AUT’s Briefing Papers initiative has kindly allowed us to syndicate their recent series on housing. The seventh briefing paper is by former Reserve Bank economist Michael Reddell:

There has been a strong sense this year that “something must be done” about high house prices, especially those in Auckland. To date, however, the policy responses display little awareness of how previous policy choices have made New Zealand housing increasingly unaffordable over the last couple of decades. Blaming investors or the tax system are largely distractions. And while non-resident Chinese purchasers may be bidding up house prices in Auckland (and Sydney, Vancouver and other cities) right now, these pressures are recent, while housing affordability problems are not. Increased demand for houses, whatever the source, doesn’t create problems if it is easy to bring new houses to market.

In 2013 Anthony King and Ivor Crewe published The Blunders of our Governments, which traced Britain’s experiences with various policy disasters over the previous thirty years. Unaffordable New Zealand house prices, especially in Auckland, are the predictable outcome of a similar blunder: the collision of two sets of perhaps well-intentioned policy choices.

In New Zealand, urban areas cover only around 0.7 per cent of our land. And land used for agriculture is just not that expensive. Prime dairy land at the peak of the boom in 2008/09 was selling for around $50,000 a hectare. Residential sections in Auckland, of less than a tenth of a hectare, are selling for ten times that much. The comparison isn’t precise – subdivisions need streets and footpaths etc, so one hectare of farmland doesn’t generate one hectare of residential land. But in research published in 2007, Arthur Grimes and Andrew Aitken found that land just inside the Auckland Metropolitan Urban Limit was selling for 10 times the price of land just outside the limit. Even if we had a construction industry that was as efficient and productive as any in the world, our house (house plus land) prices would still be very high just because central and local government together have created artificial scarcity. Auckland’s geography is certainly difficult, but that is all the more reason for having as few restrictions as possible on the ability of owners to use and develop land for housing. Land use restrictions appear to have become a much more serious constraint in the last 25 years. By contrast, in the 1950s and 60s, with a much stronger policy emphasis on home-ownership and house-building, planning restrictions (and especially land use restrictions) had a much less serious impact.

Of course, there are land use restrictions in place in local authority areas all over the country. Councils – staff and councillors – seem to feel a need to plan and central government legislation allows them to do so. But house prices in Auckland are four times those in Invercargill. That difference is really down to population growth differences.

In places where there is no population growth, land use restrictions still impose costs, but they don’t have much impact on house and urban land prices. And where land use restrictions aren’t very important, rapid population growth also won’t do much to boost house and urban land prices. An example is the big US city of Houston, where there are few land use restrictions. Over the last 35 years, Houston’s population has more than doubled while real house prices have actually fallen a little.

New Zealand’s land use restrictions are similar, in effect, to those in a range of other Anglo countries. House and land prices are extraordinarily high in places as diverse as Auckland, Sydney, Melbourne, London, Vancouver, and San Francisco. In each of these places, the inability to easily bring new land to market and use it intensively for housing runs into the pressures of rising demand from a growing population. When those two pressures collide, house prices rise. High house prices in these cities are not the result of aggressive and unwise lending by banks. They aren’t the result of speculation. They are just what happens when land use restrictions run head-on into population pressures. And housing demand is, in the jargon, quite inelastic. Existing residents of high-priced cities mostly don’t move somewhere else, partly because the big cities are where the jobs are. Sometimes the resulting high house prices are discussed as some sort of market failure, when it is a sustained failure of governments to allow markets to operate.

Where does the population pressure arise from? In the post-war decades most advanced countries experienced a high birth rate. Demand for housing rose strongly on account of this natural increase in population. Government policy choices didn’t have much to do with that source of demand, and governments in free societies generally don’t attempt to influence the rate of natural increase.

But natural population increase is now quite small in most advanced countries. Even in New Zealand, which has a relatively high birth rate by advanced country standards, the birth rate is only around replacement level. And for the last 40 years or so, New Zealand has had a large net outflow of New Zealanders, pursuing a better life abroad for themselves and their families. This outflow swings around a lot from year to year, but in total around 870,000 New Zealanders (net) have left in the last 40 years.

