There are not many people these days that would say the Northern Busway has been a bad investment with most people wanting to see it extended, both at the CBD end and also to Albany and eventually Silverdale in the North. The busway has become so popular that now about 30% of people crossing the bridge towards the city in the morning peak are doing so on a bus which is pretty impressive and for one thing has already helped to delay the need for an extremely expensive additional crossing. One question that I have been keen to get an answer to though is just how successful it has been and the NZTA as part of my OIA have kindly provided me with the the original business case for the project (1.1 MB) which makes for some interesting reading.

The business case was completed in 2004 and investigated three options, the busway, the busway but also with T3 vehicles and the busway with T3 vehicles initially but then changing to T2 from 2015 once other projects like VPT had been completed. It was estimated to cost $214.6 million although they note that the North Shore City council planned to spend some more money to improve the stations which would help build patronage. Here is a summary of the costs and benefits of each option:

The cost of the busway ended up being about $294m of which $210m was for the busway itself and $84m was for the stations. As you can see though, comparing the costs to the benefits the option for the busway actually came out with the lowest score with a benefit cost ratio (BCR) of only 1.2 which was much lower than I had expected. So the next question I had was what figures went into the business case to arrive at these benefits, one factor that would make quite a difference is the discount rate used. Our former admin has done a number of posts that explain discount rates and the impact they have on investments we make. The standard NZTA rate currently used is 8% however for the busway a figure of 10% was used (I assume that was the standard rate back then). Reviewing the business case using current NZTA methodology would most likely result in higher BCR numbers across all three scenarios.

The biggest impacts though come from the number of people using the busway as well as the benefits that accrue to other road users who aren’t using the busway. The results of the modelling for two years are included in the paper and these are 2011 and 2021 and this is where we can see how many people were expected to use the busway.

So they suggested that in 2011 we would have about 1630 people using the busway to head to town in the AM peak and in 2021 that would rise to 2174. Of course these are likely to have been fairly conservative numbers but how do they compare to what we actually had. We don’t have direct numbers for the busway but we do have numbers for the harbour bridge which would also include buses from Onewa Rd, Auckland Transports annual report for 2011 tells us that over 8600 people crossed the harbour bridge by public transport, while the report we saw last week shows that at the height of the peak about 5000 people cross the harbour by bus. Of course a decent chunk of these numbers will come from Onewa Rd but the majority have likely travelled along the busway meaning that we are already far exceeding the numbers forecast above.

What this means is that the busway has already delivered significantly greater than expected and will continue to do so. If we were to re-evaluate the the project with what we have learned so far the BCR would likely be considerably higher. Interestingly the NZTA might be doing just that right now, in responding to my request they advised me that they are just finalising a post implementation review of the Busway which should be out very soon, possibly by the end of the month. They have said they will send me a copy so I am looking forward to seeing what comes out.

I think there are also lessons we can learn from this when thinking about other PT projects, the main one is that it seems that our models consistently under estimate the demand for high quality PT services like the busway. I think it would be a good idea for politicians to keep that in mind when thinking about the CRL today as that is another project where I feel the patronage projections underestimate what will really happen.

Share this


  1. Without getting into the benefits or otherwise of PT, doesn’t this discussion mean that the BCR of some of the RONS could be underestimated? It’s interesting that the primary benefit of the buslane is assessed as travel time — this is hard to quantify. If travel time savings are really that good (and worth paying for), then there is an argument that we should pay car owners who become cyclists and motorcyclists as it gets them off the road (or reduces the road space)

    Arguing on BCR’s is a slippery slope… economic effects are even harder to measure (see any stadium built by a local council, ever)

    1. Sorry Pierre I disagree.

      We are here because we believe more balanced transport infrastructure development is better for NZ.

      If we chicken out of seeing some actual BCRs of past roading projects, then critics would say we have a blinkered vision, and they would be correct.

      I for one would be most interested in seeing some post-construction analysis of actual BCRs of projects like the Busway as above, and the original CMJ (including but not limited to loss of land use, CBD fringe residences and the decimation of K Road retail as costs) and its recent upgrade, as well as the BCR of the Ian McKinnon Dr / Dominion Rd flyover.

