Chapter 34: Attraction in the Workplace — Power, Policy, and Professional Boundaries

The paper arrived in Organizational Psychology Review in late spring, and by the time the campus listservs lit up, Dr. Adaeze Okafor had already fielded three interview requests and two accusatory emails from strangers. The paper — "Cross-Cultural Norms and Experiences of Workplace Attraction: Findings from the Global Attraction Project, Year 3" — was, as she explained to her graduate students with a mixture of pride and exhaustion, "completely normal social science." It documented variation in how employees across twelve countries understood, reported, and managed feelings of attraction toward colleagues. It did not endorse harassment. It did not celebrate workplace romance. It carefully distinguished attraction as a psychological phenomenon from the ethical and legal dimensions of acting on that attraction.

None of that mattered to the people who read the headline.

"New Study Says Workplace Attraction Is Normal and Cross-Cultural," wrote one blog. A think-piece in a major online magazine argued that the study gave "scientific cover" to harassers by framing predatory behavior as a natural human drive. An opposing op-ed accused critics of the study of wanting to pretend that adult human beings magically ceased being attracted to one another when they crossed a building threshold. Okafor and Reyes, watching the debate unfold from opposite hemispheres, found themselves engaged in the unglamorous but necessary work of science communication in the age of outrage.

Dr. Carlos Reyes, calling from Buenos Aires, was characteristically measured. "We documented what people experience," he said. "The problem is that acknowledging experience feels to some people like endorsing behavior. That is a category error, but it is an extremely common one." Okafor agreed, and then spent the next three months writing and rewriting the kind of public-facing clarifications that most researchers never anticipate having to produce when they design a study.

This chapter takes the controversy as its starting point — not because the controversy was particularly important in itself, but because it reveals something true and useful about the territory we are entering. Workplace attraction sits at the intersection of the deeply personal and the intensely institutional. It involves feelings that people have involuntarily and choices they make deliberately. It operates within power structures that make "free choice" a more complicated concept than it might first appear. And it is governed by legal frameworks, organizational policies, and cultural norms that vary enormously in their assumptions, their enforcement, and their effects.

This chapter does not tell you whether it is acceptable to date a coworker. That is not a question social science can answer for any individual in any specific situation. What it can do — what we will work to do here — is map the terrain carefully enough that you can think clearly about the forces at play, the stakes involved, and the ethical obligations that arise when attraction enters professional life.

34.1 Why the Workplace Is an Attraction Context

Before we get to policy and power, it is worth pausing over a foundational question that is surprisingly rarely asked: why is the workplace a site of attraction at all? The question seems naive at first — obviously people have feelings wherever they are — but when you examine it carefully, it leads to a richer picture of how workplace environments systematically generate the psychological conditions that produce attraction.

Proximity and the Mere Exposure Effect

Robert Zajonc's classic 1968 research on the mere exposure effect established one of social psychology's most reliable findings: repeated exposure to a stimulus — any stimulus — tends to increase liking for it. The phenomenon operates across categories from consumer products to photographs of strangers; it appears to work through a combination of reduced perceptual processing effort (familiarity feels smooth and unthreatening) and reduced uncertainty (you know what to expect from a familiar face). People like what they see often, all else equal.

Workplaces engineer repeated exposure with remarkable efficiency. You see the same faces at the same desks, in the same break room, at the same Monday morning meeting, week after week, month after month. The colleague who barely registered in your first week becomes familiar, familiar becomes comfortable, and comfortable becomes — for many people — genuinely attractive in ways that were not present at first encounter.

The residential proximity studies of Festinger, Schachter, and Back (1950) provided early evidence that the mechanisms generalize across contexts. In their study of friendship formation in student housing, residents who lived physically closer — who passed each other more often in the hallways — were reliably more likely to become friends. Workplace studies have replicated this pattern: in a classic analysis of seating arrangements and romantic relationships, Segal (1974) found that police trainees were significantly more likely to form friendships, and subsequently romantic relationships, with partners whose surnames were adjacent in the alphabet — because they were assigned adjacent seats during training. The mechanism was almost certainly proximity: alphabetical seat neighbors simply saw each other more.

The critical point here is that proximity-generated attraction is not "fake." The attraction that develops from repeated exposure involves real familiarity, real mutual knowledge, and real comfort. It is not reducible to mere exposure effects alone — but those effects are part of what initiates the process.

Shared Goals, Arousal, and Collaborative Bonding

Beyond proximity, workplaces create another powerful attraction amplifier: the experience of working toward something together under conditions of meaningful challenge. The research on misattribution of arousal — established by Dutton and Aron (1974) and discussed in Chapter 9 — suggests that physiological arousal from any source can be partially attributed to a salient interpersonal stimulus present at the moment of arousal. The adrenaline of a difficult project deadline, the tension of a high-stakes client presentation, the relief of a crisis successfully resolved — all of these generate arousal that can attach, with varying degrees of awareness, to the person working beside you.

This does not mean workplace attraction is mere misattribution. But it does mean that the emotional intensity of professional collaboration — particularly in high-stakes, deadline-driven environments — creates conditions where the arousal-attraction linkage is systematically active. People who have weathered difficult moments together often feel closer and more positively oriented toward one another than the objective quality of the relationship might independently warrant. Anyone who has experienced the particular intimacy of a team that has survived a crisis together will recognize this phenomenon.

The broader research on task-based bonding adds another layer. Work by Baumeister and colleagues on the "need to belong" suggests that productive joint effort is itself a bonding mechanism — that creating something together activates feelings of connection that are experienced as interpersonal warmth. The colleague with whom you have built something, solved a difficult problem, or navigated a genuinely hard situation is not merely a familiar face; they are someone with whom you share a specific kind of shared experience that carries relational weight.

