Auckland Rapid Transit Network

This is AT’s official future vision for  the Rapid Transit Network in Auckland. I feel the need to show this again in the context of a number of uninformed views about the CRL popping up again, as one of the chief misunderstandings is to treat the City Rail Link as a single route outside of the network it serves.

All successful transport systems are designed through network thinking and not just as a bunch of individual routes, this is true of our existing and extensive motorway network just as it is true for our rapidly growing Rapid Transit one. The Waterview tunnel is not being built just so people can drive from Mt Roskill to Pt Chev, and nor is the CRL just to connect Mt Eden to downtown.

The CRL is but one project on the way to a whole city-wide network, as is clearly shown below, and as such it doesn’t do everything on its own.

But then having said that because it is at the heart of  the current and future city-wide network it is the most crucial and valuable point of the whole system. That is true today and will continue to true for as long as there is a city on this Isthmus. In fact it is hard to overstate the value of the CRL as by through-routing the current rail system it is as if it gives Auckland a full 100km Metro system for the cost of a pair of 3.4km tunnels and a couple of stations. This is simply the best bargain going in infrastructure in probably any city of Auckland’s size anywhere in the world and is certainly the best value transport project of scale in New Zealand. Because it is transformational* for the city and complementary to all our existing systems, especially the near complete urban motorway network.

Additionally the capacity it adds to the region’s whole travel supply is immense: taking up to 48 trains an hour this can move the equivalent of 12 motorway lanes of car traffic. All without flattening any place nor need to park or circulate those vehicles on local roads and streets. And all powered by our own renewably generated electricity. This is how the city grows both in scale and quality without also growing traffic congestion.

AT Rapid Transit Network - 2041

This map will evolve over time as each addition is examined in detail. For example I expect the cost-effectiveness and efficiency a rail system over the harbour, up the busway and to Takapuna to become increasingly apparent well before this time period. In fact as the next harbour crossing, so we are likely to see that in the next decade, otherwise this is that pattern that both the physical and social geography of Auckland calls for. Additionally Light Rail on high quality right-of-ways, although not true Rapid Transit, will also likely be added in the near term.

Welcome to Auckland: City.

* = transformational because it substantially changes not only our movement options, the quality of accessibility between places throughout the city and without the use of a car, but also Auckland’s very idea of itself; we have not been a Metro city before: It is doing things differently.


Matt suggested adding this more recent version. I agree this is a good idea, it shows just how quickly ideas are changing in Auckland right now. This is a very fluid and exciting time for the city as the new possibilities are becoming acknowledged by all sorts of significant players. It remains my view that extending our existing rail system is better for Mangere and the Airport, but that taking AT’s proposed LR across the harbour in its own new crossing is a really good option:

RPTP potential LRT + RTN Map

And just this morning we get wind of these very big changes for those making plans for Auckland. It looks like the funding roadblocks [pun intended] for the necessary urban infrastructure that the growing and shifting Auckland needs may be melting away….?

Treasury Tweets


97 comments to Auckland Rapid Transit Network

  • Sorry, I’m confused. I thought the Newton station was nixed in favour of just using Mt Eden. Is it back or are they just renaming the Mt Eden station Newton?

    • jezza

      Further to Nick’s comment this looks like the CFN running pattern for trains, I was under the impression that AT had lost interest in the Mt Roskill spur, to instead be able to run 5 minute frequency all the way to Henderson. Do you know how old this plan is Patrick? PS: Completely agree with this article you have written.

    • Stu Donovan

      Yes, I underztand the Newton station was abandoned due to costs and instead replaced with Mt Eden. This must be an old map.

  • Yes it is an older map; as I say the details will alter as each section is actually added, post is about Networks and Rapid Transit and the work in progress that is Auckland…..

  • Gary Young

    Thanks Patrick for such a clear and concise statement of the need for the CRL. I’ve noticed in recent times much less use of the term ‘loop’ in the media and in comments so perhaps the message is getting through.

    As an aside, has the plan for a full rail link between Mt. Albert and Onehunga been officially abandoned?

  • AnalogKid

    There’s a big blank space, just above left of centre. When is Kaipatiki going to be served by any kind of rapid transit? 2050 perhaps?

    There are plenty of people on the Shore who don’t live next to the motorway corridor. And the most congested part of their journey is getting to the motorway corridor.

    I don’t know what the solution is. Perhaps in the absence of rail on the Shore and light rail down Onewa Road, express bus services from Highbury and Glenfied would be a start, and proper enforcement of the T3 lane.

  • Warren S

    Hi Patrick – How did you get on with Jonathan Milne of Sunday Star Times yesterday?

  • Ben T

    remember this doesn’t include the ferry network they are expanding which accommodates places like Devonport, Birkenhead and Half moon bay already. this is further connection of the network that doesn’t meet congestion when travelling.

