Included with the City Centre Future Access Study documentation released late last year were the answers to a number of questions that the previous Minister of Transport, Steven Joyce, had asked in mid 2011. As well as requesting the preparation of what turned into the CCFAS, Joyce requested the following:

  1. “Finalisation of the spatial plan and master plan including establishing achievable growth projections for the CBD
  2. Demonstration of a commitment to resolving current CBD issues, for example by improving bus operations and addressing capacity issues
  3. Evidence of rail patronage increases, particularly in the morning peak, residential intensification and CBD regeneration as a result of current investment
  4. Beginning implementation of large scale residential developments along the rail corridors
  5. Implementation of additional park and ride sites, and changes to bus feeder services” 

There’s a lot of really interesting information in Auckland Council and Auckland Transport’s response to these questions, but for this post I’m going to look at an element of the third question: evidence of residential intensification as a result of current investment.

To help get an understanding of the level of intensification past/current investment in the rail network might have stimulated, the report compares growth since 2001 in areas with good proximity to the rail network with growth in other areas that were already urbanised in 2001 These are the areas looked at: rail-caus

A couple of years back Steven Joyce suggested that most intensification in recent years had actually occurred away from the rail corridors, by saying this:

Yep, we should allow the city to increase in density (watch councillors run a mile when it comes time for the district plan changes), and we should support cost-effective transport options that support that. But we also have to understand that people like to live where they want to live, and provide cost-effective transport options (roads even!) for those people too. Amusingly, Auckland has increased in density in recent times. But largely not where the central planners said it would, (along the transport corridors) and instead in the beach-side suburbs. Fancy that.

I suspect that Joyce asked Auckland Council the question about where intensification occurred because he thought he’d be able to say “gotcha!” then the results showed that investment in rail had not had any impact on development patterns.

Comparing the level of growth in Census Area Units near the rail corridors with the level of growth elsewhere in Auckland would test Joyce’s assumption and also help answer his question – of whether the investment in rail upgrades had coincided with higher rates of growth in nearby areas (recognising of course that correlation does not necessarily mean causation.

The results are quite interesting: intensification-rail-corridor

These results are reinforced by analysis of consents by type (looking just at the Auckland City Council area this time): consent-types

This shows that almost all “higher intensity” housing typologies like apartments and terraced houses constructed within the old Auckland City Council area over the past five years have been in the rail CAUs.

The report concludes that it would seem Steven Joyce was wrong and intensification has definitely been occurring at a higher rate in areas close to the rail system compared to other parts of Auckland:

Both population trends and building consent data indicate that intensification has occurred at a faster rate within the rail CAUs than elsewhere in Auckland since 2001. Development within the rail CAUs has occurred in a variety of different ways, reflecting different development markets in different parts of Auckland. Some important trends include:

  • Significant construction of apartments in the city centre.
  • A number of larger-scale intensive developments around inner parts of the rail network (e.g. Newmarket, Mt Eden, Kingsland and New Lynn stations).
  • New development areas around stations in the outer parts of the network (e.g. Sturges Road and Takanini stations).
  • General infill and small-scale intensification across the network.
  • Increasing household sizes in the outer Southern Line part of the network.

Combining analysis of population trends and building consent information with economic analysis suggests that improved rail services boost property values and therefore makes intensification more economically viable. It would appear that this process has happened in Auckland over the past decade, although in different ways in different areas – reflecting varying market characteristics across Auckland.

I think the way in which the report notes how intensification has occurred differently in different areas is quite important. I suspect the assumptions made by Steven Joyce (and others) around ‘apartments next to railway lines’ is an over-simplification of how intensification can occur. Apartments are clearly only market attractive in some locations and it’s unrealistic to expect them to be built right throughout the rail corridors any time particularly soon. However, other forms of growth such as general infill, new growth around some stations, small-scale intensification as well as apartments and terraces have happened and overall it seems clear that there has at least been a correlation between the rail corridors and a higher level of development over the past decade.

Not for the first time, it seems that Steven Joyce was wrong.

