encompass other forms of discrimination as well. The discussion here groups these as factors that affect 1) individuals’ decision to report harassment; 2) others’ willingness to step in, which is often called “bystander intervention”;30Bystander intervention, sometimes called “upstander” intervention, has, like workplace culture, become a popular focal point for harassment-prevention efforts. See generally Sarah L. Swan, Bystander Interventions, 2015 Wis. L. Rev. 975, 977–78 (2015) (describing the growing use of bystander intervention trainings and initiatives in educational and employment settings). My own view is that “stepping in” is a better characterization. It sounds easier and less clinical than “bystander intervention” and avoids the at-times incorrect suggestion that those nearby had no role in creating the conditions that supported the troubling behavior. For use of this term in another context, see Stepping In: How You Can Help to Keep Columbia Safe, Colum. Univ., https://covid19.columbia.edu/stepping-in (last visited Dec. 21, 2020), describing strategies for stepping in to help prevent COVID-19 transmission. and 3) a potential perpetrator’s likelihood of engaging in harassing behaviors in a particular setting or toward a particular coworker.

1. To report or not to report

As coverage of #MeToo and earlier literature on sexual harassment make clear, individuals who are subjected to harassment—even egregious harassment over a long period of time—often share little about their experience with others, and when they do, it is more often with a friend or family member than a colleague or supervisor.31See Chai R. Feldblum & Victoria A. Lipnic, Select Task Force on the Study of Harassment in the Workplace, U.S. Equal Emp’t Opportunity Comm’n, Report of the Co-Chairs 16 (2016), https://www.eeoc.gov/sites/default/files/migrated_files/eeoc/task_force/harassment/report.pdf [https://perma.cc/X69Y-Q3C5] (reviewing low rates of formal reporting); see also Brake, supra note 16, at 51 (discussing reports and caselaw on reluctance to report); Andrew Tae-Hyun Kim, Culture Matters: Cultural Differences in the Reporting of Employment Discrimination Claims, 20 Wm. & Mary Bill Rts. J. 405, 435, 455 (2011) (examining the role of culture in affecting the willingness of Asian‑Americans to report discrimination); cf. David Cantor et al., Report on the AAU Campus Climate Survey on Sexual Assault and Sexual Misconduct, at iv (2015) (reviewing survey data on campus sexual assault, including on reporting rates and reasons students did not file formal complaints). Although this is changing somewhat as the #MeToo movement and social media have provided support and new mechanisms for people to tell their stories, disclosing harassment in the workplace remains difficult for many.32See Brake, supra note 16, at 51. While the discussion here focuses on coworker and supervisor harassment, harassment and other discrimination by customers, clients, vendors, and others can also have a significant effect on employees. Nearly every point in this Section carries over to these contexts as well. For broader discussion of discrimination by customers, see Katharine T. Bartlett & Mitu Gulati, Discrimination by Customers, 102 Iowa L. Rev. 223, 224–25 (2016), analyzing the effects of customer discrimination and the absence of legal prohibitions.

Consequently, to be effective, any efforts to encourage employees to report incidents, particularly to a supervisor or human resources manager, will need to take account of factors that inhibit or support employees in disclosing harassment. As will be apparent from the discussion below, these factors are often mutually reinforcing, underscoring the complexity of the landscape that harassment law and policy aim to affect.

The first is perhaps the most deeply personal: shame and embarrassment. Many individuals who have experienced harassment indicate in studies and surveys that they felt ashamed or embarrassed about what happened.33See Pamela J. Foster & Clive J. Fullagar, Why Don’t We Report Sexual Harassment? An Application of the Theory of Planned Behavior, 40 Basic & Applied Psychol. 148, 157 (2018) (“[S]urvey participants reported that being embarrassed would stop them from reporting an incident of sexual harassment.”); Kim, supra note 31, at 426–27, 431 (summarizing structural barriers to reporting sexual harassment claims, including the desire to avoid embarrassment as well as concerns about social relationships, power dynamics, low self-esteem, and fear of retaliation). Lower hardiness, lower optimism, and lower trait activism have also been identified as factors affecting the willingness of “stigmatized individuals” to confront prejudice in the workplace. Aneeta Rattan & Carol S. Dweck, What Happens After Prejudice Is Confronted in the Workplace? How Mindsets Affect Minorities’ and Women’s Outlook on Future Social Relations, 103 J. Applied Psychol. 676, 677 (2018). Some say they felt they brought the experience on themselves (i.e. that they were “asking for it” in some way) because of a friendship they cultivated with the colleague or because they drank too much at an office gathering, did not speak up about what was happening or did not repeatedly reject a colleague’s sexualized touching.34See, e.g., Lauren P. Daley et al., Catalyst, Sexual Harassment in the Workplace: How Companies Can Prepare, Prevent, Respond, and Transform Their Culture 3 (2018) https://www.catalyst.org/wp-content/uploads/2019/01/​sexual_​harassment​_in_the​_workplace_report.pdf [https://perma.cc/X23Q-M4TP] (noting that “[b]oth fear and shame can contribute to a culture of silence in the workplace”); Jacey Fortin, #WhyIDidntReport: Survivors of Sexual Assault Share Their Stories After Trump Tweet, N.Y. Times (Sept. 23, 2018), https://www.nytimes.com/2018/09/23/us/why-i-didnt-report-assault-stories.html [https://perma.cc/5KE7-NHCP] (reporting stories of women who “felt that people would not believe her because she was in a relationship when she was assaulted”; “felt shame because she had been drinking at a party before it happened”; and “felt vulnerable[,] . . . humiliated[,] . . . [and worried that if she said anything, her] career would be over”); see also, e.g., Salma Hayek, Opinion, Harvey Weinstein Is My Monster Too, N.Y. Times (Dec. 12, 2017), https://nyti.ms/2nXntXL (describing being “ashamed” about Harvey Weinstein’s abuse toward her); cf. Foster & Fullagar, supra note 33, at 157 (explaining that “[t]argets are often confused about what just happened to them, they fear they will not be believed, others will think poorly about them, they may experience shame and guilt, and they often do not recognize the incident as serious enough to report”). On college campuses this manifests in data showing substantial numbers of students who do not disclose a sexual assault to anyone or do not disclose to a school official, citing embarrassment as the reason. See Cantor et al., supra note 31, at iv (“A significant percentage of [college] students say they did not report because they were ‘embarrassed, ashamed or that it would be too emotionally difficult.’”); see also Caroline Mellgren et al., “It Happens All the Time”: Women’s Experiences and Normalization of Sexual Harassment in Public Space, 28 Women & Crim. Just. 262, 262–81 (2018) (discussing research on female university students in Sweden showing a reluctance to report and worries about revictimization). Self-blame has also been identified as a reason some women are reluctant to seek help after experiencing domestic violence. See, e.g., Alison J. Towns & Peter J. Adams, “I Didn’t Know Whether I Was Right or Wrong or Just Bewildered”: Ambiguity, Responsibility, and Silencing Women’s Talk of Men’s Domestic Violence, 22 Violence Against Women 496, 514 (2015) (describing research suggesting “that at the height of physical and emotional abuse the woman may be most vulnerable to constructing herself as responsible for initiating the man’s violence”).

Closely related to shame and embarrassment is a fear of being negatively judged by others, whether peers, supervisors, or human resources staff.35As one recent study observed, “having some assurance that an individual’s reputation would not be ruined” upon reporting is a significant factor for individuals deciding whether to report harassment. Foster & Fullagar, supra note 33, at 156. Closely related is the fear of relinquishing privacy. See, e.g., Margaret E. Bell et al., Victims’ Psychosocial Well-Being After Reporting Sexual Harassment in the Military, 15 J. Trauma & Dissociation 133, 135 (2014) (discussing negative career consequences and loss of peer support as consequences of reporting). For many, this reflects a realistic reputational concern about being perceived as disruptive or, in certain settings, as unprofessional.36See L. Camille Hébert, Why Don’t “Reasonable Women” Complain About Sexual Harassment?, 82 Ind. L.J. 711, 731, 741 (2007) (describing reasons why employees are reluctant to report sexual harassment, including fear of others’ reactions); see also Rattan & Dweck, supra note 33, at 677 (observing that “[t]hose targeted by prejudice consistently report wanting to confront, but their behavior is often constrained by situational factors (e.g., potential costs, a public context, power dynamics)” (citations omitted)); cf. Heather McLaughlin et al., Sexual Harassment, Workplace Authority, and the Paradox of Power, 77 Am. Soc. Rev. 625, 641 (2012) (describing women supervisors who “repeatedly spoke about feeling isolated and of harassment by co-workers and subordinates directed toward putting them ‘in their place’” and observing that “[w]hether attempting to prove they could lead a team of workers or prove themselves as women in masculine fields, women’s isolation in these positions repeatedly left them vulnerable to harassment”). For discussion of similar concerns in the higher-education setting, see Shamus R. Khan et al., “I Didn’t Want to Be ‘That Girl’”: The Social Risks of Labeling, Telling, and Reporting Sexual Assault, 5 Soc. Sci. 432, 433–34 (2018). Indeed, social retaliation, including negative gossip and ostracism by coworkers, which “can come from individuals at any level of the organization—peers, superiors, and subordinates[—] . . . occurs at roughly twice the rate of professional retaliation, and it carries equivalent professional and psychological harms.”37Louise F. Fitzgerald & Lilia M. Cortina, Sexual Harassment in Work Organizations: A View from the 21st Century, in 2 American Psychological Association Handbook of the Psychology of Women: Perspectives on Women’s Private and Public Lives 228 (C.B. Travis & C.W. White eds., 2018) (citing Lilia M. Cortina & S. Arzu Wasti, Profiles in Coping: Responses to Sexual Harassment Across Persons, Organizations, and Cultures, 90 J. Applied Psychol. 182 (2005)).

Conversely, employees who expect that their peers will be supportive may be more likely to report38Foster & Fullagar, supra note 33, at 157.—a data-point that has clear policy relevance. As one recent study observed, “[m]any people have a strong reliance on their friends and members of their social organizations when they find themselves in an uncomfortable situation such as hostile environment sexual harassment.”39Id. A widely cited analysis of many studies also found that the influence of friends and peers emerged as a significant factor. Paula McDonald et al., Developing a Framework of Effective Prevention and Response Strategies in Workplace Sexual Harassment, 53 Asia Pac. J. Hum. Resources 41, 52–53 (2015). This insight might be operationalized by, for example, allowing employees to register for trainings with peer groups or friends so that each knows the others might remember something about the employer’s policies and how and where to find help.40Id.; see also Nicole T. Buchanan et al., A Review of Organizational Strategies for Reducing Sexual Harassment: Insights from the U.S. Military, 70 J. Soc. Issues 687, 692 (2014) (“[T]raining many employees within a workgroup resulted in improved recognition of harassing behaviors and attitudes toward harassment, over and above the individual effects of training.” (citing Heather Antecol & Deborah Cobb-Clark, Does Sexual Harassment Training Change Attitudes? A View from the Federal Level, 84 Soc. Sci. Q. 826 (2003))). I will return to this in Part IV.

Retaliation by the perpetrator is also well-established as a reason for non-disclosure.41Many sources discuss retaliation. See, e.g., Theresa M. Beiner, Using Evidence of Women’s Stories in Sexual Harassment Cases, 24 U. Ark. Little Rock L. Rev. 117, 124–25 (2001) (“[M]any plaintiffs’ lawyers would tell you that once an employee complains about discrimination on the job, he or she can usually consider that employment relationship over.”); Mindy E. Bergman et al., The (Un)reasonableness of Reporting: Antecedents and Consequences of Reporting Sexual Harassment, 87 J. Applied Psychol. 230, 232 (2002) (discussing evidence showing that women suffered adverse consequences from reporting sexual harassment); Deborah L. Brake, Retaliation, 90 Minn. L. Rev. 18, 32–42 (2005) (describing studies of retaliation); Mizrahi, supra note 16, at 150 (2018) (discussing the importance of protection against retaliation). For recent reports discussing retaliation in academic and judicial environments, see infra note 72. Retaliation became increasingly difficult to prove in litigation following the Supreme Court’s ruling in University of Texas Southwestern Medical Center v. Nassar, 570 U.S. 338 (2013), which requires plaintiffs to prove that retaliation was the but-for cause of the adverse employment action. Id. at 339; cf. id. at 363 (Ginsburg, J., dissenting) (“‘[F]ear of retaliation is the leading reason why people stay silent’ about the discrimination they have encountered or observed.” (quoting Crawford v. Metro. Gov’t of Nashville & Davidson Cty., 555 U.S. 271, 279 (2009))). This may take the form of direct threats to personal safety, the safety of others, or workplace opportunities, including, at the extreme, termination and interference with employment elsewhere.42See Brake, supra note 41, at 44–46, 93-98 (describing retaliation claims).

It also bears noting, as an individual-level factor, that some employees do not label what is happening to them or to others as harassment, even when an outside observer might see the conduct as severe or pervasive.43For a meta-analysis of research examining the ways that race and sex affect perceptions of harassment and discrimination in the workplace, see Mallory A. McCord et al., A Meta-Analysis of Sex and Race Differences in Perceived Workplace Mistreatment, 103 J. Applied Psychol. 137 (2018). See also Justine E. Tinkler, “People Are Too Quick to Take Offense”: The Effects of Legal Information and Beliefs on Definitions of Sexual Harassment, 33 L. & Soc. Inquiry 417, 425 (2008) (“A vast empirical literature has shown that men are less likely than women to consider verbal comments, sexual humor, severe forms of quid pro quo harassment, and unwanted pressure for sexual relationships as harassment.” (citation omitted)). For additional discussion of race and sex as factors affecting perceptions of harassment, see Tanya Katerí Hernández, A Critical Race Feminism Empirical Research Project: Sexual Harassment & the Internal Complaints Black Box, 39 U.C. Davis L. Rev. 1235, 1240–41 (2006); Tanya Katerí Hernández, Sexual Harassment and Racial Disparity: The Mutual Construction of Gender and Race, 4 J. Gender Race & Just. 183, 191–92 (2001). Cf. Jeanne Murphy et al., “They Talk like that, but We Keep Working”: Sexual Harassment and Sexual Assault Experiences Among Mexican Indigenous Farmworker Women in Oregon, 17 J. Immigrant Minority Health 1834, 1838 (2015) (discussing migrant farmworkers who “reported widespread awareness of sexual harassment behaviors that they might not label as ‘sexual harassment’”). Race, sex, age, national origin, and other aspects of identity and personal backgrounds, along with individuals’ perceptions of what is to be expected in particular workplaces, all have been identified as potential influences.44See supra note 43. For additional discussion of factors that influence employee perceptions of harassment, see, for example, Richard C. Sorenson et al., Solving the Chronic Problem of Sexual Harassment in the Workplace: An Empirical Study of Factors Affecting Employee Perceptions and Consequences of Sexual Harassment, 34 Cal. W. L. Rev. 457, 460–61 (1998) (footnotes and citations omitted), identifying “demographic variables such as gender, occupational status, and education; personal experience with harassment; and sex-role identity” along with “personality and attitude variables such as attitude towards women and sex, religiosity, and locus of control.”

Perceptions of law and policy may matter as well. Employees may know that it is hard to “win” a lawsuit45On the low levels of litigation success experienced by most discrimination claimants, see, for example, Kevin M. Clermont & Stewart J. Schwab, How Employment Discrimination Plaintiffs Fare in Federal Court, 1 J. Empirical Legal Stud. 429, 449–50 (2004), showing disproportionately high reversal rate for discrimination plaintiffs who prevail at trial as compared to employers who prevail at trial; see also Kevin M. Clermont & Theodore Eisenberg, Plaintiphobia in the Appellate Courts: Civil Rights Really Do Differ from Negotiable Instruments, 2002 U. Ill. L. Rev. 947, 958 (2002), describing employment discrimination plaintiffs as “one of the least successful classes of plaintiffs at the trial court level” as well as on appeal. For a brief video that summarizes the unfavorable outcomes for plaintiffs in employment discrimination litigation, drawn from an extended study of cases, see American Bar Foundation, Rights on Trial Book Trailer, Vimeo (July 19, 2017), https://vimeo.com/226216019, discussing Berrey, Nielsen & Nelson, supra note 24. or to prevail with human resources, even if they do not know the doctrinal details. And this belief may feed into their understanding of what counts as harassing behavior.46See Tinkler, supra note 43, at 440 (describing “a complexity in the way people reconcile their knowledge of the law with their personal views about power and social interaction in the workplace”).

Even without taking law into consideration, many employees do not report incidents because they believe that meaningful redress by their employer is unlikely.47See, e.g., Jim Rutenberg et al., At Fox News, Kisses, Innuendo, Propositions and Fears of Reprisal, N.Y. Times (July 23, 2016), https://www.nytimes.com/2016/07/24/business/at-fox-news-kisses-innuendo-propositions-and-fears-of-reprisal.html [https://perma.cc/G3ZU-LSMW] (“Almost all the women said they were reluctant to go to the human resources department with their complaints for fear that they would be fired.”); see also Samantha Cooney, A Timeline of Sexual Harassment Allegations at Fox News, Time (May 2, 2017, 11:12 AM), https://time.com/4757734/timeline-sexual-harassment-allegations-fox-news [https://perma.cc/ZN2Q-B9SA]. Sometimes they may have seen or heard others’ concerns dismissed as trivial (e.g. “He’s just that way with everyone” or “Don’t be so sensitive”) or illegitimate (“He wouldn’t do that” or “You must have misunderstood”).48See Foster & Fullagar, supra note 33, at 187 (describing employees’ concern about not being believed). This is especially—though not only—an issue when the alleged harasser is more senior or otherwise more powerful within the environment.49See, e.g., Rob Copeland, Google Agreed to Pay $135 Million to Two Executives Accused of Sexual Harassment, Wall St. J. (Mar. 11, 2019, 8:52 PM), https://www.wsj.com/articles/google-agreed-to-pay-135-million-to-two-executives-accused-of-sexual-harassment-11552334653 [https://perma.cc/N2ZZ-8BCZ]; Tiffany Hsu & Mohammed Hadi, Wynn Leaders Helped Hide Sexual Misconduct Allegations Against Company’s Founder, Report Says, N.Y. Times (Apr. 2, 2019), https://www.nytimes.com/2019/04/02/business/wynn-resorts-sexual-misconduct-steve-wynn.html [https://perma.cc/3P3D-NZD7]; Mike Isaac, Inside Uber’s Aggressive, Unrestrained Workplace Culture, N.Y. Times (Feb. 22, 2017), https://www.nytimes.com/2017/02/22/technology/uber-workplace-culture.html [https://perma.cc/S55B-TJ6E] (“[H]uman resources often made excuses for top performers because of their ability to improve the health of the business. . . . [E]xecutives who were personally close to [the CEO] were shielded from much accountability over their actions.”). Doubts about the effectiveness or judgment of staff responsible for addressing concerns can also inhibit reporting, as can skepticism about their ability to protect the employee against retaliation.50See Foster & Fullagar, supra note 33, at 156 (indicating that employees are more likely to report when they have confidence that perpetrators of harassment will be sanctioned appropriately and that protections against retaliation are meaningful). And if the alleged harasser is a client or a vendor, or the incident took place at a conference or other offsite venue, the employee may sense, perhaps correctly, that little can or will be done to address the problem.51On sexual harassment by third parties, see, for example, Einat Albin, Customer Domination at Work: A New Paradigm for the Sexual Harassment of Employees by Customers, 24 Mich. J. Gender & L. 167 (2017), describing the harmful consequences of sexual harassment by customers and clients and observing that “the prevailing legal paradigm” governing harassment by third parties leads to “a weaker form of liability that provides limited protection to employees suffering from harassment.” Id. at 169–71; see also Lea B. Vaughn, The Customer Is Always Right . . . Not! Employer Liability for Third Party Sexual Harassment, 9 Mich. J. Gender & L. 1, 3 (2002) (describing “third party sexual harassment [as] a prevalent form of harassment that the legal system does not currently nor energetically pursue”). Personnel privacy policies and non-disclosure agreements that conceal what actually happens to those who violate workplace policy may also reinforce employees’ beliefs that complaints are unlikely to prompt action.52See, e.g., Schultz, Open Statement, supra note 16, at 46; see also Ian Ayres, Targeting Repeat Offender NDAs, 71 Stan. L. Rev. Online 76, 78 (2018); Orly Lobel, NDAs Are out of Control. Here’s What Needs to Change, Harv. Bus. Rev. (Jan. 30, 2018), https://hbr.org/2018/01/ndas-are-out-of-control-heres-what-needs-to-change. Part III will suggest a practice of issuing de-identified, aggregate reports to address this problem.

