dueling academics on cost-benefit of rail

Two new academic abstracts set up a striking debate, on the question of the cost-benefit analysis of rail projects.  From Peter Gordon and Paige Elise Kolesar:

Rail transit systems in modern American cities typically underperform. In light of high costs and low ridership, the cost-benefit results have been poor. But advocates often suggest that external (non-rider) benefits could soften these conclusions. In this paper we include recently published estimates of such non-rider benefits in the cost-benefit analysis. Adding these to recently published data for costs and ridership, we examine 34 post-World War II U.S. rail transit systems (8 commuter rail, 6 heavy rail and 20 light rail). The inclusion of the non-rider benefits does not change the negative assessment. In fact, sensitivity analyses that double the estimated non-rider benefits and/or double transit ridership also leave us with poor performance readings. Advocates who suggest that there are still other benefits that we have not included (always a possibility) have a high hurdle to clear.

Meanwhile, from Robert Cervero and Erick Guerra, both of UC Berkeley:

The debate over the costs and benefits of rail passenger transit is lively, deep, and often ideological. As with most polemical debates, the truth probably lies somewhere in the middle of extreme views. Some rail systems have benefits that outweigh their costs, while others do not. Applying a commonly used transit-fare price elasticity to 24 of the largest light and heavy rail systems in the United States and Puerto Rico, assuming a linear demand curve, and accounting for a counterfactual scenario, we find that just over half of the systems have net social benefits. Although Los Angeles’ rail system does not “pass” our back-of-the-envelope cost—benefit analysis, as the network expands, it will begin to mimic the regional spatial coverage and connectivity of its chief competitor—the auto-freeway system—and approach the fare recovery rates of other large, dense American cities.

I don't have time to dig into the papers, but I think the difference in tone of the abstracts is interesting in itself.  The issue of network completeness identified by the second abstract is of course critical.  Discuss.

(Hat tip: Murray Henman.  Post composed via wifi on the Amtrak Cascades, which is excellent!)

 

 

34 Responses to dueling academics on cost-benefit of rail

  1. annon August 31, 2011 at 5:37 am #

    The same issue also has a commentary on the debate by Lisa Schweitzer of USC
    Lisa Schweitzer
    Benefit-Cost Analysis of Rail Projects: A Commentary
    Public Works Management & Policy April 2011 16: 129-131, doi:10.1177/1087724X11401035

  2. teme August 31, 2011 at 7:23 am #

    I’d say it is bit beside the point: In European context light rail tends to be a good investment, and this is whre most pro side examples come form, in the States not so much. Now you can dwell deep into geographic, cultural, socioeconomical difference and do complex statistical analysis. Or you could just look up the costs of the projects and find that European light rail projects tend to be up to ten times cheaper to build.

  3. Dan W. in Los Angeles, California, USA August 31, 2011 at 10:25 am #

    I was not surprised to see Prof. Gordon was from USC. Whenever the Los Angeles Times wants an anti-transit quote from an “expert” for an article they grab someone from USC.
    The UC schools tend to be much more pro-transit in philosophy.
    Whether that is because one is a private university and one is a public university is I guess an interesting hypothesis.

  4. TransitPlannerMunich August 31, 2011 at 10:32 am #

    I agree with teme. The question is already answered, at least regarding Europe. Why rail projects must be so expensive in the US, that is something I do not understand either. In Spain or Germany the kilometer light rail comes for 10 million Euros or below, total costs including landscaping, power systems etc. – and per km subway 80 million Euros are realistic, including stations.
    But anyways, I guess the US prefers to spend their money more on wars, weapons and tax breaks for the rich, that is certainly more efficient regarding cost benefits.

  5. cph August 31, 2011 at 12:16 pm #

    What about costs such as EIR’s, consultants, etc? I don’t know European practice or how much it adds to costs of projects in the US, but it may be a factor….
    Is there any way to get the budgets of some of these European projects and compare them with US budgets?
    Are construction costs that much more in the US? Is there such a difference in constrcution costs for other projects, say a high-rise building?

