sign uplog in
Want to go ad-free? Find out how, here.

If the best measure for housing demand is land value, what people will pay to live in particular areas, Auckland Council's economists find there is a mismatch in some areas between land that is high value, and current residential development potential

If the best measure for housing demand is land value, what people will pay to live in particular areas, Auckland Council's economists find there is a mismatch in some areas between land that is high value, and current residential development potential
Low density housing close to city centre.

By David Norman and Shane Martin*

• Auckland will be required by central government’s new National Policy Statement on Urban Development (NPS-UD) to allow more development than the current Unitary Plan allows.

• This upzoning will be focused around the city and metropolitan centres, rapid transit networks, and in all other locations commensurate with the accessibility and relative demand for housing and business there.

• The best measure of demand is land values – what people pay to live in particular areas.

• In Auckland, there is a mismatch in some areas between land that is high value, and current residential development potential. This is true across distances from the city centre and current zonings.

• Land values show that more upzoning is demanded closer to the city, where the highest paying jobs, best transport links and access to other amenities exists.


How did we get here?

Large swathes of Auckland were massively upzoned (which means the rules changed to allow much more efficient use of that land) in 2016. In aggregate, this landmark study showed there is sufficient development activity available inside the region’s Rural Urban Boundary such that land prices are not being skewed up by that boundary.

Not all areas inside the Rural Urban Boundary were upzoned equally. Areas a long way from the city centre or major metropolitan centres were generally not upzoned much at all, which makes sense given the relatively low demand for property that far from jobs, transport links and other amenities. A ring much closer to the city centre, close to jobs, with the best public transport links in the city, and close to other amenities, also did not receive much upzoning.

The NPS-UD requires local governments to further upzone in and around city and metropolitan centres, and around rapid transit nodes (which typically means train stations and the northern busway in Auckland). It also requires upzoning in all other locations commensurate with the greater of accessibility and relative demand for housing and business use there.

How do we best measure accessibility? How do we measure relative demand? The Chief Economist Unit was asked to assist in analysing how the latter question pertains to all these other locations outside city centre, metropolitan centre and rapid transit catchments. Our analysis is the subject of this paper.

Measuring what people want

Some have suggested using number of sales or number of new dwellings consented in an area as measures of relative demand. But these indicators are already skewed by the very factors we want to strip out of our analysis: lots of homes get consented and sold where we have already zoned for more housing. The bigger question we really want to answer is if the right zoning is in the places people most want to be.

The NPS-UD recommends using land values as an indicator of demand. We agree. Land values are the best indicator of where, without budget constraints, people would prefer to be. That does not mean that no one wants to live in areas with lower land values. People often have links to neighbourhoods that may lack the location or amenity that make some areas more expensive. But the best indication of what area people value most on average and in aggregate is land prices there.

Two things make land valuable. One is its proximity to amenities that people value (location). The second is what you can do with the land. We want to determine the amenity that people derive on average from different locations, as if zoning was not a factor. That means a fair comparison of land values must strip out zoning and consider only the other attributes that drive land value.

Creating the triangle we need

Our work over the last four years has shown repeatedly that proximity to Auckland’s city centre is one of the strongest and most certain pointers to value. The city centre is a great proxy for where the high paying jobs are, where public transport is best, and where a huge range of goods and services are available in a small walkable area, close to the water.

It will not shock our readers then that land values outside of the immediate city centre are highest in areas right next to the city centre. Cities around the world broadly follow a pattern where land that is valuable has the highest buildings, and at lower values you get lower buildings (the dark blue triangle in the stylised graphic). But this is not the pattern in Auckland, shown in red.

The dark blue triangle is simply not possible in Auckland under current zoning.

Where your wallet is, there lies your heart

So what does the empirical data say about relative demand for land across the region? Zeroing in on the city centre, we see that all four major residential zones, from THAB (Terraced Housing and Apartment Buildings) through to the least dense zoning (Single House) are massively more valuable closest to the city. At 10km from the city centre, land values are typically 40%-50% lower than at 2km from the city centre across the four residential zones.

This demonstrates what we already know anecdotally – relative demand is far higher close to the city.

Looking across zones, the patterns are similarly obvious. On average, across all distances from the city centre, the highest value land is land that allows more efficient use (THAB). The lowest value land is typically that with the least efficient use (Single House) even closest to the city, where that zoning is quite dominant.

Given this pattern, are we currently delivering most development potential where it is most valued and where relative demand is highest? Short answer: No.

