In a recent post on the failures of toll roads, I argued that marginal analysis can be a powerful tool for understanding transport markets:

Economists understand the importance of marginal analysis when making decisions about what to build and how to charge for it. Businesses typically make pricing and production decisions “on the margin”. In other words, they look around for the next potential customer and ask: “Can I produce one additional unit and sell it to that person for a profit?” If the answer is yes, they produce it; if it’s no, they don’t as it would reduce their overall profits.

While transport infrastructure isn’t (and probably can’t be) run as a business, transport agencies could stand to take a few lessons from businesses. In particular, they should be asking: What does the next transport user want, and how can we best serve them?

Businesses that don’t ask this question don’t grow, because they don’t look for new customers. Over time, they will fail as their existing customers age out or lose interest. I’d argue that the same can be true for cities as well – cities that fail to respond to the needs and desires of the next potential resident (or business) will eventually decline. This is eventually bad for existing residents, as the tax base shrivels up and businesses close.

With that in mind, what does the data tell us about the transport desires of the next Aucklander?

Stu Donovan and Matt L have already taken a good look at the ongoing decline in vehicle kilometres travelled in Auckland (and lots of other places) – essentially, people are no longer willing to drive longer and longer distances every year. Rather than covering that issue again, I’d like to take a look at Aucklanders’ choices about travel mode using data from the 2006 and 2013 Censuses.

I used Census data to take a look at changes in changes in mode choice for commute trips in recent years. The analysis is summarised in the chart below. While growth in car use has been pretty anemic – a mere 2.3% increase – demand on all other modes is growing like crazy. There have been double-digit increases in bus trips (up 18.8%), train trips (up an astonishing 67.3%), and bike trips (up 26.4%). Ferry and walking trips have also done extremely well.

Census Auckland marginal transport user analysis

While most Aucklanders still commute in cars, alternative modes have captured the overwhelming majority of the growth in recent years. 61% of new commute journeys between 2006 and 2013 were made on public transport, on the bike, or on foot. (Broken down as follows: 22% of new demand was on the bus, 16% on trains, 7% on ferry/other, 6% on the bike, and 10% walking or jogging.)

This is an amazingly strong market signal. It’s remarkable that people are eagerly taking up alternative modes even with Auckland’s underdeveloped public transport options and dangerous walking and cycling infrastructure. Transport agencies need to recognise this market signal and build for the next Aucklander. That means:

  • Unlocking the potential of Auckland’s rail network by building the City Rail Link
  • Recognising the runaway success of the Northern Busway and accelerating investment in similar projects in underserved areas of south-east and west Auckland
  • Installing a safe, complete cycle network
  • Rolling out Auckland Transport’s new frequent bus network and complementing it with integrated fares.

The next Aucklander is likely to want to travel by bus, train, or bike. Will we invest to give them that choice? Or will we ignore all the market signals?

Share this


  1. Nice one Peter! The chart speaks for itself – Aucklanders are migrating quickly to public transport and away from commuting in cars.

    In fact I would argue that the chart significantly understates the demand for public transport, because train, ferry and bicycle are all artificially constrained by a shortage of trains and railway lines, by expensive and infrequent ferry services, and by a shortage of safe cycling routes. Once we get these constraints lifted we’ll really see public transport fly!

  2. Yes great post Peter.

    And David It is pretty clear that PT and Active mode use is suppressed in Auckland because the moment we improve any of it even a bit demand responds strongly. Contrast this with driving where despite multiple billions being spent that should make it a much easier option the response is muted.

    Basically we are looking at a saturated market. Driving amenity supply increase does not effect demand because we already have so much. This is a mature market. Invest where there is growth.

    Also the best way to improve the driving experience now in Auckland is to invest away from it: those new PT users were driving.

    1. There is a parking building on Park road – is there cycle parking in there as well as car parking (in the corners usually?) or is it staff only there.

      I guess the hospital doesn’t plan for cycle parking historically as they would consider the only cyclists who will ever visit the place come without any choice in the back of ambulances 🙂 rather than arrive on cycle by choice.

      I can say that the only way to access the complex full without the one way system is to enter it from the Park Road entrance as going in from Grafton Road entrance means you can’t get near the main building because of the one way system.

    2. Try looking on the University of Auckland website. Their Grafton campus is just over the road from the hospital and I would be surprised if they didn’t have bike parking facilities there.

  3. If you enter from Park Road and go through the drop-off area you will find a steel fence on the left hand side of the road which is perfect for locking your bike to. You’ll find plenty of other bikes there.

  4. It would be good to see this graph with the area of each bar representative of the actual numbers of trips, so ferry would be very thin while passenger/driver would be very broad.
    Perhaps also an indication of max capacity for each model.

