A Step Backwards: Why Scrapping “Good Employer” Requirements Will Cost the Public Service Its Best Disabled Talent

For more than a decade, New Zealand’s public service has worked to build a reputation as one of the few genuinely safe places for disabled people to build a career.

That reputation didn’t happen by accident. It was built deliberately, through “good employer” obligations written into law — obligations that required chief executives to actively remove barriers, foster inclusive workplaces, and recognise the distinct employment needs of groups, including disabled people, who have historically been locked out of the labour market.

ACT’s proposed “reset” of the Public Service Act would tear that framework out.

Alongside removing what leader David Seymour calls “race-based hiring requirements,” the policy would strip out the legislative provisions recognising the employment needs of disabled people, women and Māori, and replace them with a single instruction: hire on merit alone.

Seymour’s justification is that “health and safety, non-discrimination, and disability rights are already covered elsewhere in the law” and that public service jobs should go to “the best person for the job, regardless” of background.

It sounds reasonable right?. It is also, in practice, a policy that will make it measurably harder for disabled New Zealanders — especially disabled graduates just starting out — to get a foot in the door of the public service at all.

The problem with a purely merit-based hiring model is not that merit is a bad idea.

It’s that standard recruitment processes are not neutral instruments for measuring merit — they are built around the assumptions, timelines and communication styles of non-disabled applicants.

Timed written tests disadvantage people with certain cognitive or sensory disabilities regardless of their actual capability.

Interview panels routinely misread neurodivergent communication styles as a lack of confidence or competence.

Job descriptions loaded with vague, unnecessary requirements filter out capable disabled candidates before they even get to demonstrate what they can do.

None of this is deliberate discrimination in the way discrimination law is designed to catch — which is exactly why “non-discrimination law already covers it” is a weaker safeguard than it sounds.

Formal non-discrimination law prohibits an employer from rejecting someone because of disability; it does nothing to fix a selection process that quietly filters disabled candidates out before a human ever consciously discriminates against them.

Good employer provisions exist precisely to close that gap — not by lowering the bar, but by requiring agencies to actively design recruitment processes that let disabled candidates show their actual ability against the job’s real requirements, rather than against a set of unstated, inaccessible defaults.

New Zealand doesn’t have to guess what happens when this scaffolding is removed or weakened — Australia’s experience with disability employment in its own public service is instructive, and cuts directly against the idea that “pure merit” produces fairer or better outcomes for disabled applicants.

The Australian Public Service spent years missing its own targets for disabled employment under standard merit-based recruitment, to the point where the Commonwealth concluded that:

ordinary, unmodified merit selection was itself a structural barrier.

The Australian Public Service Commission’s own strategy work found that agencies had to shift from a “passive” merit process to an “active” one specifically because agencies needed to present themselves directly to prospective disabled employees, who often didn’t see the APS as a genuine possibility for them under standard hiring approaches.

That is why Australia built the RecruitAbility in the first place: an affirmative measure that lets disabled applicants who meet the minimum requirements of a role progress to the next stage, rather than being screened out by a competitive shortlisting process that, left to run on its own, systematically filtered them out before they could demonstrate their skills.

Notably, RecruitAbility does not override merit, which remains the basis for actual engagement and promotion — it simply ensures disabled candidates get to compete on it in the first place.

The existence of that scheme is itself evidence that unmodified “best person for the job” recruitment was not delivering fair outcomes for disabled applicants; it was delivering outcomes skewed by inaccessible process design.

Australia’s Disability Royal Commission went further, examining the public sector’s merit principle directly and recommending it be clarified precisely because governments recognised that public services have a crucial national role as leaders in modelling inclusive employment practice, and that this requires deliberate action rather than a hands-off approach to recruitment.

Even now, Australian federal agencies openly state that their goal is to be an “employer of choice” for disabled people — language that implicitly acknowledges disabled jobseekers do not automatically see, or get a fair shot at, public sector roles without active outreach and process redesign.

The broader Australian evidence on disability employment tells the same story from a different angle.

