When Judith Collins and Julie-Anne Genter argue about expensive Auckland transport projects, I worry that they’re both right about the problems in the other’s preferred alternative.

And it points to the need for better ways of deciding on transport investments.

Last week, Finance Minister Grant Robertson was reported to be considering scrapping the proposed walking and cycling bridge over Auckland harbour.

Collins has previously been sharply critical about the proposed bridge, pointing to initial estimates that the bridge would deliver $40 to $60 in benefits for every $100 spent building the thing. She was very likely correct.

But Minister Robertson was also reported as considering bringing forward a potential Auckland harbour tunnel project instead, which might accommodate vehicles as well as bicycles.

Genter, in debating the relative merits of bicycle bridges and harbour tunnels with Collins, had pointed to even worse benefit-to-cost estimates for a Waitematā tunnel. The NZ Transport Agency had costed the tunnel at some $10 billion, with estimated benefits of $2 billion. Genter was very likely correct as well.

Whipsawing from a bicycle bridge projected to waste about half of its $785 million cost, to a tunnel projected to lose $8 billion on a $10 billion spend, seems a poor way of setting transport priorities.

Part of the problem seems to be that nobody entirely trusts the cost-benefit calculations underpinning transport analysis.

The most compelling benefit of any urban transport project, currently, is likely to be the new housing it might enable: better transport connections make it easier to build denser housing in places people want to live.

But these wider economic benefits are contentious. In 2013, NZIER recommended incorporating the benefits of land use change as part of transport cost-benefit assessment. By late 2019, the NZ Transport Agency reported that such benefits remained generally outside of the scope of transport cost-benefit assessment as they required fairly sophisticated analysis.

For a bicycle bridge, how many people would really shift to cycling across the harbour to get to work? How much would they value that option? If the bridge were successful, could it enable greater housing density on Auckland’s North Shore? How should we think about that in a desperate housing shortage? Could Council really upzone Devonport and the rest of the North Shore if a bicycle bridge made it an attractive place for urban bicycle commuters to live, or would local opposition or other infrastructure constraints prevent it?

Cycling fans tend to be optimistic that people would shift to cycling if only the infrastructure were in place to support it. Perhaps a cycle bridge would be just the lynchpin necessary to enable tens of thousands more people to live on the North Shore and enjoy an easy commute to town. But it is hard to tell.

For a harbour tunnel, how many more trips would be enabled? What parts of the city might be able to accommodate more housing because of the new transport route? How should we value that? And would the second crossing even be necessary if Auckland implemented congestion charging?

Transport planners can make their best guesses about the answers to all of these kinds of questions, but when transport decisions are made by politicians prone to wishful thinking about the merits of their own preferred form of travel, it is hard to build a coherent transport system.

So how do we get spending on the infrastructure we need while avoiding wasting money on white elephants, when the main political parties seem to really just like spending money their own preferred modes of transport?

The best solution may be to take these decisions out of the hands of politicians.

Imagine if the proposed bicycle bridge, or the proposed tunnel, were forced to go to external investors to find financing, rather than beg politicians for it. The Auckland Harbour Bridge Board issued bonds to investors in the 1950s; those bonds funded the bridge. Tolls paid by drivers paid off the bonds. It is an important and forgotten part of New Zealand’s transport heritage.

If infrastructure were funded this way, business cases would need to convince investors with real money on the line. Would either the bridge or the tunnel really see enough paying customers to make the venture worthwhile? That would depend not only on the attractiveness of the transport option but also on the new housing opportunities it might unlock. Or maybe a multimodal bridge would prove an even better prospect.

Hopefully, some option for a new harbour crossing would prove to be a viable commercial proposition.

But if none did, we would at least avoid wasting considerable amounts of money on projects delivering far less than their cost.