  6. TransitPlannerMunich August 31, 2011 at 12:26 pm #

    @cph: I guess the costs of consultants are the critical factor. For example I wonder how much must be paid to Jarrett so that he would advocate a rail project… 🙂

  7. Jarrett at HumanTransit.org August 31, 2011 at 12:33 pm #

    TPM.  My job is to help a client identify projects and strategies that advance their own stated goals, not mine.  Be very careful, even in jest, about accusing "consultants" in general of being paid to say things they don't believe.  Some consultants do that.  I don't.

  8. Yuri August 31, 2011 at 12:39 pm #

    Peter Gordon and others in the Transportation Department at USC have repeatedly argued against light rail for years. Ironically, the new Exposition Line goes right by the campus, literally thumbing its nose at them. This line might eventually have the highest ridership in the Los Angeles light rail system, beating even the Blue Line which has the second highest ridership in the country (about 90,000 per weekday). When it opened in 1990, the Blue Line was predicted to have a low ridership of 5,000 per day. These academics conveniently ignore this great success and deny the possibility of any future successes in order to keep promulgating their anti-rail bias.

  9. Jeffrey Jakucyk August 31, 2011 at 1:17 pm #

    While the cost-benefit of rail projects might not look so good, the same analysis must be done on road and highway projects too. One can’t say that rail projects shouldn’t be built because their ratio is bad, but then ignore potentially worse ratios for roads or other projects.

  10. Zoltán August 31, 2011 at 1:29 pm #

    @Jeffrey Jakucyk
    Assuming that mobility is necessary, and it’s been decided that mobility must be expanded in one way or another, then bearing in mind the many disbenefits of road projects, it may well be the case that relatively poorly-performing rail is the least bad option.

  11. Zoltán August 31, 2011 at 2:08 pm #

    I’d agree that the network effect is entirely crucial. Any project built as one stage in a larger aim of creating a city that is considerably less dependant on automobiles, and that has a transit system that provides good levels of citywide mobility, will gain much of its value long after it’s built, when far more connections are available, and hence it is part of far more through trips by rail, or by rail plus good bus transit.
    I generally reckon that rail lines are a bad priority for cities that can’t put together a decent comprehensive bus network, for precisely that reason – the rail line won’t benefit from connection to a good wider transit network any time soon.

  12. Eric Doherty August 31, 2011 at 6:17 pm #

    If network connectivity is a crucial factor, then creating a low-capital cost rapid transit network using existing rails and roads may be the way forward for automobile dominated cities.
    I think Ottawa provides a good example of this, their great success in building ridership was done with a network of temporary bus lanes. Later, the expensive and largely grade separated bus rapid transit lines built on the success of the cheap and simple.
    (Yes I know there is a lot more to this story, including a dominant employer who started charging for parking while reducing parking supply.)
    A can of paint and a stencil that says “bus lane” can achieve a lot. So can a train on an existing track. This is not about bus vs rail, it is about networks and good policy.

  13. Nathanael August 31, 2011 at 7:25 pm #

    Suspected the USC people were simply biased; good to know they have a track record of bias. I bet they haven’t used their cost-benefit methodology on anything but rail, have they? If so, they’ll probably find that transportation is never worth it, and we’ll see exactly what’s wrong with their computation of benefits.

  14. Danny September 1, 2011 at 4:56 am #

    There is a huge culture problem that is involved here. Rail, being an inherently more politically involved process due to its higher initial investment and construction disruptions, has been hijacked by our terrible government culture. It doesn’t have to be expensive or disruptive with a low benefit/cost ratio, as we have seen in other countries (including our neighbors to the north).
    It is a culture problem though, because we make it that way with the way we perceive government. You see, in countries where the political debate is often framed as differing opinions on what constitutes good government, political sides can come to a consensus when certain actions are obvious improvements.
    Not us. We frame the debate as government vs no government. There is no room for compromise in a binary argument. And the result is terrible governance because, just like the private market, that is the kind of performance you get when job security is subject to the political whims of people who either hate you or try to take advantage of you (vote for me!).

  15. Alan Kandel September 1, 2011 at 9:24 am #

    Here’s a citation from the book: “The New Transit Town – Best Practices in Transit-Oriented Development.”
    “…An analysis by David Wohlwill of the Port Authority of Allegheny County found that fifty-four development projects valued at $302 million had occurred along Pittsburgh’s four busways. The study, like many looking at development along rail lines, did not differentiate between development attracted by the transit line and development that would have occurred anyway. A similar study of the economic impact of the Dallas light-rail system found that ‘Between 1997 and 2001, the mean value of 47 office properties near DART increased 24.7%, compared with an increase of 11.5% for 121 properties not near the stations, giving the DART office buildings a 53% advantage.'”
    Authors Hank Dittmar and Shelley Poticha sum it up this way: “The right answer to the bus versus light-rail debate is this: It depends on what the region wants to accomplish.”
    (Source: “Defining Transit-Oriented Development: The New Regional Building Block,” Dittmar, Hank and Poticha, Shelley, “The New Transit Town – Best Practices in Transit-Oriented Development,” edited by Hank Dittmar and Gloria Ohland, 2004, p. 37).
    I think Dittmar and Poticha might just be on to something here.

  16. EngineerScotty September 1, 2011 at 10:56 am #

    Danny,
    I think a big issue with rail is this: Because it attracts such vehement opposition (including opposition not on merits, such as from automobile interests looking to exclude competition and from cultural conservatives who hate urban demographics and environmentalists–don’t forget that in many parts of the country where there has been significant economic destitution due to outsourcing, many blame the environmental movement for allegedly “shutting down” industry), rail projects need to assemble fragile political coalitions to succeed; even in liberal places like Portland. And when you have to do that, there’s more fingers in the pie. Compare this with roadbuilding in the US, which tends not to be controversial; the US is very efficient at pouring concrete.

  17. Lauri Kangas September 1, 2011 at 12:42 pm #

    Property values are an interesting subject. In a Finnish whole society cost benefit calculation they are not benefits. Society as whole does not benefit from properly value increases. Money is just transferred from one party to another, but this is zero sum.
    Travel time is calculated and valued based on the assumption that time is actually freed for other uses and thus does generate a total benefit. The possibility of this leading to longer trips is generally ignored. Property values are closely related to travel times, so some of the same benefit is calculated this way.
    If you are only looking at local or regional finances, property values can of course be important. They help in financing a project. This is a different issue from the wider cost benefit analysis. In this case one cannot directly include travel times into the same calculation as there is an obvious risk of double counting some the same benefits.
    How are cost-benefit calculations for infrastructure projects made in the US anyway?

  18. Danny September 1, 2011 at 1:03 pm #

    Scotty,
    I think what you are saying is correct for most cities in the US. But there have been a few cities where those once-fragile political coalitions have become incredibly strong.
    The two strongest examples I have seen are Salt Lake City and Houston. Both cities trend consistently liberal, but with a sizable conservative base outside the core. Both started with extreme opposition to the projects, and with fragile coalitions that were barely able to push the project through. And both cities saw opposition effectively disappear for future rail expansions. Most important to consider is that these expansions are moving outside the core into conservative suburbs.
    I think these two cities show us how to break the political opposition on rail projects. It is quite simple to see that if rail projects A) Are built cost effectively and on time, and B) are operated well with ridership/FRR growth, that the opposition disappears.
    Well, maybe not disappears. But it does go from people picketing/demonstrating/rioting in the streets to merely grumbling in their living rooms. When a system is run well, ideology doesn’t really motivate opposition anymore.

  19. EngineerScotty September 1, 2011 at 2:10 pm #

    It’s interesting to compare SLC and Houston with Portland. All three have growing LRT systems and generally-well-run transit agencies, but the in-construction line to Milwaukie (in suburban, and largely conservative, Clackamas County) seems to be garnering quite a bit MORE opposition, at least from the county, than prior expansions of the system. Some of it may just be resurgent tea party activism, but there are attempts underway to use the initiative process to rescind the county’s share of local match funding. The rest of the system is popular and successful, but the backlash is there.
    One potential difference is that the line is expensive–right now over $200M /mile, albeit including a major new bridge over the Willamette River. SLC’s suburban extensions seem to be coming in around $50M/mile. Another difference might be state politics; both Utah and Texas are solidly Republican states, which permits a lot more deviation from political orthodoxy where it makes local since. State government in Oregon is dominated by Democrats, OTOH, so we may be seeing more of a similar dynamic as in Washington DC, where conservatives are pretty much opposed to anything that the Obama Administration suggests. (Moderates were essentially purged from the Oregon GOP over a decade ago).

  20. MaxUtil September 1, 2011 at 2:45 pm #

    I would just note to a few of the commentors that just because a particular report or institution does not arrive at the conclusion you want is not evidence of “bias”.
    I’m not defending the report or institution, but unless the person writing is paid by the American Petroleum Institute or something similar, just crying “bias” isn’t a convincing rebuttal.

  21. Phil September 1, 2011 at 3:11 pm #

    Yes, it would be great if we could improve frequencies, consolidate routes into 15 minute service, have bus lanes and signal prioritization. However, this could only happen if we escaped the grasp of a car-culture that dislikes any attempt to challenge its dominance. From a purely operational and service standpoint, buses can be absolutely fabulous, but you have to have a culture that wants to make the changes to support such sweeping improvements.
    What most Americans need to see is that transit can be clean, fast and frequent. Light rail promises all these things, but without carrying the cultural stigma of buses. It gives people a better idea of what public transit can be, and by doing so, encourages more citizens to lobby for and desire better bus transit.

  22. Thor September 1, 2011 at 5:25 pm #

    I think some people in this thread are setting up a false dichotomy, they are seeing things strictly through the lenses of auto infrastructure vs transit infrastructure instead of from probably the more useful framework of what types of transit infrastructure in a given area is most cost effective in attaining a specific goal. One of the problems of getting people to use transit in this country is that the cost of transit infrastructure is so expensive, that it is tough to roll it out on a widespread basis.
    Los Angeles was never built out in the same fashion as New York. There is a certain logic in New York building lots of high capacity rail into Manhattan because Manhattan has an incredibly high population density that can utilize that capacity. But Los Angeles doesn’t have one really strong central business district, it has multiple central business districts spread through out the region. Moreover a lot of transit destinations are office parks that were built up along the existing freeway network so overlaying a rail network is a weak fit at best.
    I always suspected that expanding the busway network along the freeways, along the fashion of the El Monte Busway in Los Angeles probably made a lot more sense in LA than building subways in LA. LA has lots of earthquake faults, it has lots of area with large pockets of natural gas, it also lacks the high population gradient that New York has. That means the costs of building underground are likely higher in LA while the benefits are lower. I suspect that the most cost effective way of building out transit in Los Angeles is taking away freeway lanes currently dedicated to cars and re-purposing them into bus only lanes. The freeway infrastructure in LA is already built out, the right of ways have already been acquired. Its just a matter of repainting lanes currently dedicated to cars (and possibly redoing some freeway on and off ramps) and turning them over buses. Moreover that is a project that could be rolled out widely, fairly rapidly avoiding the problem of having places like Santa Clarita where voters are upset that they are paying significantly higher taxes for better transit, but not receiving appreciable benefits to them, because the big rail projects being planned are not coming to them anytime in the foreseeable future.

