A Tale of Two Ridership Forecasts

Just a comment, a comparison of two forecasts.. The California HSR website says:
"Questions & Answers


How many riders are expected and how many passengers are forecast to be diverted from airplanes and cars? The most recent ridership forecasts for the California High-Speed Train Project estimate between 88 – 117 million passengers annually by 2030 for the entire 800-mile high-speed train network connecting Sacramento, the San Francisco Bay Area, Central Valley, Los Angeles, Orange County, the Inland Empire, and San Diego. Of the 33 million air trips forecast to be made in the year 2030, over a third or 12 million would be attracted to high-speed trains, bringing the level of air traffic in the state back to the levels of 2000, slightly higher than it is today. In other words most of the growth in air traffic would be diverted, leaving airport capacity for international and out-of-state flights."

In contrast:
The Amtrak concept plan, A Vision for High-Speed Rail in the Northeast Corridor (NEC) (pdf), shows a financially viable route could be developed. Upon its full build-out in 2040, high-speed train ridership would approach 18 million passengers with room to accommodate up to 80 million annually as demand increases in the years and decades that follow.
Somehow, I don't think Amtrak's Northeast Corridor in 2040 will draw 70 - 100 million fewer riders than California's HSR in 2030 (even if California's system is somewhat longer). Given Amtrak has no incentive to underestimate, this suggests some issues with California's forecasts. Can you say "Strategic Misrepresentation", good, I knew you could. See: Curbing Optimism Bias and Strategic Misrepresentation in Planning: Reference Class Forecasting in Practice.


Your comparison isn't fair, and it doesn't reflect what either of the studies suggest.

One, Amtrak's study predicts total intercity rail ridership of between 37.5 and 51.5 million on the Northeast Corridor by 2050. Just because all of these riders won't be on "Next-Gen" high-speed trains doesn't mean they'll be taking trips much different from those being offered on the California system.

Second, the California system is being designed to serve commuting trips within metropolitan areas as well as intercity travel -- it will significantly expand the number of people taking the train between San Jose and San Francisco; between Los Angeles and Orange County, etc. If you take that into account, it's worth looking at all the rail passengers in the Northeast Corridor that are taking other rail systems along the same routes as Amtrak. Today, commuter rail systems along the NEC (including Boston's MBTA, New Haven's Shore Line East, New York's Metro-North and NJT, Philadelphia's SEPTA, and Baltimore's MARC transport about 200 million riders a year (http://www.apta.com/resources/statistics/Documents/Ridership/2010_q2_ridership_APTA.pdf), a significant percentage of whom are using the same tracks as Amtrak's. California's 88-117 million doesn't look so unreasonable in that context.

That doesn't really wash. HSR has a limited number of stations, Commuter rail stations are frequent. You are now essentially making the claim that CAHSR's forecast includes CalTrain etc. Sure there may be some trips between downtown San Jose and downtown San Francisco served by CAHSR, just as there will be some trips from downtown Baltimore to downtown DC (a comparable distance) by Amtrak's new system. Similarly someone might even use the system to go from BWI to either city, just as they may go from SFO. But where does CAHSR say it is taking over the commuter rail system in its forecasts? MARC et al. have lots of stations between Baltimore and Washington (see http://mta.maryland.gov/services/marc/schedulesSystemMaps/marcTrainSystemMap.cfm e.g.), HSR will have one (BWI). Similarly CalTrain vs. CAHSR. They may share tracks (my understanding is this is not resolved), but you can't compare the riderships. Face it, the forecasts are unreasonable.

I think ridership outcomes depend a lot on how CAHSR markets and serves its routes. You're right that Amtrak serves many equivalent services as those of the commuter rail operations, but it does so at far higher price per passenger and with little time savings. In the Northeast Corridor, the commuter railroads take the large majority of commuting trips from cities far away enough that they arguably should be served by intercity rail rather than commuter rail -- New Jersey Transit's most-used stops on the NEC are Princeton (48.4 miles from Penn Station-NYC), Metropark (24.6 miles), Trenton (58.1 miles), and New Brunswick (32.7 miles). To see why people are choosing commuter rail over Amtrak, take a trip between Trenton and New York. On NJ Transit, the trip takes 1h08 and costs $15.50, day of purchase. On Amtrak, the trip takes 0h50 to 0h58 and costs $28.00 -- a month's purchase in advance (day-of, it's more like $63.00).

For comparison's sake, Anaheim is 29 miles from L.A.; Escondido is 32 miles from San Diego; Redwood City is 26 miles from San Francisco. Currently, Coaster provides 1h00 service between Escondido and San Diego, for example, whereas CAHSR expects to be able to offer 0h23 service, a serious improvement. Travel between cities like these are predicted to be large markets for CAHSR, which is clearly banking on (a) Getting a lot of people to give up their cars in favor of the train (now much faster) and (b) Switching some customers out of the existing commuter trains and into intercity trains.

And indeed, perhaps commuters would switch from NJ Transit to Amtrak if they were offered a much faster, slightly more expensive option. Amtrak's ridership estimates could be raised substantially if the company were able to offer cheaper and faster trips than currently -- so perhaps it is underestimating potential ridership.

On the other hand, I don't know if these ridership predictions are unreasonable -- I'm not a forecaster -- But I don't think the evidence is there right now to portray CAHSR's estimates as "unreasonable."

David Levinson

Network Reliability in Practice

Evolving Transportation Networks

Place and Plexus

The Transportation Experience

Access to Destinations

Assessing the Benefits and Costs of Intelligent Transportation Systems

Financing Transportation Networks

View David Levinson's profile on LinkedIn

Subscribe to RSS headline updates from:

About this Entry

This page contains a single entry by David Levinson published on October 2, 2010 9:47 AM.

Google CEO on Cars ... was the previous entry in this blog.

How to obtain truth in forecasting ... is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.


Monthly Archives


Powered by Movable Type 4.31-en