6 ms·

The main reason US cities rely on automobiles is because the "job watershed" of a private automobile exceeds the job watershed of train systems everywhere in th

The main reason US cities rely on automobiles is because the "job watershed" of a private automobile exceeds the job watershed of train systems everywhere in the US, even New York City. The job watershed is the number of jobs within commuting distance, whatever your tolerance for a commute is, given a transportation modality. In some cities the automobile job watershed is 100x larger than the transit job watershed. Transit systems, especially train systems, in most US cities are hub and spoke systems, centered around terminals in places like midtown Manhattan and the Chicago Loop. If you superimpose that hub and spoke model on top of an employment "contour map" that seeks a level (due to the desire to minimize rents), you'll see that there's a fundamental mismatch between what employers and residents want and the transit network. Rather than a unipolar peak at the transit hub, both employers and residents want to minimize rents which drives sprawl and transportation webs, as opposed to hub and spoke models. Transit has a future with ride sharing and bus rapid transit since those modes use a cheap, flexible network as opposed to an expensive, rigid rail network. Unfortunately, if you wish rail to become more important, its use is actually declining and has been for decades.

tomjakubowski 8y ago
All of that is interesting but the article, and GP, were talking about interurban transportation, not commuting to work. The flexibility of bus networks (using road infrastructure that would needed anyway for commerce) is desirable to be sure in metropolitan areas, but not as applicable or useful to inter-city routes.
- jackcosgrove 8y ago
  Yes, rail is suitable for arterial transportation. However intercity transportation by rail is crippled in the US because there are very few intracity rail networks to receive you at your destination, for the reasons I outlined.
rtpg 8y ago
>both employers and residents want to minimize rents which drives sprawl and transportation webs, as opposed to hub and spoke models. I don't think this holds up empirically? The US is home to some of the highest rents in the world. Mid-sized US cities have way higher rents than major cities like Tokyo or Paris. Or maybe this is the idea, but if so it has failed miserably at letting people on the lower end of the spectrum be able to rent stably
- jackcosgrove 8y ago
  I don't know where you're getting your information, but US real estate rents and prices are low by developed economy standards. Here is a link: http://www.newgeography.com/content/006201-15th-annual-demographia-international-housing-affordability-survey-2019 http://www.newgeography.com/content/006201-15th-annual-demog... "The most affordable major housing markets are in the United States, with a moderately unaffordable Median Multiple of 3.9, followed by Canada (4.3) and Singapore (4.6). Ireland and the United Kingdom both have Median Multiples of 4.8. The major markets of Australia (6.9), New Zealand (9.0) and China (20.9) are severely unaffordable (Table ES-2)."
  - jakecopp 8y ago
    Heads up, Sydney's median multiple was 13 at the end of 2017! I think it's come down a bit but not that much. We do have a pretty awesome train network though, I don't own a car and it's wonderful. Have a look at this graph for different world cities: https://www.abc.net.au/news/2018-01-22/australian-housing-unaffordability-experts-disagree-on-extent/9349796 https://www.abc.net.au/news/2018-01-22/australian-housing-un... [Edit: fixed link]
    - jpatokal 8y ago
      Your link goes back to this page?
      - jakecopp 8y ago
        Oops pasted the wrong link, too much coffee. Thanks.
  - rtpg 8y ago
    This study is focused on price to buy, where a huge segment of the population are renters. This is the core issue with so many of these reports. Focusing on buying already rules out a huge segment of the population (incidentally the ones who have trouble surviving!), and ignores the fact that rent/buy ratios are very different between markets. US buying prices are relatively low because of how the entire economy is based around home ownership. But almost half the population rents.
kylek 8y ago
I think this also has to do with EU cities being much older, and "grew up" prior to automobiles even existing/being wide-spread. American cities/sprawls are designed around the automobile (and maybe some sick idea of "the american dream" pushed by early industrialists)
- helen___keller 8y ago
  You are correct but there are also plenty of case studies in automobile-dependent cities that went all-in on transit, but that was usually due to a crushing and unbearable level of crowding thats simply not legal in any US city.
bobthepanda 8y ago
The reason of automobile dominance is more that unlike other countries, we went all in on highways and just actively dismantled existing rail infrastructure. And then to salt the wounds, our parking requirements are so high that in many suburbs I've been in, the ten minute walk from the front door of Walmart to the bus stop is ridiculous. Rail does better when congestion is high and density is high, but in most of America high density or even medium density is either illegal to build or requires so much parking because of zoning that it does not pencil out financially. The quickest thing America could do to encourage dense construction is to stop the requirement of parking with new construction. Developers do not need to be told a minimum to build, they will build what the market will bear because if they underpark their buildings no one will rent units or office space.
