The decisions a merger will not fix
Amalgamation changes the structure. It does not change judgement.
We've been talking a lot about structures, mergers and reform. But I've been wondering whether we're asking the right question. Changing an organisation's structure doesn't necessarily change how decisions are made. And in the end, that's what will determine whether reform succeeds. I've explored this in the first article of a three-part series on governance, systems and people, and what really shapes successful organisational change.
Note: First in a three-part series on what actually determines whether these mergers work. Part one looks at governance and decision quality. Part two looks at the system, and what breaks when you join things that were never designed to connect. Part three looks at meaning, and why people accept or reject a change long before they understand it. Structure, system, people. In that order, because that is the order the questions arrive in and the reverse of the order they get answered.
There is a particular kind of meeting happening around the country at the moment. Mayors and chief executives from neighbouring councils, in a room together, with a whiteboard and a deadline, working out what they are going to become. In the Wellington region that room is the Mayoral Forum, eight mayors and the chair of Greater Wellington. Outside it, a public survey ran for under four weeks, asking residents what they thought about a proposition that was still being drafted.
Councils have been told to lead their own reform or have it done for them. Proposals for new unitary authorities are due on 9 August. Regional councillors will not be elected in 2028, and that part is settled. At the same time the public service is being asked to reduce its number of agencies substantially, with the new Ministry for Cities, Environment, Regions and Transport as the first live test of the idea. Thousands of roles will go over the next few years.
The structural questions are getting all the airtime. Who merges with whom. How many councils. Which functions sit where. What the new organisation chart looks like.
Those are real questions. They are also the easier ones.
The harder question is this. When the boxes change, does the organisation actually make better decisions?
Because that is the only thing that would justify any of it.
What the record actually shows
Amalgamation is sold as efficiency. Fewer back offices, less duplication, one set of systems instead of four.
It is worth being clear about how that has gone before. Auckland's super city is the only comparable domestic example, and the record there is mixed in an instructive way. The council points to billions in savings. Academic reviews, including work by Andy Asquith, have found the efficiency gains hard to pin down, noted that staff numbers climbed back above pre-merger levels, and observed household rates rising sharply since 2010. The same reviews found the super city did well on strategic direction and leadership, and poorly on democratic engagement.
That is a useful result, and not the one usually quoted. It suggests amalgamation is not primarily a cost lever. It is a decision-making lever. It succeeded where it improved the ability to think and act at scale. It failed where it broke the connection between decisions and the people affected by them.
Tasman's mayor made a related point when the ultimatum landed. Amalgamation does not necessarily save money and does not necessarily make everything better.
So what does make it better?
Governance is the system that decides how you decide
This word has been worn thin by overuse, so it is worth reclaiming.
Governance is not the committee. It is not the terms of reference, the assurance report, the risk register, or the quarterly pack. Those are artefacts. They are the exhaust, not the engine. Governance is the system an organisation uses to make consequential decisions well and repeatedly. It answers a small set of questions:
- Who has the right to decide this?
- On what information, and how good does that information have to be?
- Who has standing to challenge it before it is final?
- How will anyone know later whether it was any good?
- What happens when it turns out it wasn't?
Everything else is administration.
The research here is settled and unglamorous. Governance studies consistently find that board and committee effectiveness has more to do with the quality of debate, the willingness to dissent, and the information reaching the table than with structure, size or composition. The large-scale programme literature, of which Bent Flyvbjerg's work on megaprojects is the best known, keeps landing in the same place. Big initiatives rarely fail on execution. They fail on the decision made at the front, on optimism about benefits and silence about risk. Delivery only makes that decision visible.
Which means that if you merge four organisations and end up with one that decides the way the four did, nothing has transformed. The same judgement is now being exercised at greater scale, with fewer people able to see it happening.
Three things that break during a merger
The pattern is consistent and rarely dramatic. It is usually quiet.
Decision rights go fuzzy exactly when they matter most
In transition, everyone knows the old delegations are dying and nobody quite knows what the new ones are. So decisions drift. They get made in corridors, in transition team meetings, by whoever has momentum and confidence. Six months later somebody asks who approved that, and the honest answer is that the process did.
The councils working through Head Start proposals are making decisions with enormous downstream consequences. About representation. About which communities keep a voice. About what gets centralised. All under a three-month clock, using delegations designed for a different organisation.
A question worth sitting with. What is the largest decision your transition has made in the last month, and can you name the person accountable for it?
If that takes thinking about, that is the finding.
Challenge disappears
This one is subtle, and it is the one that costs the most. During a merger, dissent gets recoded. Ask a hard question about the benefits case and you are no longer a governor doing your job. You are not on board. You are relitigating. You are a risk to the timeline. Particularly when a minister has said the deadline is non-negotiable and the alternative is having the whole thing imposed on you.