    • Trundler

      The ferries aren’t really a rapid transit (i.e. turn up and go, 10min service). Well certainly not Birkenhead at the weekends! Need light rail up Onewa road and then up towards Glenfield. Huge catchment.

      • Ben T

        Agreed, they are slower than a train or bus on a bus only lane, but with improved network connections further up the waitemata harbour you would end up getting more regular service during the weekdays and less spacing on the weekend schedules.

        I guess another thing is making sure you know the timetable and planning around it – you see people running for trains in the morning even though they know when they are scheduled to leave. (albeit some trains arrive later and these people could be early for the next train)

  • I have updated the post with an additional map and some news…

  • Angus Robertson

    Light Rail is rapid mass transit, if people live close enough. In Melbourne the tram network connects increasingly densely populated areas and provides great quick mass transit.

    It is unfortunate that our current Auckland City council plan to keep the inner isthmus single house dwelling until 2041.

    • To be properly considered Rapid Transit requires a separate right of way. This is not the case for most of Melbourne’s tram routes, and won’t be for Auckland’s. I have sat on trams stuck in traffic in Melbourne; that ain’t Rapid. Auckland’s proposed LR network is likely to have better right of ways than much of Melbourne’s because they will be new, and will have to to justify the capital investment. But still, they fall outside of the definition.

      Here’s a taxonomy from Jarrett Walker’s book. The key is not the vehicle but the quality of the RoW:

      • Konradk

        Yeah, a good ROW or some sort of separation is vital for speed and reliability – even if it isn’t truly ‘rapid’. Segments like St.Kilda Road (semi-separate ROW, now with tram stops placed further apart) should be the goal for Auckland light rail if there’s no room for dedicated ROW.

      • Steve Cable

        the keys are right of way and level of service, one without the other is limping

    • Stu Donovan

      yes, most of the isthmus needs to be allowed to intensify at least to Melbourne and Sydney levels of inner urban density. That means zoning that alllows for development of at least 3 storeys and preferably 6 storeys. This would allow development comparable to many of Melbourne and Sydney’s glorious inner city areas.

      I mean, given Auckland’s growth and geography that’s where the intensification has to happen.

      • The zoning debate is set for early next year in the UP. Basically yes, the single house zone has to substantially reduced in favour of the urban mixed zone, but then even that is only 3 stories, i think?

        • conan gorbey

          Even the size of the single house zone needs to reduce. 600sqm? It applies to half of Ponsonby and Grey Lynn with their 300-400sqm sections and people seem to survive and the occasion sale seems to go through.

      • Angus Robertson

        Perhaps Treasury might start “actively considering” advice to the Auckland Council to free up a bit of this land for development.

        • conan gorbey

          The issue Angus is that most of this land is already taken. What really needs to happen is for their to be less restrictive development (and heritage) controls in this area so people can make the choice as to whether they want to redevelop their property. At the moment I can’t for example do anything meaningful to my property apart from the standard choice that people around here seem to do and simply make their single dwelling larger which adds nothing in terms of extra density.

    • If you crop this map to just the North Shore, you’d think the city centre is in the north, rather than in the south. Similarly, the isthmus has much more of this low density zone than the area across the Tamaki River. Actually the biggest concentration is in the areas close to the CBD.

      I am quite curious about the rationale behind this.

  • Matthew W

    If rail is to be funded on the same basis as highways – will this be from a fuel tax on trains?

  • mfwic

    Lots of things are claimed as transformational. Some are and some are not. The problem in evaluating any project that leads to a fundamental change means you have to abandon the data you have you end up with little more than faith. But CRL should be justified even if it just continues the linear growth trend. Modelling the impact of any form of alternative in a congested network is a really hard thing to do and because it is hard you tend to leave benefits out. You can only count what you can actually model. It is hard to imagine CRL not giving a good return on investment. First in getting people into the CBD but second in allowing more offices to be built there.

    • Matthew W

      So, to paraphrase, we are investing 2.8 billion on gut feeling.

      • Warren S

        No, on clearly evident statistical trends and the clear evidence of similar cities overseas.

        • Matthew W

          Well we know the BCR of this project has been estimated as low as 0.2 so the evidence is the project is not worthwhile. Mwfic’s comment was – you can’t hope to estimate the benefits, but they will be big enough (apparently based on gut instinct).

          • Since when have made up numbers with no source become a thing here?

          • Matthew W

            Sorry I did make that up, it was 0.3. Same diff when it’s away below unity though right.


            I started realising things were getting made up around here when I pointed out an issue in some analysis in a blog post the other week (that potentially negated the core conclusions), saw no acknowledgment of my comment or retraction/amendment of said analysis, but saw that the issue had been acknowledged (I assume privately) to another NZ blog host.

          • What blog post. There are a lot of comments that get made and most of us don’t get a chance to read them all.

