Public Sector Reform: Te Kahu Tuatini: The State of the Public Service Briefing 2025

10/12/2025

The Public Service Commissioner’s second three-yearly briefing on the state of the public service, released this month, warrants close attention. Sir Brian Roche has produced a document that names structural problems with his usual frankness and proposes architectural change rather than incremental adjustment.

The statutory architecture

Schedule 3, clause 16 of the Public Service Act 2020 requires the Commissioner to brief the Minister on the state of the public service at least once every three years, with the Commissioner determining the subject matter based on issues considered to be in the public interest. The Minister must then present the briefing to Parliament. This mechanism sits alongside the separate Long-term Insights Briefings required of departmental chief executives under Schedule 6.

The three-year cadence is less frequent than comparable Westminster arrangements. Australia’s Public Service Commissioner must provide an annual State of the Service report to Parliament under section 44 of the Public Service Act 1999: a tradition extending back over a century. Canada’s Clerk of the Privy Council produces an annual report to the Prime Minister under the Public Service Employment Act, a requirement established in 1992 when the Clerk’s role as Head of the Public Service was given a statutory foundation; the report must be tabled in Parliament, and the current edition is the thirty-second. The United Kingdom has no direct equivalent, but the Civil Service Commission’s annual report focuses on recruitment and merit-based appointments under the Constitutional Reform and Governance Act 2010, with broader system assessment left to capability reviews and parliamentary inquiries.

A shift in register

The contrast with the first briefing, produced by Peter Hughes in December 2022, is instructive. That document emphasised progress toward “a leading edge, unified, trusted Public Service” and reported on diversity initiatives. It discussed the institutional tools enabled by the 2020 Act, including the interdepartmental executive boards, system leadership roles, and functional leads, as mechanisms for improved coordination.

Roche’s 2025 briefing adopts a different tone. He assesses the system as “not as well positioned as it needs to be for the future” and, significantly, names the source: the reforms of the 1980s and 1990s, which he argues “hard-wired silos into our operating model, delivering specialisation at the time, but at a cost.”

This matters. For three decades, official discourse has tended to treat the NPM reforms as foundational achievements requiring minor refinement rather than reconsideration. Matheson’s 1998 analysis had already identified “atomization and short-termism” as consequences of the reforms, noting that the sharp distinction between policy advice and service delivery, combined with output-focused accountability, created strategic coordination problems. The 2001 Review of the Centre confirmed these concerns, finding that fragmentation had created significant barriers to addressing cross-cutting policy issues. The response preserved the essential contractual architecture whilst adding coordination mechanisms. The 2006 Review of Central Agencies built on this work, representing a genuine effort to strengthen the system’s capacity for integrated investment advice and whole-of-government perspectives and definitions.

Yet each successive adjustment worked within the existing framework rather than reconsidering its foundations. Roche’s briefing departs from this pattern. Rather than proposing another layer of whole-of-government initiatives atop a fragmented base, it calls for reorganising the base itself.

The proposed direction

The briefing identifies 43 departments, departmental agencies, and interdepartmental executive boards currently operating: more, the briefing notes, than most comparable small advanced economies. The proposed remedy involves consolidating agencies around common platforms, citizens, businesses, or markets, with some agencies becoming delivery units within larger portfolio departments whilst retaining distinct identities and functions.

The digital transformation agenda provides concrete commitments: a Government app serving as a unified digital front door, a trusted digital identity ecosystem, and centralised digital investment and procurement under an expanded Government Chief Digital Officer role modelled on Singapore’s GovTech. The Cabinet has agreed to this centralisation, with projected savings of $3.9 billion over five years. Indeed, that app launches today.

On artificial intelligence, the briefing acknowledges that Aotearoa has been slower than leading jurisdictions to adopt AI systematically. The national AI strategy was launched in July 2025, and the GCDO’s coordinating role is positioned as a response.

What comes next

The diagnosis is sound and the direction appropriate. What the briefing establishes is strategic intent; what follows must be implementation specificity.

There is no indication yet of which agencies might consolidate, what the target configuration looks like, or what transition costs and timeline are anticipated. This is appropriate for a direction-setting document: implementation detail belongs in subsequent Cabinet papers and machinery of government decisions. But momentum matters. Though in the briefing itself, the Commissioner acknowledges: “… the complexity, and the courage required, for these challenges to actually be addressed.”

The current fiscal environment may create conditions that would otherwise encounter greater resistance. Agency consolidation is never straightforward: it affects chief executives, organisational identities, and ministerial relationships, but the pressure to deliver more with less creates both necessity and opportunity. I have written elsewhere that change for change’s sake is not good government, but if form follows function, and some functions are better aggregated, and the business case makes sense, then Wellington must push on. The briefing identifies that structural alignment offers economies of scale: shared corporate services, reduced duplication, integrated digital platforms, and coordinated procurement. These are the right places to look for sustainable efficiency, rather than headcount reductions that risk hollowing out capability. At this point, cutting FTEs without addressing underlying fragmentation distributes the same workload across fewer people; genuine efficiency comes from working differently, not just working harder with less.

Looking ahead

One area where subsequent work might usefully develop concerns the relationship between structural reform and the constitutional foundations the briefing invokes. “Integrity, political neutrality, and free and frank advice” appear as principles to be preserved through transformation. How agencies are configured will shape what advice they can provide; how the system is organised affects the conditions under which free and frank advice is possible. The proposed Public Service Amendment Bill and its implications for chief executive responsibilities deserve attention as this reform programme develops.

Te Kahu Tuatini 2025 represents a welcome development in official thinking about public administration in Aotearoa. While it is a modest document, it names the structural source of coordination difficulties, and an architectural change has been proposed. The task now is implementation: moving from diagnosis to delivery with the momentum the moment provides.

References

Central Agencies. (2006). Review of central agencies: Technical report. New Zealand Government.

Matheson, A. (1998). Governing strategically: The New Zealand experience. Public Administration and Development, 18(4), 349–363

Ministerial Advisory Group. (2001). Report of the Advisory Group on the Review of the Centre. New Zealand Government.