  23. Eric O September 2, 2011 at 7:25 am #

    The difference in tone between the abstracts is interesting. I’m more familiar with Robert Cervero, and I hafta say the language in the Cervero/Guerra abstract reflects him well. He is one of those wide-brimmed harmonists, everything is about fit, cohesion and networks. I suspect Cervero fits that planning personality of, hmmm, non-deliminators? (maybe there’s a better word) – essentially, folks who prefer to keep determinations open-ended and option-filled. It’s a view that has in mind the reflexive, fluid and multi-tentacled nature of all relationships. This abstract, fittingly, starts with the global “situation” (the debate is too polemical) and, in contrast to the hard-asses, ends with a laid-back and unapologetic repartee. “Our back-of-the-envelope cost-benefit analysis…”.
    I’m not familiar with Gordon and Kolesar, but, needless to say, it has the language of tried and subtle “deliminators”, aka economists. Economists pride themselves in their ability to delimit to hard-ass conclusions (that is, in fact, what their art is about). It is a mistake to think that this tone is betraying an “anti-rail” bias. I don’t think that the tone of the abstract necessarily tells us that. If you think that, I think you are missing the artful tightness these authors are used to crafting in presenting analytic arguments. It is “hey… here’s the problem”, deal with it.

  24. Dan W. in Los Angeles, California, USA September 2, 2011 at 10:09 am #

    Los Angeles is actually far more dense than Thor gives it credit for.
    It wasn’t the dream of a bus-only transit version of Los Angeles County that caused 68% of the county to vote for a half-cent sales tax increase in the middle of the greatest recession since the great depression.
    Also, the transit-only freeway lane idea just isn’t practical or relative to Los Angeles as it actually exists in reality. The places and destinations with the most density to justify transit-only lanes aren’t necessarily anywhere near a freeway.

  25. EngineerScotty September 2, 2011 at 10:28 am #

    The big problem with transit-only freeway lanes in LA is that the freeways themselves are so wide (12- and 14-lane freeways are commonplace) that getting to the stops is a pain in the butt. The Harbor Transitway, so called because it lies within the Harbor (I-110) Freeway, is mainly a series of freeway-median bus stops connected by HOV-2 lanes. And the HOV lanes on most LA freeways aren’t separated from the general-purpose lanes by anything more than a painted divider, so traffic jams in the mainline frequently snarl up the HOV lanes as well.
    LA has many nice wide boulevards (6 lanes or more) that, a la Paris, could stand to lose a general-purpose traffic lane to busses; that probably would be a better solution than median busways.

  26. cph September 2, 2011 at 10:55 am #

    Although LA’s freeway network goes to a lot of places, there are some areas that it doesn’t get particularly close to. Wilshire Blvd, for example.
    Since this boulevard is essentially a “linear downtown” along its length, high-capacity transit in the corridor makes sense.
    Buses can only do so much, dedicated lanes or not. A couple of weeks ago, I was on Wilshire Bl in Santa Monica, deboarding one of the Rapid buses (limited-stop service with signal pre-emption). There were at least two buses in the stop ahead of us. All three buses were 60-foot, articulated models.
    Surface rail might be appropriate for the more suburban parts of LA County, and for the Salt Lake’s, Denvers and Portlands of the US. But for Wilshire Blvd? Grade separation is a must, and there’s only two ways to go: Subway or elevated….and elevated is a political non-starter.

  27. anonymouse September 2, 2011 at 11:05 am #

    Underground construction in LA is actually much easier than in NYC, because in Manhattan at least, you have to dig through solid granite unless you’re digging shallow cut-and-cover trenches just below street level, whereas LA has much better geology that makes it pretty easy to dig the moderately-deep tunnels that work best for subways. Also, LA is not, for the most part, the sort of suburban sprawl that has “office parks along freeways”. It has plenty of tall buildings and high density nodes, and these are not always very well connected by freeways. Beverly Hills, for example, does not have particularly good freeway access. Many of the high density nodes happen to be in a fairly straight line mostly along Wilshire Boulevard too, which would make great sense for a subway line. And this spread out nature actually makes for a more efficient subway network: trains run full both ways, which is more efficient than the NYC case with full trains going to Manhattan in the morning and empty trains leaving, which wastes 50% of the capacity.
    The freeway network is actually quite small for a city of its size, especially in the denser western parts of the city, and is thus very overloaded. And taking freeway lanes away when the roads are already at stop and go levels of traffic for most of the day is politically unpalatable to say the least. Where bus-only lanes have been built, they were built on old rail ROWs (El Monte busway, Orange Line) or as part of a massive billion-dollar freeway expansion (Harbor Transitway). And LA used to have a fairly extensive network of freeway express buses, which it has been slowly dismantling over the course of the past few decades, partly as they’ve been replaced with rail lines, but partly just because they’re not an effective way of providing service.
    The general conclusion is that a fairly small but carefully designed subway system could actually greatly help the mobility in LA, and indeed the Red/Purple Line subway has half the ridership of BART on a system 5 times smaller (by total length). And a subway along Wilshire will likely be the fastest way to get from one point on the street to another, by a huge margin at rush hour, and a small one outside it.

