Thinking Methodically About Interstate Rail

Jan 30th, 2009 | Posted by

Over at The Transport Politic Yonah Freemark has a very comprehensive proposal for an interstate rail network that goes beyond merely figuring out which corridors should have HSR, and weighting these corridors using a specific methodology that prioritizes their funding and construction. Freemark’s goal is to bring some order and sense to the growing nationwide movement for improved intercity rail, and provide a reasoned way of determining what ought to be built when:

In order to evaluate the different lines, the transport politic developed a system by which it could examine the cost effectiveness of each line both in terms of travel within the corridor alone (the Corridor Score) and within the system as a whole (the Overall Score). Travel between every city pair in the system between 50 and 500 miles apart was evaluated, and the results were compiled by corridor, whereupon they were divided by route mile to appraise potential ridership by mile of new construction. The results provide the basis for prioritizing routes and suggest a method by which the federal government could begin to imagine how such a high-speed rail system might be developed. (PDF with description of methodology, evaluation of every city pair, and scores for each corridor or here)

According to this methodology the California HSR proposal scores less than some of the Northeastern and Midwestern corridors, and within the California HSR corridor Fresno-Sacramento and LA-San Diego score higher than the LA-SF corridor prioritized by Prop 1A. The analysis also shows that, while popular with some, the LA-Vegas corridor is relatively “marginal” in importance though still worth building.

As a thought exercise in how to plan implementation of an HSR network without resorting to which state has the most pull in Congress, it’s a very good discussion starter. But I have to question some of the assumptions that went into this. From the source data:

The equation is designed to allow for a simple comparison between different routes; while it does not calculate ridership, it provides a good estimate for which routes would be more or less used. It should not be taken as an exact formula or one that has been heavily researched, but it provides a good jumping-off point for more research on where to place new high-speed rail routes.

The calculation is based on the following assumptions:
•People are less likely to take a journey as the distance of the journey increases;
•Given a choice, more people would choose to travel to a bigger city than a smaller one;
•Travel choices are based entirely on city size and distance between cities;
•Density of metropolitan areas does not affect travel;
•There are no regional differences in travel preferences;
•No two city pairs compete with one another – as new routes open, new ridership is generated.

These assumptions are of course quite untrue in many ways. However, given time and data limitations, this formula provides a quick method of comparison.

I think the first assumption is questionable at best. LA-SF has a further distance than some other pairs, but compares well to Madrid-Barcelona, two cities with some similarity to LA and SF and with a similar distance. The new AVE line between the two cities has eaten deeply into air travel on what was one of Europe’s busiest air corridors, just as LA-SF is one of North America’s busiest air corridors. HSR compares well with door to door times using air travel on the corridor, and because it offers particular amenities that airplanes do not, it is likely that LA-SF HSR has some of the best growth potential in the country.

I know this site is inherently biased, but of all the corridors in America, there are few better suited to HSR than LA-SF. The Northeast Corridor surely outranks ours in most criteria, which is why it has a successful quasi-HSR system already in the Acela (and that deserves to be upgraded to full and true HSR). But I am hard pressed to think of many other corridors that make a more compelling case than our own.

The Transport Politic plan is a concept and a starting point of course, not a final plan, and so my comments are offered in a constructive and not critical or defensive manner. (Besides, CA is the only corridor that brings non-federal money to the table, guaranteeing us a privileged spot in terms of federal HSR funding – I don’t think we have much to worry about on that front.) We need to begin treating HSR like a national network, to be built in phases according to objective criteria. Measuring that criteria becomes key, and it will necessarily involve a complex set of factors.

So, go over to The Transport Politic and read the plan. What do you all think about it?

  1. thetransportpolitic
    Jan 31st, 2009 at 04:04

    Thanks for the helpful comments. I tend to agree with you – actually, in many ways, I would have liked to have ranked the LA-SF line as the first in my plan, but I tried a few equations to measure potential ridership, and couldn’t get it to come out on top! This isn’t to say that it isn’t the best line in the country with which to begin, it’s just to say that I couldn’t find a formula that shows that. (This is an inherent problem with any kind of empirical research… it requires making a lot of assumptions that may or may not be based on reality!)

    And, of course, as you said, California is the only state that’s actually putting in the money…

  2. Rafael
    Jan 31st, 2009 at 05:05

    I like the idea of a more comprehensive national plan based on formal analysis of ridership potential for two reasons:

    a) it re-establishes passenger rail as a strategic component of the nation’s infrastructure and,

    b) it provides a political basis for Congress to spend federal dollars by including virtually all major population centers.

    The effort is of course a first cut, lovingly prepared by someone in their free time. It is intended as an introduction of the concept, now that it’s becoming ever-clearer that major spending on infrastructure is both popular and necessary to get the economy out of its present deep funk.

    The electorate also appears to be ahead of politicians in terms of its willingness to diversify the transportation sector away from oil and toward domestic sources of primary energy. In practice, that means electricity because it can be generated in any number of ways and is easily distributed. It’s no accident that renewables and a new electric grid are top priorities for the Obama administration.

    Consumers have recently experienced that oil prices can rise rapidly and wreak havoc with the personal finances of millions of homeowners, triggering massive real or paper losses in their principal piggy banks. Meanwhile, big oil keeps raking in the profits and doling out the campaign contributions. I suspect this is the main reason why improving passenger rail is still largely a grassroots effort in the US, unlike most other countries. California and the NEC are the exceptions that prove the rule.

    However, the formula Yonah used to produce a first cut at ridership estimates and associated prioritization is misleading IMHO. For example, the following assumption is simply nonsensical:

    “Given a choice, more people would choose to travel to a bigger city than a smaller one”

    People need to go back home as well, so annual traffic volume will be virtually identical in both directions.

    Also, the formula only looks at pairs of cities. In e.g. California, the starter line will run from SF to Anaheim. Any meaningful ridership analysis has to take into account that riders will board the trains at multiple transit hubs at either end and even a couple in the middle. Rail lines can support a mix of express, semi-express and local trains, whereas most flights are between pairs of large cities.

    Finally, the ranking assumes that potential ridership is the only driver behind passenger rail. This disregards the substantial problems many cities have in expanding their highways and adding runways to their principal airports to accommodate future growth in population and affluence. Some, e.g. California, have simply run out of real estate and/or public acceptance of increased environmental burdens in key location. In other words, rail is a virtue reborn out of planning necessity.

  3. thetransportpolitic
    Jan 31st, 2009 at 05:57

    Rafael – in response to your points:

    1."Given a choice, more people would choose to travel to a bigger city than a smaller one…People need to go back home as well, so annual traffic volume will be virtually identical in both directions."

    >> The calculations I used are based on travel both ways between two cities. In other words, people in City X are more likely to travel to City Y than City Z if City Y has a larger population than City Z. This is may sound obvious but it is important because without this assumption, one might assume that people in City X are equally likely to travel to City Y and City Z, no matter their respective populations. The equation is based around two-way travel.

    2. "Also, the formula only looks at pairs of cities. In e.g. California, the starter line will run from SF to Anaheim. Any meaningful ridership analysis has to take into account that riders will board the trains at multiple transit hubs at either end and even a couple in the middle."

    >> The way I looked at it was thus: for the corridor from San Francisco to LA, I looked at SF-LA, but also SF-Fresno, SF-Bakersfield, SJ-LA, SJ-Fresno, SJ-Bakersfield, etc, etc. So the formula recognizes the idea of multiple destinations and origins within each corridor.

    3. "Finally, the ranking assumes that potential ridership is the only driver behind passenger rail. This disregards the substantial problems many cities have in expanding their highways and adding runways to their principal airports to accommodate future growth in population and affluence."

    >> You're absolutely right, though I don't have the resources or time to look at those issues for each corridor.

