Autonomous Vehicles

Posted by on . 36 comments.

I see that while I was in China, Tyler Cowen wrote a column on the promise of driverless vehicles for the New York Times. This is a topic I’ve been banging on about for a while. I do appreciate him drawing attention to the obstacles to innovation, however, since I’ve got money riding on the eventual adoption of autonomous vehicles.

Comments

  1. Brett says:
    May 31, 2011 at 10:22 am

    He barely touches on the potential for Driverless Cars.

    Think of the possibilities for deliveries and pick-ups alone, and I’m not talking about delivery trucks. You could send your automated car to pick up your laundry and groceries at a drive-through pick-up window, which itself might be automated.

  2. Alex B. says:
    May 31, 2011 at 11:11 am

    Autonomous vehicles solve some issues with congestion, but they still can’t match the raw density-enabling capacity of mass transportation.

    Some of the problems of the car are just inherent to the mode – cars are diffuse, spatially inefficient, and decentralized – none of that changes with automation.

  3. Mike Fogel says:
    May 31, 2011 at 11:36 am

    +1 to Alex B’s comment above.

    Another thought – autonomous vehicles may enable our health care costs to reach a new highs, by enabling a state of lazyness previously out of reach. Sitting in the car watching cable TV is less engaging than at least driving.

    Not to say they aren’t coming. 15 yrs sounds aggressive though.

  4. low-tech cyclist says:
    June 3, 2011 at 10:40 am

    Some of the problems of the car are just inherent to the mode – cars are diffuse, spatially inefficient, and decentralized – none of that changes with automation.

    No, some of that changes with automation. As Ron S. said in the November thread that Ryan links to, “With a zero-accident rate, there is no need for RoboCars to weigh 2,500 pounds.” There’s also little reason to have multiple-passenger robocars; the natural outcome would be cars with a motorcycle-sized traffic footprint.

    That would still be more spatially inefficient than mass transit, but it would be a lot less spatially inefficient than what we’ve got now.

    The same thing would hold for fuel efficiency: near-universal robocars would burn considerably less gasoline than we do now, but still wouldn’t be as fuel-efficient as mass transit.

    Alex, I’m not sure what problem you’re trying to define with respect to cars being diffuse and decentralized. Could you expand a bit on that?

  5. Alex B. says:
    June 3, 2011 at 2:30 pm

    It’s simple, really. Every mode of transport needs vehicles, rights of way, and terminals.

    For cars, the terminal is the parking space. And if you want to have a system where you can go from anywhere to anywhere with a one seat ride, you need far more terminal capacity (i.e. parking spaces) to accommodate those trips.

    Sure, smaller automated vehicles would be more spatially efficient than the cars we have now, but they’d still very extremely spatially inefficient compared to pedestrians and mass transit.

    Terminal capacity isn’t a problem with pedestrians, since for pedestrians, you are the vehicle.

  6. Mixner says:
    June 3, 2011 at 6:01 pm

    “Autonomous vehicles solve some issues with congestion, but they still can’t match the raw density-enabling capacity of mass transportation”

    Mass transit ridership is limited by demand, not capacity. Buses and trains are rarely full. On average, 7 out of 10 seats are empty. Very few, if any, conventional mass transit services will survive in a world of autonomous cars.

  7. Alex B. says:
    June 6, 2011 at 12:24 pm

    I said density-enabling capacity. Mass transit has a higher capacity than cars do.

    Simply put, you could not build New York on cars alone – you’d have to dedicate far too much space to parking to enable that kind of density. With robocars, you could have them drive themselves out of the city to park during the day, but then you’re not talking about a car-based, decentralized, on-demand system – you’re essentially talking about a type of transit.

    In short, I’m not talking about transport modes for mode’s sake, I’m talking about what those transportation modes enable the city to do.

  8. EngineerScotty says:
    June 6, 2011 at 3:17 pm

    Driverless cars–specifically, driverless taxis–might make “personal rapid transit” schemes a possibility. Right now, taxis are economically inefficient due to the need to pay a driver, and thus suffer from the somewhat paradoxical problems that a) they are expensive to use, and b) the hack behind the wheel, more often than not, is earning a poverty wage. But driverless taxis would share one advantage that transit has over personal automobiles, in that the vehicles need not be stored at the workplace (or at home overnight); instead they can serve other trips–or if parking them is needed, they can be stowed at an out-of-the-way terminal maintenance facility.

