*This paper was presented at the 2018 ASSA Conference, hosted by the American Economics Association.
This violence…sits on our backs like oppressive weights
Now that the Harvey Weinstein revelations have seared into public consciousness, the country has been thrown into a turbulent process of confronting and rapidly reassessing an insidious form of abuse that haunts American workers, often silently and mostly unchecked. As targets have come forward to reveal how sexual harassment has blighted and shaped their working lives and economic trajectories, we are reconsidering how to value and measure the experiences of those who have been targeted and beginning to reckon the costs to the broader economy and society.
The cultural shifts and new information appearing in connection to the #MeToo movement provide a golden opportunity for social scientists to do the critical work of illuminating the vast economic dimensions of a mammoth problem that has so far lumbered largely under the radar. Researchers, particularly in the male-dominated field of economics, also have a chance to synchronize their work with the needs and concerns of the broader society and to help rectify wrongs — not the least of which are problems in their own arenas.
What is sexual harassment?
I wasn’t even a person…I was a thing: not a nobody, but a body. – Salma Hayek, actor
While the reality is of course much older, the law only caught up to sexual harassment a few decades ago — slowly and by fits and starts. 1964 marked a leap forward in civil rights and workplace law in the U.S. with the passage of the Civil Rights Act, with Title VII laying down prohibitions on employment discrimination on the basis of race, color, religion, sex or national origin. The Equal Employment Opportunity Commission (EEOC) was created to enforce and administer the law, but federal courts typically refused to view sexual harassment as a form of employment discrimination until the 1980s, when it came to be perceived under the umbrella of gender discrimination.
Under the current federal legal framework, sexual harassment is defined as “unwelcome verbal, visual, or physical conduct of a sexual nature that is severe or pervasive and affects working conditions or creates a hostile work environment.” Activities can take myriad forms: a perpetrator may use physical violence as well as coercion or the creation of a workplace made hostile by such conduct as unwelcome sexual language or unwanted touching. But precision is tricky: as Stanford Law School professor Deborah Rhode has noted, interpretations of what constitutes harassment vary widely. What exactly is meant by “hostile?” What constitutes “unwelcome”? Some courts use the so-called “reasonable woman standard” to determine if the defendant’s behavior would be considered sexual harassment by a fictitious, reasonable female, a framework developed to help correct for the tendencies of men to perceive harassment differently (and less frequently) than women. Others use the older “reasonable man” or “reasonable person” standard. Confusion abounds.
There is certainly a need for more extensive and finely tuned assessments of prevalence, but polls and studies across a variety of settings and contexts suggest that sexual harassment in the American workplace is common. In 2013, the combined number of sexual harassment charges filed with the EEOC and state and local Fair Employment Practices Agencies was over 10,000, but the 2016 EEOC task force report on the topic notes that the problem is underreported. It gives a range of 25-85 percent of women who have experienced sexual harassment on the job. The agency attributed the broadness of the range to a number of factors, including how questions were asked, how sexual harassment was defined, and the type of sample. Among surveys and polls distributed across research literature and journalistic accounts, a range of one quarter to one third of workers seems fairly consistent.
For example, a recent poll conducted by MSN, in partnership with Business Insider, revealed that one in three people (31 percent) in the U.S. admitted to having experienced sexual harassment on the job. Altogether, 45 percent of women surveyed cited sexual harassment problems at work — around 33.6 million women. The group experiencing the most abuse was women between the ages 30 and 44, of whom close to half (49 percent) said they had undergone unwanted sexual attention while working. 47 percent of women ages 45 to 64 said they were targeted, followed by 41 percent of women ages 18 to 20, and 40 percent of women 65 or older. In addition, 15 percent of men said they had experienced workplace sexual harassment.
Data compiled by the EEOC reveals that half of all formal claims of sexual harassment in the U.S. do not result in any charges. Cases are hard to win, and victims may be afraid to come forward, anticipating defamation lawsuits, intimidation, and the burden of coming up with corroborating witnesses. Time is not on the side of targets: the EEOC reports that the average wait time for a complaint to be addressed was 295 days in 2017. That problem is likely to get worse: the agency has seen a surge in complaints filed in the wake of the Weinstein revelations, but its budget is expected to shrink in the current political climate.