The organizational implications of this mechanism deserve emphasis. Employers who maximize collaborative intensity — open-plan offices, cross-functional teams, intensive on-site work cultures, mandatory social events — are not just optimizing for productivity. They are also systematically engineering conditions that generate interpersonal bonding. This is not necessarily a problem; bonded teams often do perform better. But it creates a particular organizational responsibility: environments that are deliberately designed to intensify interpersonal connection cannot be simultaneously indifferent to the interpersonal consequences that connection sometimes generates.

Similarity, Professional Culture, and Selective Sorting

Chapter 14 reviewed the robust finding that similarity promotes attraction across multiple dimensions — attitudinal, demographic, and dispositional. Workplaces sort people by a distinctive package of similarities. Colleagues in the same organization share professional culture, vocabulary, and values about work. People in the same field or industry share educational trajectories, intellectual interests, and — often — social class positioning. The research director and the junior analyst at the same policy institute likely share more cognitive frameworks, more reference points, and more basic assumptions about what counts as interesting or important than either shares with most people they could encounter on a dating app.

This is not a trivial observation. The similarity-attraction link is strongest for attitudinal and value similarity — precisely the kind of similarity that professional culture generates in abundance. People who care about the same problems, use the same analytical frameworks, and find the same kinds of work rewarding are genuinely similar in ways that matter for attraction. That this similarity is contextually produced does not make it less real.

The Extended Character Observation Window

Perhaps the most important and least often discussed reason workplaces generate attraction is that they provide something rare in contemporary social life: the opportunity to observe a person's actual character across a wide range of situations over an extended period, without the distortions of impression management that characterize specifically romantic or sexual contexts.

You see how a colleague handles conflict with a difficult client. You see whether they give credit generously or hoard it. You see how they talk about people who are not in the room. You see whether they stand up for team members or leave them exposed. You see how they behave under pressure, under recognition, and under disappointment. This is extraordinarily rich information for the purpose of character assessment — and character assessment, as the research on long-term relationship satisfaction consistently confirms, is more predictive of enduring attraction than physical appearance.

The irony is that the workplace reveals character in precisely the way that dating contexts often conceal it. Early romantic contexts activate impression management with particular force: people present idealized versions of themselves. Workplaces are not immune to impression management, but they subject it to a longer and more varied test. The character that survives years of varied professional observation is, in an important sense, more reliable as a basis for attraction than the character performed over a first dinner.

34.2 The Prevalence of Workplace Romance: What the Data Say

📊 Research Spotlight: How Common Is Workplace Romance?

Survey data on the prevalence of workplace romance has been collected with reasonable consistency across several decades, though question wording and sample selection produce notable variation in estimates. The picture that emerges is one of surprisingly high prevalence.

The Society for Human Resource Management (SHRM) has surveyed American workers on workplace romance annually since the 1990s. Their 2019 survey found that approximately 27 percent of American workers had been in a workplace romance at some point in their careers — a rate that represents some decline from the 36–41 percent range found in CareerBuilder's annual workplace romance surveys between 2004 and 2018. Whether this decline reflects genuine behavioral change (possibly related to #MeToo culture shifts), methodological differences between survey instruments, or sampling variation is an open empirical question.

When questions are extended from "relationship" to include attraction that was never acted upon, rates rise substantially. Studies using broader definitions typically find that 50–60 percent of workers report having experienced workplace attraction at some point in their careers. The Okafor-Reyes Year 3 data offers important cross-cultural texture: in their twelve-country sample, self-reported rates of workplace attraction (defined as "noticing or experiencing attraction toward a colleague that you were aware of at the time") ranged from 54 percent in the Swedish sample (where, Okafor noted, respondents showed the highest comfort with direct disclosure on sensitive topics) to 31 percent in the Japanese sample (where the combination of professional discretion norms and survey culture involving significant underreporting of personally sensitive experiences likely suppressed responses below behavioral reality).

The U.S. Okafor-Reyes sample registered 44 percent for attraction and 21 percent for "acted upon in some way" — broadly consistent with SHRM data, with the caveat that survey question wording was not identical across instruments.

Several demographic patterns appear consistently across studies:

• Age effects: Younger workers report higher rates of workplace attraction and workplace romance than older workers, even controlling for relationship status. The effect is partly compositional (younger workers are more often single) but persists after controlling for relationship status.
• Industry effects: Industries with longer work hours, higher stress, and more intense team collaboration — finance, medicine, law, technology at crunch time — tend to show higher rates than industries with more routinized, shift-based work.
• Hierarchy effects: Supervisor-subordinate romances are consistently reported at lower rates than peer romances, but carry disproportionate risk of harassment complaints and negative organizational outcomes.

⚠️ Critical Caveat: Self-Report Limitations

All of the above data are based on self-report surveys, which are subject to the standard vulnerabilities: social desirability bias (underreporting stigmatized behaviors), recall bias (inaccurate memory for past events), and question framing effects (what "workplace romance" means varies by respondent). Behavioral observation data on workplace attraction is — for obvious reasons — extremely difficult to collect. These estimates should be understood as approximations rather than precise measurements.

34.3 Power Asymmetries: The Heart of the Problem

If workplace attraction were simply a matter of mutual feelings between equals in a context with no consequences, the ethical analysis would be considerably simpler. The reason we need a full chapter on this topic is that workplaces are not structured around equals. They are structured around hierarchies — formal and informal, explicit and invisible — and those hierarchies transform the meaning of attraction in ways that cannot be analytically separated from the feeling itself.

Consider the clearest case: a supervisor and a direct-report subordinate who experience what appears to be mutual attraction. Assume, for purposes of argument, the most charitable possible interpretation: both individuals are genuinely interested, neither is fabricating attraction, the feelings are real and reciprocal. Even in this best-case scenario, a series of structural complications arise that make the situation categorically different from the same mutual feelings arising between two people who met in a running club.

The Consent Problem

Genuine consent — the kind of consent that is ethically meaningful rather than merely formal — requires the ability to refuse without suffering negative consequences for the refusal. When one party controls the other's performance evaluations, project assignments, scheduling, promotion recommendations, and (directly or indirectly) continued employment, the ability to decline an advance is structurally compromised.

