Jarrett Walker recently posed an interesting question, that he was after some more in depth research on than the usual ‘reckons’, “why is public transport ridership in the US falling so much?”

This builds on a tweet from Kirk Hovenkotter, showing that ridership had fallen in most large US cities over the past year, which was reported on CityLab:

A number of “culprits” are suggested. The obvious one being that oil prices have fallen significantly over the past few years, alongside an ongoing process where immigrants – or poorer Americans in general – are being priced out of transit rich inner cities and into more car dependent suburban areas:

Some of the factors behind these declines are national, as the transportation scholar David Levinson points out via email. The economy is expanding, and oil prices are plunging. People are buying more cars and driving them more often, both to work and to weekend activities that are better served by vehicles. American cities continue to suburbanize, and as they do, taking transit often becomes a less attractive option. Immigrants, long a strong base of ridership for agencies, are increasingly moving out of urban centers… and buying and driving their own vehicles.

Other suggestions are that people are shifting to using ride-hailing services like Uber and Lyft – but the evidence is pretty thin on the ground when it comes to that:

This argument probably holds truest for weekend boardings. But the best research out there (and there isn’t much yet) suggests most workers don’t rely on Uber and Lyft for regular daily commutes. Ride-hailing may even be more supportive of transit than competitive, at least in the biggest cities (smaller cities might be another question). At the very least, it doesn’t seem to be siphoning a significant number of riders away. When Uber and Lyft left Austin, mass transit saw a very modest one percent bump in ridership, according to the transportation consultant Jarrett Walker.

Of course, meanwhile in Auckland public transport ridership continues to grow strongly, as I noted yesterday February ridership was up 8.6% on February 2016, with both the rail network and the Northern Busway again registering double-digit increases.

Even across Canadian cities that we usually enviously compare ourselves, ridership growth is much slower than Auckland – although they start from a higher base. This leads to an interesting question of why Auckland is bucking the trends seen elsewhere so strongly. I think there are a few possible suggestions:

  • Auckland’s recent rapid growth and the growing congestion it has created, means that PT offers a pretty competitive travel choice for many people – especially when using the rail network or the Northern Busway.
  • We’re still seeing the benefits of recent investment in rail electification and integrated ticketing, as well as the improved “value for money” offering that came with zone-based fares last year.
  • Service network improvements, mainly in the south so far, have also helped increase ridership – some of that is a result of us shifting to a system that encourages greater transfers although indications are that overall journeys have increased too.
  • The NZ/US dollar exchange rate usually offsets fluctuations in oil prices so we don’t see as rapid increases/decreases in fuel prices at the pump as is the case in the US. Also a higher proportion of what we pay is tax when compared to the US.

All up we are doing well to buck the international trends and it should give us ongoing confidence in investing in public transport – that people will continue to flock to where improvements are made even when fuel prices are relatively low. We’ve still got a long way to go but perhaps one day soon we can be envied as a city used in case studies of what to do to make public transport better rather than our history of the opposite.

Share this


  1. Auckland is doing well because it is starting off a very low base. A number of trips that could be taken by PT ‘competitive’ was very small under the older model of running PT. Under the new model, more people can use PT.

    As services improve, the growth rate will be very high until it approaches a saturation point – where adding more services does not induce much more patronage.

    In cities that have a matured PT, one could expect only modest single-digit growth due to proportion (i.e. a high base).

    Mohring Effect https://en.wikipedia.org/wiki/Mohring_effect
    Sigmoid Function https://en.wikipedia.org/wiki/Sigmoid_function

    1. Auckland actually has higher PT patronage per capita than most US cities. For instance, Salt Lake City and Denver have also done quite a lot to turn their PT systems around and invest in rapid transit. Based on the above table, SLC has around 40 annual PT trips per capita and Denver 35. Seattle, which is one of the two cities with growing PT patronage over the last year, has around 59 annual PT trips per capita.

