Frequency and Freedom on Driverless Rapid Transit

DSCN1552If you’ve seen much of Vancouver on television the last few days, you’ve probably seen a shot of a small train gliding along an elevated guideway.  It’s SkyTrain, the world’s largest system of fully automated (driverless) metros.  Perhaps you’ve ridden driverless trains that shuttle between airport terminals.  SkyTrain is the same principle, at a citywide scale.

Driverless trains raise all kinds of anxieties. Many people like knowing there’s someone in charge on the vehicle, and imagine that this person will be useful in emergencies.  But on most subways, you can only talk to this person by pushing an intercom button.  There’s very little he can do if there’s an emergency in your car other than call for help. 

DSCN1540SkyTrain provides exactly the same emergency intercom button, but the person you’re talking to is in a central control center instead of on the train.  As a result, they are likely to have a better view of the whole emergency, and the resources at hand, than a train driver could have.  There are still staff roaming the system, checking fares and providing assistance, but not one on every train.

Of course, if zoom and whoosh matter to you, there’s nothing like the experience of riding in the single seat at the front of the train, enjoying a view that’s usually reserved for a train driver.  And it’s quite a view.

None of these, however, are the real reason to consider driverless metros.  The real reason is that they break the connection between frequency and labor costs.  Look at SkyTrain’s frequency tables:

Expo frequency

On the Expo and Millennium lines, you’ll never wait more than 8 minutes for a train.  On the combined portions of these lines, which connect some of the densest centers in greater Vancouver, you’ll never wait more than 4 minutes — not even at 7:00 AM on Sunday or 11:45 PM on a Tuesday.    In a year of living in Vancouver and riding Skytrain every day, I never waited more than 5 minutes.

DSCN1549 Some non-automated subways do run as frequently as SkyTrain, but they’re in places like Manhattan, London, and Paris, cities vastly larger than greater Vancouver.  (Paris, by the way, now has a driverless line, and is beginning to convert existing lines to driverless.)  No non-automated system could have delivered such high frequencies late into the night in a city of Vancouver’s scale.

The lack of a driver is the key to those extreme frequencies.  When you have a driver on every vehicle, the labor cost is the dominant cost of operations.  So when you have to cut service, as many North American agencies are doing this year, you end up cutting frequencies, starting with late night and weekend.  Many North American light rail systems are dropping below even a 15-minute frequency in the evening, making themselves increasingly useless for the spontaneous trips that are essential to freedom in urban life.

As I’ve followed the debates between driverless metro and light rail, most recently in Honolulu, I’ve been struck by how rarely frequency is mentioned.  Reading the coverage of Honolulu’s debate between elevated driverless metro and light rail, you’d think that the primary question is profile — elevated vs surface — and that this is a question of running fast on an ugly elevated structure or running slowly but more accessibly on street.  Few on either side make the crucial point that a driverless system will be able to run frequently even very late at night, early in the morning, whenever you want to travel.  Light rail, bound by labor costs, will always be pressured to cut frequency outside the peak period.

This is not, in any sense, a blanket endorsement of driverless metros over light rail.  They are different tools for different situations.  The stable operating cost of driverless metros comes at a capital cost.  Vancouver’s next project, the Evergreen Line, is expected to cost $1.4 billion (CAD) for 11 km, well over twice the per-km cost of typical light rail.

Light rail is wonderfully flexible, able to run onstreet with signalized intersections, and across pedestrian zones, as well as in conventional elevated or underground  profiles.  Driverless metro must be totally grade-separated, which in practice usually means elevated or underground.  SkyTrain got its name because the original lines were mostly elevated, though the newest, the Canada Line, has a long underground segment. (I’ll do another post on the local community impacts of the SkyTrain elevated structure, and the serious problems (and opportunites) they create.  I’ll do yet another on Vancouver’s current SkyTrain vs light rail debate, in its Broadway corridor.)

As you’d expect, SkyTrain has higher average speeds and reliability than surface light rail.  As always in urban transit, speed and reliability are mostly not features of the technology; they are mainly about how much you stop and what can get in your way.  Not much can get in the way of a fully grade-separated system like SkyTrain.

But I am struck that in these debates there’s so little talk of off-peak frequency.  As an urbanist, I want to see vibrant cities that are full of activity for 18 if not 24 hours a day.  That means I want nightlife.  Isn’t that a good reason to be interested in a service that really can run every 4 minutes at midnight?Granville-lights-3

Last Photo:  Gordon Price, Price Tags

84 Responses to Frequency and Freedom on Driverless Rapid Transit

  1. Danny February 18, 2010 at 6:23 am #

    I have commented about labor costs in many transit blogs, and they all seem to come back with the same comments. Among the most popular:
    1) The government needs to set an example to the private market in setting wages
    2) I hate the middle class and want everybody to be poor
    3) We should be subsidizing transit anyway because of the positive externalities it brings to drivers
    Unfortunately, all of these arguments are incredibly naive and in the end they do nothing more than work to disable transit effectiveness. When we keep operating costs low, we get more transit out of less money. This post is an excellent example of exactly that. I wonder how transit advocates would feel about eliminating transit jobs completely, as opposed to merely keeping transit workers at market wages?
    Operating costs matter much more than people realize. Economists understand the rationality of costs quite well. When revenue drops below total costs, it is rational to continue to operate as long as your revenues are greater than operating costs. Furthermore, it is rational to continue to expand service until marginal revenue equals marginal cost. This means that if you are making an operational profit, even at a net loss, it is still rational to continue to expand your service until the extra revenue you get does not make up for the extra cost impose. With high labor costs, this means that you can not expand your service as much for a given level of funding.
    Of course expanded service levels can come in many forms, including higher frequency, longer route lengths, faster speeds, or even more stations. But if you choose any one, or combination of the above results, I’m pretty sure a transit advocate would be delighted. So why is it that transit advocates never seem to mind obscenely high wages?

  2. smably February 18, 2010 at 9:11 am #

    The TTC does almost as well on its subways (“trains run every 2-3 minutes during the rush hours and every 4-5 minutes outside the rush hours”) — but at a much greater cost than the SkyTrain, I’m sure.

  3. jfruh February 18, 2010 at 10:13 am #

    I really think that people who don’t ride transit regularly (and sadly, the people in power making these decisions almost never ride transit regularly) don’t get how much more appealing frequent service is, and what a killer to frequent use long headways are.
    Here in Baltimore, the Light Rail network routing is such that Penn Station, which could be a major transit hub, is served by trains only ever *30* minutes, all day. The connection to the airport is also on 30 minute headways outside of peak weekday commuting hours (when it’s at the barely acceptable 20 minute level).
    I’ve been told that the driving force behind these schedules is the local transit agency’s phobia of seeing empty trains. Their thinkng is that there is only a set number of people who will be riding transit, so if they come less frequently, they’ll be making better use of their resources. It doesn’t seem to occur to them that more frequent service would attract more riders.

  4. Mike B February 18, 2010 at 11:13 am #

    It’s worth noting that SkyTrain’s Evergreen line will cost so much because it has to use a costly proprietary linear induction motor technology developed by Bombardier. The trains on the Millennium and Expo lines do not have traction motors on the trains but are propelled though a maglev-like strip in the middle of the track. This adds a significant cost to adding onto the Ex and Ml lines. It’s not too fair to say the driverless system adds a significant cost to the construction of the Evergreen line when its the propulsion technology that really inflates the cost.
    The new Canada Line is driverless, uses conventional technology, and cost $2 billion CDN for 19.2km. About 60% of the line is underground.

  5. Nathan Williams February 18, 2010 at 11:16 am #

    Up here in Boston, two of the three heavy rail lines and the light rail line operate with two operators per train. There has been a lot of resistance on going down to 1 per train, even though one line already did it. The idea of going driverless seems like fantasy from here.

  6. Steven Vance February 18, 2010 at 12:12 pm #

    Jarrett, the automated train in Singapore (MRT) has almost double the track length as SkyTrain and 5 times as many daily riders.

  7. Pantheon February 18, 2010 at 12:53 pm #

    I would rather have a high upfront cost and a lower operating cost than the reverse. That way if there is a recession, service won’t be cut as much. Transit cuts are death to transit. They make you feel like you can’t rely on transit and that you are better off owning a car.

  8. Watson February 18, 2010 at 1:46 pm #

    Their thinkng is that there is only a set number of people who will be riding transit, so if they come less frequently, they’ll be making better use of their resources. It doesn’t seem to occur to them that more frequent service would attract more riders.
    I’m sure that transit planners realize that more frequent service tends to attract more riders. But it won’t necessarily attract enough extra riders to cover the additional operating costs. And more frequent service might increase capital costs too. The bus route that runs near my house has only two buses an hour. Even at that low frequency, the buses usually seem to be mostly empty. It doesn’t seem likely to me that increasing the frequency would generate enough additional riders to cover the additional costs.