Violent sexual assault, an extreme form of sexual harassment, also happens in the workplace, with immigrant women and undocumented workers at particularly high risk. A recent open letter written by Alianza Nacional de Campesinas, the first national women’s organization of farm workers, testifies to the prevalence and violence of sexual harassment in the industry. The Bureau of Justice Statistics reports that from 1993 to 1999, workplace rapes and sexual assaults numbered 36,500, though it is safe to say that this number falls significantly short of actual occurrences.
Where does sexual harassment occur?
Upper management told me that he “was a high performer”… and they wouldn’t feel comfortable punishing him for what was probably just an innocent mistake on his part. – Susan Fowler, former Uber employee
Workers across the socio-economic spectrum seem to run a gauntlet of sexual abuse in exchange for the privilege of earning a living. It happens from the top to the bottom of the workplace hierarchy and across all industries and sectors.
Certain risk factors appear to be common in workplaces with high rates. In 2015, the EEOC found that it occurred most frequently in situations where diversity is lacking among employees and also those where workers are highly diverse but segregated across job types. Problems showed up more often when there is a star in the workplace who flouts the rules, as well as those with high numbers of young or immigrant workers or monotonous duties. Places where customer service and client recommendations are key to job performance and tied to compensation were more likely to have higher rates of sexual harassment, as did jobs where workers perform duties in isolated places where management is far away and those where alcohol is flowing.
According to an analysis by Jocelyn Frye of new, unpublished data collected by the EEOC, sexual harassment appears to happen more frequently in industries dominated by low-wage workers, with minority women working in services industries especially vulnerable. In the service-related industries, especially food service and retail, women filed more than three times as many claims as those in higher-paying fields such as finance and insurance. The EEOC data reveals that between 2005 and 2015, around 25 percent of complaints came from the service sector. A new poll by Hart Research indicates that 40 percent of females working in the fast food industry experience sexual harassment — abuse they often feel they have no way to combat. Women surveyed often responded to the harassment by cutting back hours, changing their shifts, or quitting. But 42 percent said they felt compelled to accept it out of fear of losing jobs they could not afford to lose. These are sectors that disproportionately hire women of color, who may face additional burdens of racial stereotyping and lack of support systems if they speak out against harassment.
Too many workers in today’s precarious job market can’t count on solid protections from sexual harassment: The EEOC’s authority only covers workplaces where there are 15 or more employees, which leaves many workers, notably domestic workers like house cleaners or home healthcare aids stuck with inadequate channels for reporting abuse. Beginning in 2010, eight states passed laws known as the Domestic Worker Bill of Rights, which include protections from sexual harassment, but that leaves 42 states where workers may face unclear and burdensome processes for filing charges under various state statues involving harassment and discrimination.
Freelancers and workers in the gig economy also find sexual harassment difficult to navigate. While a few states, such as California, have protections against sexual harassment for independent contractors, most do not offer targets clear pathways for addressing abuse.
Blue-collar workers face their own sexual harassment challenges. A recent New York Times account drawing on information from the EEOC, employment lawyers, academics, and employees found that women in blue-collar jobs, a sector where the share of women has fallen, face rampant sexual harassment and may view it as a condition of employment. They may face particular risks of physical danger when they are targeted as well as ambivalent responses from unions and blacklisting in their industry if they report misconduct. The impact of sexual harassment may hold back lower-income women from reaching the middle class.
Union responses to sexual harassment have been insufficient and in some cases counterproductive. Among the revelations of the recent exposure of pervasive sexual harassment at Ford Motor Company factories in Chicago were allegations that the United Automobile Workers union not only discouraged women from filing complaints but that some union representatives actually harassed women themselves. Labor organizer and author Jane McAlevey has highlighted troubling patterns within the labor movement of sexist, predominantly white male leadership that hinders progress. Such imbalances, she writes, lead to an emphasis on protecting male workers alleged to harass, pervasive disregard for sectors like education, healthcare, and the public sector where women predominate, and a lack of support for organizing strategies that would help both unionized and non-union workers targeted by sexual harassment.
