Skip to main content

INSIGHTS BLOG > Playing the Sexual Harassment & Misbehavior Game

Playing the Sexual Harassment & Misbehavior Game

Written on 08 December 2017

Ruth Fisher, PhD. by Ruth Fisher, PhD

Definition of Sexual Harassment

Scenarios in the Sexual Harassment & Misbehavior Game

Cost to Firms of Sexual Harassment & Misbehavior

The Sexual Harassment & Misbehavior Game

Musings about the Game

Outcomes of the Model


The scourge of the day is sexual harassment. There’s nothing new about sexual harassment. It has always existed. What has changed is the fact that society is now overtly condemning the practice, rather than covertly condoning it.

On the other hand, the issue is a more complicated than simply determining whether or not harassment is acceptable.

Of course, sexual harassment does occur. And from what I’ve personally heard and experienced, there’s much more actual harassment that happens than just those cases that make the news.

However, there are complications.

One complication is the fact that relatively benign behavior can be misinterpreted as harassment. Someone tells a joke in bad taste. Or someone makes an inappropriate comment or gesture. An overly sensitive worker may then “misinterpret” or overreact to the inappropriate behavior.

A second complication is that the alleged victim may be intentionally falsely accusing the alleged harasser. The intention of the accuser may be to extort gains (payment, promotion, etc.) from the alleged harasser. Alternatively, the accuser may want the alleged harasser to be punished (have him fired, ruin his reputation, etc.) for whatever reason (personal, political, ideological). I would call this latter event sexual misbehavior, where the accuser has bad intentions. The classic case is the “woman who sleeps her way to the top.”

A third complication is that the accuser and the alleged harasser may have actually both willingly engaged in sexual activity. But after-the-fact, the accuser may have regretted his or her actions and thus have alleged harassment. This is another of case of what I would call sexual misbehavior, but where the accuser does not necessarily (initially) have bad intentions.

All these potential scenarios create complications for firms. When one employee accuses another employee of harassment, it may be

  • An actual case of sexual harassment;
  • Inappropriate behavior misinterpreted by the accuser as sexual harassment;
  • A case of sexual misbehavior on the part of the accuser, trying to benefit at the expense of the accused; or
  • A case of sexual misbehavior by the accuser who regrets his or her actions.

So then given an accusation of sexual harassment, what should the firm do? Should the firm punish the accused, given the risk that

  • The allegation may be true, but the accused is a rainmaker for the firm, and firing him or her will cost the firm a substantial amount in lost revenues?
  • The alleged harassment was simply a misinterpretation? Or worse, given the risk that the alleged harassment is actually sexual misbehavior by the accuser? In this latter case, firing the accused means you would be firing an innocent—and perhaps valuable—employee.
  • If the firm does not punish the employee and information about the incident leaks to the public, then the firm will be suffer the wrath of society?

We see, then, that when a firm is presented with an accusation of sexual harassment, it’s not so clear which actions the firm should take.

This analysis examines the Sexual Harassment & Misbehavior Game. It seeks to answer such questions as:

  • How do the courts define sexual harassment?
  • What are some of the different scenarios involving sexual harassment?
  • What are the costs to firms of sexual harassment and misbehavior?
  • What does the Sexual Harassment & Misbehavior Game look like?
  • What does the model predict?

My thanks to Lloyd Nirenberg, PhD, for providing significant direction for the analysis.

Definition of Sexual Harassment

Let’s start with the definition of sexual harassment. The legal definition is defines it as follows.

Courts and employers generally use the definition of sexual harassment contained in the guidelines of the U.S. Equal Employment Opportunity Commission (EEOC). This language has also formed the basis for most state laws prohibiting sexual harassment. The guidelines state:

Unwelcome sexual advances, requests for sexual favors, and other verbal or physical conduct of a sexual nature constitute sexual harassment when

  1. submission to such conduct is made either explicitly or implicitly a term or condition of an individual's employment, 

  2. submission to or rejection of such conduct by an individual is used as the basis for employment decisions affecting such individuals, or 

  3. such conduct has the purpose or effect of unreasonably interfering with an individual's work performance or creating an intimidating, hostile, or offensive working environment. 