          • Matthew W


            Actually the latest on the BCR for the CRL from AT is the CCFAS which has it at <1. Of course they say this wasn't a business case but that doesn't take anything away from the BCR calculated.

          • I was made aware of the issue of watercare figures not being included and tweeted it straight away from memory and I thought I commented in the post but it doesn’t seem to be there. We don’t hide from things if we’re wrong but it’s not always easy to immediately correct things when we’re all busy doing other stuff too.

            FYI it was council who alerted us to the changes

          • Harriet

            City Centre Future Access Study is the latest I believe

            0.4 when you don’t account for wider economic benefits
            0.9 when you do

            Also this is based on NZTA current models which are based on historical data modelling (Though don’t quote me on it, I dislike stats in social sciences I find them about as useless as a screen door on a submarine) what we have seen however is a massive trend shift since the late 1990s that puts these models well out of place. For example these models have greatly underestimated trends for Brit, Northern Busway & etc. while current motorway projects are finding themselves not meeting the models.

            Page 15.

          • Peter Nunns

            “Well we know the BCR of this project has been estimated as low as 0.2 so the evidence is the project is not worthwhile.”

            Your figure is both out of date and wrong. You referenced a 2011 report in which AT and MoT were arguing over how to calculate transport and wider economic benefits. Even though MoT took a more conservative view on the benefits – an overly conservative view, in my personal opinion – it was still enough to get central and local government to commit to further investigation of the project.

            Following on that, the City Centre Future Access Study was commissioned by both local and central government to investigate options for addressing constraints on access to the CBD. That report found that the CRL was the most economically efficient alternative – simply running more buses from the west and south would stuff up key access roads. CCFAS estimated a BCR of 0.9 for the CRL – my understanding is that this is best understood as a minimum estimate of the BCR.

            Based on the CCFAS, local and central government agreed to progress with route protection and a full business case, but they still disagree on the appropriate timing for CRL. Central government thinks that you could optimise the BCR by waiting until 2018-2021 to start construction, while Auckland Council thinks that it would be better to start today. (Hence their decision to pay for early works on Albert St.) These decisions were made in light of the available economic analysis, which suggests that the CRL’s BCR is considerably higher than the 0.2 figure that you cite, and almost definitely higher than 1.

          • Matthew, FYI the MoT review you cite did raise a few valid points in relation to the original case but in the main it was bogus and staff were deliberately trying to find ways to make CRL look bad. At the same time as MoT were reviewing it AT had a team of overseas experts doing the same who raised some of the same valid points mentioned but had very different outcomes about most of it compared to MoT. One example of where the MoT deliberately tried to sabotage it was in the inputs for the modelling. The original business case suggested parking prices would rise to ~$30 per day but MoT said that would never happen and put in figures half that. Well we’re already over the figures the MoT stated wouldn’t happen till 2041.

          • Matthew W

            “That report found that the CRL was the most economically efficient alternative – simply running more buses from the west and south would stuff up key access roads.”

            A BCR of less than 1 can’t possibly be the most economically efficient alternative – do nothing has a BCR of 1 (or around 1.2 if the deadweight loss of taxation isn’t being explicitly included).

          • Matthew W

            Yes I got the BCR out by 0.1 apologies. Hard to keep up with all the negative net benefit transport projects these days. The point I was making is, once you get An economic evaluation with negative benefits, you are effectively justifying the project on opinion.

            Once you get BCRs of less that 1, BCR becomes a poor desicion making metric. The most economically efficient option would be the one that minimises negative net losses. Which is always going to be do nothing.

          • But there is no real BCR below one.

          • Matthew W

            CCFAS estimated a BCR of 0.9 for the CRL – my understanding is that this is best understood as a minimum estimate of the BCR

            Just reviewing that – the 0.9 includes WEBs so I think can hardly be called a minimum estimate. In fact fully 50% of the benefits are in WEBs. I am not saying that is wrong, but that is a fairly “bold” proportion of WEBs when the consensus around if and how to include them in a CBA is still up for debate.

          • Matthew W

            “But there is no real BCR below one”

            I am not sure what you mean – BCRs below 1 arent real by definition? That isn’t correct, it is a real, meaningful result of a CBA.

          • Peter Nunns

            “Just reviewing that – the 0.9 includes WEBs so I think can hardly be called a minimum estimate. In fact fully 50% of the benefits are in WEBs. I am not saying that is wrong, but that is a fairly “bold” proportion of WEBs when the consensus around if and how to include them in a CBA is still up for debate.”

            I’d describe it as a minimum estimate for methodological reasons – the modellers made some choices that probably resulted in under-estimates of benefits. Including, but not limited to:
            * Assumptions about parking prices in the city centre – they assumed that average prices in 2041 would be lower than today’s earlybird prices. Yikes.
            * Annualisation of PT benefits – the approach used tends to over-value benefits from peak-only services and under-value benefits from all-day services provided by CRL.
            * Interactions between ART (regional traffic model) and SATURN (CBD traffic model) – results appear to have been combined in such a way as to exclude a significant chunk of decongestion benefits.
            * Air quality benefits resulting from fewer buses running on busy streets have not been calculated, although an evaluation procedure exists. In my experience, these benefits can be surprisingly large.