In the corporate world, patterns of abuse are often hidden behind a wall of obfuscation. In Hollywood, the Weinstein Company’s use of nondisclosure agreements to silence targets has illuminated a problem that stretches across many industries where women from middle and higher socioeconomic strata can be found, including Silicon Valley and Wall Street, where various nondisclosure agreements, forced arbitration agreements (a trend driven by a series of Supreme Court decisions dating back to 1991 and now affecting over half of all workers), and confidentiality provisions have helped cast a shroud of secrecy over sexual harassment and allowed predators to stay in their jobs. For example, recent revelations at Uber suggest that nondisclosure and nondisparagement agreements served to silence complaints. The National Women’s Law Center points out that forced arbitration and nondisclosure agreements have emerged as a tactic to prevent workers from pursuing legal action or disclosing sexual harassment charges. Employees are made to sign away rights in order to work and have to take grievances through a dubious and hidden negotiation process in which management and lawyers are in control.
Rising to high levels in their field does not necessarily protect women from sexual harassment. According to one 2012 study by sociologists at the University of Maine and University of Minnesota, women who obtain power in the workplace, particularly in male-dominated environments, may be even more likely to be harassed than less powerful females, contrary to popular perception. They found that 58 percent of female supervisors in predominantly male work environments are likely to experience harassment, while 42 percent might expect harassment in female dominated workplaces. Researchers theorized that harassers were driven less by sexual desire than an urge to control and dominate women who were viewed as a threat to male privilege. The “power threat” theory suggests that women supervising men can be seen as undeserving of their positions, and their isolation makes them vulnerable to targeting. Wall Street executive Sallie Krawcheck’s vivid descriptions of pervasive sexual harassment in finance, where 90 percent of traders are male as well as 86 percent of financial advisers, underscores the problem.
Some research, such as a 2013 paper by Anne Maas, Silvia Galdi and Mara Cadinu, suggests that workplaces that feature hierarchical structures and significant power imbalances are more prone to sexual harassment. Unsurprisingly, it is a serious problem in academia, with graduate students who work closely with supervisors wielding substantial power over their trajectories and job prospects appearing to be particularly vulnerable. A recent crowd-sourced survey created by Karen Kelsky, who runs an academic job consulting business, focused on higher education and gained widespread media attention. Though it does not claim to be scientific, the survey nonetheless contains illuminating descriptions of how and under what circumstances harassment may occur, giving potential researchers useful clues about where to investigate. One 2017 study by Nancy Chi Cantalupo and William C. Kidder analyzing nearly 300 faculty-graduate student harassment cases for commonalities found that most faculty harassers are accused of physical rather than verbal harassment, and that over half of cases involve alleged serial perpetrators. The authors found that one in ten female graduate students at major research universities said that they had been sexually harassed by a male faculty member. They noted a paucity of scholarship on the topic, due in part to hindrances like confidentiality restrictions that keep harassment occurring on campus out of public view.
How to measure the costs of sexual harassment?
You don’t want to have this become you or your brand…In some ways, the higher the profile of the person you’re accusing is, the more likely that is to happen.
– Anita Hill, attorney and academic
Like a stealthy virus, sexual harassment impacts the wellbeing of society at every level. We know something about the public health toll and the effects on workers’ bodies over time — how it can cause chronic depression, PTSD, or amplify pre-existing mental health problems, as well as create physical issues like sleep disorders or even long-term problems like heart disease. In 2003, economist Kaushik Basu of Cornell University published a paper arguing that exposure to sexual harassment has something in common with exposure to excessive health hazards and working excessive hours due to its biological impact.
But what is the big picture of the economic impact? Part of the resistance to taking the issue of sexual harassment seriously in the workplace has surely been a lack of understanding of the economic consequences. In surveying the scholarship landscape on the subject, over and over one finds the statement: “significant research is lacking.” Trying to get the big picture can mean dealing with a puzzle of anecdotes, inferences, and studies that are not up-to-date.
Until recently, much of the attention to the economic impact of sexual harassment has centered on the cost to companies. Beyond the expense of providing sexual harassment training and protocols, the most obvious costs are those incurred in legal settlements and lawsuits. A 2010 study released by the Society for Human Resource Managers found that one in three companies had dealt with sexual harassment claims within the past two years. The EEOC task force reports that since 2010, employers have paid a total of $699 million to employees claiming that they were harassed on the basis of sex, race, disability, age, ethnicity/national origin, color, and religion through its pre-litigation process, and cited an estimate of settlements and court judgments in 2012 that amounted to more than $356 million in costs.