Scenarios in the Sexual Harassment & Misbehavior Game

Order of Actions

I assume the order of actions and events in the Game proceed as follows (see Figure 1).

1. One employee (the Accuser) files a claim of sexual harassment by another employee (the Accused) with Firm Management. Faced with an accusation of sexual harassment, Firm Management believes that:

  • The Accused is guilty of sexual harassment with some probability α.
  • The Accused is not guilty of sexual harassment with some probability β; the Accuser simply misinterpreted the actions of the Accused.
  • The Accused is actually a victim of sexual misbehavior on the part of the Accuser Believes with some probability (1 - α - β).

2. Firm Management chooses whether or not to investigate the accusation. An investigation comes at some cost, C, to Firm Management. If Firm Management investigates the accusation, it will discover the truth of the situation.

3. Firm Management decides whether or not to punish the Accused (in the case of no investigation) or the Guilty Party (in the case of investigation).

4. If Firm Management decides not to punish the Accused, then the Accuser may sue the firm with some probability SUE. If the Accuser sues the firm, the information will be made public.

5. If the Accuser chooses not to sue the firm, then Firm Management decides whether or not to make the accusation and state of punishment public.

6. If Firm Management decides not to make the accusation and state of punishment public, then this information might still leak out to the public. In this case, despite Firm Management’s effort to keep the information within the firm, the public still becomes aware of the situation. The firm might be sanctioned by the public for withholding the information1

Figure 1

1 order actions

Harassment vs. Misbehavior

I distinguish sexual harassment from sexual misbehavior as follows.

Harassment: The Harasser sexually harasses the Harassed as per the definition of sexual harassment in the previous section. Situations include:

  • The Harasser threatens either to have the Harassed fired or to prevent the Harassed from being promoted if he or she complains.
  • The Harasser promises to grant the Harassed benefits, such as a part, or a promotion, if he or she “consents” to engage in sexual activity.

Misbehavior: The person Accused of sexual harassment is actually being used or manipulated by the Accuser. That is, claims of harassment are false accusations. Situations include:

  • The Accuser wants to the Accused to be socially condemned, lose his or her job, and/or not earn a promotion.
  • The Accuser hopes to “earn” a part, a promotion, or some other benefit by engaging in (if not actually initiating) sexual activity with the Accused.
  • The Accuser plans to blackmail the Accused by engaging in (if not actually initiating) sexual activity with the Accused and then extorting the Accused for money or a promotion.
  • The Accuser regrets having voluntarily engaged in sexual activity with the Accused and falsely accuses him or her of harassment after-the-fact.

Misbehavior: Undiscovered vs. Discovered: The Accuser falsely accuses the Accused of sexual harassment. The false accusation will eventually either remain undiscovered or become discovered.

  • Undiscovered: The Accuser is never discovered to be the actual Harasser. The Accused may end up wrongly suffering punishment. The Accuser may be wrongly martyred, promoted, or paid off.
  • Discovered: The Accuser is discovered to be the actual Harasser. The Accused is vindicated, though might still be blemished. The Accuser suffers punishment and/or loss of reputation.

Not Investigate vs. Investigate

Faced with an allegation of sexual harassment, Firm Management can choose whether or not to investigate the accusation.

  • Don’t Investigate: Firm Management chooses to not investigate the accusation. Firm Management doesn’t know the truth of the situation. This gives Firm Management plausible deniability of the allegation.
  • Investigate: Firm Management chooses to investigate the accusation at a cost C to the firm. The investigation reveals the true situation. Firm Management has documented proof of the harassment or misbehavior.

Not Punish vs. Punish

Faced with an allegation of sexual harassment, Firm Management can choose whether or not to punish the Accused.

  • Not Punish: Firm Management chooses to not fire the Accused or Guilty Party.
  • Punish: Firm Management chooses to fire the Accused or Guilty Party.

Not Make Public vs. Make Public

Faced with an allegation of sexual harassment, Firm Management can choose whether or not to make the accusation and punishment public.

  • Not Make Public: Firm Management chooses to keep the accusation of harassment and any subsequent punishment confidential within the firm. In this case, Firm Management is potentially promoting a hostile work environment for its employees.
  • Make Public: : Firm Management chooses to make the accusation of harassment and any subsequent punishment public. In this case, Firm Management is virtue signaling to the public that it promotes a safe work environment for employees.