            Lastly, WEBs are no longer under debate for transport projects. NZTA included them in the core Economic Evaluation Manual in 2013, and CCFAS calculated them using approved procedures.

            Basically, when you look at the numbers in more detail, the case for CRL starts to get stronger, not weaker. The opposite tends to be true for highway projects, partly because they receive less critical scrutiny from central government. For example, see Stu’s take on the Mill Road evaluation:

          • There are indeed, and rightly, debates around the use of agglomeration benefits in transport project evaluations. And a significant part of this is about attempts to add them to projects and modes that are essentially dispersive [ie are anti-agglomerative in nature] particularly motorways.

            If ever there was a project that most closely fits the criteria for enabling agglomeration it is clearly this one. If radically improving the access to employment, education, and commerce for tens and tens of thousands of people everyday, to, from, and through the nation’s biggest concentration of jobs, fastest growing residential area, biggest tertiary market, and premier retail precinct, isn’t an absolute perfect example of agglomeration I’m not sure what is. Furthermore this is a project that connects the nation’s fastest growing place and the nation’s fastest growing transport system. And one that, because of it’s spatial efficiency and huge capacity, concentrates rather than separates. Urban economics are all about agglomeration, and that is what the very nature of the CRL; urban and agglomerative. These benefits are most likely understated in the CCFAS especially as it wouldn’t include resent ridership or city growth data, but also because CCFAS had all sorts of strange assumptions at its core.

            Your ‘concern’ about agglomeration benefits should be taken to those projects where they really don’t pass the sniff test; the RoNS.

            Furthermore you seem to be trying to pooh-pooh agglomeration benefits because they are hard to quantify. That’s right they are, but that is a dangerous trap; insisting only on the easily quantifiable be included in evaluations is to mischaracterise the world. You know: ‘Not everything that counts can be counted, and not everything that can be counted, counts’. Or easily so.

          • Matthew W

            “Lastly, WEBs are no longer under debate for transport projects. NZTA included them in the core Economic Evaluation Manual in 2013, and CCFAS calculated them using approved procedures.”

            I didn’t realise NZTA was the global authority on economic evaluation! These are the people who brought us the three part evaluation method of efficiency effectiveness and strategic fit (where efficiency is the CBA, effectiveness is just a restatement of the benefits in the CBA and strategic fit is just about whether the politicians like it). I am not saying including WEBs is wrong, but there is debate about adding them to transport benefits and potential double counting.

            Anyway the fact remains even 0.9 is well below a BCR of say 1.2 required for the project to be viable let alone the best thing since sliced bread (or in other words better than other projects in a constrained funding environment), so those conservatisms you have alluded to (link?) still have a heck of a hill to climb.

          • Matthew W

            Hi Patrick,

            I am not disputnig the use of WEBs, I am just saying that an assessment that gets fully 50% of its benefits from WEBs is unlikely to be particularly conservative.

            Regarding agglomeration effects, I am sure you are right, the CRL will provide significant benefits in this regard. However my understanding is that while there is evidence for agglomeration effects at the metro scale, agglomeration effects at the level you seem to be implying are more valuable have not been proven. Agglomeration effects relate to the size of the metro area but not necessarily to the density of its core. So, for example, urban motorways would also bring agglomeration benefits.

          • Stu Donovan

            Matthew W – if I am interpreting your comment correctly, then my understanding of agglomeration would seem to differ somewhat from yours.

            Specifically, my understanding of the literature on agglomeration economies is that it identifies several microeconomic channels through which agglomeration can benefit producers and consumers (must studies, but not all, focus solely on the former). These micro-economic channels attentuate over space at different rates. Put another way, agglomeration economies have different spatial scales.

            Some agglomeration economies, such as labour market pooling, tend to function at regional scales. Literature suggests that these agglomeration attenuate to zero after about one hours travel-time, which is more-or-less what you’d expect based on people’s willingness-to-commute to work on a regular basis. Other agglomeration economies, such as knowledge spillovers, attenuate more rapidly. Arzhagi and Henderson, for example, found that they decay to zero after about 15-30 minutes walk. I.e. within centres.

            But just because localised agglomeration economies, such as knowledge spillovers, attentuate rapidly is not equivalent to suggesting they’re less economically significant. In fact the literature I have read suggests quite the opposite. Ciccone and Hall, for example, find that productivity is more strongly related to urban density than urban scale. This may be what Patrick is referring to. Venables developed a nice microeconomic model which considered the additional fiscal externalities introduced by progressive income tax scales (which tend to suppress agglomeration).