2. See something, say something?

Some of the same factors that influence an employee’s decision whether to report an experience of harassment—a sense of self-efficacy53See, e.g., Timothy A. Judge et al., Self-Efficacy and Work-Related Performance: The Integral Role of Individual Differences, 92 J. Applied Psychol. 107 (2007). and a view about what constitutes harassment—also affect whether coworkers help when they learn about concerning behavior, either by addressing the situation themselves or reporting an incident to someone who can take action.54See Jane Adams-Roy & Julian Barlin, Predicting the Decision to Confront or Report Sexual Harassment, 19 J. Organizational Behav. 329, 330 (1998). Reputational concerns are similarly influential—in the bystander context, the question is whether an individual who steps into a situation is seen by others as helpful or meddling.55Id. Knowledge about where to go for help and how the employer is likely to respond may also influence whether coworkers report incidents and concerns.56Id.; see also Tinkler, supra note 43, at 439 (observing that “the extent that people view sexual harassment rules as ambiguous and threatening to workplace norms affects how they define sexual harassment”).

Workplace policies and culture have a particular role to play here. Even individuals who are not temperamentally inclined to address a colleague’s behavior may overcome their reluctance if they are required or expected to act because of their role as a supervisor or manager.57Many higher-education settings, for example, instituted broad-based required reporting of sexual harassment involving students under the Obama-era Title IX guidance from the Department of Education. Although this guidance has since been withdrawn, anecdotal evidence suggests that many colleges and universities retained this required-reporting policy, at least in the immediate aftermath of the policy change. For the history of these regulations and an alternate approach that requires information-sharing but limits the number of individuals with reporting obligations, see Merle H. Weiner, A Principled and Legal Approach to Title IX Reporting, 85 Tenn. L. Rev. 71 (2017). Weiner argues that broad required-reporting policies are ineffective for a variety of reasons, including that they are frequently not followed and may deter students from seeking help from faculty members and administrators whom they know. Id. at 73. My own views have evolved over time. Although there are benefits to having faculty members serve as confidential resources for students, accounts of persistent but unreported harassment have persuaded me that, on balance, required reporting better serves the well-being of students and others. For this reason, accountability measures that require managers to report concerns—and impose consequences if they do not—are central to the Equal Employment Opportunity Commission’s recommendations for reducing workplace harassment.58Feldblum & Lipnic, supra note 31, at 31, 34.

Skills training on how to step in and help a colleague may also enhance an individual’s capacity and willingness to act, particularly if the employer regularly reinforces “stepping in” as a workplace norm.59Id. On the use of “stepping in” rather than “bystander intervention” to describe expectations for employees, see supra note 30. This is another area where legal accountability requirements can provide an assist, as Part III will address.

3. Why do people sexually harass others at work?

Understanding why someone might sexually harass a coworker or supervisee is a third individual-level factor worthy of consideration in developing interventions to support harassment prevention and response.

The characteristics of individuals who are likely to engage in sexual harassment at work have not been studied deeply, though there are insights to be gleaned from the research that does exist. One literature review has noted studies suggesting, for example, “that harassers lack social conscience, are naïve about heterosexual relationships, and engage in immature, irresponsible, manipulative and exploitative [behaviors].”60Paula McDonald, Workplace Sexual Harassment 30 Years on: A Review of the Literature, 14 Int’l J. Mgmt. Revs. 1, 8 (2012). Other studies have shown that “[h]arassers are also thought to over-infer women’s criticism and rejection, supporting the view that [harassment] is related to aggression rather than seduction.”61Id. Looking to social identity theory, scholars have also observed that men who strongly identify with other men in traditionally male-dominated workplaces may carry over an “us versus them” perception into hostile or harassing behaviors.62See, e.g., Anne Maass et al., Sexual Harassment Under Social Identity Threat: The Computer Harassment Paradigm, 85 J. Personality & Soc. Psychol. 853, 854, 866, 867 (2003) (applying social identity’s theory that “one’s self-concept derives in part from the status of the groups to which one belongs” and concluding that sexual harassment can be understood “as an attempt to protect or restore a threatened [masculine] gender identity” (citations omitted)). A growing literature on counterproductive workplace behavior of all types similarly finds a connection between individuals’ internal capacity to regulate their emotions and situational factors in the work environment.63See, e.g., Al Karim Samnani et al., Negative Affect and Counterproductive Workplace Behavior: The Moderating Role of Moral Disengagement and Gender, 119 J. Bus. Ethics 235 (2014). Samnani and co-authors explain that “perceived injustice/unfairness, desire for revenge, and abusive supervision are associated with CWB [counterproductive workplace behavior]. Since situational stressors typically evoke negative emotions among employees, the experience of negative emotions appears to be a significant precursor of CWB.” Id. (citations omitted). Anti-social personality disorders may present similar challenges. See Paul Moran, Antisocial Personality Disorder: An Epidemiological Perspective 54, 85, 98 (1999); Stuart D. Sidle, Personality Disorders and Dysfunctional Employee Behavior: How Can Managers Cope?, 25 Acad. Mgmt. Persp. 76, 76–77 (2011) (identifying how few managers are likely to feel equipped to respond effectively to employees whose antisocial behaviors create significant challenges for workplace relationships). As one study put the point, “some people may be predisposed to sexually harass and some social situations may be conducive to sexual harassment.”64Robert T. Hitlan et al., Antecedents of Gender Harassment: An Analysis of Person and Situation Factors, 61 Sex Roles 794, 795 (2009).

A more extensive literature on sexual assault identifies additional factors that may affect whether an individual will attempt to coerce a sexual interaction. One major study, commissioned by the U.S. Air Force and produced by the RAND Corporation, examined these oft-studied factors: “experience of child abuse, previous sexual behavior, interpersonal-skill deficits, gender-related attitudes, perceptions of peer behavior, and substance abuse.”65Sarah M. Greathouse et al., RAND Corp., A Review of the Literature on Sexual Assault Perpetrator Characteristics and Behaviors, at ix (2015). Considering these together, the RAND report concluded that “sexual assault perpetration is a complex behavior that is likely influenced by a combination of factors, including an individual’s developmental and family history, personality, and environmental and societal influences.”66Id. at xi.

As is surely apparent, workplace culture, law, and policy are unlikely to be influential unless they can affect a potential perpetrator’s sense of how others would regard the behavior.67See Paula M. Popovich & Michael A. Warren, The Role of Power in Sexual Harassment as a Counterproductive Behavior in Organizations, 20 Hum. Resource Mgmt. Rev. 45, 52 (2010) (observing that “while behavior is usually the most observable outcome, social influence also operates through the transmission of norms (i.e., behavioral expectations) and values”). Some research has found individuals with a predisposition toward engaging in harassing behavior were more likely to do so when “exposed to an authority figure who displayed such behavior himself.”68Hitlan et al., supra note 64, at 796. In the context of sexual assault, the RAND report concluded, too, that “individuals who perceive their peers as approving of sexual assault are more likely to commit sexual assault.”69Greathouse et al., supra note 65, at x. The report also noted a smaller number of studies that have found a “link between sexual assault perpetration and perceptions of peer pressure to engage in sexual activity.” Id.

Similarly, individuals’ perceptions of their own power and the power of others within an organization may affect whether they engage in unwanted sexual conduct and comments.70See Popovich & Warren, supra note 67, at 51–52 (describing how situational dynamics can affect individuals’ power in a workplace). In the workplace, this power may be understood in a variety of ways, including an individual’s formal authority or perceived authority to control the status and opportunities of another person, as well as individual’s physical or social power within a group.71Id. The counterpart to this point is the targeted person’s belief that the harassing individual has this authority and any challenge to that authority will prompt additional harm. See supra text accompanying notes 47–52. These perceptions of power are informed and influenced, in turn, by organizational and societal-level factors that I will turn to below and in Parts II and III.

B. Organizational Structure

Individuals work within an organizational structure, and the choices an employer makes about structuring the work environment can themselves influence the effectiveness of efforts to reduce harassment.72The National Academies of Sciences, Engineering, and Medicine has explored in depth the organizational challenges in academia. Paula A. Johnson et al., Sexual Harassment of Women: Climate, Culture, and Consequences in Academic Sciences, Engineering, and Medicine, Nat’l Acads. of Sci., Eng’g, & Med. 134, 134 (2018) (finding that “environments where people are isolated because of significant differences in power are more likely to foster and sustain sexual harassment” and that “[t]his power isolation occurs when there is a significant power imbalance—one party holds enough power and authority over the other that the former isolates the latter from being able to go to others for help without risking potentially serious retaliation”). A substantial analysis of risk factors within the federal judicial branch highlights many similar issues and concerns. See Fed. Judiciary Workplace Conduct Working Grp., Report of the Federal Judiciary Workplace Conduct Working Group to the Judicial Conference of the United States 3 (2018), https://www.uscourts.gov/sites/default/files/​workplace_conduct_working_​group_​final_report_0.pdf [https://perma.cc/6TR7-UYPX] (observing that power disparities and the obligation of confidentiality are factors that “increase the risk of misconduct or impose obstacles to addressing inappropriate behavior effectively”). Small hierarchical departments with little outside oversight, for example, can render employees more vulnerable to abusive behavior and less able to access protection if the manager disregards complaints or perpetrates the abuse.73Johnson et al., supra note 72, at 16. Structures that make an employee’s advancement dependent on a relationship with one individual exacerbate the risks associated with reporting abuse by that person. Physically isolated work areas also can enhance risk because there is less likely to be a coworker present to witness harassing behavior, much less intervene to stop it.74Id. at 43.

In addition to formal structure, special risks also have been identified in settings where individual employees depend on their employment for immigration status or basic subsistence and, as a result, are less likely to seek help.75Feldblum & Lipnic, supra note 31, at 26–27; see also Johnson et al., supra note 72, at 18 (noting that immigrants are one of the populations that faces “additional systems of oppression, domination, or discrimination”); Marissa Ditkowsky, #UsToo: The Disparate Impact of and Ineffective Response to Sexual Harassment of Low-Wage Workers, 26 UCLA Women’s L.J. 69 (2019) (addressing a variety of challenges faced by low-wage workers, including in service industries and farming). Workplaces with “high value” employees, young workforces, and cultural or language differences, among others, may pose outsized risks as well. Feldblum & Lipnic, supra note 31, at 26–27. Environments where alcohol or drug use is pervasive, either in the workplace (as is true for some restaurants and nightclubs) or during off-site parties, tend also to present greater risks that must be accounted for if employer efforts and legal interventions are to be effective in reducing the incidence of harassment.76Feldblum & Lipnic, supra note 31, at 88.

C. The External Environment

In addition to the individuals and organizational structures that define a workplace, external influences—both societal and governmental—can substantially affect whether harassment prevention and response efforts gain traction inside a workplace.77Although societal and governmental influences are often treated as separate sites for analysis and intervention, I consider them together here as my focus is on their collective effect on employee interactions. As Iyiola Solanke has observed, “the language that is used to communicate at an interpersonal level draws upon assumptions and ‘common sense’ provided by the surrounding culture.” Solanke, supra note 27, at 8. Critical social psychologists have recognized that “there is no escape from this.” Id. (citing Brendan Gough & Majella McFadden, Critical Social Psychology: An Introduction 13 (2001)). As with individual-level factors, an employer has control over how it responds to external influences but, ordinarily, not over the influences themselves. Still, just as employers must consider the market in which they operate if they are to succeed, so too must they take account of the conditions outside the workplace that affect their employees.

Shifts in popular culture, large-scale social commitments, and the legal/regulatory environment all have the potential to affect employee expectations of their work environments and, by extension, the obstacles or openings for prevention and response efforts.78The EEOC Select Task Force Report observes that “events and coarse social discourse that happen outside the workplace may make harassment inside a workplace more likely or perceived as more acceptable.” See Feldblum & Lipnic, supra note 31, at 27; see also Jean M. Twenge & Stacy M. Campbell, Generational Differences in Psychological Traits and Their Impact on the Workplace, 23 J. Managerial Psychol. 862, 873 (2008) (observing that psychological differences across age groups can have a significant effect on employees’ interactions and expectations from their workplace). Interactions and comments that were considered by many as acceptable or funny in popular entertainment, for example, come across today as jarringly out of step.79Rob Owen, Sexual Harassment Has a Long History as a Comedic Punchline on TV, Pitt. Post-Gazette (Nov. 30, 2017, 11:00 AM), https://www.post-gazette.com/ae/tv-radio/2017/11/30/Sexual-harassment-has-a-long-history-as-a-comedic-punchline-on-TV/stories/201711300085 [https://perma.cc/2CN5-PCBH] (describing a “turning point for when such jokes [about sexual harassment] went from funny to cringe worthy”). There is a legal analogue to changes in generally acceptable argumentation regarding the exclusion of women from opportunities open to men. Compare, for example, Justice Bradley’s concurrence supporting Illinois’s refusal to allow Myra Bradwell to practice law, Bradwell v. Illinois, 83 U.S. 130, 141, 142 (1872) (Bradley, J., concurring) (“The paramount destiny and mission of woman are to fulfil the noble and benign offices of wife and mother.”), with Justice Scalia’s argument for why the Virginia Military Institute should not be required to admit women, United States v. Virginia, 518 U.S. 515, 580, 566 (1996) (Scalia, J., dissenting) (relying in part on educational diversity as a rationale for upholding VMI’s exclusion of women and describing a VMI committee’s study as “utterly refut[ing] the claim that VMI has elected to maintain its all-male student-body composition for some misogynistic reason”). Cf. Suzanne B. Goldberg, Constitutional Tipping Points: Civil Rights, Social Change, and Fact-Based Adjudication, 106 Colum. L. Rev. 1955 (2006) (reviewing the ways in which courts change their descriptions of social groups over time). These kinds of changes in popular culture both reflect and inform what happens in real workplaces.80See generally John Hassard & Ruth Holliday, Introduction to Organization-Representation: Work and Organization in Popular Culture 1 (John Hassard & Ruth Holliday eds., 1998) (noting that media portrayals of organizations “remark on and inform current organization theory and practice”). In similar ways, societal expectations about interpersonal behavior also may affect whether complaints or concerns will be taken seriously by employers.81Cf. Lynne Andersson & Christine M. Pearson, Tit for Tat? The Spiraling Effect of Incivility in the Workplace, 24 Acad. Mgmt. Rev. 452, 468 (1999) (describing how changing work patterns related to technology, globalization, and other factors contribute to the spread of incivility in the workplace); Carla S. Fugas et al., The “Is” and the “Ought”: How Do Perceived Social Norms Influence Safety Behaviors at Work?, 16 J. Occupational Health Psychol. 67, 67 (2011) (explaining that social norms “have an important contextual influence on attitudes and health-related behaviors”).

Cultural inflection points also seep into, or sometimes abruptly unsettle, dominant patterns of behavior. In the 1960s and 70s, for example, the civil rights and women’s movements reached into the workplace and delegitimized, or at least called into question, some longstanding forms of differential treatment based on race and sex.82See generally Catharine A. MacKinnon, Sexual Harassment of Working Women (1979) (analyzing cases involving sexual harassment and setting out a theory for why sexual harassment should be treated as legally actionable sex discrimination rather than as misplaced sexuality at work); Karen Lindsey, Sexual Harassment on the Job and How to Stop It, Ms., Nov. 1977 (explaining and condemning sexual harassment at work). Cultural change regarding the rights and dignity of lesbians and gay men, which began decades ago but entered the mainstream through the marriage equality movement, similarly caused individuals and employers to reconsider the acceptability of anti-gay hostility in the workplace. Justin McCarthy, Slim Majority in U.S. Favors New LGBT Civil Rights Laws, Gallup (June 13, 2019), https://news.gallup.com/poll/258176/slim-majority-favors-new-lgbt-civil-rights-laws.aspx?utm_source=alert&utm_medium=email&utm_content=morelink&utm_campaign=syndication [https://perma.cc/W7LN-7GG9] (finding that nearly all (93%) of Americans polled by Gallup in 2019 “are highly supportive of ensuring LGBT individuals have equal employment opportunities”). See generally Suzanne B. Goldberg, Sticky Intuitions and the Future of Sexual Orientation Discrimination, 57 UCLA L. Rev. 1375, 1407–15 (2010) (considering strategies to challenge negative intuitions about lesbians and gay men, including those that support employment discrimination). More recently, increased attention to the lives and dignity of transgender individuals has prompted shifts that have carried over to many, though certainly not all, workplaces. See Bostock v. Clayton Cty., 140 S. Ct. 1731, 1741 (2020) (finding unlawful sex discrimination when “an employer . . . fires a transgender person who was identified as a male at birth but who now identifies as a female” because “the employer intentionally penalizes a person identified as male at birth for traits or actions that it tolerates in an employee identified as female at birth”); see also Jennifer C. Pizer et al., Evidence of Persistent and Pervasive Workplace Discrimination Against LGBT People: The Need for Federal Legislation Prohibiting Discrimination and Providing for Equal Employment Benefits, 45 Loy. L.A. L. Rev. 715 (2012); Emily K. Crawford, Comment, America’s Finally Beginning to Talk About It—Transgender Individuals’ Rights in the Workplace, 18 Duq. Bus. L. Rev. 45 (2016). By contrast, while passage of the Americans with Disabilities Act marked an important shift in the rights of people with disabilities, there has been less momentum toward change in social dynamics in and outside of workplaces. See, e.g., Samuel R. Bagenstos, The ADA Amendments Act and the Projects of the American Disability Rights Movement, 23 U. D.C. L. Rev. 139, 139 (2020) (describing the American disability rights movement’s success in obtaining the passage of the Americans with Disabilities Act as “outpac[ing] the changes in social attitudes toward people with disabilities” and resulting in backlash). More recently, the #MeToo movement’s transformative effect on discourse about sexual harassment has prompted sustained national conversation and debate,83Significant attention has been paid to sexual harassment in earlier times, though less pervasively than what the #MeToo movement has achieved. Violent attacks on women at the 1993 Tailhook Annual Symposium on sea-based aviation, for example, led to national news coverage and a major investigation and condemnation of the behavior by the Department of the Navy. See, e.g., Norman Kempster, What Really Happened at Tailhook Convention: Scandal: The Pentagon Report Graphically Describes How Fraternity-Style Hi-Jinks Turned into Hall of Horrors, L.A. Times (Apr. 24, 1993, 12:00 AM), https://www.latimes.com/archives/la-xpm-1993-04-24-mn-26672-story.html [https://perma.cc/L9YZ-HJXG] (describing extensive assaults and abusive behavior by servicemembers at the Tailhook Convention); Michael Winerip, Revisiting the Military’s Tailhook Scandal, N.Y. Times (May 13, 2013), https://www.nytimes.com/2013/05/13/booming/revisiting-the-militarys-tailhook-scandal-video.html [https://perma.cc/9PHG-CJV6] (reporting that 83 women and 7 men “were later found to have been assaulted during the [incident]”). as has the Black Lives Matter movement with respect to anti-Black racism.84Nate Cohn & Kevin Quealy, How Public Opinion Has Moved on Black Lives Matter, N.Y. Times (June 10, 2020), https://www.nytimes.com/​interactive/​2020/​06/10/upshot/black-lives-matter-attitudes.html [https://perma.cc/T3GM-KNJZ] (describing significant increase in public support for the Black Lives Matter movement). As Catharine MacKinnon observed regarding #MeToo, “[t]his mass mobilization against sexual abuse, through an unprecedented wave of speaking out in conventional and social media, is eroding the two biggest barriers to ending sexual harassment in law and in life: the disbelief and trivializing dehumanization of its victims.”85Catharine A. MacKinnon, Opinion, #MeToo Has Done What the Law Could Not, N.Y. Times (Feb. 4, 2018), https://www.nytimes.com/2018/02/04/opinion/metoo-law-legal-system.html [https://perma.cc/RL44-5AR5]. Whatever one thinks about the changes underway, there can be no question that these social movements are influencing the expectations and interactions of both employers and employees.