      Auckland, by contrast, now has 52 annual PT trips per capita, which would put it near the top of the US cities. This isn’t to say that we’re awesome, but that the US cities are really, really shit.

      1. People simply use the mode of transport that best suits them, time-wise, $ -wise, flexible-wise, etc. There will be some who use PT, some who use cars, some who walk or cycle. Nothing or no group is ‘shit’ at it, which is a particularly offensive thing to say without understanding what those groups of people do, or their individual needs. Not everyone has to use any particular form of transport. We are all individuals who have differing transport needs at differing times for differing reasons. For example transporting a family is cheaper by car in Auckland than PT, transporting an individual is cheaper by PT than car, etc. PT won’t cut it going to Bunnings t pick up DIY materials.

      2. By comparison, Australian cities are around 100 trips per capita / year, with commuter rail doing the heavy lifting (although in Melbourne tram ridership is now approaching rail).

    2. Yeah hold the phone, compared to the figure above our 86m trips a year puts us above half the cities on that list. Hardly a low base compared to those comparisons.

      For example, Dallas-Fort Worth-Arlington. Less boardings each year than Auckland and what, eight, ten million people in the urban area?

      Sorry, I don’t buy the low base argument.

  2. In general it is safe to say that the US does not offer a good model for ordering cities. The truly enormous subsidies for driving are so widespread and distortive that it makes looking to US for any kind of benchmark a mistake.

    Though there is one thing I think we could learn from US cities and that’s using small percentages of sales tax to fund infrastructure. To me this looks like the one of the better answers to NZ’s local authority budget woes, especially in the face of the pressures faced by the growth of the tourism industry. Tourists pay gst, how much they pay is in proportion to the demands they place on the areas they visit. If say 1c of of the 15c gst was distributed pro-rata to all regional councils to help fund infrastructure up and down the nation then this would greatly help returning a measure of balance to this problem.

    Currently all gst goes straight to central gov, so a tourism boom swells government coffers while the costs of the boom largely fall locally; whether it’s public loos in Southland or Light Rail in Auckland. The government jealously hoards this revenue boom, soon to be smoked on election bribes, while all regional councils spend far too much time and energy working out ways to slash half a librarian’s time, or not picking up rubbish. Roads go un-repaired etc.

  3. Perhaps we shouldn’t overlook what Gil Pennelosa called “dignify the experience”, of in this case bus or rail or ferry travel. And here I think the shinny new toys really help – the smart new trains, the cool new stations and so on help to make people feel they are not making a second class choice.

    1. I think the relevant equation is:

      Experience * time * cost

      People will use a service if two of these are good. They’ll cram on a crowded bus for a short time if it’s cheap. They’ll pay a reasonable price for a train that allows them to sit and read at comparable commute times, or a bit more (less than a taxi) for a fast airport train, etc.

  4. I don’t buy the low base argument to a degree, it depends on who we are comparing ourselves to. Some of those cities with similar ridership rates around 85m have similar or larger populations than Auckland. Minneapolis has twice our population with similar ridership. Milwaukee has about the same population and half our ridership. Even Seattle has twice our population but only twice the ridership. That means we are doing quite well in comparison to those US cities. It may also just mean that most US cities are really crap at doing PT and we should avoid doing anything they do because it is killing their ridership.

    Auckland is benefiting from years of investment into trains, larger buses, boomers retiring and spending their days on free PT etc. We have been slowly improving capacity and people choose to use it.

    1. I don’t buy that explanation. The majority of Auckland’s patronage increase has happened on rail and busway – ie the high-quality, highly reliable elements of the network. And there’s a lot of people of all walks of life on PT.

      A more plausible story is that when you provide something that works well and gives people a good experience, they will use it. Full stop.

      1. It will be difficult to tell who is right in the future Peter. The new network requires more people to change mode so all of the ridership number will increase even though the same number of people will make the same trip. Do you think that might be the intention question mark.