            From my reading the NZTA’s WEB processes focus on agglomeration economies at the metropolitan scale, i.e. they are not that well-suited to considering localised agglomeration economies. A few years ago I contributed to an NZTA research report which considered these issues, which you can find here:

            There’s a spreadsheet model which you can use as well.

          • Peter Nunns

            Matthew, at this point you seem to be arguing for the sake of arguing. Several people have provided in-depth responses, including pointing out that you had the numbers wrong. (You originally said CRL had a BCR of 0.2.) You’ve restated your criticisms but don’t seem to have taken the time to consider what other people are saying. While I often appreciate your take on things, I think you’re being needlessly stubborn on this one.

          • Matthew W

            I shouldnt have been so precise in my quoting of BCRs, and I was always well aware of the CCFAS. But my central point remains, I could have simply said studies have shown the BCR to be less than 1.2. I assume everyone is arguing about my central point which is what the post is about, (not my misquoting of a BCR by 0.1 which would just be nitpicking) that is that the economic case for the CRL has hardly been proven.

            Your argument, which I have seen in various forms again and again, is that the lack of an economic case doesn’t matter because the assessments were conservative in terms of benefits. Well thats fine but it seems a dubious basis (that the project might be viable with more optimistic assumptions) on which to champion a project as the greatest thing ever. I’m not trying to argue with you about WEBs or whatever, but please accept that making a few unquantified and unreferenced points that the a previous evaluation may have been conservative does not rebut the argument that the economic case is unproven.

            Of course at least your argument addresses the issue directly. The post above contains the even more common argument that because the project has benefits it is therefore worthwhile. One would hope spending billions on a project would result in gross benefits. How could a rail tunnel under a CBD not have benefits? Among people who are serious about thinking about this project that is a self evident thing to say but it is not an argument for the worthiness of the project.

          • Peter Nunns

            “Your argument, which I have seen in various forms again and again, is that the lack of an economic case doesn’t matter because the assessments were conservative in terms of benefits.”

            You misconstrue the process. An economic case exists, and a BCR has been calculated. The decision to progress the project has taken into account both the number and the limitations of the number.

            “the project might be viable with more optimistic assumptions”

            Not “optimistic” assumptions – realistic assumptions. How on earth does it make sense to assume that city centre parking prices will be lower in 2041 than they are today? (For those assumptions, see e.g. page 40-41 of the traffic modelling report in the AWHC business case.) Parking prices have considerable impacts on mode choice and hence transport user benefits. The use of more realistic parking price assumptions in AC/AT’s original study was a large part of the reason why it estimated a BCR in the range of 1.1-2.3.

            “The post above contains the even more common argument that because the project has benefits it is therefore worthwhile.”

            Nobody has said this. You are arguing against a straw-man.

          • Matthew W

            “. How on earth does it make sense to assume that city centre parking prices will be lower in 2041 than they are today? ”

            By that time we may well have driverless cars in which case the market for parking in the CBD may have basically disappeared altogether.

            “You misconstrue the process. An economic case exists, and a BCR has been calculated. The decision to progress the project has taken into account both the number and the limitations of the number.”

            I dont think so. I think the decision to progress the project is because politics.

          • ‘By that time we may well have driverless cars in which case the market for parking in the CBD may have basically disappeared altogether.’

            Wow; now this is pure fantasyland by the man claiming we should only measure solid and visible things. So many assumptions in that sentence. You might as well say; ‘don’t build anything cos stuff might happen’.

            Why are you so focussed on this project with your doubts about capital spending? Every ‘concern’ you’ve raised could be leaved ay any investment at all. Especially the RoNS. It is clear you are relying on your personal dislike of this project, while trying to attack my arguments as being similarly based, but with the opposite conclusion.

            Do you work in the Auckland MoT office perchance? Some there have dedicated their careers to these sort of arguments against the CRL, the very people that stack the assumptions in the models with absurdities like falling parking costs in a growing and space constrained city.

          • Matthew W

            Hi Patrick,

            I would argue against the RONs too if I could find anyone to argue with (the ones with negative net benefits anyway). I guess I could write to the politicians but they are perpetual liars and spinners so what would the point be – I get my fill of being patronised by them via the media. You guys on the other hand argue (often very well and I more ofthen than not agree with you) from an objective, logical point of view. I dont have a dog in this fight so to speak, just an interested observer.

          • Matthew W

            Hi Stu,

            You obvioulsy know more about this than me, I was just paraphrasing Eric Crampton on the subject:


            “And, fortunately, while the economic literature points strongly to the benefits of urban agglomeration and of having lots of people in a city, it is far from clear that those benefits require having a single dense centre. “

          • Interesting quote from Crampton. Pretty sure Glaeser would disagree, well depending on what is meant. Certainly cities can have centres [plural] of density, or solid and consistent levels of density that don’t peak too much. But even those that do, and LA comes to mind, still have an area that is denser than the rest that gets called the centre. Houston the poster boy for sprawl even has a big ‘ol centre that’s denser than everywhere else.