With little or no natural increase, and a substantial average annual outflow of New Zealanders, New Zealand’s population would now be falling slightly even with the modest rate of inward migration of non-New Zealanders we had in the 1980s. There would be no population pressure on the housing stock in the country as a whole, and probably only modest pressures even in Auckland.

But around 25 years ago, New Zealand immigration policy was reformed to encourage a much larger annual net inflow of non-citizens. The current annual target, reconfirmed by Cabinet only last year, is around 45,000 to 50,000 permanent residence approvals each and every year. That is one of the largest rates of non-citizen immigration (as a share of population) of any advanced economy. As one would expect, a disproportionate share of the migrants settle in our largest and most diverse city.

No one envisaged the impact on house prices when the land use restrictions became progressively more binding, or when the more expansive immigration policy was adopted. With hindsight the contribution of these two directly contradictory sets of policies is pretty clear. Land use restrictions might do little harm in a country with low or no population growth (the situation in many OECD countries today). And rapid rates of non-citizen immigration would have no adverse impact on housing affordability if land could be as freely used for housing as in Houston. As it is, the young and the poor, disproportionately of Māori and Pacific origins, find it almost impossible to purchase a house in our largest city simply because of the choices – blunders – of our governments. It is time for our government to confront that responsibility and to bring about change. If sufficient reform of land use restrictions is not possible – and the overseas precedents are not encouraging – the case for a significant reduction in the target rate of non-citizen immigration is pretty clear.

Share this


  1. I’d be interested to know how this is intended to work. A developer can just popup 10,000 houses wherever they like and the council has to pay for the necessary transport, pipes, etc? Does the developer have to build any parks, libraries, etc?
    Maybe in the old days when sewerage just went into septic tanks or the ocean and simple narrow roads of low quality were good enough it would have been fine, I can’t see it working now…

    1. Remember that no one is paying the real cost of their services in Auckland. Waterpark charges hardly cover O and M let alone the cost of capital and hence capital spend. Central interceptor is being paid for by rates and council debt not by Watercare charges. If Watercare charged rates that gave it a reasonable return on capital there wouldn’t be an issue. The same goes for roading.

  2. Surely this paper overlooks the disadvantage that accrues to those who live on the far outskirts of a city, and the very high costs of giving those outskirts people proper access to what makes the city worth living in – the whole ‘compact city’ discussion? Do the increased numbers in the city provide the money needed to build infrastructure – the stormwater collection system, the railways, etc?

    1. Many work on the “far outskirts of the city” and have quite different ideas to you about what makes a city “worth living”. Many US compact cities come with a very high crime rate, and UK’s attempts have met with similar issues. High rise apartments cost nearly twice as much per sqm as a simple single level house to build. Singapore and Hong Kong manage the cost problem by building very high apartment blocks so the land cost per unit becomes very low. Utility services including stormwater are quite cheap if planned before building starts.

  3. Extending urban boundary will bring our city back to sprawl, which goes back to auto dependant and congestion.

    Increasing density within urban limit is preferred, and I agree council should encourage more density without introduce expensive compliance cost.

    Current cost of sub-division $120k, and with construction 3k per m2.

    Also numbers of sub-dividable land is limited, with many land zoned for single housing, and suburban land having a relatively large minimum size (1 per 400).

    All of these drives up price.

    1. The pinnacle of this is the inner suburbs. Why would anyone in his right mind put single house zoning with a minimal plot size of 600m² on land worth $2000/m²?

  4. The article lazily refers to “landuse restrictions” when its clear he is referring to metropolitan limits. Not all landuse controls should be tarred with the same brush.

    If everyone (read: developers) are so keen on sprawl then they should pay the FULL cost of infrastructure. None of then currently do this, they look to the Council to provide the bulk infrastructure and then complain of the impact of development contributions on the prices of the houses they wish to sell.

    Corporate welfare – still alive and kicking!

  5. ” land just inside the Auckland Metropolitan Urban Limit was selling for 10 times the price of land just outside the limit”
    Well yes of course it does; the reason is quite obvious to someone who doesn’t have an ideological viewpoint to push (such as the authors of this article). Put the land outside the metro limits inside those limits and it will increase to around the same price.
    The reason land outside is cheap is precisely because it is outside.
    These people must know that; disingenuous or just dishonest?

    1. So if wee extended the boundary to Wellsford do you expect farmland on the inside of the line somewhere in the Kaipara Flats will be worth ten times the farm land just outside?