  28. Thor September 2, 2011 at 11:53 am #

    The issue is cost effectiveness of building out a mass transit network that covers the REGION. The current plan is so extraordinarily capital intensive that the projected build out of the rail network probably won’t occur during my life time. Moreover the costs are so great that large areas will never be covered yet they are being asked to fund the system with the transit sales tax. The high cost (and huge cost overruns) of tunneling on Redline means that we probably won’t see a rail connection from the West side of the Valley to West LA ever, nor are there plans to link West LA with the LAX and South Bay employment centers.
    If one wants to argue that the freeway network in Los Angeles is limited, it is still relevant to point out that the current freeway network in Los Angeles is still much bigger than even the most optimistic projections for the rail network in Los Angeles. Moreover the freeway network is already built out. The right of ways have already been secured. The private sector has already build lots of the potential transit destinations along the existing freeway corridors. The question to me is why not increase the intensity of the existing network by shifting from mostly single occupant vehicles to higher capacity vehicles like buses.
    During the 84 Olympics this is what was tried and what worked amazingly well with almost no additional capital investment.
    A busway (not shared with carpoolers) down the 405 (among others) seems like a critical need. It would speed up transit alternatives from lots of trip orgins and destinations.
    Moreover the other key issue for making mass transit transit everywhere is removing car capacity from the city, that has been true on streets that have gone on road diets, that has been true in dutch cities where traffic lanes were taken away to expand bike paths and it will be true in LA. Taking away car lines from freeways is a big part of reducing car capacity and reducing congestion in Los Angeles.

  29. Alon Levy September 2, 2011 at 4:44 pm #

    Just one note: the comment about USC’s anti-rail bias is motivated by the polemics of James Moore, who’s done research funded by Reason. This is the exact equivalent of someone funded by the American Petroleum Institute.

  30. Wad September 3, 2011 at 12:54 am #

    @Thor, the rail network’s success or failure doesn’t depend on you. Metro can and does model where millions of people can take trips along certain corridors.
    Appeals to “the region” discount the unfairness to high-ridership neighborhoods and corridors. Transit boardings aren’t equal, even when equal service is provided. Transit capital and operating dollars can’t be treated like relief grain scattered from a helicopter.
    The question to me is why not increase the intensity of the existing network by shifting from mostly single occupant vehicles to higher capacity vehicles like buses.
    You’re not going to like the answer.
    L.A. tried that and it has not worked. L.A. used to run an extensive suburb-to-CBD bus network in which most riders could pick up a bus outside of their home.
    The San Fernando Valley had a 400-line practically on every street. The Transit Insider maintains an archaeology of the lines. This was repeated throughout the Southern California area.
    By the mid-1990s, these had mostly disappeared. One reason is because they were underutilized. They were too expensive to operate, and they hogged up resources. At around the same time, these buses had gone to improve service on local lines and add many more limited-stop services.
    It may not be fair to the region, but those capital-intensive rail lines performed far better than buses that served everyone everywhere. If the express buses had been able to carry 300,000 boardings a day — as the rail lines do now — they’d still be around.