Legal and regulatory factors are another potentially powerful source of external influence on organizations and the individuals within them. Part III will address workplace regulations in depth, as they, unlike societal norms and cultural change, require direct engagement by the employer. Other legal changes outside of employment, especially those related to civil rights such as marriage equality for same-sex couples and rejection of race discrimination in education and marriage, are more akin to the cultural influences just described, as they permeate society in ways that reverberate in workplaces as well.86See, e.g., Mark Z. Barabak, Gays May Have the Fastest of All Civil Rights Movements, L.A. Times (May 20, 2012, 12:00 AM), https://www.latimes.com/style/la-xpm-2012-may-20-la-na-gay-rights-movement-20120521-story.html (reporting on shifts in attitudes and related legal changes in access to marriage for same-sex and interracial couples); cf. Lani Guinier & Gerald Torres, Changing the Wind: Notes Toward a Demosprudence of Law and Social Movements, 123 Yale L.J. 2740 (2014) (considering the interaction between legal and social change).

In short, external influences on workplaces add yet another layer of complexity as we consider the power and limits of law in addressing harassment and other discrimination.

II. Organizational Culture—The Basics

The fact that there are so many individual and external factors over which an employer has little control raises the stakes for organizational culture as a tool for harassment prevention and response, and likewise for legal interventions to support effective use of organizational culture in this regard. Yet much of what we think of as organizational culture has little apparent connection with harassment policies and processes. This Part addresses the overarching function of workplace culture and some of the constraints on employers in developing that culture. The next Part will turn in depth to the choices that employers make about legal accountability and the effect of those choices on organizational values and employee interactions.

Most often, organizational culture is associated, from a business perspective, with an organization’s ability to achieve its goals. As a Harvard Business Review analysis put the point, “[w]hen aligned with strategy and leadership, a strong culture drives positive organizational outcomes.”87Boris Groysberg et al., The Leader’s Guide to Corporate Culture, Harv. Bus. Rev. (Jan. 2018), https://hbr.org/2018/01/the-culture-factor#:~:text=The%20Leader’s​%20Guide%20to%20Corporate%20Culture. In addition to the academic literature discussed below, an extensive popular business commentary also defines and analyzes the elements of corporate and other organizational cultures. See, e.g., John Coleman, Six Components of a Great Corporate Culture, Harv. Bus. Rev. (May 6, 2013), https://hbr.org/2013/05/six-components-of-culture (surveying mission statements, consulting reports, and other sources to theorize foundational elements of positive business culture). For discussion of compliance culture in corporate settings, see infra notes 93–95 and accompanying text. Whether those desired outcomes are for profit or other ends, a culture that encourages productivity, innovation, regulatory compliance, quality control, and other goals tied to the relevant market is likely to be considered successful.88To the extent sexual harassment impedes employees in their work, efforts to prevent harassment are arguably as market-oriented as other efforts to retain quality employees and support their productivity. See, e.g., David N. Laband & Bernard F. Lentz, The Effects of Sexual Harassment on Job Satisfaction, Earnings, and Turnover Among Female Lawyers, 51 Indus. & Lab. Rel. Rev. 594, 606 (finding “evidence that female lawyers who had experienced or observed sexual harassment in the legal workplace reported lower overall job satisfaction and a greater intention to voluntarily exit their current employment context than did other female lawyers”); Chelsea R. Willness et al., A Meta-Analysis of the Antecedents and Consequences of Workplace Sexual Harassment, 60 Personnel Psychol. 127, 151 (2007) (describing studies linking sexual harassment to reduced productivity and increased turnover).

Organizations take a variety of approaches to developing these shared “beliefs, values, behavior patterns, and understanding” that comprise workplace culture,89Linstead, supra note 4, at 10930; see also J. Yo-Jud Cheng & Boris Groysberg, How Corporate Cultures Differ Around the World, Harv. Bus. Rev. (Jan. 8, 2020), https://hbr.org/2020/01/how-corporate-cultures-differ-around-the-world (describing organizational culture as “the shared, pervasive, enduring, and implicit behaviors and norms that permeate an organization (rather than individual employees’ own culture styles)”). and popular business literature is flooded with guidance on how to develop a culture that works.90See, e.g., Alan Kohll, How to Build a Positive Company Culture, Forbes (Aug. 14, 2018, 1:44 PM), https://www.forbes.com/sites/alankohll/2018/08/14/how-to-build-a-positive-company-culture/#16bf8c3049b5 [https://perma.cc/GN3E-N7M5] (noting that companies can build a more positive culture with attention to employee wellness, goal-setting, and listening to employees); Monica Zent, The 8 Essential Steps to Building a Winning Company Culture, Entrepreneur (Nov. 7, 2014), https://www.entrepreneur.com/article/239475 [https://perma.cc/SXU7-BLXD] (providing strategies to improve company culture that include reflecting on the organization’s core values, opening up communication, and having fun). Whether through banners, wellness resources, and rewards programs, or mission statements, newsletters, and social gatherings, or the absence of any deliberate culture-creation efforts, employees come to understand the organization’s values, modes of working, and types of activities and behaviors that are well-regarded, tolerated, or rejected.91Coleman, supra note 87.

A vast literature examines the scope and constitutive elements of organizational culture in employment settings, going back many decades and cutting across a variety of disciplines.92The word “culture” was first used in connection with work organization in the mid-20th century. See Linstead, supra note 4, at 10931 (referencing Elliott Jaques, The Changing Culture of a Factory 251 (1952), which studied organizational management in a UK company and coined the term “culture” as an important element to its success). Organizational psychologists, organizational development specialists, and scholars in myriad other disciplines have since added their insights to the field. Id. at 10930. For illustrations of various disciplinary approaches to the concept of corporate culture, see, for example, Charles A. O’Reilly III et al., The Promise and Problems of Organizational Culture: CEO Personality, Culture and Firm Performance, 39 Group & Org. Mgmt. 595, 599–601 (2014); Greg Urban, Corporations in the Flow of Culture, 39 Seattle U. L. Rev. 321, 322 (2016). To the extent law has been linked to organizational culture, it is mostly through the lens of compliance culture, where the aim is to generate a common understanding “about the importance or legitimacy of legal compliance vis-à-vis other pressures and goals.”93Donald C. Langevoort, Cultures of Compliance, 54 Am. Crim. L. Rev. 933, 944 (2017) (citing Karl E. Weick, Sense Making in Organizations (1995)). This is often accompanied by considerations of “tone at the top,”94See, e.g., Harvey L. Pitt & Karl A. Groskaufmanis, Minimizing Corporate Civil and Criminal Liability: A Second Look at Corporate Codes of Conduct, 78 Geo. L.J. 1559, 1634 (1990) (discussing the role of corporate codes in “set[ting] a serious tone from the top” and “communicating an intent to stay well within the law”). and concerns about financial fraud and other white-collar crime.95See generally Langevoort, supra note 93, at 933–35 (discussing the emergence of a “culture of compliance” in connection with white-collar crime). Langevoort dates an increase in academic interest in compliance culture to the 1990s and notes that by the early 2000s, “‘tone at the top’ and other invocations of ethical culture by regulators were becoming more common.” Id. at 942. In 2004, the federal Organizational Sentencing Guidelines were revised to provide that firms should “promote an organizational culture that encourages ethical conduct and a commitment to compliance with the law.” U.S. Sentencing Guidelines Manual § 8B2.1(A)–(B) (U.S. Sentencing Comm’n 2004); cf. Jennifer Arlen, The Failure of the Organizational Sentencing Guidelines, 66 U. Miami L. Rev. 321, 326, 328 (2012) (recognizing the Guidelines’ positive effect on corporate conduct but also expressing skepticism about their effectiveness). Yet relatively little attention has been given to the relationship between workplace culture, discrimination, and harassment, until very recently.96See, e.g., Feldblum & Lipnic, supra note 31, at v (observing that “[w]orkplace culture has the greatest impact on allowing harassment to flourish, or conversely, in preventing harassment” and “[t]he importance of leadership cannot be overstated—effective harassment prevention efforts, and a workplace culture in which harassment is not tolerated, must start with and involve the highest level of management of the company”). For recent cases addressing the link between workplace culture and harassment or other discrimination, see, for example, Brown v. Nucor Corp., 785 F.3d 895, 912 (4th Cir. 2015), discussing corporate culture in the context of a class-decertification motion in a race-discrimination suit regarding failure to promote, observing that “[i]t strains the intellect to posit an equitable promotions system set against that cultural backdrop [of a racially hostile environment], particularly in light of the other evidence presented”; Davis v. Packer Eng’g, Inc., No. 11-CV-07923, 2018 WL 1784131, at *7 (N.D. Ill. Apr. 12, 2018), denying post-trial motion to reverse hostile-environment finding in Title VII sexual harassment case and noting testimony “about a company-wide practice of condoning the workplace culture and failing to respond to complaints.”

A starting premise of this Article is that workplace culture can be a significant determinant of whether harassment and other discriminatory behaviors are likely to thrive. More specifically, employees in a collegial workplace will be more likely to identify and address harassment as outlier and unacceptable behavior than those in a hostile or contentious workplace where abusive modes of interaction are commonplace.97This picks up on the idea in corporate compliance literature that an environment that stresses excellence, in addition to legal compliance, may be more likely to achieve both. See infra note 113; cf. McDonald, supra note 60, at 12 (identifying weaknesses in a model that relies on legal compliance based on the interrelatedness of other workplace phenomena). At the same time, too strong an insistence on collegiality may pose concerns, both for the organization and individuals within it. See, e.g., Devon W. Carbado & Mitu Gulati, Working Identity, 85 Cornell L. Rev. 1259, 1262 (2000) (analyzing how “outsider groups” face pressure to conform so that they are not perceived as disruptive and “to behave in particular ways to avoid discrimination”); Lisa Interligi, Compliance Culture: A Conceptual Framework, 16 J. Mgmt. & Org. 235, 246 (2010) (discussing research showing that “[o]rganizations with homogenous or strong cultures can be susceptible to ‘group think’, a phenomenon in which critical thinking is also compromised and diverse opinions are discouraged” (citation omitted)). A burgeoning workplace-civility literature bolsters this view, indicating that greater attention to respectful relations among employees is likely to reduce tolerance for harassment.98See, e.g., Michael P. Leiter et al., The Impact of Civility Interventions on Employee Social Behavior, Distress, and Attitudes, 96 J. Applied Psychol. 1258, 1272 (2011) (finding that improving civility in the workplace leads to positive social behavior and resolution of workplace environment concerns). See infra note 107 and accompanying text for discussion of possible employee objections to collegiality-enhancing efforts.

Closely related is an organization’s commitment to diversity, equity, and inclusion (DEI) as a set of values and practices that inform the ways in which employees interact. In addition to its broader benefits, a DEI commitment has two crossover points related specifically to workplace harassment. First, vulnerable populations, whatever those are within the organization, need to know that the employer will take their concerns seriously, especially given individual-level barriers to reporting.99See supra notes 31–52 and accompanying text (discussing individual-level barriers to reporting). And second, without express attention, individuals in the dominant group may miss issues of importance to other employees because they are not affected directly.100Cf. Barbara J. Flagg, Was Blind but Now I See: White Race Consciousness & the Law 1 (1998) (“The most striking characteristic of whites’ consciousness of whiteness is that most of the time we don’t have any.”).

Yet a workplace culture that encourages collegiality and mutual respect among employees at all levels of the enterprise is easier to promise than to create.101Perhaps needless to say, even strong leadership from the top will not result in identically styled interactions across all levels, units, and locations within an organization. A comprehensive evaluation would look to how the elements here take hold within each subunit of a firm. See generally Erin Mayer, When Culture Doesn’t Translate, Harv. Bus. Rev. (Oct. 2015), https://hbr.org/2015/10/when-culture-doesnt-translate. Many employers express an interest in having a strong, cohesive, and equitable culture, for example, but the gap between expression and reality can be a large one.102See Kate Conger et al., Google Faces Backlash over Handling of Sexual Harassment, N.Y. Times (Oct. 31, 2018), https://www.nytimes.com/​2018/10/31/technology​/google-​sexual-harassment-walkout.html [https://perma.cc/63AZ-EN5L] (noting employee reactions to Google’s efforts to promote equity in the workplace, including one manager’s comment that “Google’s famous for its culture. But in reality we’re not even meeting the basics of respect, justice and fairness for every single person here”). Consider, for example, the major corporations that recently have faced serious sexual harassment accusations notwithstanding the slogans (e.g. Alphabet’s “Do the right thing” motto, for example),103This slogan replaced Google’s “don’t be evil” motto. Kate Conger, Google Removes ‘Don’t Be Evil’ Clause from Its Code of Conduct, Gizmodo (May 18, 2018, 5:31 PM), https://gizmodo.com/google-removes-nearly-all-mentions-of-dont-be-evil-from-1826153393. Google’s parent company recently reached a large-scale settlement of shareholder actions based on high-profile sexual harassment allegations at Google. See Daisuke Wakabayashi, Alphabet Settles Shareholder Suits over Sexual Harassment Claims, N.Y. Times (Sept. 25, 2020), https://www.nytimes.com/​2020/09/25/technology/google-sexual-harassment-lawsuit-settlement.html [https://perma.cc/FW5Y-YS7E] (describing alleged misconduct by senior leaders at the company). I served as an expert for the plaintiffs and helped develop the non-monetary provisions of the settlement agreement in this case. employee handbooks, and other culture and equity-promoting tools that declare an intolerance for harassing behavior.104See, e.g., Jaclyn Jaeger, Opinion, Lessons from Wynn Resorts’ Sexual Harassment Scandal, Compliance Wk. (Aug. 28, 2018, 6:00 AM), https://www.complianceweek.com/​opinion/lessons-from-wynn-resorts-sexual-harassment-scandal/2163.article [https://perma.cc/AE6Z-THWC] (describing the sexual harassment scandal at Wynn Resorts where, despite having a policy on preventing harassment in the company’s Code of Business Conduct and Ethics, the board ignored complaints of sexual harassment and assault “and allowed the behavior to continue for decades as they turned a blind eye”). The same is true for the policies and procedures that may demonstrate an employer’s interest in legal compliance but not in creating a culture that rejects harassment or other abusive behaviors, as discussed in more depth below.105See infra Section III.A.

Even when employers genuinely seek to create a workplace culture that fosters mutual respect and collegiality, the task itself is challenging. The sheer diversity of personalities and skill sets in any workplace, especially among supervisors and managers, can pose significant difficulties for achieving and maintaining an organizational culture that has a positive influence on the way people interact with each other.106Feldblum & Lipnic, supra note 31, at ii, 26–27; cf. Ron Carucci, 4 Ways Lying Becomes the Norm at a Company, Harv. Bus. Rev. (Feb. 18, 2019), https://hbr.org/2019/02/4-ways-lying-becomes-the-norm-at-a-company (noting that systemic factors “might influence whether or not people in organizations distort or withhold truth from one another”). Efforts to enhance collegiality and equity as part of workplace culture may also face objections from employees who argue that their free expression, creativity, or capacity to be themselves at work is impaired.107See, e.g., Lyle v. Warner Bros. Television Prods., 132 P.3d 211, 225 (Cal. 2006) (rejecting discrimination and harassment claims regarding discussions that took place among writers for the television show Friends and observing “[t]hat the writers commonly engaged in discussions of personal sexual experiences and preferences and used physical gesturing while brainstorming and generating script ideas for this particular show was neither surprising nor unreasonable from a creative standpoint”). For critique of Lyle, see, for example, Sarah Pahnke Reisert, Let’s Talk About Sex Baby: Lyle v. Warner Brothers Television Productions and the California Court of Appeal’s Creative Necessity Defense to Hostile Work Environment Sexual Harassment, 15 Am. U. J. Gender Soc. Pol’y & L. 111, 114 (2006), arguing that “the creative necessity defense is both inconsistent with the legislative intent of federal and state anti-discrimination laws and unnecessary”; Catharine A. MacKinnon, Directions in Sexual Harassment Law, 31 Nova L. Rev. 225, 236 (2007), stating “[g]ranted, the kind of workplace it is, is part of the totality of the circumstances, but whether that should permit it to be unequal is the question.” The question whether antidiscrimination law infringes free expression rights or values has been explored extensively elsewhere and is beyond the scope here, other than to recognize the possible tension with some workplace-culture initiatives aimed at preventing harassment. For discussion of the underlying legal questions, see, for example, Deborah Epstein, Can a ‘Dumb Ass Woman’ Achieve Equality in the Workplace? Running the Gauntlet of Hostile Environment Harassing Speech, 84 Geo. L.J. 399, 400 (1996); Eugene Volokh, How Harassment Law Restricts Free Speech, 47 Rutgers L. Rev. 563, 563–64 (1995). At the other end, some employees may urge the employer to go further, objecting to any willingness to tolerate communications or behavior that they find offensive.108Cf. Jason Wingard, Employee Activism Is the New Normal. So Why Is Amazon Leadership Freaking out?, Forbes (Jan. 10, 2020, 8:16 AM), https://www.forbes.com/sites/jasonwingard/2020/01/10/employee-activism-is-the-new-normal-so-why-is-amazon-leadership-freaking-out/?sh=7ca20e2027f1 [https://perma.cc/PZ7Z-PC8X] (describing research by a global public relations firm finding that nearly “4 in 10 employees (38%) say they have ‘spoken up to support or criticize their employers’ actions over a controversial issue that affects society’”). Employee advocacy that focuses on internal workplace culture can sometimes be a mechanism for penalizing individuals whose identity or self-expression is in the minority. See generally Carbado & Gulati, supra note 97.