This is not a claim that the subordinate is necessarily unable to say no, or that every subordinate who says yes to a supervisor's interest is doing so under coercion. It is a claim that saying no has potential costs — professional costs — that would not exist in the absence of the power differential. Those costs, real or merely perceived, alter the voluntariness of the consent calculus in ways that cannot be neutralized by individual good intentions.

Legal scholars and ethicists use the concept of coercive context to describe situations where formal consent is present but structural features make that consent morally problematic. A subordinate who agrees to a date with their supervisor may be exercising genuine autonomous choice — or may be calculating, consciously or not, that declining could be professionally costly. From the outside, these two situations are in many cases indistinguishable. That is precisely why institutional policies exist: not to adjudicate individual cases, but to eliminate the coercive context itself by separating the romantic dynamic from the professional power relationship.

The Perception Problem

Even when a workplace relationship is genuinely mutual and genuinely voluntary, colleagues observing it from the outside cannot know this with certainty. The result is a perception problem with real organizational consequences. The subordinate in a relationship with their supervisor may receive preferential treatment in assignments, schedules, or evaluations — or may simply be perceived as receiving it — affecting team dynamics, morale, and the credibility of both parties' professional accomplishments.

Research by Pierce and colleagues (2004) found that observer perceptions of favoritism in supervisor-subordinate romances were associated with decreased morale and increased turnover intent among other team members, independent of whether actual favoritism occurred. The perception of favoritism is in some ways as organizationally damaging as actual favoritism: it degrades the legitimacy of institutional processes even when those processes are functioning correctly.

The Exit Problem

Unlike romantic relationships that begin in purely social contexts, workplace relationships must continue after they end. The termination of a workplace romance typically does not end the professional relationship: the former partners continue to share meetings, emails, performance review processes, and the physical space of the workplace. Research by Dillard and Witteman (1985) on post-romantic workplace dynamics found that about half of workplace romances that ended were reported as having "gone sour in a way that affected work," with reported negative effects including reduced productivity, increased interpersonal conflict, and in some cases the exit of one party from the organization entirely.

The exit costs of a failed workplace romance are not symmetrically distributed. In supervisor-subordinate relationships, the lower-power party typically bears a disproportionate share of the exit costs: they are more likely to transfer, to change departments, or to leave the organization entirely. The supervisor, whose organizational position is typically more secure, is more likely to remain in place. This asymmetry of exit costs is part of what makes the power differential ethically significant beyond the immediate question of consent. A person who enters the relationship knowing they will bear a disproportionate share of exit risk cannot be said to have agreed to those risks fully freely if the power differential itself shaped their ability to assess and negotiate the terms of involvement.

34.4 Legal Frameworks: Quid Pro Quo and Hostile Work Environment

The legal architecture governing workplace harassment in the United States rests primarily on Title VII of the Civil Rights Act of 1964, which prohibits employment discrimination based on sex. The application of Title VII to sexual harassment was not settled until the Supreme Court's landmark ruling in Meritor Savings Bank v. Vinson (1986), which held that sexual harassment that creates a hostile or abusive work environment constitutes sex discrimination under Title VII. Two categories of harassment have been legally recognized since Meritor:

Quid Pro Quo Harassment

Quid pro quo harassment — from the Latin "this for that" — occurs when employment benefits are made contingent on sexual favors, or when the refusal of sexual advances results in negative employment consequences. It takes the form of an explicit or implicit exchange: submit to this, and you will be promoted / retained / assigned the desirable project; refuse, and you will face consequences. Because quid pro quo harassment by definition requires the ability to confer or withhold employment benefits, it can only be committed by someone with formal authority over the victim's employment — a supervisor or manager, not typically a peer.

The legal standard for quid pro quo is relatively straightforward: was a tangible employment action — hiring, firing, promotion, demotion, assignment — made contingent on submission to sexual conduct? When this standard is met, employer liability is essentially strict: the employer is liable regardless of whether it knew about the conduct or had anti-harassment policies in place.

It is worth noting that quid pro quo harassment is not limited to explicit propositions. Courts have found the standard met in cases where the contingency was communicated through implication, atmosphere, and pattern — where a target was placed in a situation where submission was the understood expectation even if nothing was ever stated in direct transactional terms. This has the important practical implication that "I never explicitly said anything" is not a defense when the contextual signals communicated an implicit expectation clearly enough to a reasonable person.

Hostile Work Environment Harassment

Hostile work environment harassment does not require a formal power differential, and it does not require an explicit transactional element. It occurs when unwelcome sexual conduct is "sufficiently severe or pervasive to alter the conditions of the victim's employment and create an abusive working environment" (Meritor). The legal standard requires both an objective element (a reasonable person would find the environment hostile) and a subjective element (the plaintiff actually found it hostile). The conduct need not be physically threatening; verbal conduct alone can meet the standard.

⚖️ Debate Point: The "Reasonable Person" Standard and Its Critics

The "reasonable person" standard embedded in hostile work environment law has been contested since its inception. Feminist legal scholars — including Kathryn Abrams, Catharine MacKinnon, and Susan Estrich — argued that the standard encodes a masculine perspective: what a reasonable man finds tolerable may systematically differ from what a reasonable woman finds hostile, and applying a gender-neutral standard effectively privileges the perspective of the dominant group. The Ninth Circuit adopted a "reasonable woman" standard in Ellison v. Brady (1991), reasoning that harassment must be evaluated from the perspective of the person it most commonly affects. Other circuits rejected this approach, concerned that it essentializes gender differences in ways that might backfire.

The current legal landscape is unsettled. The EEOC uses "reasonable person in the complainant's situation" — an attempt to account for context without locking in gendered assumptions. The debate reflects a genuine tension between the universalizing aspirations of law and the particularity of lived experience that any adequate harassment standard must engage.

34.5 The #MeToo Movement and Its Effects on Workplace Norms

The #MeToo movement — which entered mainstream public discourse in October 2017 following reporting on Harvey Weinstein and was rapidly amplified by social media — produced the most significant rapid shift in workplace romance and harassment norms in living memory. The sociological effects were multidimensional and, in some respects, contradictory.