            Crampton says this earlier in the piece on Christchurch: ‘Sure, downtown was more spread out than suited some people’s aesthetic sensibilities but it generally worked’ which I think is both true but also missing the point; generally working is not thriving. The evidence does show cities with dense cores cities flourishing way better than blah spread out ones like Christchurch was and is; a doughnut city with a weak centre.

          • Matthew W


            If you want me to take your points seriously I shall, however I note that the authors of these reports would have done things for a reason and picking holes in analysis is an easy thing to do, and I dont have a lot of information to go on based on your bullet points:

            * Assumptions about parking prices in the city centre – they assumed that average prices in 2041 would be lower than today’s earlybird prices. Yikes.
            Is this the average cost of parking faced by users for all carparks? In which case a number will be at 0 which will bring down the average. Also driverless cars may have a big effect here and are anticipated to be coming on stream by 2041.

            * Annualisation of PT benefits – the approach used tends to over-value benefits from peak-only services and under-value benefits from all-day services provided by CRL.
            I dont know what approach was used so can’t comment. I’d need a better explanation

            * Interactions between ART (regional traffic model) and SATURN (CBD traffic model) – results appear to have been combined in such a way as to exclude a significant chunk of decongestion benefits.
            Again I would need more information on that one. However decongestion benefits are a bit of a claytons benefit – if we dont want congestion we only have to price road use properly. If we dont have road pricing the question is why. Until we answer the why it is difficult to justify including these benefits. If people dont want road pricing then perhaps it follows that they don’t care that much about congestion and therefore dont experience the purported benefits from it being reduced.

            * Air quality benefits resulting from fewer buses running on busy streets have not been calculated, although an evaluation procedure exists. In my experience, these benefits can be surprisingly large.
            Buses may well be all electric or hybrid by 2041. If we want better air quality (or at least to internalise the externality) we would introduce pigouvean taxes on emissions. Not hard particularly for transport emmisions. Claiming benefits from a very indirect method of subsidising rail seems like a long bow. Similarly to the comment on decongestion above, if we dont want to internalise this externality (when we can), is the externality actually as large as claimed and can we rely on counting it?

          • Matthew – You’ve frequently raised the CCFAS and effectively said it’s the form of truth we should follow. Peter has said it is very conservative and you don’t seem to believe that. Perhaps you might believe the Ministry of Transport who themselves said the modelling was wrong and likely to be underestimating PT trips

            From this post where I went through docs obtained from the ministry via an OIA

            There’s also this which looked at how the modelling worked

          • Matthew W

            Thanks Matt,

            Yes certainly some gross assumptions with that model! The main issue I have with it is they are assuming people will get crowded off buses. Why cant they put on more buses? Because of a lack of road space and congestion. Why is there congestion and a lack of road space for more buses? Because they are assuming road space will be unpriced. So the whole model is predicated on an irrational policy for allocating road space.

            Anyway my main point is the economic case for the CRL isnt yet settled which I think we can agree on. The government have promised a business case in 2017 so I guess we will have to wait until then. Hopefully they do not assume irrationally unpriced road space will persist until 2041.

          • MW, hoping for road pricing in a world where the government have formally killed attempts by the Council to consider it. But also any road pricing would also impact very positively on the case for the CRL. There will still be buses post CRL, and it is absolutely clear, including from the CCFAS, that there is no workable or cheaper bus only solution. Also as a fan of road pricing then you will understand that as close as we have got to outside of fuel tax is parking price, yet the impacts of market pricing of this resource you want to wish away with fantasy future technology, that is by no means certain to have that specific outcome at all. Let’s be clear; the market pricing of finite urban road and parking supply in a growing city can only improve the case for the one project that uses neither but delivers a huge access solution.

            So still turning yourself into funny shapes to try to cast doubt on an increasingly more solid looking project.

            But we agree, every project needs good sceptical critics, if we are a little testy it’s because this is the only one that gets them. Have you considered the Additional Harbour Crossing?

            We’ed be happy to run an analysis of the value of that from a capex-sceptic like yourself? It has a lot of ill-informed fans, so is ripe for myth busting…

          • Additionally:

            ‘The main issue I have with it is they are assuming people will get crowded off buses. Why cant they put on more buses?’

            Assuming people will get off crowded buses when offered a more comfortable, quicker, direct to the city, frequent, clean and modern alternative at the same price is pretty reasonable assumption really, don’t you think? Oh and it isn’t even an assumption, it’s an observation here and now in Auckland; rail use growing at over 20% bus use outside of the northern busway flat [crowded buses, slower buses]. People are rational actors; they know a better option when they see one.

