KiwiBuild – the government programme to deliver 100,000 down-sized homes in 10 years – is a waste of time and money.

It serves no useful material public purpose and absorbs time and resources that would be far better devoted to addressing the real problem – housing affordability.

That is the thrust of the New Zealand Initiative’s research report KiwiBuild: Twyford’s Tar Baby, which was published this week.

KiwiBuild has become a policy in search of a policy justification.

Labour first proposed it in 2012. In so doing, it baldly asserted a “lack of entry-level new houses” was “one of the main reasons” housing is unaffordable.

How that assertion, even if true, could possibly explain sky-rocketing prices for all houses in Auckland, new and old, big and small, was unasked and unanswered.

Given this misdiagnosis, the remedy was, of course, a no-brainer – in the sense you would only come to this conclusion if you did not think at all. The conclusion being the state should step in to build 100,000 shoe-box houses for aspiring first-home buyers.

Why they would want a new house any more than a first-car buyer would want a new car was also left unasked and unanswered.

For those who might think these justifications are a little thin, the policy offered a supporting justification: Increase the proportion of owner-occupied homes.

Here the naive idea seems to be that a first-home buyer will be moving from rented accommodation that will stay rented and that is all there is to it. Of course, that is not all there is to it.

Economists Andrew Coleman and Grant Scobie have demonstrated that the proportion of owner-occupied houses is largely insensitive to the rate of government housebuilding.

The proportion is likely to be much more influenced by the benefits of renting relative to owner-occupying for existing dwellings.

Not adding materially

A third ‘justification’ for KiwiBuild was the unsupported assertion that it would increase the housing stock by 50,0000 homes compared to what would have occurred otherwise. If that were so, it could certainly ease the affordability problem for a time but is it so?

It is almost certainly not so in the short run. The construction industry was already up against capacity constraints. This forced the government to change the objective to delivering 100,000 homes rather than building them.

Inducing developers to build a KiwiBuild home on land they were going to build on anyway is not adding materially to the housing stock.

The Treasury gave deep consideration to the additionality issue in forecasting residential construction activity through to 2021 after the adoption of KiwiBuild. The Initiative’s report compares the Treasury’s forecasts to those in its pre-election economic and fiscal update. There appears to be no material difference in trend activity.

In the long run, only something that permanently lowers the cost of housing of a given quality at the margin can increase the housing stock above what would occur otherwise given income and population growth. That requires some combination of lower land prices and lower construction costs across the board.

In itself, KiwiBuild lowers neither the cost of land nor industry construction costs. (It could even raise both along the way because of capacity constraints.)

Messrs Coleman and Scobie showed that if industry supply costs are unchanged, public investment does not change the housing stock in the long run.

Why then make a fuss? After all, the government has said it will not subsidise KiwiBuild buyers.

If the programme is not subsidised and does not make a material difference to anything that matters, who cares?

No subsidy, no need

The problem is that if there really were no subsidy, there would be no need for KiwiBuild. Developers exist to satisfy unsubsidised demand. They and buyers could profitably cut out the middleman – the government.

So, to entice developers and would-be buyers to participate, KiwiBuild must promise them a financial inducement implicit or explicit. That incentive is a subsidy, by any other name.

The government’s decisions to allocate by ballot rather than by auction, to restrict the range of buyers, and to impose special resale restrictions would not be necessary if there were no subsidy.

However, as the experience of the first KiwiBuild sale in Papakura demonstrated, to subsidise the people who can afford a KiwiBuild home is deeply inequitable.

Yet the dilemma for the government is that the smaller the subsidy, the greater the risk it will be repeatedly embarrassed by unsold KiwiBuild homes, as has already occurred in Wanaka. After all, politicians lack the commercial incentive to design and build homes in locations that best meet buyers’ preferences.

This government’s commitment to KiwiBuild means it is stuck to this tar baby. The more it does to detach itself from the subsidy problem, the greater the risk of failing to sell.

Another argument for KiwiBuild is that associated measures to increase the supply of land for housing and ease the path to lowering unit costs through fabrication and facilitate intensification and improving infrastructure supply will make an enduring difference to housing affordability.

Indeed, they could. But the rub is that all this can and should be done without KiwiBuild.