There are two sets of maps at the end of this paper. The first set shows in greater detail where the most valuable THAB land is, and how much of each area’s residential zoned land is zoned THAB (from 0% to greater than 75%). The most valuable THAB land, worth over $4,000/m2, is close to the city, but several areas close to the city have little or no zoning for THAB at all. In contrast, the second set of maps shows lots of single house zoned land closest to the city, where once again, land values are highest.

Being careful what we wish for

But the stroke of a pen increasing development rights closer to the city is not without its challenges and it would be naïve to ignore them. Key among these challenges are what infrastructure will be required to support the increased density clearly demanded, and who pays for that infrastructure. Property owners who receive upzoning, whether they choose to redevelop or not, will receive a huge windfall gain in property value from upzoning, based on the expectation that infrastructure will be provided to service a more efficient use of their land. It is good economics that those who receive the windfall value gain should contribute a commensurate share to the cost of that infrastructure.

An appropriate funding mechanism will need to be in place in advance of zoning changes. If not, a changed zone may just be an idea on paper, as no meaningful development could ensue.


* David Norman was chief economist at the Auckland Council when this article was written. Shane Martin is a senior economits at Auckland Councit. This article was first published here. It is here with permission.

We welcome your comments below. If you are not already registered, please register to comment.

Remember we welcome robust, respectful and insightful debate. We don't welcome abusive or defamatory comments and will de-register those repeatedly making such comments. Our current comment policy is here.

59 Comments

Where are the maps?

Click on the link: first published here - take you to the .pdf file.

Omitted maps are now in the article.

Great analysis - the maps referred to can be seen by clicking the link to the originally published report.

Yes, great analysis on how to get to a 10x median income multiple, up from 3x in 1992.

What were interest rates in 1992?

I assume it is a rhetorical question as you know interest rates were higher than they are today, implying of course that the median multiple was low then because interest rates were high and the median multiple is high today because interest rates are low.

Then riddle me this. In jurisdictions where the median multiple was 3x median multiple in 1992 and still is 3x today, why did it not matter that interest rates were also higher then and low now?

Are you saying interest rates don’t matter and house prices would still be at 10x multiple in Auckland even if rates went back up to 10% or even 20%?
I agree that the RUB does increase prices, but I doubt we would be anywhere near 3x without it.

If rates went up that much, the economy would crash and you would be at a multiple a lot less than 10x very quickly.

Interest rates ONLY matter when land supply is restricted.

When we have restrictive land policies the tag gets to wag the dog.

In jurisdictions where land supply has fewer restrictions, the median multiple is approx. 3x income, irrespective of geography, immigration and interest rates.

NZ house prices, right up until the early 1990s, had always been around a 3x median multiple, irrespective of what interest rates had done up until then.

Are those places cheap because there is plenty of land, or are they cheap because the endless sprawl makes them not very desirable places? I’m guessing people go to those places because they are cheap, not because they are good.

If plenty of land per se was the driver then Australia would be giving land away and they nearly have the same median multiple as we do. Singapore, with a pop. of 6 million, fits into the size of Lake Taupo and they are only 6x median multiple.

It's the Govt. policy that dictates what and when you can do with the land and who restrictive it is to demand.

Check out Demographia Report on affordability in other places. http://www.demographia.com/dhi.pdf

What do I know but the mindsets and work ethics of singaporeans is a little bit different to us kiwis. Goes to show its not what you have but what you do with it that matters

Yes exactly.

We have unaffordable housing in NZ because that is what our Govt. policy sets the rules to, and which we all play to the rules.

My God man read some Evans and Bertuad.

Your earlier study that shows sufficient land release with the RUB is flawed due to the land-use restrictions extending so far out that any rural land on the fringe that has future potential to be upzoned has already been land banked and valued at this potential. So you are using a false highly inflated value rural fringe datum point on which to make all your other comparisons. The fringe land would not be valued at this price if the RUB boundary was removed as was promised by the present Labour Govt.

All that study shows is there is sufficient land available to continue the 10x median income to house price Ponzi, not how to make land more affordable, irrespective of the demand.

The graph in this article that shows the declining land value from the city center to further out is the same for any city in the world.

The main difference in the graph in cities that have less restrictive land policies, and the likes of Auckland, is that the fringe land in less restrictive cities is set at a lot lower price (due to competition because of greater ease of availability on the fringe) and increases at a lower price back into the city center, giving the same shape to the curve but at a lot lower price at any point on the curve.

Thus if people truly want to live closer in, they can afford to do so and are not forced out to the fringe and beyond like they are in Auckland.

Thus all land prices, anywhere on the graph line in Auckland, that are higher than those cities with less restricted land policies, are due to Auckland City Councils' ideology of trying to restrict supply and force people to live in the areas your article wants them to, at a 10x median income multiple.