- yellowapple 8y ago
  "Developers do not need to be told a minimum to build, they will build what the market will bear because if they underpark their buildings no one will rent units or office space." The problem with this boils down to 1) That ain't exactly true; in sufficiently-dense cities people will still rent spaces in underparked buildings 2) If you underestimate the demand for parking, it's way more difficult to add parking after-the-fact I do agree that parking lots are absurd wastes of space, though. They're cheaper than parking garages, sure, but they suck on pretty much every other metric. Parking garages are much more space-efficient, and if every building was built atop a parking garage (even just one or two levels), that'd cut down significantly on both the need for parking lots and the need for street parking.
  - bobthepanda 8y ago
    1 and 2 are fine, because if enough people need parking spaces then someone might develop their lot with a parking garage that there is market demand for. Structured parking is extremely expensive; 19K per spot for an aboveground structure [1]. I've heard 30K per spot for underground parking. And parking garages are generally terrible for the streetscape and building, and are difficult to convert to other uses if say, autonomous cars get rid of the need for it later on. If you're building parking based on demand for free parking you'll never build enough unless you want buildings too far apart to walk to the bus stop or their house, which just drives the need for parking up. Pricing parking is a very effective way of managing spaces so that some proportion of them is always free; in my current city, Seattle, new buildings cannot hand parking to residents or workers free with the unit or job, since many people are car-free and all bundling does is socialize the cost of parking. As a result even with no parking requirements many garages are overbuilt (out of town developers not used to developing for local conditions) and the well located ones do a brisk trade in monthly passes and day parking while the residents are out parking their cars at their job. [1] - http://denver.streetsblog.org/wp-content/uploads/sites/14/2017/10/2017-Cost-Article.pdf http://denver.streetsblog.org/wp-content/uploads/sites/14/20...
yellowapple 8y ago
The problem with buses is that they're typically slow compared to even subway/metro rail (let alone inter-city). That doesn't mean they're not the right fit in a wide variety of scenarios (for example, they're about perfect for small/medium cities that can't afford to invest a whole lot in intra-city rail, and they're the only viable option for smaller cities/towns that aren't already stops on a larger inter-city rail line), but they're not necessarily the right fit for a large city with more ground to cover (even with dedicated bus lanes, buses can only - safely - drive so fast, and that speed is typically less than rail). The phrase "bus rapid transit" is kinda nonsensical/oxymoronic; there's not really much that's "rapid" about buses. Ridesharing is slightly more promising, since the smaller vehicles can drive a bit faster and don't have to stick to specific routes. Unfortunately, because they're smaller, they don't carry as many people, so they're less suited for heavy load (i.e. lots of passengers trying to go from the same place to the same place). Rail is probably the most efficient, but also the least flexible and the most expensive (the expense can be mitigated somewhat if you're proactive about building tram rails into your roads, but not all roads warrant that, and by the time they do it's likely already too late; plus, that doesn't do much for older cities). The ideal would be something like the following: - High-speed rail between cities - Metro rail between major parts of cities (and/or between cities and suburbs/satellites) - Full buses on fixed (or at least firm) routes to transit centers in neighborhoods/suburbs/satellites that don't have metro connections of their own (connecting to nearby metro stations, the high-speed rail hub, and/or other bus stations) - Minibuses on more flexible routes for intra-(neighborhood/suburb/satellite) transit (between neighborhood minibus stops, to nearby macrobus/metro/rail stations, or to specific minibus stops outside that particular "zone" in rare cases where a point-to-point really is faster and enough people are trying to do it) All this would be controlled by a rider walking up to some sort of kiosk (or using a smartphone app), picking a destination, and getting a route calculated on-the-fly (for kiosk users: probably printed, or if we wanna be real slick, dispensed on an e-ink display that'll update if the route changes).
- jackcosgrove 8y ago
  Bus rapid transit (BRT) refers to reserving lanes along existing roads strictly for buses, and doing so along stretches with few controlled intersections, or even changing the timing on stoplights to ease the flow of the BRT vehicles. BRT is much faster than traditional bus mass transit.

tomjakubowski 8y ago

All of that is interesting but the article, and GP, were talking about interurban transportation, not commuting to work. The flexibility of bus networks (using road infrastructure that would needed anyway for commerce) is desirable to be sure in metropolitan areas, but not as applicable or useful to inter-city routes.

jackcosgrove 8y ago
Yes, rail is suitable for arterial transportation. However intercity transportation by rail is crippled in the US because there are very few intracity rail networks to receive you at your destination, for the reasons I outlined.

rtpg 8y ago

>both employers and residents want to minimize rents which drives sprawl and transportation webs, as opposed to hub and spoke models. I don't think this holds up empirically? The US is home to some of the highest rents in the world. Mid-sized US cities have way higher rents than major cities like Tokyo or Paris. Or maybe this is the idea, but if so it has failed miserably at letting people on the lower end of the spectrum be able to rent stably

jackcosgrove 8y ago
I don't know where you're getting your information, but US real estate rents and prices are low by developed economy standards. Here is a link: http://www.newgeography.com/content/006201-15th-annual-demographia-international-housing-affordability-survey-2019 http://www.newgeography.com/content/006201-15th-annual-demog... "The most affordable major housing markets are in the United States, with a moderately unaffordable Median Multiple of 3.9, followed by Canada (4.3) and Singapore (4.6). Ireland and the United Kingdom both have Median Multiples of 4.8. The major markets of Australia (6.9), New Zealand (9.0) and China (20.9) are severely unaffordable (Table ES-2)."
- jakecopp 8y ago
  Heads up, Sydney's median multiple was 13 at the end of 2017! I think it's come down a bit but not that much. We do have a pretty awesome train network though, I don't own a car and it's wonderful. Have a look at this graph for different world cities: https://www.abc.net.au/news/2018-01-22/australian-housing-unaffordability-experts-disagree-on-extent/9349796 https://www.abc.net.au/news/2018-01-22/australian-housing-un... [Edit: fixed link]
  - jpatokal 8y ago
    Your link goes back to this page?
    - jakecopp 8y ago
      Oops pasted the wrong link, too much coffee. Thanks.
- rtpg 8y ago
  This study is focused on price to buy, where a huge segment of the population are renters. This is the core issue with so many of these reports. Focusing on buying already rules out a huge segment of the population (incidentally the ones who have trouble surviving!), and ignores the fact that rent/buy ratios are very different between markets. US buying prices are relatively low because of how the entire economy is based around home ownership. But almost half the population rents.

kylek 8y ago

I think this also has to do with EU cities being much older, and "grew up" prior to automobiles even existing/being wide-spread. American cities/sprawls are designed around the automobile (and maybe some sick idea of "the american dream" pushed by early industrialists)

helen___keller 8y ago
You are correct but there are also plenty of case studies in automobile-dependent cities that went all-in on transit, but that was usually due to a crushing and unbearable level of crowding thats simply not legal in any US city.

bobthepanda 8y ago

The reason of automobile dominance is more that unlike other countries, we went all in on highways and just actively dismantled existing rail infrastructure. And then to salt the wounds, our parking requirements are so high that in many suburbs I've been in, the ten minute walk from the front door of Walmart to the bus stop is ridiculous. Rail does better when congestion is high and density is high, but in most of America high density or even medium density is either illegal to build or requires so much parking because of zoning that it does not pencil out financially. The quickest thing America could do to encourage dense construction is to stop the requirement of parking with new construction. Developers do not need to be told a minimum to build, they will build what the market will bear because if they underpark their buildings no one will rent units or office space.