These general challenges are amplified at various points in an organization’s lifespan. In a start-up phase, for example, all attention may be devoted to bringing a service or product to market, and those leading the enterprise may not have expertise, much less staff dedicated to working on these issues.109For discussion of the scant response to sexual harassment complaints in start-up companies, see Jennifer S. Fan, Innovating Inclusion: The Impact of Women on Private Company Boards, 46 Fla. St. U. L. Rev. 345, 380 (2019). Notably, the prevalence of sexual harassment complaints has led some start-ups to reconsider “their loose, entrepreneurial cultures [and] now consult[] with human resource experts.” Maureen Minehan, Startup Companies: Waking up to HR’s Importance?, HR Wire (Apr. 24, 2000). Indeed, as numerous accounts of “bro culture” at start-ups have indicated, issues associated with organizational culture and how coworkers treat each other may not be on the radar at all.110Sarah Benstead, ‘Bro Culture’ and Why It’s an Issue for Startups, Breathe (Oct. 18, 2018), https://www.breathehr.com/en-gb/blog/topic/company-culture/bro-culture-and-why-its-an-issue-for-startups[https://perma.cc/MZ58-GERS]; Dan Lyons, Opinion, Jerks and the Start-Ups They Ruin, N.Y. Times (Apr. 1, 2017), https://www.nytimes.com/2017/​04/01/opinion/sunday/jerks-and-the-start-ups-they-ruin.html?_r=0 [https://perma.cc/LLV4-2AR2] (describing “a [bro] culture built on reckless spending and excessive partying, where bad behavior is not just tolerated but even encouraged”). For organizations that are struggling to survive or are severely resource-constrained even during a period of growth, a focus on workplace culture may sound like a diversion from core issues.111Benstead, supra note 110. Even for stable or growing organizations, significant attention to coworker interactions may seem to detract from the mission or be seen as risky because of demonstrated or anticipated resistance.112Benstead, supra note 110.

Yet as is true for most human enterprises, workplaces can be rife with conflict, and cultural expectations regarding interpersonal behavior and approaches to conflict resolution will emerge, whether or not they are stated expressly by the organization’s leadership. The next Part will turn to the ways in which these expectations and approaches are themselves cultural choices.

III. The Culture-Shaping Implications of Legal Accountability Choices

Choices that employers make about how they will satisfy legal requirements related to harassment prevention and response are not often thought of as part of workplace culture. Instead, “checking the box” is an apt descriptor for what many employers do, with counsel and human resources staff focused on ensuring that compliance requirements have been met.113Lauren Edelman has written extensively on this issue, including on the ways in which human resources departments have taken an “expert” role within organizations and exerted substantial influence on how governing law and regulation are interpreted and applied. See, e.g., Lauren B. Edelman, Working Law: Courts, Corporations, and Symbolic Civil Rights 78 (2016) (arguing that human resources professionals are central in “making the legal environment salient to organizations and various organizational actors and procedures”); Lauren B. Edelman & Shauhin A. Talesh, To Comply or Not to Comply—That Isn’t the Question: How Organizations Construct the Meaning of Compliance, in Explaining Compliance: Business Responses to Regulation 103 (Christine Parker & Vibkek Lehmann Nielson eds., 2009) (showing how employers’ internal practices in the corporate context influence judicial interpretations of governing law). Too rigid and technical a focus on legal compliance at the expense of attention to other values can be costly for firms. See, e.g., Interligi, supra note 97, at 246 (noting that “employees in normative compliance cultures are more likely to comply with organizational expectations if they expect that their colleagues also would comply, and if they believe compliance behaviour is expected of them” and that corporate compliance involves more than “just meeting regulatory demands”); Donald C. Langevoort, Monitoring: The Behavioral Economics of Corporate Compliance with Law, 2002 Colum. Bus. L. Rev. 71, 73 (2002) (examining compliance strategies outside the law, which might take “a more ethics or integrity-based effort at building compliance cultures within firms”). As this literature suggests, an important question is whether law is treated as the primary source of the leadership’s fiduciary obligations or as one of many, with markets and broader social contexts serving as additional sources of authority. Notably, if legal liability is treated as the sole touchstone for determining acceptable workplace behavior, objections to harassment by coworkers or even employees may be recharacterized legalistically by those doing the harassing as interfering with their freedom of speech as discussed supra note 107. Thanks to Martha Chamallas for this observation.

But this approach misses the mark, both theoretically and practically. When we remove the “human resources/legal” overlay, we can see that choices about harassment policy, reporting procedures, communication and employee trainings, and even decisions in individual cases are fundamentally about culture—that is, they are about guiding individuals and teams to support the organization’s mission. As with other aspects of workplace culture, these choices can impact productivity, effectiveness, reputation, and, ultimately, the organization’s stability and survival.114See Jordan Ellis & Kathryn Brown, 5 Strategies for In-House Counsel to Enhance Their Organization’s Workplace Culture in the Era of #MeToo, ACC Docket (July/Aug. 2019) at 46, 48 (stating that “as recent headlines show, mere allegations of a toxic work environment can tarnish an organization’s brand, driving down morale, productivity, and retention, not to mention share prices”). Put another way, these choices communicate organizational values to employees, whether intentionally or not, and will influence the ways that employees interact with each other, with spillover effects on the overall culture of the workplace, and vice versa.115See Steven H. Appelbaum et al., Positive and Negative Deviant Workplace Behaviors: Causes, Impacts, and Solutions, 7 Corp. Governance 586, 595–96 (2007) (discussing the role of organizational factors, such as fair procedures and respectful treatment of employees, in curbing misconduct by employees).

Importantly, legal-accountability culture is distinct from compliance culture, though the two may seem similar because both have implications for an organization’s exposure to liability. The idea of a compliance-oriented culture is, in essence, to deter cheating or other unlawful conduct in service of work-related aims.116See supra note 95 and accompanying text. For discussion of the law’s influence on other aspects of corporate compliance, see generally Langevoort, supra note 93, at 935, surveying research in law, economics, and other social sciences. These issues arise most often when employees might be tempted or even incentivized to violate policy or law as a means of achieving or exceeding performance-related goals that are tied to their success.117See supra note 95 and accompanying text. For discussion of the law’s influence on other aspects of corporate compliance, see generally Langevoort, supra note 93, at 935, surveying research in law, economics, and other social sciences. By contrast, the abuse of power involved in harassment is not typically in service of individual performance goals118One exception to this proposition would be a situation where employees are competing with each other and one uses harassment as a strategy to diminish the effectiveness of the other. More typically, though, the inclination to abuse power by harassing a coworker is associated with a sense of impunity that may itself be a byproduct of workplace success. See infra note 186 and accompanying text. or organizational success.119See supra notes 102–04.

The discussion that follows will apply this theoretical foundation and look closely at how harassment policy, procedures, and operations embody choices about workplace culture. The nine decision points discussed here include: policy design, communication, reporting options, employee training and “local” knowledge, privacy and retaliation protections, capacity-building for staff who handle complaints, alternate forms of conflict resolution, tailored sanctions, and transparency about how the process works. This will further set the foundation for Part IV’s discussion of the relationship between these choices and existing sources of law, and Part V’s consideration of whether law can do more to affect these culture-shaping choices.

A. Policy Drafting and Review

The starting point for most employer decisions about harassment and discrimination is the organizational policy. In a sense, this is a straightforward choice in that most employers focus on tracking legal prohibitions against discrimination,120A pair of 1998 Supreme Court sexual harassment rulings made clear that an employer risks exposing itself to liability if it lacks a policy and process for addressing complaints. See Burlington Indus., Inc. v. Ellerth, 524 U.S. 742, 765 (1998) (holding that an employer may offer, as an affirmative defense to liability for sexual harassment, that it has an antiharassment policy and procedure for receiving complaints and that the employee “unreasonably failed to take advantage of any preventive or corrective opportunities provided by the employer or to avoid harm otherwise”); Faragher v. City of Boca Raton, 524 U.S. 775, 807 (1998) (same). though of course they are free to go beyond in ways that will be discussed in Part IV.

But from an organizational culture standpoint, two other choices are significant: how the policy will be written and how it will be reviewed and revised over time.

While policy writing is frequently seen as a lawyer’s or human resources task to ensure that compliance requirements are met,121See supra note 113. the policy’s on-the-ground meaning—its influence on workplace culture—derives from the understanding and experience of the employees who are its “end users.”122See Interligi, supra note 97, at 237–38 (explaining that “compliance should be addressed at two main interfaces: at the interface between the organization and its environment where stakeholder expectations attempt to influence the behaviour of the organization; and at the interface between the organization and its employees where organizations aim to influence and shape the behaviour of employees”); see also Feldblum & Lipnic, supra note 31, at 80 (highlighting the importance to having “[a]n easy-to-understand description of prohibited conduct, including examples” in an anti-harassment policy). If the primary driver of the writing process is compliance, the language choices are likely to sound legalistic by contrast to communications that are designed to build organizational culture. There also may be less incentive to make the policy user-friendly; compliance needs can be satisfied without the extra work that goes into having the policy be a vehicle for advancing the organization’s values.

An approach that recognizes a harassment policy’s cultural relevance would, for example, incorporate into the drafting process questions about clarity and accessibility. Some examples:

Are the prohibited acts described in a way that is easy for employees to understand?

Are there examples to help employees draw lines between permissible and impermissible conduct?

Are the procedures for filing and processing a complaint made clear to potential users of the system and those implementing the policy?

By contrast, treating the policy as “off to the side” and designed primarily to ensure compliance rather than as an integral element of organizational culture will lead predictably to a policy that is less accessible to employees, signaling in turn the diminished value they should give to it.

A related set of choices concerns the policy review process. The default, compliance-oriented approach would lead an organization to review a policy only in response to changes in the law. Treating a policy as a tool for building organizational culture would suggest a more participatory approach. Among the questions an employer might consider:

Is the policy subject to regular review and revision to ensure that it remains vital in codifying or reinforcing organizational culture?

Do people who have been part of a policy enforcement process, whether as complainants, respondents, witnesses, or advisers, have a meaningful opportunity to offer input?

Do other employees have an opportunity to pose questions and provide comments?

While these forms of engagement are not required by law, skipping them is a lost opportunity for culture-building and potentially a signal that the employer does not see its policy as embodying commitments at the heart of the organization’s relationship to its employees.123See, e.g., Deloitte, Corporate Culture: The Second Ingredient in a World-Class Ethics & Compliance Program 2–3 (2015), https://www2.deloitte.com/content/dam/Deloitte/us/Documents/risk/us-aers-corporate-culture-112514.pdf [https://perma.cc/LB82-3F3M] (asserting that “without a culture of integrity, organizations are likely to view their ethics and compliance programs as a set of check-the-box activities” outside the heart of their business practice); PwC, PwC State of Compliance Study 2016: Laying a Strategic Foundation for Strong Compliance Risk Management 2 (2016), https://www.pwc.se/sv/pdf-reports/state-of-compliance-study-2016.pdf [https://perma.cc/9T4D-Y696] (noting that “[w]ithout strong alignment of compliance management to business strategy, it is difficult to efficiently integrate compliance into business processes”). Creating an open and ongoing policy-evaluation process, on the other hand, communicates a depth and seriousness of cultural commitment, unlike the reactive bursts of policy-related activity that often follow an incident but are not sustained after attention moves on.

B. Communication

Even a well-crafted policy will have little influence in the organizational environment unless employees are familiar with it.124Even a well-known policy is rarely sufficient, on its own, to constrain employee misconduct, as the #MeToo movement’s narratives have shown. See supra note 104 and accompanying text. From a compliance perspective, an onboarding training and a periodic email or online refresher may suffice. But from a workplace-culture standpoint, both of those, without more, will convey to employees that the policy is a minor player in the life of the organization.125Cf. Deloitte, supra note 123, at 4 (“Values—with ethics and integrity at their core—must be clearly and consistently communicated.”). Here, questions that might be considered include:

How do employees learn about the policy? Is it part of a generic package of employment policies or is it linked to the organization’s goals and values?

Is the policy easy to find? Is it clear where employees can go if they have questions?

What kinds of reminders do employees receive?

What is the focus of and purpose conveyed by announcements about training? That is, are communications about trainings focused on compliance or framed in terms of organizational values?

Are messages tested with employees and evaluated for clarity and effectiveness?

In short, the central issue from a workplace-culture standpoint is not whether the employer has a policy and issues reminders, but rather how that employer links its policy communications to the organization’s shared beliefs and values and whether employees understand what the employer is trying to say.

C. Systems for Reporting and Responding to Complaints

As with policy, a culture-oriented analysis of an employer’s system for receiving and responding to complaints and concerns will add considerations that a compliance focus is likely to miss. The key, again, is not the words on the page but the meaning of those words within the workplace.

One foundational choice concerns the options given to employees about where to go with concerns about harassment or other discriminatory or abusive behavior. To satisfy compliance requirements, organizations must provide a clear path by which employees can report complaints about harassment or other discrimination and seek redress.126See supra note 120 (discussing Burlington Indus., Inc. v. Ellerth, 524 U.S. 742 (1998), and Faragher v. City of Boca Raton, 524 U.S. 775 (1998)). From a risk-management perspective, facilitating reporting can also be helpful as an early warning sign of potential problems that may expose the employer to liability. Most often, this is accomplished by directing employees to a supervisor or human resources staff.

Yet think about the factors elaborated in Part I that influence employees’ decisions about whether to come forward with a concern about themselves or a coworker. While many of those are deeply personal and outside of an employer’s control, a system that does not take account of individual-level barriers to reporting is unlikely to pick up much of the harassment or other problems on the ground. Design choices that are relevant here include, for example:

Are there multiple entry points to the system so that an employee can choose to go either to someone they know or to someone who is not in their circles?

Is it easy for an employee to report online or in person, so that individuals can choose between face-to-face and less direct ways of sharing their experiences or concerns?

Are the reporting options structured in a way that makes them accessible to employees of all ranks and roles within the organization? Reporting options may look different (or may need to be communicated differently, as discussed below), for employees in the field as compared to those in office locations, and for those in lower-level or more senior positions.

Does the online reporting option explain clearly what will happen after someone submits a report? Do the supervisors or others charged with receiving in-person reports know enough to explain the process to an employee?

How clearly does the system address questions about confidentiality and retaliation? And at what point in the process (before or after reporting) are these concerns addressed?

As discussed at the outset, reporting harassment remains difficult for many people, and individuals will vary in their preference for reporting a concern to a known colleague, a stranger in another office, or via an online form or telephone hotline.127See supra Section I.A; see also Kim, supra note 31, at 426–27, 431 (describing social relationships, power dynamics, low self-esteem, embarrassment, and fear of retaliation as barriers to reporting). Having a range of access points, ideally with input from employees themselves, increases the likelihood that concerns will come to the attention of the organization’s leadership rather than being stymied because employees distrust or are uncomfortable with the one or two options on offer.128Johnson et al., supra note 72, at 3 (recognizing that advisors and mentors may be involved in diverting employees from reporting); see also Foster & Fullagar, supra note 33, at 156 (discussing the role of reputational concerns in decisions about whether to report harassment).

Importantly, these multiple access points can, if they are implemented effectively, become a positive part of the fabric of workplace culture. By placing accountability mechanisms throughout the organization, an employer may be able to reinforce that behavioral norms in the workplace are not solely the responsibility of human resources or other “back of the house” staff.129Cf. Deloitte, supra note 123, at 3 (urging that a culture of integrity involves “[s]enior leaders hold themselves and those reporting to them accountable for complying with the law and organizational policy” (emphasis added)). Alternately, when seen solely through a compliance lens, reporting becomes a side activity, distinct from culture, and may be understood to be aberrational rather than integral to the employer’s expectations and employees’ shared understandings of their roles.130Id. at 4 (noting that “[s]ome organizations even make adhering to values part of the goal-planning process by setting objectives that are tied to specific cultural elements” to reinforce organizational values throughout the workplace).

D. Training and Local Knowledge

Training is another employer-sponsored activity that often falls into the “check the box” category, with responsibility for training design and implementation typically sitting with human resources staff or an organization’s lawyers.131See supra note 113 for discussion of Lauren Edelman’s work on this point. See also Joanna L. Grossman, The Culture of Compliance: The Final Triumph of Form over Substance in Sexual Harassment Law, 26 Harv. Women’s L.J. 3, 13–14, 42–49 (2003) (arguing that the incentive underlying employer training programs is to prevent legal liability, not to prevent harassment). Yet, as should be apparent at this point in the argument, training is essentially a cultural intervention, even if it is not understood in that way by the organization. As with policy, approaching training as a “must do” compliance requirement, rather than as core to the organizational mission, will send a cultural signal that the organization views the covered topics as outside of what matters most.132Cf. Susan Bisom-Rapp, An Ounce of Prevention Is a Poor Substitute for a Pound of Cure: Confronting the Developing Jurisprudence of Education and Prevention in Employment Discrimination Law, 22 Berkeley J. Emp. & Lab. L. 1, 6 (2001) [hereinafter Bisom-Rapp, An Ounce of Prevention] (observing that “no matter how well intentioned, [trainings] are poor substitutes for searching inquiry into the particulars of a given workplace”); Frank Dobbin & Alexandra Kalev, Why Diversity Programs Fail, Harv. Bus. Rev. (Aug. 2016), https://hbr.org/2016/07/why-diversity-programs-fail [https://perma.cc/WAT3-CCQ6] (discussing how—at least with diversity training—“laboratory studies show that this kind of force-feeding [mandatory training] can activate bias rather than stamp it out”). When training is viewed strictly through a compliance lens, employers may also be disincentivized to evaluate its effectiveness. See Susan Bisom-Rapp, Fixing Watches with Sledgehammers: The Questionable Embrace of Employee Sexual Harassment Training by the Legal Profession, 24 T. Jefferson L. Rev. 125, 146 (2002) (noting that employers may be reluctant to evaluate their training programs out of concern that those findings could be used against them in discrimination litigation).

Culture-oriented questions here would be similar to those noted above, with a focus on the link between training, organizational values and the employee experience at all levels of the organization:133These are in addition to questions employers should consider to ensure they have incorporated practices in training design so that employees remember the information provided. On research and best practices in workplace harassment training, see, for example, Feldblum & Lipnic, supra note 31, at 44–53. For proposals to achieve effective employee engagement in the area of diversity training, see, for example, Katharine T. Bartlett, Making Good on Good Intentions: The Critical Role of Motivation in Reducing Implicit Workplace Discrimination, 95 Va. L. Rev. 1893, 1960–71 (2009), discussing motivational principles that “support employee autonomy, competence, and relatedness.” For discussion of adult learning theory and its application to employee training, see generally David M. Rosch & Corey Seemiller, An Integrative SixDomain Model of Employee Training and Development, 31 New Horizons Adult Educ. & Hum. Resource Dev. 25 (2019), identifying and applying elements of adult learning theory models to employee training. As discussed supra notes 131–32, some scholars have questioned whether training can ever be effective particularly in the context of diversity. See, e.g., Alexandra Kalev et al., Best Practices or Best Guesses? Assessing the Efficacy of Corporate Affirmative Action and Diversity Policies, 71 Am. Soc. Rev. 589, 611 (2006) (studying employer training in the context of diversity-related efforts and finding that “[p]ractices that target managerial bias through . . . education (diversity training) show virtually no effect in the aggregate”).

Does the training express values in addition to behavioral rules and reporting information?

How does the training account for the ways in which shame, embarrassment, fear of social and job-related retaliation, and other individual-level factors may be barriers to reporting?

What do evaluations show about employee absorption of the training content, and does the evaluation process include meaningful engagement by employees and refinement of future training?

Related but often missed when thinking about training is the importance of ensuring that there are enough “local experts” who understand the organization’s policy along with the basics of reporting and the complaint-resolution process.134See, e.g., Deloitte, supra note 123, at 3 (observing the responsibility of “front-line and mid-level supervisors [to] turn principles into practice”). This draws again from the individual-level factors discussed in Part I, and the idea that many people are most comfortable turning to someone they know and trust for help and information.135See supra notes 38–40 and accompanying text (discussing how employees with supportive peers may be more likely to report workplace harassment). The focus on local experts also builds on the concept of “herd immunity”—that not every individual needs inoculation for the community to be protected from a particular disease.136Rhea Boyd, It Takes a Herd, Am. Acad. Pediatrics: AAP Voices (Apr. 8, 2016), https://www.aap.org/en-us/aap-voices/Pages/It-Takes-a-Herd.aspx [https://perma.cc/BW9J-E2PK]. Carried over to the employment context, the idea is that it is essential to diffuse enough knowledge about harassment policy and procedures so that the information permeates a workplace and anyone in the community can find what they need from a coworker if they do not have the information themselves.

E. Privacy and Retaliation Protections

Protection against retaliation and policies for protecting the privacy of employee complaints also sound in the register of legal compliance, not workplace culture. But again, both are indicators of an employer’s values and expectations, and each conveys whether the employer is sensitive to the challenges individuals might face when dealing with harassment.137See Feldblum & Lipnic, supra note 31, at 16; see also Comm’n on Gender & Comm’n on Race & Ethnicity, Report of the Third Circuit Task Force on Equal Treatment in the Courts, 42 Vill. L. Rev. 1355, 1505 (1997) (discussing study’s observations that “[r]elatively few employees use the formal complaint or grievance system” and that “[t]he lack of use may be due to a perception among employees that the complaint system is . . . retaliatory”). As the individual-level factors discussed in Part I indicate, employees may hesitate to report a concern if they believe their coworkers will hear about it.138See supra notes 35–37 and accompanying text (discussing reputational concerns associated with being perceived as disruptive or unprofessional). A compliance-oriented policy typically will remind employees about limits on confidentiality and that information will be shared with the alleged harasser as part of an investigation. When approached through a workplace-culture lens, the same information can be provided not as a stand-alone safeguard of the employer’s needs but instead with a link to broader organizational values, making clear that the employer understands the complaining employee’s interests in privacy and ongoing ability to work and will do all it can to protect those interests.139Because reputational concerns can present a significant barrier to bringing forward a complaint, a responsive policy might, for example, underscore that information will be shared only on a “need to know” basis.

F. Capacity-Building for Staff who Handle Complaints

At the end of the day, policies are only as effective as the people who implement them and the resources available to support those people, which means that staff who address harassment will themselves shape workplace culture in addition to ensuring compliance. For this reason, training for those who implement the employer’s policy is significant not just for avoiding liability but also for communicating the employer’s values related to fairness, impartiality, and sensitivity to the issues involved.140Feldblum and Lipnic expand on this point in analyzing the elements of well-functioning systems, discussing, inter alia, a finding that federal managers “often recast harassment complaints as personality clashes or interpersonal difficulties.” Feldblum & Lipnic, supra note 31, at 24 n.182 (citing Lauren B. Edelman, Howard S. Erlanger & John Lande, Internal Dispute Resolution: The Transformation of Civil Rights in the Workplace, 27 Law & Soc’y Rev. 497 (1993)). The issue of training on policy implementation has been much discussed in the Title IX context for students. See, e.g., Michele Landis Dauber & Meghan O. Warner, Legal and Political Responses to Campus Sexual Assault, 15 Ann. Rev. L. & Soc. Sci. 311, 323 (2019); Jacob Gersen & Jeannie Suk, The Sex Bureaucracy, 104 Calif. L. Rev. 881, 897–911 (2016). I have also commented in this area. See Goldberg, supra note 13, at 112. Ill-trained staff who are not capable of educating employees about the process and responding effectively to questions and concerns will reveal, even if unintentionally, how the employer values its employees.141See Bell et al., supra note 35, at 147 (showing that some meaningful response to a report leads to improvements in the employee’s well-being); see also Feldblum & Lipnic, supra note 31, at 35 (“When the right behaviors . . . are rewarded, that sends a message about what an organization’s leadership cares about.”). A case manager or intake specialist who does not have investigative responsibilities but does have specialized knowledge can play a critical role in informing both complainants and respondents about their options and keeping them apprised of steps in the process. For an example from an academic setting, see Columbia Univ., Gender-Based Misconduct Policy and Procedures for Students 25–26 (2019), http://www.columbia.edu/cu/studentconduct/documents/GBMPolicyandProceduresforStudents.pdf [https://perma.cc/YE3Q-8278]. By contrast, if those who handle complaints, whether in human resources or other parts of the enterprise, earn a reputation for being effective, they may be among an employer’s most valuable resources for both addressing harassment and reinforcing the employer’s core mission.

G. Alternate Dispute Resolution

When reporting complaints or concerns, many employees want only to have the conduct end so they can work unimpeded by harassment. But in some settings, a full investigation can take time and cause disruption in the local environment, triggering reputational concerns and other individual-level barriers to reporting.142See Feldblum & Lipnic, supra note 31, at 16. Decisions about whether and how to offer expedited, context-sensitive options for addressing incidents are thus, like others already discussed, not only part of the policy-and-procedure world of human resources and legal compliance but also part of the workplace culture.143The use of referees in contract disputes to enable less formal resolution of conflicts where parties are in an ongoing relationship offers a promising reference point. See Ronald J. Gilson, Charles F. Sabel & Robert E. Scott, Braiding: The Interaction of Formal and Informal Contracting in Theory, Practice, and Doctrine, 110 Colum. L. Rev. 1377, 1403 (2010). Methods of dispute resolution with reduced formality also carry risks, including that power imbalances may impede fair resolutions. See generally Eric K. Yamamoto, ADR: Where Have All the Critics Gone?, 36 Santa Clara L. Rev. 1055, 1058–62 (1996) (discussing critiques of alternate dispute resolution). But see Carrie Menkel-Meadow, The Trouble with the Adversary System in a Postmodern, Multicultural World, 38 Wm. & Mary L. Rev. 5, 12–24 (1996) (arguing that the adversarial system is inadequate for many types of dispute resolution). If record-keeping is not systematic for informal resolution processes, there is also the risk that repeat offenders will be more difficult to identify. Cf. Ditkowsky, supra note 75, at 101–02 (discussing limitations of information escrows that enable an individual to record an experience of harassment but not disclose that experience directly to the employer).

H. Sanctions and Other Responses to Policy Violations

The penalty and response structure for harassment-policy violations also may influence an individual’s willingness to make a report or even express a concern. As in the criminal law context, workplace penalties that are seen as disproportionately harsh may lead employees to avoid reporting out of concern for themselves or their colleague.144See Feldblum & Lipnic, supra note 31, at 40. The EEOC Co-Chairs of the Select Task Force on the Study of Harassment in the Workplace recommended against “zero tolerance” messaging or “one-size-fits-all” sanctions because of concerns about “employee under-reporting of harassment, particularly where they do not want a colleague or co-worker to lose their job over relatively minor harassing behavior—they simply want the harassment to stop.” Id. At the same time, sanctions that are seen as inconsequential or unduly light may also lead employees to conclude that reporting an incident is not worth their time or reputational risk.145See, e.g., Comm’n on Gender & Comm’n on Race & Ethnicity, supra note 137, at 1505 (hypothesizing that the complaint system was underutilized because employees believed it was “useless” and “futile”).

Calibrating sanctions to the nature of workplace misconduct is a law-driven activity in that unduly weak sanctions may create liability for an organization. Yet through the workplace-culture lens, it becomes easier to see how sanctions express organizational values, too. Treating sanctions as part of workplace culture also widens the frame for thinking about types of interventions that might be effective for culture-building as well as for penalizing wrongful behavior. In a local environment where misconduct has taken place, for example, an effective response would address not only the individual responsible for a policy violation but also the surrounding team. This could include attention to dynamics that preceded or supported the wrongful behavior as well as the effects of the misconduct, investigation, and penalty on the broader group.146For discussion of the impact of sexual harassment on teams, see, for example, Jana L. Raver & Michele J. Gelfand, Beyond the Individual Victim: Linking Sexual Harassment, Team Processes, and Team Performance, 48 Acad. Mgmt. J. 387 (2005).

I. Transparency/Information-Sharing

Information is essential to culture-building as well—what is provided and what is withheld is another mode of communicating organizational values. With respect to harassment in particular, an employer’s choices will signal whether an employer is serious about generating trust in its system for handling complaints.147Cf. Feldblum & Lipnic, supra note 31, at 35–36, 35 n.165 (indicating that higher reporting levels reveal employees’ trust in the employer’s system). Without information showing that the system is prompt and effective in addressing concerns, employees may not be motivated to bring complaints given the potential individual-level costs. Yet too much information—such as detailed reporting that can be traced back to individual cases—may reduce employees’ willingness to come forward with complaints or to participate in the process as a witness or even a respondent. Tensions exist as well with the privacy protections that apply ordinarily to personnel matters.148See id. at 42. Reporting aggregate data can help strike the balance and, in doing so, convey an employer’s seriousness and sense of accountability to its workforce.149Because employees frequently look to close colleagues for support, as noted earlier, matters may become known within a unit or department even without disclosure by the employer. In these instances, when employees seek additional information, an employer will have to balance many considerations in deciding what, if anything, to share about a specific case. A non-comprehensive list of considerations would include: a) Are all of the employees who are directly involved in the conflict sharing information publicly?; b) Does that information sharing accurately reflect the employer’s response?; c) Has the accused employee left the workplace, either voluntarily or not?; d) Is there discussion of the issue on social media or in other public settings? Each of these questions, along with others, informs the balance between the benefit to parties from privacy protection and the harm that can be caused by the spread of information that, in the employer’s view, is inaccurate and may discourage reporting of additional misconduct.

IV. Legal Accountability and the Challenge of Extra-Legal Harms

This Part turns more closely to the ways in which law interacts with the culture-shaping legal-accountability choices just described. Traditional legal authorities, such as statutes, regulations, and court-made doctrine, are important, of course, and I will note them briefly here.

But there is an additional area where law—or, really, the absence of enforceable law—exerts significant influence on workplace culture and the prevalence of harassment and other forms of discrimination.150Cf. Michel Foucault, Discipline and Punish (Alan Sheridan trans., Vintage Books 2d ed. 1995) (1977) (characterizing the law’s refusal to recognize certain harms as generating its own form of power and control). I will reserve a fuller discussion of the shadow category of extra-legal harms for later work. This is the area of extra-legal harms—that is, the behaviors that are not legally actionable and hence not the subject of compliance requirements but can still be as detrimental as cognizable discrimination to employees who experience them.

Most forms of implicit bias and microaggression fall into this category. They do not demonstrate the invidious intent required to satisfy the disparate treatment standard applied to most employment discrimination claims and are regularly deemed insufficient to prove the existence of a hostile environment.151The standards for establishing hostile-environment and disparate-treatment claims are distinct. As one court explained: “[T]o state a claim for a hostile work environment, a plaintiff must plead facts that would tend to show that the complained of conduct: (1) “is objectively severe or pervasive—that is, . . . creates an environment that a reasonable person would find hostile or abusive”; (2) creates an environment “that the plaintiff subjectively perceives as hostile or abusive”; and (3) “creates such an environment because of the plaintiff’s sex.” On the other hand, a gender discrimination claim based on disparate treatment requires a plaintiff to plead facts that would tend to show “that the defendant had a discriminatory intent or motive for taking a job-related action.” Desouza v. Off. of Child. & Fam. Servs., No. 18-CV-2463, 2019 WL 2477796, at *4 (E.D.N.Y. June 12, 2019) (citations omitted).

In addition there is a subcategory that I call “low-grade harassment,” meaning the intentional acts that are similar to cognizable misconduct but insufficiently severe or pervasive to give rise to liability.152In the sexual harassment context, the Supreme Court has held that “[c]onduct that is not severe or pervasive enough to create an objectively hostile or abusive work environment—an environment that a reasonable person would find hostile or abusive—is beyond Title VII’s purview.” Oncale v. Sundowner Offshore Servs., Inc., 523 U.S. 75, 81 (1998) (quoting Harris v. Forklift Sys., Inc., 510 U.S. 17, 21 (1993)). The category of low-grade harassment is akin to what is sometimes described as “microassault,” meaning “attacks meant to harm the victim that are similar to traditional forms of discrimination.” Eden B. King et al., Discrimination in the 21st Century: Are Science and the Law Aligned?, 17 Psychol., Pub. Pol’y & L. 54, 56 (2011); see also Goldberg, supra note 16, at 793 n.216 (defining the analogous category of “low-grade discrimination” as “discriminatory acts that the law has been construed not to prohibit”). These acts, in other words, are different in degree but not in kind from conduct that would be considered unlawful.

From a workplace-culture standpoint, the critical question is whether and how an employer addresses these kinds of conduct that may look and feel like harassment but are not covered by antidiscrimination law. The concern is not only for the affected employees but also for the workplace as a whole. Ambiguity at the border of permissible and impermissible conduct can spill over to confusion about what kinds of conduct the organization deems acceptable. This in turn may undermine efforts to protect against liability and, further, cut against other efforts to build a culture where employees feel valued and comfortable bringing forward concerns.

A. Traditional Sources of Law

The body of law governing workplace harassment is itself complex.153See, e.g., Cynthia Estlund, Rebuilding the Law of the Workplace in an Era of Self-Regulation, 105 Colum. L. Rev. 319, 326 (2005) (observing that post-New Deal employment law is comprised of “a growing array of statutes and common law doctrines governing various terms and conditions of employment”). In addition to statutes and case law that directly address discrimination,154Id. at 331–32 (discussing evolution of federal and state antidiscrimination law). there is criminal and tort law regarding nonconsensual touching;155See Joanna Stromberg, Sexual Harassment: Discrimination or Tort?, 12 UCLA Women’s L.J. 317, 318 (2003) (describing assault and battery claims to challenge workplace sexual harassment). labor law that governs unionized workforces;156See generally 29 U.S.C. §§ 151–169 (2018) (setting out the provisions of the National Labor Relations Act); see also Estlund, supra note 153, at 328–30 (describing the development of labor standards laws). industry-specific statutory and regulatory oversight;157See Olatunde C.A. Johnson, Overreach and Innovation in Equality Regulation, 66 Duke L.J. 1771, 1789–1800 (2017) (exploring the role that agencies and regulated entities, including employers, play in crafting solutions to pervasive civil rights concerns). labor and tax laws that affect the classification of employees and contractors;158See Shu-Yi Oei & Diane M. Ring, Tax Law’s Workplace Shift, 100 B.U. L. Rev. 651, 685 (2020) (commenting on the relationship between legal tests in tax and labor law to determine whether a worker is an employee or independent contractor). immigration law that can make some employees more or less vulnerable to exploitation;159See Lori A. Nessel, Undocumented Immigrants in the Workplace: The Fallacy of Labor Protection and the Need for Reform, 36 Harv. C.R.-C.L. L. Rev. 345, 348 (2001). family and medical leave laws that may become a point for negotiation or coercion;160Jessica Snorgrass, Waiving the Effectiveness of the FMLA: The Anti-Waiver Approach to Enforceability of FMLA Severance Agreement Waivers, 45 San Diego L. Rev. 163, 202 (2008) (“The pervasive history of gender discrimination against mothers in the workplace raises concerns that women may be particularly susceptible to coercion in agreeing to sign [Family Medical Leave Act] waivers, or may be targeted by employers to sign such agreements.”). But see Family Medical Leave Act of 1993 (FMLA), 5 U.S.C. § 6385(a) (2018) (“An employee shall not directly or indirectly intimidate, threaten, or coerce, or attempt to intimidate, threaten, or coerce, any other employee for the purpose of interfering with the exercise of any rights which such other employee may have under this subchapter.”). and the list goes on.161See Estlund, supra note 153, at 321 (“[T]he role of external law—of courts, of legislation, and of regulatory bodies—has burgeoned . . . .”).

Beyond these formal bodies of law there are, to use Robert Cover’s words, the “narratives that . . . give [them] meaning.”162Robert M. Cover, The Supreme Court, 1982 Term—Foreword: Nomos and Narrative, 97 Harv. L. Rev. 4, 4 (1983). These narratives may be especially influential in the workplace where few people will ever file formal complaints, much less bring litigation,163See Lauren B. Edelman, Legal Ambiguity and Symbolic Structures: Organizational Mediation of Civil Rights Law, 97 Am. J. Soc. 1531, 1540 (1992) (noting that “[l]awsuits are a primary means of defining the boundaries of compliance, yet research shows that people who see themselves as victimized by employment discrimination rarely pursue legal redress” (first citing Kirstin Bumiller, Victims in the Shadow of the Law: A Critique of the Model of Legal Protection, 12 J. Women Culture & Soc’y 421 (1987); and then citing Richard E. Miller & Austin Sarat, Grievances, Claims, and Disputes: Assessing the Adversary Culture, 15 Law & Soc’y Rev. 525 (1981))). but many will have views about the kinds of interactions among coworkers the law permits and forbids, and about how managers ought to respond when conflicts arise.164See supra notes 45–46 and accompanying text.

B. The “Law” of Extra-Legal Harms: Microaggression, Implicit Bias, and Low-grade Harassment

In addition to defining prohibited conduct, the sources of law just described also signal what kinds of behaviors and comments are permissible. Yet some of this permissible activity is experienced as harassment by targeted employees and their coworkers, negatively affecting workplace terms and conditions as much as if the conduct was unlawful. But the law leaves a vacuum here, and for employers focused primarily on legal compliance, the idea of addressing these harmful but not unlawful behaviors and comments may not register at all.

From a workplace-culture standpoint, the challenge is that by leaving microaggressions, implicit bias, and low-grade harassment unaddressed, employers may convey that experiences of these behaviors do not “count” as concerns. To the extent these interactions look and feel similar to those prohibited by formal policy, an employer’s disregard of them also may signal disinterest in addressing conduct that is prohibited.165See Feldblum & Lipnic, supra note 31, at 32 (noting that “leaders who do not model respectful behavior, who are tolerant of demeaning conduct or remarks by others, or who fail to support anti-harassment policies with necessary resources, may foster a culture conducive to harassment”). In this way, the category of extra-legal harms itself may be a potent barrier to addressing workplace harassment.