Recent noteworthy cases involving sexual harassment suggest that the cost to companies can be quite high: In 2016, Fox News Channel’s payment of $20 million in 2016 agreed to settle a lawsuit filed by former broadcaster Gretchen Carlson against former Fox News CEO Roger Ailes. In 2012, a federal jury awarded surgical aid Ani Chopourian $167,720,488 against Catholic Healthcare West when she alleged termination for filing harassment complaints. (Court records show the award was later vacated when attorneys reached a negotiated settlement). In 2011, Ashley Alford, an employee at a lease-to-own retail chain Aaron’s Rents alleged sexual harassment and assault by her manager and won a $95 million federal court verdict in an EEOC case. (The jury verdict was later reduced to $41.3 million, and court records show attorneys later reached an out-of-court settlement totaling $6 million).
Beyond these costs are the indirect costs of problems like lower productivity, higher turnover, and reputational harm. An oft-cited study from 1988 found that a typical Fortune 500 company lost $6.7 million a year owing to absenteeism, increased health-care costs, poor morale, low productivity and staff turnover resulting from sexual harassment — over $14 million in 2017 dollars. As Fox New recently learned when advertisers fled former host Bill O’Reilly’s show after harassment allegations, and the Weinstein Company also found when it had to sell distribution rights to one of its films in the wake of accusations against producer Harvey Weinstein, the ripple effects can be significant. (It is important to note that the stories that make the headlines are not an accurate sample of what occurs across the American business landscape: more information is needed in what happens in ordinary companies where the spotlight of media attention does not reach).
In a 1994 study, the cost of sexual harassment in the federal government was estimated to be $327 million, which includes the expense of job turnover ($24.7 million), workers taking sick leave because of harassment ($14.9 million), and diminished productivity ($287.5 million).
The financial and economic burdens born by women themselves, while undoubtedly significant, are difficult to quantify. We need to know more the effects of sexual harassment (and retaliation for reporting it) on things like wages, job performance, and career opportunities, not least because targets who pursue legal action are asked to show measurable harm, which requires reliable data.
Sociologists Heather McLaughlin, Christopher Uggen, and Amy Blackstone have produced one of the few recent pieces of research on how sexual harassment impacts women economically. The researchers, focusing specifically on women who are targeted early in their careers, found that women who had been harassed are far more likely to change jobs than those who didn’t. Overall, 80 percent of women who experienced severe sexual harassment left their jobs within two years.
Such women were also more likely to move to a different industry and reduce their work hours following incidents. These kinds of shifts can derail a career path, especially, the researchers note, when disruptions happen early in the women’s career. They also observed that women, compared to men, experience far more serious effects from interruptions to their work path.
According to the study, sexually harassed women reported greater financial distress two years later than those who were not targeted. The authors cited a willingness among targets to accept negative economic consequences in order to escape sexual harassment, saying that they often endured financial strain due to unemployment, career uncertainty, diminished hours or pay and anxiety about starting a new job. They found that overall impact is on par with that of serious injury or illness, incarceration or assault.
They also found that many women have suffered long-term career effects as they lowered their aspirations and narrowed their field of opportunity to avoid a repeat of the degrading experience. Those who stood up to hostile work environments, meanwhile, were often penalized with career stagnation and ostracization — even if they were not themselves the targets of the harassment.
In a recent interview, McLaughlin acknowledged the difficulty of putting a dollar figure on the economic losses to women, citing the problem of capturing what women experience in research models. She noted, as an example, discrepancies in the conduct women said they experienced and whether or not they would characterize such behavior as sexual harassment.
Researchers know that they face murky waters when diving into the subject. The difficulty of obtaining true rates of the problem and dealing with variations in impact, the potential differences in losses among industries, the conflation of broader gender discrimination and sexual harassment, and the problem of parsing cases in which multiple forms of discrimination are at play, such as sexual harassment together with racial discrimination, all create challenges. The work of distinguishing sexual harassment and its specific impact on women of color, trans people, from other forms of harassment they are subject to, and accounting for their effects is critical.
