Follow us on Twitter

Rail Patronage vs Projections

With the release of the station boarding data in the post yesterday by Mr Anderson, and the recent sluggish patronage growth, its perhaps worthwhile looking at how things are going compared to projections. Again we can look at the supporting report for a lot of information as AT/AC attempted to answer one part of Steven Joyce’s question relating to evidence of patronage growth, particularly in the morning peak.

First of all, anyone who has followed this blog for long enough will know about how rail patronage has increased. In 2003 before Britomart opened rail patronage was sitting at about very low base of 2.5 million trips per annum, since that time it has shot up to over 10 million trips per annum now. While it is off a low base, it has easily outstripped population growth and the report says it has seen per capita usage jump from 2.2 trips per person to 7.3. over the same time morning peak patronage has increased very strongly going from 1,012 people arriving at Britomart during the two hour peak to 6055 in 2012. What the report notes is interesting however is that the percentage of off peak patronage growth,¬†particularly¬†in recent years has outstripped the level seen during the peak which is useful for showing that the improvements aren’t just about getting more peak users but that there are benefits off peak too.

It is pretty clear that that there has been some pretty big increases but the next question is if the increases were a result of a cannibalising of bus patronage. So the reported looked at the AM peak numbers and compared those with private vehicle numbers over time and from that we can see that both rail and buses have increased there share when it comes to accessing the city centre although for some reason ferry patronage wasn’t included which would explain why private vehicle usage is at 56% when the 2012 numbers were 50:50 between PT and car.

CCFAS Modeshare 1990-2012

But what about those projections, well each of the projects over the years have had some form of projections attached to them. In 2001 when the council of the time was putting its case together for Britomart the report notes that the current transport model used was still being developed however it was working enough to get some basic results. There appear to have been two years that were modelled forecasting 10 and 20 years out (2011 and 2021) and the numbers are below:

Britomart Projection Numbers

Britomart Projection Numbers Graph

So with the exception of outbound passengers in the morning peak, the actual numbers are considerably higher than even the 2021 projections (40% higher than 2011 prediction and 15% higher than 2021). In fact the change in daily patronage out of the station just between 2010 and 2011 was greater than what was predicted to occur between 2011 and 2021. What’s more the author of the report notes that the projections included the network and frequencies being in a similar state to what they are in now and included electrification.

The next set of projections come from the 2006 rail development plan which is what really kicked off project DART and electrification. While it appears we are tracking at less than what was forecast, there are some fairly important reasons for that. The forecast was based on by now there being more frequent services and importantly was based on electrification being started almost straight away and completed in 2011. It was however delayed by both the previous and current governments and we are now not expecting all the trains to roll out to the network till 2015/16. The report notes that the 2016 forecast patronage has now been revised up higher from 15.7 million trips to 17.3 million.

2006 Projection Numbers Graph

Electrification had been forecast to be completed in 2011

We also have projections for Onehunga, which ended up opening later than expected (from memory it was initially said that it would open in 2009 but didn’t open till late 2010. Even so AM actual numbers seem to be a little bit ahead of projections while the all day numbers are not far off.

Onehunga Line Projections

This post is starting to get a little long so tomorrow I will look at some of the reasons that have been identified as being responsible for the increase in patronage but it is pretty clear from reading through the report so far that we seem to constantly underestimate rail patronage. As the same modelling is being used, that means it is quite likely that projections in the CCFAS and the original business case have also been underestimated and that could have significant impact on the outcome of the economic analysis.

8 comments to Rail Patronage vs Projections

  • Greg N

    Theres a looming problem with the numbers here. for sure, and you can see if you compare between the Figure 4.5 graph and the Figure 4.7 graphs.

    And its this: the actual daily boardings and alighting for trains at Britomart in 2011 is shown as 25,000 (more or less),

    If I then mutiply that number by 220 – being about the number of working days in a year, allowing for statute holidays and assuming no rail shutdowns (a best case scenario), I get about 5.5 million annual alighting and boardings a year will occur Britomart. Figure 4.7 shows that total figure is 11 million alighting and boardings for the same year (the black actual line) for the entire network.

    So what these two graphs show is that PT users using Britomart for starting or ending journeys represents about 50% of the trips made on the network currently.

    Which is probably right given the figures from the earlier posts for per-station alightings and boardings.

    But, when PTOM and fully integrated fares come into force, won’t the percentage “share” for Birtomart ending/starting trips number change dramatically (down) while the total number of trips using Britomart will still increase markedly?

    PTOM alone will drive a big upswing in train journeys. let alone the additional drivers of the EMUs and integrated ticketing.

    And if so, isn’t this also showing that even the current (revised) projections in figure 4.7 are probably way, way too low, once we move the bulk of the long spaghetti journeys from the buses onto the trains as is planned?

    As I see that 4.7 figure, if you mentally slide the red line 4 years to the right (into the future) – which is what has occurrred by the delays to electrification, EMUs etc which has in effect delayed completion of the 2006 rail plan from say mid-2011 to mid 2015 (to be optimistic) – while still allowing PT usage to build.

    You can then see that even now actual usage (the black line) is tracking ahead of the red projection line and has been for many years until recently and even if that growth flat-lined from now on until post-EMUs, it would still just behind/on the red line by then. A no-one expects the growth will flatline – it will still tick upwards albeit much more slowly than it has been until EMUs come onstream.

    So yeah, we are definitely ahead of the curve (demand wise), but not ahead of the curve supply wise.

    Furthermore once PTOM becomes the norm, then the demand comes on for the switching of trains to complete journeys. Instead of catching a bus to Britomart, to catch another one back out of Britomart to your destination as occurs now – that will still occur to some extent, but the bulk of the switching will occur at the trains not the buses. And there are really only two places you can do this switching.
    So this means the pre-CRL Newmarket and Britomart will become major pinch points in the network as people have to use them to change trains.

    This alone may mean that pre-CRL – around 2018 or so the rail network may well grind to a halt on congestion at Newmarket and Britomart. Even with improvements at Britomart this bottleneck may rapidly choke the entire network unless its managed.

    That success is what you call induced (PT) demand I guess – it shows if you build it (better) they will come.
    But we also need the CRL to make them stay using PT longer term as its no good just getting folks on the PT – we have to deliver the service they require to keep them using it longer term.

    • Mr Anderson

      With much longer trains from electrification Britomart’s capacity will be substantially larger than it is now. Which should hopefully get us through to when CRL opens.

      • Greg N

        Hmm, maybe longer trains with EMUs will help – but its going to be a “damn close run thing” even so.

        And any delays to CRL completion from now on may prove disastrous longer term.
        While the delays to EMUs meant we got a better deal – it has come at a cost and not just in monetary terms and the same argument can be made for the HOP card too – we got the better system, but not without a lot of pain along the way.

  • bbc

    Figure 4.7 is meaningless then as basically none of what it was predicated on actually happened, I don’t see why they included it in the report as more than likely this is the sort of figure Brownlee will pluck out without reading anything else and say aha train patronage is well below what was forecast.

    • Richard D

      My take on figure 4.7 is that it reinforces the message from figure 4.5 – that the underlying patronage growth was underestimated, and that by 2015, even if there were no electrification, we might expect to be at the patronage levels the APT model was predicting with electrification. ,

      • Greg N

        “and that by 2015, even if there were no electrification, we might expect to be at the patronage levels the APT model was predicting with electrification”

        Except, that its becoming clear to AT at least (why else is Mike Lee and AT panicking about “falling passenger numbers” and calling in the so-called experts) – that the system is pretty much maxing out about now in 2012/13 – which is pre EMU by 2-3 years – so we won’t have the same level of growth we have had in the past over the next while.

        So we are losing momentum, more on the supply side, than on the demand side – but losing without a doubt.

        And if the falling passenger numbers are used by naysayers to justify delays in the CRL or futher post-EMU improvements (e.g. double track Onehunga, link to Western line from Onehunga, Electrification to Pukokohe etc), then we may well be still suffering the long term consequences 10-20 years from now.

        If you look at figure 4.2 graph, at the very left you can see a declining mode share of buses (basically PT as it was then), leading across the early 90′s – it took some 20 years!! – until about 2009/10 on that graph to get buses back to the same mode share they had in 1990 (and for trains to come back from zero to a somewhat respectible figure)
        - and even that mode share in 1990 was the tail end of a ongoing falling trend throughout the ’80s of falling PT use – caused by a pattern of road investment and cheap second hand car imports from Japan.

        You can see that short term decisions and delays in building infrastructure can have a major knock on effect that lasts for decades.

    • Greg N

      bbc
      Brownlee, Joyce and co will use whatever graphs or numbers they like to justify their positions.
      Whether or not the figures are correct is irrelevant.

      Witness Joyce’s comment in a recent post about a opinion piece from 2 years ago that “more people a re living near the beach these days – so the planners have got it wrong”.
      Sure, more people are living in Auckland, but as a percentage no more are living near the beach than anywhere else, but he takes an absolute figure and makes it justify his “roads for Africa” approach.
      And Brownless just echoes the same belief.

      “Facts, I don’t need facts where I’m going!”.

  • Stu Donovan

    why was not ferry patronage included in the city centre mode share? That seems a bizarre omission because it’s a relatively easy thing to count.

Leave a Reply