  9. gap February 18, 2010 at 1:51 pm #

    I have always drooled over Vancouver’s system, and yes frequency has always been the most attractive element to me. Skytrain really feels like a horizontal elevator to me in that it is practically on-demand – I think that goes a long way to explaining its high ridership in a city of Vancouver’s size. A question though – any idea when and if skytrain has recouped the additional capital costs through its much reduced operating costs?

  10. Jarrett at HumanTransit.org February 18, 2010 at 1:54 pm #

    Re the last 2 comments:  I don't have the details at hand, and all my Vancouver contacts are out on the streets helping with the Olympics.  But I don't think anyone in Vancouver thinks in terms of recouping capital costs, and certainly not in terms of breaking even on fares.  The real purpose of SkyTrain is to drive the creation of more sustainable urban form, and the "horizontal elevator" effect of extreme frequency is a key to not only triggering such development but being useful to the people who end up living there.

  11. EngineerScotty February 18, 2010 at 2:27 pm #

    Has your local freeway recouped its capital costs?
    Unless it used tolls to finance construction and the bonds are now paid off, chances are the answer is “no”.

  12. Watson February 18, 2010 at 2:41 pm #

    EngineerScotty,
    Studies have found that direct public subsidies to motor vehicle users amount to about 1 cent per vehicle-mile, or about 2% of the average total cost of owning and operating an automobile.
    Jarrett,
    The money to pay for more frequent service has to come from somewhere. If fares won’t cover it, it’ll have to come out of some other part of the budget. You can argue that it’s justified on the grounds of some long-term goal of promoting a more “sustainable” urban form, but the people who actually have to pay the money may not be convinced.

  13. EngineerScotty February 18, 2010 at 3:04 pm #

    Watson,
    What do you think of the figures reported (and the studies cited) by this page at the Sierra Club website? The cited studies are a bit older (and include one 1996 study by Mark Delucchi of UC Davis, one of the co-authors of the 2008 study you are undoubtedly referring to), but take into account quite a few factors not accounted for in the 2008 study.

  14. gap February 18, 2010 at 3:13 pm #

    Let me clarify……when I asked whether skytrain recouped costs, I meant specifically the additional upfront costs of automation only. I think that this piece of information IS valuable when making cases for future installations of automated transit. Like it or not, it is easier to sell a x% increase in capital costs if you can demonstrate that that additional x% will not only be paid for but pay dividends down the road. I would even imagine that in some borderline cases where a decision between full grade separation and say 90% grade separation is to be made, one could potentially argue that full grade separation with automation could actually be cheaper in the long run.

  15. Watson February 18, 2010 at 4:03 pm #

    Scotty,
    I wrote “direct public subsidies.” Direct public subsidies to mass transit are much higher — around 70 cents on the dollar. So to answer your question, highways do not recover their capital costs entirely, but the evidence indicates that they come much, much closer to doing so than transit services. Eliminating these subsidies would have only a marginal effect on the cost of driving, but a huge effect on the cost of using transit.
    As for your Sierra Club link, only a few of the items on its “Sources of Subsidies” list are direct public subsidies, many others are arguably not subsidies at all, and most apply to many kinds of industrial activity, including mass transit, not just motor vehicles. Environmental and military costs associated with the production and consumption of petroleum, for example, apply to diesel-powered buses and trains as well as automobiles. “Sprawl” is listed as a subsidy for motor vehicle users, apparently on the grounds that motor vehicles promote sprawl and that sprawl causes a “loss of transportation options.” But you could argue that high-density development is a subsidy to transit users for the same reason. Transit promotes density, and density makes it harder to choose the “transportation option” of driving. “Parking” is listed as a subsidy on the grounds that it causes “more expensive goods or services.” But those expenses are mostly paid by the same drivers who get the parking benefit. They’re just paying for their parking indirectly. So it’s not clear that it should be counted as a subsidy at all.
    The bottom line is that I think the Sierra Club’s analysis is basically worthless.

  16. mezzanine February 18, 2010 at 4:21 pm #

    This link has a lot of details – this is a presentation by Translink (vancouver’s regional transit authority) to the City of Ottawa (who is trying to decide on their LRT line.
    http://www.city.ottawa.on.ca/residents/public_consult/tmp/lrt/vancouver_en.pdf
    Some highlights:
    – ~100% operations and maintenace cost recovery from allocated fare revenue.
    -33 stations (pre c-line) and ~ 30-55 staff. ~ 38 of these are roving attendants (ie, not involved with central train control, but do ‘patrol’. Even with that, 60% of costs are labour
    and we are seeing crazy ridership with the olympics. the airport metro is carrying up to ~ 210,000 daily ridership. (not very comfortable with ~ 40 metre long trains, but somehow we are doing it…)
    “With thousands of Olympic tourists in town, and Vancouverites urged to abandon their vehicles in favour of public transportation, Kelsey said a record-setting 500,000 people rode the Expo and Millennium Lines on Sunday. Another 210,000 rode the Canada Line.”
    http://www.ctvbc.ctv.ca/servlet/an/local/CTVNews/20100215/bc_transit_records_100215/20100215?hub=BritishColumbiaHome

  17. Daniel February 18, 2010 at 4:55 pm #

    Certainly often in the political debate, there seems to be a fixation on infrastructure, and little talk of frequency, yet the later is critical to making the service attractive to passengers.
    So we get the situation such as in Melbourne where billions are being spent on new rail lines, often with no specific details on what services will run on them, while simple, relatively cheap upgrades like improving connecting feeder bus services to areas that don’t (and will never) have rail are not funded.

  18. Mixner Spotter February 18, 2010 at 5:10 pm #

    You’re not fooling anyone, Watson.

  19. Richard February 18, 2010 at 6:05 pm #

    The frequency of SkyTrain is great. It is really nice not to have wait for long periods of time at night. Regarding the cost Evergreen Line, as a good portion has to be in a tunnel due to steep grades and a road with a narrow right-of-way and that a good portion of the planned SkyTrain route is at-grade along a railway corridor, the cost of SkyTrain is only 20-25% greater than light rail would have been. It really is not that helpful comparing the cost of a line against the average cost of other lines in total different contexts especially when estimates for the different technologies are for the corridor are readily available.
    @Mike B
    The cost of the linear induction motors and the rail really doesn’t add much to the cost of the project. It is the grade separation that makes SkyTrain more expensive.

  20. Rhywun February 18, 2010 at 6:34 pm #

    The idea of going driverless seems like fantasy from here.

    Same here in NYC. The concept is really only possible for new construction.
    As for subsidies, we could argue all day about it, but the fact remains that public transit is largely seen as a “benefit” and priced accordingly, at the lowest possible level such that those of even the most modest means can afford it. I mean c’mon, my entire transportation needs are covered at 89 bucks a month. Buying a car, insuring it, repairing it, paying to store it while on the job would cost I don’t even know how many times that per month. I’m amazed that people even consider it worthwhile, but that’s just me.

  21. Watson February 18, 2010 at 7:04 pm #

    As for subsidies, we could argue all day about it, but the fact remains that public transit is largely seen as a “benefit” and priced accordingly, at the lowest possible level such that those of even the most modest means can afford it.
    But there’s very little means-testing of the benefit. So affluent New York bankers, for example, get massively-subsidized commutes to work courtesy of ordinary taxpayers. If the goal is to subsidize transportation costs for the poor, a better policy would be to give them a means-tested transportation credit and let them use it in whatever way makes the most sense for their personal situation.
    Buying a car, insuring it, repairing it, paying to store it while on the job would cost I don’t even know how many times that per month. I’m amazed that people even consider it worthwhile, but that’s just me.
    You say you live in NYC. For many New Yorkers, and probably most people who live in Manhattan and the denser parts of the other boroughs, it’s not worthwhile to own a car, because driving in New York tends to be difficult and expensive and because New York is dense enough for mass transit to provide an effective alternative. Elsewhere in the country, the situation is very different.

  22. Corey Burger February 18, 2010 at 9:33 pm #

    All transportation mode are subsidies of some kind. The problem with automobile & road subsidies is that they have externalities such as pollution, increased dependence on oil, etc. Contrast that with transit/bike/pedestrian subsidies which have many positive externalities such as health benefits, lower energy costs, high community involvement, etc.

  23. Mixner Spotter February 18, 2010 at 10:04 pm #

    You don’t like trains, Mixner. We get it. Now please go away.