            More buses at a certain point congest other buses, there comes a time, ie Dominion Rd, where without a full busway and stations, more buses can not be made to work [of course fulltime bus lanes would be a good start there now].

          • Peter Nunns

            Hi Matthew – thanks very much for coming back and engaging with the details! We appreciate your critical perspective on things.

            I think the whole parking pricing thing bears a post on its own. (Which I will hopefully have time to write before the end of the year.) According to a table on page 7 of this report, priced public parking accounts for roughly half of the city centre parking supply. Current earlybird prices are in the range of $20. So even if the rest of the parking supply wasn’t priced, that would imply an average price of $10. As I said, that’s higher than the AWHC modelling has for 2041.

            Annualisation is the process by which modelled estimates for a single AM peak period are factored up to annual outcomes. If you look at the CCFAS report, you’ll see that a single annualisation factor has been applied to all public transport model results, even though some PT services only run in the peak while others run all day. This will tend to underestimate the benefits of all-day service patterns enabled by the New Network and CRL.

            Your last two points seem to both take the same perspective – that it’s only ever worth implementing the first-best policy. While I agree that congestion pricing is probably a good idea, I also think that second-best policies are worth pursuing. To that end, are you aware that parking prices can be set to simulate a congestion price? My suspicion is that, as Patrick said, a congestion price or higher parking charges would tend to strengthen demand for PT – meaning that any reductions in decongestion benefits from those policies would be balanced by an increase in PT user benefits.

            On the emissions, I’d agree that new vehicle technologies will play a role, but we’re also going to have a substantial proportion of diesel buses in the fleet through the 2040s. Buses tend to be used for 15-20 years, with a bit of reburbishment. I’d be a bit surprised if electric buses were price-competitive with diesel buses before the mid-late 2020s.

          • Matthew – the preferred CCFAS option was the CRL and enhanced bus infrastructure. It wasn’t made entirely clear but that included a full four lane busway along Wellesley St and lots of bus ramps and lanes etc. In this case it isn’t a car congestion issue but a spatial one, there simply isn’t the space for all the buses that would be needed. The outcome from the CCFAS was that more investigation needed to be done to find how to get better capacity on the street level.

            That is what has lead to CCFAS2 from which light rail came about. Other than the outcome of LRT we haven’t seen any report from this though.

            Also I understand some of the new business case may have already been done and is sitting with the minister

          • Matthew W

            Thanks for the responses Matt, Peter and Patrick,

            I think you have convinced me the CBA for the CRL is likely to have been fairly conservative. The CCFAS modelling frankly sounds bizarre. I guess I would like to see a business case that has some consensus that gives a number showing it is a viable project. Sounds like that may be coming soon.

          • Peter Nunns

            Thanks for the debate Matthew! I think it’s really important to challenge these things so I’m glad you’re doing so.

      • Mike

        No, Matthew W – you’re thinking of the RONS projects.

      • mfwic

        No Matthew not gut feeling. I am saying you model everything you can model and look at the result you get and understand it represents everything you were able to include or able to enumerate. But there will always be “off-model effects” or those things that you know from theory or experience with other projects are likely to occur but are not in the model. The biggest problem is when a project can cause a “fundamental change” to use the jargon. Patrick has talked elsewhere about projects that were transformational. (He is a bit negative about some). But if you are close with the model result and you think the off-model effects are worth it you go with the project. And please remember a big project with a marginal B/C can still be a better choice than a whole heap of small projects that each has a high B/C but are not going to work together well. I have done heaps of small projects that just end up moving a problem but they met the cut-off.

    • I agree mfwic, few projects are transformational, it’s a big ask. Most can only hope to nudge the status quo along a bit. And many can’t even manage that.

      Looking back, I would say the Harbour Bridge was transformational, clearly for the Shore, the CMJ was certainly and violently transformational for its host areas, and the entire relationship of the City to its inner suburbs. It certainly destroyed value of our premier shopping strip, killed Newton etc… made living further out and moving by car significantly better, but greatly stressed the whole viability of the city centre, but was great for suburbanites. Ripping out the trams was clearly a big transformation. Earlier; Tamaki Drive?, Grafton Bridge maybe not; I don’t have such a good handle on what life was like either side of that?

      If I’m right about the CRL qualifying to join this list then the interesting thing is that it would be the first time since the war that we’ed found a non-destructive transformational transport project. Those others all came at huge cost, especially to place. I think this aspect is one of a number of misunderstood values that this mostly misunderstood project has [being misunderstood, even by its promoters, is a classic sign of likely transformation, IMO].

    • Harriet

      Also I heard that the big problems with old school NZ transport BCR’s is that they;

      A) Don’t take into account increases in Agglomeration;
      B) Don’t take into account potential Transit Orientated development;
      C) Only count the benefits of a project for 30 years, when anybody knows that the CRL will provide benefits for more than 30 years, the London Underground original line opened in 1863 and the tunnel while of course been extended & upgraded over the years is still in use.