Share this


  1. Interesting and potentially useful analysis.

    Unfortunately I have to question the spatial resolution: It seems a bit “coarse” to provide a definitive answer to the question. Rather than using area units as the base unit of their analysis I wonder why Auckland Council did not calculate a 1km walk buffer around rail stations and then pro-rata the change in population for the underlying census meshblocks? That would be more accurate than simply aggregating area units. Personally I am sceptical of land use analyses based on area units, as the latter are often too coarse to tell you what’s going on. It may be that all that growth came from one “rail area unit”, e.g. the big block by the airport.

    Not that Steven Joyce is right. Which brings me to the next question; what does Joyce mean exactly when he says: “central planners”? I’ve never met someone with the job title “central planner”. Precisely who is he talking about? Central government planners? One would think not because land use planning is a local issue, as he notes elsewhere in his comment. No, I think Joyce’s use of this phrase actually says more about his simple/dated world-view than it does about his understanding of development patterns in Auckland (he’s very wrong). This world-view sees him invoking memes of Soviet style “central planning” when he is discussing local land use regulations/planning, as if we’re somehow making a choice between markets and communism (gasp! The “c” word).

    Which brings me to a personal bug-bear of mine. Maybe it’s just me not paying attention but I have never heard Joyce (or any other minister in this government for that matter) publicly acknowledge, let alone seem to understand, when markets may not be performing efficiently due to the presence of externalities, market power, transaction costs, or equity issues (especially inter-generational issues associated with selling assets, ETS subsidies for agriculture, and PPPs for transport funding …). This would in turn imply that Joyce views any “market”, no matter how imperfectly it is performing, as being better than “central planners” and their dastardly regulations.

    For Joyce’s benefit, and anyone out there who has been sucked into that tempting right-wing anti-planning meme, “central planners” are simply a group of people, let’s call them a “community”, who negotiate with each other according to agreed legislative processes (“the rules of the game”) to coordinate desirable outcomes that would be difficult and/or inefficient to resolve through pure market mechanisms. By extension, planning rules cannot simply be viewed as an encroachment on property rights, as people like Joyce would have you believe. In many places they can seek to preserve rights by, for example, preventing your neighbour building a sewage treatment plant next door and exposing you to noise, odour, etc.

    While I believe that planning in New Zealand has, in many cases, made poor decisions, e.g. minimum parking requirements, there is no need to throw the planning baby out with the communist bath water by questioning the whole profession/practise. Put simply, planning is what communities do to enable them to move forward in a way that is mutually beneficial. Now that Joyce is Minister of MBIE you would hope he starts to get to grips with some of the “finer” (some would say essential) economic concepts related to planning.

    1. Interesting question is whether people only walk to rail stations at the moment. Could rail be spurring intensification in areas that are close to rail by way of bus or car (and how the hell to we measure that?).

    2. Actually Stu, people will walk further than 1 km for a predictable or very high frequency PT service, the lower risk and certainty of getting a ride in these instances give people the incentive to walk further, up to 1.5 km from (possibly faulty) memory

      with regard to evidence and understanding of market forces and society in general, as we’ve seen with Brownlee and the CRL the Nats keep putting up straw men until it gets to the point that they just say b*gger it, I don’t like that answer

      1. Yes SteveC I know.

        In this case however the precise walk distance is not very important. All we are trying to show is that the relative (%) growth of locations with proximity (however defined) to rail stations is greater than those that do not, i.e. the last column in the table 4.15 above. Defining proximity in terms of area units is an extremely haphazard way to go about answering the question, in my opinion. Not saying the answer is wrong, just that it may be biased.

        As an example of potential sources of bias arising from the use of area units, we know that area units are generally defined by Statistics NZ using a combination of residential population (SNZ tried to make them broadly the same size) and natural boundaries. Because Auckland’s rail lines were historically used more for freight purposes, they tend to traverse more industrial areas away from the city centre that have low population densities. This in turn may mean that rail area units are larger on average than non-rail CAUs.