        1. Some of the people I work with helped design the New Network. According to them, raising patronage by forcing transfers was never a design objective. Rather, the aim was to get a public transport network that provided more frequent service to more parts of the city, which in some cases does require terminating services at logical transfer points like Panmure Station.

      1. I get your point, but I imagine taking public transport in Detroit is dangerous. Also high unemployment = low number of commuters. In the case of severe poverty, public transport isn’t an inferior good. Walking for two hours in the rain is…

    2. While the examples of inferior goods on the Wikipedia page lists busses as an inferior good, what’s defined as an inferior good in any given context can change, and can even differ from person to person.

      If the desirability of a good increases in a context, it could become a ‘normal good’ in economic terms, as is the case for high quality rapid transit systems where they are developed. The reverse can also happen, where driving in central London is so undesirable it could be considered an inferior good.

      Your argument against PT in this case just doesn’t stack up.

      If we create a PT system that is high quality, fast, on time, and relatively comfortable, the less people will subconsciously see PT as an inferior good, and the more people will use it.

  5. I put this comment on Jarrett’s site, but for some reason it’s still in moderation.

    An international perspective here. Some but not all NZ cities are experiencing falling PT patronage. Auckland is up quite a lot (total patronage rising at 6-7% per annum), Wellington is up a little (~3% per annum), but Christchurch is down and most of the smaller towns are down as well over the last couple of years. Some data here: http://greaterakl.wpengine.com/tag/patronage/

    This is happening against a backdrop of rapid population growth, rising traffic and congestion, and lower petrol prices. The differentiating factors seem to be:
    * Rapid transit availability / investment – Auckland has recently invested to significantly improve its rapid transit system, while Wellington has a long-established (but fairly static) regional rail system. A majority of Auckland’s patronage growth has been on rail or busway.
    * Geographic constraints – Auckland and Wellington have geographies that funnel a lot of travel demand through pinch-points, which experience concentrated traffic congestion as a result. Where rapid transit or bus lanes are available, this is an incentive to shift modes. Christchurch is a flat city on a plain.
    * Land use disruptions – Christchurch is still dealing with the impacts of a major earthquake that hollowed out the city centre.

    The big Australian cities seem to be on the same trend as Auckland and Wellington: https://chartingtransport.com/2017/01/28/update-on-trends-in-australian-transport/

    So if you’re looking into the US experience, I’d start by investigating some cities where PT patronage *hasn’t* declined, and see how they differ from the rest. That might give some preliminary indication of what factors matter, and what factors don’t, and motivate a more in-depth analysis.

    1. I think a transport orientated blog such as this should be investigating much more why Christchurch has had no PT uptake.

      1. I’d love to see some analysis on Christchurch PT, but understand it’s not the focus of this blog. I think the Christchurch PT situation is unique and complex to untangle. For the decade before the earthquakes Christchurch PT was high and growing. Even in the first 2 years after the quakes it was growing fast and looking like it would get back to pre-quake levels quite quick. Only in the last couple of years has patronage flattened off. I think there are many contributing factors; CBD still only a quarter full, network restructure in 2014 resulting in overall fewer service-kms, dysfunctional governance (currently in the process of being reshaped), dispersed development patterns, deferred investment in PT infrastructure. I suspect the next few years will be different though – the governance issues have now been addressed and just need some time to bed in, and the shift of workers and residents back into the CBD is going to start ramping up.

        1. There’s also a growing storyline I’m hearing from the public (and even politicians) that Chch public transport is broken/buggered/dysfunctional. This has mainly come from the stories of those who were adversely affected by the last round of bus route changes (i.e. they now have to walk further to reach a bus or take two services instead of one). At the same time however, the high-frequency lines provide 10-15 min services on eight corridors into the city plus the orbital route, although they are still hampered in places by a lack of bus priority corridors (the busiest corridor in Chch, Riccarton Rd, is only now getting some bus priority). This is also coupled with ongoing confusion about how fares work in Chch (e.g. thinking they will have to pay more for taking two buses, not knowing about the maximum daily fares), despite it being by far the cheapest and simplest bus system in the major cities. While there are certainly network routing and bus priority issues that need to be addressed, the biggest problem seems to be a failure to take the lead on the narrative of what buses are actually like, rather than the perception.

