Design to Fail: Of Supertrains and Sprawl


We must face the prospect of changing our basic ways of living. This change will either be made on our own initiative in a planned way, or forced on us with chaos and suffering by the inexorable laws of nature.” – Jimmy Carter (1976)

Cities built as sprawl require every trip of significance be in an auto. If CO2 pollution and foreign oil consumption increase, then every commute is a failure. Why design such failure into the fabric of our cities? Says one sprawl critic (Jane Jacobs): “The pseudo science of planning seems almost neurotic in its determination to imitate empiric failure and ignore empiric success….to put it bluntly, [sprawl planners] are all in the same stage of elaborately learned superstition as medical science was early in the last century, when physicians put their faith in bloodletting.”

One recent such sprawl-enabling designed-to-fail project is California’s proposed high speed “supertrain.” This would take years and cost billions to build. After that, it might provide expensive rides between northern and southern California for a few passengers, but would still require auto commutes to and from the terminal, and rental cars when people arrived at their destination, at least outside San Francisco.

The alternative would be increasing local transit and building the land-use to support it. I say do that first, then build high-speed rail. This alternative is far cheaper and more effective. How expensive would local transit amenities be, compared to sprawl infrastructure? Portland built its entire 300-mile network of bikeways for the cost of a single mile of urban freeway.

Curtiba bus stop

We could even emulate Curtiba (Brazil). Curitiba received funding for a “heavy rail” project like high-speed rail, but discovered that heavy rail is ten times more expensive than light rail, and light rail is ten times more expensive than buses, even with dedicated busways. They wanted the 100-times larger transit coverage buses offered, but were concerned about the downside of buses: the expense of hiring drivers. While trains can add capacity for a single driver by adding more cars, buses were confined to 40 – 80 passengers per driver. Not having learned “design to fail,” Curtiba worked with bus manufacturers to make “accordion” buses–that triple the length of a bus for a single driver, carrying nearly 300 passengers.

As part of their design-to-succeed, Curitiba’s land use planning also supports ridership. The stops are nice, not just a sign stuck in the side of the road. City planning insures many potential transit customers are near the stops too, and have sidewalks that reach them — none of which is true for sprawl

Incidentally, far from being a “big government” solution, Speedybus itself is a public-private partnership. The public owns the routes and sets fares; private investors own the buses. Far from being subsidized, the system generates nearly two million fares a day from a city of 1.8 million, and turns a profit.

Designing cities to succeed, and support transit is not rocket science; we did so before 1950. The elements of success include pedestrian access, particularly to transit stops, so … sidewalks and mixed-use neighborhoods. Mixed-use means stores, residences and offices, perhaps even light industry, all coexist in a single neighborhood. So when you get off the bus, there’s actually someplace to go.

The good news is that recent moves by California’s legislature like AB 32 make such mixed-use, pedestrian-friendly planning something we’ll see more of….if we don’t spend all our money on expensive politically “sexy” design failures like high speed rail first.

Adam Eran


  1. says

    I’m looking forward to high-speed rail. There’s supposed to be a link between LA and Riverside. It would make it feasible to visit relatives without driving (or with less driving).

    It’s probably not economical if you have several bodies to transport, but for solos and couples, I bet it will be popular. Metrolink is already quite popular.

    HSR is totally different from buses. We need more buses too, but that’s a different thing.

  2. Joe Weinstein says

    Eran’s conclusion is not wrong, but it is absurdly argued.

    As Stevan T. observes, long-distance transit options (hi-speed rail or air or bus or whatever) do NOT compete with local transit options.

    Eran utterly misrepresents the stakes by labeling the proposed hi-speed rail system as ‘sprawl infrastructure’. Indeed, thanks to the limited number of stations and their distance apart in space and time, the system will not provide or even promote new large-scale everyday commuting options.

    Eran complains that to get to or from the few hi-speed-rail stations you would need other modes – rental cars, buses, or whatever else locally is available. True, but so what? The same is true of every other limited-station long-distance transit mode – e.g. air or even inter-metropolitan bus. Like these other modes, hi-speed rail gets you from one station to another on its network. The rest is up to you in any event.

    However it IS true (but NOT because hi-speed rail is ‘sprawl infrastructure’) that a given amount of bucks for better transit could do a lot more for more people for more days of the year if invested in local intra-metropolitan rather than long-distance inter-metropolitan transit. That fact is precisely why I parted company with many fellow enviros a couple years ago and voted against the hi-speed rail measure.

    But no one has shown that hi-speed rail is a worse investment than other stuff we have been spending massive moneys on anyhow – e.g. the California penal system and substances prohibition. Or that local transit is a better investment than sectors we are under-spending in, arguably education or even nutrition.

    Contra a Harvard biz school theory (whose widespread adoption during the 1990s and early 2000s led to the info-tech craze and then the dot-com bust), no person – or society – should be so smartly stupid as to put all her discretionary investment capital in precisely the one and only highest-return-du-jour sector.

    In the real world of personal or social investment choices, it is folly to focus all investment on what looks like the ‘highest-yield’ ‘best’ or most ‘important’ sectors. In a reasonably diversified social investment portfolio, both transit sectors – intra-metro and inter-metro – arguably merit some investment.

  3. George says

    Given the choice between a fast train and the TSA, I would take the train anytime. However, it is true that fast rail is probably not worthwhile unless it can share tracks with freight, as in most of Europe. Then the major problem is attracting enough riders without improvements to urban transportation. It takes us 1.5 hours to reach Union Station by the clumsily designed Metro (a product of arrogant county government) and about as long by taxi and airport bus and long walk through the station. That provides the best argument for using buses on the already available subsidized highways full of trucks hauling stuff that should be on trains.

    But who would ride a bus to San Francisco, even matching the total transit time of TSA-infected air travel? Few use urban buses in America because they have complex routes and fare structures, and are inconvenient, slow, irregular and noisy. I had the experience years ago of walking a block before using a transfer and having it refused because I did not get on at the correct stop, and have never ridden one since. Now I only use unavoidable Amtrak buses that fill gaps in rail travel. A trip to San Francisco would probably include a stop somewhere with a choice between McDonalds and Burger King, or equivalent. Yuck! Even American trains have better food.

  4. Stevan Thomas says

    High speed rail is not intended to replace local transportation, it is a link between Northern and Southern California, e.g., between San Francisco and Los Angeles. To conflate this with local light rail and bike routes is absurd.

  5. Steve Lamb says

    Well I didnt read the whole article because the first sentence is BS, so I assume what follows must be also.. The FIRST AMERICAN SUBURB WAS BUILT IN CONNETICUT IN THE 1840’s, a WEE BIT BEFORE THE AUTOMOBILE and was accessed BY RAIL.

    The Los Angeles/Southern California region was laid out in suburbs accessed BY RAIL and serviced locally by streetcar. The writer EITHER is uniformed totally about the history of American urbanism and suburbanism, or he is a shill for developers seeking dense urban downtown projects for their profits, but in any case, he is just plain dead wrong. Suburbs DO NOT REQUIRE the automobile.

    Suburbs do of course imply things that some social planners fear: Freedom, autonomy, less control by planning elites, but they do not REQUIRE the automobile….Go back to school.

    • Adam Eran says

      Sorry dude, but street car suburbs do not qualify as sprawl. Sprawl suburbs are distinct since they offer no pedestrian access to the train stop, and insufficient compactness around transit stops, so transit can never attract enough ridership to be economically viable. Sprawl requires autos by definition — and excludes transit.

      Sprawl is also single-use (all residences, all commerce, all office) whereas streetcar suburbs at least have access to multiple uses because the streetcar goes there, and typically have small stores and offices integrated into the fabric of residential neighborhoods. The attraction of light rail is that (in theory) the stops will attract the kind of improvements that make transit viable.

      Meanwhile, the “freedom, autonomy, and less control by planning elites” is anything but what’s available in suburbia (where I live). I’m not free to do without an auto, nor is my child. As for “autonomy,” what does that mean? I treat my own sewage? We live in a complex society, where trash removal, road maintenance and lighting districts abound. Denial is not just a river in Egypt, here.

      But cruelest of all, the “planning elites” are absolutely in charge of sprawl planning, and have resisted genuine urbanism for the nearly three decades that I’ve been paying attention. The sprawl zoning codes are obscure, and require a priesthood of interpreters. These codes are typically as thick as a phone book, and about as clearly written. Contrast this to the plan with which Haussmann built Paris. It was six rather large pages…and he got Paris!

      The epitome of public disempowerment is suburban sprawl. It actively discourages people from getting together and doing anything as a society. Is that important? Think a moment about how little you can accomplish by your lonesome: Do you treat your own water or sewage, vet your own medications, or even sew your own clothes? These little creature comforts, and I’d say your survival, are dependent on others. The fact that you can manage to prize “autonomy” but never mention “cooperation” is as clueless as the contention that streetcar suburbs are the same as suburban sprawl.

      Meanwhile, more bad news about high-speed rail:

Leave a Reply

Your email address will not be published. Required fields are marked *