      If that weakens our case, then so be it, but I am confident it will not.

      1. I agree about the balanced transport infrastructure. I guess I’m wanting to see a proper comparison – the Puhoi-Wellsford and other RONS are attacked for having low BCR, but they may not be too different to that calculated for the busway! Hence the observation.

        In terms of assessing whether projects actually deliver their calculated BCR, that is another really good question. There are a lot of ‘interesting’ local projects that cost a lot and (I think) don’t really deliver many benefits.

        1. The tendency (in NZ) has been that BCRs for road projects are (generally) overstated because of cost-overruns and over optimistic assumptions about traffic growth. From what I have seen the exact opposite is true for public transport.

        2. I’d like to see the assessment as the irony is that the Busway had cost overruns but appeared to underpredict growth. The assumptions about traffic growth also feed into the time saving cost estimation for the PT (and this reduces if traffic volumes don’t grow), so this is a bit of a double edged sword.

          What this emphasises is that modelling this is very difficult.

        3. If the modelling is so hard to get right (which if obviously is) I guess the question becomes do we continue to use it? If not then how do we decide on what projects to build and what not to as leaving it up to the politicians saying “We are building X because I think it is good” leaves us open to all sorts of potential issues.

        4. I believe the problem relates to how we use the modelling. At the moment it seems we abdicate any responsibility for thinking to the modelling process. We say “hey that must be a good project because the transport modelling outputs say so.

          It would be much better to use the modelling as simply telling us what the outputs are from different scenarios of assumptions. There would never be a single BCR but rather about 10 of them based on the different mix of assumptions. Then we actually think about which assumptions we believe.

  2. It is clear that NZTA and the MoT are entirely populated by traffic engineers, who, like the current government, believe deep to their bones that no one really wants to use any form of movement except cars. This deeply held yet unexamined assumption [like the one that claims that every new road ‘creates’ economic growth] means that they consistently underestimate demand for public transit, refuse to see declines in driving despite the numbers being right in front of them, and have to ignore the negative externalities of their programmes.

    Perhaps despite the countless billions having been spent and still being spent on the auto-dependant model the fact that demand for alternatives is growing is like a nagging reminder that this brave dream from the 1950s hasn’t, doesn’t work….?

    The place we have to help these troubled professionals to is to help them to recognise that the best hope for their beloved motorway system’s continued functionality is to relieve the pressure on it by building real joined up and quality alternatives to it. It isn’t about abandoning their life’s work, but complimenting it with a supporting system. The One System idea was only ever viable in a smaller city and was really the vain hope of cheapskate politicians and not real transport professionals anyhow. This political view still has sway in the current government swayed by the self-delusion that auto-dependency is self funding.

    1. When people discuss PT here I think it would be safe to observe it is normally Auckland they have in mind. Bearing that in mind, it seems to me the two statements “…no one really wants to use any form of movement except cars…” and “…every new road ‘creates’ economic growth…” may even actually still be true in the New Zealand that exists outside Auckland. I think it is important when considering the views of non-Auckland based agencies the fact that they usually consider Auckland to have the same transport issues as, say, Tauranga but suitably scaled up. If you don’t live in Auckland it’s size is hard to grasp when in this country the next largest “city” barely exceeds a paltry 360,000 people. I think it isn’t an unfair observation to make that outside Auckland people seldom grasp the essential “otherness” that size and scale imposes between planning for a fast growing city of 1.5 million vs. a town (by any other countries estimation) with a (largely static) population of 360,000. It would be interesting, for example, to see what the different spending priorities would be if the portion of the NZTA budget being spent in Auckland was transferred to Auckland Transport. To me, this raises intersting questions around some larger issues, like for example the appropriateness of keeping the capital (and therefore key decision making) in Wellington when Auckland is racing to a population of two million plus.

      1. Agreed.
        I’ve often thought the Public Transport problems Auckland faces are not so much due to NZTA/etc being ideologically opposed to PT, but because they’re all based in the little country village called Wellington have never grasped how different Auckland is to anywhere else in NZ.
        In most places in NZ inc Christchurch cars IMHO are the most efficient way to get around, but that all changes when you get to a “real” city like Auckland. All of a sudden you can’t just “build roads” because there is a whole level of complexity that goes with it which you don’t get anywhere else in NZ.