Research into barriers facing disabled jobseekers has found that the likelihood of finding work falls as the number of structural and non-vocational barriers a person faces increases — barriers that have nothing to do with a candidate’s actual ability to do a job, and everything to do with systems designed around non-disabled norms.

A 2025 Grattan Institute analysis similarly found that despite more than a billion dollars a year in public investment, outcomes for disabled jobseekers in Australia remain persistently weak, particularly for those with higher support needs —

A reminder that formal equality of opportunity, on its own, does not close employment gaps that were built up over decades.

The New Zealand public service’s good employer obligations are the domestic equivalent of Australia’s affirmative measures: they require agencies to actually design recruitment, onboarding and workplace practices around disabled people’s needs, rather than assuming a “neutral” process will naturally produce fair outcomes.

Remove the legal requirement, and there is no reason to expect agencies experiencing constrained budgets and political pressure to prioritise workplace adjustment programmes, accessible recruitment pipelines, or graduate schemes aimed at disabled candidates.

These programmes exist because they were required, resourced and monitored. Take away the requirement, and the resourcing quietly follows it out the door — not necessarily through any single dramatic decision, but through years of agencies deprioritising work that is no longer mandated.

That matters enormously for disabled graduates specifically. Graduate programmes are often the only route into professional public sector careers for young disabled people who lack the kind of “networked” pathway into work that non-disabled graduates can rely on.

Removing the statutory requirement to actively recognise and address disabled New Zealanders’ employment needs sends a clear signal to a graduate weighing up whether the public service is worth the effort: the institution is no longer required to make room for you.

None of this means disabled people can’t succeed under a merit system — plainly, many do.

The point is narrower and better supported by the evidence: an unmodified merit process, stripped of the active measures that “good employer” provisions require, has consistently been shown — in Australia and elsewhere — to filter out capable disabled candidates for reasons that have nothing to do with their ability to do the job.

Fixing that took years of deliberate policy work. Repealing the legal requirement to keep doing it puts all of that at risk.

Seymour’s central defence of the policy is that disability rights don’t need to live in the Public Service Act because they’re already protected elsewhere — in the Human Rights Act and general non-discrimination law.

On its face this sounds like a technicality rather than a real reduction in protection.

It isn’t.

It misunderstands what “good employer” provisions actually do, and the distinction matters enormously for disabled jobseekers.

General non-discrimination law is a prohibition. It tells an employer what it cannot do: it cannot refuse to hire someone, or treat them worse, because they are disabled. It is enforced reactively, after the fact, and it requires a disabled applicant to prove that disability was the reason they were rejected — a genuinely difficult thing to establish when a hiring panel can point to any number of “objective” criteria a candidate technically didn’t meet.

Non-discrimination law does nothing to fix a selection process that was never built with disabled applicants in mind in the first place. A timed online test, an assessment centre with no sensory accommodations, a job ad stacked with unnecessary “must haves” — none of these breach the Human Rights Act, because none of them involve anyone consciously deciding to exclude a disabled person. They exclude disabled candidates anyway, quietly, by design.

“Good employer” provisions are a duty — they require agencies to proactively build recruitment, onboarding and workplace processes that work for disabled people, rather than waiting for a disabled candidate to be excluded and then prove it in a legal complaint.

That is precisely the gap Australia’s RecruitAbility scheme was built to close, and it is telling that Australia felt the need to build it at all: the APS already had, and still has, exactly the kind of general non-discrimination protections Seymour points to.

Those protections did not, on their own, get disabled Australians into public service jobs at rates anywhere near proportional to their share of the population. That is why the affirmative, process-level scheme exists on top of the general law, not instead of it.

Seymour’s claim that non-discrimination law “already covers” disability rights is therefore true only in very narrowest sense — and precisely the sense that doesn’t do the practical work of getting disabled graduates hired.

It covers the rare case of a hiring manager who explicitly and provably rejects someone because they are disabled.

It does nothing for the far more common case: a capable disabled graduate who never gets shortlisted because the process itself wasn’t built for them, and who has no legal claim because nobody ever said the word “disability” out loud.

That is exactly the gap “good employer” duties exist to close — and exactly the gap this policy would reopen.