1. Microagression and implicit bias

Microaggressions in the workplace include behaviors, comments, and other interactions that employees may experience as harassing even if not intended that way by a supervisor or coworker.166See Peggy C. Davis, Law as Microaggression, 98 Yale L.J. 1559, 1560 (1989) (observing that discrimination law’s intent requirement does not effectively serve the purpose of eradicating racism because “American racism is pervasive and largely unconscious”); id. at 1565–66 (quoting Chester Pierce, who coined the term “microaggression,” defining microaggressions as “subtle, stunning, often automatic, and non-verbal exchanges which are ‘put downs’ of blacks by offenders” that “stem from the mental attitude of presumed superiority.” (first quoting Chester M. Pierce et al., An Experiment in Racism: TV Commercials, 10 Educ. & Urb. Soc’y 61, 66 (1977); and then quoting Chester Pierce, Psychiatric Problems of the Black Minority, in American Handbook of Psychiatry 512, 515 (S. Ariel ed. 1974))). An additional, widely cited definition comes from Derald Wing Sue, who describes microaggressions as, “[c]ommonplace verbal or behavioral indignities, whether intentional or unintentional, which communicate hostile, derogatory, or negative racial slights and insults.” Derald Wing Sue et al., Racial Microaggressions in Everyday Life, 62 Am. Psychol. 271, 278 (2007). Sue later extended his work to address microaggressions based on aspects of identity in addition to race. See Derald Wing Sue, Microaggressions in Everyday Life: Race, Gender, and Sexual Orientation 137–59 (2010); see also Derald Wing Sue, Microaggressions: More than Just Race, Psychol. Today (Nov. 17, 2010), https://www.psychologytoday.com/us/blog/microaggressions-in-everyday-life/201011/microaggressions-more-just-race [https://perma.cc/45G4-9ZFW]. An extensive literature documents their harmful effects, including on individuals’ health and workplace participation.167See, e.g., Kevin L. Nadal et al., The Injurious Relationship Between Racial Microaggressions and Physical Health: Implications for Social Work, 26 J. Ethnic & Cultural Diversity Soc. Work 6, 12 (2017) (identifying correlation between racial microaggressions and poor general health, pain, lower energy levels, fatigue, and other health issues); Suchitra Shenoy-Packer, Immigrant Professionals, Microaggressions, and Critical Sensemaking in the U.S. Workplace, 29 Mgmt. Comm. Q. 257, 259, 260, 262–65 (2015) (describing negative impacts of microaggressions on immigrant professionals as including learned helplessness, unequal power relationships, persistent aggravation, invalidation, and demoralization); Tinkler, supra note 43, at 419 (“While research has found that people tend to believe that sexual jokes or remarks are a part of normal work life, evidence suggests that those who encounter it are often troubled by their experiences, consider it inappropriate, and do not know whether or how to stop it.”); cf. Tessa E. Basford, Do You See What I See? Perceptions of Gender Microaggressions in the Workplace, 38 Psychol. Women Q. 340, 342, 345–46 (2013) (observing that women who experience discrimination in the form of microaggressions reported declines in well-being, reduced work performance, and lower organizational commitment than men who experienced similar amounts of discrimination). Yet absent a dramatic change in the law, an employee’s experience of microaggressions will not give rise to a cognizable claim.168

See, e.g., Nichols v. Mich. City Plant Planning Dep’t, 755 F.3d 594, 600–05 (7th Cir. 2014) (affirming that a Black school janitor who alleged microaggressions from students and staff presented insufficient evidence for a reasonable juror to conclude that the janitor was subjected to severe or pervasive harassment); Arizona v. Metro. Transp. Auth., No. 17CV4491, 2019 WL 2613476, at *6 (S.D.N.Y. June 26, 2019) (finding that two race-based comments and unspecified microaggressions do not constitute a “steady barrage of opprobrious racial comments” sufficient to establish a hostile work environment claim (quoting Morrison v. United Parcel Serv., Inc., No. 17CV2885, 2019 WL 109401, at *3 (S.D.N.Y. Jan. 4, 2019))); Kiani v. Huha, No. 27-CV-17-9003, 2018 WL 4855437, at *4 (Minn. Ct. App. Oct. 8, 2018) (holding that “[m]erely labeling the request with the nebulous term ‘microaggression,’ which can involve purely unintentional communication, adds nothing to [an] unsupported accusation of bias”); cf. King et al., supra note 152, at 69 (discussing microaggressions and observing findings that “seem to indicate a disconnect between the experiences of targets of discrimination and the legal system in which recourse is sought”).

Implicit biases likewise may result in practices or decisions that reduce opportunities for subgroups of employees based on race, gender, and other aspects of identity.169See, e.g., Susan Sturm, Activating Systemic Change Toward Full Participation: The Pivotal Role of Boundary Spanning Institutional Intermediaries, 54 St. Louis U. L.J. 1117, 1122 (2010) [hereinafter Sturm, Activating Systemic Change Toward Full Participation] (observing that “[u]nderrepresented group members experience implicit bias [and] exclusion from opportunity networks” (footnote omitted)). See generally Jerry Kang, Trojan Horses of Race, 118 Harv. L. Rev. 1489 (2005) (discussing sources and effects of implicit bias). Yet evidence of the link between implicit bias and discriminatory acts has been difficult to establish,170See, e.g., Gregory Mitchell, An Implicit Bias Primer, 25 Va. J. Soc. Pol’y & L. 27, 53 (2018) (noting “the tenuous relationship between implicit bias and behavior”). and although courts have acknowledged the existence and potential effects of unexpressed biases,171See, e.g., Watson v. Fort Worth Bank & Tr., 487 U.S. 977, 990 (1988) (observing “the problem of subconscious stereotypes and prejudices” in workplaces). implicit bias, without more, is not generally understood as grounds for employer liability.172See, e.g., Samuel R. Bagenstos, Implicit Bias’s Failure, 39 Berkeley J. Emp. & Lab. L. 37 (2018) (arguing that efforts to increase awareness of and challenge implicit bias have not succeeded); Michael Selmi, The Paradox of Implicit Bias and a Plea for a New Narrative, 50 Ariz. St. L.J. 193, 198 (2018) (“Courts have long had difficulty addressing unconscious bias . . . [and] have begun to reject expert testimony regarding implicit bias, in large part because the general message that it is pervasive and unconscious is difficult to square with traditional notions of legal proof.” (footnotes omitted)).

2. Low-grade harassment

Conduct in the category I call “low-grade harassment” is similar to implicit bias and microaggression in its potential for detrimental effect on others and in its non-cognizability as a legal harm but different because the behavior is understood to be intentional. Think here of the unwanted, intentional comments, physical contact, and other behaviors (such as staring at a colleague’s body parts) by coworkers, supervisors, or others that negatively affect the targeted individual’s experience at work based on sex, race, or other protected aspects of identity but are ineligible for a court-ordered remedy because they are infrequent or come from someone without direct authority over the targeted employee.173I intend also to include coworkers whose workplace terms and conditions are affected negatively by these behaviors. Low-grade harassment thus occupies a place somewhere in between what is clearly prohibited and what is widely acceptable.174Cf. Kathleen Kelley Reardon, It’s Not Always Clear What Constitutes Sexual Harassment. Use This Tool to Navigate the Gray Areas, Harv. Bus. Rev. (June 19, 2018), https://hbr.org/2018/06/its-not-always-clear-what-constitutes-sexual-harassment-use-this-tool-to-navigate-the-gray-areas (setting out a six-point spectrum of sexual misconduct ranging from “generally not offensive conduct” to “egregious sexual misconduct” for employers to use as a framework in workplace-culture trainings). In this sense, and because it looks so much like prohibited behavior, low-grade harassment has even greater potential than microaggressions and implicit bias to cause confusion for employers and employees.

The Supreme Court arguably created the category of “low-grade” harassment when it concluded that sexual harassment, to be actionable based on a hostile-environment theory, must be “severe or pervasive.”175Meritor Sav. Bank, FSB v. Vinson, 477 U.S. 57, 67 (1986) (finding an employee’s harassment of a supervisee to be sufficiently severe or pervasive “to alter the conditions of [the victim’s] employment and create an abusive working environment” (alteration in original) (quoting Henson v. City of Dundee, 682 F.2d 897, 904 (11th Cir. 1982))). Title IX jurisprudence requires the misconduct to be both severe and pervasive. See Davis v. Monroe Cty. Bd. of Educ., 526 U.S. 629, 651 (1999) (“[A] plaintiff must establish sexual harassment of students that is so severe, pervasive, and objectively offensive, and that so undermines and detracts from the victims’ educational experience, that the victim-students are effectively denied equal access to an institution’s resources and opportunities.”). The Department of Education’s 2020 Title IX regulations reinforce the federal government’s commitment to “severe, pervasive, and objectively offensive” as the standard by which sexual harassment claims in educational settings should be evaluated. 85 Fed. Reg. 30,026, 30,136 (May 19, 2020) (to be codified at 34 C.F.R. pt. 106). A small number of jurisdictions, including New York City and State apply a different, more encompassing standard to hostile-environment claims. See Abe v. N.Y. Univ., 169 A.D.3d 445 (N.Y. App. Div. 2019) (holding that the court below “incorrectly cited the ‘severe and pervasive’ standard in evaluating plaintiff’s hostile work environment claim, instead of applying the more liberal standard under the New York City Human Rights Law (HRL)”), appeal dismissed, 139 N.E.3d 846 (2020); Robert H. Bernstein et al., Attention New York Employers: When It Comes to Workplace Harassment, Times Are Changing, Nat’l L. Rev. (Aug. 20, 2019), https://www.natlawreview.com/article/attention-new-york-employers-when-it-comes-to-workplace-harassment-times-are [https://perma.cc/2E3Q-PH5K] (“Setting aside settled federal precedent, the new state law, like the city law, expands the kinds of behavior within its reach to include harassment ‘regardless of whether’ it ‘would be considered severe or pervasive.’ Such conduct will now constitute an unlawful discriminatory practice if the affected employee experiences ‘inferior terms, conditions or privileges of employment.’”). Other jurisdictions that have moved away from the “severe or pervasive” standard include California, A single incident of harassing conduct is sufficient to create a triable issue regarding the existence of a hostile work environment if the harassing conduct has unreasonably interfered with the plaintiff’s work performance or created an intimidating, hostile, or offensive working environment.”); Delaware,conduct that “has the purpose or effect of unreasonably interfering with an employee’s work performance or creating an intimidating, hostile, or offensive working environment”); Washington, Wash. Admin. Code § 392-190-056(1)(c) (2019) (prohibiting “conduct or communication [that] has the purpose or effect of substantially interfering with an individual’s educational or work performance, or of creating an intimidating, hostile, or offensive educational or work environment”). That is, some amount of unwanted touching and comments would be permissible but a larger quantum of the same conduct would cross the line and be subject to judicial remedy.176This approach has produced numerous cases in which courts rejected claims because the amount of sexual misconduct shown by the plaintiff did not meet the “severe or pervasive” threshold. Typical is a case in which a federal district court granted summary judgment to the employer even after the plaintiff-employee submitted evidence showing that a coworker asked the plaintiff out on multiple occasions, told the plaintiff he had her phone number, and gave the plaintiff his phone number saying that “he was a nice guy and to give him a try.” High v. R & R Transp., Inc., 242 F. Supp. 3d 433, 444 (M.D.N.C. 2017). In this case, the plaintiff had asked her supervisor to tell the coworker to stop asking her out, and the supervisor “told Plaintiff there was nothing he could say.” Id. at 441. The court in High relied on a Fourth Circuit case in which an administrative assistant alleged a hostile environment based on comments by a manager and salespeople in her office, including “[w]e’ve made every female in this office cry like a baby”; a supervisor’s question to a female colleague about “whether she would be a ‘mini van driving mommy’ or ‘be a salesperson and play with the big boys’”; and a comment that the employee should “go home and fetch [her] husband’s slippers like a good little wife.” Hartsell v. Duplex Prods., Inc., 123 F.3d 766, 773 (4th Cir. 1997). That court observed that to allow the employee’s claim to go to trial “would countenance a federal cause of action for mere unpleasantness.” Id. at 773. By anchoring the standard in this way, a non-trivial number of cases reject claims involving unwanted and explicitly sexualized interactions.177See generally Sperino & Thomas, supra note 16 (analyzing numerous cases in which courts reject harassment claims); see also, e.g., Weiss v. Coca-Cola Bottling Co. of Chi., 990 F.2d 333, 337 (7th Cir. 1993) (concluding that a supervisor’s actions were not sufficiently severe or pervasive where he asked the plaintiff repeatedly out on dates, “called her a ‘dumb blond,’ put his hand on her shoulder several times, placed ‘I love you’ signs in her work area and attempted to kiss her in a bar”); Scott v. Sears, Roebuck & Co., 798 F.2d 210, 211–12, 214 (7th Cir. 1986) (sustaining a grant of summary judgment where the plaintiff was subjected to propositions, lewd comments, and a slap on the buttock because the incidents were “relatively isolated”); Berrey et al., supra note 24, at 78–85 (providing examples).

Even in cases involving less severe incidents, a comment or touch that some would find funny and others would call unfortunate or rude (e.g. “Nice skirt, it would look better on my bedroom floor”) may be experienced as disparaging and discriminatory by the person on the receiving end.178See ‘Nice Skirt, It Would Look Better on My Bedroom Floor’—Your Sexual Harassment Stories, Guardian (Sept. 11, 2017, 2:00 AM), https://www.theguardian.com/​careers/​2017/sep/11/sexual-harassment-at-work-stories-readers-stories-workplace [https://perma.cc/EC75-8STB]; see also Robin Pogrebin & Zachary Small, He Left a Museum After Women Complained; His Next Job Was Bigger, N.Y. Times (Jan. 10, 2020), https://www.nytimes.com/​2020/01/10/​arts/​design​/joshua-helmer-philadelphia-museum-art-erie-art-museum.html [https://perma.cc/XTL3-MXGV] (quoting a text from a museum director inviting an intern to coffee and suggesting they get together on his “cool back deck” rather than at a coffee shop). For an extensive collection of examples, see Sperino & Thomas, supra note 16. See also Diane P. Wood, Sexual Harassment Litigation with a Dose of Reality, 2019 U. Chi. Legal F. 395, 411 (2019) (reviewing numerous sexual harassment cases in which plaintiffs lose and expressing “serious concerns about the effectiveness of the legal system in addressing claims of sexual harassment in the workplace”). For compilations and discussion of race-based comments and related incidents in academic and other settings, see generally Charles R. Lawrence III, If He Hollers Let Him Go: Regulating Racist Speech on Campus, 1990 Duke L.J. 431, 431–34 (1990); Daniel Solórzano et al., Keeping Race in Place: Racial Microaggressions and Campus Racial Climate at the University of California, Berkeley, 23 Chicano-Latino L. Rev. 15, 26–27, 34–44, 47–48, 55–56 (2002); Catharine Wells, Microaggressions in the Context of Academic Communities, 12 Seattle J. Soc. Just. 319, 320–23 (2013). Yet, as the Supreme Court has explained, “‘simple teasing’, offhand comments, and isolated incidents (unless extremely serious) will not amount to discriminatory changes in the ‘terms and conditions of employment.’”179Faragher v. City of Boca Raton, 524 U.S. 775, 788 (1998) (quoting Oncale v. Sundowner Offshore Servs., Inc., 523 U.S. 75, 82 (1998)). Although the Supreme Court has described the hostile-environment standard as taking “a middle path between making actionable any conduct that is merely offensive and requiring the conduct to cause a tangible psychological injury,” Harris v. Forklift Sys., Inc., 510 U.S. 17, 21 (1993), in application the path seems to veer strongly to one side. As the Court has clarified repeatedly, “‘mere utterance of an . . . epithet which engenders offensive feelings in an employee’ does not sufficiently affect the conditions of employment to implicate Title VII.” Id. (alteration in original) (citation omitted) (quoting Meritor Savings Bank, FSB v. Vinson, 477 U.S. 57, 67 (1986)).

The Court’s reluctance to step in appears to be tied to perceptions that judicial intervention may chill permissible behavior or amount to enforcement of a civility code at work.180See, e.g., Vance v. Ball State Univ., 570 U.S. 421, 452 (2013) (“Title VII imposes no ‘general civility code.’” (citation omitted)); Faragher, 524 U.S. at 788 (“These standards for judging hostility are sufficiently demanding to ensure that Title VII does not become a ‘general civility code.’” (citation and some internal punctuation omitted)). Hostile-environment doctrine, the Court has emphasized, “does not reach ‘the ordinary tribulations of the workplace, [including] ‘sporadic use of abusive language’ or generally boorish conduct.”181Vance, 570 U.S. at 452; see also Oncale, 523 U.S. at 81 (reiterating in the context of a same-sex sexual harassment claim that Title VII “forbids only behavior so objectively offensive as to alter the ‘conditions’ of the victim’s employment”).

Lower courts have echoed this reluctance, highlighting the limits on their capacity and legitimacy when reviewing internal workplace dynamics. As many have written, “[f]ederal courts do not sit as a super-personnel department that reexamines an entity’s business decisions.”182Torlowei v. Target, 401 F.3d 933, 935 (8th Cir. 2005) (per curiam) (internal punctuation and citation omitted); Charles A. Sullivan, Circling Back to the Obvious: The Convergence of Traditional and Reverse Discrimination in Title VII Proof, 46 Wm. & Mary L. Rev. 1031, 1115–16 (2004) (“[L]iterally hundreds of cases recite some version of the slogan that courts do not sit as ‘super-personnel departments.’” (footnote omitted)); cf. Sandra F. Sperino, Disbelief Doctrines, 39 Berkeley J. Emp. & Lab. L. 231, 240 (2018) (“[C]ourts improperly use the super-personnel department justification to undergird their dismissal decisions in a wide-range of contexts.”). In education cases, courts likewise regularly state their reluctance to override the judgment of schools in internal disciplinary proceedings. See, e.g., Rosenthal v. N.Y. Univ., 482 F. App’x 609, 612 (2d Cir. 2012) (“Courts must defer to the university’s effort to substantially observe the rules, regulations, and procedures it has announced in advance, and will disturb their decisions only if their actions are arbitrary, irrational, or in bad faith.”). To the extent courts are asked to evaluate the meaning of particular remarks or conduct in context, they often shy away from what may seem like a sociological enterprise.183See Suzanne B. Goldberg, On Making Anti-Essentialist and Social Constructionist Arguments in Court, 81 Or. L. Rev. 629, 642–43 (2002) (exploring reluctance of judges to engage in what can be a sociological task of assessing dynamics in a work environment). Scholars have observed, too, that courts “lack the local knowledge to craft effective responses to the deep and complex equality problems that arise in individual workplaces.”184Bagenstos, supra note 16, at 25.

C. Options for Responding to Extra-Legal Harms

The similarity to legally impermissible conduct of low-grade harassment, in particular, presents special challenges for individual workers and organizations in deciding whether to report or respond to concerns. It may, for example, prompt potential complainants toward confusion or self-doubt (“is what happened bad enough?” “am I being too sensitive?”). It may also provide cover and reinforcement for those who seek to discriminate or harass (“just so long as I don’t cross the line” or “I didn’t do what Harvey Weinstein/Bill O’Reilly/others did so I’m fine.”).185See Rachel Sklar, Now Being ‘as Bad as Harvey Weinstein’ Means Being a Rapist, Wash. Post (Feb. 25, 2020, 6:17 PM), https://www.washingtonpost.com/​outlook/2020/02/25/now-being-bad-harvey-weinstein-means-being-rapist [https://perma.cc/ZVE6-BC4C] (“Since 2017, As Bad As Harvey Weinstein has been the unofficial bar for the ‘Should we ruin a man’s life over this?’ conversation about when misconduct—or worse—should be discussed publicly.”). Arguably, too, the expansiveness of legally permissible harassment may have led numerous corporate boards to discount liability risks and harms caused by “high-value” employees, endorsing a sliding scale that created opportunities not only for the Harvey Weinsteins and Bill O’Reillys of the world but also for senior executives who engage in disturbing but less obviously egregious harassment and lower-ranked purveyors of unwelcome comments and conduct.186Cf. Feldblum & Lipnic, supra note 31, at 86 (identifying risk factors in workplaces with high-value employees, with significant power disparities, and where compensation is tied to customer satisfaction); Alana Semuels, LowWage Workers Aren’t Getting Justice for Sexual Harassment, Atlantic (Dec. 27, 2017), https://www.theatlantic​.com/​business/archive/2017/12/low-wage-workers-sexual-harassment/549158 [https://perma.cc/A73T-XZZP].

These challenges generate difficult, though often-overlooked, questions for employers at the organizational-policy level. Even employers inclined toward a strict legal compliance approach must wrestle with how to draw lines between comments and actions that are permissible if they are occasional but not if they are frequent.187Cf. Faragher v. City of Boca Raton, 524 U.S. 775, 788 (1998) (affirming that “offhand comments, and isolated incidents (unless extremely serious) will not amount to discriminat[ion]”). And as discussed in Part III, they must communicate those lines clearly to employees and to those responsible for handling complaints so that workplace policy is not mistakenly—or deliberately—over- or under-enforced.

One response to this difficulty could be for legislatures and courts to broaden the scope of discrimination law to encompass more harassing conduct. Yet while many have argued,188See, e.g., Sperino & Thomas, supra note 16. and I would agree, that lines should be moved in that direction, the problems caused by low-grade harassment will not be resolved fully by doctrinal adjustment. Line-drawing at the margins is always challenging. With courts already reluctant to oversee workplace interactions, some amount of harassing and discriminatory conduct will remain permissible, with the ambiguity-generating effects just described.

This context further illuminates how employer responses to permissible and impermissible conduct creates an accountability culture that affects the broader workplace culture. As described earlier, an overly legalistic approach will miss the opportunity through policy and training to address local-level behaviors that do not generate liability risks but do cause harm to employees and create dysfunction on workplace teams.