There is also the difficulty of precisely defining the damage and tracking its more insidious effects. A 2002 study found strong evidence that among active duty personnel in the U.S. armed forces, sexual harassment resulted in lowered job satisfaction. Similar evidence was found by psychology researchers in a 1997 study on female private sector and university employees who experienced both high-level and frequent, low-level sexual harassment. But calculating the cost of such experience is daunting. Does lower job satisfaction cause workers to apply less effort? What are the effects on other employees? What measures of profitability and performance could capture the impact?
Sociologists like Heather MacLauglin and anthropologists like Kate Clancey, professor of anthropology at the University of Illinois and lead author of a study of the extent of sexual harassment in scientific fields, have provided valuable qualitative and quantitative research on the prevalence and characteristics of harassment for economists to draw on for their own analyses.
One issue under debate among economists is the existence of “danger pay,” or pay for performing a hazardous duty or work involving physical hardship. Joni Hersch, an economist and professor of law and economics at Vanderbilt University, looked at the measurement of sexual harassment risks at work by industry, age group and sex and found that female workers are six times more likely than males to experience sexual harassment on the job. In her 2011 study, she found that women in workplaces where sexual harassment is common earn slightly more than they would in jobs with a lower risk of abuse, concluding that the difference was attributable to danger pay. Economist Elise Gould of the Economic Policy Institute disagrees with this interpretation, arguing that women not willing to put up with harassment are being pushed out into lower-paying jobs. Whether or not the concept of danger pay is applicable to sexual harassment, it appears that trying escape it may be economically burdensome to workers.
Historically, a lack of attention to the subject of sexual harassment, along with attendant issues of funding and support for research, has been a barrier to proper assessment. Lack of scholarship on the economic cost of sexual harassment also likely reflects the biases of an economics field heavily dominated by men and closely aligned with power structures that reinforce gender imbalances and inequality. The prevalence and cost of sexual harassment to researchers themselves is an area that also demands more investigation: reports of problems plaguing scientific communities add to the evidence that students and researchers are commonly targeted.
In calculating the overall cost to the economy, an avenue economists might explore would be to identify firms in which the incidence of sexual harassment appears to be low and compare their economic performance with companies with higher rates. There is also more work to be done on assessing the role of sexual harassment in the gender wage gap, which, according to a 2016 Institute for Women’s Policy Research analysis, widened among full-time workers between 2014 and 2015, with the ratio of median weekly earnings for women working full time compared to men decreasing by 1.4 percent. Sexual harassment must certainly be part of the story, but research specifying its role is lacking. Future study could also measure the effect of various forms of compensation strategies in ameliorating the effects of sexual harassment.
Remedies and reparations
So even if the people who did target me were punished, I still feel like I deserve some sort of compensation. I don’t want them to release a public apology — I want them to send me a check. – Kristen Gwynne, journalist
In October 2017, as news of his alleged activities exploded in the media, Harvey Weinstein announced that he would start a $5 million foundation devoted to providing scholarships to female directors at the University of Southern California. The USC School of Cinematic Arts quickly rejected the proposal. Obviously, a predator starting a scholarship fund is not the answer to repairing the damage of sexual harassment, which appears to be a kind of hidden tax many women must pay in their working lives.
The #MeToo movement has something in common with the Civil Rights Movement: it is not ultimately about taking down a few monstrous sexual predators any more than the Civil Rights Movement was about bringing justice to a few violent racists. That is part of the story, to be sure, but the harm brought about by sexism, power imbalances, and inequality usually occurs more slowly and insidiously than what happens in a few dramatic, high profile cases. Just as violent racial attacks are part of a larger systemic pattern of oppression and unfairness, sexual harassment unfolds within an architecture of systemic abuse in which the aspirations and dignity of large swaths of the population are regularly disrespected and violated.
We are beginning to understand and acknowledge that relying on such remedies as training videos that have become ubiquitous in U.S. organizations falls short of what is needed to correct such a large and persistent wrong. The efficacy of such training has been called into question by many, including Shannon Rawski of University of Wisconsin at Oshkosh, who notes that very few scientific studies have tested the effectiveness of sexual harassment training and points out that such instruction is often intended to minimize an employer’s own legal and financial liability. Rawski has found that researchers don’t have much evidence that sexual harassment training works to lower the number of incidents in a workplace or helping to shift its culture toward one that takes the issue seriously. The EEOC has reported that the training may even have negative effects in some cases. A 2007 study shows that men who underwent training tended to come out of it with their gender biases reinforced, associating men with power and women with a lack of it.