        Whether this potential bias has an impact on the results or not I just don’t know and it’s impossible to tell, but it’s generally good scientific practice to make any potential sources of bias as transparent assumptions in the experimental design (i.e. choose a somewhat arbitrary distance of 1km), rather than implicitly embedded into the analysis itself. That way you could, for example, repeat the analysis with 400m, 800m, and 1200m buffers to check that your findings are independent from the walk distance you have chosen.

        Nic, I’m very dubious about whether people driving to rail would spur “intensification”. Park and ride (especially when it’s free) simply allows people to make long distance trips by road and then rail and thereby live in areas that they would be less likely to live in otherwise.

        1. Perhaps CAUs were chosen due to a lack of recent data at meshblock level?

          I am looking forward to getting data from the 2013 census and seeing what that says.

          1. Maybe.

            Even so you could take the area units population in 2011 and pro-rata it across the meshblocks based on their 2006 populations. We have done that at work for the whole of New Zealand for both populationand employment.

            This in turn allows us to estimate 2012 population/employment densities at the meshblock level. It only took half a day’s work …

        2. What I was thinking is infill style intensification. I know on the North Shore most park n riders at the busway come from the east coast bays. Not walking distance to the busway at all, but neither is it a long distance trip. But yes, isn’t allowing people to make trips that let them live in areas they were less likely to live otherwise exactly what we are looking at?

          1. That is “sprawl” in my book. A more intensive form of it, for sure, but still sprawl. The reason being that if you consider the counter-factual (i.e. what would people do in the absence of park and ride?) you would probably find that we would have an even more intensive city – because many of them simply would not chose to live farther away.

            In the case of ECB that probably would mean fewer students living comfortably at home with Ma and Pa while studying at UoA ;).

  2. if there were land available I would expect there would be strong demand for apartments around the Greenlane / Market road area of the southern line. Market conditions are ripe for this given the current high prices of the predominantly ‘house’ property type in this area reflecting the area’s demand.
    I for one would be a keen buyer of a good quality apartment in the area close to the rail line and given the lack of park ‘n ride facility.

    I might have to take it to be that this development is not occurring because of a lack of available sites or that the market is catching up to this situation.

    Would love to prove our unelected tsar Steven Joyce wrong on the lack of rail line intensification.

      1. Yes I agree it will take time .. just look at New Lynn. And it will cross a rather large threshold, methinks, with electrification. I still don’t think that properties located close to train stations have fully capitalised these benefits – mainly because the average punter probably does not even know the electrification is happening.

  3. Interesting as those numbers are, they only represent Dwellings not population.
    While each Dwelling will have at least one person in it, I’d be keen to see the numbers for population in those dwellings by CAU.
    As I recall the census does track number of people per dwelling as part of the dwelling census form filled in by one of the residents on census night.

    What I expect you may see is that the “population” growth of CAU’s near the railway line is even higher than the non-CAUs.
    And after all, its population in the CAU not dwellings that actually counts as far as completely rebutting Jocy’es arguments is it not?

    1. Indeed, and at the most basic level the easiest form of intensification is actually just fully utilising the dwellings we already have, and the second most is probably splitting larger dwellings into flats to suit smaller household sizes.

      I can recall one study in Melbourne that claimed that at an occupancy of one person per bedroom across the whole city (which is fairly low considering that couples usually share and children often do to) would not only fix their housing crisis, they’d have about ten years further growth met without a single extra house. I.e, there are enough spare bedrooms across Melbourne to house the projected 2022 population. Probably the same here.

  4. Things are tracking in the right direction. But – transport is just of a range of factors influencing where people might want to live, and consequently where the residential property development goes. Add in factors such as the current old diesel fleet, and shocking punctuality figures, lack of holiday and poor weekend services, and I think it is just too early to see the major changes…..though as I commented, there does seem to be the beginnings of a positive trend there. And this is great news.

    A complicating factor for the eastern rail corridor, and southern rail corridors south of Penrose, is rail freight which are frequent and with significant noise. Overseas rail networks are starting to have to implement a range of noise mitigation measures above and below the track and these have been reported in several of the professional international rail journals. I can see that Auckland too will need to consider such measures if it is serious about high quality residential and office development alongside the tracks.

