The other week Australian planning expert Greg Vann came to Auckland to talk about his experience developing the South-East Queensland urban growth strategy, ShapingSEQ. A lot of what he had to say was transferrable to Auckland. While Queensland faces different environmental challenges that often result in different decisions about built form, Brisbane and Auckland are both mid-sized New World cities experiencing rapid growth.

(Unfortunately, Greg’s Auckland Conversations talk isn’t online yet.)

Among other things, Greg argued that urban planning had to provide affordable living, not simply affordable housing. Essentially, consider transport costs (including the cost of car ownership) as well as housing costs when assessing affordability.

This is a topic I’ve looked at in the past using Census data on household incomes, rents, commuting choices, and car ownership from Auckland, Wellington, and Christchurch. The data suggests that there is a ‘spatial equilibrium’ between housing costs and commuting costs: While rents tend to decline with increasing distance from the city centre, commuting and car ownership costs rise in almost exact proportion to offset them.

However, there’s a further wrinkle to this: Lower-income households tend to locate further out. This results in a situation where (on average) Aucklanders pay a similar proportion of their household income throughout most of the city:

Auckland map 1 Rent share

But where overall housing plus transport costs tend to take up a higher share of household incomes in outer suburban areas than on the isthmus:

Auckland map 2 HT share

Talking about similar patterns in Brisbane, Greg Vann made a really good point that I had never considered before: This pattern of population distribution can have significant effects on long-run wealth distribution.

Cities that encourage low-income households to save money on housing by spending more money on transport (car ownership, commuting costs) encourage them to invest in depreciating assets rather than appreciating ones. Over time, the resale value of cars goes down – eventually to zero – while homes tend to hold their value or even increase in value over the long run.

Consequently, well-off people who live in areas that are rich in public transport and cycling options will tend to save money on car ownership and invest more on housing ownership, while low-income people who tend to live in places without transport choice will do the opposite. Over time, this can contribute to rising wealth inequality.

Here’s a chart based on the data from my 2014 paper. Housing costs, as proxied by median rents, are higher in the inner suburbs (0-10km from the city centre). Households in these areas spend around $1000 more per annum paying the rent than households 20-30km from the city centre.

However, this is offset by higher transport costs in the outer suburban areas: Households 20-30km from the city centre spend around $700 more on car ownership per annum, because they have to own more cars, and around $1300 more on commuting costs per annum.


That extra $2000 a year is money that disappears from household savings and investment. People in the inner suburbs may be paying more for housing, but when they do so they buy an asset.

What can (or should) we do about this?

From my perspective, there are three basic answers.

First, we need to give everybody opportunities to live in locations where they can save money on transport and car ownership and invest it in housing instead. Basically, that means allowing people to build more homes in most suburbs in the city, and especially in neighbourhoods with better public transport, walking, and cycling options. (Other policies may also play a role, such as development of new social housing in accessible areas.)

In a lot of neighbourhoods, this will mean enabling options for ‘invisible density’, like low-rise blocks of apartments, terraced housing, or backyard granny flats. In other contexts, it might mean making the jump to mid-rise apartment buildings. But the point is: if this kind of thing isn’t happening, low-income households and young people will be at the back of a long queue for a limited supply of housing with good transport choices.

Second, we need to significantly expand the reach and speed of the city’s public transport network, as well as complementary options like safe cycling facilities. Giving people in more parts of the city these choices will enable them to spend less on depreciating assets like cars and save and invest instead. As Anthony Downs observed in his tome on congestion, Still Stuck in Traffic, this may be one of the more significant effects of public transport investment:

…public policies should help low-income households spend more on ownership of housing by providing more transit opportunities. (p132)

Getting there isn’t necessarily easy, though, as Auckland is trying to do a lot of things at once in the transport investment space. The Auckland Transport Alignment Project sets out a potentially transformative rapid transit network plan, including committed projects like the City Rail Link, Northern Busway extension to Albany and AMETI busway and near-future priorities like the Northwestern Busway and mass transit (eg light rail) to the isthmus and airport. But we can only build so many of these at a time!


That brings me on to the third point: Transport and land use planning needs to be better integrated. For example, when rapid transit upgrades go ahead, it would also be useful to rezone those areas to allow more housing to get built. This will mean that more people get the opportunity to benefit from increased accessibility and the option to save money by giving up their second (or third) car. Pretty simple concept, really.

