Let’s say you’ve bought non-refundable tickets for a movie, and then hear scathing reviews. It’s better to tear up the tickets and make other plans for the evening rather than bloody-mindedly watch the film just because you’ve already paid to go.

That’s according to one of the more obvious-when-you-hear-it economics propositions. If you have invested resources in something and cannot recover the investment if you change your mind, you should not let that earlier investment influence your decisions as the costs of that investment are sunk.

This week, Labour finally sunk its considerable political costs and tore up its KiwiBuild plans – although after it had kept doubling down for a long time on its fundamentally defective policy.

If KiwiBuild could have been effective, KiwiBuild would not have been necessary.

If the zoning and infrastructure were in place for the Government to build houses, there would have been no housing crisis – private developers would already have been building.

But if KiwiBuild was necessary because there weren’t enough houses, then it would be an impossible policy: what prevented private builders from building would also prevent the Government from building.

So if we needed the Government to build houses, KiwiBuild wouldn’t work because of those other constraints, and if it could work, it wouldn’t have been needed.

None of this should be surprising. The policy was never thought through, the targeted numbers were picked because they sounded appropriately impressive for an election campaign rather than feasible, and even Treasury’s warnings of the project’s limitations proved too optimistic.

KiwiBuild never addressed the real root of the housing crisis. Although I was always a KiwiBuild sceptic, even I thought that given the magnitude of the housing shortage, a government building programme would have a hard time doing real harm. Maybe the houses built under its auspices wouldn’t quite be of the style or size or configuration buyers might ideally want, but how badly could you mess up building houses in the middle of a housing crisis?

Well, in their increasingly desperate quest to hit nonsensical targets, the Government started building houses in places where it was easy to build rather than where houses were neded.


The failure was even more spectacular given the very, very easy route the Government had to get out of the mess.

Phil Twyford, better than just about anyone in the country, understands the underlying issues causing the housing crisis. He has been pushing a supply agenda that would strike at the root of the crisis and prevent it from happening again. That agenda has made far less progress than it should have because KiwiBuild has sucked too many officials into a black hole of futility.

But the Government could have easily changed tack upon hearing the advice of Treasury on KiwiBuild. Imagine if instead of insulting Treasury analysts, Minister Twyford in May of 2018 had announced the following:

“This Government was elected on a promise to end the housing crisis. We thought that a massive government-led building effort was the best way to achieve that. On forming government, we consulted broadly and sought the views of officials. They’ve told us that the most important thing we can do to end the housing crisis is quickly advance a broad agenda of local government reform, infrastructure financing reform, changes in urban planning, changes in local council incentives, and transport reform. We will advance this agenda. The supply agenda we will soon announce will enable the building, over a decade, of tens of thousands more homes, townhouses, and apartments than ever could otherwise have been built. Most of those homes will be delivered by the private sector, thanks to the reforms our Government is putting in place letting our builders build. Some houses, in some places, may be built by the Government as well. Every one of those homes, no matter who builds it, will be a KiwiBuild house. For too long, the National Party denied that there even was a housing crisis. Our supply agenda will end the crisis and make sure it never returns.”

But enough of I-told-you-sos. The point of sunk costs is to ignore them and look forward. KiwiBuild is behind us – a sunk cost. And Labour should be commended for ending it rather than damned for taking so very long to reach an obvious conclusion.

The path ahead remains the broader supply agenda, parts of which have been quietly working away in the background by those officials not sucked into the black hole of KiwiBuild, and parts of which have yet to hit the Government’s radar.

Let’s walk it through.

Fundamentally, the housing crisis emerged because not enough houses were being built. Not enough houses were being built because council zoning rules prevented sufficient building. This had systematic effects across the whole building industry. Because building vast new subdivisions, or substantial new dense and intensive brownfield developments, was effectively impossible, the construction sector geared up for the task it was allowed to undertake: bespoke small-scale construction and renovation.

Why do many councils facing growth pressures set rules that make it tough to build? Because growth is costly for councils. Councillors face persistent NIMBY opposition to new development. Density advocates work to block surburban expansion while downtown residents opposed to development next door blocked density. The outcome in too many cities was that no one was allowed to build anything anywhere.

… growth has to be in councils’ interest. If it is not, everything else is futile.

But it is worse than that. When an urban council at its debt limit has to accommodate growth, the costs simply outweigh the benefits. Central government gets more income tax revenue when a city’s population expands; local government gets an infrastructure bill. Central government gets more GST and company tax revenue when it allows more commercial and industrial development; local government gets the complaints about tall buildings blocking views and faces the bill for trunk infrastructure upgrades downtown. More rates revenue simply is not enough for councils to make growth viable for them.