  5. The next adult Aucklander is likely to be a child who grew up here. That person is less likely to want to drive than the average adult now – the figures are very clear on this trend, here and abroad: (Your own articles have illustrated this well.)

    Planning for current Aucklander is a way of cementing us in the past, and blocking the preferences of actual human being Aucklanders.

    1. I agree.
      There is a lot of latent demand for alternative transportation that is not being explored.
      As this graph shows that whenever an alternative for cars pops up, people pounce and use it.

      Alternative transport (to private vehicles) should receive the majority of funding I reckon.

  6. Patrick just to clarify, the total commuters column (the last one) is the overall change from 2006 to 2013 censuses on the number of commuters (by all modes) across these two census?

    So that means over the 7 years (to 2013) the total number of commuters who commuted to paid work on census day increased by 3.6% – but the change in “car driver” or “car passengers” only increased by 2.3% – so “Car” increase was actually less than the actual growth of commuters (“the rate of inflation”) – this means that the share of “car” in the commuting mode mix from 2006 to 2013 actually dropped by 1% relative the other modes (in “real terms”).

    Considering the high modal share of “car” in Aucklands commuting population that is a huge drop.

    i.e. if mode share for car was 83% in 2006 (which I recall it was about that sort of a percentage), then it would be only 84.9% in 2013 (83*1.023) , when it should have been expected to be at least 85.9% in 2013 (83*1.035) if it was keeping up with the general growth in commuting numbers.

    To be flat is bad enough considering how it always be up and up every census for as long as they have been held for car and passengers – to be declining shows that people are getting fed up with the traffic and voting to use the other modes.

  7. Although I don’t disagree with your conclusion, I do disagree with the way you got there. You seem to be inferring what the next consumer would buy from what consumers are buying now from what is available now. One could reasonably argue that the next consumer would want to travel by car, and that current consumers are not traveling by car because all produced “car travel opportunities” are being consumed by others. In the same way, if there was less internet bandwidth available than Aucklanders want to consume, and so young people started buying more radios as a cheaper alternative, it wouldn’t mean that they actually wanted to own those radios rather than an internet connection.

    1. Reasonable Q, (but to use your analogy), if there is now internet bandwidth surplus (as shown by the fact that existing “consumers” of the bandwidth “drivers” – are no longer growing as fast as the capacity is being increased, i.e. the growth in “Car” usage is decreasing – because the provision of internet bandwidth/road space is increasing more than the any increase in car usage over the same time).

      And yet radios (aka “PT journeys”) are selling like hot cakes – even as the available (radio) “stations” that make experience better are increasing only relatively slowly as compared to the internet bandwidth growth.

      Either, there is “imperfect knowledge” in the market place (i.e. those who would like to drive don’t as they either don’t know – when and/or – where to drive, or the cost to particiapte is too high).
      Or has the market itself fundamentally changed? And people are showing now they simply don’t want the same old range of “Hobsons choice” products they were previously offered at whatever price?

      The consumer can only buy products (or use modes of transport) from what is on offer – true they can wish for personal teleports – but they aren’t available, at any price, any time soon.

      There is no point assuming that because “next” consumer would like to buy a teleport – you should base all your future travel planning on that – as the existence or otherwise of this “latent demand” for Teleports does not change one iota the future date when said Teleports will be for sale in any practical sense – because the technology doesn’t exist yet.

      [For Teleports, read “driverless cars” and the analogy is equally the same].

      However, you know that for the forseeable future, what you have now will form the basis of what people will buy in the future then its also a safe bet to assume that change is going to happen, so just because the “cars” are it doesn’t mean it always will be so, and that it is also false to assume that everyone not buying cars is really wishing they could.

      The problem comes when you assume (as our forebears do/did/and still do) that because cars are what you want yourself, that is what you will offer, and that that is all anyone will *ever* want. We have that fallen in that trap with a “Cars for Africa” based travel mode focus time for a change.

      The people have spoken and continue to speak with their feet.

      “Bread and Circuses” – will only stave off the revolution, not prevent it.

    2. This is a really good question! We’ve taken a look at it in the past. As you have no doubt noticed, an analysis of transport data shows that there has been a sustained New Zealand-wide (or really, developed-world-wide) reduction in VKT per capita over the last decade. While this _could_ in theory be happening due to limitations on road space, it’s a bit unlikely given the fact that this slowdown in VKT has coincided with an unprecedented increase in road-building. (See some data on this at:

      Moreover, when we look at demand on some _specific_ roads, we find that traffic volumes have often stayed flat or even fallen over the same period. The Auckland Harbour Bridge is a classic example. Average daily traffic volumes on the bridge peaked in 2006 and have fallen by about 10% since then. They’ve stayed low since then. The bridge has some spare capacity – more cars crossed it in 2006, and it’s not like the clip-ons have been knocked off since then! It’s just that people don’t want to drive across it any more, especially since a high-quality PT alternative like the busway is available. (See a quite good chart of traffic volume on the bridge here:

      Honestly, it is a bit surprising to see all the market signals in the transport sector pointing towards strong demand for PT and weak demand for roads, but that’s the way it is now.