      When you take into account these your BCR improves massively

      • mfwic

        I used to think about how the discount rate means we only value benefits for a short period while walking to work on a road the Romans built. I think the answer to that is we have limited money so we have to focus on things that give immediate benefit. The people who built the great cathedrals took a different view, they knew they would never see the job finished but didnt care about that. The difference is the social structure. They had a few people paying for everything so monuments gave those people something to leave behind. We tax everyone and many people struggle and most of us could think of better things to do with the money ourselves so we tax low and spend on public essentials.

        • MFD

          There is more to it than that. The further in the future the benefits are the less likely it is the that the beneficiaries were the ones paying the cost.
          Everybody dies.

          • That is precisely why it makes sense to borrow for long term capital investments.

            This attitude, that we should only invest for the short term, also explains why Auckland has such a shitty built environment. This was never the attitude of the Victorians and Edwardians, who believed in leaving a better city then the one they inherited, who built so much of the infrastructure that is still admired today in first world cities. We can do better than this version of institutionalised selfishness.

        • Interesting argument. Does tend to support wealth taxes or other methods targetting our most wealthy to fund long-term investments they are no longer morally committed to as part of their citizenship.

  • Steve Withers

    Hands up everyone who will probably be dead (or almost) in 2041. I’ll be 83. 🙁

  • RJ

    Can I suggest that as the population grows we change some of thise rapid bus routes into train routes? I don’t think it will manage in the future if buses are the only ones. in the far future, some here might have already passed but having trains would be able to accomodate more especially if we get an influx of people that we reach megapolis population say the same as London or Tokyo etc.

  • AK

    Where is the rail ROW designation through Owairaka and Mt Roskill? Just down Mt Albert Rd?

  • Harriet

    1. Extend Rail to Puke adding two stations in between by 2017.
    2. Complete all four LRT routes by 2020.
    3. Complete CRL 2020 (If Sisi can build an Suez bypass in one year I can build the CRL in 3.
    4. Southwest-Airport Line (Via Otahuhu I prefer the running pattens due to just running the crosstown to the Airport giving by a CBD and West easy access to Airport) by 2021
    5. Northern Metro (Silverdale – CBD [Over the bridge I have no time for extra tunnels I need the spare cash :D] – Westgate – Constellation) – Possible Stage Two would be Huipai – Westgate – CBD – Takapuna – 2022
    6. Do a business case on Eastern Suburbs comparing three options Heavy rail spuring GI-Botany,Manakau then Airport & LRT Panmure-Pakaranga-Howick, just the Heavy Rail route, or both routes LRT then do by 2023

    • Harriet

      How do I fund simple (At least most of it)
      Third Crossing – 6 Billion
      Puhoi-Walkworth (I wont call it Wellsford till it actually goes their :D)
      East-West Link – 1.5 Billion
      Cut other motorway upgrades for Auckland – Conservative 1B
      Cut upgrades to Mills Road, GSR and other roads like this – Conservative 1B
      Redevelop Golf Courses into Urban Villages (Medium to High Density PPP developments I.e AC & Ockham partnership I suggest) 1.2B not including the rates these places would pay
      AMETI – Don’t really need the rest if we do 6 – 1B
      TOD around all new stations – Unknown

      So I have raised $12.9 Billion for my projects and I haven’t even raised future spending only shifted it & sold a non performing asset. Add a Congestion Charge that starts low then increases on each area as they get their rapid transport & a regional fuel tax then you are looking very good just shows how inefficient our transport budget is.

    • Ted F

      I like the whole post Harriet. Hope that you’re able to get involved.
      Some here worry about being dead by the time these come to fruition,but don’t worry we are really thinking about what is good for our Grandchildren.

    • Yeah much more realistic timeframes in terms of requirement, 2041 is ridiculous, that’s 25+ years away, by then we will likely be needing even more infrastructure than shown.

  • bbqroast


    What about a “Crescent Line” connecting Howick to Albany? The idea is to merge a number of the lines shown here, into one continuous route.

    – Urban busway Howick/Pakurange – Panmure – Onehunga

    – Busway following motorway Onehunga – Tunnel

    Then I’m actually in favour of a urban busway shadowing the tunnel using the road reserve. With a transfer station on to the Western line by moving Avondale Station a little East further into the road reserve.

    – Joining the motorway busway at the northern Tunnel Portal.

    – Joining the western busway through Massey, out to Westgate, then turning North all the way to Albany to meet the Northern Busway.

    If you could achieve an average speed of 50-60 km/h (with a express bus, which should be possible with the motorway busways) then you could easily beat car traffic on the Pakuranga – Albany route, and be competitive at other times. At the same time you connect a ton of otherwise poorly served regions (North of the Western line, the wedge between the Western and Southern Lines, and the Howick region).

Leave a Reply