Summary of Scenarios

The configuration of the different scenarios is presented in Figure 2. We have 42 different scenarios depending on whether:

[A] A reported case of sexual harassment is harassment, misinterpretation or misbehavior. The probabilities of each scenario, a, b, and (1 – a – b), are discussed in more detail below.

[B] Firm Management chooses not to investigate or to investigate the allegation.

[C] If Firm Management chooses not to investigate, it will not know what the true situation is.

[C] If Firm Management chooses to investigate, it will have documented proof that the situation is one of either sexual harassment, misinterpretation, or sexual misbehavior.

[D] Firm Management chooses to not punish or to punish the Accused or Guilty Party, depending on whether or not the truth was discovered.

[E] The Accuser chooses to not sue or to sue Firm Management and the Accused. If the Accuser chooses to sue, the situation will be made public.

[F] If the Accuser chooses to not sue, then Firm Management chooses to not make public or to make public the situation and punishment.

[G] Subsequent to Firm Management’s actions, situations not made public through a lawsuit or by Firm Management may become leaked to the public. The probability that a situation not made public is subsequently leaked to the public is L.

Figure 2

2 scenarios2 

Costs to Firms of Sexual Harassment & Misbehavior

Costs of Sexual Harassment

Catey Hill in “Preventing sexual harassment can save companies millions” and Josh Young in “The Effects of Sexual Harassment in the Workplace,” describe the potential costs to firms of having employees who engage in sexual harassment. According to these sources, the largest costs to firms of sexual harassment is the possibility that victims of sexual harassment will sue the firm. Firms that harbor sexually harassing employees also risk high turnover and lower productivity from the associated hostile work environment. Firms may suffer these losses from the victims of sexual harassment themselves, as well as from other employees aware of the harassment. Finally, other costs the firms may suffer include problems obtaining VC funding, loss of customers, and problems recruiting new employees to join the firm.

There’s also the potential for social condemnation, which may or may not cause the firm to lose investors and/or customers.

As for total damages, based on past events, my sense is that total damages to the Firm will be larger when the harassment:

  • Lasts for a longer period of time,
  • Is more egregious in nature,
  • Reflects a pattern of misbehavior on the part of the Harasser,
  • Is condoned my management, and/or
  • Is supported by hard evidence.

Costs of Sexual Misbehavior

Firms that develop a reputation for condoning employees who “sleep their way to the top” will suffer a loss of reputation. Having employees who benefit from sexual misbehavior signals that the firm supports misbehavior over merit. The firm also signals that it harbors a potentially hostile work environment. As such, the firm may lose actual and potential future employees, customers and investors.

Firms that publicly punish sexual misbehavior will develop a reputation for promoting a safe work environment for employees. The firm will generate loyalty by actual and potential employees, customers and investors.

Matrix of Costs Outcomes

Figure 3 presents the scenarios in Figure 2, together with the potential private and social costs to each of the key players associated with each scenario.


Private costs of outcomes include

  • Costs of investigating the allegation, C; and
  • Loss to the firm under the Punish scenario associated with firing the employee at issue. In particular, the firm could lose any power, connections, clients, and/or projects attributable to the fired employee.
  • Costs (or benefits) of condoning (or punishing) sexual harassment or misbehavior, in the form of higher (or lower) turnover and/or lower (or higher) productivity.

Social costs of outcomes include gain or loss in social reputation, which affects current and future potential sales (Customers) and investment opportunities (Investors).

Accused, Accuser

Private outcomes include

  • Emotional and monetary costs (benefits) of being fired under the Punish scenario (keeping one’s job under the Not Punish scenario), and
  • Emotional costs and benefits of having opposing party either not held accountable (under the Not Punish scenario) or held accountable (under the Punish scenario).

Social outcomes include

  • Reputation within firm if scenario is Not Make Public and/or Not Leak, and
  • Reputation in society if scenario is Make Public and/or Leak

Figure 3

3 scenarios full2 

The Sexual Harassment & Misbehavior Game

Combining information from the previous sections, we get the Sexual Misbehavior Game as illustrated in Figure 4.

Figure 4

4 sh game 

Musings about the Game

Relative Ranks of Workers

When I look at Figure 4, a thought that comes to mind is: are the Accused necessarily higher up in the firm hierarchy than their alleged victims? Equivalently, are the Accusing necessarily subordinate to the Accused?

In “Sex and Power in Washington” Elizabeth Drew asserts that it's the power the Harasser has over the Harassed that makes the harassment egregious. She downplays acts of harassment when the Harasser does not have such power over the Harassed.

Unwanted sexual aggression is never a good thing, but it’s doubly objectionable when the victim’s career is at stake. This distinction is what puts a question mark over the head of Glenn Thrush, a breakout reporter in the Washington bureau of the Times. Thrush, who has been “suspended” by his paper pending an investigation, was accused of making sexual advances on colleagues during his previous job at Politico, but he had no direct power over them, which puts him in a different category.

Without power, the Harasser can create a hostile work environment for the Harassed. But he or she cannot also be a threat to the career of the Harassed.

Does the Firm Want Harassment Made Public?

My initial tendency was to think that Firms surely wouldn’t want any alleged and/or proven cases of sexual harassment or misbehavior made public. Airing one’s dirty laundry in public is never a good thing. Or is it?

Actually, Firms can use cases of sexual harassment or misbehavior to establish a reputation for not tolerating bad behavior from employees. Firms accomplish this “virtue signaling” by publicizing cases of sexual harassment or misbehavior, together with specifics on the punishment meted out to the perpetrators.

Firms may also publicize employee misconduct together with punishment to avoid public condemnation, especially once the cat’s out of the bag. I think this is what happened in the Entertainment Industry after Harvey Weinstein’s behavior became known outside the industry. What everyone had supposedly known to be true for decades—yet failed to condemn—all of a sudden became behavior worthy of immediate public condemnation. I think this is also what’s happening with Conyers. From Elizabeth Drew:

As for Representative John Conyers, House Democratic leaders are trying to ease him out of office, apparently on the grounds of senility—the 88-year-old now-former chairman of the House Judiciary Committee has shown up for some occasions in his pajamas. At first the House circulated for years Democratic leadership (mainly Minority Leader Nancy Pelosi) felt that Conyers surrendering his committee chairmanship would suffice... But the fourth harassment charge against Conyers, brought to light the day after Pelosi’s appearance, pushed leaders into pressing for Conyers’s resignation from the House, though a significant portion of the Congressional Black Caucus disagreed with this.

Company Management may also use the opportunity to oust a high-ranked employee (the Harasser) who has power others want. When the Donald Trump and Billy Bush recording was released, many people used this as yet another justification for removing Trump from office.

When Is Harassment More Likely to Occur?

Firms that provide work environments more conducive to sexual harassment or misbehavior will be more likely to face such accusations by workers. Define g as the probability a firm receives an accusation from an employee that he or she is being harassed by another employee.

(1)      γ is the situational contribution to the probability of sexual harassment or misbehavior.

If Firm Management receives an accusation from an employee that he or she is being harassed by another employee, the Firm doesn’t know for sure whether the accusation is true or not. There is a probability a that the accusation is true, that is, the Accuser is a victim of sexual misbehavior by the alleged Harasser.

(2)      α is the individual contribution to the probability of sexual harassment or misbehavior (propensity to harass).

There is a probability β that the accusation is false, where the Accuser misinterpreted the actions of the Accused.

(3)       β is the individual contribution to the probability of being a victim of sexual harassment or misbehavior (propensity to be harassed).

There is a probability (1 - α - β) that the accusation is false and the Accuser is actually the perpetrator of sexual misbehavior against the alleged Harasser.

(4)       (1 - α - β) is the individual contribution to the probability of engaging in sexual misbehavior (propensity to engage in misbehavior).

By defining these probabilities, we are then led to ask:

  • When will a firm be more likely to suffer from sexual harassment or misbehavior? That is, when will γ be larger?
  • When will an Accuser alleging sexual harassment be more likely to
    • Be a victim of sexual harassment? That is, when will α be larger?
    • Be a victim of sexual misinterpretation? That is, when will β be larger?
    • Be a perpetrator of sexual misbehavior? That is, when will (1 - αβ) be larger?