Quantitative Evidence for Behavioral Change

Survey data from 2018–2020 documented widespread changes in workplace interaction. A 2018 LeanIn.org survey found that 60 percent of male managers reported discomfort with one-on-one mentoring of women — a rate substantially elevated from pre-#MeToo baselines. A 2019 Pew Research Center survey found that 31 percent of women and 19 percent of men reported that workplace interactions between men and women had become more careful or cautious.

Workplace harassment complaint rates showed an initial spike: EEOC data showed a 12 percent increase in sexual harassment charges filed in fiscal year 2018 compared to 2017 — a pattern consistent with increased willingness to report rather than necessarily an increase in the underlying behavior. Longitudinal data on harassment incidence (rather than reporting) is harder to obtain and more mixed in its conclusions.

The Chilling Effect Debate

The survey evidence for changed workplace behavior has been interpreted through two competing framings that are worth examining carefully.

The chilling effect narrative holds that #MeToo has caused men to reduce legitimate mentoring relationships and professional interactions with women out of excessive fear of misinterpretation, depriving women of valuable career opportunities. This concern is real and has been documented: when senior men avoid solo mentoring relationships with junior women, the costs fall on the junior women who are most in need of sponsorship.

The accountability narrative holds that the "chilling effect" framing misrepresents what changed. The behaviors being reduced were not uniformly legitimate professional conduct; many of the changes involved the reduction of conduct that was genuinely problematic — casual physical contact without consent, comments on appearance, "jokes" that required the target to accept degradation to maintain team harmony. Treating all behavioral reduction as equivalent ignores the substantial proportion of reduced behavior that should have been reduced.

Sociologist Cecilia Ridgeway's analysis of post-#MeToo dynamics offers a useful synthesis. She argues that #MeToo shifted the cultural "accounting" around ambiguous workplace conduct — it increased the probability that conduct previously silently absorbed would be named and responded to. This shift in accounting, she argues, is precisely the function that cultural norm shifts serve: not creating new harassment (most of which was occurring before 2017) but changing the social calculus around acknowledgment and response.

💡 Key Insight: "Normal" Is Not "Acceptable"

A recurring rhetorical move in discussions of workplace attraction — amplified by the Okafor-Reyes controversy — is the inference from prevalence to permissibility: if attraction is widespread and cross-cultural, it must be "natural," and if it is natural, imposing restrictions on it must be overreach. This inference fails at every step. Prevalence does not confer normative status. "Natural" in the evolutionary sense (a behavior that served reproductive functions in ancestral environments) tells us nothing about ethical permissibility — many things that are naturally occurring are ethically impermissible. And acknowledging that feelings are widespread does not prescribe any particular response to them.

34.6 The Okafor-Reyes Study: Findings and the Science-Policy Interface

The findings that generated the Okafor-Reyes controversy deserve examination on their own terms, because they are genuinely interesting and the controversy around them illustrates important issues about how scientific findings enter public discourse.

Okafor and Reyes documented what they called attraction acknowledgment norms — culturally specific expectations about whether, when, and how workplace attraction should be recognized, expressed, or disclosed. Their analysis focused on three contrasting samples: the United States, France, and Japan.

The American Pattern: Compartmentalization

In the U.S. sample, respondents described a norm of compartmentalization: attraction is an internal experience to be managed privately and not expressed, and the burden of management falls largely on the individual experiencing the attraction. Disclosure was widely viewed as creating an unwelcome obligation on the part of the person disclosed to — placing them in the position of either reciprocating or explicitly rejecting, both of which were seen as socially costly. The norm, as respondents described it, involved a high threshold for disclosure paired with relatively little cultural infrastructure for graceful decline.

This pattern has features that are protective from a harassment perspective: it reduces the frequency with which unwanted attraction is expressed in the first place. But it also has costs: the suppression norm creates no particular cultural support for the targets of unwelcome attention in declining gracefully, because the norm's implicit assumption is that such situations should not arise at all rather than that they should be navigated well when they do. When attraction is expressed despite the suppression norm — as it inevitably sometimes is — the targets are often poorly supported by the social scripts available to them.

The French Pattern: Social Acknowledgment

In the French sample, a norm of social acknowledgment was more prevalent: low-key expressions of attraction — a compliment, an invitation to dinner, sustained social attention — were viewed as acceptable forms of professional sociality without necessarily implying romantic expectations. Importantly, French respondents also showed substantially higher comfort with declining such acknowledgments without negative social consequence — the norm, as it operated in their accounts, involved both expression and refusal being treated as socially normal rather than momentous.

This combination — higher baseline acknowledgment paired with more normalized refusal — is, in principle, consistent with strong consent norms. Reyes noted this carefully: the French pattern has features that are appealing from a consent-values perspective, because it normalizes the "no" alongside the "yes." But he was equally careful to note that this pattern depends on a dense set of shared social assumptions about what expressions mean and what refusals mean — assumptions that are not transportable across cultural contexts or organizational settings.

The Japanese Pattern: Professional Discretion and Its Complexities

The Japanese sample revealed the most internally complex picture. Strong norms of professional discretion meant that workplace attraction was rarely acknowledged in explicit form — expressed rates of workplace attraction disclosure were among the lowest in the twelve-country sample. But qualitative interviews substantially complicated this surface picture.

Women in the Japanese sample described navigating implicit expressions of attraction from male colleagues with considerably less freedom to respond than surface norms suggested. The norm of professional indirection, they reported, created a context where unwelcome attention was expressed in forms that were deniable as attraction — putting recipients in the position of either accepting conduct they found uncomfortable under the social fiction that it was not what it was, or naming the subtext in ways that violated the norm of professional face-saving.

Okafor found this pattern particularly important for the broader debate: "Silence is not acceptance. Discretion norms don't eliminate unwanted attention — they eliminate the social permission to name it. That's not a more ethical situation; it's a less transparent one."

