NZ has no idea how many houses it needs

There will be 6.8 million people living in New Zealand in 2072, or so Treasury says in its latest set of forecasts released a fortnight ago. With all due respect to Treasury’s forecasters, your humble columnist can say with some degree of certainty that this number will be wrong. Not just wrong in the way that most forecasts going out 50 years will be wrong, but quite drastically wrong.

Treasury is notoriously bad at forecasting population growth, having drastically undershot in its forecasts for more than a decade. According to Stats NZ, our population hit 5.1 million in December – a bit slower than might have been expected as a result of the border being closed for most of the year.

If you’d asked Treasury a decade ago when the population would reach that milestone, it would have said some time in 2030, for that’s what its 2011 forecasts said – it was a decade late.

There’s a bipartisan but private frustration at Treasury’s population forecasts, which seem to anticipate the country snapping back towards the bucolic backwater that it was in the 1970s, rather than the rapidly urbanising globalised New Zealand it is now – and which seems to be our future.

Stats NZ does a better job, forecasting, with 90 per cent certainty, that the population will be somewhere between 5.27 million and 8.48m in 2073.

To be fair to Treasury, at least it puts in the effort to forecast at all. Pity the poor Ministry of Housing and Urban Development. When Nicola Willis used Thursday’s Question Time to ask about the gap between the supply of houses and demand (essentially, how many houses we must build to solve the housing crisis), Housing Minister Megan Woods said that current advice put the gap at anywhere between 28,000 and 200,000 homes.

Now, 28,000 and 200,000 are very different numbers. They’re very, very different numbers when you’re talking about dwellings – the difference between them is the difference between the township of Timaru (about 20,000 dwellings) and the entire region of Wellington (about 200,000).

The gap between them says almost nothing about the housing crisis, but quite a lot about our inability to forecast and model how big it might be.

Woods said that the Government – to its credit – was working up ‘‘place-based’’ assessments; that is, what we need to build in any given area.

‘‘We can understand not only the housing shortfall in a given geographical location but we can also understand what the shortfall is in terms of rental properties, what the shortfall is in terms of home ownership, what the shortfall might be in terms of public housing,’’ Woods said of the assessments.

That’s a promising start and recognises that the housing crisis has metastasised from an urban to a rural problem for owner-occupiers and renters.

But what we need is to realistically forecast population growth and where those people might be heading. Currently, the housing strategy is a bureaucratic translation of that saccharine inspirational dictum more commonly found in someone’s loo than in a policy document: ‘‘Shoot for the Moon. Even if you miss, you’ll land among the stars.’’

There’s at least some chatter in Wellington about using a housing model to better estimate housing demand, and ministers would do well to listen to it. The second of the three laws that will replace the RMA, the Spatial Planning Act, will force the Government to come up with 30-year plans for the infrastructure it plans to build and the land it plans to develop.

Perhaps the biggest single input into these plans will be how fast the population will grow. The obvious advantage of 30-year plans – planning and building infrastructure as it’s needed rather than after – will be completely undone if the Government isn’t realistic about the country it’s planning for. Noone expects the Government to be bang on with its forecasting, but current population estimates are barely credible.

There are strong political crosswinds. Policies to lower immigration and thereby slow population growth have been popular at recent elections. Being upfront about the fact that New Zealand is in the midst of a rapid growth phase risks making a government that orders improvements to these forecasts somewhat unpopular.

As for a housing model, modelling housing need essentially creates a rod for a government’s own back. That, of course, is precisely why we need such a model. But the current Government, still bruised from its spectacular failure to meet the KiwiBuild targets, may think there are quite enough rods for its back already.

Any model of housing demand shouldn’t absolve this Government or any other from measures to dial back some superfluous demand. Reserve Bank figures show that in the month to December, 45 per cent of the $9.5 billion of new residential housing lending went to investors, while just $1.7b went to first-home buyers. There’s no point in freeing up extra land for development only for a handful of investors to snap up the homes on offer.

Modelling housing need essentially creates a rod for a government’s own back. That, of course, is precisely why we need such a model.