  4. Anonymous
    Jan 31st, 2009 at 07:56

    Would better source data than population be to use airline passenger data plus existing rail passenger data plus freeway usage data plus bus ridership between two points? There you have an ability to take total passenger mile trips between two destinations as well as a price spectrum to weight peoples preferences for speed versus price. This could allow you to look at the substitution that people would make for rail. The whole point of the HSR is to get people off planes and cars. I think the airline substitution is the most apt, since similar trips (business, weekend getaway, family visits) that people now make by plane could easily be replaced by HSR.

  5. Rafael
    Jan 31st, 2009 at 08:09

    @ thetransportpolitic –

    thank you for your clarifications, especially on how you computed the scores. I fully understand that you cannot possibly be expected to factor in everything a professional (i.e. paid) analyst would have to, but that doesn’t mean doing it on a shoestring will yields meaningful results.

    Your core message is excellent: creating a web of interconnected regional passenger rail services linking (almost) all of the country’s larger cities should be a national objective.

    IMHO, this message would be more powerful if you articulated why the current policy of propping up Amtrak plus strictly regional efforts in disjoint corridors is insufficient.

    The argument that we need to keep up with the Joneses is well taken but not all that strong. The geography and population density of the US and Canada (why exclude Mexico?) is different from that of Europe, Japan and China. Surely, the case for reviving high-quality passenger rail service in North America should rest on economic, environmental and strategic policy objectives specific to the continent.

    Also, many people still associate “intercity passenger rail” with interstate Amtrak trains that take days to reach their destination, often with long delays. Your map appears to suggest more of the same, i.e. routes of several thousand miles, yet your text (appropriately) emphasizes trips of 50-500 miles. More detail on your ideas regarding how the proposed infrastructure would be used, e.g. typical routes, transfers, punctuality, sleeper/auto trains etc. would be useful.

    Finally, you may want to address if creating new or upgraded intercity passenger rail links should be prioritized over highway and airport maintenance and/or expansion.

    There’s only so much money to go around, so should any part of the interstate system be abandoned rather than fixed?

    Should the EIR/EIS rules for adding regular or carpool lanes to freeway medians include an analysis of constructing a passenger and/or freight rail line instead?

    Should secondary airports in cities served by rapid rail/HSR be closed to free up land for transit-oriented development?

    Would it be counter-productive if rail freight lost market share to road freight on account of the new passenger rail services? If so, how do you propose that narrow rights of way in built-up areas be shared, given the very different business models and infrastructure needs of freight and passenger rail?

    And perhaps most importantly: should intercity passenger rail be expected to service the debt for its infrastructure through fares, to cover only operational expenses or, should both construction and operations be subsidized as a public service? The answer determines to what extent, if any, the private sector should be involved.

  6. mike
    Jan 31st, 2009 at 09:41

    The first assumption sounds plausible but in fact is not. For all modes of transport combined it makes sense, but that’s not what we’re looking at; we’re looking at just HSR.

    For true HSR, the optimal trip distance is probably between 175 to 400 miles. Shorter than that, its speed advantage over driving doesn’t always compensate for the access/egress time for getting to the station. Longer than that, it will not always be faster than flying.

    In other words, total trips should decrease with distance, but HSR market share can increase with distance.

  7. arcady
    Jan 31st, 2009 at 10:29

    Since we’re making grand plans, I can’t help but wonder if it would be feasible to have HSR night trains. That way, the feasible journey time becomes 8 to 11 hours, and it’s still more convenient than flying since you’re not wasting much valuable day time on travelling. Of course, it also means that your high speed lines are longer and require more investment, so I think this sort of service works best when you have an existing HSR line for part of the way, and upgraded conventional line for the rest. You’d have, for example, trains from Boston and New York to Atlanta and Chicago. It might even be possible to link the Bay Area and Portland. This sort of thing hasn’t really been done in Europe or Japan, but China has recently introduced a high speed (155 mph) night train between Beijing and Shanghai.

  8. Alon Levy
    Jan 31st, 2009 at 10:55

    I think a lot of the corridors in the map are just too weird, like Toronto to Montreal via both Ottawa and Kingston, or New York to Albany with intermediate stops only in Poughkeepsie and Kingston but none in Westchester.