Remove the RUB, build houses anywhere... result, no land for vegetables to feed the inhabitants

It doesn't work like that. The present system makes housing far more expensive so forces people to the fringe ie encourages sprawl. This is why people are moving out to places like Pukekohe onto elite soils. Which they can only do by council allowing zoning change on that land.

All the present policies do, is just make everything more expensive, which makes it difficult to buy what you really need.

If you make it cheaper on the fringe, then it becomes cheaper (relative) going in, so those people that would like to move closer in can now afford to do so.

Yeah sure mate... build in no mans land without amenities or jobs as far as the eye can see. You dont get cheaper housing in Mt Roskill Mt Eden Mt Wellington or Mt Albert where people actually want to live. You get empty spec homes that developers cant sell. Has building in pokeno reduced the cost of homes elsewhere, of course not. What about all of the sections in Te Kawhata they cant sell

Never said about building in no man's land without amenities. And the examples you are giving are ones that are built under the present restrictive system, which proves my point.

When they try to build for affordability alone, they have to buy on the far side of the RUB then they have built too far away for people that have to work closer in.

And in Pokenos case they only have to make the price that little bit cheaper than their competition closer in to get their market. The developer is going to charge whatever they can get away with, even if they could sell cheaper. If the RUB had been removed and land had become cheaper closer in, then there would be no point in building at Pokeno.

One of Auckland's problems is its Super City size also incorporates a lot of rural lands, which they are restricting. Whereas Christhchuch is made up of three councils, two of which come in very close to Christchurch city itself and provide competition with Christchurch City Council to keep prices more affordable.

You need to read Alan Evans, Alain Bertuad, etc. to understand how land economics really works. A quicker read of the 2021 Demographia report gives some good pointers http://www.demographia.com/dhi.pdf

Is Mt Roskill really a desirable place to live

Location wise it’s pretty good. Hard to know if the state house rebuild will make it better or worse.

Are you suggesting it's not?

I live there... It's a mixed bag, but I quite like it really. Good parks, good public transport. Cheerful and very diverse community-- a lot of sirens and the odd cooked individual, but also the best neighbours I've ever had. Really good soil too, the garden's cranking. I'm a renter but I'd be happy to buy there.

I live there... It's a mixed bag, but I quite like it really. Good parks, good public transport. Cheerful and very diverse community-- a lot of sirens and the odd cooked individual, but also the best neighbours I've ever had. Really good soil too, the garden's cranking. I'm a renter but I'd be happy to buy there.

Table 6 is interesting, the way to make land 67 times more valuable is to just re-zone it to to residential. Probably speaks to the ongoing extreme shortage of developable land whereas no one wants farmland.

Given the seismic lessons of the past decade in Christchurch and Wellington as well as the ongoing pandemic, is pushing multi-story and high density building development prudent? While cities must be livable there must be a better way than creating future deathtraps.

Given the seismic lessons of the past decade in Christchurch and Wellington as well as the ongoing pandemic, is pushing multi-story and high density building development prudent? While cities must be livable there must be a better way than creating future deathtraps.

Modern high-rise construction, built to (and actually inspected to meet) 100% of the earthquake provisions in the building code are not death traps.

It also requires the interior fitout to be done properly as well though, and even now it still often isn't.

There is only one reason increasing density close to the city has challenges: NIMBYs. The council can pretend it is infrastructure but they didn’t seem to have any trouble zoning density everywhere else.

Way back in the 90s land development around southern suburbs like papatoe was curtailed.... inadequate pipes. You could not even have 2 houses on 1000 to 1200 sqm. Different story now when you drive through there.... every new building is multi storey and multi unit

A great example of the council wanting your money. At sections that size, there is no need to be connected to a central system at all, there are far cheaper and environmentally friendly STEP systems available.

Council did the same when they ran water pipes past your property and made you pull out your rainwater tank and connect to their system.

Well, the National Policy Statement Urban Development has solved that wee problem.
There will be an outcry later this year when the council changes the Unitary Plan to enable much more density near train stations and centers.

No train stations in most of the central isthmus unfortunately. Maybe if they actually build the light rail it would have effect.

Mt Eden, Kingsland, Morningside, Newmarket, Remuera, Ellerslie, Greenlane...
Parts of Ponsonby are walking distance to the CBD so should also see some rezoning

Mt Eden station is at the very north of Mt Eden. So it still leaves the big area between the western line and southern line as fairly low density (most of Mt Eden, Epsom, Royal Oak, One Tree Hill, Onehunga, etc).

Yes, but an awful lot of land will be rezoned high density when you think of an 800m radius of each station.
Also, if light rail goes ahead (quite likely), there will need to be high density rezoning around each stop, that will capture quite large chunks of Mt Eden.