  31. anonymouse September 4, 2011 at 11:51 pm #

    @Thor, we can argue about the future of LA transit till the cows come home, so I’ll just try to address a few key points. First of all, you mention “a transit system for the REGION”. But how do you define the region? If you just take the contiguous urbanized area, you get a vast expanse of Moorpark to Redlands and Santa Clarita to San Clemente, both about 100 miles, and a population of around 16 million, larger than quite a few European countries, for example Belgium or Hungary or Sweden. The region is too big, and too diverse for just one mode of transportation or a single level of service to be the ideal solution. If, on the other hand, you draw some smaller boundary to define your “region”, then you have the tough question of where do you draw the line and why? There are inevitably going to be plenty of commuters crossing that line constantly, because one of the big advantages to living in LA is having access to a huge regional job market.
    Another point I want to address is a big inherent weakness in running transit lines on freeways: those are by definition the places with the easiest automobile access and thus the places where it’s hardest for transit to compete. Even with congested freeways and exclusive busways, it’s by no means guaranteed that the bus will have the advantage, and practically guaranteed that ridership will fall off steeply along with road congestion. Meanwhile, on corridors like Wilshire, a subway can actually be the fastest option, even without congestion, and can do much, much better competing for ridership against cars, even if it’s ultimately not any faster, or even noticeably slower than the freeway buses you advocate. It’s a matter of comparative advantage.

  32. tomtakt September 12, 2011 at 7:40 am #

    I didn’t have time to read through all the comments, but it seems to me that these policy analyses of transit likely don’t differentiate much based on transit design and land use characteristics, instead opting to distinguish by service mode, which doesn’t necessarily correlate at all with outcome/benefit, as Jarrett so often tries to explain. These study sound like they were designed to be easily carried out, but not to provide meaningful results. Yes, a study that did that would be an massive undertaking, but I think it would be better than trying to use only half of the information.
    Also, I don’t think cost-benefit problems in the U.S. are just explained by high costs. No matter where you go in the U.S., the inclusion of transit consideration in land use, if it exists at all, pales in comparison to practice in most of western Europe. Even most purported “TOD” examples fall short, often stuffed with extra parking instead of destinations. The very idea of TOD is dubious to me, because it implies the notion that the rest of urban development doesn’t need to be transit oriented or at least transit-conscious, which is absurd if you’re trying to provide an effective system for moving people and goods (which transportation agencies should be). The same is true of “context-sensitive design” for roadways–so which roadways are the ones where we’re not taking local context into consideration during design?
    Anyway, my point is that most of these high-level policy analyses of “all heavy rail in the U.S.” feel rather empty, regardless. The question should never be, whether we build a train or not based on cost-benefit analysis of all other trains in countless cities around the country. The question is how do we build an efficient transportation system in a given location that provides many options, redundancy, and reliability. And if we’re thinking about adding a transit line (because we shouldn’t be stuck on trains from the get-go), the question should be how will we build the transit line, what will it look like, what will it cost, and are there alternatives (transit or not transit) for providing the needed transportation.

  33. Nathanael September 16, 2011 at 5:48 pm #

    Thor, LA was built around a set of interurban streetcar lines (Red Cars! or formally Pacific Electric) interconnecting former towns, and local streetcar lines (Yellow Cars!). (I’ve left out some of the other companies.)
    There’s a reason why surface light rail is working so incredibly well for LA. LA was *built* around it. Hell, the Blue Line is built on one of the Red Car lines.
    (Having less street running than the old system, with almost all exclusive reservation, is an important improvement, though; car traffic is one of the things which hurt the old system.)

  34. Andrew September 17, 2011 at 8:12 pm #

    The reason that many rail projects in the US are not all that successful is sort of a chicken or egg problem: a single rail line is never very useful because the vast majority of residents, and the vast majority of employers are not accessible by rail so it will never have terribly high ridership. Making the problem worse is that in many US cities, the bus systems which connect to the rail system are terrible, or do not exist at all, making the rail system useless for most people. Washington is a particularly bad example of this, with many outer Metro stations being little more than park and rides with little or no connecting bus service, and many suburbs having no bus service at all. On the other hand, as the rail and bus network expands, each successive extension to the rail system becomes more and more useful.
    In LA proper (but not the suburbs) rail makes quite a bit of sense, and there are many dense areas which can justify it. In many other US cities, the money is better spent on expanding the bus system.