🧪 Methodology Note: The Science-Policy Interface

The controversy about the Okafor-Reyes paper illustrates a recurring challenge in social science: the gap between what a study actually claims and what stakeholders on both sides of policy debates want it to claim. Researchers studying sensitive social phenomena — harassment, attraction, consent — face systematic pressures to produce findings that can be cleanly deployed in policy arguments. The complexity and conditionality of scientific findings — the "it depends," "in this context," "for these respondents" caveats — are precisely what get stripped out in translation to headlines.

Okafor and Reyes addressed this in their public statement directly: "Our findings describe what people experience across cultures. They are not a prescription for what policies should permit. The relationship between scientific description and normative evaluation is not automatic. Researchers can help by being clear about what they are claiming; public communicators can help by resisting the temptation to flatten that clarity into usable simplicity."

34.7 Organizational Policies: What Companies Actually Do

Organizations have developed several distinct policy approaches to managing workplace romance. Each reflects different assumptions about the role of organizations in regulating employees' personal lives.

Anti-Fraternization Policies

Anti-fraternization policies prohibit romantic or sexual relationships between certain categories of employees. The scope varies considerably: the most restrictive versions prohibit any romantic relationship between any two employees; the most common versions prohibit supervisor-subordinate relationships within the same reporting chain. Broad policies — extending to peer-peer relationships — are increasingly rare and legally contested in jurisdictions with strong privacy protections for employees' off-duty conduct. California, for instance, has extensive statutory protections for off-duty conduct that limit employers' ability to prohibit peer romantic relationships.

Narrow, hierarchy-focused policies are both more legally defensible and more organizationally functional: they address the specific structural problem (the coercive context created by power asymmetry) without claiming regulatory authority over the full range of employees' personal lives.

Disclosure Policies and "Love Contracts"

Disclosure policies do not prohibit workplace relationships but require that relationships be reported to HR. Some organizations pair disclosure requirements with relationship agreements — colloquially known as "love contracts" — in which the disclosing parties affirm that the relationship is voluntary and that they understand the company's harassment policy and its applicability during and after the relationship.

The legal purpose of these agreements is liability management: if a relationship later generates a harassment complaint, the organization can demonstrate that it was put on notice, that the parties affirmed consent, and that the relevant policies were acknowledged. The practical limitations of this approach are significant. A signature on a document at the beginning of a relationship establishes initial consent; it does nothing to address the ongoing consent dynamics of a relationship that continues while the power differential continues. Nor does it alter the structural reality that the lower-power party faces exit costs that the higher-power party does not.

Supervisor Transfer Policies

The most organizationally sophisticated approach — and the one most consistently recommended by organizational psychologists and employment attorneys — is the supervisor transfer policy: when a supervisor-subordinate relationship is disclosed, one of the parties must transfer to a different reporting chain. This approach addresses the structural problem directly. Rather than trying to manage a coercive context through documentation, it eliminates the coercive context by separating the romantic relationship from the professional power differential.

Transfer policies require organizational flexibility — multiple departments or reporting lines into which transfers can be made — that smaller employers may genuinely lack. But where feasible, they are the most coherent policy response because they directly address the specific feature of supervisor-subordinate relationships that makes them ethically problematic.

📊 Research Spotlight: Policy Effects

Koen and colleagues (2010) examined harassment complaint rates in organizations before and after implementation of formal disclosure policies. They found that disclosure policies were associated with modest reductions in complaints in large organizations (500+ employees) but no significant effect in organizations with fewer than 200 employees — a finding the researchers attributed to the reduced anonymity of disclosure in smaller settings, where disclosure requirements may paradoxically deter reporting. Organizations with supervisor transfer policies showed the most consistent reductions in harassment complaints over time.

34.8 HR and the Structural Management of Attraction

Human Resources departments occupy a structurally uncomfortable position in the governance of workplace attraction. They are formally charged with protecting employee well-being and ensuring legal compliance — but their organizational position makes the first of these commitments genuinely difficult to fulfill, because their primary client is the organization, not the individual employee. This creates predictable tensions when employee interests and organizational liability interests diverge.

Research on employees' actual use of HR harassment reporting mechanisms produces sobering findings. Studies consistently find that the large majority of harassment targets do not file formal complaints — estimates of the proportion who report range from 5 to 30 percent depending on the severity of the conduct and the demographic characteristics of the sample. The reasons cited most frequently are fear of retaliation (well-founded: studies consistently find that retaliation against harassment complainants is both common and difficult to prove), skepticism about HR impartiality, and concern about the professional costs of being labeled a "difficult" employee or "troublemaker."

Sociologist Frank Dobbin's research on organizational anti-discrimination programs adds a counterintuitive dimension. His analysis found that some formal programs — particularly mandatory online training modules and complaint procedures presented as the organization's official response to harassment — may paradoxically reduce the perceived legitimacy of informal complaints about ambiguous conduct. The implicit message, Dobbin argues, is that the organization has "done its part," creating a threshold effect: conduct that does not rise to the level of formal-policy violation may be absorbed without response in ways that collectively maintain a problematic culture.

This is not an argument against organizational anti-harassment programs. It is an argument for designing them with attention to their actual behavioral effects — which requires rigorous evaluation — rather than their symbolic function, which is to demonstrate institutional compliance with legal norms.

34.9 When Attraction Becomes Harassment: The Pathway and the Gray Zones

Most popular conversations about the line between attraction and harassment treat it as a binary. You are on one side or the other. The research literature on perpetration tells a substantially more complicated story.

The transition from attraction to harassment typically unfolds gradually, through a series of incremental steps that feel internally justified at each stage. The pathway researchers have documented most consistently begins with what they call boundary testing behavior: action that is constructed to be ambiguous, that can be plausibly interpreted as innocent while simultaneously serving as a probe of the target's receptivity. A prolonged touch on the shoulder. A comment that is just within the range of professional, but carries a charge. An invitation to coffee that could be professional or could be something else. The ambiguity is often functional: it allows the initiator to escalate if the response is positive and to retreat to the "innocent" interpretation if it is not.