        1. Thanks for the shout out Patrick. I hope some Cantabrians take up the offer. I am trying to find some time for a transport type article.

      2. Christchurch PT was buggered long before the ‘quakes.

        Splitting the management of PT planning and delivering across two or more separate organisations/councils has not worked well here or in NZ generally.

        Just like it didn’t work in Auckland before the SuperCity was formed.

        And while, some say that Auckland PT is not working any better now under AT’s control, at least there is only one organisation to blame unlike elsewhere.

        Still the present Christchurch system as it stands or falls, definitely has Joyce and Brownlees grubby, interfering, fingerprints all over it.

      3. I’ve been meaning to do that for a while, but my intentions keep getting overtaken by events. I’m also not well enough in touch with the publicly-available Christchurch data to come to firm conclusions. (Where I have better data, it’s stuff that’s come to me in a work context, so I don’t feel comfortable blogging it.)

        With that caveat, here are some qualitative observations:
        1. You can explain a large share (think 25-30%) of the reduction in patronage by looking at the disruption in PT commuting to the city centre. This has been persistent, as city centre employment recovery has lagged the forecasts that were developed immediately after the quakes.

        2. Christchurch’s high-frequency network performs well, but there is also a lot of resource being devoted to fairly circuitous suburban routes. The bus interchange is a constraint to raising frequencies on the core cross-town routes, which in turn puts a floor under average waiting time for transferring between buses at the centre.

        3. From my limited experiences riding Christchurch buses, there appear to be issues with reliability and on-time performance, even during the interpeak periods. You could fix this with more bus lanes and priority at lights.

        4. City centre parking prices are looooowwwwww and they won’t get off the floor without more development to take up the empty lots.

  6. Growth in PT is good but as other comments have said it’s off a very low base and growth will be capped by capacity of current options.
    The fundamental problem, based on the last census, is that there is a ratio of 20 drivers for each passenger in peak hour traffic in Auckland.
    Rather than waiting for infrastructure to be built, people have the power to significantly improve traffic by sharing seats in their own cars – 50% of traffic is going to 2.7% of Auckland’s land area. Less cars with more people is the short term answer.

    1. How would that significantly improve traffic?

      Doubling the number of people sharing rides would raise the average vehicle occupancy from 1.1 to 1.2 people per car. If the number of people travelling by car was fixed, that would be an 8% decline in vehicle traffic.

      But remember we have traffic congestion, which simply means there are already more people trying to drive at the same time than the roads can handle. So you wouldn’t get a decline in vehicle traffic, because traffic isn’t fixed and it’s already oversubscribed, you’d just get more people on the road at the same time.

      In that regard there is very little reason for anyone to carpool, It just means more inconvenience for the individuals concerned but exactly the same traffic. So there is no mechanism for it to work. Compared to driving alone it just makes your life harder for no benefit.

      One thing to note, the goal of proper public transport isnt to reduce traffic, its to let people travel without being part of traffic in the first place.

      1. Hi Nick,

        We’re not talking about timid incrementalism. We’re talking about engaging people on a wide scale.

        Of course there will be less cars on roads if significant numbers of sole drivers start sharing – they need to be supported with tools to make this easier – we’re part of the help – not the only ones.

        Increased public transport usage is great but a 9% increase on an 8% base is only a shift of 0.72% – timid incrementalism at best – it’s heading in the right direction but without massive increases in options and capacity it’ll remain low.

        Wellington is having the most success with both public transport and exercise – around 15% now walk, jog or cycle to work – Auckland is lowest on this and going backwards.

        Again all modes of transport are important in addressing traffic – Localift is multimodal and we’ve had interest from cyclists, runners, public transport users and even walking school buses.