        I do believe that a lot of Aucklands transport problems can be traced back to Parliament being in Wellington not Auckland.

        1. …this would be the same Wellington where 18% of commuters use public transport (as opposed to 8% of Aucklanders)? And the same Parliament comprised of about 1/3 of MPs who LIVE in Auckland?

        2. I always thought the reason for so much historical spend on PT in Wellington, and therefore greater use, was because Wellington is essentially connected to the outside world through a single motorway. Imagine evryone trying to come in through that by car every morning.

        3. Glen K – Agree completely. As someone who worked for years in the Wellington transport community (MoT, Transit New Zealand as-was and then Tranzrail), can I add:

          * Wellington has some specific transport challenges of its own, both the region getting into the CBD and (separately) the city getting into the CBD. Far more so than Christchurch, which has the same population and has had the same-sized CBD.

          * The high net use rates for PT in the region have a lot to do with a very large share of the region’s jobs being in the CBD – far higher than in Auckland. As a result, the rate of bus use for Wellington city proper (~100 trips/person/year) is far higher than for the Auckland isthmus (currently around 55 trips/person/year).

          * It’s changing now, but for most of the post-war period Auckland was hampered by a Balkanised local council structure which basically had the peripheral councils in the region, and even more the separate Isthmus “rotten boroughs”, pulling against the centre. Hence the dispersal of jobs out of the downtown, which is fatal for public transport demand.

          * In its time, the Regional Council did not work all that well, and for the same factors of “balkanised” politics. This was not really an issue in Wellington.

          * The other thing that I found frustrating when I worked as a budgets officer for Transit, was that despite all the rhetoric from local councillors about wanting public transport, when it came to them fronting up with their own money, there was a near-total lack of action. OK, things are changing now, but at the time the gap between rhetoric and actions was significant.

      2. It’s not uncommon for the capital to not be the largest city, with Washington D.C. being a prime example. In fact, in the English-speaking west the UK is the outlier by having London as largest city and capital. Australian, Canada, us, and the US all have the capital city a fair way down the “largest city” chart.
        The problem for Auckland arises from our exceptionally flat government structure, where the national government is involved with decisions that would ordinarily be left to a state government under the federal model existing in those countries. Decisions on a local transport project (which the CRL is) would normally be filtered through a state-level government with its own powers of taxation before going to central government for a top-up. What we have here, though, is a central government that has absolute control over the ways in which lower levels of government may raise money in addition to having absolute control over how they use that money.

        1. Yeah, but part of the reason for the “flat” structure is because we only have the population in total of one of the smaller American states, and it is less than NSW or Victoria. Scotland’s devolved Government has a lot of freedom in its land and sea transport arrangements, but it has a population of 5m. That’s a good question for debate – how much government can we reasonably devolve in NZ to our sub-national Governments?

        2. NZ is more like an Australian State than the whole country. Like each Au state NZ has one dominant primary city, but unlike them that city is not the capital, does not house the institutions of state gov, nor the cultural ones. Or at least not the ones funded by the whole state equivalent.

          And because we currently have the party of the provinces in power (led by two nominal Aucklanders; Key & Joyce, actually a Hawaiian wannabe tax exile and a semi-rural life style blocker and small town boy) the true nature of the city is ignored and in fact attacked by the governing party and it’s cowering apparatchiks. And this wilful ignorance of the needs of a third of he population is all the more possible because the institutions of state are domiciled in a charming little fishing village at the other end of the North Island.

          The Super City is, in fact, a challenge to this state of affairs. As we shall see as time and demographics unfold.

        3. However all the Australian states have two houses of Parliament (and Victoria is close in size to where NZ is projected to be within the next 30 years), so I’m not buying any suggestion that we could not support two layers of real legislative authority given that they then have TLAs below their state governments.

          If we’re going to talk about federal model, the majority (33) of US states don’t have the most-populous city as the state capital, and that’s states of all population (and economy) sizes, with most also giving significant devolved powers to their sub state-level government bodies.

