Lies, damned lies, and patronage statistics

Let it be known that I’m not a great fan of how our regional council, ECan, presents their patronage statistics. In August 2018, they reported the following to the Greater Christchurch Public Transport Joint Committee:

Patronage for the year ending 30 June 2018 rose by just under one percent to 13,572,240 boardings for the year. While only a modest increase, it should be noted that this is the first time in four years (and only the fourth time in the last 10 years) that yearly patronage has increased.

 Agenda for Greater Christchurch Public Transport Joint Committee (page 13), 15 August 2018

That’s got to be good news, eh? Maybe. But only if you overlook that we experience massive population growth. The following graph shows population growth for Waimakariri, Christchurch and Selwyn, based on revised population data issued by StatsNZ in October 2018:

Greater Chch population

The population growth from mid-2017 to mid-2018 was exactly 2.24%, i.e. patronage per capita has in fact fallen. Maybe the news, as presented to the committee, wasn’t so good after all?

The eagle-eyed amongst you may have spotted that the graph above shows a 2018 population well above half a million people and I should point out that Greater Christchurch as defined for planning purposes excludes some areas from all three councils. The actual population within Greater Christchurch sits at just over 480,000. The population growth within the true Greater Christchurch area was 2.32% over the same period so it’s basically the same. I’ve chosen that simplification for all the analysis as it’s much easier to get population time series for the district council areas.

Map of Greater Christchurch

I gave a presentation to our regional council, known as Environment Canterbury but commonly abbreviated to ECan, in April 2018 as part of their Long Term Plan consultation (sample slide below). One of my eight recommendations to them was that they change their reporting to patronage per capita. It’s simply misleading if you don’t do so when you have big changes in population. The graphs shown to the ECan decision makers may imply that ECan is doing well when it is not. Leaders cannot make good decisions when data reporting is not showing the true picture.

LTP-slide

So how are we doing with patronage per capita? The following graph shows monthly data between July 1999 and June 2018. Looking at the blue line, there are big seasonal variations and that makes it hard to read the graph. For that reason, one uses a data-smoothing technique that creates a 12-month rolling average (the orange line).

When you leave out the spin and look at hard data, it can be clearly seen that patronage per capita has been falling since 2014. In June 2017, the average person took 27.19 trips on public transport and a year later, the average person took 26.55 trips. A “rise by just under one percent” is fake news; that’s rather a drop in patronage of just under one percent.

And it isn’t hard to explain why that has happened; the “Hubs and Spokes model” that has been gradually phased in since December 2012 forces many passengers to change bus. Forcing people to change bus represents a drop in level of service. Forcing people to change from a high-frequency route to a low-frequency route without a guaranteed connection is a recipe to really piss people off. Maybe Christchurch’s sky-rocketing cycling number can, at least in part, be explained by public transport having become much less attractive over the last few years?

Hubs-and-spokes, with four core routes going through the central city (spokes) and other buses connecting to those at suburban interchanges (hubs)

So how are we doing compared to the other two big cities? Before I answer that, I offer the observation that while Auckland Transport and Greater Wellington publish their raw data on their websites, ECan does not. In Christchurch, you have to go begging to get monthly user numbers. Why? The other cities are transparent; why is that different with ECan? Do they want to hide what’s going on? I’ve asked those questions and they haven’t been answered.

The chart below compares patronage per capita for Greater Christchurch, Greater Wellington, and Auckland.

A few things stand out:

  • Patronage in Wellington is relatively high; it’s currently about 2.8 times as high as it is in Christchurch. On average, each Wellingtonian makes 74 trips on public transport per year.
  • Patronage in Wellington is static, and has been for a good decade. That said, their latest data series goes up to June 2018 but they implemented a hubs-and-spokes network in July, and combined with other ‘improvements’ their patronage will have experienced a subsequent dip. [editorial note: my suspicion was unfounded, their patronage is steady, and I’ve eaten my hat]
  • In 2006, Auckland had the same level of patronage as Christchurch. Auckland patronage has grown ever since.
  • Christchurch has experienced an earthquake (which doesn’t help) and after some initial recovery patronage has been falling since 2014. On average, each person in Greater Christchurch makes 26.5 trips on public transport per year.

If you were to linearly project Wellington and Auckland’s patronage performance over the last decade into the future, you would see Auckland catching Wellington during 2030. It will not take that long for these reasons:

  • Auckland is on a roll. They are improving their public transport provisions all the time. There is no end in sight with innovations. When they open the City Rail Link currently under construction their patronage will make a massive jump. And they are about to start building a light-rail network.
  • Wellington’s July 2018 bus restructure hasn’t gone too well. Despite my suspicion that they would have experienced a drop in patronage, this turned out to not be the case; their numbers remain static.

My guess is that Auckland patronage numbers will meet the Wellington patronage somewhere between 2025 and 2027.

Let me state that there’s nothing principally wrong with having a hubs-and-spokes network. These things can work. I’m not familiar enough with the Wellington network and bus frequencies to be able to judge whether or not it could work there. What I can say is that it will never work in Christchurch as there simply isn’t the bus priority infrastructure in place that is needed to ensure that the frequent core routes arrive at the hubs on time. If you cannot ensure reliable connections to services that run at a lower frequency, you will soon get rid of that part of the ridership with transport choices. And that is why we are seeing declining patronage in Christchurch.