Situational Determinants of Sexual Harassment or Misbehavior (γ)

In “Sexual Harassment in the Workplace: How does it affect Firm Performance and Profits?” Drusila Brown and Xirong Lin discuss the situational conditions that are more supportive of harassment. These authors indicate that a Firm environment is more conducive to harassment when:

  • Organizational Awareness: There is relatively little organizational awareness of the issue. There is a low perception by management that sexual harassment is a concern, there is less employee training on the subject of sexual harassment, etc.
  • Organizational Tolerance: There is relatively more organizational tolerance of “loose” behavior.
  • Competition: There is less competition for workers from outside employers.
  • Supervisor Power: Supervisors have more power over workers.
  • Supervisor Alignment: There is less alignment of supervisors’ interests with those of the firm. This will be the case, for example, when supervisor pay does not depend on the performance of subordinates.

Propensity to Be a Sexual Harasser (α) or Misbehaver (1 - α - β)

Employees will have a greater incentive or propensity to engage in sexual harassment or misconduct when the potential gains from doing so are larger.

Potential gains will be larger when there is a large difference in power between the two employees. In the case of sexual harassment, this means employees are more likely to be harassed when the Harasser has a much higher rank than the Victim. In the case of sexual misbehavior, this means employees are more likely to be a target of misbehavior when the Misbehaver has a much lower rank than the Victim.

Potential gains will be larger when the favor being sought (a role or promotion):

  • Is less likely to be obtained by the seeker on his or her own (low probability event) and/or
  • Has high value (high value event)

Expected Value of Favor = [(probability seeker obtains job with help)

– (probability seeker obtains job without help)] x [value of job]

Propensity to Be a Victim of Sexual Harassment or Misbehavior (β)

Employees will have a greater propensity to be victims of sexual harassment or misbehavior when they are more “thin-skinned” or otherwise less tolerant. Some younger people today have a reputation for being less tolerant than others. It would thus not be surprising if the number of allegations of sexual harassment by these “snowflakes” were greater than reports by others. Unfortunately, because much sexual harassment goes unreported, it would be difficult to use information on reported cases over time to test this hypothesis.

How Do We Decrease the Incidence of Sexual Harassment?

There’s nothing new about sexual harassment. It has always existed. But recently society has shone the spotlight on a practice that has traditionally been relegated to the shadows. The move from shadows to spotlight was swift. Rose McGowan served as a “Missionary,” converting public knowledge into common knowledge.

Pretty much the entire entertainment industry knew that Harvey Weinstein was a long-time sexual predator. But it took Rose McGowan’s tweet to convert this public knowledge to common knowledge. That tweet shattered the shield that had been protecting the predators. People have been coming forth ever since to accuse other well-known personalities of sexual harassment.

In Harvey Weinstein and the Common Knowledge Game, W. Ben Hunt, Ph.D., explains this dynamic.

The core dynamic of the CK [Common Knowledge] Game is this: how does private knowledge become — not public knowledge — but common knowledge? Common knowledge is something that we all believe everyone else believes. Common knowledge is usually also public knowledge, but it doesn’t have to be. It may still be private information, locked inside our own heads. But so long as we believe that everyone else believes this trapped piece of private information, that’s enough for it to become common knowledge.

The reason this dynamic — the transformation of private knowledge into common knowledge — is so important is that the social behavior of individuals does not change on the basis of private knowledge, no matter how pervasive it might be. Even if everyone in the world believes a certain piece of private information, no one will alter their behavior. Behavior changes ONLY when we believe that everyone else believes the information. THAT’S what changes behavior. And when that transition to common knowledge happens, behavior changes fast.

So then one way to decrease the incidence of sexual harassment in society is to encourage Missionaries to come forth and publicize allegations of harassment. To the extent we can encourage Missionaries to reveal true cases of harassment, while not encouraging accusers to publicize false allegations, society will be better off.  

Outcomes of the Model

We can plug some numbers into the model captured in Figures 2, 3, and 4 in to see what happens. Playing with the model revealed some juicy tidbits.