        1. I am trying to take Harrymc through a thought process that will enable him to understand that moving the supply curve will affect the market price. This is not a red herring it is looking at the situation of unconstained land supply.

      1. Pretty much yes. It’s funny how some people think the line makes urban land expensive. The reality is that being able to build on land makes it valuable, and not being able to build on it makes it far less valuable. The boundary keeps rural land outside the line cheap, not the other way around.

        1. So you seriously believe if you extended the boundary to Matamata ( let’s pretend it is part of Auckland) the price of dairy land would increase to $1m a hectare? With hundreds of thousands of undeveloped hectares between Auckland and Matamata? Don’t be absurd!

          1. Yes of course, for the places people want to build and live the price will not be dairy farm prices. For a home buyer anywhere where a developer had put money into roading, infrastructure, sewers, power etc to put land on the market, and where the council had collected developer contributions for the highways and bus stops and parks and community facilities… would cost approximately the same as any other subdivision in kumeu or millwater or wherever.

            If you are talking about the raw land market, Well the first round of developers would make money buying land at dairy farm prices on the edge of urban Auckland and selling it in 600m2 suburban blocks, which is what they do know by betting on/lobbying for land release. But it wouldn’t take long for those dairy farmers to not sell off their goldmine estates for sfa, when they realise what they are worth. as for the other areas, you can buy a hectare by Matamata for $200k right now. $20 a metre. It’s cheap because it can’t be built on sure, but more so because it’s remote from jobs and infrastructure and remote from transport to access jobs or infrastructure.

            Consider what you really mean when you talk about a supply curve. Is it the supply of unimproved and largely inaccessible rural land in the arse end of the Kaipara or deep in the Waikato that is restricted, or is it improved ready to build on suburban land handy to the motorway that is in short supply?

            The urban boundary keeps the land outside the line cheap. It’s just daft to talk about land “just over the border” being so much cheaper as an argument against the border. Remove the line and it won’t be.

          2. No one is suggesting that the land just outside the boundary is the “natural” price absent restrictions and that the price of buildable land will fall to it. That is a red herring. The price of the land hitherto outside the boundary will rise just as the price of the land inside will fall as supply increases. This is basic stuff.

          3. My view is the price of land inside the urban area won’t fall because you’re not increasing supply within the urban area. Ok you might get a bit of equalising at the edge where land in dairy flat is somewhat substitutable for land in Albany, for example, but land in dairy flat isn’t going to do anything for the price of land in Northcote or Mt Albert.

          4. I think it will, because the land/property markets in Auckland are highly correlated. i.e. we dont have a bunch of independent property markets they are all interelated through substitutability. And indeed Peter’s post this morning discusses this in terms of spatial equilibrium.

            While not many of the residents of Northcote might be prepared to move to Kaukapakapa, as you say, potential residents of Albany may well be. And potential residents of Sunnynook may well be prepared to live in Albany. And residents of Northcote may well be prepared to live in Sunnynook. And residents of Mt Albert may be prepared to live in Northcote. And residents of Mt Eden may be prepared to live in Mt Albert. etc etc.

            That substitutability can be thought about in terms of contours of equal price radiating out from the city centre. And it is reasonable to think that the gradient of price is something that is fundamental (for a given set of inputs e.g. transport options, amenities etc). This is because the price gradient between places represents the trade off between living in different places. The gradient is the value of time costs between living in Henderson and Morningside as per Peter’s post this morning. So if the gradient is fixed, and if we are able to pull the absolute value of land down at the outskirts, then this change will pull down the value everywhere, as the gradient between different places remains the same.

          5. But to pull the price down you need to get ahead of the curve somehow. If your land in dairy flat is cheaper than Albany, but adds an extra 15 minutes to your cost+time outcome, you’re probsbly still better off in Albany. Ergo that extra land doesn’t do anything for existing prices.

            For edge land to have major impacts within the urban area it ever needs to be ridiculously cheap, or have very good access/services. Given that people currently don’t value places like Matamata that have very cheap land but poor access and facilities, it seems you’d need the impossible, somewhere that is both well connected and serviced but under the curve on price, I.e, it needs to be desirable/valuable and cheap at the same time.