What do you think about the impacts of urban form and household location on wealth distribution?

Share this


  1. Fourthly, it shows that decentralisation should be a key policy. Decentralising jobs both to suburban centres, and to regional cities. The all the eggs in one basket, late Victorian period built form of a too strong CBD at the expense of everywhere else isn’t equitable.

    1. That’s a pretty tough thing to make happen. Even if you wanted to.

      Agglomeration theories suggest it’s also likely to harm productivity significantly.

      1. Agglomeration, with newer communication technologies, is overrated.

        And keeping all the jobs in Auckland might be popular in Auckland, but here in the provinces we don’t agree. Best thing any Auckland based manager needing new staff can do is consider where in the provinces to rent new office space. Conversely my lovely Palmy job was moved to Ponsonby because the Auckland based managers thought we might like to join the Auckland real estate ponzi scheme. 12 out of 13 us thought otherwise.

        1. A paradox of the last 30 years is that ICTs tend to be a complement to face-to-face contact, not a substitute. Better communication and information processing has enabled people to do a lot *more*, but doing it in teams usually requires physical proximity. This means that ICTs have strengthened agglomeration economies in cities, not undermined them.

          I’ve found this when managing projects across multiple offices: When dealing with programmers / data analysts who are producing things that can be sent anywhere, instantaneously, it’s still easiest to sit down in a room with them to work through goals and issues. Trying to do this via email or over the phone can easily lead to costly rework.

        2. If there are no advantages to agglomeration then businesses will work out that it is much cheaper to be outside the CBD or even Auckland. There is no need for any policies, no one is holding a gun to CEO’s heads.

    2. It is a key policy (read the Auckland Plan), and has been a key policy for the last 40 years. That’s the policy that has given us Manukau, Albany and Westgate, and the rest of the metropolitan centers. Unfortunately it doesn’t work especially well. Turns out we still have transport and housing problems.

      One thing you might want to consider is the difference between the CBD and the centre of the region. You can try and move, augment or replace ‘downtown’, but you can’t change the centre of the region.

      The distinction is important because the closer you are to the centre of the region the closer your access to everywhere else across the region. That’s simple geometry, and trying to disperse demand centers around the place doesn’t change that.

    3. Auckland is a decentralised city. Only one in five jobs is located in the city centre and fringe. The commuting cost and housing price gradients shown in this data *already* reflect the impact of decentralisation of employment.

      Decentralisation of employment flattens the commuting distance gradient somewhat, but it doesn’t change the fact that the centre of the region is the most accessible place with the lowest average travel distances. I explained why in this previous post.

    4. Strong centres are hardly a Victorian invention. They’ve been around for Millenia, and aren’t going away quickly. Decentralisation is a cute idea, but it needs a lot more than just a policy push. Just take a look at Manukau and the Hutt to see how slow and how problematic this can be.

    5. Agree with you Matthew. Moving businesses out of central Auckland to where the personnel are is a sensible thing. The need to be central with modern communications is no longer necessary. Witness the changing scene from after work Friday drinks a few years back to people now declining and joining the exodus to the suburbs as quickly as possible to minimize the pain and stress of their journeys. The other thing that would make these people’s lives easier is to have roading that flows so they spend less time and money on commuting.

      1. I think the CBD bars would disagree with you, there as chocker as they have ever been on a Friday. More people catching the train or Northern Busway, means more people who don’t have to worry about either congestion or having to drive.

      2. Ha, the exodus! Neither the data nor simple observation on the ground agrees with you. The city is literally busier than it has every been with jobs and retailing. Never in Auckland’s history have there been more jobs, shops or residents living in the city centre, and to add to it they are building what, six new skyscrapers this season?

        And the bars are chockers on a Friday afternoon, you can trust me on that one ;).

      3. ‘…joining the exodus to the suburbs as quickly as possible to minimize the pain and stress of their journeys.”

        So decentralisation is great, but the direct effects of attempted decentralisation (and terrible PT provision) are destroying our society? You can’t be serious. Methinks this is a Patrick Reynolds parody account! 😉

      4. ‘Move to where the personnel are’ – The problem is that for any business with a skilled workforce the personnel are likely to be widely scattered throughout the metro area. The centre may be the best location for minimising their *average* commute time.

      5. So you think the govt or council should have policies to tell businesses where they must be located? Should they also tell them what they must sell, how they must make it, etc? I get the feeling your ideas might not work too well…

      6. Ummm think you should check your stats, CBD is still stuggling to provide more office space due to demand!!! Bar’s are also thriving.

