The name on the letterhead

Invercargill's Head Start submission lists the principles it wants any Southland proposal judged against. A united voice. Fair representation. A strong local identity.

That last one is not a technical criterion. Nobody can model it, cost it, or put it in a business case. It sits alongside deliverability and value for money because the council knows, correctly, that without it the other two will not save the proposal.

KPMG's assessment of the Head Start pathway makes the same point from the outside. Loss of identity is one of the biggest concerns for councils and their communities, and international experience shows that where it is not managed, the result is reduced civic engagement and, in some cases, outright reversal of the merger years later.

This is the part of the reform that structure cannot fix and systems work cannot fix either. Parts one and two dealt with judgement and connection. This one deals with belonging, and it runs on a different clock. People do not wait for the detailed design phase to decide how they feel about becoming part of something larger. They decide in the first few weeks, on far less information than any of the frameworks in this series assume they will use, and the decision is difficult to reverse once made.

The verdict arrives before the evidence

Social identity research gives a plain account of why. Belonging to a group is part of how people understand who they are, not an incidental fact about them. When a merger recategorises that group into something larger, it does not just change an org chart. It puts identity itself under threat, and threat produces fast judgement, not careful evaluation.

The research on intergroup mergers is consistent on what follows. Members of the smaller or lower-status partner in a merger tend to feel less represented in the combined entity and experience a weaker sense of continuity with who they were before. That perception forms early, largely on symbolic cues rather than substantive ones, and it strongly predicts whether people go on to identify with the new organisation at all.

Which is a way of saying that the business case was never going to be what decided this. People were always going to ask a prior question first. Do I still recognise myself in this. Everything in the proposal gets filtered through the answer.

What the record shows when this goes wrong

New Zealand does not need to look far for the risk. The Government's forced amalgamations in the late 1980s carried a political cost that outlasted the reform itself, and the 2023 pushback against Three Waters consolidation ran on the same fuel: a sense that something local was being taken away, expressed long before most people had engaged with the technical case for or against it.

Overseas, the pattern is stronger still. New South Wales forced 44 council mergers through in 2016. The backlash was immediate and did not fade. By 2021, residents of the amalgamated Inner West Council voted in a non-binding poll to de-merge back into its three original councils, five years after the change and well after any transition costs had been absorbed. Quebec ran the opposite experiment in reverse: after forced municipal mergers around Montreal, a change of government allowed referendums, and 28 municipalities voted to de-merge, some citing better fiscal control but many citing simple attachment to the place they had voted to leave. In Norway's regional amalgamations, more than eighty percent of Finnmark residents opposed merging with neighbouring Troms county, largely over fears of losing a distinct northern and Sámi identity, and polling years afterward still showed a majority in favour of splitting back apart.'

None of these reversals happened because the merged organisations performed badly on the metrics in part one or the systems in part two. Some performed adequately by both. They happened because enough people never accepted that the new entity was theirs, and that judgement, once settled, proved close to impossible to unsettle with better service delivery.

There is a smaller, sharper version of the same finding from Japan. Researchers studying municipal amalgamations found residents willing to accept a real, quantifiable loss in local services and spending efficiency in exchange for keeping their original municipality's name. Identity was not a preference people weighed against efficiency. For a meaningful number of them, it was the thing efficiency had to be weighed against.

Four things meaning does during a merger

The verdict forms before the design does. By the time a Head Start proposal reaches detailed design, most residents in the affected area will already have an emotional position on it. Not a considered one. A fast one, formed on a handful of signals: whether their community was consulted or informed, whether their area's name survived the proposal, whether the mayor or leader they know is still visibly in the room. Detailed design cannot undo a verdict that formed before it existed. It can only confirm or complicate one already held.

Silence gets filled with the worst available story. Sensemaking research is clear that people cannot tolerate an information vacuum. When leaders do not supply a coherent account of what is changing and why, one gets supplied anyway, usually by whoever is most anxious or most opposed, and it tends to default to loss rather than gain because loss is easier to make vivid.

Wellington's Let's Talk process, open from 24 June to 19 July, is at least an attempt to get ahead of this by naming the constraint honestly: that this is not yet formal consultation, that a great deal will be decided during detailed design, and that a low response rate will read as consent. Whether that framing lands as reassuring candour or confirmation of a decision already made will depend entirely on what residents were primed to expect going in.

Symbols carry weight that is easy to underrate and expensive to dismiss. A name, a flag, a logo, the address on the letterhead. To a spreadsheet these are decoration. To identity they are load-bearing, in the same sense a data standard is load-bearing to a consenting system. Change them carelessly, or let them default to whichever partner had the bigger population, and you have delivered a clear signal about whose merger this was, regardless of what the representation model actually says.

Once the frame sets, facts stop moving people and start confirming them. This is not stubbornness. It is how identity-protective reasoning works. Once someone has decided a merger is a loss, information that would once have been persuasive gets read as spin, and information that confirms the loss gets accepted uncritically. This is precisely why the early weeks matter disproportionately. The frame that forms in the vacuum of May and June is far cheaper to shape than the one that has to be dislodged in 2027.

What good looks like

Name the loss before someone else does. Every merger costs something real: a name, a level of local control, a council you could ring directly. Say what it is, plainly, before asking anyone to accept what they gain. A proposal that only lists benefits reads as either naive or dishonest, and either reading costs trust.

Give continuity signals early, and make them real. A retained name for a community board, a local office that still exists, a face people recognise still visibly accountable. These cost little and do more for acceptance than most of the substantive design work, because they answer the identity question before the structural one has even been asked.

Build local decision rights, not local consultation slots. The KPMG assessment is direct about this: without genuine devolved decision-making, larger councils feel distant no matter how well they perform. Invercargill's proposal to push real authority to community boards is a test of whether this can be done deliberately rather than promised and then quietly walked back once the new organisation exists.

Communicate meaning before you communicate structure. Most transition communication starts with the org chart and works outward. It should start with what does not change, who residents can still call, what their community will still be recognised as, and only then move to governance. The order matters because the order is the order people process it in.

Measure identification, not just satisfaction. Service satisfaction surveys will not tell you whether people see themselves in the new organisation. That requires asking directly, repeatedly, and paying attention when the answer is no even while satisfaction scores hold up. Auckland's super city scored well on strategic leadership and poorly on democratic engagement, and the two numbers coexisted for over a decade before anyone treated the second one as the more urgent finding.

The through-line

Part one argued that structure does not fix judgement. Part two argued that structure does not fix connection. This part argues that structure does not fix belonging, and of the three, belonging is the one that forms fastest and lasts longest. A governance model can be redesigned. A system can be re-integrated, however slowly. A verdict about whether this still feels like home is much harder to move once it has set, and by the numbers above, it sets early.

The councils writing Head Start proposals this month are, understandably, spending most of their energy on the parts that can be modelled: boundaries, delegations, data standards, cost lines. Those are the tractable problems, and they matter. But the thing that will determine whether any of it holds together in 2030 is happening in parallel, in a register none of the assessment criteria measure, in the few weeks when residents decide, largely without waiting to be asked, whether this is still theirs.

One question worth taking into the next communications meeting, not the next governance one. If someone in your community had to explain, in one sentence, what they lost and what they kept, would they be able to answer both halves, and would either answer surprise you?

That is the actual test. Everything else is paperwork.

---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

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