          6. The study cited in the post shows exactly that – there is ridculously cheap land just outside the boundary. So the question is, in an unconstrained environment without an RUB, how far away from the current metropolitan limits would be have to go for land prices to remain very low.

            I dont have time to do the calcs (but I might later), so I will have to use made up, (but trying to be realistic) numbers for now. How thick a ring of land would you need to provide, say a 25 years of land supply? 6km? I think that is a reasonable first order guess based on the amount of future urban land proposed under the PAUP.

            OK so we have land 6km from Albany (say) that will remain “second best” to other land for 25 years. How much is it worth? A reasonable estimate is that it will be worth the value of rural land plus the net present value of the difference between the value of rural and about to be subdivided land in 25 years time. If you discount the value of that difference over 25 years you end up with a value that is about a fifth of the undiscounted value depending on your discount rate.

            So lets do the math! Lets denote the value of just about to be subdivided land right next to the metropolitan area “BU” for boundary urban.

            And the value at 6km away we call “CR” for close rural land. Hows that?

            And the value of rural land unaffected by any potential for future urban development is R

            So based on the method described above,

            CR= R + BU/5

            And the value of the urban land BU = CR + 6G

            where G is the price gradient I talked about in my previous post, expressed in units of $/km

            So CR = R + (CR + 6G)/5

            Rearranging, 4CR/5 = R + 6G/5

            CR = 5R/4 + 6G/4

            =1.25R + 1.5G

            So at 6km from the urban boundary prices are 1.25x the rural price plus 1.5 x the per km price gradient. So depending on the value of the price gradient relative to the rural price ( and accepting that things will asymptote out at the edge), we can say that at 7.5km the price might be 1.25xR and at say 10k it will be back to R.

            So 10k from the existing boundary you will have rural prices. So the price of land at the boundary will be R + 10G. So we arent going too far away before we are back at rural prices.

            More generally however I think the onus of proof is on those who say that a policy that restricts supply doesn’t affect the price!

    2. I think all the parties here have a point. Matthew’s right that regulations that prevent land from being developed can artificially reduce the value of that land. (For what it’s worth, I think that Grimes and Liang’s estimate is basically credible – it controls for other determinants of land prices, such as proximity to major centres and the coast.)

      However, Nick’s also got a valid point, which is that the location of the MUL may coincide with other, unobserved factors that influence land prices. To give two examples:
      * Auckland Council (and its predecessors) have in some cases drawn the MUL to prevent development on flood-prone land, which is likely to be less valuable than adjacent land even in the absence of the MUL
      * Land just inside the MUL is more likely to be serviced with transport and water infrastructure, or already subdivided. Once again, this will tend to push up land prices inside the MUL.

      I suspect that these unobserved factors partly explain Grimes and Liang’s estimate. However, I’d be surprised if they were a full explanation.

      Finally, I’d also observe that other market imperfections, such as the lack of efficient congestion pricing and regulations that prevent intensification, will also contribute to the “MUL effect” by artificially inflating demand for greenfield land.

  6. “That is one of the largest rates of non-citizen immigration (as a share of population) of any advanced economy.”
    So cut the rate of immigration to a sustainable level (particularly immigrants from non-advanced economies…particularly China, India, Africa, Middle East). This would help relieve pressure on housing (especially since a majority of immigrants from developing countries choose Auckland to settle in).

      1. Brits, Europeans etc are more likely to settle outside of Auckland than people from developing countries. In NZ we have 2 population issues: 1) Auckland is currently growing too fast for it’s housing ability. 2) Most of the rest of the country has fairly stagnant or very moderate levels of population growth.
        and yes I would rather have immigrants from developed countries rather than from developing countries that lower our standards of living and wages etc.

    1. China is a non-advanced economy now is it? That’ll be news to all the high-tech manufacturers there.
      We’re the ones that sell raw commodities rather than value-add products. That’s non-advanced.

      1. Yes that is correct. Sure it has a lot of developed features and has parts that are more developed than NZ but overall it is still a lot less developed on average, it has half of it’s population barely above poverty levels, has the biggest pollution problems in the world. Most of our manufacturers moved there to take advantage of their cheap labour and cheap costs. Sure we send them raw commodities (which is due to idiotic government policy) so does Australia, Canada… are they also non-advanced?

Leave a Reply