From Virtue to System Design: What the new Code of Conduct tells us

10/02/2026

Last week I traced the way cabinet committees function as the coordination machinery of executive government: the institutional furniture through which political intent becomes administrative action. Across all of their iterations, the committee system addressed a persistent problem: how to get coherent decision-making across fragmented portfolios, competing priorities, and the relentless pace of the policy cycle. That same problem now appears, less visibly but no less consequentially, in the domain of integrity.

On 30 March 2026, the new Code of Conduct for the Public Sector | Te Tauākī Whanonga mō te Rāngai Tūmatanui comes into force, replacing the Standards of Integrity and Conduct that have governed behaviour in the state services since 2007.

This post is about what sits beneath that change. It is as I argue below, an overdue reckoning. This post begins with the assumptions embedded in the 2007 settlement: a framework that placed its confidence in individual professional judgement to hold the kawanatanga together. I then read the 2026 Code as a structural response: not a change in values, but a re-housing of them within system architecture, leadership obligations, and stewardship over time. I end with the question that neither document adequately confronts: what it now means to be a public servant when the locus of integrity shifts from the person to the system.

The 2007 settlement: integrity as personal obligation

To understand what is changing, one needs to read the 2007 framework properly. Most people did not. The visible artefact was the poster: Fair, Impartial, Responsible, Trustworthy. A mnemonic. A wall decoration in every Wellington reception area. Four words doing the work of an entire constitutional settlement.

But the poster was never the code. The substance sat in Understanding the Code of Conduct: Guidance for State Servants, published by the State Services Commission in 2007 and updated in 2010. Grounded in the State Sector Act 1988, it offered explanations rather than instructions: general principles to be applied with judgement, addressed to the reader as a reasoning professional. The tone was deliberative throughout. It assumed you could be trusted to work things out.

That assumption ran deep. Political neutrality, free and frank advice, the “no surprises” convention, conflicts of interest: each was acknowledged as genuinely difficult, and each was left to be navigated through professional skill and good faith rather than prescribed in detail. The organising logic was consistent: individual judgement, exercised conscientiously, was sufficient to hold the system together.

That confidence rested on a further assumption: that there was time. Time to prepare advice, to reason through conflicts, to disclose difficulties, to let good faith do its quiet work. Leadership obligations were lightly specified. Senior officials were expected to model the standards, but leadership was not treated as a distinct institutional category with its own architecture of accountability. The system imagined that its people were, by and large, reliable, and that what they needed was space rather than scaffolding.

In the conditions of its time, this was not naïve. It worked, or at least well enough that its limitations were not yet legible as systemic risks. But individual virtue, distributed unevenly across a large and complex set of policy and delivery systems, does not reliably produce coherence at the system level. It produces pockets of excellence surrounded by inconsistency. Under contemporary conditions of speed, scrutiny, and collapsing boundaries between public and private life, that inconsistency has become a vulnerability rather than a tolerable cost.

The 2026 Code: integrity as system asset

The 2026 Code does not abandon political neutrality, impartiality, honesty, or service. It re-houses them. Where the 2007 framework addressed itself to individual state servants exercising judgement, the 2026 Code addresses itself to a system: to organisations, leaders, and the institutional conditions under which a culture of integrity must be sustained over time.

The taxonomic changes signal this. The four standards give way to five values and five statutory principles drawn from the Public Service Act 2020, each paired with a te reo Māori descriptor. These pairings are not decorative. They function as constitutional markers of a bicultural state articulating its expectations with increasing institutional confidence. “Responsible” becomes “Accountable”, shifting emphasis from personal conscientiousness to answerability within a system. The addition of “Respectful” and “Responsive” signals that legitimacy now turns not only on lawful action, but on how the state listens and engages with those it serves.

But the deeper shift sits in two places.

Stewardship (kaitiakitanga) is introduced as a formal principle: not a vague injunction to think about the future, but a claim about the temporality of the state. Public servants bear obligations not only to the government of the day but to the continuity of institutional capability across time: preserving knowledge, developing workforce capability, and sustaining the capacity to engage with Māori and understand Māori perspectives. Where the 2007 guidance gestured towards these concerns, the 2026 Code makes them architectural.

Leadership is reconceived in similar terms. The 2007 framework assumed that modelling good behaviour would suffice. The 2026 Code specifies obligations: fostering ethical cultures, ensuring staff understand the Code, encouraging speaking up, responding consistently to integrity concerns, and aligning organisational policies with the Code’s requirements. This is what an institutional or regulatory designer does when exhortation and expectation setting is no longer enough.

