Proposed Amendments Public Service Act (2020): Restoring Roles or Removing Safeguards?

20/05/2025

This first paper in the Government’s reform package promises clarity but risks control. By reviving the sharp lines of the 1988 Act, it restores focus on ministerial direction and trims back sprawling structures. But it also strips out equity clauses, weakens long-term stewardship, and risks reducing neutrality to obedience. The question for Parliament is whether this is genuine accountability or just a narrower, more brittle public service.

The first paper in the government’s suite of reforms to the Public Service Act 2020 sets the tone. Titled “Clarifying the role and responsibilities of the Public Service”, it claims to tidy up the Act and bring focus back to what matters: supporting ministers, delivering efficiently, and doing so with competence and neutrality.

That’s the sales pitch. But before Parliament gives the Bill the green light, it may need to be clear about three things: what the Bill (when we finally see it) restores, what it risks, and what must be asked if we’re serious about democratic accountability.

What it restores: A sharper edge and a clearer line of sight

At face value, the paper brings back a kind of clarity. It restores the old language of “principal responsibilities” for chief executives, a throwback to the State Sector Act 1988, and rearranges their duties to foreground giving advice, following ministerial instructions, and delivering services efficiently.

It also calls time on sprawling interdepartmental structures, requiring end dates and reviews for Executive Boards, Functional Chief Executives, and Departmental Agencies. That’s a sound instinct. It reins in complexity and reminds the system that machinery-of-government design should not be permanent by default. A close look at the now-disclosed workings of the various executive boards suggests they became forums for information-sharing rather than joint decision-making. They became the embodiment of “this could have been an email.” Under these reforms, such bodies would be subject to regular review and potential dissolution if they no longer serve their original purpose.

And it tries, in parts, to remove duplication. While well-intentioned, the Long-Term Insights Briefings were over-prescribed, underwhelming, and barely free or frank. In practice, most departments produced these documents with minimal public impact, with several briefings receiving fewer than 30 public submissions. The proposal to centralise them under DPMC once per term isn’t radical, and may improve coherence, if done well.

In all these areas, there’s an attempt to tighten the line between ministerial direction and public service delivery. That’s the restoration: a cleaner institutional frame with fewer side missions. It’s a tightening of the knots, a sharpening of the pattern. That ought to be welcomed.

But it’s only half the story.

What it risks: Stripping back the parts that hold democratic value

The same changes that claim to streamline could also narrow the scope of the public service. This paper doesn’t just remove clutter. It potentially removes safeguards, and safeguards are not decorative in the tukutuku of good government. They hold the pattern.

First, diversity, inclusion, and pay equity are cut entirely from the legislation. The reasoning is that future governments can pick these priorities up via policy tools, including workforce strategies, letters of expectations, and so on. I accept that these clauses may have appeared performative in practice: the public service workforce profile has only marginally improved since their inclusion, with Māori representation increasing from 16.4% to 17.3% between 2020 and 2023, according to Te Kawa Mataaho data. However, removing statutory obligations makes equity optional again, and optionality is how inequality survives in systems built for sameness. Policy can change overnight. Law creates a floor.

Second, the emphasis on “serving the government of the day” is sharpened to a point. The role of advice is reaffirmed, but only insofar as it supports lawful instructions. In the Bill, I will be looking to see if the broader constitutional function of the public service, including its role in maintaining continuity, long-term thinking, and contestable knowledge, is acknowledged. This is an appropriate correction. As I’ve written elsewhere, the theory of public value has been widely embraced in Wellington, often without interrogation. Its appeal lies in positioning the public servant as a mission-driven actor, pursuing value on behalf of the community. But in practice, it has had the unintended consequence of displacing political accountability and allowing officials to claim legitimacy through stakeholder engagement and consultation, rather than through transparent alignment with elected authority. The result has been a fraying of constitutional threads. This clarification, therefore, is at least trying to restitch those threads. The risk is that it does so by pulling the weave too tight: restoring ministerial control, but at the cost of community, of pluralism, of those who don’t have access to ministers, of foresight, and of professional independence. In our system of governance, the Cabinet is expected to govern for all, not just those it regards as its base.

I also note some repositioning of the “spirit of service.” Previously a standalone clause, it’s now tucked into the purpose section and stripped of its symbolic force. That may seem trivial, but symbolism isn’t fluff. It shapes culture. The “spirit of service” has been invoked during crises like the Christchurch earthquakes and COVID-19 to rally public servants beyond narrow job descriptions. If service is reduced to execution, not ethos, we lose the very thing that anchors public trust. We’ll see what the Bill says. I am mindful, though, that the spirit of service has leaned too heavily on the theory of public service motivation and was unbundled from outcomes, accountability, and performance. Even so, it is a thread that should not be quietly unpicked.