Introducing hubs-and-spokes in Christchurch was never about making public transport better or attracting more passengers. It was about saving cost, with the commissioners aiming to meet the national fare box recovery policy introduced by Steven Joyce as quickly as possible (see my Metrocard post for more context). At the time this was implemented, ECan was run by a group of commissioners installed by the fifth National Government. Bus kilometres were cut by 10% to 15% and the planners were keeping their fingers crossed that they wouldn’t lose patronage in the process. Well, their wish didn’t come true. One of my recommendations at the LTP hearing was that councillors and commissioners should acknowledge that introducing hubs-and-spokes in Christchurch was a mistake.  There’s a video recording of my presentation; I think I had their undivided attention for 12 minutes.

We need to do better. Patronage ought to be increasing, not decreasing. I told ECan councillors and commissioners that they ought to aim for a “superb level of service”. What do you think ought to be done?

16 thoughts on “Lies, damned lies, and patronage statistics

  1. The PT networks of all these cities depend on central government funding to be developed. Significant capital expenditure has been made by government in both Auckland and Wellington in recent years. Whereas in Christchurch all we really have is the spending on the new bus exchange in the city. So both Auckland and Wellington have benefited from the capital grants put into their systems that has not been committed to down here. For example the costs in Auckland to double their rail tracks, build a lot of new stations and electrify with new trains. In Wellington they have got new trains and various other things.
    So the answer would be to ask those politicians in Wellington why there is no money yet. Even the election promise for the feasibility study would be a good start, instead we have a cop out, CCC got to choose and they decided $200 million should go on a stadium and public transport shoud get $50 million of ratepayer support over 10 years, that is tiny.
    Public transport provision is heavily regulated by central government. The new government after 1 year has not done a whole lot to change the negative regulation and subsidy cuts from the previous government. This also affects Wellington’s bus network and blaming GWRC for something that the government dictated they had to follow isn’t that reasonable. There have been a lot of promises and select committees and other government grandstanding but I don’t see a whole lot of solutions.

    Like

    1. You suggest to “ask those politicians in Wellington why there is no money yet”. I can answer that easily. The way transport funding gets dished out is by the relevant authority making a case and putting it to the funding agency. The blockage isn’t in Wellington; the blockage is in Canterbury. Phil Twyford mentioned the other day that he would welcome a Canterbury request for serious public transport investment. But the request isn’t coming. No request – no dosh from central government. Quite simple really.

      Like

  2. So little money is spent on PT in Christchurch compared to Wellington and Auckland that I am not at all surprised by the appalling stats you have outlined. You only get out what put in. Unfortunately it appears to be easier for most local body politicians to peddle the fantasy that Christchurch is a “car city” and that people won’t use good PT, so it’s not worth investing in. Auckland’s story over the last 10-15 years should put that fallacy well and truly to bed.

    Liked by 1 person

  3. Pretty bleak picture.These issues can only really be fixed by returning ECAN to full democracy as soon as possible. Public transport, the environment and the people of Canterbury have suffered enough now. It should be a priority of this government to return Canterbury to a democracy. It is time to come up with more ambitious plans to fix our public transport, clean up our rivers, end the water bottling and get serious about conserve our local wildlife. It should also be assessed whether ECAN is the right organisation to run public transport in Canterbury.

    Like

    1. Mayor Dalziel tweeted the other day that she wants one integrated system. I concur. That said, I don’t think that the city council is doing anywhere near enough to provide enough bus priority. It’s their responsibility to ensure that buses can get to places in an efficient manner. Instead, buses are stuck in general traffic. They’ve left a lot of room for improvements. And if you look through the city’s Long Term Plan, you sadly won’t be able to identify much action on that front either. There is a lot more to do than amalgamate the transport functions of a few councils.

      Like

  4. That’s really good that you’ve provided a proper per capita graph, and absolutely ECan should be doing so themselves and reporting the passenger numbers better.

    It’s funny that people are referring to Auckland’s network as a hub and spoke network, because how it has been so successful for me has been because it’s more like a grid than it was. When I gave up my car 10 years ago, to go almost anywhere, I had to take a spoke into the hub (the city) and out again. Now I have a number of frequent services going across town as well as into town, and each of these cross other frequent services that open up the whole city for me in a way that I’ve never experienced before.

    The reason it works is because of the frequency of the buses. And the reason they can be frequent is that there aren’t as many routes, meaning I need to transfer. And transferring can be an issue where there’s poor pedestrian amenity between the routes. It’s not a problem on its own. I actually love transferring. It means that on the same fare, I get to stop and do an errand in a place I might not have been able to get to recently. I can often choose between several different combinations of routes to get to my destination, and I choose them based on which location for a stop off en route suits me that day. For a mum who relies on the bus for transport (I only cycle locally), and with the family’s errands mainly on my shoulders, the transfers are a positive, not a negative.

    Christchurch’s falling passenger numbers needs urgent attention. And investment. And road reallocation and bus priority and all that. It may even be due to a hub and spoke model – I don’t know. But I don’t think it’s due to transfers.

    Liked by 1 person

    1. It isn’t so much the transfers as a concept that are the problem, but rather that is where the failures are happening. I’m all for transfers, but when you are going from a frequent service to a rather infrequent one (or vice versa), and much of your system’s reach depends on this, it can make the whole thing fall down.

      The difference with Auckland, to my mind, is that the hub and spoke model happens to have been implemented much better than either Christchurch or Wellington.

      Like

  5. 1) hub and spoke of fine but ignores the last mile. In the commuter peak most origins are from home and low density cities like Christchurch have a vast spread of origins. There has to be a comprehensive coverage network underlying hub and spoke, which could possibly be provided by on-demand transit.
    “Public transport” is there to serve a public purpose in the first instance – to provide transport mobility for those that don’t have other transport options. Comprehensive coverage must come ahead of any hub and spoke overlays.

    Like

Leave a comment