  24. EngineerScotty February 18, 2010 at 11:01 pm #

    Watson, a few points.
    1) Given that the vast majority of fossil fuels burned in vehicles are NOT burned in transit vehicles, dismissing the military costs of oil usage on the grounds that transit benefits too would be like excluding the cost of roads from the equation on the grounds that busses use them. And its worth noting that US naval expenditures benefit OTHER countries as well.
    2) Sprawl has legitimate and real costs compared to density, especially for infrastructure and services–the cost of which scales linearily with the distance that needs to be covered to provide the service. Utilities, roads, bus service, police and fire services, the post office–all of these things get more expensive when density gets lower. (Of course, at certain high levels of density, when building heights start reaching a certain level, other costs come into place). But in general, it costs a heckuva lot more to provide urban services to a collection of quarter-acre lots than it does to denser urban forms.
    3) Your observation about means-testing applies equally to roads.
    In general, my issue with the line of argument that “transit is more heavily subsidized, thus its a bad investment” is that it’s almost a tautology. Trillions have been invested in roads and highways over the years; there are around 4 million miles of roads in the US, compared to dedicated transit mileage in the thousands, for the whole country. (I’m obviously excluding freight rail lines which see passenger service). Of course transit service is going to be expensive to provide, and low quality, given the current state of affairs.
    The argument you seem to be making is, that since we have a developed road network and lots of infrastructure and development which caters to the automobile, we shouldn’t bother with transit (or should limit it to areas of high density and low-quality social service routes); and should focus our resources extending what we already have. The trouble with what we already have is that its highly dependent on a resource that may soon enter short supply, and which has numerous deleterious environmental effects.

  25. EngineerScotty February 18, 2010 at 11:08 pm #

    Whether Watson is the same fellow who posts as Mixner at thinkprogress, or as garyg at streetsblog, is probably irrelevant.

  26. Mixner Spotter February 18, 2010 at 11:08 pm #

    Mixner will retort that since we can’t put an absolute precise number on private auto externalities, they don’t exist or matter. He’ll then state highly auto dependent sprawl development is simply due to preferences, not bad policy.

  27. Alon Levy February 19, 2010 at 12:53 am #

    Danny: you’re misstating how economists think. Economists tell people to think at the margin: if marginal revenue is higher than marginal cost, then a company should expand; if it’s lower, it should contract. Average revenues and costs may be quite lower.
    And Watson, “studies show” is a line used by virtually everyone on the Internet. I’ve seen it used by creationists, who don’t even have hack thinktank-published studies to fall back on, and can’t even do numbers games using two sets of data. So either refute the Texas numbers saying highways pay at most 50% of their total construction and maintenance cost and some pay 16%, or go invent a new name for yourself and troll another blog.

  28. Pantheon February 19, 2010 at 1:15 am #

    You guys are being trolled. This is supposed to be a thread about driverless rapid transit. Instead you all are getting drawn in to an endless debate that we have covered before, and that has nothing to do with the topic at hand. I’m not saying Watson’s points aren’t worthy of discussion, or that he doesn’t express them well. But in effect this is a hijacking of the thread.
    I personally would like to see a lot more of the driverless model, and I’m curious as to why it hasn’t taken off in the U.S. Anyone have any thoughts on that? Maybe it’s the opposition to elevated structures. Or since Democrats are the only ones in America who seem to want transit, I wonder if the unions push them to build systems that require drivers…?

  29. Vonny February 19, 2010 at 1:53 am #

    To be sure, it is certainly possible to upgrade an existing segregated system to driverless mode.
    It is currently underdone in Paris subway (line 1), as well as in Stockholm.
    Also, A line doesn’t need to be driverless on all its length.
    You can pretty well imagine a driverless sytem becoming manual driven, to continue its route at grade (by the way in the yard the train are usually manually driven)
    for the third line in Vancouver (Canada line), this option had been briefly considered, but the bidders didn’t think it was a wise choice, eventually because you have some other constraint like operational one, what to do with the third rail, frequency limitation…
    Consider that if the automatic train run fully automated in a segregated way, it is more for economic consideration that by technical limitation.

  30. spyone February 19, 2010 at 7:31 am #

    @gap
    I agree entirely that a more expensive alternative is easier to sell when you can say the operating cost savings will repay the additional construction cost in x-number of years.
    Unfortunately, the biggest advantage of fully automated is a little harder to quantify: it isn’t the lower cost of operating at times you were already planning on operating, it is the new times of operation that become possible due to the lower cost. It is the ability to say “rather than running once an hour from 9pm to 1 am, this line will run every 15 minutes just like it does during rush hour, and will keep running past 2am to take home bar-goers.” when those vehicles/times will not generate enough revenue to pay a driver.
    Not sure how exactly to put a dollar value on that.

  31. Danny February 19, 2010 at 7:32 am #

    Alon,
    I did not misstate how economists think. In fact, I didn’t even contradict anything you said. I think what happened is you didn’t understand what I said.

  32. spyone February 19, 2010 at 7:54 am #

    I read a blog, essentially, posted by a fellow who lives and works in Vancouver. Once a week he posts a new entry that is the epic saga of his work-related stress: crazy people at work, out of snacks at the vending machine, crazy encounters on his commute.
    He takes the bus to the Skytrain, or sometimes he just walks to the Skytrain. He has mentioned missing his bus, but never missing the train. It took some time for me to realize that this is because he is not planning to catch any particular train: he arrives at the station inside a given window, and trusts that a train will be along soon enough to get him to work on time. Occasionally he makes better time and stops at a 7-Eleven to eat.
    That alone surprised me: that he was not trying to catch a specific train, that his commute had enough slack in it that his bus being late and missing the connecting train wouldn’t make him late for work (and conversely that he wasn’t usually getting to work significantly early and having to wait until his shift starts).
    The second thing that surprised me was that his shift runs 11pm to 7am.
    In a city of about half a million, in a metro area about 4 times that size, they run the transit every few minutes even at 11pm, and not just when there is a hockey game. A man with a roughly minimum-wage job could work the graveyard shift somewhere far from his home, not own a car, and have no problem getting to and from work.
    I live in a city of similar size, in a metro area of similar size, and I work a similar shift. There is only one bus route in this city that runs after 7pm, and it stops at 1am. The buses start again somewhere between 6am and 7am, so I could conceivably take the bus home, but my relief could not take the bus in. Of the 3 shifts at our store, none could take the bus more than one way: at either the beginning or the end of their shift, there is simply no bus to take.
    Driverless Metro cannot replace, for example, bus service. But it can allow a city that is not nearly as large as New York to provide service frequent enough to be useful over most of the day without prohibitive cost.
    And as to Boston and other such cities: drivers on completely grade-separated lines are rapidly becoming obsolete. I understand the Unions’ alarm at losing all those jobs, and their responsibility to their members’ futures, but I imagine a lot of buggywhip makers were put out of work by the automobile. At some point you need to stop demanding that your members be hired to perform a job that requires no human involvement and instead demand that they be retrained to do something useful. (IIRC, the second “driver”on Boston’s Red Line only opens and closes the doors.)

  33. Watson February 19, 2010 at 9:01 am #

    Scotty,
    To address your numbered points:
    1. The relevant costs are not the total cost by mode but the cost by mode per unit of transportation benefit, which is usually measured in passenger-miles. If you think you can show that transit has lower environmental and military costs per passenger-mile than automobiles please do so.
    2. You seem to be conflating user costs and subsidies. If sprawl has higher “urban services” costs than dense development, those costs are relevant to this discussion only to the extent that they are paid by people other than the consumers of those services. That is, only to the extent that they are subsidies. Your speculations about these costs also seem to me highly dubious. As far as I’m aware, per capita crime rates do not tend to decrease as density increases, so I wouldn’t expect police costs to decrease as density increases either. As far as I’m aware, firefighting costs tend to increase with building height, and the risk of a fire in one building or home spreading to others tends to increase as building density increases. So if anything I would expect fire-related subsidies to decrease with density rather than the reverse. Again, if you think you can show that subsidies for urban services are higher for sprawl than for dense development then please do so. Other external costs that tend to increase with density are congestion, crowding, noise, pollution and loss of privacy. The difficulty of internalizing these costs is one reason why people tend to support laws that limit density.
    3. I don’t really understand your point here. I’m not arguing that improving mobility for the poor is a primary purpose of motor vehicle user subsidies. I don’t think it’s a significant purpose of those subsidies at all. But it is often cited as a primary purpose of mass transit subsidies. My point is that mass transit subsidies are a very ineffective way of improving mobility for the poor, because so many of the beneficiaries of those subsidies are not poor. A clear illustration of this is the Consent Decree in Los Angeles in the 1990s. The LA bus riders union successfully sued the government for spending so much public money on rail transit that mostly served wealthier Angelenos at the expense of bus services that mostly served poorer and minority communities.

  34. Watson February 19, 2010 at 9:10 am #

    Alon Levy,
    And Watson, “studies show” is a line used by virtually everyone on the Internet. I’ve seen it used by creationists, who don’t even have hack thinktank-published studies to fall back on, and can’t even do numbers games using two sets of data. So either refute the Texas numbers saying highways pay at most 50% of their total construction and maintenance cost and some pay 16%,
    The study that concluded that direct public subsidies to motor vehicle users are about 1 cent per vehicle-mile is Do motor-vehicle users in the US pay their way? by Mark Delucchi of the Institute of Transportation Studies at UC Davis. The author notes that this finding is similar to that of studies by the Transportion Research Board and the FHWA.
    Please show me the study supporting your claims about highway subsidies in Texas. A “newsletter” is not a study.

  35. Watson February 19, 2010 at 9:21 am #

    Pantheon,
    EngineerScotty complained about road subsidies in his comment of 09:27. He brought up the issue, not me.

  36. EngineerScotty February 19, 2010 at 10:55 am #

    Watson,
    1) Well-designed and well-used transit is massively more energy-efficient than autos are. With the current state of US transit–with many routes being low-ridership “coverage” routes–autos can look competitive; but even the most energy-efficient automobile can’t compete with a bus full of passengers.
    2) Some of these costs are paid for by users–but many of these costs are borne by users independent of whether or not they drive. If my water rates are higher because the city water bureau has to recoup higher construction costs for servicing sprawl, or my police taxes are higher because the local police force has to put more officers in patrol cars and distribute their operations vs having more cops walking beats (things like walking beats simply don’t work in surburbia); that has nothing to do with my transportation choices.
    I agree that very high density has its own costs (and suggested as such in my original post); though some of the costs you cite aren’t an issue unless you are cramming lots of internal combustion engines among the density. There seems to be a “sweet spot” for density, I’d guess between 4,000 and 8,000 persons/square mile, at which urban services (including transit) can be provided at lower expense, but where the downsides of high density aren’t as much of an issue. (Instead, it seems, many cities tend to have islands of high density in the urban city, surrounded by a sea of sprawl).
    3) I have no objection in principle to means-testing; other than thinking it ought to apply to all modes, or none. The problem with charging everyone $5 for a bus ride but giving the poor $4 off is that this ensures that only the poor will ride the bus–enforcing economic mode segregation even worse than is experienced today.
    A common argument among anti-transit advocates is essentially that roads are a public good due to universal use, so its OK to socialize their costs (i.e. everyone benefits from the roads–even non-drivers–so its OK for them to be financed generally); but transit is not a public good due to low(er) use–thus expenditures on transit should face higher scrutiny. A textbook Catch-22, if you ask me–and argument ad inertium.

  37. dejv February 19, 2010 at 12:33 pm #

    My city developed an efficient and user friendly model of night-time transit with long headways: it drops all daytime lines and runs special night time buses. There is virtually no traffic so they don’t have to wait at intersections and most of the stops are request stops reducing travel times to two thirds of day-time transit. All of these lines meet every half-hour to hour at main station, providing convenient transfer and proximity to historic district, the center of nightlife. In addition, there are other transfer points at places where bus routes meet or intersect.

    (IIRC, the second “driver”on Boston’s Red Line only opens and closes the doors.)

    Wow, I’m amazed. I wouldn’t expect such obsolete jobs over the big pond.

  38. Betty February 19, 2010 at 1:10 pm #

    SkyTrain is my idea of progressive. That, along with TransLink’s hybrid artics.
    The transit agency I work for in Portland loves to beat its chest about how progressive it is.
    I saw a Photo Shop’d photo a couple years ago that displayed a 19th century rail car next to a MAX car on Morrison. I remember thinking ‘Not much progress there’. And, our roads are being torn up by the City of Portland to extend StreetCar of all things.
    Deluding the public to believe the past was better than the present and that we’re a European city is not progressive. I thought I’d be able to look forward to traveling more like George Jetson.

  39. David February 19, 2010 at 4:05 pm #

    Just a note on night life.
    Skytrain does seem favourable compared to other options and in theory it would be better in promoting night life than light rail or buses. But as always the agencies involved need to be committed to this for it to work, and I wouldn’t say that Vancouver’s transit agency or the various Municipalities involved are committed to encouraging a 24 hour city (or even much of an 18 hour one for that matter… because seriously what kind of Friday night is it when you go home at 10-12pm?!?). Vancouver is a bit of a backwater when it comes to urban culture, especially night life.
    As for what this has to do with driverless rapid transit one should note that the last Canada Line train from Downtown to the burbs is at 1:15am everyday except Sundays/Holidays when it’s about 15 minutes earlier. The last Expo Line train from Downtown to the burbs leaves at 12:38am everyday except Sundays/Holidays when it leaves at 11:38pm! As for the Millennium Line that’s even worse is that the last one leaves Waterfront at 12:09am everyday except Sundays/Holidays when it leaves at 11:09pm! For the Millennium and Expo Lines there’s limited service on limited sections for about an hour longer from the advertised times, but it doesn’t serve much of a purpose if you live further down the line since you can’t utilize the entire system and make a full trip. This has caused a lot of hassle for some of my suburban friends, including unwanted sleep overs like we’re still teenagers or something. It’s one of the mesmerizing things about the Vancouver, the almost conspiratorial attempts (along with the poor night bus service and anti-Liquor ordinances from the City itself that discourage night life outside of the Granville strip you’ve pictured, Yaletown and Gastown) designed to prevent a vibrant and varied night life.

  40. Watson February 19, 2010 at 4:08 pm #

    Scotty,
    Your discussion of the issue is still confused and uninformed. First, it’s important to clearly distinguish user costs from subsidies. If low-density suburbs have higher water utility costs, but those higher costs are covered by higher water rates on suburban homes, that’s not a subsidy to car users. The higher cost of the service is paid by the consumer of the service, not by a third party.
    Again, if you think you can show that subsidies to automobile users are higher than subsidies to transit users, please do so. With respect to direct public subsidies, I think the evidence is pretty conclusive that transit users receive much, much higher subsidies than automobile users per passenger-mile. You speculate that automobile users receive higher indirect subsidies than transit users, but you haven’t produced any serious analysis to support that speculation. You assert, for example, that autos promote sprawl and that sprawl causes higher policing costs. But you haven’t produced any evidence that that’s true at all, let alone how large the effect is. It seems to me at least as plausible that higher density causes higher crime rates and that higher crime rates cause higher policing costs. In any case, the questions cannot be settled by GUESSING.

  41. Dave 2 February 19, 2010 at 5:59 pm #

    Having used Skytrain since 1986, and probably 1000 times since 2005 alone, I can assert that there is no anxiety over ‘driverless’ trains. An occasional panic stop, where the train brakes from 80 km/h to 0 in about three seconds reminds us that there is a computer program ensuring that two trains never collide.

  42. Richard February 19, 2010 at 6:15 pm #

    @Watson
    I suspect you already know about this site but just in case, http://www.vtpi.org/tca/

  43. Dave 2 February 19, 2010 at 7:11 pm #

    David, just a note that you’re reading the schedule wrong. The 12:38 PM last departure is King George to Waterfront, the last departure from Waterfront to King George is actually 1:15am. Not that that’s much better compared to New York or Berlin with their 24 hour service, but it is comparable to the last train times in London or Paris.
    PS, I have on occasion made use of the “limited service on limited sections”, if your destination is North Burnaby then these M Line trains that don’t go all the way around the horn to Waterfront can be useful.

  44. Vonny February 19, 2010 at 9:04 pm #

    David, just a note that you have took the skytrain in the wrong direction, and have ended to king George, that is the reason why you find “Vancouver is a bit of a backwater when it comes to urban culture, especially night life” 😉

  45. spyone February 19, 2010 at 9:46 pm #

    @ Watson:
    You are definitely wrong to characterize Engineer Scotty’s post of 9:24 as a “complaint” about road subsidies.
    Gap had asked how quickly the SkyTrain’s reduced operating cost would offset the increased construction cost, And Scotty appears to have misunderstood him to have asked how long before those savings would offset the entire construction cost, so he pointed out that roads rarely achieve that goal, implying that it would be unreasonable to expect rail to do so. (In fact, he offered no factual support for that assertion.)
    You brought up road subsidies in your post immediately after that, as a reply to what he had said although I do not see anything in what he said that would make you think of them. Given what he’d said basically implied that most roads do not pay for themselves, data on the economic benefits of roads would be relevant. Data on increased tax revenue resulting from the economic benefits of roads would be especially relevant.
    What percentage of the cost of operating a motor vehicle is paid by the owner of that vehicle has no direct bearing on the question of how long it takes for whatever agency built a given road to realize enough economic benefit from its existence to cover the cost of its construction, so your “reply” to Scotty was not really relevant, and it does appear to me that you raised the subject first, at least within this thread.
    Perhaps Jarrett could make a short post about “The great road subsidy debate” so there would be a place where the discussion could continue and be entirely on-topic.

  46. EngineerScotty February 19, 2010 at 10:12 pm #

    Probably a fair summary, spyone. (And I think costing would be a great topic for a separate post, given the wide variety of opinions, including in the literature, on the topic).
    Of course, if we bring the economic value of roads (which I obviously ignored) into the question, then we also need to consider the economic value of transit as a benefit as well, not just how much is recovered in fares.
    A higher water bill paid by suburban residents IS an external cost, given that its paid by those residents who don’t drive equally as those who do. Granted, most suburban residents of driving age are motorists–the burbs are not conducive to car-free living–but simply conflating two groups on the basis of a large conjunction, is a bit sloppy. Since someone posted the Victoria Transport Institute study above, Chapter 8 (http://www.vtpi.org/tca/tca08.pdf) is particularly relevant to this discussion.
    Per-capita subsidy comparisons are, ultimately I think, not terribly useful, when comparing one mode of transport which is ubiquitious and fully developed, with another which is in many ways in its embryionic stage. As I stated above, it’s essentially a demand that transit achieve financial efficiencies comparable to autos without comparable investment–in order to justify further investment necessary to achieve the economies of scale necessary for efficient investment.
    The argument that we shouldn’t build more transit because its inefficient because few people use it because it’s inconvenient and incomplete because we haven’t built enough of it to be truly useful–strikes me as more than a bit circular. If you think that because we’ve built a century worth of infrastructure based on the automobile, we should sleep in the bed we made rather than trying to build up reasonable transit infrastructure, make that argument. But complaints about transit subsidies when much of our transit infrastructure is social-equity coverage service, strikes me as misguided.