The target, facing genuinely ambiguous behavior, is placed in an impossible interpretive situation. Responding positively risks encouraging escalation. Responding negatively risks a confrontation that may have professional consequences, and may feel disproportionate if the behavior was genuinely innocent. Many targets resolve this impossible choice by doing neither — by absorbing the behavior neutrally and hoping it stops. Research on boundary testing perpetrators indicates that neutral responses are typically read as a weak green light: as not actively discouraging, and therefore as license to continue.

What follows this pattern, in documented harassment cases, is typically escalation: behavior that becomes progressively less ambiguous, and in which the initiator becomes less careful about maintaining deniability. By the time the behavior is unambiguously harassing — clearly sexual, clearly unwelcome, clearly beyond the bounds of professional conduct — it has often been preceded by months or years of behavior that was absorbed without formal response. This history is important for understanding why harassment is so difficult to address: by the time the behavior is unambiguous enough to report, the target has typically been silent through a long period of escalation, which can be used to undermine the credibility of their account ("if it was really that bad, why didn't you say something earlier?").

The Gray Zones Are Real — and Limited

It would be dishonest to pretend that every uncomfortable workplace interaction is harassment or that every expression of attraction that is not perfectly reciprocated constitutes a harm. The gray zones between attraction, awkwardness, poor social judgment, and harassment are real, and they contain genuinely ambiguous cases that resist clean categorization.

But the gray zones should not be used to cast general doubt on the existence of the pattern, or to generate so much uncertainty about every specific case that institutional responses become impossible. Research on perpetration consistently finds that harassers rarely commit single, isolated ambiguous acts; they typically engage in escalating patterns that target multiple individuals over time. Institutional processes that treat each incident as isolated and categorically ambiguous consistently fail to account for this pattern evidence — which is one reason why peer reports, organizational memory, and documentation practices matter so much in addressing harassment effectively.

34.10 Institutional Power Contexts: Professors, Doctors, Therapists

The workplace attraction analysis applies with particular intensity in contexts where institutional power is especially pronounced. Three professional contexts deserve attention.

Academia

The professor-student relationship is among the most structurally asymmetric professional relationships that most working adults will encounter. Faculty members can control grades, degree progress, recommendation letters, research opportunities, and informal network connections that substantially determine career trajectories. Graduate students and advanced undergraduates — the populations where faculty-student attraction most frequently arises — are particularly vulnerable, because their professional futures may be substantially dependent on single mentoring relationships with the very person who has expressed interest in them.

Institutional policies on faculty-student relationships have evolved significantly over the past three decades. Many major research universities — including Harvard, Stanford, the University of Michigan, and UCLA — now have explicit policies prohibiting romantic or sexual relationships between faculty members and students over whom they have direct supervisory authority, regardless of the students' age or stated consent. Several have gone further, prohibiting relationships with any current students in the same department or program, even absent direct supervisory authority. The rationale for these broader prohibitions — which go beyond the structural coercion argument and into a more expansive claim about the corrosive effects of romantic dynamics on academic communities — will be examined in Case Study 1.

Medicine

The American Medical Association's Code of Ethics prohibits sexual relationships between physicians and their current patients. The prohibition is grounded in the same coercive context analysis that applies to supervisor-subordinate workplace relationships: the clinical relationship creates a power dynamic that is incompatible with genuine consent, and the fiduciary duty of the physician to the patient requires that the physician actively protect against exploitation of the therapeutic relationship.

The AMA guidance extends the prohibition to recently terminated doctor-patient relationships, though the operationalization of "recently" involves some professional judgment. Several medical specialty organizations have adopted stricter standards — particularly in psychiatry and psychotherapy, where the power dynamics of therapeutic relationships are understood to be especially intense and durable. The AMA's underlying reasoning is worth stating clearly: the fiduciary duty of a physician to their patient is incompatible with a sexual or romantic relationship, not because such feelings are impossible or immoral per se, but because the fiduciary relationship — which requires the physician to place the patient's interests above their own — cannot be coherently maintained alongside a romantic dynamic. The roles are structurally incompatible.

Psychotherapy

The prohibition on therapist-client sexual contact is among the most absolute in professional ethics. The American Psychological Association's ethics code prohibits sexual intimacy with current clients and with former clients for a minimum of two years after termination of the therapeutic relationship. Even after the two-year period has elapsed, the therapist bears the burden of demonstrating that the relationship is not an exploitation of the former therapeutic relationship — a burden that most ethics scholars regard as effectively impossible to meet in most cases.

The strictness of this prohibition reflects decades of clinical and research evidence on therapeutic transference — the process by which clients develop intense feelings toward therapists that are rooted in the therapeutic dynamics (the therapeutic relationship as a context for examining one's most vulnerable self with an attentive, non-judgmental other) rather than the therapist's personal qualities. Transference makes the client's expressed feelings toward the therapist an unreliable guide to the genuinely voluntary character of any subsequent relationship — a point that is professionally non-controversial among trained clinicians, whatever the client may believe at the time.

🔵 Ethical Lens: The Obligations of the Powerful

Discussions of power-asymmetric attraction in workplace and professional contexts often focus on what the less powerful party should do: how to respond, how to report, how to protect themselves. This framing — however well-intentioned — locates the responsibility for managing the consequences of power in the wrong party.

The more useful ethical question is: what obligations does holding power confer?

The answer that emerges from professional ethics frameworks across medicine, law, therapy, and academia is something like this: those who hold institutional power over others are obligated to actively protect the conditions under which genuine consent is possible, and to refrain from actions that compromise those conditions even when they believe the less powerful party is a willing participant. The willingness of the less powerful party does not discharge the obligation of the more powerful party to maintain appropriate professional relationships — because the less powerful party's apparent willingness may itself be a product of the power differential that the professional is obligated not to exploit.