It gets even worse. Construction costs are higher than they need to be because of a nest of regulations, liability rules, and incentives. There are, in theory, ways of importing a containerload of gib board from places whose standards we trust and whose challenges are comparable – like Vancouver, Seattle and Tokyo – places that face earthquake risk and wet weather issues.

But council has to sign off on the final building certificate. If council signs off on a building certificate and something goes wrong a decade from now, when every one of the construction companies has flipped from Bill’s Construction (2019) Ltd to Bill’s Construction (2025) Ltd, council is the one left holding the bag under joint-and-several liability. So councils try to limit risk by being highly risk averse about allowable building methods. Standard ways of fixing walls to frames are defined in terms of standard materials. And good luck getting your new build signed off if you haven’t used those standard methods.

Once you recognise the incentives at play, you start recognising the difficulty of the problem – and start seeing the ways of unwinding the mess.

Fundamentally, growth has to be in councils’ interest. If it is not, everything else is futile.

You can set a National Policy Statement on urban growth that prohibits some of the deplorable things councils do that increase housing costs, like minimum parking restrictions. But changing the rules without changing the incentives is pouring old planning wine into the new NPS bottles. There are always ways of restricting new development as long as councils see growth as a cost rather than as a benefit.

While Minister Twyford has spoken eloquently on the harms caused by metropolitan urban limits that constrain growth at the fringes, the National Policy Statement on sensitive soils practically invites councils wishing to constrain growth to declare land at the fringes to be sensitive. Councils are not stupid. We get the outcomes we get because of the incentives councils face – even if the outcomes are often mindbogglingly stupid.

There are minor tweaks that could improve councils’ incentives and outcomes. Removing councils from joint-and-several liability, and making them only liable for their proportionate share of any problems down the track, could make councils less risk-averse. That would allow more innovation in building techniques to reduce costs.

Alternatively, the Government could use its upcoming Urban Development Authorities as consenting authorities able to sign off on final building inspection certificates – effectively, a competitive issuer of building consents.

But, more fundamentally, our financing infrastructure frameworks have to change so the beneficiaries of the infrastructure cover the infrastructure’s cost, over time – rather than dumping that cost on council’s general balance sheet.

There are lots of ways of achieving that kind of outcome.

Texas uses municipal utility districts, where a developer sets a utility district that owns the infrastructure, finances it through a bond issue, and pays off the bond through a ratings levy on the properties in the development. Ownership and governance of the district transfers from the developer to the homeowners as properties are sold. That mechanism has enabled new communities to develop wherever they meet a need – and house prices consequently never get out of step with underlying construction costs and the underlying costs of bare land.

Phoenix does it through a special ratings area and revenue bonds. Council issues the debt for infrastructure for a new development, but the debt is ring-fenced. The bond is paid off by a special ratings levy on the properties benefitted by the bond. If the new development fails, there is no recourse to council. That means every bond can have its own credit rating, and none of them are seen on council’s core balance sheet.

Twyford is progressing legislation that will have a similar effect. It would allow for a special debt vehicle for infrastructure for new subdivisions, with the debt paid off by a levy on the properties benefitting from the infrastructure, and no recourse to council if the development should fail. This mechanism is inferior to American financing vehicles in a few ways, but it is still better than what we have now – and removes one cost that councils face in accommodating growth.

The Government needs to progress, and quickly, its legislative work on infrastructure financing vehicles.

Unfortunately, there seems little appetite in central government for aiming at the root of the problem. Fundamentally changing councils’ incentives would see direct payments to central government to councils that enable growth. That could be done by providing grants to councils commensurate with the GST take from new housing development in their area, or by sharing the increase in tax revenue that central government enjoys when local governments facilitate growth.

That path requires central government being willing to trust local government a bit more, but we are in a vicious circle where central government trusts local government with nothing because it has no expectation of local government competence, and local government then too often attracts the calibre of officials you might expect when they are not trusted to do anything.

From where we are, there is a good path forward.

The Government needs to progress, and quickly, its legislative work on infrastructure financing vehicles.

It needs to consider whether to extract councils from joint-and-several liability, or enable its urban development authorities to be competitive issuers of consents, or both.

And it needs to reconsider its national policy statements in light of the incentives councils face and either provide more positive encouragement for councils to enable growth, or assign more explicit performance targets to guide behaviour. It is not hard to imagine, for example, development-friendly provisions in a National Policy Statement on Urban Development that would come into effect for any council where the median house price is more than five times the median household income, and would be in abeyance otherwise.

After victory in the Battle of El Alamein, Winston Churchill said it was not the beginning of the end, but that it was, perhaps, the end of the beginning. The victory of Labour over itself in abandoning the sunk cost of KiwiBuild, we hope, marks the end of the beginning of the war on the housing crisis.

We look forward with hope to the success of Twyford’s coming campaign. There remains a lot of blood, tears, toil and sweat in the path ahead – but there is a promising gleam.

Leave a comment