  47. Paul February 20, 2010 at 2:41 pm #

    @David in regards to very late night service. The people who run skytrain would love to run all night service. The problem is they need a window to do track maintenance. And they can’t do it with the trains running. Although I have thought that they could possibly run single track sections in areas they might be inspecting or working on. But that would mean that trains may have to run at 30 min frequencies.
    In regards to the trains being automated. I feel during these 2010 Olympics with the huge crush load of people the system is experiencing. That the all day rush hour service is the biggest reason the system as run so well. If it had drivers, Translink would of had to decide when to have more trains based on scheduled events. So you would of see increases in service at certain times. But because it was driver less it allowed them to run all the trains all day non-stop. Even if there were times when the passenger load dropped off for a time.

  48. Watson February 20, 2010 at 9:30 pm #

    I suspect you already know about this site but just in case, “>http://www.vtpi.org/tca/
    I don’t think much of the Victoria Transport Policy Institute. This “institute” appears to consist of one man with a website producing self-published reports extolling the supposed virtues of transit and urbanism. The “institute’s” address is a house on a residential street. There’s no peer review. No government or academic affiliation. No recognized expertise.
    The report you link to is rather confusing. It purports to be a “Transportation Cost and Benefit Analysis,” but it doesn’t appear to contain any clear estimates of benefits. There’s a section that discusses methods for estimating benefits, but I don’t see any actual estimates.
    That said, with respect to costs, the author concludes that both the external costs and total costs of the two forms of mass transit he mentions (diesel buses and electric bus/trolley) are HIGHER than the equivalent costs for the average automobile, and that substituting these forms of transit for the average automobile would increase costs to society rather than reduce them.

  49. EngineerScotty February 21, 2010 at 12:26 am #

    The CV of VTPI principal Todd Litman is here, and his contributions to the literature (including numerous publications in peer-reviewed journals, and numerous references to the above-referenced paper by peer-reviewd journals) can be found at this Google Scholar query. Bottom line–Litman is a serious researcher and is qualified to speak on the subject; he’s not an amateur publishing bogus “research” from his basement. Lots of authors in the subject area (on all sides) are affiliated with think tanks and professional firms rather than universities and public institutions, many experts in the field have vast professional experience rather than doctorates and academic criteria. I’d be careful casting that particular stone.
    Given that your ad hominem attack on VTPI’s credentials appears to lack merit; do you have any methodological criticism of his cost estimates for autos?
    With regard to the transit cost estimate–yeah, we know already busses are subsidized. For the US, at current levels of ridership, per-rider costs are indeed high–due to many services running empty much of the time. Litman addresses this in the report; noting that the incremental cost per additional rider is quite low for transit. This only re-iterates the point I (and others) have been making–complaining about per-rider subsidies for low-ridership transit services misses the point–efficient transit operation requires higher levels of ridership, which is difficult to achieve given chronic under-investment in transit infrastructure and the underpricing of driving. The incremental cost (to society) for an additional riders is negative in most cases–having one more person ride the bus doesn’t increase the agency’s costs, but does increase the agency’s revenue. Each additional car on the road, OTOH, places additional burdens on society.

  50. Danny February 21, 2010 at 8:52 am #

    There is no doubt that cars have a level of flexibility that no transit system can replicate, and even at a cost that we consider ridiculous, many people are still willing to pay (Even at high tax rates, England still has a large share of car use).
    Cars are desirable without a doubt. But there is a HUGE fallacy in Watson’s idea about relative subsidies. It has to do with two major economic relationships between transit, cars, and their passenger efficiencies.
    The first economic relationship is that cars are a normal good (increased prosperity equals increased use), whereas transit is an inferior good (decreased prosperity equals increased use). The second relationship refers to efficiency. In both transit and cars, efficiency per passenger goes up with ridership…but only transit is intended for high ridership. For the most part, car costs are proportional to use, whereas transit costs are inversely proportional to use.
    Now that we know these two relationships, we can understand Watson’s fallacy. But right now, I am going to resort to hypothetical numbers, because I don’t have the time to research and pull up actual numbers. I’ll estimate, but in the case of failed estimates, the principles still stand but the numbers will be a little different.
    Lets say in our hypothetical city that cars have a 95% modal share and transit has a 2% modal share, the rest filled with minority modes like biking or walking. Lets also say that transit is currently functioning at an average 25% load factor, and at a 50% load factor transit is capable of breaking even, meaning that currently it is 50% subsidized. Lets also assume that cars are subsidized, but only 5% or so.
    You see, because cars are a normal good and transit is an inferior good, if the costs of cars go up, people will switch to transit. And since 2% modal share of transit results in a 25% load factor, we can assume that 4% modal share can result in a 50% load factor, which will put the transit subsidy at zero (fully funded by user fees).
    So if the elasticity of demand for car use is greater than o.4, then eliminating the small subsidies (5%) for car ownership will result in an elimination of the large subsidies (50%) of transit use. And that is Watson’s fallacy.

  51. John February 21, 2010 at 10:04 am #

    Grade separation is a good thing in almost every application. At-grade systems (light rail mostly) have the issue of being at the whim of traffic, which is not exactly a good way of drawing people out of their cars. Grade separation and automation are good things for transit.

  52. Watson February 21, 2010 at 11:46 am #

    Scotty,
    Sorry, but whatever work Litman may have done elsewhere, the VTPI does indeed appear to be one man with a website working out of his home, and the report in question is self-published on that website. If it has been published in a peer-reviewed academic journal or by a government agency, please provide the citation.
    And my point is not about “low ridership” transit, but about transit overall. As I said before, the evidence is pretty conclusive that, per passenger-mile, direct public subsidies to transit users are much, much higher than direct public subsidies to motor vehicle users. In response, you suggested that motor vehicle users receive higher indirect subsidies than transit users, due to external costs of motor vehicle use related to pollution, land use, public services, etc. But you have produced no evidence to support that claim, let alone to show that indirect subsidies to motor vehicle users are so much higher than indirect subsidies to transit users as to offset the much higher direct subsidies to transit users. The VTPI report’s finding on external costs undermines rather than supports your claim.
    Nor have you produced any evidence that transit would be more efficient, or transit subsidies lower either in total or per passenger-mile, if ridership were higher. How do you propose to produce a significant increase in transit ridership without significantly greater public spending on transit?

  53. Watson February 21, 2010 at 11:59 am #

    Danny,
    I have no idea why you think increasing transit’s modal share from 2% to 4% would necessarily, or even just probably, “result in” a proportional increase in transit’s load factor, or indeed any increase in load factor at all. Or why you think eliminating a 5% subsidy for motor vehicle users would somehow result in the elimination of a 50% subsidy for transit users. What policies, exactly, do you think would achieve these outcomes?

  54. danny February 21, 2010 at 12:16 pm #

    Not a single policy. Its a simple relationship between two types of substitute goods, one normal and one inferior.
    If transit has a two percent modal share, and increases to 4 percent, then that is a doubling of transit passenger miles. The resulting jump from 25% to 50% is the result of doubling the number of transit passenger miles, assuming that total passenger miles has stayed the same, and assuming that transit capacity has stayed the same.
    The price elasticity of demand for car travel shows how quantity demanded changes with price. If price elasticity is .4, that means that due to a 5% increase in price, 2% will switch modes. Any price elasticity higher than that will result in even larger modal switch.

  55. Watson February 21, 2010 at 1:03 pm #

    If transit has a two percent modal share, and increases to 4 percent, then that is a doubling of transit passenger miles. The resulting jump from 25% to 50% is the result of doubling the number of transit passenger miles, assuming that total passenger miles has stayed the same, and assuming that transit capacity has stayed the same.
    But you can’t assume that. How do you propose to double transit passenger-miles without any increase in transit capacity? More generally, how do you propose to increase transit passenger-miles without a proportional increase in transit capacity?
    If price elasticity is .4, that means that due to a 5% increase in price, 2% will switch modes.
    More unjustified assumptions. What is the actual price elasticity of car travel? And what is the basis for your assumption that the response to a price increase for car travel would be mode-switching to transit rather than any number of other potential responses — switching to a cheaper car, carpooling, combining trips, walking, bike-riding, telecommuting, etc., or simply foregoing travel?