    And that brings up an interesting facet about Google’s choice of state for its research: While Nevada has many advantages for Google–its close to California, but lacks the Golden State’s notorious gridlocked politics, it also is home to one of the fiercer taxi cartels in the nation. I’m certain that the Vegas taxi industry will fight tooth and nail to keep driverless taxis from ever becoming legal in Nevada; and it wouldn’t surprise me if this extends to all driverless vehicles, even those not used for providing public transport services.

  9. Mixner says:
    June 6, 2011 at 5:12 pm

    “I said density-enabling capacity.”

    And I replied that ridership is limited by demand, not capacity. Higher capacity isn’t going to produce higher density if there isn’t the demand to use it.

    “Simply put, you could not build New York on cars alone”

    We’re not building any more New Yorks. New York’s density is a legacy of its age. Almost all new urban development is low density and car-oriented. Autonomous cars will make it even more car-oriented. No one is going to bother with buses and trains when they can get a (autonomous) taxi ride for the same price. The only transit services that are likely to survive in a world of autonomous cars are the very few services for which demand is so high that even autonomous cars couldn’t meet it. But even in New York, even in Manhattan, autonomous vehicles are likely to have a devastating impact on demand for mass transit.

  10. Alex B. says:
    June 7, 2011 at 10:42 am

    @Engineer Scotty

    I agree, driverless cars make PRT a potential reality (at least in concept). I also agree that for it to work in an urban environment, it would be more like a taxi than a personal vehicle (i.e. a shared transportation resource).

    @Mixner

    You’re simply wrong about the character of urban development.

  11. EngineerScotty says:
    June 7, 2011 at 1:09 pm

    Of course, I suspect that any autonomous vehicle plying the streets of Manhattan, if it isn’t being ridden in by its owner, will need a medallion. :)

  12. low-tech cyclist says:
    June 9, 2011 at 12:47 pm

    Mixner: Mass transit ridership is limited by demand, not capacity. Buses and trains are rarely full.

    Similarly, roads are rarely full. So we’ve got excess capacity coming out the wazoo. So why would there be any need for driverless cars, since it’s so easy for everybody to get from Point A to Point B on these largely empty roads?

    That would be because you’re failing to make a distinction between peak loads and average loads. At peak times, in the cities I’m familar with at least, capacity IS the limiting factor.

    At the times when people want to get downtown, you really can’t get many more people into DC via Metro (for instance) without increasing the number of cars and trains, and ultimately the number of subway lines.

    Sure, you can always get more people on the trains and buses at 10pm. And I could do calisthenics in the middle of Fifth Avenue in midtown Manhattan at sunrise on a Sunday morning.

    Almost all new urban development is low density and car-oriented. Autonomous cars will make it even more car-oriented.

    I doubt it. Autonomous cars would allow new urban development to be more high-density and walkable because you could get rid of parking requirements. If your autonomous car takes you downtown, you don’t need to park it where you’re going; you can send it to a parking lot on the edge of the built-up area, and send for it when you are about to need it again.

    (This also resolves the problem Alex brings up about terminal capacity. The terminal capacity doesn’t need to be where all the people are.)

    And more high-density and more walkable = more advantageous for mass transit.

    And there will still be plenty of demand for more mass transit in a world of autonomous cars, because unless car sizes shrink even more than I’d expect in such a world, you’ll still have plenty of congestion on the roads at peak hours, and it will still make sense to take a subway downtown if you put a premium on getting there fast.

  13. Alex B. says:
    June 9, 2011 at 5:49 pm

    Well, it doesn’t really solve the terminal capacity issue.

    Again, if this is going to be a car as we know the car to be (i.e. it is my personal space, I control the music, how cold the a/c is, etc) and it’s something that is owned by an individual, if my car has to be automatically driven far outside of the city to park, that means I have to summon it back when I want to leave – and that takes time. Ergo, the on-demand “freedom” that the car offers is gone.

    Now, if I just hop in any old car, then we’re talking about a system more like a taxi than a personal car.

    That’s the key difference – owning the vehicle vs. sharing that space with others.

  14. