The Linking Melbourne Authority (LMA) has released publicly a swag of technical material in support of its case for building the East-West Link motorway in inner suburban Melbourne.
There’s a lot there but unfortunately the key documents that most observers want to see haven’t been released. There’s no sign of either the full business case (we’ve previously seen a “short-form” version e.g. see here and here) or the detailed traffic modelling work done by Veitch Lister Consulting (VLC).
We do however get to see the results of the forecasting VLC’s done for the LMA (Appendix E). That’s noteworthy in its own right, but it’s especially interesting because it can be compared with the modelling done by Public Transport Victoria (PTV) for its long-run vision for the city’s rail system.
Both modelling exercises should broadly concur on the current level of total motorised travel in metropolitan Melbourne. They should also generally agree on the future level of travel in the absence of any interventions (like the East-West Link or the Melbourne Metro).
The LMA’s modelling counts a total of 12.2 million trips across the metropolitan area by all motorised modes on a weekday in the base year (2011) and the PTV’s exercise puts it at 12.9 million. The respective figures for 2031 are 16.4 million and 17 million.
There’s a potentially more important difference in the forecast growth of public transport patronage.
The LMA’s modelling indicates public transport patronage across the metropolitan area will increase by 75% over 2011-2031. That’s more than double the 30% increase expected in the number of light and heavy vehicle trips, albeit from a smaller base.
The PTV’s modelling however predicts a much larger 105% increase in public transport patronage and only a 21% increase in vehicle trips. It expects public transport to capture 19.5% of all metropolitan motorised trips in 2031 versus LMA’s estimate of 14.7%.
The hands-on work for PTV was done using the VITM model and the LMA’s was done by VLC using its Zenith model. This is a firm that enjoys a well-earned reputation for the technical quality of its work.
Get Crikey FREE to your inbox every weekday morning with the Crikey Worm.
Perhaps there’s a purely technical explanation for the difference in modelled mode shares, but if there is it isn’t explained in the publicly available documents (1).
I expect much of the difference can be explained by the fact that the modelling exercises were done for separate organisations with different missions. One wants to build roads, especially the East-West Link; the other wants to build public transport, especially the Metro rail line. I expect both are very committed to what they do.
The output of any model will depend on the assumptions underlying the data fed into it. These aren’t ‘black and white’ issues; they’re essentially matters of opinion. It’s likely these two organisations have quite different views at the margin on what are or aren’t reasonable assumptions.
They both report to the same Minister, Terry Mulder, who’s Minister for Public Transport as well as Minister for Roads. Even so, I expect the ‘public transport’ version of the model is substantially different to the ‘East-West Link’ version of the model.
Whether or not one is better than the other is impossible to say without further information. Nor is it possible to say if the business case for the East-West Link is critically dependent on a less optimistic view of the outlook for public transport.
The answer to these and many other questions would be clearer if the Government released the full business case for the East-West Link, as well as VLC’s modelling. It’s wonderful to have lots of documents available publicly, but what’s really needed for informed public consideration is the salient information.
- A “technical explanation” might be that the LMA appears to define public transport mode share by trips whereas PTV uses boardings. If so, it leaves unanswered the question of why they’d use different definitions. In any event, although boardings are expected to grow faster than trips, this still wouldn’t explain most of the difference in the growth of public transport patronage between the two modelling exercises.