This is a demanding standard. It is demanding by design. Institutional power is not merely a personal quality; it is a structural feature conferred by the institution and revocable by the institution. The institution has a legitimate — some would say a constitutive — interest in ensuring that its power is not turned against the people it is intended to serve.

34.11 Returning to the Okafor-Reyes Controversy

Several weeks after the paper was published, Okafor and Reyes issued a joint public statement. Several passages model the kind of careful science communication that the public discussion of research findings rarely involves but consistently needs.

"Our findings document the prevalence and cross-cultural variation of workplace attraction as a psychological experience," they wrote. "They do not imply that attraction should be acted upon, that existing power differentials should be ignored, or that legal and ethical frameworks governing workplace conduct should be loosened. The existence of a feeling is not a recommendation for action. Our study describes what people experience; it does not prescribe what people should do."

They continued: "Several critics have suggested that acknowledging the prevalence of workplace attraction normalizes harassment. We believe the opposite is closer to the truth: treating workplace attraction as unspeakable prevents the development of institutional frameworks for managing it ethically. Organizations that pretend the feelings don't exist are not better positioned to address their consequences — they are worse positioned. Silence does not eliminate attraction; it eliminates the social and institutional infrastructure for responding to it responsibly."

This argument — that clear-eyed acknowledgment of a phenomenon is a precondition for managing it responsibly — is, in many ways, the thesis of this chapter. Workplace attraction is real. Power differentials are real. Legal frameworks are real. Ethical obligations are real. Managing the intersection of all four requires engaging honestly with each of them, which means being willing to describe what exists before rushing to either condemn or permit it.

The controversy itself, Okafor reflected later, was also data: "The fact that people immediately tried to use our description of reality as a weapon in a policy argument tells you something about how badly we need better public literacy around the science-policy interface. That's not the researchers' problem to solve alone. But it means we have to be much more careful about how we communicate."

34.11b Intersectionality and the Workplace Attraction Landscape

Any serious analysis of workplace attraction must contend with the fact that its risks and consequences are not evenly distributed. The research literature is consistent on a point that deserves emphasis: who experiences unwanted workplace attention, who is disbelieved when they report it, and who pays the steepest professional price for either reporting or not reporting are systematically patterned by race, gender, sexuality, and class.

Gender and the Asymmetry of Risk

The most extensively documented asymmetry concerns gender. Large-scale studies consistently find that women experience unwanted workplace sexual attention at substantially higher rates than men. The 2018 Stop Street Harassment survey found that 38 percent of women reported experiencing workplace sexual harassment, compared to 13 percent of men. These rates vary substantially by industry and occupational context: research on restaurant workers, domestic workers, and agricultural workers documents rates of sexual harassment that dwarf those found in office-based professions, a disparity that reflects the intersection of gender with class, employment precarity, and isolation from formal HR systems.

The asymmetry of risk is not merely about incidence; it also involves response and consequence. Research on harassment disclosure consistently documents that women who report harassment fare worse professionally than men who report equivalent conduct — they are more likely to be disbelieved, more likely to face formal or informal retaliation, and more likely to exit the organization as a result of the reporting process rather than the initial harassment. This disparity in response creates a rational deterrent to reporting that has nothing to do with the severity of the original conduct.

Race, Ethnicity, and Compounding Vulnerabilities

Black women and women of color face a distinctive intersection of vulnerabilities in workplace harassment contexts that has been documented with increasing precision in recent years. Sociologist Jennifer Berdahl and colleagues have extended their work on harassment to examine how racial minority status intersects with gender to produce distinctive patterns of unwanted attention. Their findings suggest that Black women and Latinas in professional settings face harassment that simultaneously sexualizes and racializes — conduct that draws on racist stereotypes about hypersexuality in ways that leave targets uncertain whether to frame their experience as racial discrimination, sexual harassment, or both.

The ambiguity this creates is not incidental. Legal frameworks for addressing harassment are structured to evaluate discrete categories — racial discrimination, sex discrimination — in ways that may inadequately capture conduct that operates through both simultaneously. Research by Kimberlé Crenshaw on intersectionality (the foundational theoretical contribution that gives this concept its name) specifically examined this limitation in legal contexts, arguing that discrimination that operates through the intersection of categories may be invisible to frameworks that examine categories one at a time.

For our purposes, the practical implication is clear: the experience of unwanted workplace attraction and harassment is not a generic experience that varies only in its severity. It is shaped by the specific social identities of those involved, in ways that produce systematically different patterns of risk, response, and consequence.

Sexual Minority Workers

LGBTQ+ workers face a distinctive set of challenges in workplace attraction contexts. In environments where heterosexuality is assumed — which describes a substantial portion of workplaces — gay, lesbian, bisexual, and queer workers must make ongoing decisions about disclosure, concealment, and the management of their romantic and sexual lives in professional contexts where those decisions carry professional consequences.

Research on sexual minority workers' workplace experiences documents both the costs of concealment (increased psychological distress, reduced authentic engagement, difficulty forming genuine workplace relationships) and the costs of disclosure (in non-affirming environments, disclosure can trigger discrimination, social exclusion, and hostile treatment that does not qualify legally as harassment but functionally creates exactly the hostile work environment the law is designed to prevent).

Transgender and nonbinary workers face an additional layer of complexity: their gender expression itself is frequently the target of workplace attention that ranges from invasive curiosity to overt hostility. Research by the National Center for Transgender Equality consistently documents high rates of workplace discrimination and harassment among transgender respondents — rates that, in the most recent surveys, remain substantially higher than those for cisgender LGB workers.

Class, Employment Precarity, and the Power to Say No

A dimension of the workplace attraction analysis that is often underexamined is class and employment precarity. The "power to say no" that is foundational to consent-based analysis of workplace attraction is not uniformly distributed — it is substantially determined by the alternative options available to the person being approached. A tenured professor has far more power to decline an advance from a department chair than an adjunct instructor who teaches on renewable one-semester contracts. A full-time employee with strong performance reviews and institutional ties has far more power to decline an advance from a supervisor than a contract worker on a three-month engagement with no guaranteed renewal.