  56. danny February 21, 2010 at 2:56 pm #

    Transit capacity expansion isn’t needed for most US cities. Most are running hopelessly below capacity, meaning that we can have large increases in ridership without expanding capacity.
    Price elasticity isn’t something that can be calculated stochastically, because elasticity changes at different price levels. There have been multiple studies of price elasticity at different times and for the most part the numbers range between .2 and 1.5.
    Transit has long shown to be the most common substitute for car transportation. Walking and cycling substitutions have only taken place in measurable amounts during oil price spikes, and when they do happen, it is often in places with inadequate transit. But you are right that it might not a perfect assumption, just the most likely by a margin that is at the very least measurable.

  57. Watson February 21, 2010 at 3:37 pm #

    Transit capacity expansion isn’t needed for most US cities. Most are running hopelessly below capacity, meaning that we can have large increases in ridership without expanding capacity.
    You’re not answering the question. Merely observing that there is unused capacity in the current system does not explain how to increase ridership without increasing capacity. So, again, how do you propose to do this? And specifically, how do you propose to DOUBLE transit ridership without any increase in transit capacity?
    Transit has long shown to be the most common substitute for car transportation.
    More unsupported assertion. Where has this been shown? Give us some links. And you seem to be assuming that transit would be substituted not merely for “most” of the passenger-miles lost from car travel, but all of them, which seems even more implausible.
    Price elasticity isn’t something that can be calculated stochastically, because elasticity changes at different price levels. There have been multiple studies of price elasticity at different times and for the most part the numbers range between .2 and 1.5.
    Well, let’s use your scenario of a doubling of transit ridership from the current level. How much of an increase in the price of driving do you claim would be necessary to induce this level of increase in transit usage (again, per your assumption, without any increase in transit capacity)? Whatever your answer is, show us the evidence on which it is based.

  58. Danny February 21, 2010 at 5:51 pm #

    I already told you, my numbers were hypothetical. They are meant to show the relationship between two different types of transportation substitute goods, a normal good with costs that scale proportionally with use, and an inferior good whose costs scale inversely proportional to use. Given current modal shares which are heavily skewed towards car use, it would take a very small amount of people to switch to increase ridership of transit to sustainable levels. The hypothetical situation was meant to illustrate a simple concept: That a small amount of subsidy for the normal good will result in a huge subsidy of the inferior good…and likewise, making cars only slightly more sustainable will go a long way toward making transit sustainable.
    So yeah, if you don’t like subsidies for transit, get rid of subsidies for its substitute (cars), no matter how big or small those subsidies are, and then we can talk about transit subsidies.
    If you want to see real numbers, find the transit capacity utilization rates of cities where the cost of driving is high, and compare to any city where the cost of driving is low.
    Anyone who frequents the comments of this blog will know I am a huge critic of transit operations across the country. I hate their cost structures, I hate their subsidies, and I hate the governments that operate them. I can sympathize with your hatred of subsidies. But the answer to subsidies for transit lies at least partially in the reduction of subsidies for its substitute, and you are pushing the cognitive dissonance pretty hard to be able to deny a simple concept.

  59. EngineerScotty February 21, 2010 at 6:18 pm #

    Watson,
    Much of the work done by the ITS at UC-Davis is also self-published by the university–if you are to suggest that only journal articles are credible evidence of anything, a whole lot of stuff is off the table. Delucchi, in particular, has lots of self-published (by the University, rather than in a journal) work–and both he and Litman have sufficient academic credentials and experience that their self-published works deserve the benefit of the doubt. (Not to its accuracy, of course–feel free to debate the merits of the paper; but to its provenance and the fact that it represents a good-faith contribution to the intellectual record–Litman is neither a crackpot or an amateur; his work is serious work and merits a serious response).
    The report in question has numerous cites in the literature–some of them by Litman in other works, but many of them by independent authors.
    Given the nature of the report, there are several reasons why self-publication is not inappropriate. 1) It is written largely for a lay audience–activists, decision-makers, and the general public. 2) The author wanted to maintain copyright and make the report freely available–most journals which conduct formal peer-review require surrender of copyright as a condition of publication, and many journals keep their articles behind a paywall. 3) It’s too long for most journals to accept. 4) It doesn’t contain much in the way of original research–instead it summarizes and analyzes research that has already been done.
    So–pretty please with sugar, can you comment on the merits of the report, rather than trying to dismiss it as unworthy of consideration?

  60. CroMagnon February 21, 2010 at 8:32 pm #

    Finally, the Winter Olympics in Vancouver are showcasing a technology and modality I’ve been promoting for years as viable regional or rapid transit system; best used in cases where light rail cannot acheive the speed, capacity, or reliability and where full manual metro with long trains might not be justified.
    Light Rail proponents in this country are an honest to god political machine with the one size fits all mentality. The LRT governt bureaucrats can never come up with a convincing reason why automated metro isn’t a better idea than LRT in some cities. In Baltimore, they never considered it in the scoping for Alternatives Analysis of a recent transit proposal!
    It really is all about time savings: Speed is time savings; reliability is time savings; frequency is time savings. And “just show up whenever” WILL get the ridership densities to blast LRT advocates unjustified “preference” for surface service. End rant;) Thank you. I repeat myself.

  61. Alon Levy February 21, 2010 at 9:08 pm #

    The study that concluded that direct public subsidies to motor vehicle users are about 1 cent per vehicle-mile is Do motor-vehicle users in the US pay their way? by Mark Delucchi of the Institute of Transportation Studies at UC Davis.
    That study says 20-70 cents per gallon. Cars don’t get 70 mpg.
    GTE.

  62. EngineerScotty February 21, 2010 at 9:36 pm #

    Light rail proponents an honest-to-God political machine?
    Huh?
    There are certainly some urbanists who prefer surface rail to grade-separated; but I’d wager that the deciding factors between LRT and grade-separated metro (driverless or no) are, in order:
    1) Sticker shock–much US transit is built on the cheap. We see this now in Honolulu, where there have been calls to convert a $5 billion elevated line to a surface LRT in order to save money on capital costs.
    2) Opposition from transit unions. For obvious reasons, bargaining units representing transit workers–who generally tolerate driver-operated rail lines (despite one advantage of rail being lower labor costs), will fight any proposal for driverless metro tooth and nail.
    3) Opposition to elevated rail for cosmetic reasons. Els have a bit of a bad rap in the US, given that quite a few American els are old, ugly, noisy installations through blighted neighborhoods. More intense coverage of SkyTrain may help here.
    Compared to THAT, the romantic notions of streetcar-loving urbanists probably don’t amount to a bucket of warm spit. And the companies which manufacture light rail trains (Bombardier, Siemens, etc) are also more than happy to manufucture heavy rail, including driverless systems; unlike the bus-vs-rail case, where you have competing manufacturing lobbies (not to mention oil companies fighting against anything that’s electric-powered), that simply isn’t the case for LRT vs metro.

  63. CroMagnon February 21, 2010 at 10:23 pm #

    EngineerScotty, the actions of the current urban planning establishment in the US involved in transportation planning resemble a political machine.
    This is about the groups [big developer leagues who hire planners/architects] who are trying to sell transit to cities (citizens, really)–and it’s almost always rail transit and surface transit of the LRT/streetcar variety, b/c they likely figure it’s the fastest, cheapest way to get choice riders on transit and generate market interest and government subsidizes to build the new urbanism the urban planners want next to the line. Many in the urban planning field are rather glorified architects than anything else. Certain present company aside, of course, but there are far too many people promoting transportation plans and projects who don’t understand or care to understand math, space, physics, etc.
    Obviously, the manufacturers don’t care what they build. Politicians not worth their salt will want to build anything that gets done fast enough to occur while they’re still in office and not have to raise taxes too much if any, ergo more LRT. LRT is sold too much by the Machine because they claim it’s good, but it’s just a cover for easier and faster, so the developers/planners/architects can move on to the next money making project faster. I guess my rant wasn’t over.

  64. David Baker February 21, 2010 at 10:59 pm #

    Copenhagen’s new subway is also driverless, with very frequent service. Also the trains, and thus the station platforms, can be shorter, thus simpler and less exprensive. Very smart and the way to go all around.
    Obviously these systems need to be grade separated. A very good thing, much faster, and it also avoids the issue of the significant hazard to cyclists that LRV tracks in the street represent.

  65. EngineerScotty February 21, 2010 at 11:25 pm #

    While I do see urbanists having love affairs with streetcars and similar technologists in certain cities (including Portland, where I live), I don’t see a mega-conspiracy to hijack the public transport business in the US for the benefit of urban developers and a few yuppies who long for Paris.
    Light-rail, after all, is a rather diverse technology–it can in some places approach Metro-like service, in others resemble streetcars. That often makes sense, as a solution to the last-stop problem (lines with dense stop spacing and slower speeds at the end, and rapid-transit characteristics in the middle, effectively distribute passenger loads along the length of the line, especially in the absence of an anchor destination). There are many places where LRT provides adequate speed and capacity to meet transit needs, and grade-separated metro is overkill. Of course, the potential of going driverless can change the cost/benefit equation–but the insistence that “the only rail that’s worth building is metro” is like saying “the only roads that are building are freeways”. There are many places where you don’t want a freeway, or an el.
    I had an interesting conversation (on a public blog, so I don’t mind sharing it here) with one of the former directors of Portland Streetcar–he’s an urbanist, but a fair-minded fellow. The subject was the possibility of integrating Streetcar operations with TriMet–and he indicated that it likely wouldn’t happen; but was reluctant to get into detail why. My guess, reading between the lines, is that TriMet views the Streetcar as a bit of a boondoggle–a viewpoint that is somewhat understandable (Portland Streetcar is up-front about their urbanist intentions, and that they view streetcars as placemaking technology moreso than transit technology).
    On the subject of TOD in general, a few good questions to ask concerning any TOD proposal:
    * Who pays? If the developer or the neighborhood is contributing most of the money, its less of an issue.
    * How much of the project is TOD-driven? Lines that pass through greenfields or brownfields on the way to existing destinations don’t bother me so much; its when the entire point of the line (or a segment thereof) is to service speculative development that’s a major red flag.

