Next Auckland harbour crossing needs a rethink
- Grant McLachlan
- 3 days ago
- 3 min read
Chris Bishop's new business case looks like more delay. It is actually the government admitting the data behind the last one was never good enough.
Transport Minister Chris Bishop's new Detailed Business Case for the Waitemata Harbour Crossing has been dismissed as delay dressed up as diligence. Labour leader Chris Hipkins called it an announcement about an announcement about a study about a study. He is not wrong about the theatre. He is wrong about the substance.
The business case Bishop has ordered is necessary because the process behind the last one was flawed. NZTA's board endorsed a tunnel option on 15 May 2026, straight off the back of its Investment Case. That recommendation came from a model, not a measurement. Nobody has established, with any precision, how much of the traffic entering the Auckland Harbour Bridge from Lake Road, the Northern Motorway, or the wider North Shore arterial network is genuinely crossing the harbour, and how much is short-hop local traffic a smarter network could intercept before it ever reaches the bridge. I set out the case for that missing data in Drowning in its own harbour in March.
The numbers Bishop cites in his announcement are not in dispute. The bridge carries 170,000 vehicles a day, underpins close to $1 billion in annual economic activity projected to reach $3.9 billion by 2051, and faces an estimated $4.8 billion disruption cost from renewal works if nothing changes. None of that needed another study to establish. What it does not tell us is what the second crossing should actually connect, because nobody has bothered to measure where the traffic comes from and where it is going.
The case for urgency is real. Weight restrictions already constrain freight and buses on the existing bridge, and the lack of dedicated public transport lanes stops Auckland making full use of the Northern Busway, which carries 40 percent of all people crossing at peak. Major renewals are coming regardless of what gets built next. But urgency is precisely why the crossing has to be sized to the traffic that actually needs to cross, not to a modelled approximation of it.
This matters because the price tag keeps climbing on assumptions rather than evidence. A Treasury report, reported by RNZ in November 2024 and cited again this month, put the capital cost of the first stage at between $22.9 billion and $27.2 billion. More than $36 million has already gone into the preliminary business case, three-quarters of it on consultants. NZ First leader Winston Peters puts the wider consultant and contractor spend on the project at more than $106 million, with still no plan for what gets built. Every dollar spent modelling a guess is a dollar not spent measuring a fact.
Bishop's decision to strip NZTA of ownership and hand the project to an independently appointed Senior Responsible Owner, answering directly to Cabinet rather than the agency that has run the process for a decade, is the right instinct. It concedes what critics have argued for years: leaving delivery inside an agency whose institutional interest lies in perpetuating the study cycle guarantees the cycle continues.
Cabinet's parallel decisions to fund a fresh look at Wayne Brown's preferred Meola Reef option, examined repeatedly in the past and previously dismissed as unachievable, and to weigh a market-led proposal to upgrade the existing bridge rather than build anew, both deserve genuine testing rather than reflexive dismissal. Auckland has form on discarding cheaper options for expensive ones that never get built. The Eastern Motorway died in 2004 partly because nobody could show what traffic it would actually carry.
Sydney answered this same question in the 1980s, privately financing and building its harbour tunnel in under five years. Auckland has instead spent fifty years commissioning studies while the cost estimate climbed from billions to tens of billions. An independently led business case, freed from NZTA's stake in the outcome, is a genuine chance to break that pattern, but only if it starts by measuring the harbour's traffic rather than modelling it again.
Bishop is right that this project cannot afford another wrong start. But getting it right starts with counting cars, not modelling them.
