Whether or not having a ‘limit’ around the edge of Auckland (call it a MUL or a RUB of whatever) drives up the cost of housing is a fairly long-running debate. The argument in favour of this assumption is encapsulated by an opinion piece in the NZ Herald a week or so back by Don Brash:
At the moment, less than 1 per cent of New Zealand’s area is urbanised. We are one of the least densely populated countries in the world. The council has quite deliberately chosen to make land expensive.
And the consequences of that decision are disastrous, socially and economically.
It’s disastrous socially because for most low and middle-income families, buying a house in Auckland is now not even remotely possible, and for those families who do make the attempt, it almost inevitably means both parents working outside the home. Most low and middle-income families can’t even make the attempt, and often live in over-crowded, poor quality rental accommodation.
At the moment, the median house price in Auckland is some seven times the median household income. It should be about three times, as it was 20 years ago and is now in many of the fast-growing cities in, for example, the United States.
Why is it possible to buy 500sq m sections on the outskirts of Houston for $40,000, whereas 400sq m sections on the outskirts of Auckland cost $400,000? The answer lies simply in the fact that in Houston there are relatively relaxed attitudes towards using land on the outskirts of the city, whereas in Auckland that has been prohibited.
We’ve recently discussed the question of whether or not sprawl is a good idea, and I think that’s largely a separate debate to this one. What’s particularly worth testing here is the question of whether opening up large amounts of greenfield land will improve housing affordability or not. The Auckland Housing Accord and its accompanying (although rather contradictory) legislation means that we may well be testing this question in the relatively near future – but recent evidence suggests that standalone housing built on the urban periphery is generally very expensive. Let’s take a look at what was noted in a Central Government report on housing and affordability recently:So standalone houses on the urban periphery generally aren’t that cheap to build, whereas most higher density housing types are comparatively affordable to construct on a per unit basis. And it seems that the availability of affordable housing units is very much linked to the number of higher density units being built:Now one of the reasons why recent standalone houses might be so expensive is because there’s (apparently) not a lot of greenfield land going around – which means that it’s fairly scarce and therefore high value. Expensive land means that a developer needs to put a big house on the site (or two or three smaller ones, which is generally banned through planning rules) in order to make a profit. So maybe opening up more greenfield land could lower the price of that land (more supply leading to lower prices at a set level of demand) and therefore result in it being “economically sensible” to build cheaper standalone dwellings on that greenfield land.
An interesting question then, however, is “how much extra greenfield land would be needed to make a difference?” Plus, perhaps, also a question around whether there really is a shortage of greenfield land being supplied. Let’s refer once again to the Central Government report, this time in relation to land supply:
The key ‘takeaway’ point from the above slide is that only a small proportion of land which is zoned for development and served with infrastructure has actually taken the next step to being developed. As “boosting housing supply” requires a lot of extra houses to actually end up being built, we find ourselves with two further questions:
- If we want to go from around 2,000 sections being “ready to go” (i.e. subdivided) up to around say 10,000 – does that mean we need five times as much land zoned and serviced as we have now? In other words, is the market always going to drip-feed the subdividing of zoned and serviced land into sections, no matters how much land is available?
- Do we instead need to focus on what’s holding up the 15,000 sections worth of land that is zoned and serviced from actually being split up into subdivided sections and eliminate those barriers?
Option 1 may well lead to achieving the desired amount of increasing housing supply, but at a truly staggering cost to ratepayers and taxpayers in the form of new roads, new schools, new water pipes, new wastewater pipes and so on. Furthermore, if it’s only around 13% of zoned and serviced land which is actually going to be split up for development to occur on it, then we’re going to be building an absolutely vast amount of roads, pipes, schools and all the rest which are totally surplus to requirements – at least in the short term.
Undoubtedly this post raises far more questions than it has answers, which is unsurprising. There just doesn’t appear to be much evidence that the best way to improve housing affordability is by building more houses on the edge of the city. Or that it’s a lack of zoned and serviced land which is holding up the release of new greenfield sections onto the market. Or that increasing the supply of land zoned for greenfield development will actually do anything – especially not in the short or medium term, to improve housing affordability.
It’s all a bunch of questions, a bunch of very expensive questions if the Auckland Housing Accord does decide to open up greenfield land in the vague hope of improving housing affordability.