With the country now at Level 1 and ‘back to normal’ it means all of the restrictions have now lifted. That means that many more people will soon be heading back to the office. Universities will likely take a little longer and not be back till the start of the next semester.
As I wrote last week, the big question now is how long it will take for public transport demand to get back to some semblance of normalcy. Given the trends we’ve seen so far, I’m fairly optimistic this will happen relatively quickly. For example last week the four working days saw an average of 179k boardings a day. That’s about 52% of what usage was at the same time last year and is up from about 154k trips the week before. It’s also not too bad when you consider that physical distancing requirements had reduced capacity by about 50%.
But a return to normalcy also means we need Auckland Transport to get back to focusing on how to get a lot more people to use public transport more often. Doing so is needed to play an important contribution in reducing emissions and improving urban mobility.
To get more people using PT there are many things we need to do. These range from small improvements and tweaks to improve the customer experience to fare policy to large infrastructure projects to improve the network.
Many of those large infrastructure projects are already underway and in the past, particularly around the time when Auckland Transport raise fares, we’ve often seen debates about fare policy, including the suggestion of free fares.
To test the impact of large interventions we worked with the Auckland Forecasting Centre to model a range of scenarios that allow us to see the impact April 2020 of i) planned and proposed infrastructure improvements to the network ii) more radical fare policy changes, and iii) a combination of the two. In all cases, the counterfactual is that no changes are made to either today’s PT infrastructure or policy.
It’s worth noting that this is a model – it won’t be a perfect reflection of reality – but is a reasonable guide for the scale of likely impacts.
For the network improvement option, they include the City Rail Link, the Eastern Busway, more bus lanes around the city and Light Rail to the airport and to Westgate. This largely sounds like the network that is expected to be delivered as part of the current ATAP.
What I find strange about this scenario is that they estimate ridership would only increase by 22% over what would occur without them. It’s not clear what timeframe this analysis covers but it doesn’t really pass the sniff test. There were just over 103 million boardings on the network in the 12-months to the end of February and growth had been slowing – you can see on the graph that this year prior to COVID ridership had largely been tracking last year.
The City Rail Link alone is expected to roughly double usage of the rail network within a few years of opening to around from 22 million to 50 million trips. That alone could potentially deliver the 22% increase but add in those other projects and if we only achieved a 22% growth, that would be a significant underachievement. Also of note, ATAP suggests usage will rise to 170 million by 2028, about an 83% increase on what it was in 2018.
By comparison, if instead of improving the network, we made fares free, the model suggests we’d seen an increase of 23% in ridership. The problem with this option is there simply wouldn’t be the capacity on existing services to accommodate that kind of increase unless it all occurred off-peak. While fares will play a part for some, for most PT is simply slower and less convinent and so addressing that is key. This means network improvements, like those above, would need to be made.
Doing both network improvements and free fares is expected to deliver a 54% increase in ridership.
Translating these changes to changes in emissions and car trips they estimate:
- the improved PT network would only reduce car trips by about 1.2% but have almost no impact on emissions as the gains would be swallowed by induced demand.
- free fares would have a greater impact with a 4% reduction in car trips and a 3% in emissions.
- both options combined see car trips drop by 6% with about a 4% drop in emissions.
They don’t explain why, under their model, similar amounts of increased ridership generates quite different levels of car use and emissions. I also suspect these outcomes will be heavily skewed by the earlier issue on the impact on usage but th
After putting some high-level costs and benefits around the the free fares proposal they note.
These scenarios reveal a few things that can help inform the conversation on good transport policy. At a high level, we learn that emissions reductions alone are unable to justify a policy as radical as free PT, with a value of the reduction in the tens of millions of dollars per year but a likely cost in the hundreds of millions of dollars.
This does not imply that reducing emissions is a bad thing, but rather points out that emissions cannot make the case for free fares on their own. Naturally, there are other important reasons to consider making PT free – congestion, social equity, better access to jobs, and a nudge toward sustainable transport – but as always, the value of these impacts needs to be weighed against the costs.
We also learn that PT users respond to both changes in service quality and changes in fares. Put simply, this is because fares are not the only or even necessarily the dominant cost in someone’s decision to use the service. The other costs of travelling are largely time costs – waiting, delays, transfers, headways, and total in-vehicle travel time – costs that an improved network can lower. The modelled results suggest that fare reductions and the service quality improvements we looked at are equally important to PT users.
It is important in all this thinking to remember that “bums on seats” is not the goal. PT is a means to achieving the other outcomes we have highlighted. One further implication of the analysis is that improved PT itself won’t eliminate congestion and emissions. Something that cuts more to the behaviour of motorists, such as congestion charging, is still likely to be needed.
A separate presentation I was provided breaks down the impacts across a number of other metrics
And the change based on the type of trip
Overall it is an interesting report but there does seem to be some flaws with it. Furthermore, it doesn’t line up with other research we’ve seen that suggests frequency, reliability and speed are larger drivers of PT us than price. I even recall seeing a chart showing this from Auckland Transport a few years ago but have never been able to find it since. They did send me the results of a a survey in December of nearly 13k customers though showing the drivers for current PT users.
It also includes these charts showing how people ranked the attributes of each mode. It does suggest value for money is important but as per above, value for money could just mean it’s cheaper than the alternative.