What it means to work inside the new architecture

Here is the question that the transition opens but neither document adequately addresses. If the 2007 framework said, in effect, you are a professional; act accordingly, and the 2026 Code says you are part of a system; the system must be designed to produce integrity, what does it now mean to be a public servant inside that second architecture?

This is not an abstract concern. Codes do not govern by themselves. They are mediated through position descriptions, leadership frameworks, recruitment material, performance systems, and the quiet scripts through which people learn what their role permits and requires. These are the places where a conception of integrity is either lived or merely professed. If integrity is now a system property, that understanding must be translated into differentiated role obligations — common where they must be common, explicitly varied by seniority, function, and proximity to ministers. At present, those artefacts appear only weakly aligned with either settlement. The coherence that the 2026 Code demands at the level of principle will need to be built, agency by agency, in the architecture of everyday work.

And there is something more personal at stake. For a generation of public servants formed under the 2007 settlement, integrity was something you carried with you: a mark of professional identity, not a system output. The 2026 Code asks them to accept that their judgement, however good, is no longer treated as sufficient. That is not a comfortable message. It may not even be a welcome one. But it reflects a recognition that has been a long time coming: that coherence at the system level does not emerge from the accumulated good faith and good work of individuals. It has to be designed, maintained, and led. The cost of failing to do so is borne not by institutional designers, but by the people the system is meant to serve.

References

State Services Commissioner. (2007). Standards of Integrity and Conduct. Wellington: State Services Commission. ISBN 978-0-478-30309-4. Available at:
https://www.publicservice.govt.nz/guidance/standards-of-integrity-and-conduct

State Services Commission. (2007; revised 2010). Understanding the Code of Conduct: Guidance for State Servants. Wellington: State Services Commission. ISBN 978-0-478-30313-1. Available at:
https://www.publicservice.govt.nz/assets/DirectoryFile/Guidance-Understanding-the-Code-of-Conduct.pdf

Public Service Commissioner. (2026). Te Tauākī Whanonga mō te Rāngai Tūmatanui | The Code of Conduct for the Public Sector. Wellington: Te Kawa Mataaho Public Service Commission. ISBN 978-0-473-77623-7. Effective 30 March 2026. Available at:
https://www.publicservice.govt.nz/guidance/the-code-of-conduct-for-the-public-sector

Notes

Institutional nomenclature and legislative context

The institutional nomenclature itself traces the shift this post describes. The 2007 documents were issued by the State Services Commission, under the authority of the State Services Commissioner and the State Sector Act 1988. The 2026 Code is issued by Te Kawa Mataaho Public Service Commission, under the authority of the Public Service Commissioner and the Public Service Act 2020. The change is not merely cosmetic. The State Sector Act 1988 was the legislative expression of the New Public Management reforms: it conceived of the state services as a collection of discrete agencies, each managed by a chief executive or a board on something approaching a contractual basis with a responsible Minister. The Public Service Act 2020 reasserted a unified public service, introduced the principles that now underpin the 2026 Code, and gave statutory expression to concepts that the earlier legislation had either left to convention or not contemplated at all. Even the shift from “State Services” to “Public Service” carries institutional weight: “state services” described an administrative apparatus; “public service” makes a claim about role and purpose. The language tells the same story as the documents themselves: a state moving from a managerial settlement organised around agency-level accountability to a constitutional settlement organised around system-level coherence. Whether that coherence has in fact been achieved remains an open question.

A note of caution on reach

The title Code of Conduct for the Public Sector invites a broader reading than the statutory framework may sustain. Under section 17(2) of the Public Service Act 2020, the Code applies to public service agencies, Crown agents, certain other Crown entities (excluding tertiary education institutions, school boards, Crown research institutes and their subsidiaries), Schedule 4A companies, and the Parliamentary Counsel Office. It does not, on its face, extend to state-owned enterprises, wānanga, or local government. The 2007 Standards drew broadly similar boundaries under section 57 of the State Sector Act 1988, stopping at the state services perimeter and leaving the wider state sector to determine their own codes. The shift in terminology from “state services” to “public sector” might reasonably be expected to signal an expansion of scope, not least because in earlier and common usage “public sector” encompassed local government. Whether it does so here requires closer examination than I have undertaken. What can be said is that the Code has acquired the character of a regulatory system instrument. The question is whether we yet fully understand who the regulated parties are within that system. More analysis will be needed as the instrument goes live.