  2. The distance from the city centre is only relevant if you work in the city centre but nearly 70% of Auckland’s workers don’t.

    1. Incorrect. Areas closer to the city centre have easier access too all of Auckland’s jobs because they’re in the middle of the region and… geometry.

      1. Absolutely right Brian, curiously people seem to get this back to front. The CBD exists in the middle of the region because it is the place with the greatest access to the regional population and economy. It’s not because someone arbitrarily decided to put all the jobs and the shops there. Even if the CBD didn’t exist at all, your closeness to the centre would still describe how many regional jobs you could get to readily.

        1. Nicker, I’m not sure that your statement is historically or factually correct. I’d put it to you that Auckland CBD is not there because it is the best, most centralised place, or the most convenient: after all, Queen St was a small creek down a valley between two hilly ridges: but instead it is probably there because there was a good anchorage there 160 years ago, and once property conditions get established, it is hard to alter their course.

          Geographically, a more useful centralised location might be Sandringham or Balmoral. If the city planners had thought far enough ahead of course…

        1. You just went full retard. I should post a meme. Sorry for the personal attack everyone, but seriously… I’ll go now.

        2. I’m amazed it’s that high (30 % working in the CBD), that’s a huge proportion for what must be a tiny proportion of the region’s land area.

        3. It’s not quite that high, from the last census data its 24% in the wider central area, including the CBD, Newmarket, Grafton, Parnell Freemans Bay and Newton.

          But still, a quarter of all the job in the Auckland region (1,086km2) being in 19km2 in the middle is quite remarkable, thats 24% of the jobs in 1.7% of the area, very telling. Must be some good reason businesses and workers keep wanting to be in the central city despite all the attempts to make them set up elsewhere.

          Anyway I didn’t want to debate this because it is completely not the point of Peters post, but it is fascinating stuff regardless.

        4. yes, 30% is what you hit when you cast the net a bit wider than the downtown to include places like Freemans Bay, Ponsonby, Kingslan, Newton, Newmarket and Parnell. And these areas would generate an even higher proportion of wages / GDP / tax revenue.

          I’d actually expect approximately 30% of Auckland’s regional GDP (which by extension is about 10% of national GDP) would happen within a fairly short late night drunken stumble of the city centre. That’s quite a lot of economic activity in quite a small area.

          Must be something good about it eh?

    2. See comment above. These maps and charts are based on data on how people *actually* commute – I’m not assuming that everyone travels into the city centre.

        1. The Auckland city centre is, as its name implies, at the geographic *centre* of the region. It’s also at the confluence of the city’s principal transport networks. It’s a logical point to measure from.

          The estimates of average commuting cost are based on Census journey to work data, which includes information on where people live and work and which allows us to map journeys onto the road or public transport networks. A key finding from this analysis is that people who live further away from the city centre tend to commute longer distances, on average. Even accounting for the fact that most jobs aren’t located in the city centre.

    3. imagine the size of the CBD if we had even 40% of the population working there!!! Let alone 70%. I don’t think our Unitary Plan would allow that at its full capacity would it?

        1. Is that 90% of WCC area or 90% WRC area as only having 10% of the regions work force outside the CBD sounds a little far fetched.

        2. It’s pretty simple Ted: Wellington has one centre, and of the people who work, most work in the centre. Centralised PT works well here.
          Other centres in the Wellington region, like Porirua or the Hutt, are not included in that, as they are separate cities.
          But in Wellington itself, if you live in a Wellington suburb, you pretty much work in the CBD.

          Which is one reason why having an earthquake in Wellington is a scary thing if someone declares a Red Zone – the entire city fails on the spot. If you close the CBD you kill the city. In Chch it didn’t matter so much as most of their retail activity takes place in suburban malls, but we have none (well, ok, one small failing mall in J ville) and so we have nowhere for office workers to relocate to.

        3. ” if you live in a Wellington suburb, you pretty much work in the CBD” is that accurate or is that just the people you know? It is hard to believe as not everyone works in an office but maybe just because you don’t know anyone who doesn’t you just assume everyone does.

        4. I find it hard to believe 90% in Wellington CBD. There seem to be plenty of jobs in lower hutt, airport, miramar etc. I wouldn’t be surprised if Wellington is sitting 40-50% though (which is quite high for a center).