    To say nothing that you could get completely different results by applying different formulas – for example, measuring HSR demand by the size of the air travel market instead of distance.

  9. Jim
    Jan 31st, 2009 at 11:15

    It’s unfortunate that the US is no longer a place where we can pull off something similar to the interstate system. The country is too socially. politically and ideologically fractured for such a national project. It will happen very slowly though, with state and regional plans and gradual piece by piece upgrading of Amtrak. In many parts of the country this comment from the Dallas Morning News sums up the attitude.. “‘It gives virtually free transportation to a few people at the expense of the rest of us, while providing zero benefit to the non-riders.'” It’s a different country out there when you leave california.

  10. arcady
    Jan 31st, 2009 at 11:28

    “It gives virtually free transportation to a few people at the expense of the rest of us, while providing zero benefit to the non-drivers.” Are they talking about the Interstate Highway System?

  11. yeson1a
    Jan 31st, 2009 at 12:28

    I agree with Jim..people since the early Regan era have looked at anything thats a large public works as a waste stealing their money and no direct use to them.Lets hope that our HSR system will go along way to changing that way outdated mindset

  12. Rafael
    Jan 31st, 2009 at 12:33

    @ arcady –

    China Railways just launched a bullet train service with
    sleeper cars.

    @ Jim –

    to paraphrase Donald Rumsfeld: you’re talking about the old America.

    Cheer up, the US is only just beginning to re-invent itself as a “post-superpower”. I’m not quite sure what that means yet, but in spite of the dire economic situation it’s almost certain to be a massive improvement over the past eight years.

    All the country needs to do is sharply scale back the military-industrial complex to free up human and financial resources for civil engineering projects. The Cold War is over and its US-centric institutions need to be replaced by a single new one for defending the free world, with each member nation or supranational body contributing both leadership and burden in rough proportion to its economic and other capabilities.

  13. BruceMcF
    Jan 31st, 2009 at 16:22

    A difference between California and the Great Lakes States is that the Great Lakes States could gain tremendous benefit from Rapid Rail … indeed, the top shelf options of the Midwest Hub and Ohio Hub proposals are Rapid Rail systems.

    IOW, how many 1m+ metro areas are within 300 miles of metro Chicago … and how many 1m+ cities are within 300 miles of the Bay or the LA Basin.

  14. Jim
    Jan 31st, 2009 at 18:26

    “All the country needs to do is sharply scale back the military-industrial complex to free up human and financial resources” he he no one wants rail and high speed rail more than me since it’s my job, but this above statement isn’t going to happen. AGIn, americans are way beyond the days of the 30 40s and 50s. or even the 70s, which was the last decade where there was any sense of people getting along or being on the same page. No one wants those days back more than I but they will never come back.
    America post 80’s is a “everyone for themselves/Ive got mine/ what’s in it it for me/ how much will it cost/” kinda place now. Sure, for the moment we’re on the post bush high, but that is going to last about 10 minutes. The most we can hope for is to bring california into the 21st century. I wouldn’t waste my time worrying about the other 49.

  15. Jim
    Jan 31st, 2009 at 18:31

    or as it pertains to cali vs the US there’s the old saying – “you can lead a horse to water but you can’t make him drink” and the US is one lame horse… (secession now!!!)

  16. 無名 – wu ming
    Feb 2nd, 2009 at 03:58

    the US was also divided in many ways back when the interstate system was built. moreso, when you think about it, given the centralizing and unifying effects of the interstate highways and television in the 50s.

    i’m surprised that the sac-bay area corridor, one of the most successful amtrak ones, doesn’t rate much. I-80 is packed most of the time, and the amtrak line there is to capacity increasingly, every time they expand the number of trains.

  17. TomW
    Feb 2nd, 2009 at 11:49

    The first assumpion (travel demand decreases with distance) is a valid one. Put it another way: if LA and SF were on oposite sides of the continent, would there be as many trips between them? I think not.
    You seem to have mixed up “lwoer demand” with “low demand”

Comments are closed.