    1. Sure, but its sustained growth – and it been doing that sort of ramp up consistently year on year for 7 years and looks like doing so for longer again.

      I don’t think growth of car driving would have *ever* trended up anywhere in its time, over 7 years like PT use has over the last 7 years, and cars have had 90+ years to get where they are today, with 60+ of those years with policies to specifically promote their use over all other modes.

    2. Most definitely! That’s why I also reported the figures a different way, showing that the majority (61%) of growth in commute trips from 2006-2013 came on PT or active modes.

      Given the low base that PT and active modes are starting on, that’s an even more impressive accomplishment. But it didn’t make quite as compelling a graph 😉

    3. I guess the devil’s advocate did not have the time to read the entire post:

      “While most Aucklanders still commute in cars, alternative modes have captured the overwhelming majority of the growth in recent years. 61% of new commute journeys between 2006 and 2013 were made on public transport, on the bike, or on foot. (Broken down as follows: 22% of new demand was on the bus, 16% on trains, 7% on ferry/other, 6% on the bike, and 10% walking or jogging.)”

      1. No doubt that there has been significant growth in PT usage in this time. This can be attributed to the completion of the DART project, opening of the Northern Busway and the availability of integrated ticket products that allowed for mode transfers. Some of these journeys would have been multi-modal e.g. car to Albany and then bus. The new network with its forced transfers will see a significant increase in “patronage” but not necessarily an increase in revenue. Doesn’t mean we shouldn’t continue to invest in PT projects but we do need to ensure that they add value for the money spent.

        1. With frequent all-day every-day service and a connective grid that goes everywhere, the New Network will be vastly more useful to far more Aucklanders. It will absolutely increase revenue. And what’s more, the New Network costs the same to run, it’s just far more eficient. So it also means an enormous increase in value for money.

          1. Nick I accept that providing a better connected network and more opportunity will see an increase in usage. I’m not convinced that this will result in reduced subsidies (which was the primary aim of PTOM) as the perfect storm of integrated fares , likely increased tender costs due to gross contracts and likely PVR increase will not result in increased net revenue. My point was more about the investments we make in all projects PT and roads generated value for the cost incurred.

          2. Foreigner, the new network is service-km, and service-hour neutral and actuallt PVR drops slightly allowing operators to drop the worst of the worst of the fleet. Contracting will be cheaper because PTOM does away with the worst of the privatize the profits and subsidise the loses. Model.

        2. Foreigner. The additional 2 million train users over the last year will have added at least an additional $5 million to the operational income of the PT networks at no additional cost. We’re running those train anyhow. Adult fares with HOP discounts ranging from $1.60 to $8.40, and the average train journey being 16km or about 4 stages or $4. Of course there are monthly passes, student and child discounts and fare evaders. So to be safe let’s say an average $3 = $6million.

          Bus use has been growing too, so there are definitely new uses on the PT Network and they must have been drivers, as it is all growing way faster than population growth in AK which is at 2.3% and anyway is mostly made up of babies, who don’t do a lot of travel on their own.

          Financial this great, but economically it’s even better.

  8. I am confused! Is the graph showing the percentage change in mode percentages? ie a % of a %? If so does that have any meaning to anyone outside of the PR industry?

    1. See I told you devil’s advocate was devilish…FWIW I too am confused! when people choose handles that indicate both staunchness and insecurity at the same time!

  9. “dangerous walking and cycling infrastructure” – I really wish we could move away from the stereotype that cycling is dangerous. It really isn’t – the problem is that people THINK it is dangerous – subjective safety. Therefore they don’t ride – reducing the number of people cycling and making it even more dangerous. As NY has shown, the more people cycle the safer it is:

    Objectively cycling is not that dangerous if assessed on a per hour basis – the only really fair way to do it as a per km would massively skew in favour of cars:

    “The ongoing national Household Travel Survey tells us that we collectively cycle more than 25 million hours each year in NZ. This means that, on average, we have one fatal crash for every 2.5 million hours of riding or more.”

    The reason why cycling infrastructure is so important is that it will encourage lots of new people to cycle. The more people cycle, the safer the roads are for everyone – people in cars, trucks, on bicycles and walking. You can see in the Cycling Christchurch article I linked to above that the Netherlands has half the number of cyclists per capita killed. But it also has half the number of everyone killed on roads.

    Less than one person a month is killed on a bicycle while at least one person a day is killed in a car.

Leave a Reply

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