  3. A more relevant point and one that is easily fixed is the UP minimum subdivision sizes. 400 sq m in mixed housing suburban. Thats a net size so add on driveways etc. So although there is no density control you cannot subdivide below 400M2. The net site area plus the 400m2 means that in reality many parts of auckland are harder to develop now under the UP than under the legacy plans.
    4 years wasted to get an inferior plan.

    1. That is for a ‘vacant lot’ subdivsion yes. However, if you do have a 700m2 section, you can build two dwellings on it and only deal with building consent. 3 or more then yes, RC consent is required.
      The plan was hoping to enable development before subdivision, rather than people subdivide then sit on that empty lot waiting for land values to increase.

      1. Thanks Anthony, So am I right in saying that on the 700m2 section someone can build 2,3 or more dwellings which is how I read the UP but I am not sure if you are saying they can or cannot subdivide. Because for most people having 2,3 or more dwellings they cannot subdivide is not really feasible.

        1. Yes you can subdivide if you consent the houses to be built on the new subdivision at the same time. What the limit refers to is how small you can do a empty lot subdivision, i.e. The minimum for slicing off a piece of land and selling it empty.

          The purpose of this rule is to ensure that council can check the small sites can actually be built on in a meaningful way. Basically if you want to have sub 300m2 you just have to show a design for houses that fit and meet the requirements.

    2. UP subdivision rules apply to the creation of *vacant* lots – they are large because the seller has designed anything to integrate and you need to provide sufficient area for the purchaser to establish a reasonably compliant development (setback from all boundaries that are now in seperate ownership, yadda yadda). If you build first THEN subdivide the rules are different, as you have ‘proved’ it all works together (including probably by having some ‘rules’ broken as the dwellings are on a common site to start with) and then the subdivision is literally ‘lines on a map’ defining the ownership. The rules implicitly incentivise integrated ‘urban’ redevelopment, not slice and dice suburban infill site development.

  4. All a moot point in most NZ towns and cities, where you can generally drive across town in 10-15 minutes, and the sorts of ongoing concerns you like to raise just don’t exist. That’s why NZ is so desireable.

    Auckland would still be like that too if employment was spread out with the populace. You are basically just highlighting once again the fact that centralised large cities are expensive by nature. One wonders why you bother to promote the concept so much, when you spend your days complaining about the outcomes.

    Go live in Wanganui and be amazed at how you’re left with nothing to write about, because everything is fully functional. Maybe then you’ll realise how backward Auckland is becoming, and then in turn start to see the real solutions to Auckland’s woes.

    Auckland needs to stop trying to be Manhatten. It worked well as a network of towns, and that’s what it needs to return to.

    Also, you say people who live further out have to drive more. You don’t seem to get it – they CHOOSE to drive more. That’s the lifestyle they want. The freedom of the road, with trips to the beach, into the ranges, up the coast, across Auckland to friends, etc. Perhaps if you lock yourself into CBD development dependency, you can’t see that?

    1. I couldn’t think of anything worse than Wanganui, but then again I like Auckland. The more obvious question is why don’t you live in Wanganui, given it is very obvious that you really don’t like Auckland? Set the trend, I’m sure everyone will quickly follow you.

    2. Setting aside the fact that you don’t seem to have properly read the post or the supplemental comments explaining the data, I have to ask: Why don’t you live in Whanganui if it’s so wonderful?

      Serious question.

    3. Geoff you constantly come here with this argument, so I guess you really think it is an important piece of news. But all you are saying is that small places [Whanganui pop ~44k on a busy day] and bigger places [Auckland 1.6m and rising solidly] are different.

      So true! They not the same scale, we get it. They exhibit different problems and opportunities.


    4. “they CHOOSE to drive more”. Yeah, that’s bull****. A five person family on a combined income of $100,000 lives where they can afford to pay rent.

    5. So what you are saying is Auckland should become a city like Manhattan or Melbourne as we have already got plenty of Wanganui’s right? Then we can provide people the advantages of city life (restaurants, bars, cafe’s, large events etc.), while if they want they can live in Wanganui and enjoy the quite life.

      I’m sure that what you were saying, were you not?

  5. The median rent figure for the 0-10km is distorted by the high number of apartments in the city centre and surrounds.A more correct analysis would be to use the median rent of a 3 bedroom free standing house in the 3 zones and you would get a different picture.