So what would help? Appropriate remedies and reparations come in many forms, including compensation payments, enhancing worker protections, and addressing the consequences of the incidents on the lives of targets even years after they occurred. Beyond these responses, a broader strategy of addressing gender inequality and power imbalances between workers and employers are critical to tackling the problem.
Recognizing that wrongdoing took place and identifying who is responsible is a crucial component of reparations. So is lifting the veil of secrecy by limiting the use of nondisclosures and ending forced arbitration. Lawmakers in New York, New Jersey and California are considering legislation that would limit the use of both nondisclosure agreements and out-of-court settlements with confidentiality clauses in cases of sexual harassment and assault. Current federal laws, including Title VII of the Civil Rights Act and the Congressional Accountability Act, could be amended to require companies to file disclosures of the number of sexual harassment claims settled, the amounts paid and the remedial actions taken in response. Recently, a bipartisan group of lawmaker took a positive step in introducing legislation that would end forced arbitration clauses in employment agreements, which would empower targets of sexual harassment.
Promoting more women into leadership roles is also widely cited as a remedy to conditions that allow harassment to flourish. Among the largest 3,000 U.S. public companies, women hold only 16 percent of board seats, while 624 companies have no women at all on their boards, which may explain why a recent survey reveals that 80 percent of boards have not discussed the issue of sexual harassment in the aftermath of the Weinstein revelations. In a Harvard Business Review article, Frank Dobbin and Alexandra Kalev note that that harassment is more common in workplaces where men hold most managerial roles or “core” jobs. “We already know how to reduce sexual harassment at work, and the answer is actually pretty simple,” they observed: “Hire and promote more women.” Many researchers suggest that obtaining gender balance throughout the organization is a crucial part of the solution. A 2015 report from researchers at Kent State University and the University of Texas at Tyler found that the “prevalence of male norms in the male-dominated environment may result in a more hostile workplace for women who are perceived by men as violators of the gender norms.”
Research indicates that the majority of sexual harassment charges filed, mostly by women, contain allegations of retaliation, which, although illegal, is both pervasive and subversive. The need for establishing secure mechanisms for reporting abuse and robust efforts to combat retaliation, including clear anti-retaliation policies, are needed. The tendency for businesses and organizations to cast human resources departments, which are incentivized to operate on behalf of management rather than the employee, in the role of handling sexual harassment complaints is inadequate, and often harmful to the target (see a recent journalistic discussion of sexual harassment in the media and the lack of channels for addressing it). There is a need for studies to gauge the effectiveness of alternatives for companies to address the problem.
Compensatory payments for sexual harassment are critical, but a single payment to a target such as a settlement check may often be an inadequate response and not proportional to consequences. Depending on the severity of the abuse, medical, psychological, social or legal services should be considered along with other forms of restitution or compensation. They can be particularly important for targets attempting to recover from the long-term effects of abuse.
More broadly, sexual harassment is part of the general trend of unchecked corporate power and a labor landscape in which workers face heightened job insecurity and few protections. Accountability for sexual harassment across class dimensions demands that the most vulnerable potential targets, particularly low-income women, women of color, immigrants and undocumented workers, are the focus of updating legal protections and the enforcement of labor law. As McAlevey has noted, worker supports such as universal childcare, free healthcare, free university education, paid maternity and paternity leave are crucial components to ensuring that all employees, particularly women, can participate in the workplace in a way in which they are not unduly burdened, insecure, and intimidated.
Finally, there must be a dedicated focus among social scientists to research the scope of the problem and its costs to society in every dimension. Social scientists can help do the valuable work of rectifying these wrongs by providing clear, detailed, and reliable information about how it happens, where it happens, and how costly it is to individuals and organizations. The good news is that going forward, researchers will have the benefit of new information that has been revealed since the launch of the #MeToo movement and can take advantage of a heightened awareness of the corrosive effects and widespread nature of sexual harassment. They will be much more likely to receive support, funding, attention, and outlets for their work, which will help us to chart a course forward to a more equitable world.