          NZ is actually in a pretty unique situation globally with how we’re governed. Unitary states are not uncommon, nor are unicameral parliaments (though they’re much, much less common). It is, however, somewhat less common to find a country that is a unitary state with a unicameral parliament and a head of state whose powers are almost entirely ceremonial. It means that we’re particularly vulnerable to the vagaries of poor-quality governance.

        4. Matt, QLD is unicameral, although not exactly an example to follow after Sir Joh. The territories, NT & ACT are also unicameral but they can be overruled by the Feds.

        5. SimonL, yes but there are still the two houses of Federal gov above that. This is Matt’s point, we have few checks and balances, and as this gov now seeks to further limit the already minor power of local gov., this can only get worse.

          The interesting question raised by the creation of the new big Auckland Council is this if it were to get more independence particularly to raise revenue, particularly for transport infrastructure, would this be a good thing? Looking at this gov. and this council I would say; of course. But with the miserable C&R ticket always poised to take over and shut everything down, sell everything off, and fund retrograde roads, absolutely no culture and certainly nothing intergenerational or transformative there is cause for caution….

  3. Not only underestimating the demand for quality public transport but also, it would seem from Matt L’s recent post on funding for the new EMUs, systemically overestimating the construction costs of rail infrastructure.

  4. Thinking about this a little more, I wonder if there is a study to be done on PT vs roading projects and the effectiveness of the business cases. This could allow systematic statements to be made, for instance new roading construction estimates and the measurement of induced demand, the patronage estimates for PT projects, the time savings that do (or don’t) manifest and so on.

    This could possibly improve our assessment, but the difficulty for planners is always how do you compare the benefits of (for instance) a pedestrian crossing with the disbenefits of slowing cars down and throttling roads.

    What is readily apparent is that the car centred world we live in could be drastically improved, particularly in Auckland CBD

    1. I agree that we could be underestimating the results for roading projects but we are more likely to be overestimating them due to traffic volumes in recent years not rising in line with expectations.

      I also agree that the economic analysis is only one part of the equation but at the moment it is used to fight projects that the government doesn’t like while is ignored for projects it does where a higher priority is placed on “Stragetic Fit” which is shorthand for does it match what the government says it wants regardless of whether that investment makes sense on any other measure.

      A comparison of projects before and after should be a mandatory requirement. I’m please that they are doing an assessment of the busway but it would be good for that to happen for other projects as well.

      1. Might be an interesting thesis for some University student – go back and revisit old BCR costings and compare them to reality.
        Nobody else seems to be interested in doing it for obvious reasons.

  5. Good post, good comments. Reading this from a QR train heading to a meeting on the Gold Coast, so am in need of some entertainment.

  6. the BCR system in 2004 valued the time of bus passengers (poor people? or obviously “less time sensitive” because they didn’t use cars) much less than it does now, possibly faulty memory says the time value for PT was around half that of a car driver, note how the travel time benefits jump from $143m to $236m with the T2 option

    the HOV addition was developed at the time by BECAs as a “fill up spare capacity” ploy to boost the benefit side of the equation, which also (opinion here, they may think differently) helped Transit accept the busway project as it better fit their mindset if it had cars on it

    admittedly, the number of HOVs allowed onto the busway was always viewed as a variable, capped and governed by the number of buses, so that bus operations were paramount and the HOVs were not to cause delay to bus traffic

    1. I have often wondered if there was a business case to allow minivans with >6 or 8 pax to use the busway (or bus lanes). They are essentially ‘mini buses’ (or mini vans, if you will :P). If we are focussed on moving people quickly and efficiently, then this sort of operation seems to make sense.

      HOV2+ I find insane. HOV4+ I have time for. HOV6+ or 8+ strikes me as a no brainer to encourage car pooling. In fact, you could even offer discounted/free parking for HOV8+. Would love to see the cost estimation per passenger time saving for that.