    Rents are generally a poor proxy for housing costs

    1. That doesn’t tell you anything about rental cost, but it tells you a lot about land cost. The assumption that only a freestanding house counts is a very peculiar part of the New Zealand psyche that I have never been able to buy into. Why does an apartment not count as a dwelling in your books?

      1. I’m not saying an apartment doesn’t count as a dwelling, just that it distorts the median price being quoted so there appears to be little difference in housing costs between inner and outer Auckland when we all know that is not is not true.

        Some sort of average capital cost would be a better proxy for “housing cost” but even then you’d need to net off the “cost” of owning the house with the capital gain achieved in a year as well.

        It’s bogus analysis

  6. Now let’s demand AT/NZTA carry out an OBJECTIVE study on rapid rail to Auckland airport. It has not be done yet, so in my opinion temporary ATAP should be reviewed with the next Government.

    Rapid Transit – Heavy Rail – connected to the entire heavy rail network.
    Mass Transit (and slower) – LRT – not what passengers need after a 15 hour flight. Perfect for Dominion Road.

    1. Jon – that’s really odd to be focusing on what someone coming off a 15 hour flight would need in a post that is so very clearly about the patterns of people’s commute from home to work and back.

    2. John, the air travellers are only a few percent of the demand for the line. You need to focus on the residents and the workers, think about what suits their needs. If you build a train line just for people catching planes you’ll guarantee yourself a white elephant that costs plenty and fixes little.

    3. To be honest after a long flight I’d rather jump on an LRT that runs every 5 minutes than a train that runs every 15 minutes, even if the LRT takes longer.
      I really can’t see heavy rail being more than every 15 mins to Auckland airport off peak, in fact it would probably be more like every 30 minutes.

        1. Well with per-unit marginal operating and staffing costs at around 1/3rd of an EMU, I am absolutely expecting that LR will be three times more frequent off peak. We are certainly talking about 1/2 hourly EMUs, or ten minute headways on LRT.

        2. Nick off peak HR would be the same frequency as the other main lines (currently every 20 minutes but it appears that might be increasing to every 15 soon) as it would be the other end of one of the current ones. LR would possibly run every 10 minutes off peak (I wouldn’t expect they would be much more frequent than 3-5 minutes during peak) on Dominion rd but I very much doubt more than every second or third LRV would continue to the airport during off peak time.

        3. Obviously you would run every LR vehicle CBD to airport on every run. The airport is an anchor station that ensures that your trains are consistently full along the line with people travelling both ways all day. Running every train full length is also far more legible for passengers.

        4. While it’s hard to predict frequencies given how far off either would be from construction, I think it is reasonable to expect LR to have twice the frequency given the vehicles would be roughly half the size, therefore for an equivalent level of loading they would need to be run twice as often. It’s probably mainly an issue later in the evening and early in the morning, as by the time either of these are constructed off-peak rail frequencies would likely be 10-minutes on main lines anyway.

        5. Twenty minutes during the interpeak yes, but the off-peak covers later in the evenings and all weekend too. Check out the gold coast light rail frequency and span:

          Compare that to the heavy rail frequency on the Gold Coast heavy rail line:

          The light rail frequency at it’s worst (excluding the bit where it runs 24 hours on weekends) is the same as the heavy rail at its best during the weekday peak!

        6. Why 1/2 hour headways off peak for rapid transit heavy rail to the airport when the Southern and Eastern lines currently have 20 minute off peak services? Are you exaggerating the headways of LRT over Rapid Transit? We’ve seen a lot of exaggeration in the benefits and costs of LRT over heavy rail by Auckland Transport.

          In anycase, let’s hope the next Govt takes over this issue as AT have not managed to secure a route in six years of existance even though previous studies by ARTA/ARC highlighted Rapid Transit as the best option over LRT.

        7. The Southern and Eastern have 20 minute headways *interpeak* Jon, they are 30 mins off peak. Simple fact is the operating cost of one light rail vehicle is substantially lower than one EMU. It stands to reason that with a given budget the LRT can achieve much better off peak frequencies. No matter how you want to cut it, EMUs are two to three times more expensive to run per hour. They have a lot more capacity and work very well when busy, but interpeak and off peak they are simply more expensive per unit or per passenger. This isn’t ideology or personal preference, this is reality.

          Please don’t start up with your pathetic conspiracy theories about secret cabals plotting to discredit your favoured flavour of trains, it really just makes you sound like a ranting crackpot. Don’t you want people to take you seriously?