So people stop asking. Not because anyone silenced them, but because the cost of asking quietly went up and the benefit quietly went down. Meanwhile the assurance function, where one exists, gets busy confirming that milestones are green rather than asking whether they are the right milestones.
Wellington City Council deserves some credit here for saying the quiet part out loud. Its Let's Talk consultation on amalgamation (letstalk.wellington.govt.nz/amalgamation), open from 24 June and closing at 11.59pm on 19 July, states plainly that given the Government's tight timelines this is not formal consultation on a detailed proposal, and that formal consultation on specific options will come later, if a proposal is accepted and reaches detailed design. The council decides whether to submit on 6 August. Government makes an in-principle decision in September. Detailed design runs to March 2027.
If you live in Wellington and you are reading this before Sunday, the survey is open and it takes a few minutes. Public forums with elected members are also being held in the last week of July. Whatever the limits of the process, the feedback goes to the mayor and councillors before the 6 August decision, and a low response rate will be read as consent.
That is an honest description of a real constraint, and it is worth reading carefully, because it describes the sequence most councils are in. The community is being asked for a view on a direction before the proposition exists. The proposition then gets built during the phase where the choices actually get made. By the time formal consultation arrives, an in-principle decision has been taken and a great deal has hardened.
None of that is anyone acting in bad faith. It is what a three-month clock produces. But it does mean the challenge has to come from somewhere other than the public, and it has to come during detailed design rather than before it. If the only structured scrutiny in the calendar is a survey that closed in July, the scrutiny is finished before the decisions are made.
Independent review earns its keep precisely here. Not as a compliance tick. As the mechanism that makes it safe and ordinary for someone to say the savings number is wrong, without that being a career event. A second question. When did someone last change a significant decision in your organisation because of something a review raised? If the answer is never, what is the review actually for?
Alignment gets confused with agreement
Everyone signs the proposal. Everyone nods. Then implementation starts, and it emerges that the four parties held four different pictures of what they agreed.
This is already visible. Invercargill's mayor wants two unitary authorities because he believes one would become city-centric. Nelson has wanted to merge with Tasman for over a decade and Tasman has not. Environment Southland's chair says the community has to own it or the back end will not work. Wairarapa may go alone. These are not disagreements about structure. They are disagreements about what the thing is for, and structural agreement will paper over them right up to the moment it cannot.
Alignment is not everyone wanting the same thing. It is everyone able to describe the same thing, including the parts they dislike. That is a higher bar, and it is worth testing before signing rather than after.
What good looks like
Not a longer framework. A shorter list.
Decide what you are optimising for, and say it in one sentence. Cost. Environmental consistency. Regional strategic weight. Better planning under the new resource management system. Local voice. These pull against each other. An organisation that cannot say which one wins when they conflict has not made a decision, it has made a wish. The conflict will surface in year two, in public.
Build the decision architecture before the organisation chart. Delegations, thresholds, what reaches the board and what does not, who can stop things. Do it early, while it is still abstract and therefore still negotiable. Once real people hold real roles, every delegation conversation becomes a conversation about status, and the design loses.
Put a real challenge mechanism in place and give it teeth. Independent, resourced, with a route to decision-makers that does not run through the person whose plan is under review. Then act on something it says, early. That is the only way anyone will believe it is real.
Design for the thing Auckland got wrong. Auckland was capable on strategy and weak on connection. If you are building something larger, the mechanism for staying attached to the communities you serve is not a refinement. It is the design constraint. Invercargill's proposal to push genuine authority to community boards is at least an attempt at an answer. Whether it works depends entirely on whether those boards hold real decision rights or a consultation slot.
Write down what you expect, so it can be checked. What will be true in three years if this worked? Put a number on it. Put a date on it. A falsifiable expectation is the cheapest governance improvement available to any organisation, and almost nobody writes one, because it means someone might come back and look.
Where the attention is going
The structural question is absorbing nearly all of the available attention, and it is the tractable one. Boundaries can be drawn. Functions can be combined. All of that can be done competently, and the result can still be an organisation that makes poor decisions slowly, in a larger building.
The organisations that come out of this well will not be the ones with the cleverest structure. They will be the ones that used the disruption, and it is real disruption for a lot of people, many of whom did not choose it, to rebuild how judgement gets exercised. Who decides. On what basis. With what challenge. Checked how.
That is what governance is for. Not the paperwork. The judgement. The window for getting it right is open now, while everything is fluid and nobody has hardened into a new position. It closes faster than people expect. One question to take into the next meeting. What is one decision your organisation makes badly, and repeatedly, that a new structure will not fix on its own? Start there.
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Axis of Action works with organisations at the point where strategy meets delivery. Transformation advisory is the core of the practice: shaping programmes so they hold together from mandate to delivery, testing whether the intended change is actually being built, and helping leaders make the calls that keep it on course. Alongside this sits programme assurance, independent review, strategic alignment, governance services, and executive coaching for individuals and leaders. See the full range of services