For years, Auckland’s City Rail Link was treated as a kind of civic extravagance, a tunnel-shaped affront to fiscal sobriety. Every revision of its budget invited the same small chorus of disbelief. At about $5.5 billion, the CRL became evidence, for some, that public transport had grown too ambitious for its own good.

Then came the second Waitematā Harbour crossing.

Suddenly, sums of $10 billion, $15 billion, even $25 billion entered the conversation with remarkably little moral alarm. Depending on what is built, we may spend the equivalent of two, three, perhaps five City Rail Links.

The comparison is worth dwelling on.

A second harbour crossing is necessary. The existing bridge opened in 1959 and remains too important, too vulnerable and too singular a link for a city of Auckland’s size. Freight needs resilience. Emergency services need redundancy. Northland cannot remain tied to the upper North Island through one ageing piece of infrastructure.

But a second crossing is not the same thing as a second motorway.

Six new lanes over the harbour do not conjure six new lanes through the city. The cars still have to arrive somewhere. Auckland’s motorway junctions and central streets are already constrained. The bottleneck does not vanish. It merely crosses the water.

This is where the arithmetic of cities becomes awkward for the motor car. Cars are marvellous at giving individuals freedom, but poor at moving very large numbers of people through very small amounts of space. Dense cities eventually learn this whether they wish to or not.

The CRL is an example of the alternative. At full capacity, Auckland’s rail network will be able to move up to 54,000 people an hour at peak through two underground tracks. To shift comparable numbers by private car would require not another lane or two, but something approaching dozens of motorway lanes.

Seen from that angle, $5.5 billion looks less like extravagance and more like scale.

The Northern Busway has already taught Auckland the same lesson. Public transport now carries a substantial share of peak travel across the harbour, especially into the central city. But buses, too, have limits. The North Shore will eventually need a higher-capacity rapid-transit spine.

That is why, for more than a decade, Auckland’s own transport planning has treated a new Waitematā crossing and a rapid-transit connection to the North Shore as part of the same problem. Successive studies have consistently examined multi-modal options like rail, light rail, bus rapid transit, and road capacity together. Separating them produces poor outcomes. Importantly, those same studies have not justified removing rapid transit in order to reduce costs; rather, they have shown that excluding it simply shifts congestion and undermines the long-term value of the investment.

That is why rapid transit cannot be treated as an ornamental extra in a future harbour crossing, something to be added once the serious business of roads has been paid for.

Road capacity will still matter. Freight will still matter. Cars will not disappear, nor should they. But if Aotearoa New Zealand is willing to spend perhaps $15 billion or $20 billion on a new crossing, it becomes rather difficult to argue that rail or rapid transit is somehow the unaffordable part.

Major infrastructure is expensive. The real question is what we buy with it.

A second Waitematā crossing should give Auckland resilience for road and freight traffic, but it should also establish a permanent high-capacity rapid-transit connection to the North Shore. The exact technology can be argued over. What should not be optional is mass transit itself.

Otherwise, Auckland risks spending twenty-first-century billions on a twentieth-century idea: more lanes, more cars, more distant suburbs and, before long, another queue.

The case for another crossing is strong.

The case for building another 1959 is not.