Reviewing the Peer Review Report on Ridership
There are two discussions that happen regarding ridership on the California high speed rail project. One is absurd and one is necessary. The absurd discussion is the one that dominates conversation and media coverage. It revolves around the question “will anyone ride the trains?” And the necessary discussion is the one that is generally ignored, revolving around the question “what are the exact details of who will ride the trains when and at what cost, and what impact does this have on financing and operations?”
The “will anyone ride?” discussion is absurd because there is no evidence – none whatsoever – to indicate the California HSR trains will go empty. Instead, the evidence indicates, quite clearly, that the trains will have high ridership. A quick review of that evidence is in order:
• Poll shows 79% of American travelers would use HSR if it was an option
• California intercity passenger trains – the closest example the state has to the HSR system – have high ridership and are quite popular
• Californians have shown they will find other ways to get around if freeways are not a viable option
• The Acela, the USA’s only HSR-like system, is profitable and has high ridership
• HSR systems are generally considered viable if they serve a market between 300 and 500 miles apart. Madrid and Barcelona are 386 miles apart via the AVE, and SF and LA would be about 420 miles apart.
• California’s demography compares favorably to other HSR routes both in terms of raw population (SF and LA are both much larger than Madrid and Barcelona, at least in terms of metro area population) and in terms of population density.
• Gas prices are going to keep rising for the foreseeable future
• A great shift away from driving is under way, with digital devices, traffic congestion, and high costs of car ownership and gas driving people to choose alternatives
• A warming planet will eventually force a shift away from burning fossil fuels and toward more efficient forms of travel.
Considering the above, arguing that Californians won’t ride the bullet trains is equivalent to claiming the earth is flat or that global warming is a lie – it’s a claim that flies in the face of the available evidence.
The usual argument against trains, that Californians somehow have a cultural affinity for driving, also ignores the above evidence as well as the fact that the “cultural affinity” is actually the result of deliberate policy choices in the second half of the 20th century to limit alternatives to driving.
Even still, transit proposals remain popular. Richard Mlynarik can rage against BART all he wants, but Bay Area voters still love it. Over 2/3rds of LA County voters agreed to tax themselves to expand their rail system.
The reason that absurd discussion matters is because ALL discussion of HSR ridership happens in the context of the question of “will anyone ride the trains at all?” That is particularly true of the media’s coverage of the Independent Peer Review of the California High-Speed Rail Ridership and Revenue Forecasting Process.
That review did not at all question whether the trains would have high ridership. Instead it asks technical questions about the precise nature of the projections used, which are relevant because the financing model used is dependent on precise projections. Even in a scenario of “high ridership” with packed trains by 2030 – which we should expect based on the evidence I listed above – that ridership might not be enough to cover the financing if the price points aren’t correct, or if there’s more money borrowed than even high ridership can cover. (Which, after all, was the Taiwan problem).
So what does the Peer Review actually say? Their findings can be summed up as follows: The Cambridge Systematics study was pretty good, since nobody else has ever really done an HSR study in the USA, but even so there are some specific things they ought to improve to get the most reliable numbers possible. The report does NOT claim the HSR ridership numbers done by CS are invalid, does NOT claim the system will fail to attract riders, and does NOT suggest the project is a boondoggle. Instead it is a highly technical document making highly technical suggestions within the framework of assuming the HSR project is still generally a good idea.
Their findings fall into several overall categories:
More documentation needed. The panel wanted to see evidence of the reasoning behind some key assumptions, including levels of service on the system, at airports, fuel prices, and such.
Update models to match current practices in transportation planning. Because some of the HSR ridership projections are based in studies conducted in 2000, it doesn’t always reflect the most current thinking.
Make the models more conservative. There’s a long discussion in Section 4 about variable and constraints used in the 2005 study that warrant modification.
Over the long-term, try to remove Stated Preference bias via several methodological changes. Specifically, they suggested doing a study in the LA-SD and SF-Sacramento corridors, where intercity rail is already heavily used, to help deal with this problem.
And their conclusion:
The current model system represents an ambitious step towards defining the best practice in North America, replacing ad hoc and closed proprietary models used in many previous HSR feasibility studies. In many ways the model is generally well founded and implemented. How- ever, in order to have full confidence in it the issues identified in Section 4 must be addressed quickly. Moreover, the incomplete, unclear, or out-of-date elements of the documentation discussed in Section 3 must be completed as part of the short-term actions. Once these issues are addressed the Panel will be in a position to make a more definitive determination about the model and forecasts derived from it.
Another thing worth noting, this from page 6 (section 4.2):
The Panel found no evidence that these results are biased in aggregate or that any differences are in a particular direction as a consequence, but believes it is a relatively simple improvement that will make the model more reliable.
Overall the panel’s suggestions appear reasonable, especially from the perspective of getting the most accurate projections possible based on the evidence we have before us. I don’t see anything in their recommendations that is unreasonable.
And more importantly, I don’t see anything in here to suggest using the review to make an attack on the HSR project itself. Of course, I fully expect people like Senator Alan Lowenthal, oil industry ally Gary Patton, and other HSR opponents to do exactly that – to claim that “omg the Peer Review found that the ridership numbers are flawed, and therefore the project is DOOMED!”
I fully expect the media to also make the same error. Instead of treating this like the technical debate between experts that this actually is, I would predict we see more articles like the LA Times published, which titled its article on the review as Report casts doubt on forecasts for California high speed rail. The most accurate title would be “Report disagrees with several assumptions made and recommends changes in forecast for California high speed rail,” but that’s boring and god forbid readers be treated to anything boring, even if it’s also accurate.
This is a high-level overview of the report; feel free to dig into the details in the comments.