  66. EngineerScotty February 21, 2010 at 11:31 pm #

    One other comment on the Canada Line, that nobody has mentioned.
    The southernmost section of the line, through the Vancouver suburb of Richmond, is an elevated (and single-track, though its a relatively short stub) line along No. 3 road, the main commercial strip in Richmond. The line essentially replaces a busway that was there before (two exclusive bus lanes, one in each direction, down the median of the street). The street configuration has changed for the SkyTrain line–rather than running down the median, the train route passes along the eastern side of the street, with a pedestrian plaza (albeit bisected by numerous driveways–the place denser-than-average sprawl with strip-malls galore) underneath the tracks.
    Of course, one wonders why they built the busway in the first place–given that any bus heading north to Vancouver would hit mixed traffic as soon as they crossed the northern channel of the Fraser River.

  67. Richard February 21, 2010 at 11:38 pm #

    On a slightly different topic, the Olympics are highlighting the advantages of grade-separated transit in a downtown core. Several downtown streets have been opened up to pedestrians only (closed to cars in other words), which has created basically a giant street party. That would be difficult to do while maintaining service for surface based transit. Both the transit and the massive numbers of people in streets are two of the biggest successes of the games. Hopefully, several will remain car-free after the Olympics.

  68. Pantheon February 22, 2010 at 1:31 am #

    I agree with CroMagnon. Transit in America is about the same thing everything in America is about: greed, money, and power. It’s not even intended to be about what’s best for the people. The notion that public policy should serve powerful, entrenched interests is so ingrained in the national psyche that Americans don’t even see it. And the vast majority of Americans don’t even own passports so they don’t know how bad they have it. That kind of ignorance is exactly what the powerful interests want. If transit truly served the people, grade separation, automation, and extreme frequency would be the norm everywhere.

  69. dejv February 22, 2010 at 1:50 am #

    Several downtown streets have been opened up to pedestrians only (closed to cars in other words), which has created basically a giant street party. That would be difficult to do while maintaining service for surface based transit.

    In fact, surface-based transit mixes really well with pedestrians. There’s only one problem – it must crawl at approximately twice the pedestrian speed so long routes through pedestrian streets kill average speeds.

  70. Vonny February 22, 2010 at 2:27 am #

    Dejv,
    I am not sure you encompass what was meaning Richard: so here a link to give you a better idea of what is a “pedestrian street” those day in Vancouver:
    http://www.youtube.com/watch?v=FzVuLTPYtk4
    The link show Robson street which in the original ancouver transit agency plan was supposed to be open to transit service…
    here is another link showing a blocked trolley at Robson and thurlow due to crowd in the street
    http://vimeo.com/9448409

  71. EngineerScotty February 22, 2010 at 11:07 am #

    A lot of stuff to unpack in Pantheon’s comment. While lots of things in the US are certainly about “greed, money, and power”–transit is hardly unique in this regard, and neither, unfortunately, is the US. There’s plenty of greed and corruption to go around in London, Paris, Berlin, Beijing, Tokyo, Mumbai, or anywhere else–some of it is camouflaged by an economy not in recession, or by a less free press, or by imperialist behavior (which causes the have-nots to be located apart from the haves). The quality of a region’s transit, however, is not a reliable indicator of corruption or lack thereof. (Chicago has a fairly robust transit system, after all…)
    I DO own a passport, have travelled abroad quite a bit, and have experienced quite a few country’s transit systems; please don’t engage in patronizing “dumb yankee” arguments. Perhaps that wasn’t your intent, but this is a serious forum here.
    For better or worse, the US has lots of grade separation–unfortunately, we’ve chosen to grade-separate highways rather than rail in most cases. If you take your last sentence, Pantheon, and translate it to the highway context, it becomes “If highways truly served the people, grade separation and extreme speeds would be the norm everywhere.” Guess what? That’s exactly what we did. Like the result? Neither do I.
    This has generally been done with the full support of the majority of the electorate. If you think the public is brainwashed you are entitled to that opinion–I generally consider “the public r stupid” arguments to be a sign you’ve lost the debate already. That said, in this country there is a tremendous amount of inertia for autos, and for most voters, the individually-rational choice is more of the same–at least until the next oil shock hits. Some anti-transit critics push this as the collectively rational choice as well–Watson’s arguments above essentially reduce to “we’ve build auto infrastructure (low density land use, highways everywhere, refueling stations everywhere) for the past 100 years. It’s too expensive to build a competing comprehensive transit network”.
    Obviously, neither you or I consider that the COLLECTIVELY rational choice (for many reasons), and persuading the public to suspend self-interest for the greater good is a hard thing to do. (With good reason–many scoundrels and demagogues make similar arguments concerning their plans–you’ll notice that the current public mood in the US is HIGHLY skeptical of proposed reforms; they’re well aware of the corruption of power, and suspect that the current administration is little different than the previous one). But right now, the world we live in is one where public support for transit is weak. In the bizarro-federalist world of US politics, where most tax dollars go to Washington rather than to local governments local transport officials have to go begging in DC for transit dollars (and jump through the numerous hoops, including quite a few which seemingly exist to limit the amount of transit, particularly rail, that does get built), the ability of local governments to spend local funds is severely constrained. It’s not surprising high-quality grade-separated rail seldom gets built; the rules in DC are designed to discourage it.
    You’re right, Pantheon, that we’ve got major problems. It’s a far larger systemic issue, though, than local developers angling for a trolley past their land; or transit unions fighting to keep jobs. Transit interests frequently fight over crumbs, and your comment is joining the scrum down on the floor. You–we–would be better off trying to get a seat at the table where the steak is being served.
    Until that happens, there is no mix of transit technology that will successfully displace the automobile.

  72. Watson February 22, 2010 at 6:27 pm #

    Scotty,
    I’m not really interested in a lengthy debate about the merits of the VTPI. I explained why I think its work should be treated with skepticism. But the more important point is that the VTPI report in question doesn’t support your position anyway. Litman found that transit has HIGHER external costs than autos, not lower. As far as I can tell, you still haven’t produced any serious analysis indicating that motor vehicle use is subsidized anywhere near as much as transit use.
    Alon Levy,
    The figure for direct public subsidies is the lower one – about 20 cents gallon, which is about 1 cent per vehicle-mile at 20 mpg, and about 0.7 cents per passenger-mile at an average vehicle occupancy of about 1.6.

  73. Watson February 22, 2010 at 6:39 pm #

    Danny,
    I don’t “hate” transit subsidies. I think a decent case can be made for some subsidies to transit in general on social justice grounds, and to transit in particular places and times on other grounds (congestion relief, etc.).
    I just get tired of complaints from transit proponents about subsidies to automobile users, given all the evidence that transit subsidies are much, much greater.

  74. David February 22, 2010 at 10:04 pm #

    @Paul,
    Fair point about maintenance. That being said I’m curious as to whether or not it’s possible for that work to be completed during the downtime on Sunday through Thursday nights when the night life crowd won’t (generally) be as large. That way the service could still be 24 hours on Friday’s, Saturday’s and the occasional other night like New Years or Halloween.

