I think of our construction sector as a flywheel. It takes enormous energy to get it spinning, but once there it can tick over easily. Give it a steady flow of work, with certainty, and there will be investment in people, capital and technology.
But when there are sudden stops and the pipeline is uncertain there are real costs.
Business closures and job losses often mean we lose people to other sectors – often to Australia where wages are nearly 50% higher. We also lose their know-how and connections. The latter may seem frivolous, but in an industry that works with a large network of specialist suppliers, those networks are a key ingredient of efficiency and productivity.
Designs and consents expire, so restarting means paying twice for the same work. And construction costs have risen 3.5% a year over the past quarter century, against general inflation of 2.7%, so the same project costs more every year we wait.
Deferring a project can make the books look better today, or remove a political thorn. But, these have real costs on the sector, it has real consequences for communities who need that infrastructure and, from my analysis, we end up paying more, often for less, later.
Shamubeel Eaqub
There are also costs I haven’t counted here: the sunk costs on projects that were never fully started, because they aren’t easily visible in the economic data.
But a couple of case studies make the point well and give shape to my frustration at silly political bickering.
The iReX ferry replacement was cancelled in December 2023, two and a half years after the ships were contracted. The write-off was $643 million. Of the $671m spent, only $85m produced anything physical. The replacement ferries will now arrive in 2029, three years later than the ones we cancelled.
Auckland Light Rail spent $228m over seven years and three institutional resets without producing a completed business case or laying any track.
No political party has clean hands. The damage to the sector runs deep. About 95% of firms in civil construction have fewer than 10 staff, with thin buffers and no forward book to fall back on. They often hold large amounts of specialist equipment, and fixed costs sink them if the work dries up. After the GFC the sector lost 2500 jobs and 900 businesses closed. The current downturn has already cost 1800 jobs.
Rebuilding those teams takes up to five years, and we will pay for that lost capacity and capability for years to come.
We know what to do.
Five ideas that I think would help
· Ring-fence maintenance and renewals. Deferred maintenance costs two to five times more to fix later. Cutting renewals in a fiscal squeeze is a false economy.
· Fund the pipeline, don’t just publish it. The UK has committed £725b over 10 years. Australia gives its industry five to seven years of funded visibility. Our unfunded lists give firms nothing to invest against.
· Publish the cost of stopping. Any minister or council proposing to pause, review or cancel a project should first publish the expected exit fees, remobilisation costs and lost benefits. Then decide, in the open.
· Build in the downturns. When private work retreats, capacity is available and prices moderate. Cramming projects into boom windows inflates procurement costs by up to 30%.
· Bring in long-term capital. KiwiSaver holds $142b currently and will be around $1t by 2050. At the average infrastructure allocation of Australian pension funds, that could be around $50b of patient money looking for exactly these assets.
None of this is technically hard. The barrier is political: the temptation to treat every inherited project as the previous government’s mistake. We have spent 25 years and nearly $12b repeating the same mistakes. The cheapest infrastructure policy is to finish what we start and lift our sights beyond the political term.