Peter M’s recent post on “Making light rail make sense” threw up a lot of interesting comments.
One of the more interesting arguments went something along the lines of “many people will never catch a bus so we should invest in rail“. I find this line of argument interesting because it expresses a view (about how to grow public transport patronage) that differs substantially from my own. Having spent last night tossing and turning with this issue rolling around in my head, I thought it would be worthwhile to fritter away more time on Sunday writing this blog post.
Anyhoo, when people argue that the best way to grow public transport patronage is to invest in rail they tend to invoke the following logic:
- Premise 1: There’s a large group of people out there who don’t use public transport (true)
- Premise 2: Many of these people have preconceived notions of the merits of rail versus bus technologies, which typically favour the former (also true)
- Conclusion: Therefore growing public transport patronage is best achieved through investment in rail
While I agree with both premises, I don’t agree with the conclusion for the following two reasons:
- Buses can adopt many of the characteristics that have historically been associated with rail. For example, if people appreciate ride quality, then maybe we should invest in higher quality buses and better road surface and geometry? If people appreciate the speed/reliability of rail, then maybe we should provide buses with similar levels of priority? If there are policies that help rail systems run faster, e.g. pre-paying for tickets, wider stop spacing, and all-door boarding then why not adopt them on buses too? If people are attracted to simpler operating patterns that are more legible then let’s operate our bus system the same way. Basically all of these factors, which people typically associate with rail, are not intrinsic to any particular technology, but instead reflect mutable decisions. The design of the Northern Busway and the Central Connector, for example, exemplify how bus corridors can adopt “rail” characteristics where desirable.
- It presumes the best way to grow patronage is to attract new users. The above argument contains an implicit assumption that the best way to grow patronage is to attract people that are not currently using public transport. But what if your current public transport system is only meeting a proportion of the travel demands of its existing users? In that case would there not be an opportunity to grow patronage by increasing the degree to which public transport meets the travel needs of existing users? And if you consider that younger people tend to use public transport more, does this create the opportunity to grow patronage simply by slowing the rate at which current users stop using PT as they grow older? Stated differently, could we grow public transport patronage by slowing the rate at which existing users becoming non-users.
The first point noted above is, I think, fairly self-evident and does not require much further discussion. All I’d like to clarify is that I’m not suggesting buses can adopt all of rail’s attributes, only pointing out that it is possible for buses to adopt many of the attributes that crop-up most frequently in “mode centric” debates. And by adopting these characteristics buses can effectively narrow (not eliminate) the “quality gap” that many people have in mind when they express a preference for rail.
The second point, I think, is more interesting and deserves further discussion. What I am suggesting here is that attracting “new users” to public transport is only one way of growing patronage. Another way, which would seem to involve less effort, is to expand the degree to which PT meets the needs of existing users and, in particular, slow the rate at which they stop using public transport as they grow older. That is, to focus on patronage retention rather than simple expansion.
Indeed, evidence seems to suggest that in the wake of a PT improvement the most immediate patronage growth is attributable to existing users subsequently choosing to use public transport more. And given that public transport is more popular among younger people, in the long run such improvements may increase the degree to which public transport is incorporated into their future decisions, such as whether to get a drivers license, buy a vehicle, and – perhaps most importantly – the relative locations of their home/work. The logic of this argument could be summarised as follows:
- Premise 1: A large number of relatively young people are currently using public transport to meet a proportion of their travel needs (true)
- Premise 2: These young people will, in the future, have to make many significant transport/lifestyle decisions that will impact on their future demand for public transport (true)
- Conclusion: Therefore growing public transport patronage may be best achieved by focusing on meeting the needs of existing users which, in turn, will slow the rate at which existing users stop using public transport
Some numbers might also help to make this argument a little clearer.
First, consider a stylised world that consists of only three groups of people who are able to be completely characterised by their age-group. In the figure below the x-axis (horizontal) shows these three age-groups, whereas the y-axis (vertical) shows the percentage of trips made by car and public transport for each age-group (the MoT’s HTS hints at similar differences in use of public transport between age-groups, even if the mode share is different).
Now if we know the population weightings for the 0-19, 20-65, and 65+ age groups – and we assume that all groups have the same daily need for travel (which may be approximately true for the 0-19 and 19-65 age groups) – then we can calculate the contribution of each age group to total PT trips, as illustrated below.
The first thing to note is that while the 0-19 age-group is only 30% of the population, they contribute 60% of all PT trips, which is more than twice as much as the contribution made by the 19-65 age-group. Similarly, while the 65+ age group is only one-sixth of the 19-65 age-group, the former contributes 2/3 of the patronage of the latter.
Using this little hypothetical quantitative framework we can now test a few different scenarios.
Let’s first consider a scenario where we invested heavily in rail with the aim of getting more people in the 19-65 age-group onto public transport. And let’s say that this investment was spectacularly successful, such that the public transport mode share for this age-group doubled from 10% to 20% (highlighted in red), while patronage in the other groups was unaffected (for the purposes of this exercise it’s useful to hold mode shares in other age groups constant). The impact on overall travel demands is summarised in the table below, where the important figure is in the right-hand column, second row from the bottom. This shows that public transport’s share of all trips has increased from 25% to 31% as a result of this investment.