      1. A slow tram every 5 minutes on a completely unbuilt line or rapid trains every 10 minutes on a line almost completed to the airport?

        Heathrow express leaves every 10 minutes… the slow Tube rattles on too.. But in Auckland the heavy rail system is already in place and the CRL is being built to cope with airport rail services.

        1. “A [light rail vehicle] every 5 minutes on a [] line [which can be completed for $1b on top of already committed projects] or [] trains [that will be no faster] every 10 minutes on a line almost [that will cost over $2b to be] completed to the airport?”

          It took an inordinate amount of work to fix the factual errors in your statement, but I think that it was worth it.

        2. The $1B LR only gets you from Onehunga or possibly at the most the end of Dominion Rd (SH20 end) to the airport as it assumes that the Dominion rd to Britomart section, maintenance and stabling facilities are already built and the LRVs are already in service so yes it is very misleading.

        3. But while the $1 billion price tag for lrt excludes the dominion road section, the benefits also exclude the dominion road section. Combined the lrt will give decent pt to a large number of people for the same price as heavy rail which will only give additional pt to mangere and airport.

        4. How is that misleading? HR assumes the $2.5 billion city rail link is built FFS. City Rail Link isn’t going to be open until 2023, there is a good chance light rail opens at the same time, or even slightly earlier.

        5. Well Nick the HR option doesn’t assume the CRL is there as it has nothing to do with it (the CRL increases the capacity of the network making an airport connection easier but with a little smarter use of the existing network it would be possible without it). HR does for some reason assume via Onehunga is the best option and includes upgrading the existing Onehunga line as well as the airport section. The LR option is priced just from the airport to I believe Onehunga but it is possible in includes up the hill to the SH20 Dominion Rd intersection. It does not include the connection to Britomart or the required facilities that a new mode will require or that LRVs will also be required to run on it. HR includes everything it needs to connect to the current network along with a huge overkill in requirements around Kirkbride Rd and requirements for the airport end to be underground while it is ok for LR to be at ground level creating chaos with the already congested airport roads.

        6. Are you seriously suggesting the build heavy rail to the airport then run it at two trains an hour? That is the worst idea I have every heard. If you want anything close to a decent frequency you need the CRL first.

        7. Nick I hope your reply is not to me and in the wrong place as I’ve never suggested 2TPH to the airport, I’m picking it would be the same frequency as the other three main lines. If you are referring to my comment about using the existing network better (the CRL is being built so it is irrelevant anyway) that means not sending everything to Britomart as the congested part of the network is Newmarket to Quay to Britomart, the rest of the network has plenty of capacity until it gets tangled up there.

        8. @BigTed, either the HR business case included the CRL, or it only ran 2 tph to the airport, or it took trains off of other lines.

          If you want to lump the cost of Dominion Road onto the LR option then you need to lump the benefits in too, then you need to put the cost of the CRL on the HR option too. Apples and Apples, you know?

        9. Heathrow express every ten minutes would probably be the equivalent of a train every hour to Auckland airport if you consider passenger numbers.
          I always took the tube because it cost about $40 less.

        10. Heavy rail should have been based on a rail spur from Puhinui. Cost would then be a new rail line through what is currently fields (Puhinui Peninsula) and a bridge into airport precinct. You could possibly run it as the Manukau – Airport shuttle to keep frequencies low – 15 min would be quite possible.

        11. You could analyse that option and get far fewer benefits for not much less cost, and then discard it. Or as a competent transport planner you could recognise that it’s a terrible option and discard it between your longlisting and shortlisting.

        12. Interesting Sailor Boy. Not much of substance to in your comment other than an insult and some derisory statement about benefits. The fact we are building East West link shows we have a dearth of competent transport planners. What went wrong with the short listing there!

  7. Forgive me for stating the obvious Peter. But you have compared rent with travel costs. Households closer in spend an extra $1000 in rent while households further out spend an extra $2000 on travel. That makes the inner area households better off by $1000 as rent money is lost to them in the same way as petrol costs are. If they buy a house then they will lose the money as higher finance costs. The question is why are people willing to pay an additional $1000 per year in total costs to live further out? They must be getting a benefit that to them exceeds that sum. A bigger house, a yard, better lifestyle for their family, access to schools. Or maybe they were going to buy that car anyway for other reasons.

