More than one in five people in employment have experienced violence or harassment at work. The harm is not gender-neutral.

ILO Convention No. 190 defines violence and harassment as a range of unacceptable behaviours and practices, or threats of them, whether a single occurrence or repeated, that aim at, result in, or are likely to result in physical, psychological, sexual or economic harm. Gender-based violence and harassment is not a separate, narrow list of offences: it describes how sex and gender shape who is targeted, how harm is expressed and whose exposure becomes normalised.

It can include assault and unwanted touching; sexual comments, images or requests; threats, intimidation, humiliation and bullying; stalking or harassment through work-related messages; or demands for sexual compliance tied to hiring, shifts, pay, training, promotion or contract renewal. Economic harm matters because ordinary management powers can be weaponised: overtime removed after refusal, wages withheld to force compliance, a contract threatened, or housing and work used together as leverage.

Not every disagreement, unpopular decision or moment of rudeness is automatically violence or harassment. The question is whether the conduct or threat is unacceptable and aims at, results in, or is likely to result in harm, read in its context and power relationship. Intent is relevant, but it is not the only test. Local legal definitions and procedures vary; this page states a rights-based frame and is not a legal determination of an individual case.

Under C190, the world of work includes more than the production floor or office. It reaches places where a worker is paid, rests, eats, changes or uses sanitation facilities; work-related travel, training, events and social activities; communications enabled by technology; employer-provided accommodation; and the commute to and from work. Protection also extends beyond standard employees to people working regardless of contractual status, including trainees and apprentices, workers whose employment has ended, volunteers, jobseekers and applicants, and people exercising an employer’s authority.

This wider frame matters in supply chains. A supervisor’s message sent after the shift, harassment in a recruitment process, coercion in a dormitory, or abuse on employer-organised transport does not become private merely because it occurs away from the line.

Violence and harassment are behaviours, but the conditions that allow them to continue are structural. A person may control another worker’s shifts, overtime, leave, performance record, line assignment, contract renewal or access to employer-provided housing. If that same chain of command also receives the complaint, refusing the conduct may carry an economic and safety cost before any formal retaliation is recorded.

The concentration of women in low-paid production work makes this power structure visible. An ILO regional study estimated that Asia’s garment, textile and footwear sector employed around 60 million workers. In the latest country data used by the study, women represented 86% of garment workers in Myanmar, 76% in Cambodia, and around 70–73% in Thailand and Viet Nam. These are employment-structure figures, not estimates of GBVH prevalence. They show why gender, job position and managerial authority have to be analysed together. Source: ILO, 2022.

Commercial pressure does not mechanically cause violence. But short lead times, sudden order changes, price pressure and aggressive production targets can intensify the power held by supervisors and managers. When that pressure passes through insecure contracts, gendered hierarchies and weak worker voice, abuse can become a way of enforcing pace and obedience. Responsibility therefore reaches beyond one person’s conduct to purchasing practices, job design, supervision, worker representation and access to remedy.

Harm can also come from co-workers, subordinates, recruiters, contractors, customers, clients or other third parties. That is why a policy focused only on individual misconduct by direct employees will miss part of the workplace system. Prevention has to examine job design, supervision, production pressure, recruitment, accommodation, transport and digital communication — not only whether a prohibited-behaviour poster exists.

Calling some workers “vulnerable” as if vulnerability belongs to their identity hides the source of risk. Risk is produced when discrimination meets unequal power, insecure work, isolation or dependence on a person or institution that controls income, status, movement or care. These conditions can intersect rather than simply add up.

The global survey found that migrant women were almost twice as likely as non-migrant women to report sexual violence and harassment at work during their working life (10.0% compared with 5.4%). It also found that people who had experienced discrimination related to gender, disability, nationality or ethnicity, skin colour or religion during their life were more likely to report violence and harassment at work. These associations do not make identity the cause of harm; they point towards the structures around it.

In practice, the same workplace can be experienced differently by a migrant woman whose housing or documents are tied to work, a woman with a disability who encounters inaccessible information or communication, a migrant woman with a disability facing both sets of controls, or a worker whose stigmatised health status can be exposed or used against them. A gender lens therefore cannot stand alone: it must be able to see migration, disability, health, age, race and ethnicity, sexuality and gender identity, and contractual status without reducing any person to a category.