This is not to suggest that employers should prohibit any behavior that an employee might find objectionable or require every misstep to be formally investigated and sanctioned. Workplace norms and culture are established in many ways, and a fully disciplinary approach may crowd out alternate responses to harassment and other harmful behavior.189See supra notes 95, 97 (discussing the interaction between compliance and other values that support ethical behavior). Also, as noted above, the risks of disciplinary overreach can be significant, particularly for individuals who are outside the mainstream of a given workplace. See supra note 97 and accompanying text. For early discussions of “crowding out” in the context of tradeoffs between economic and altruistic or other-regarding motivations, see Bruno S. Frey & Reto Jegen, Motivation Crowding Theory: A Survey of Empirical Evidence, 15 J. Econ. Surv. 589 (2001); Bruno S. Frey & Felix Oberholzer-Gee, The Cost of Price Incentives: An Empirical Analysis of Motivation Crowding-out, 87 Am. Econ. Rev. 746 (1997). For more recent discussion of crowding-out in diverse regulatory contexts, see, for example, Kristen Underhill, Money that Costs Too Much: Regulating Financial Incentives, 94 Ind. L.J. 1109, 1117–18 (2019). For discussion of the balance between self- and external regulation in the context of financial self-regulatory organizations, see William A. Birdthistle & M. Todd Henderson, Becoming a Fifth Branch, 99 Cornell L. Rev. 1, 3–5 (2013). Put another way, if formal law or policy violations are the sole or primary prompt for addressing workplace interactions, employers may miss opportunities to address low-grade harassment and may disincentivize employees, including supervisors, from stepping in to address concerns.

V. The Power and Limits of Law

Building on the landscape set out thus far, this Part takes a closer look at the power and limits of law to influence workplace culture in relation to harassment prevention and response. I focus especially on how legally driven interventions affect the local workplace norms and practices that shape most employees’ day-to-day experience. This is not to say that “tone at the top” is unimportant; to the contrary, employer messages can be a valuable way to set expectations for how employees will conduct themselves, as suggested by compliance literature and doctrine discussed above.190See supra notes 94–95 and accompanying text. Yet good policies and strong leadership statements do not necessarily translate to effective implementation on the ground, as shown repeatedly in #MeToo stories of corporations mismanaging harassment allegations.191See, e.g., supra notes 47–49 (discussing disregard of harassment allegations by large employers that had human resources staff and anti-harassment policies in place). Policies, trainings, and accessible reporting options similarly are unlikely to have their full intended effect unless they account for the individual-level barriers that inhibit many employees from disclosing misconduct or helping a colleague.192See supra Sections I.A.1–2. And all of these interventions require sustained attention and effort if they are to be integrated meaningfully into the workplace culture.193See supra notes 89–98 and accompanying text.

Given these challenges, it may seem that law has little to offer, especially relative to other kinds of employee-engagement strategies focused on creating an inclusive workplace.194For extended discussion of theory and practice toward advancing full participation, see generally Sturm, Activating Systemic Change Toward Full Participation, supra note 169, and Sturm, Architecture of Inclusion, supra note 22. Regulatory levers and policy-based incentives also have great potential to prompt organizational change, as Olatunde Johnson has shown. See, e.g., Olatunde C.A. Johnson, Equality Law Pluralism, 117 Colum. L. Rev. 1973 (2017); Johnson, supra note 157, at 1775–77. Still, for many employers, legal requirements will remain a primary touchstone and motivator for efforts to prevent and respond to sexual harassment.195This may be the result of resource constraints, limited capacity, or any number of other reasons, including that the governing law is an externally validated source of expectations for employees, clients, and other stakeholders. Yet there are also efforts to center the “business case” for employer attention to civility and nondiscrimination. See, e.g., Jennifer K. Brooke & Tom R. Tyler, Diversity and Corporate Performance: A Review of the Psychological Literature, 89 N.C. L. Rev. 715, 718, 726–34 (2011) (explaining the business case based on the benefits of diversity “lead[ing] to improvements in team functioning”). But see Jamillah Bowman Williams, Breaking down Bias: Legal Mandates vs. Corporate Interests, 92 Wash. L. Rev. 1473, 1479 (2017) (concluding that the “legal case [for diversity] is more effective than the business case” and discussing “potential drawbacks of instrumental diversity rationales”). With this in mind, the discussion that follows will examine several points of entry for legislative and doctrinal intervention.

Returning to the framework of Part I, I begin by looking briefly at doctrinal change as a potential influence on individual behavior, the surrounding environment, and employer choices regarding policies and practices. I then consider in depth whether and how regulatory interventions might prompt employers to integrate harassment prevention and response activities more fully and intentionally into workplace culture, concentrating on larger-scale observations about organizational policy and reporting systems, workplace structure, and other guideposts for employee interactions.

A. Doctrinal Change and Workplace Culture

Recall that an individual’s perception of law is only one of many influences on a person’s decision to report or confront harassment at work. Shame, fear, a limited sense of personal efficacy, reputational concerns, and many other factors are likely to be far more significant than legal requirements in determining how employees respond when they experience or observe harassment.196See supra Part I.

Against this backdrop, the question here is whether there is any room for law to shift behavior at the individual level.197For general discussion of law as a nudge to change behavior, see Richard H. Thaler & Cass R. Sunstein, Nudge: Improving Decisions About Health, Wealth, and Happiness (2008). As a thought experiment, we might imagine a world in which legal doctrine is communicated clearly enough for employees to understand when their employer is obligated to address harassing behavior.198Clear protection against retaliation would also be important. See supra notes 41–42 (recognizing the chilling effect of retaliation on reporting). With that clarity, we might expect some increased activity by those who currently hesitate to report or step in because of haziness about legal boundaries.199Prominent #MeToo settlements also may shift what individuals as well as employers and possibly courts understand to be impermissible harassment. See, e.g., Non-Monetary Relief at 2, City of Monroe Emps.’ Ret. Sys. v. Murdoch, No. 2017-0833-AGB (Del. Ch. Nov. 20, 2017), https://www.blbglaw.com/cases/twenty-first-century-fox-inc/_res/id=Attachments/​index=3/2017-11-20%20(EFILED)%​20Brief%​20to%​20Motion%20for%20Approval%​20of​%20Settlement.pdf [https://perma.cc/Q6LD-3S78] (expressing commitments to addressing all forms of harassment in settlement of a large shareholder derivative action arising from sexual harassment and race discrimination allegations at Fox News). I participated as an expert in crafting this agreement. On the other hand, the “severe or pervasive” standard in federal harassment law is likely to remain difficult to meet, even if it is relaxed somewhat, given courts’ reluctance to intervene in workplaces as described in Part IV.200But see supra note 175 (discussing the reduced threshold for harassment claims in California, Delaware, New York City and State, and Washington). Greater clarity about that high bar might then be an inhibitor to reporting, especially for employees who had assumed that all forms of harassment were prohibited.201But cf. Cynthia L. Estlund, How Wrong Are Employees About Their Rights, and Why Does It Matter?, 77 N.Y.U. L. Rev. 6, 16–17 (2002) (noting that “[r]elatively few employees benefit directly from wrongful discharge laws, but many more benefit from the precautions employers take to avoid litigation and liability”).

There is also little reason to think that more legal knowledge or a reduced threshold for claims will deter individuals whose harassing behavior stems from an anti-social personality disorder.202See supra note 63. Deeply held views about gender and power, prior life experience, and opportunities created by workplace hierarchies and physical isolation have been identified, too, as far more dominant influences than awareness of the state of the law.203See supra Section I.A.3. That said, there are those who take cues from their surrounding environment about what behaviors are acceptable.204See supra note 64 and accompanying text. It is conceivable that, for this group, clearer and more expansive doctrine will prompt stronger messages from leadership that employees should rein in any inclinations to harass. On the other hand, the doctrine is likely to continue to tolerate quite a lot of low-grade harassment, as just discussed, which may limit the power of those messages to prompt behavior change.

Still, workplaces do not exist in a vacuum and, as discussed in Part I, other law-related changes can influence what individuals tolerate from each other and expect in organizational policies and practices. Yet highly visible legal events, such as the convictions of Bill Cosby and Harvey Weinstein, are relatively few and far between.205See Bill Chappell, Bill Cosby Loses Appeal of Sexual Assault Conviction, NPR (Dec. 10, 2019), https://www.npr.org/2019/12/10/786678567/bill-cosby-loses-appeal-of-sexual-assault-conviction [https://perma.cc/TB8G-7XE5]; Jan Ransom, Harvey Weinstein’s Stunning Downfall: 23 Years in Prison, N.Y. Times (July 14, 2020), https://www.nytimes.com/2020/03/11/nyregion/harvey-weinstein-sentencing.html [https://perma.cc/HXA5-XDQY]. In most nationally prominent #MeToo stories, whether about media figures, comics, or celebrity chefs, legal doctrine has been on the sidelines.206For an illustrative list, see, for example, Audrey Carlsen et al., #MeToo Brought down 201 Powerful Men. Nearly Half of Their Replacements Are Women, N.Y. Times (Oct. 29, 2018), https://www.nytimes.com/interactive/2018/10/23/us/metoo-replacements.html [https://perma.cc/X2GE-UZJ2], noting that “a few” of the men listed faced criminal charges. When law does play a role, the focus has been largely on criminal offenses or employment termination packages (e.g. Les Moonves and CBS) and non-disclosure agreements (e.g. Fox News) rather than on changes to harassment jurisprudence that might influence workplace norms.207See Clare Duffy, Gretchen Carlson Fights Back Against Nondisclosure Agreements like the One She Signed with Fox News, CNN (Dec. 15, 2019), https://www.cnn.com/2019/12/15/media/gretchen-carlson-fox-news-nda-reliable-sources/index.html [https://perma.cc/ZG8H-2NYC]; Vanessa Romo, CBS Denies Former CEO Les Moonves $120 Million Severance Package, NPR (Dec. 17, 2018, 7:26 PM), https://www.npr.org/2018/12/17/677587813/cbs-denies-former-ceo-les-moonves-120-million-severance-package [https://perma.cc/F6EZ-4BK2]. Legal claims have also been used by individuals accused of sexual harassment to push back against accusations. See Anna Orso, In the Second Year of #MeToo, the Perpetrator Playbook Was Upended, Phila. Inquirer (Oct. 15, 2019), https://www.inquirer.com/news/me-too-second-anniversary-alleged-perpetrator-response-to-allegations-matt-lauer-20191015.html [https://perma.cc/H6FF-933L]. Indeed, most law-reform efforts in the wake of #MeToo have focused on mechanisms that affect the filing and settlement of lawsuits, such as statutes of limitations and non-disclosure agreements, or on reiterating responsibilities of corporate boards, rather than on fundamental questions of what conduct should be reached by discrimination law.208See Erik A. Christiansen, How Are the Laws Sparked by #MeToo Affecting Workplace Harassment?, A.B.A. (May 8, 2020), https://www.americanbar.org/​groups/​litigation/publications/litigation-news/featured-articles/2020​/new-state-laws-expand-workplace-protections-sexual-harassment-victims (describing legislative changes since 2017, most of which do not address the scope of sexual harassment claims, and “concerns that the new laws will do more to increase litigation than to actually reduce workplace harassment”). Even hearings on changes to the policy governing harassment complaints about members of Congress were about procedural fixes rather than the scope of conduct that would be prohibited. See, e.g., Emily Cochrane, Negotiators Strike Deal to Tighten Sexual Harassment Rules on Capitol Hill, N.Y. Times (Dec. 12, 2018), https://www.nytimes.com/2018​/12/12/us​/politics​/sexual-​harassment-capitol-hill.html [https://perma.cc/S8W5-KLMG]; Vivian Wang, New York Rewrites Harassment Laws, but Some Say the Changes Fall Short, N.Y. Times (Mar. 30, 2018), https://www.nytimes.com/2018/03/30/nyregion/new-york-revised-sexual-harassment-laws.html [https://perma.cc/8CYG-PJCA]. As is true for individuals, there is little reason to think that marginal transformations of harassment doctrine will permeate societal views in ways that reshape dynamics within workplaces.

B. Regulatory Interventions at the Organizational Level

Given the constraints on doctrinal change, a pressing question, then, is whether other legal interventions can encourage employers to move beyond the minimum required for litigation defense and develop a workplace culture that deters harassment and supports reporting and effective resolution when incidents occur.209For discussion in the antidiscrimination context of actions that go beyond the judicially required minimum, see Olatunde C.A. Johnson, Beyond the Private Attorney General: Equality Directives in American Law, 87 N.Y.U. L. Rev. 1339, 1387, 1390–92 (2012), discussing the requirement that government “affirmatively further” nondiscrimination objectives in transportation, housing, and other areas. Several of the organizational touchstones considered in Part III—policy, reporting systems, and training—are potential entry points, as are broader employer actions regarding organizational structure and workplace culture.

In essence, I argue that regulatory intervention related to policies and reporting systems is helpful but unlikely to have a substantial influence on workplace dynamics. Structural changes to an organization, by contrast, could provide meaningful protection but are largely unreachable by legal interventions. If law is to exert influence on workplace dynamics, workplace culture becomes a key site, and the discussion below will focus most attention on the possibilities there, including training, data-sharing, and more.

1. Policy and reporting systems

Employer policies, as a version of workplace “law,” are relatively straightforward to regulate. A government agency could require employers to have clear, protective policy language and post an employee bill of rights or other information about sexual harassment protections.210See infra text accompanying notes 236–41 (discussing revisions to New York City’s sexual harassment law). Taking guidance from the discussion in Part III, the agency might even recommend that employers have a regular policy-review process to seek employee input.

At the end of the day, though, policies in handbooks and on websites—and even on office bulletin boards—have not shown themselves to be a significant force in deterring sexual harassment.211See supra notes 102–04 (discussing sexual harassment allegations against large corporate firms that had protective policies and other corporate culture commitments in place). This is not to say that policies are unimportant. To the contrary, they can express important employer values and commitments, and posting them can provide public reinforcement.212It bears noting that many employers also include disclaimers in handbooks providing that their strong, employee-focused commitments do not establish legal rights. See generally Stephen F. Befort, Employee Handbooks and the Legal Effect of Disclaimers, 13 Indus. Rel. L.J. 326 (1992) (discussing the use and validity of disclaimers in employee handbooks). But as a singular feature in a workplace where most employees are unlikely to read the policy in detail (even if it is hanging on the office wall),213Cf. J. Bret Becton et al., Preventing and Correcting Workplace Harassment: Guidelines for Employers, 60 Bus. Horizons 101, 103–04 (2017) (describing policy dissemination methods and recommending that employers require their employees to “acknowledge that they have received, read, and understood the policy” as a best practice). much less bring formal complaints,214See supra Section I.A.1 (outlining individual-level barriers to reporting sexual harassment). organizational policy language is an improbable driver of change. Consequently, while legally mandated changes to harassment policies and agency-imposed posting requirements can strengthen basic protections, their impact on workplace culture is likely to be minimal.

An employer’s procedures for receiving and responding to complaints are arguably a more effective intervention point because their function is to bring problems to the employer’s attention. Having accessible and understandable reporting systems can be a positive counterweight to the many individual-level factors that may inhibit an employee from seeking help.

With this in mind, regulators might require employers to post complaint procedures widely, adopt or expand anonymous compliance hotlines, and highlight government resources as complements or alternatives to the employer’s systems.215See infra notes 236–41 and accompanying text (discussing analogous New York City provisions). Some states have imposed similar requirements on educational institutions. See, e.g., N.Y. Educ. Law §§ 6439–6449 (McKinney 2015) (codified as Article 129-B of Title VII of the New York Education Law). They could impose penalties for anonymous reporting lines that turn out not to be anonymous,216Cf. Noam Scheiber, Anonymous Harassment Hotlines Are Hard to Find and Harder to Trust, N.Y. Times (Apr. 21, 2017), https://www.nytimes.com/​2017/04/21/business/media/fox-sexual-harassment-hotline-bill-oreilly.html [https://perma.cc/JQ2M-3ZSZ]. and they could require employers to explain their procedures in multiple languages where appropriate.

These kinds of changes might influence the workplace culture by signaling the employer’s interest in hearing about concerns, though much will depend on the effectiveness of the employer’s response. Ultimately, though, complaint procedures, like policies, tend to be relatively static. As a result, they are not front of mind for most employees and changes to them are unlikely to influence daily workplace dynamics.217See Grossman, supra note 131, at 23–24 (observing that even where improvements were made to policies, reporting statistics remained relatively unchanged). It is conceivable that regulatory support for fair processes that include a meaningful opportunity to be heard, an impartial decisionmaker, and anti-retaliation and other protections discussed earlier, might engender greater trust and in turn prompt more use of the complaint system. See, e.g., Bell et al., supra note 35, at 147 (describing positive benefits of processes that are responsive to complainants’ concerns).

D. Workplace Structure

By contrast, workplace structure is a factor that has daily impact. Individuals working in small groups where power differentials are significant, or in isolated or highly decentralized workplaces, or in workplaces where heavy alcohol consumption is the norm tend to be more vulnerable to sexual harassment than workers in other settings.218See Fed. Judiciary Workplace Conduct Working Grp., supra note 72, at 3–4; Feldblum & Lipnic, supra note 31, at 25–29; Johnson et al., supra note 72, at 169, 172–73, 176–77. Yet it is difficult to imagine any branch of government mandating that an employer change its business structure to reduce these risks.219See supra notes 182–84 and accompanying text (discussing courts’ reluctance to review employers’ business decisions).

Still, the idea of government intervention in work environments has some support from existing practices. Health and safety regulations direct certain aspects of how employers structure their workplaces, and national security regulations control others.220See Adam M. Finkel et al., The NFL as a Workplace: The Prospect of Applying Occupational Health and Safety Law to Protect NFL Workers, 60 Ariz. L. Rev. 291 (2018) (describing the Occupational Health and Safety Act and its potential applicability to football players); Leslie Gielow Jacobs, A Troubling Equation in Contracts for Government Funded Scientific Research: “Sensitive but Unclassified”—Secret but Unconstitutional, 1 J. Nat’l Security L. & Pol’y 113, 126 n.65 (2005) (describing regulations that require research universities to segregate classified research from other research and noting that “several universities have stand-alone facilities for classified research” and that “a common premise is that a stand-alone facility is easier to protect than an on-campus laboratory or building” (internal punctuation and citation omitted)). One might argue that sexual harassment is an occupational health and safety hazard and should be treated as such,221See Catharine A. Mackinnon, Sexual Harassment of Working Women 159 n.48 (1979); Anita Bernstein, Law, Culture, and Harassment, 142 U. Pa. L. Rev. 1227, 1292–93 (1994) (elaborating MacKinnon’s argument about the applicability of the Occupational Safety and Health Act to sexual harassment). and that it is also a corruption-like abuse of power that could, in some instances, pose national security risks.222Corruption is frequently defined as the “misuse of public power for private gain.” Susan Rose-Ackerman, Corruption and Government: Causes, Consequences, and Reform 91 (1999); see also Bernadette Atuahene & Timothy R. Hodge, Stategraft, 91 S. Cal. L. Rev. 263, 295 (2018) (“A widely agreed-upon definition of corruption is ‘[t]he abuse of an entrusted power for private gain’” (citations omitted)). Sexual harassment meets this definition and can have similarly harmful consequences in and outside of work environments. I will develop this point further in future work. On national security risks associated with sexual harassment, see, for example, Maya Rhodan, ‘We, Too, Are Survivors.’ 223 Women in National Security Sign Open Letter on Sexual Harassment, Time (Dec. 1, 2017, 12:52 PM), https://time.com/5039104/we-too-are-survivors-223-women-in-national-​security-sign-open-letter-on-sexual-harassment/#d70876f5-3b56-4489-8eba-49cadc10bd3a [https://perma.cc/UKR2-6H6S].