What It Comes Down To

There are three value comparisons that drive Firm Management’s choices of action:

  1. The value of the Accused to the firm,
  2. The expected cost to the firm of the Accuser bringing suit against the firm, and
  3. The expected drop in firm value associated with information being leaked to the Public.

Does the value of the Accused to the firm outweigh the expected costs of (i) being sued by the Accuser and (ii) having the matter leak to the Public? If so, the Accused stays. If not, he goes.

Expectations play a major role in the decision of Firm Management. It is tempting to keep what you have—the valuable employee—rather than firing the rainmaker because of some indeterminate risks that the Accused may sue or the stock value might take a hit.

Prominent Employees

In “The damaging, incalculable price of sexual harassment” Kari Paul and Maria LaMagna report describe how companies are often willing to payoff accusers of prominent employees. Paying off the accusers is often less costly to the firm than firing the prominent employees.

When alleged perpetrators are prominent at their companies, employers sometimes decide it’s worth settling with alleged victims rather than getting rid of the perpetrator, Jennifer Drobac, a professor of law at Indiana University’s Robert H. McKinney School of Law. They’re sometimes powerful and profitable for the company that management decides it’s worth paying “a few hundred thousand dollars in settlements to make a problem go away,” he said. That decision often comes with one major problem: In many cases, these men continue to offend.

In today’s winner-take-all economy (WTA) (see my previous blog post for a discussion of the WTA game), prominent employees – a headline actor, a CEO who has achieved a company turnaround, the head rainmaker in a law firm – are often worth more to the firm than all underlings combined. In these cases, almost any costs associated with sexual harassment on the part of the prominent employee will be tolerated. It’s simply cheaper to deal with any effects of a hostile work environment (employee turnover, lower employee productivity) and pay off the Accusers than it is to lose the prominent employee. It’s also easier to keep what you have–a valuable employee–than to give it up for some small chance of loss–a lawsuit or drop in stock price.

As Kari Paul and Maria LaMagna indicate, however, the problem with condoning such behavior is that it tends to be ongoing. At some point, one of the many Accusers of the predatory employee will sue the company. Or, information about one or more harassments will leak to the public. When either of these events occur, Firm Management will be forced to pay the piper.

Investigate or Not?

Faced with an allegation of sexual harassment, Firm Management must decide whether or not to investigate the claim.

Suppose Firm Management does not plan on punishing the Accused, even if he or she is found guilty during an investigation. In this case Firm Management should not investigate. This gives Firm Management plausible deniability, in case either (i) Firm Management is sued, or (ii) the allegations are leaked to the public. In either of these two cases, Firm Management will be crucified, either in court or in the public arena. There’s nothing worse to juries or the public than to discover that Firm Management knew about a misdeed, yet overtly chose not to punish the perpetrator.

Alternatively, suppose Firm Management does plan on punishing the Accused if found guilty. In this case, if there’s a chance the Accused is innocent, then Firm Management should investigate the accusation if the Accused is valuable to the firm. Otherwise, Firm Management risks wrongly punishing (losing) a valuable employee.

Virtue Signaling

Firms who make public their punishment of sexual harassers are engaging in virtue signaling. They’re telling the world that they don’t tolerate employee misbehavior, that they provide their employees with a safe work environment.

But what does such virtue signaling get such firms? Does the public react by lavishing purchases and investments on them?

More likely, virtue signaling gets a firm a reprieve from punishment that would have been imminently forthcoming had it not fired the harasser. Presumably, the wave of firings that have taken place in the wake of the Harvey Weinstein scandal were motivated more by fear of punishment than hope of reward. Kevin Spacey. Matt Lauer. John Conyers. All these people had been known to be long-time predators. However, it wasn’t until the Harvey Weinstein story broke that they were held accountable. Why then? It was fear of sanctioning by the lynch mob—riled up by the Harvey Weinstein situation—that motivated the firings, not the expectation of rewards for doing the right thing.

Zero Tolerance

Merriam Webster defines zero tolerance as “a policy of giving the most severe punishment possible to every person who commits a crime or breaks a rule.”

Many firms have implemented a zero tolerance policy against sexual harassment, as a means of virtue signaling. Such a policy clearly disempowers predators. However, it also empowers accusers: a policy of zero tolerance weaponizes the threat of accusation by misbehaving employees (i.e., false victims).