yellowapple 8y ago
"Developers do not need to be told a minimum to build, they will build what the market will bear because if they underpark their buildings no one will rent units or office space." The problem with this boils down to 1) That ain't exactly true; in sufficiently-dense cities people will still rent spaces in underparked buildings 2) If you underestimate the demand for parking, it's way more difficult to add parking after-the-fact I do agree that parking lots are absurd wastes of space, though. They're cheaper than parking garages, sure, but they suck on pretty much every other metric. Parking garages are much more space-efficient, and if every building was built atop a parking garage (even just one or two levels), that'd cut down significantly on both the need for parking lots and the need for street parking.
- bobthepanda 8y ago
  1 and 2 are fine, because if enough people need parking spaces then someone might develop their lot with a parking garage that there is market demand for. Structured parking is extremely expensive; 19K per spot for an aboveground structure [1]. I've heard 30K per spot for underground parking. And parking garages are generally terrible for the streetscape and building, and are difficult to convert to other uses if say, autonomous cars get rid of the need for it later on. If you're building parking based on demand for free parking you'll never build enough unless you want buildings too far apart to walk to the bus stop or their house, which just drives the need for parking up. Pricing parking is a very effective way of managing spaces so that some proportion of them is always free; in my current city, Seattle, new buildings cannot hand parking to residents or workers free with the unit or job, since many people are car-free and all bundling does is socialize the cost of parking. As a result even with no parking requirements many garages are overbuilt (out of town developers not used to developing for local conditions) and the well located ones do a brisk trade in monthly passes and day parking while the residents are out parking their cars at their job. [1] - http://denver.streetsblog.org/wp-content/uploads/sites/14/2017/10/2017-Cost-Article.pdf http://denver.streetsblog.org/wp-content/uploads/sites/14/20...

yellowapple 8y ago

The problem with buses is that they're typically slow compared to even subway/metro rail (let alone inter-city). That doesn't mean they're not the right fit in a wide variety of scenarios (for example, they're about perfect for small/medium cities that can't afford to invest a whole lot in intra-city rail, and they're the only viable option for smaller cities/towns that aren't already stops on a larger inter-city rail line), but they're not necessarily the right fit for a large city with more ground to cover (even with dedicated bus lanes, buses can only - safely - drive so fast, and that speed is typically less than rail). The phrase "bus rapid transit" is kinda nonsensical/oxymoronic; there's not really much that's "rapid" about buses. Ridesharing is slightly more promising, since the smaller vehicles can drive a bit faster and don't have to stick to specific routes. Unfortunately, because they're smaller, they don't carry as many people, so they're less suited for heavy load (i.e. lots of passengers trying to go from the same place to the same place). Rail is probably the most efficient, but also the least flexible and the most expensive (the expense can be mitigated somewhat if you're proactive about building tram rails into your roads, but not all roads warrant that, and by the time they do it's likely already too late; plus, that doesn't do much for older cities). The ideal would be something like the following: - High-speed rail between cities - Metro rail between major parts of cities (and/or between cities and suburbs/satellites) - Full buses on fixed (or at least firm) routes to transit centers in neighborhoods/suburbs/satellites that don't have metro connections of their own (connecting to nearby metro stations, the high-speed rail hub, and/or other bus stations) - Minibuses on more flexible routes for intra-(neighborhood/suburb/satellite) transit (between neighborhood minibus stops, to nearby macrobus/metro/rail stations, or to specific minibus stops outside that particular "zone" in rare cases where a point-to-point really is faster and enough people are trying to do it) All this would be controlled by a rider walking up to some sort of kiosk (or using a smartphone app), picking a destination, and getting a route calculated on-the-fly (for kiosk users: probably printed, or if we wanna be real slick, dispensed on an e-ink display that'll update if the route changes).

jackcosgrove 8y ago
Bus rapid transit (BRT) refers to reserving lanes along existing roads strictly for buses, and doing so along stretches with few controlled intersections, or even changing the timing on stoplights to ease the flow of the BRT vehicles. BRT is much faster than traditional bus mass transit.