But the structural changes that might make a difference here—such as having teams of supervisors instead of individualized oversight; restrictions on alcohol consumption in entertainment businesses; video monitoring or other alterations to physical spaces where abusive behavior is more likely—are more nuanced than the standardized safety masks, rubber gloves, and other requirements typically in place for managing non-human workplace hazards or the detailed background checks and extra layers of cyber safety for managing security risks related to sensitive data and analysis. As a result, while it is interesting to consider how existing regulatory structures might carry over to sexual harassment prevention efforts, regulation to achieve structural change in the diversity of American workplaces is unlikely to gain traction.223Sector-based efforts, however, might have promise in proposing self-regulation to address risk factors particular to those environments. For example, the National Academies of Sciences, Engineering, and Medicine generated an extensive report on how higher-education institutions might reduce the incidence of sexual harassment in academia, Johnson et al., supra note 72, at 149, and some professions have developed specific rules and practices to address sexual harassment at conferences. See, e.g., Code of Conduct for UNFCC Conferences, Meetings and Events, United Nations Framework Convention on Climate Change (UNFCCC), https://unfccc.int/about-us/code-of-conduct-for-unfccc-conferences-meetings-and-events (last visited Dec. 21, 2020); Allison Torres Burtka, How Associations Can Help End Sexual Harassment, ASAE (Mar./Apr. 2018), https://www.asaecenter.org/​resources/articles/an_magazine/2018/march-april/how-associations-can-help-end-sexual-harassment [https://perma.cc/7EYH-GNHT] (recommending codes of conduct for meetings).

E. Workplace Culture

This brings us to workplace culture. But culture, whether in or outside a workplace, is not an obvious site for legal intervention. Statutes and case law, with their ambiguity and distance from interpersonal interactions, are unlikely to affect, except in the most attenuated ways, the shared understandings, values, and behaviors among an organization’s employees.224See Linstead, supra note 4, at 10930; see also supra Part I (reviewing extra-legal barriers to reporting harassment). Even further, regulation risks generating the type of culture that focuses on compliance at the cost of attention to the on-the-ground employee experience.

Still, thoughtfully crafted regulation has potential to support an anti-harassment workplace culture by linking legal accountability with employee engagement through the kinds of questions set out in Part III. Requirements that employers not only conduct regular trainings but also evaluate their effectiveness, assess employees’ experience with complaint processes, and report publicly, in the aggregate, on their handling of harassment complaints all put a focus on how employees understand and experience the employer’s efforts. Engaging the workforce in these ways may help convey that sexual harassment prevention and response is more than just a compliance obligation for the organization’s human resources or legal department. Settlements of sexual harassment suits can serve a similar function, binding employers to these and other modes of ensuring that employees have sufficient information about workplace policies and resources and linking those policies directly to organization values.225See supra note 199 for discussion of provisions in the settlement of the Fox News shareholder derivative action. For more on the settlement, see Jonathan Stempel, 21st Century Fox in $90 Million Settlement Tied to Sexual Harassment Scandal, Reuters (Nov. 20, 2017), https://www.reuters.com/​article/​us-fox-settlement​/21st-century-fox-in-90-million-settlement-tied-to-sexual-harassment-scandal-idUSKBN1DK2NI.

Before elaborating, I want to be clear about the limits of the claim here. Employer trainings, surveys, and publications are not considered generally to be behavior-change tools.226Research has been mixed on whether employer-sponsored training can reduce the incidence of harassing behavior, and many scholars have expressed skepticism. See, e.g., Bisom-Rapp, An Ounce of Prevention, supra note 132, at 6 (describing trainings as “symbolic gestures of employers”); Grossman, supra note 131, at 42–49 (reviewing research on harassment training showing mixed results regarding effectiveness). But see JoAnna Suriani, Reasonable Care to Prevent and Correct: Examining the Role of Training in Workplace Harassment Law, 21 N.Y.U. J. Legis. & Pub. Pol’y 801, 813–19 (2018) (discussing harassment prevention training, including findings that “training increases knowledge [about] harassing conduct,” that “social scientists have been unable to conclude what effect different types of training have on participants’ behaviors,” and that “certain training methods have been shown to hold more promise than others”). Diversity training has been the subject of more extensive and critical study. See, e.g., Frank Dobbin & Alexandra Kalev, Why Doesn’t Diversity Training Work? The Challenge for Industry and Academia, 10 Anthropology Now 48, 49–51 (2018) (identifying multiple limitations of diversity training, including that short-term educational interventions, in general, do not change people’s behavior; anti-bias training may activate stereotypes; and training may make employees complacent). There is also the risk, and the reality in many settings, that these requirements will be managed like other compliance requirements—by doing the minimum necessary to avoid penalty. One could argue further that this type of box-checking does more harm than good because a begrudging fulfillment of the obligation may provide an extra signal of the employer’s disinterest. Even if senior management carries out its responsibility with neutrality, it is possible that training will be administered at lower levels of the organization in ways that exacerbate ongoing harassment or other local vulnerabilities. It is similarly true that surveys might be administered in ways that deter disclosures and that entities might comply with reporting requirements in ways that obscure concerns.

Still, I want to suggest that even weakly delivered training and limited surveys and reporting may be promising, even if unglamorous, sites for law to exert influence on workplace culture. I will turn first to training requirements and then address survey and reporting requirements more briefly.

First, consider a workplace where employees are trained only on arrival if at all and the employer addresses questions about its harassment policy and procedures individually and with one-off responses. A training mandate sets a new anchoring point, requiring a company-wide plan and an eye toward prevention, even for entities that do the minimum to comply.227A regulatory requirement often prompts the emergence of sector-specific offerings to facilitate compliance “solutions,” increasing the likelihood that employers will be able to meet this new obligation effectively. By providing its own training as a model, the regulator can reduce costs for employers and present the required material with the tone and framing that are consistent with its prevention-oriented goals. See infra note 235 and accompanying text. On the costs of compliance regimes generally, see Reinier H. Kraakman, Corporate Liability Strategies and the Costs of Legal Controls, 93 Yale L.J. 857, 887–88 (1984).

In addition, a requirement can provide a foothold for employees who are interested in addressing the effects of harassment on workplace culture.228See Sturm, Activating Systemic Change Toward Full Participation, supra note 169, at 1127–29 (explaining the role of “institutional intermediaries” in pressing organizations to address structural problems); Sturm, The Architecture of Inclusion, supra note 22, at 298 (discussing how employees can be “organizational catalysts” in advocating for gender equity in the workplace). Messages from trainings might reinforce efforts that employees are already making from within, and some employers may seek employee volunteers to assist with implementation or evaluation. Trainings, especially in groups, also can provide a vehicle for employees to connect with coworkers (even if to complain about what was offered) and possibly to advocate for additional change.229See McDonald et al., supra note 39, at 45 (reviewing research showing that group trainings have been found to be more effective than individual trainings).

Importantly too, even weak training can embed knowledge locally—if all employees are trained and know that others are trained as well, the workforce’s information baseline expands. And while training may not prompt behavior change, it can increase the likelihood that more people notice when behaviors are impermissible or close to the line. This diffusion of information about how to address problems may help reduce individual-level barriers to reporting and provide a reference point for stepping in (e.g. “you remember what they said in that training”).230See supra note 136 and accompanying text (characterizing shared knowledge as a protective factor analogous to herd immunity). Even when awareness does not carry over into action, particularly in workplaces with low commitment or capacity to strengthen the overarching workplace culture as described in Part II, training requirements may provide a starting point that makes those next steps possible.

By surfacing issues related to employee misconduct, a required training can also put a fresh focus on employers’ systems for receiving and responding to complaints and other workplace conflicts. At its best, this additional attention may lead to improved systems and possibly an interest in strengthening conflict-resolution skills for managers and others in the organization. In this sense, a training requirement has direct benefits for cultural awareness and potential to catalyze additional attention and action in a workplace that otherwise handles these issues through a compliance lens.

While training may shape culture by funneling information about policies and resources directly into the workplace, employee surveys and public reporting generate data that can affect workplace culture by exposing difficult, or positive, dynamics that may not be picked up through the formal complaint process. Whether legally required or strongly encouraged as governmentally endorsed best practices, these methods follow in the tradition of information-forcing techniques that help expose troubling patterns and shift practices.231On the use of information-forcing as a technique in employment contexts, see Cynthia Estlund, Just the Facts: The Case for Workplace Transparency, 63 Stan. L. Rev. 351, 364–65 (2011). For discussion of information-forcing strategies more generally, see, for example, Richard H. Pildes & Cass R. Sunstein, Reinventing the Regulatory State, 62 U. Chi. L. Rev. 1, 102–12 (1995). In Britain, for example, companies with more than 250 employees are required to report information about compensation by sex, and while there are critiques of the requirement and its implementation, even limited data enables more rigorous evaluation and advocacy than when information is purely anecdotal.232For media coverage of the reporting requirement, see, for example, Liz Alderman, Britain Aims to Close Gender Pay Gap with Transparency and Shame, N.Y. Times (Apr. 4, 2018), https://www.nytimes.com/2018/04/04/business/britain-gender-pay-gap.html [https://perma.cc/6NRB-LECK]; Frances Perraudin, What Is Gender Pay Gap Reporting, and What Does It Mean?, Guardian (Feb. 28, 2019, 3:40 PM), https://www.theguardian.com/society/2019/feb/28/what-is-gender-pay-gap-reporting-and-what-does-it-mean [https://perma.cc/75YN-PFMX]. Several other countries have similar disclosure requirements. See Christine Aymayr-Pintar, Eurofound, Pay Transparency in Europe: First Experiences with Gender Pay Reports and Audits in Four Member States (Nov. 4 2020), https://www.eurofound.europa.eu/sites/default/files/ef_publication/field_ef_document/ef18004en.pdf [https://perma.cc/BB3Z-EACH] (reporting on company-level gender pay reports and audits in Austria, Denmark, Sweden, and Finland). An even closer example comes from higher education in the United States, where some states require schools to publish information about sexual misconduct complaints and resolutions.233See, e.g., N.Y. Educ. Law §§ 6439–6449 (McKinney 2015). The Obama administration’s White House Task Force to Protect Students from Sexual Assault also urged colleges and universities to conduct climate surveys and gather data on campus sexual misconduct. See Michelle J. Anderson, Campus Sexual Assault Adjudication and Resistance to Reform, 125 Yale L.J. 1940, 1970 (2016). This recommendation was later put into official guidance for Title IX Coordinators. See Deborah Tuerkheimer, Beyond #MeToo, 94 N.Y.U. L. Rev. 1146, 1200 n.331 (2019) (citing Bea Hanson, Best Practices: Campus Climate Surveys, Dep’t Just. (Oct. 14, 2016), https://www.justice.gov/archives/ovw/blog/best-practices-campus-climate-surveys [https://perma.cc/M8ZL-A4QV])). Neither the British nor the U.S. requirement has a direct tie to culture-change efforts but both reflect an understanding that addressing harassment and other discrimination requires more than legal compliance aimed at the most severe or pervasive misconduct.

F. Workplace-Culture Interventions, as Applied

Regulatory requirements of training, surveys, and public reports are blunt and often frustrating instruments, even with their potential to reduce workplace harassment and improve employer response. They impose costs in time and funds spent on compliance, often through cottage industries of trainers and survey administrators, that may be consequential for small or struggling enterprises.234See Michael W. Johnson, Harassment and Discrimination Prevention Training: What the Law Requires, 55 Lab. L.J. 119, 126 (2004) (discussing the need for trainers to “completely understand the complex body of harassment and discrimination laws” and “keep up-to-date with new cases that constantly change the interpretations of these laws” at a minimum); see also Robert K. Robinson et al., U.S. Sexual Harassment Law: Implications for Small Businesses, 36 J. Small Bus. Mgmt. 1, 4–5 (1998) (observing the impact of regulations on small businesses where “time and financial constraints have precluded most small firms from implementing either policies or education and training programs”). They are typically not sensitive to the contexts of diverse sectors, some of which already may face heavy training requirements and others that have employees who are not online or are otherwise not gathered easily, making information delivery difficult. They also have the potential to reinforce a compliance orientation toward harassment, which may be in tension with creating a culture of collegiality and trust among coworkers that itself deters harassment and encourages reporting.

These costs can be mitigated, however, by government actors working in partnership with employers. By providing model trainings, surveys, and reporting templates that are responsive to diverse sectors, agencies charged with addressing discrimination can contribute proactively to broader workplace-culture change, which traditional enforcement actions are less likely to accomplish.235Cf. Johnson, supra note 194, at 1993 (advocating use of the “fuller range of public law and private law regulatory tools” to address inequality). The potential for agencies proactively to provide resources to support compliance bears underscoring because it is often overshadowed by agencies’ investigation and enforcement authority. Public-private partnerships may face challenges in this area as compared to those focused on business development or historical preservation, for example, because civil-rights agencies that would work with employers to develop materials are also responsible for enforcement actions, including against potential partners. But if this dual role can be managed, there is tremendous potential for collaboration on content and delivery mechanisms.

The New York City Human Rights Commission’s work on sexual harassment provides a helpful illustration. In the wake of #MeToo disclosures and media coverage, the New York City Council enacted an ordinance requiring annual sexual harassment training.236N.Y.C. Local Law No. 96 of 2018, Council Int. No. 632-A of 2018, https://www.nyc.gov/assets/cchr/downloads/pdf​/amendments/​Local_Law_96.pdf [https://perma.cc/MCW5-M2YM]. The ordinance specifies that the training be interactive, with examples, and include coverage of both legal terms and workplace-culture concepts like bystander intervention.237Id. Notably for purposes here, the ordinance also mandates that the City’s Commission on Human Rights create and offer at no charge an online training module that would enable compliance, so long as the employer also provides employees with information about how to bring sexual harassment claims in their own workplace.238Id. (“The commission shall develop an online interactive training module that may be used by an employer as an option to satisfy the requirements of [the ordinance], provided that an employer shall inform all employees of any internal complaint process available to employees through their employer to address sexual harassment claims. Such training module shall be made publicly available at no cost on the commission’s website. Such training module shall allow for the electronic provision of certification each time any such module is accessed and completed. The commission shall update such modules as needed.”). The ordinance does not require that employers use this training, characterizing it instead as “a minimum threshold” that “shall not be construed to prohibit any private employer from providing more frequent or additional anti-sexual harassment training.”239Id. The Commission previously had held hearings in which it received testimony from a variety of employees and employers along with experts on addressing sexual harassment and other workplace challenges.240See N.Y.C. Comm’n on Human Rights & the Sexuality & Gender Law Clinic at Columbia Law Sch., Combating Sexual Harassment in the Workplace: Trends and Recommendations Based on 2017 Public Hearing Testimony (2018), https://www.nyc.gov/assets/cchr/downloads/​pdf/SexHarass_Report.pdf [https://perma.cc/JKA6-9YD2]. This provided a foundation for the diverse situations and sectors that are depicted in the city-created training, increasing the likelihood that the training would be relevant to many different local environments. Importantly, too, this training is part of a broader city-sponsored campaign to address sexual harassment, which also includes workplace-posting requirements and a social media and public advertising campaign designed to reinforce an anti-harassment norm.241Sexual Harassment in the Workplace, N.Y.C. Hum. Rts., https://www.nyc.gov/site/cchr/media/sexual-harassment-campaign.page [https://perma.cc/JG43-53NB]. As a result, neither the posting nor the training requirement functions as an isolated effort, likely to be disregarded or forgotten. Each, instead, is part of an expansive effort directed at cultural expectations as well as policy and law enforcement.

At the federal level, an education initiative called It’s On Us provides another interesting example. The project, which encourages students and others to pledge to raise awareness and fight against sexual assault, originated in the Obama administration, after the White House Task Force to Prevent Sexual Assault reported on the need for broad-based engagement efforts.242The Story of Our Movement, It’s on Us, https://www.itsonus.org/history [https://perma.cc/A34E-EACS]. It’s on Us has since transitioned into a freestanding not-for-profit organization. Id. Unlike the New York City campaign, which was linked closely to compliance requirements, this effort derived its momentum from a social movement that had arisen around campus sexual assault and relied on celebrities as well as campus-based organizations to spread the word. But similar to the City’s effort, the federal It’s On Us initiative provided easy-access materials for students and schools to generate or support climate-oriented work on campus.243It’s on Us, Campus Organizing Tools (2019), https://www.itsonus.org/​wp-content/​uploads/​2019/07/IOU_​CampaignOrganizingTools​_7.25.19.pdf [https://perma.cc/3QZX-8Q8F].

While development of toolkits, training materials, social media campaigns, and public-service announcements is often the domain of not-for-profit organizations working with their constituents, these are also vehicles available to governments that seek to further law’s impact in difficult-to-reach contexts. With respect to sexual harassment, in particular, these initiatives may be among the most important ways to support changes in workplace culture.


When we take full account of the individual, organizational, and societal factors that influence an employee’s experience at work, it is unsurprising that harassment, especially sexual harassment, persists in many workplaces. With traditional sources of law having had limited impact, the increase in attention to workplace culture as a source of change appears promising, especially as a way to address the day-to-day “local” interactions that shape employees’ experiences.244See supra notes 140–41 and accompanying text. Yet law remains influential because employers will continue to face choices about legal accountability as they develop and implement workplace policies.245See supra note 12 and accompanying text.

The problem is that an approach delinking legal accountability from workplace culture misses the ways in which choices about compliance can create additional barriers to effective harassment prevention and response. As I have argued here, harassment policy, communications, and trainings are elements of culture-creation, not just protection against liability. When they are generated and implemented through a compliance lens, employers may satisfy their general counsel but will be unlikely to improve the experience of employees, except perhaps at the margins.

By contrast, an approach that makes the employee experience its focal point will use policy and training requirements to create a more seamless set of cultural expectations for how employees will interact. This reframe of legal accountability—from obligation to an opportunity for culture-creation—recognizes that, for the targeted employee, harassing behavior will have a negative impact on productivity even if the behavior would not give rise to liability. This becomes apparent when considering low-grade harassment, as described above, as well as microaggressions and manifestations of implicit bias that do not expose a firm to liability but most certainly have consequences for employees on the receiving end.246See supra Section IV.B (“The ‘law’ of extra-legal harms.”).

This is not to say that an employer must respond in an identical way to all workplace interactions that an individual employee experiences as harassing. But if legal accountability is the sole driver, severe or pervasive misconduct will take center stage. By not declaring values and setting expectations about low-grade harassment and other behaviors that cause extra-legal harms, employees may think twice about whether what they experienced or observed was serious enough to be of interest to the employer. Given the many pre-existing barriers to reporting discussed at the outset of this Article, the result is likely to be a workplace culture in which doubts are resolved in favor of keeping quiet.

Still, even with its doctrinal limitations, law remains a powerful tool for addressing workplace harassment so long as enforcing agencies and employers take account of how legal-accountability choices shape workplace culture. Regulatory interventions can be crafted to require diffusion of policy information, interactive trainings, and disclosures that go beyond traditional check-the-box requirements.247See supra Part III, Section V.B. Even more important, government agencies that generate and enforce these measures can create templates and modules that enable compliance while being attentive to their effects on workplace culture. Through convenings, trainings, and publications, these agencies can also create opportunities for employers to assist each other in integrating their legal-accountability efforts with broader attention to how those efforts are experienced by employees. When properly understood and deployed, legal-accountability requirements thus hold great promise not only to punish the worst forms of misconduct but also to support a workplace culture that rejects harassment as outlier behavior and contrary to the organization’s shared values and commitments.