Yes light rail would change everything especially as stations would be closer together. Pretty much all of Mt Eden, Mt Roskill, Hillsborough, Onehunga Mangere Bridge and Mangere.

Watch prices escalate again, with rezoning

Mt Albert, Avondale, New Lynn, Henderson, Lincoln, Ranui
Penrose, Papatoetoe, Otahuhu and so on

Yeah.
I also forgot Orakei, Meadowbank, Glen Innes, Panmure....
A lot of the Isthmus will be covered

Pt Chev although no train line has excellent proximity and green links to the CBD via the cycleway and bus (plus motorway access). Map shows <15% THAB. Could get zoned upwards?

ie it’s majority suburban zoning to urban rather than to THAB

Nimbys in posh central and coastal areas won... they don't want no apartments, stick them somewhere far away

True. There’s a few going up on the ridge in Mission Bay but in hindsight Grimes was correct when he said the Bays should have apartment blocks on the waterfront.

Yeah, and the likes of great free marketeer David Seymour will ensure that places like Remuera and Mt Eden never get the intensification they rationally should. Property rights, don’t you know...

What a bizarre comment. Labour are in power. How’s that working for you?

They weren’t in power when the unitary plan was approved that left much of Epsom etc as single house zone despite being very close to the city.

That was the domain of the local council / mayor not central government. Phil Goff is very much a labour man.

Actually that's not quite right. The government mandated the need for the Unitary Plan, and also made submissions on it which resulted in huge changes to the final plan compared to the one the council prepared and publicly notified.

The Key government instigated the Independent Hearing Panel process for the Unitary Plan. I am no fan of what Key did with housing generally but that particular action was a positive move.

I make it because it's an example of staggering hypocrisy. ACT is all in favour of liberalising land use... oh, except when it's the wealthy electorate that Seymour lives in. It shows that behind the facade of sincerity, he doesn't even believe in libertarianism.

No, point completely missed.

You liberalize land use so that people, both as individuals and as a group can decide what they do with their own land within a neighbourhood context. That's what covenants represent on a subdivision. So if the people of Epsom as a group say they want to keep it as it is, then that is democracy. And all people should be given that right, (not just those ones that have an electorate MP that believes that), subject to environmental factors, including sunlight issues, etc. that would affect their immediate neighbours.

The biggest issue most land in most cities has is they have no covenants, so you are at the mercy of a centralized council suddenly zoning from single level to 6 storey, just so they can fulfill some ideological dream, and increase their rate base to pay for it,, and you as an individual are sacrificed for the greater good. But what that means most of the present individuals as a group are, while they think they are the greater good group, are only one of the property owners who the council wants to replace with 6 other ratepayers when they rezone.

If you have a problem with Nimbyism we can soon fix that for you
Give us your address and we'll move a couple of Comanchero and Mob families in either side of you

That can happen anywhere; in fact I would say most of those would want a big section not high density.

75 Woburn Rd, Lower Hutt, I call it the Vogel. You guys know that my fiancee is a woman for the people and would love to share her street with true kiwis

Duo of Robertson and Orr is working on Jacinda Ardens principal to remove inequality so are working to bring not only entire Auckland but while of NZ on par with Remuera / Mission.

Certainly did a job on the house across the road from her house in Sandringham

1) "Given this pattern, are we currently delivering most development potential where it is most valued and where relative demand is highest? Short answer: No."

Of course not as we have arbitrary zoning and density rules imposed by planners who don't bear the costs of those rules. There should be no zoning or density limitations and the constraints only being National Environmental Standards.

2) "It is good economics that those who receive the windfall value gain should contribute a commensurate share to the cost of that infrastructure."

No, its not. The land does not create a demand for the council services. The land should be subject to a central government land tax (resource tax), with other taxes lowered to offset them.
Councils should only be allowed to rate on improvement value, as it is the people living in the houses that create the service demands.

I’ve always said if we must have planning it should always be done with a compass (e.g everything within x km of CBD / train station / etc). Auckland’s planning seems ad hoc and almost entirely based on how loud people complain which is a really bad approach.

Well again, the national policy statement requires density near train stations and centers. Makes sense to me.

There is much to like is this analysis but Norman referring to his study being a landmark case is not true. It hasn't been peer reviewed and there are questions by the economic profession about the assumptions made in the study. I have heard some economists say privately they are surprised Norman's study shows UGBs having any effect on prices because the study assumes all the price of land is related to amenity i.e. that land bankers have no localised monopoly pricing power.

I wonder if he's left council, the way that the footnote is written.
His swansong?