Employment precarity — the condition of contingent, uncertain, easily-terminated work — amplifies the coercive context of workplace attraction in ways that are analytically separable from formal supervisory authority. A worker who cannot afford to lose income has less effective power to say no than a worker with equivalent formal authority protections, because the consequences of refusal are more devastating. Any serious analysis of consent in workplace attraction contexts must account for economic coercion alongside formal organizational power.

34.12 An Ethical Framework for Workplace Attraction

Let us close by returning to the question deferred at the beginning: whether it is acceptable to date a coworker. Social science cannot give you a universal answer. What it can give you is a framework for thinking clearly about the specific situation you are in — one that takes seriously both the genuine reality of workplace attraction and the genuine complexity of power, consent, and institutional obligation.

The Questions Worth Asking

Is there a power differential? If so, who holds the power, what is its scope, and how durable is it? The existence of a power differential does not make a relationship automatically impossible, but it transforms the ethical calculus significantly. A formal supervisory relationship — where one person evaluates the other's performance, determines their assignments, and influences their career advancement — represents the clearest case for abstention, or at minimum for immediate structural remediation (transfer to a different reporting chain) before anything else proceeds. Informal power differentials — seniority, social capital, access to networks the other person needs — are less visible but not less real.

Is genuine consent possible in this context? Not formal consent — but the voluntary, informed, non-coerced agreement that the concept of consent actually demands. If structural features of the situation compromise voluntariness in either direction — through organizational power, through financial dependency, through professional vulnerability, through the employment precarity discussed above — those features need to be addressed structurally before any relationship can be genuinely consensual. This means that "I asked, and they said yes" is a necessary but not sufficient condition for the kind of consent that is ethically meaningful in power-asymmetric workplace contexts.

What do your institution's policies require? Most organizations of any scale have policies that address workplace romance and disclosure. Those policies exist for reasons that reflect accumulated institutional experience about where workplace romances tend to create problems — not theoretical problems, but the specific documented harms of favoritism perceptions, retaliation after breakups, and coercion that leaves the lower-power party with no good options. Knowing what the policies are is a minimum of responsible conduct. Understanding why they exist — engaging with the reasoning rather than just the rules — positions you to act ethically in situations the policy may not have anticipated.

What are the obligations of the more powerful party? If you hold more institutional power, the ethical burden falls more heavily on you. This principle often provokes resistance: why should I be held to a higher standard just because of my position? The answer is that the higher standard is not about personal virtue but about structural reality. You benefit from institutional power that was conferred by the organization to serve organizational and social purposes. That power is not neutralized by your good intentions in a specific interpersonal situation; it remains structurally present and potentially distorting. Those who benefit from institutional power are the appropriate parties to bear the costs of maintaining the conditions under which that power does not become coercive — because they are the ones with the structural capacity to create or dissolve those conditions.

What This Framework Does Not Do

This framework does not resolve every case. It does not tell you with certainty whether any specific relationship is ethically permissible. It does not adjudicate the hard cases — the peer-peer relationship where the power differential is informal and contested, the relationship that began after a transfer that the organizational policies required, the attraction that arises after someone has left the department but not the organization. These cases exist, they are genuinely complex, and they require contextual judgment that no framework can fully supply.

What the framework does is ensure that contextual judgment is informed — that the people making decisions are thinking clearly about power rather than around it, attending to the less powerful party's position rather than primarily their own, and engaging with institutional policy as a tool for responsible conduct rather than an obstacle to personal desire.

The Longer View

Okafor, in a postscript to the controversy, made an observation that is worth holding onto as this chapter's closing thought. "The most interesting thing about the public response to our paper," she wrote, "was not what people argued but what they assumed we were trying to say. Critics assumed we were defending harassment. Defenders assumed we were endorsing workplace romance. We were doing neither. We were trying to describe what actually happens between people who spend forty hours a week together in a structured hierarchical environment."

"What happens is: they find each other interesting and sometimes attractive. What needs to happen next — the institutional response, the individual ethical reasoning, the legal framework that draws the limits — is a completely different question. The mistake is to collapse the first question into the second. Attraction is not harassment. Feeling is not action. Description is not prescription. If we could maintain those distinctions clearly in public discourse about workplace relationships, we would be substantially better positioned to address the real harms that occur when those distinctions are violated."

That is, perhaps, the most useful takeaway this chapter can offer: a set of distinctions clear enough to think with, and a framework careful enough to act on.

Next chapter: Chapter 35 examines how media — from Shakespeare to Netflix — both reflects and constructs our cultural scripts for romance and seduction.

Review Questions

1. Explain how the mere exposure effect, misattribution of arousal, and similarity-attraction together contribute to making workplaces contexts of attraction.
2. What does research data indicate about the prevalence of workplace romance across cultures? What methodological factors affect these estimates?
3. Distinguish quid pro quo harassment from hostile work environment harassment in legal terms. Give an original example of each.
4. What is a "coercive context"? Explain why a supervisor-subordinate relationship creates a coercive context even when both parties report genuine feelings.
5. Summarize the Okafor-Reyes findings on American, French, and Japanese workplace attraction acknowledgment norms. Why did Reyes caution against treating the French pattern as straightforwardly preferable?
6. Describe the three main types of organizational workplace romance policies. What does research say about the effectiveness of each?
7. Why do professional ethics bodies in medicine, psychotherapy, and academia prohibit romantic relationships between practitioners and those they serve, even when the less powerful party explicitly consents?
8. Describe the "boundary testing" pathway to harassment. Why does it place targets in an interpretive double-bind?
9. What does Frank Dobbin's research suggest about the unintended consequences of some formal anti-harassment programs?
10. Explain the ethical principle articulated in the "obligations of the powerful" section. Do you find this argument persuasive? What objections might be raised against it?

Next chapter: Chapter 35 examines how media — from Shakespeare to Netflix — both reflects and constructs our cultural scripts for romance and seduction.