    1. A lot of people have no idea how much their car actually costs them – I was surprised when I added it up.
      But yes a bit odd that this post says it costs less to rent close to the city and then says that people living close to the city get better capitol gain from their houses. They don’t if they are renters.
      And I can’t imagine it is cheaper to own a central house than a suburban house no matter how much transport costs are.

      1. Yeah, I can’t see how buying a house in the central suburbs and commuting less is cheaper than living further out. Doesn’t accord with my experience. Capital gains are awesome on central suburbs houses but if you can’t service the debt, you can’t get in there no matter how much you save in transport costs. And for what I could afford in the centrals suburbs, it would be pretty crappy house that I wouldn’t want to bring my kids up in…so no thanks

        1. Mwfic – do they REALLY pay the full costs?

          I would suggest taxpayers in Dunedin and ratepayers in Auckland will be picking up the real costs of infrastructure to subsidise those living in the outer burbs.

    2. Yep, that’s a fair point. The reason I did the analysis with rents rather than house prices (or mortgage servicing costs) was that was the data that was available from the Census. I’d expect rents to generally track with mortgage servicing costs, but there *does* seem to have been some divergence in rental yields between different parts of the city.

      Conversely, I also haven’t captured all transport costs – commuting and car ownership only. And I’ve only looked at the financial costs, not time costs etc, as those aren’t germane to discussions of whether you can financially afford to live somewhere. (But they do have an impact on your quality of life and children’s future prospects!)

      Basically, I’d see this as a useful but indicative bit of analysis. At some point I’d like to do some further work to close these gaps, but nobody’s offered to pay me for it!

      On a separate note, good to see you in person yesterday; sorry the format didn’t give us a chance to have a chat.

      1. Yes. I see your data as a form of revealed preference. People are paying to live in the suburbs so there willingness to pay indicates they must be getting some benefit from that choice. I am sure we will get a chance to catch up at some point. Hope you get rid of the Brontosaurus soon.

        1. In addition to that I suspect the determination to own a place plays a role, something that is definitely easier to do further out. This is an upfront cost so needs to be kept down, whereas the commuting cost is more ongoing so can be spread into the future.

          At the moment the option of a smaller affordable place closer in is not often available, if this was we might see that option being taken up. I agree though some, maybe many would still prefer a longer commute for a bigger backyard (I certainly wouldn’t).

        2. Writing the paper on housing and transport costs forced me to sharpen up my thinking on spatial equilibrium questions (and how the spatial equilibrium may be affected by various policies). Since then I’ve taken a much closer look at the determinants of house prices using hedonic methods, trying to get a quantitative sense of what local amenities people are paying higher prices for, and how planning policies may affect prices.

          That’s been fairly useful, but it’s also limited because prices are an indirect measure of subjective wellbeing. And even if you have a better measure of wellbeing, you still have the issue that people have different preferences, and so may respond differently to the features of dwellings and locations. It’s difficult to model that.

      2. Rents are next to useless as a proxy for housing cost

        For example Kelston Median house price $750k median rent $450

        Herne Bay median price $2.4m median rental $500


  8. Do they factor in additional travel time from living further out? Living near my work saves me 8 hours per week and I’m happy to spend a bit of money to save that time

    1. I’d like to see how that would be priced; door-to-door my journey home is about three times longer than my early morning commute; even though the actual moving journey part is only about 50% of the time. Hopefully the new network can improve this dramatically when it’s fully rolled out.

    2. For me, an equivalent house to what I live in now in the central suburbs would be an additional 1.5 million at least. It would probably still be a 15 minute drive (at best) to and from work. It currently takes me on average 45 minutes each way. And considering I’d have the car anyway, I’m essentially facing the argument of am I prepared to pay 1.5 million to save myself 5 hours of travel time a week and maybe $20 in gas per week…I guess everyone values things differently but at this point in time, I’m prepared to sacrifice some of my time for a peaceful sleep at night (not worrying about a massive mortgage over my head).

  9. Peter – as someone who is familiar with EEM – does this require consideration of ongoing maintenance costs of a motorway for example or just up front capital cost.

    If not,why not?

        1. Maintenance isn’t actually that expensive for a motorway. They just sit there doing nothing, occasionally you put the lights on and every 5 years you reseal them.

  10. Two issues with the argument:

    Public transport fare also increase relative to distance travelled.

    People who rent will not be benefit by the raising of property values.

Leave a Reply

Your email address will not be published. Required fields are marked *