What Parliament should ask before saying yes

Is this really about clarity, or is it about control? If the intent is to clean up confusion, that’s welcome. But if the result is a public service that speaks less, questions less, and reflects less, we’re not clarifying. We’re shrinking. A legitimate alternative might be to maintain the broad purpose statements while providing clearer operational guidance through secondary instruments like the Commissioner’s standards, preserving both values and practicality.

Who benefits from removing diversity and equity clauses? Not the workforce. Not the communities most affected by systemic gaps. Reasonable people can argue that operational policies rather than legislation are better vehicles for workforce composition goals. The Public Service Census shows relatively high engagement scores among diverse groups, suggesting some progress. But Parliament should still ask why a commitment to fairness suddenly counts as a “distraction,” and how the Public Service Commissioner ensures the public service remains representative without statutory backing. A public service that looks like the people it serves is part of our democratic weave. Remove the thread, and the panel distorts.

What happens to the long-term voice? A single insights briefing per term, run out of DPMC, risks becoming a political echo chamber. I’ve criticised the current briefings for being underwhelming, but the solution may not be centralisation. A better approach might be fewer, more targeted briefings with greater resources and independence. Parliament should ask what mechanisms will remain to ensure that anticipatory thinking, risk surfacing, and cross-agency learning persist. The Treasury, the former Productivity Commission, the Infrastructure Commission and the Climate Change Commission’s reports offer useful models. It’s not about weaving more panels, it’s about strengthening the ones we already have.

Where is the cost-benefit analysis? This paper is light on implementation details. There is no serious account of what it will take, in cost, capability, or time, to reshape chief executive responsibilities or wind down unnecessary interdepartmental bodies. It is unclear what the changes will cost. Parliament should demand regulatory impact statements before proceeding. Woven change still needs stitching.

Are we strengthening neutrality, or misunderstanding it? Real political neutrality isn’t blind loyalty to ministers. It’s institutional integrity, professional independence, and the capacity to speak inconvenient truths while still delivering in double-quick time. It’s also the ability to speak to the concerns of people not in the room. And yes, the service is becoming more political: that’s not just me vibing about it, this concern is backed by research. Chris Eichbaum and Richard Shaw’s research confirms it. Their 2017 survey of New Zealand public servants found a marked increase in what they call functional politicisation: where public servants internalise political responsiveness as a performance norm. This isn’t political advisers bullying officials. It’s officials anticipating ministerial preferences, often at the expense of contestable policy thinking. It’s a warp in the pattern: a bending of neutrality toward responsiveness. Responsiveness is not always neutrality. The 2018 State Services Commission investigation into the leaked Budget reinforced how critical the perception of neutrality is to public confidence. That trust rests not just on behaviour, but on architecture. If the law fails to defend political neutrality as something deeper than obedience, we all pay: in poor decisions, lost talent, and public distrust. That’s why I’ll be examining the Bill. To make sure it strengthens neutrality, not misreads it. Because if we get this wrong, we won’t just lose good advice, we’ll lose the frame that holds the pattern.

The real task ahead

Supporters of these reforms will argue that they restore democratic control. That elected ministers should set the agenda, and public servants should deliver. From that position, anything that blurs the line, such as long-term reports, equity clauses, expansive stewardship duties, weakens the link between the ballot box and the machinery of government.

That’s not wrong. The 2020 Act pulled chief executives in too many directions. It made their jobs harder. These proposals begin to correct that. But it’s not the whole picture.

Democracy isn’t just a handover of power every three years. It’s the daily practice of balancing competing interests, surfacing long-term risks, and holding public values that can’t be settled electorally. That work happens in the practical state where policy meets implementation, where advice is tested not for loyalty but for rigour and implementation confidence, and where institutions hold memory, complexity, and the capacity to say “yes, but not yet and not all at once.”

This paper tightens the bolts on majoritarian authority. But in doing so, it risks pulling too hard on the threads that give strength and meaning to our constitutional democracy. It builds responsiveness, but risks distorting the pattern. It clarifies roles but may unweave the very strands that link the public service to Aotearoa’s diverse people, histories, and futures.

A well-made tukutuku holds tension, but not at the expense of integrity. If the weave becomes too rigid, it can no longer flex with the weight of complexity, nor hold space for the relationships and values that give it life.

The question isn’t whether the public service should serve elected governments. Of course it should. The question is whether it can serve meaningfully, if stripped of qualities like independence, foresight, and fairness: the same qualities that give our public service depth and strength, and let officials express their love for the nation through their service to it.

That’s the task for Parliament. To ensure clarity doesn’t become control and cost. And that in strengthening one part of the frame, we don’t unpick the panel.