How to manage Sexual Harassment Risks And Live The Positive Duty

Creating safer and more respectful workplaces in Australia is no longer about reacting once a complaint lands. The Positive Duty under the Sex Discrimination Act 1984, alongside strengthened WHS obligations on psychosocial hazards, has shifted the baseline. Organisations are now expected to identify and control sexual harassment risks before harm occurs.

That shift demands a different kind of capability. Not policies, not posters, but the practical skill to spot early warning signs, understand where a workplace is vulnerable, and act quickly to prevent issues escalating into conflict, misconduct, or psychological injury.

This article is part of our Conduct & Culture series, created in collaboration with Intersection, where we break down how organisations can meet the Positive Duty with clarity and real-world practice. Here, we focus on Standard 4: Risk Management – the discipline of anticipating, assessing, and addressing psychosocial and behavioural risks as a core leadership responsibility.

Proactive Sexual Harassment Risk Management

For years, risk management sat quietly in WHS files and risk registers. Today, regulators expect something far broader: leadership behaviour that prevents harm before it appears in any data.

Safe Work Australia is clear: managing psychosocial risk is an ongoing, proactive process, not a once-a-year activity. Their model defines four continuous steps – identify hazards, assess risks, control risks, and review controls – all supported by genuine consultation and strong leadership accountability.

The Human Rights Commission reinforces the same principle. Under the Positive Duty, organisations must take “reasonable and proportionate steps” to prevent sexual harassment, discrimination, and victimisation. The key word is prevent. The duty is forward-looking, not reactive.

And this responsibility doesn’t sit with the risk team alone.
Modern people risk management is a shared function – HR, people leaders, and frontline managers all play critical roles. Identifying early signals, understanding team dynamics, supporting psychologically safe processes, and responding in a trauma-informed way are core HR responsibilities, not peripheral tasks.

Effective prevention requires everyone influencing people and culture to treat risk as part of their everyday practice not something that gets escalated to a single department.

Psychosocial Risks Can Leave Early Signals

Serious incidents often grow out of a series of small interactions and changing team dynamics that occur long before a formal report. These early signals can be subtle, but they become visible when organisations understand the patterns to look for.

Common early indicators can include:

• conflict or discomfort that lingers but isn’t resolved
• teams operating without clear boundaries or role clarity
• a manager whose behaviour regularly sits close to the line
• subtle exclusion, jokes, or micro-aggressions that start to become routine
• rising workloads or pressure that erode team safety and connection
• unexplained turnover or withdrawals from the same area
• informal norms taking precedence over respectful behaviour

Intersection describes these as micro-signals which are small shifts in tone, behaviour, or culture that can indicate increasing vulnerability. They don’t always point to harm, but they do warrant attention.

The Human Rights Commission echoes this in its Positive Duty guidance: environments where early concerns are minimised, or where low-level disrespect is allowed to persist, are the ones where more serious issues are most likely to take hold.

For HR, the task is not to assume these signals are proof of a problem, but to treat them as risk information. They help map the conditions where harassment is more likely to develop – conditions which can hopefully be changed with early, proportionate action. Spotting and taking action on the small shifts is how you prevent the big ones.

Understanding Intersectionality and Power Imbalance Risks

Sexual harassment risk isn’t distributed evenly across a workplace. It increases where power dynamics are unequal, where some employees are marginalised or underrepresented, and where people hold less influence, visibility, or job security.

Intersectionality helps identify these elevated-risk environments. Employees may face compounding vulnerabilities based on gender, age, gender identity, race, disability, cultural background, sexual orientation, or employment status. When these factors overlap, risk increases and willingness to speak up decreases.

Typical intersectional and power-imbalance signals include:

• seniority or authority gaps that go unexamined
• culturally diverse employees withdrawing or changing shifts to avoid certain colleagues
• women or minorities being one of few (or the only one) in a male-dominated team
• precarious or early-career workers reluctant to raise concerns
• “high-exposure” roles where opportunities or hours depend heavily on one individual
• team cultures built around banter, in-jokes, or stereotypes
• patterns of discomfort affecting the same groups repeatedly

Research consistently shows that unchecked power imbalances, rigid gender norms, and exclusionary culture significantly heighten harassment risk.

For HR, this shifts prevention from generic training to targeted interventions:

• extra attention to high-risk pockets
• supporting managers to recognise when power or identity shapes risk
• ensuring reporting pathways are safe for everyone, not only those with organisational confidence
• checking whether workplace culture is experienced differently across demographic groups

Intersectionality doesn’t complicate prevention. It pinpoints where harm is more likely and where support must be strongest.

“Reasonably Practicable” Isn’t a Loophole – It’s a High Bar

Safe Work Australia’s position is blunt: reasonably practicable means doing everything you reasonably can to eliminate or minimise risks to safety. It is intentionally a high standard.

Ask yourself… Given your size, resources, and risk profile, did you take all steps a prudent organisation should have taken to prevent harm?

For harassment and sexual harassment, that can include:

1. Tracking patterns, not just incidents
2. Addressing known leadership risks (not waiting for “one more concern”)
3. Consulting workers about emerging psychosocial hazards
4. Properly resourcing prevention – not delegating it to HR alone
5. Ensuring controls (training, reporting pathways, behavioural expectations) actually change behaviour
6. Ensuring responses are trauma-informed, culturally safe, and supportive for people who report
7. If more could have been done, the expectation is that it should have been done.
8. For HR, this provides clarity: proactive prevention isn’t optional. It’s the standard.

If you could have done more, the law assumes you should have.

Frontline Managers Can Be Powerful Risk Controls

Across Positive Duty cases, WHS investigations and cultural research, frontline managers regularly emerge as the strongest predictor of culture outcomes.

A manager who avoids conflict, ignores discomfort, or minimises behaviour is a risk exposure.

A manager who intervenes early with fairness, consistency, and sensitivity is a control measure.

The most effective controls are behavioural, not administrative:

• Noticing shifts in mood, performance, or team dynamics
• Addressing disrespect early, at the lowest level
• Responding consistently and without favouritism
• Rebalancing workloads and job demands that create vulnerability
• Calling out exclusion, stereotyping, and misuse of power
• Consulting workers before making changes that might increase psychosocial risk
• Closing the loop when issues are raised

Using the Hierarchy of Controls to Strengthen Cultural Safety

Safe Work Australia confirms that the hierarchy of controls applies to sexual and gender-based harassment, but it must be interpreted through a psychosocial lens. The aim is to address the organisational drivers of harm, not shift responsibility onto individuals.

Higher-order controls are generally the most effective. Below is a non-exhaustive set of examples illustrating how they translate to cultural and behavioural risk:

Elimination
Remove known drivers of harassment and disrespect. Examples includes:

• addressing harmful behaviours early
• removing or redesigning roles where power is concentrated in one individual
• eliminating unsafe norms, banter, or environments where harassment is normalised

Substitution
Replace risky practices with safer alternatives. Examples include:

• structured supervision instead of informal gatekeeping
• transparent decision-making instead of discretionary “favourites systems”
• formal mentoring instead of unmonitored one-on-one interactions

Engineering Controls (work design and systems)
Create systems that reduce opportunity for harm and increase visibility and safety. For example:

• psychological safety norms in meetings
• transparent rosters and shift allocation
• multiple safe reporting options
• redesigned workspaces that reduce isolation or vulnerability

Administrative Controls (policies, processes, education)
Policies, reporting pathways, respectful behaviour guidelines, complaints handling frameworks, and clarity about behavioural expectations.
These are critical but not sufficient on their own.

Supportive Controls (least effective alone)
Resources such as EAPs, wellbeing initiatives, or resilience programs. Useful, but not a substitute for fixing structural or behavioural risks.

Metrics Matter – But Not All Data Tells the Truth

A modern review process isn’t just about checking whether actions were taken. It’s about understanding whether your organisation is seeing reality clearly. And the uncomfortable truth is that not all data is reliable, especially when it comes to sexual harassment, psychological safety, and psychosocial harm.

Many organisations still lean heavily on lag indicators like reported incidents, complaints, or workers’ compensation claims as their measure of cultural health. But these figures often represent only a fraction of what is actually happening.

The EML Inside the Minds of Australia’s Workplaces study highlights this gap starkly. Their research shows a significant disconnect between workers’ lived experience and the claims data organisations rely on. Psychological injury claims capture only a tiny proportion of the harm occurring in workplaces not because issues aren’t happening, but because most people never escalate their experience to a formal claim.

For sexual harassment, the gap is even wider. Under-reporting is the norm, not the exception.
Barriers can include:

• fear of repercussions
• power imbalance
• cultural or gendered stigma
• lack of trust in internal processes
• past experiences of issues not being acted on
• concerns about confidentiality or being disbelieved

This means organisations that rely solely on formal data sets inadvertently create a false sense of security. A low number of reports doesn’t mean a low number of incidents. It may simply mean people don’t feel safe reporting.

So what metrics actually matter?

Effective review requires a broader data set, including:

• Sentiment indicators – pulse checks, exit patterns, team climate surveys
• Micro-signal tracking – early behavioural cues and patterns managers observe
• Work design indicators – workloads, rostering fairness, role clarity
• Managerial behaviour markers – conflict avoidance, inconsistent responses, lack of follow-through
• Hotspot analysis – teams or leaders with repeated low-level issues, high turnover, or increased informal concerns
• Intersectional risk patterns – identifying groups less likely to report but more likely to experience harm
• Pathway trust – data on which reporting channels people actually use and why

When combined, these can give a far richer and more accurate picture than claims or complaints alone…. it ensures leaders can understand the truth of people’s lived experience – not the illusion created by low reporting numbers.