  75. Pantheon February 23, 2010 at 2:38 am #

    EngineerScotty,
    1. Corruption is everywhere. But the whole system of governance in America is designed to serve the wealthy elite. This is due to numerous factors, like the supreme court defining a corporation as a person, and therefore entitled to unlimited “speech” in elections, a constitutional framework with competing branches of government, a court system that consistently interprets the constitution in a manner that favours the capitalist elite, privately funded campaigns which ensure that half the congress are the wealthy elite and the other half are owned by them. And the litigious culture that allows wealthy interests to block or delay things (the Seattle LRT was almost destoyed by lawsuits that dragged on for years). The Constitution itself is an anachronism, in that it was designed to cripple governmental power for the sake of small landowners and slaveholders in an agrarian economy, but now that benefits transnational corporations. I think the idea (so cherished by Americans) that 21st century policy should be dictated by the petty prejudices of 18th century slaveholders will be shown by history to be deeply misguided. These political and legal structural issues simply do not exist in other countries, no matter what you may wish to believe.
    (If you want a specific example that relates to transit, research Norman Braman – a billionaire owner of car dealerships in Florida – who waged a successful campaign in 1999 to kill a one cent sales tax increase that would have funded a $16 billion expansion of the public transportation system).
    2. If you find my arguments patronizing, that’s not my problem. Refute their content, don’t simply state that you are offended. It is possible for a statement to be both offensive and true. In fact, the more offensive statements are often the true ones.
    3. I never said highways served the people. In fact I believe the entire highway and road system is part of a deliberate conspiracy to serve the steel, oil, and auto companies who profit from it, and impoverish the working classes, thereby keeping them at subsistence wages and ensuring their continued status as slaves to the wealthy elite. Lower class people in America often live in places where driving is the only option, and the cost of buying, fueling, repairing, and maintaining their automobile eats up all of what would be their discretionary income or savings. It is not a coincidence that America has the most developed highway and road system in the world.
    4. The American public has a more provincial mindset than any other country in the western world. America and Australia are the most culturally isolated countries in the western world, due largely to geographic isolation. However, Australia’s cultural isolation is softened somewhat by its historical ties to Great Britain as a commonwealth country. My own subjective opinion is that I find Americans to be incurious about the wider world, to a level that I have not encountered anywhere else. You may disagree, but it is fallacious to say “if someone makes Argument X, then it is a sign they have lost the debate”. That is a rhetorical tactic, nothing more. I could just as easily say “if someone writes a seven paragraph response (as you did), then it is a sign they have lost the debate”. That is a meaningless statement. Maybe Argument X is correct!
    5. You go into great detail but sometimes I think you miss the forest for the trees. You admit “the rules in DC are designed to discourage [quality grade-separated rail]”. But did you ever consider why the rules are designed to discourage it? If you stop to think about the why, you might surprise yourself and end up agreeing with me.
    In 2008, the B.C. government overruled a prior decision to make the Evergreen line LRT, and made it ALRT (Skytrain-like) instead. They explained that ALRT would result in higher construction costs, but significantly lower annual operating costs, higher frequency, signficantly shorter travel times and 2.5 times more ridership by 2021. The B.C. government’s goal is to double transit mode use by 2020, and they found that only ALRT technology was consistent with that goal.
    http://www.coquitlam.ca/_Media+Centre/News+Releases/Transportation+and+Transit/ALRT+Technology+Recommended+For+Evergreen+Line.htm
    Why isn’t that kind of political commitment, foresight, and leadership shown south of the border?

  76. Alon Levy February 23, 2010 at 2:51 am #

    The figure for direct public subsidies is the lower one.
    What’s a direct subsidy, how does it differ from the other kind of subsidy, and which figure is comparable to your favorite farebox recovery ratio for US transit?
    For the record, even 20 cents translates to US roads having a recovery ratio in the 40s. 70 cents puts it in the high teens. And, until the magical pollution-free, accident-free flying car rolls out and dispenses ponies to every child and sells cheap bridges to every adult, the externalities to be tacked on top of the 70 cents are brutal.

  77. Greg Finley February 23, 2010 at 11:49 am #

    I wish they would adopt this in DC, but I’m not holding my breath.

  78. EngineerScotty February 23, 2010 at 1:38 pm #

    Pantheon,
    1) While the Citizens United decision is terrible in many ways–it’s also barely a month old. Which doesn’t rebut your larger point, but average citizens are capable of taking power when they want to. One of the disturbing political trends which enables the wealthy elite in this country, is the activist left’s continual tendency to drop out of meaningful political participation (i.e. not voting) whenever the D’s are in power and the progressive agenda is not satisfied. (You see this all the time on DailyKos and similar forums). There’s tons of posts along the lines of “I’m through being a democrat”, when what they should be focusing is electing better democrats.
    Parliamentary systems have a bit of an edge in better diffusing power, for two reasons–1) third parties can form more easily and be viable, where our system of elections makes it difficult; 2) the ability of the PM to call a snap election–or a vote of no confidence to cause one to happen on short notice, makes representatives more aware of the public mood, while not putting them in full campaign mode. In the US, everyone knows when the next election is, and everyone’s planning for it.
    Finally, I’ll point out that politics in the likely next superpower are far more undemocratic, corrupt, and power-benefiting than US politics can dream up. Despite the recent communist history of the People’s Republic of China–today that country is a crony capitalist state that would make Texas blush. Not to excuse anything in the US, but it’s important to remember who the competition is nowadays.
    (Finally, I’ll note that much of the red tape required for infrastructure construction, was in large part a response to the freeway buildout a generation or two ago, when neighborhoods were bulldozed and nobody could do a thing about it.)
    I know you don’t consider highways to “serve the people”–many do, though, including those who build them. And the US polity tends to view highways are a public good moreso than it does transit–transit still gets viewed as a social service or an entitlement, unfortunately.
    I won’t argue about the cultural isolation of many Americans–I was objecting to the suggestion that this American is clueless about the world–I’m not, thank you. 🙂
    I think we agree on some of the big picture more than we disagree. OTOH, you seem annoyed that TriMet is building LRT and–when the economy goes down–cutting bus service.
    I’m more annoyed that TriMet has to make that choice to begin with.

  79. Watson February 23, 2010 at 2:02 pm #

    Alon Levy,
    What’s a direct subsidy
    By “direct public subsidy” I mean public funds used for the direct provision of the goods and services consumed by the user. For motor vehicle users, this mostly means roads and highways and associated infrastructure (bridges, tunnels, signs, stoplights, etc.). For transit users, it mostly means fixed transit infrastructure (railtrack, stations, etc.), transit vehicles, and transit operating expenses. The evidence is pretty clear that transit users receive much higher direct public subsidies per passenger-mile than motor vehicle users.
    If you think you can show that motor vehicle users receive higher indirect subsidies than transit users (e.g. military protection of oil supplies, environmental costs of fuel consumption, etc.) then I invite you to do so. And by “show” I mean produce a serious, comprehensive, quantitative analysis from a reputable source. I’m not interested in more speculation and guesses.

  80. Paul February 28, 2010 at 8:59 pm #

    @David
    Fair point on the idea of possibly leaving it open for Friday and Saturday nights and possibly also festive holidays. I should of expanded on my definition of maintenance. I think of most of what they do is just preventative maintenance. Going around checking the switches and making sure they work and are fully operational. Also looking for signs that failure is possible. Sure they could possibly skip it once or twice a week. At the same time if the next day a switch or something failed. Well heads would roll if it was found out due to poor preventative maintenance. In a sense they are in a dammed if you do and dammed if you don’t dilemma
    @Cromagnon
    I fully agree on what you said about the skytrain technology and how they have worked during the Olympics. To me it wouldn’t have run so well if it not automated. The high frequency levels enabled the crowds to move much quicker and lines up to dissipate quickly.

  81. Greg March 2, 2010 at 11:33 am #

    Here’s a clip from Metro Vancouver’s The Sustainable Region TV show about how the Skytrain system works:
    http://www.youtube.com/watch?v=e1gE4aoJRw0

  82. Nathanael March 11, 2010 at 10:11 pm #

    “If you think you can show that motor vehicle users receive higher indirect subsidies than transit users (e.g. military protection of oil supplies, environmental costs of fuel consumption, etc.) then I invite you to do so. And by “show” I mean produce a serious, comprehensive, quantitative analysis from a reputable source.”
    DOE carbon dioxide emissions numbers. Next question.

  83. Jarrett at HumanTransit.org May 26, 2010 at 12:36 am #

    Curt Sampson submitted this comment:
    “Driverless trains raise all kinds of anxieties. Many people like knowing there’s someone in charge on the vehicle, and imagine that this person will be useful in emergencies.”
    On most rail systems, perhaps excepting streetcars, whether or not there is a person on the train you can talk to during operation has nothing to do with whether or not there is a driver. For fairly obvious safety reasons drivers cannot handle interactions with passengers while driving the train. Nor (as I said, with the exception of streetcars) would they be likely to stop the train between stations due to an onboard incident; it’s almost invariably safer, and you’ll get help faster, by continuing to the next station.
    If there is any on-board person with whom passengers can interact, it’s going to be the conductor. And of course whether or not you chose to have a conductor has nothing to do with whether you have a person or an automated system driving.
    “The lack of a driver is the key to those extreme frequencies.”
    Actually, the lack of a driver combined with low demand during those times. Most Tokyo subway lines have similar high frequencies during the very early morning and late night, even though every train has both a driver and a conductor who are well paid.
    Those quibbles aside, I quite agree with your article. I’ve lived in areas with and without high-frequency public transport, and it makes a world of difference not to think about checking schedules.
    — Curt Sampson http://taihendaro.blogspot.com

  84. Dave 2 July 1, 2010 at 10:50 pm #

    Vancouver’s Skytrain has no worse than 3-4 minute service on the Waterfront-Columbia line all day from 5 AM to 1 AM (ish), not many western North American LRT systems can boast such high frequency.
    (which, now that I review this post, was the main point of this entry)
    [Curt Sampson? flash back to van.general c. 1996]