    Out west, the far fewer freight trains make rail noise much less of an issue, and once the EMUs take over it should become a non-issue. So, on that basis alone, the western corridor is going to be the one to yield the biggest increase in intensification. New Lynn will be very important to get right, and after some initial concerns over the roading layout, current signs are looking pretty good. The inner west stations are also going to be early examples of transit oriented development and in fact the potential is fairly consistent right the way along the corridor all the way out to Swanson.

    One one other corridor that could work very well is the line from Onehunga through to Newmarket. Ellerslie is already something of a model station both for residential customers and for employees going to/from nearby businesses. I hope to see a similar degree of development near other stations along the Onehunga branchline in future.

    In regard to the Nats – I believe Joyce will at a point in future, listen to the Auckland business community over the CRL and intensification, if they speak loud enough as one voice. Brownlee on the other hand……..

    1. tuktuk – agree that Onehunga through to Newmarket is a corridor with growth potential. The Te Papapa station area is ripe for brownfields redevelopment, lots of low-density commercial buildings there. With the EMUs and associated service improvements looming, surely developers are getting interested…
      And there is also still potential in Ellerslie itself – e.g. every day I bike past a car yard (!) on Kalmia St. about 200 metres from the station and right across the road from Ellerslie School – it is crying out for mid-density residential redevelopment.

  5. Re-post of comment I wrote a while back:
    Something which really annoyed me was Steven Joyce’s argument that because 94% of urban journeys in Auckland are made by road (I may not have the exact figure but it’s of this order), then transport infrastructure-spending should reflect this – i.e. spend 94% of it on roads because this “caters for people’s preference”. However Joyce seemed oblivious to the reasons for PT-use being so low, and that these reasons greatly skewed people’s ability to exercise any real preference at all. Strange then that Joyce never cited Wellington’s 30% PT mode-share (or whatever it is), as justification for spending 30% of funding on PT projects there. And he never thought to ask why Wellington’s PT use is that much higher. The answer is simple: Because the service is there, and because it meets people’s needs to a greater degree than Auckland’s. Equally obvious is that if the service is not there or if it is poor, then of course people will not use it. A pity that no one ever tried to call Joyce out on this during parliamentary questions!

    And the corollary to Joyce’s insistence that funding should reflect (and therefore reinforce) current modal shares, is the assumption that it must somehow be desirable to preserve this state of affairs and not to use transport funding as any sort of tool to alter people’s travel patterns. I guess you have to expect such a paucity of analytical thinking from an average kiwi bloke who made his money through a radio station and whose only prior understanding of transport was that “Kiwis love cars and rail is a bottomless pit”. And Brownlee and Key are no better.

    1. I’ve never met Steven Joyce, so I’m basing my impressions solely on the content of his comments and the few times I have seen him talk.

      My gut feeling is that he is an arrogant and manipulative person that is not interested in “truth”. As per the example you mention, he is a master at applying dodgy statistics inconsistently to support his positions, much more so than your average politician. While some in the media think he is brilliant (e.g. Fran O’Sullivan), I suspect his duplicitous behaviour will eventually bite him in the ass.

      I should add that none of my political commentary should be seen as being inherently “anti-conservative” whinging. I have fiscal conservative tendencies up there with the best of them and seek to manage my own household’s finances with Scandinavian precision. That is – balance the books but be mindful of wider benefits and costs :).

    2. P.s. Steven often used to trot out statistics like “80% of kiwis drive to work, therefore we should spend the bulk of our transport money on roads.” But as you say if he applied this consistently then approximately 10% of the NLTF (i.e. $300 million) would be spent on walking and cycling. Fat chance – it’s more like less than 1%!!!

      1. Being as he’s unelected and unaccountable to the electorate I don’t think anything would ever stick on him politically, Stu. If National lose the next election I don’t think Joyce or Key will stick around.

        1. Agree, but if he cocks up on a grand scale then he could lose his ministerial portfolios, at which point he’d probably also depart the scene.

Leave a Reply