      1. the USA has extensive experience in car pooling, often for quite long distances and many of their freeways have “diamond” lanes, see

        when I worked on the early days of busway development at North Shore City, we piggybacked onto a tour organised by Queensland Transport for their busway team, first call was Houston which operates tidal flow segregated bus/hov lanes down the centre of their freeways, they started out with vanpools, but the uptake was poor so the numbers started coming down to HOV 4+, then HOV 2+, despite the time savings there was significant resistance to HOV formation until they introduced a guaranteed ride home scheme

        HOVs aren’t the panacaeia they appear to be! it’s hard to get people to join up and possibly the anonymity of the bus is an easier sell than sharing a car with a relative stranger

    2. That is outrageous, Steve. Really? If ever anyone doubted the existence of a pro-car bais in our transport institutions there’s your proof. It is also a case of circular reasoning. No need to improve the effectiveness of public transport because we have decided that the value of anyone using it is half that of someone driving, why, because it is slow so they obviously not going anywhere important. Why is it slow, because we won’t invest in it. Why, because the people using it are of little value. Why, because its slow. Why.?….. And so on. Where is Franz Kafka when you need him? Phaaaaark..

      The more you look a the details of these agencies’ processes the more you see how they just reflect the bias, the assumptions and clearly the narrowness of those in charge.

      1. I don’t think that’s actually the reasoning behind the differences in travel time between modes. The basic concept is that time spent travelling is lost productive time for doing something else (work at your desk, shopping, etc). If you’re driving, then generally you can’t do much more than that task (although a good many have a crack at talking to others on the phone as well…). If you’re sitting on a bus/train however you can be quite productive, getting through some reading, pottering away on a laptop/smartphone, etc. So the COST to you (or the nation) of extra travel time when on PT is not as bad, hence the $/hr value is lower. Similarly, someone biking, for example, is also getting some exercise (saves them spending additional time at the gym), so their travel time also has a lower value.

        1. No, it is correct that PT users had their travel time savings valued less than if they were driving. From memory I think this has been corrected somewhat now.

          Travel time savings are a somewhat bogus argument anyway. Commuters travel on their own time, not their employers. This recent survey found that 40% actually enjoyed their time commuting. Less than 10% said they would use any travel time savings to work more. Furthermore only 3% specified that zero minutes would be the ideal commute time, contradicting the BCR formula which states that the more travel time savings the better. Most people like having a book end to the day.

          In the end I think the true value of transport projects is the increased choice and accessibility provided. More focus should be placed on absolute capacity transport solution – at the moment this carries no weighting in the current BCR analysis as far as I’m aware.

  7. Is the Britomart business case available? It would be interesting to compare the pre-construction passenger estimates and costs to that post-construction. This may have good relevance for the consideration of the CRL…

      1. I don’t think the NZTA would have a copy as they or their predecessors’ weren’t involved due to it being an Auckland project. Would be best to ask AT or the council but they may try to charge for it which they are allowed to do if the request is a bit tricky (the NZTA tried it on for this business case saying it was in the archives somewhere until I pointed out that they would have to have had used it for the review they are conducting).

  8. I’m a bit intrigued that the accident benefits for all three options are exactly the same. If there was a likely downside to allowing T3 or T2 traffic in with buses on the busway, it would be the greater chance of conflicts – perhaps it’s balanced out by removing the T2/T3s from the general traffic stream?

    It would also be interesting to see a breakdown of the travel time benefits for the buses vs the T2/T3 traffic. I would imagine that as more private vehs are introduced into the mix, the relative savings for buses decrease (from having to mingle with more traffic on the busway). This might also have a spin-off effect of discouraging some people from using the bus service – why would you switch to a busway service if it’s perceived as not much different than a standard T2 motorway lane? Again, it’s not clear from the business case figures above whether the expected patronage numbers for each option were the same.

  9. 2 things I would like to see
    1) Timings -how long from Albany to Akoranaga then to Britomart on bus at peak (could get of GPS and average it so data has been captured )

    2) Actual numbers for Busway use. Have had about 5 people counting the numbers at each station earlier this year so AT has data- cannot get it brokendown due privacy but overall data should be avaialable . Have to remember Bridge represents Onewa, Busway + Takapuna Buses

    By the way for 1 on Tuesday May 15th the 881 bus
    Left Albany 0708. Full by Smales (Left about 5 people behind at Akoranga) no one off at Britomart.
    Got to 2 Symonds St (Auckland University)at 0740. Had 2 red lights on Busway (due road repairs) and torrential rain downtown part of trip.

Leave a Reply

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