        Attitude is the biggest thing to change – a can-do commitment to fundamental behaviour change is a prerequisite – not waiting for governments or others to fix it for us.

        Happy to share more with you.

        Cheers, Stephen

        1. Meh. Carpooling has been “the future” for decades.

          To carpool you need to find someone who
          (1) lives close by
          (2) works somewhere near your workplace
          (3) has a very similar schedule as you

          These are long odds, and most people just won’t find a match.

        2. Sweet buzzwords Steven, you’ll have us believe we’ll all be carpooling if we just want it hard enough! Just need a can do attitude!

          Here is a little hint: it’s almost impossible to change people’s behaviour. As a society we do it with laws, enforcement and punishment, or with subsidies/bribes of one form or another. But even then laws are often broken, and encouragements ignored.

          What you actually need to do is change the conditions and let people change their own behaviour accordingly. If you want to change the outcomes, change the inputs.

          This is how rapid transit has been so successful in Auckland, we didn’t use our can do attitudes to ‘behaviour change’ people to take the same old crappy slow buses stuck in traffic, we gave them a fast reliable busway that skipped the traffic entirely and worked out better than driving on the motorway. Now almost half the people crossing the harbour bridge each morning are on the bus. Not because they were conned into it, but because the bus is actually better for each individual on board.

          You are approching the issue the wrong way, you cant leave everything the same and expect people to change. People will always do what is easiest for them. If you want people to do something different, make that something different better than what they do now.

          Here is a question for you, if people are willing to run their own car and sit in traffic and risk unknown delays day to day, why would they do the same but add in extra time and hassle of coordinating longer and more circuitous trips to share with others? What benefit does your system provide to the individual? Does it make travelling significantly cheaper, i could understand that having an effect. Does it make the trip somehow faster and more reliable? Is there some other advantage for the person concerned?

  7. The main problem with PT in the US is image. It is seen as Poor Transport rather than Public Transport for the most part. In most cities in the US it is pretty much impossible to have a journey without being harassed in one form or another by someone that either lives on the street/has mental issues/is drunk or on drugs. While that can occur in Auckland it is far less common. Part of this relates to society in the US in general with a lot more street people/urban poor. I’ve had a knife pulled on me on a bus in the US (luckily it wasn’t a gun) and have seen other similar things happen there.

    1. Guess you have never heard of cities such as: LA, San Francisco, Seattle, Salt Lake City, Denver, Chicago, New York, Boston, Washington DC, Philadelphia, Atlanta, Miami, Baltimore?
      All of which have more extensive rail networks than Auckland does.

      1. Yeah, and I think you’ll find that bus ridership is down (even in Seattle), while trains are static – or growing, in the case of Seattle where a new line and improved services were provided.

  8. I think you’ll find urban rail patronage is still rising in much of the rest of the world outside North America. The US seems very sensitive to fluctuations in fuel prices. Maybe because they have a lesser tax component, price fluctuations are sharper percentage-wise.

    1. ‘because they have a lesser tax component, price fluctuations are sharper percentage-wise.’

      Absolutely right. Also nowhere on earth are the transfers to driving as great as the US… the billions that subsidise every aspect of the driving world from fuel, to vehicles, to parking, and of course freeways, are vast… it’s a testament to the persistence of the reality of geometry in cities that any Transit functions at all in the US. And why the persistent but never arriving idea that Transit is dead originates there; and of of course is repeated by dull minds elsewhere like the NSW transport minister…

  9. Because the Crown’s gas monopoly (high pricing and high tax ) on gas .
    Gas is cheaper in the states and PT is not as pleasant an experience in the US.

  10. Oh and the other more subtle reason for the US decrease in PT is the growing social unrest what with with the banker’s media conglomerate propaganda stories inducing a public fear of others (planting seeds of racism, Islamaphobia). Looks to me like its media driven ” domestic terrorism” .control the minds and you control the people.

Leave a Reply

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