Let’s consider another scenario, where PT investment focuses on users in the 0-19 age-group. This might be, for example, a doubling in the level of concessions for these users and a focus on expanding frequency and span so that public transport meets more of their all-day, all-week travel demands. Let’s also say that this investment increases PT’s market share in the 0-19 age-group by 50%, i.e. from 50% to 75% (highlighted in red), while other groups are unaffected. The impacts on overall travel demands are shown below, where the important number is “32%”. So a 50% increase in PT market share for the 0-19 age-group generates slightly more PT trips than the 100% increase in PT market share for the 19-65 age-group.
But that’s not all of course. The second part of my argument observes that investments targeted at younger age groups may impact on their future decisions about how public transport can contribute to their quality of life. Thus, let’s consider a situation where the 0-19 age-group, which is already more favourable to PT, “expands” to include people in the next bracket, i.e. the 20-24 age-group. This is basically arguing that the PT investment made in the previous scenario “slows” the rate at which people transition from the “young” to “middle aged” age-group, at least in terms of their travel patterns. This might occur, for example, if people decide to delay getting their drivers’ license or buying a vehicle, or choosing to live somewhere that enables them to access their work via public transport once they leave university.
The impact slowing down this demographic “shift” is highlighted in red in the table below, where the change in age brackets changes increases PT market share further to 37%.
While this hyper-simplified quantitative framework essentially “proves” very little, it is useful for illustrating the differences between the two arguments posed above. A focus on investment that retains existing (in this case younger) users can generate considerable patronage growth, because it combines immediate patronage growth (i.e. an increase in mode share, as per Scenario 2a) with an expansion in the size of the population that views public transport as being relevant to their lives (i.e. an increase in the size of the age-group, as per scenario 2b).
These two somewhat distinct patronage impacts are illustrated by the red arrows in the chart below, where the vertical arrow indicates the increase in mode share and the horizontal arrow indicates an expansion in the age-group.
Ultimately, I think that the investment required to achieve the shifts shown in red above are likely to be easier than getting new people (i.e. the 90% of car users in the 19-65 age-group) onto public transport, mainly because people in the middle age-group will have already made many decisions, e.g. how many cars to own and where to live and work, which reduces the need for and effectiveness of PT. This in turn suggests that shifting these people onto PT, through for example investment in rail service, will be relatively expensive and inefficient because you are effectively fighting against major socio-economic decisions that people have already made.
Some of you might baulk at these conclusions.You could point out that increasing PT mode share from 50% to 75% for the 0-19 age-group is likely to be significantly harder than increasing PT mode share from 10% to 20% for the 19-65 age group (NB: These numbers are made up). Perhaps. It is certainly true that market saturation will kick in at some point, such that the size of the vertical red arrow in the figure would reduce and the role of the horizontal arrow would increase. But on the other hand I’d suggest that public transport in Auckland is far from saturation in any age-group, or indeed any market segment that you can think of. For example, while many young people use PT for travelling to/from school and university, the remainder of their travel demands are likely to be met by cars. Hence even though younger age-groups already use PT proportionally more, I suspect that there remains considerable room for further growth.
As an aside it seems to me that the best way to influence the travel choices of the 19-65 age-group is not by tailoring our public transport improvements to align with their preferences, but instead to change the way that we price and manage the use of private vehicles. This means targeting vehicle use directly through – for example through time-of-use pricing and parking pricing. Of course, in the process of improving the public transport service for existing users we are likely to attract some new users from outside our target markets, and these people definitely should be catered for as required. But they are a positive second-order patronage benefit, rather than the primary driver of investment in the public transport system.
In many ways this message is a relatively non-glamorous one of getting “back to basics”, i.e. placing a bigger emphasis on getting the existing network operating efficiently, rather than worrying too much about the technology being used to deliver it. On a simple level, thinking about public transport patronage this way encourages you to place a higher focus on existing users, rather than potential new users. More specifically, growing public transport patronage starts from having a detailed understanding of (and appreciation for) existing users. This alone can tell you a lot about the types of markets (and PT services) that are likely to generate the most patronage in the future. In my mind the “rail only” approach discussed above unintentionally demeans, or at least undervalues, the preferences of existing users – the majority of whom are using buses.
Before wrapping this up, I think it’s also worth mentioning that some aspects of this discussion are related to an earlier post on generational differences. That is, because most of our transport decision makers (including myself) fall into the 19-65 age-group there is a natural tendency for us to propose solutions that address our needs, rather than the needs of our users. This can result, for example, in a undue focus on high-speed services. For their part, PT users seem to not value speed – or more accurately “travel-time” – as much as other attributes, such as frequency, reliability, simplicity, and affordability. And in my experience relatively few people currently on the bus will complain that the bus is “not a train“.
None of this is to suggest that the attributes associated with rail-based solutions are irrelevant, or that investment in rail won’t attract new passengers – experience shows that it certainly will. But it is intended to highlight that the best way to grow our patronage is probably to focus less on large infrastructure projects and more on small to medium scale improvements to the existing network (whether it be rail, bus, or ferry). These will in turn help to generate more trips from existing users for more years. When you think about public transport from this perspective you realise that passenger retention equates to patronage growth; that the views of existing users, especially the young, are significantly more important than the views of non-users.
Having said all that, what would be my advice for the people at Auckland Transport and Auckland Council who are currently grappling with a slowdown in patronage growth, especially on rail? The first thing I would do, I think, is to get to know my existing customers better; learn to cherish their patronage and respect their choices; and – perhaps most importantly – consider initiatives that will encourage young people to keep using PT even as they get older. Some targeted market research into secondary and tertiary students who are currently using public transport, for example, might be a good place to start. Tertiary students that are about to graduate probably offer the most potential for future patronage growth.
I get the feeling that’s what a good business would do.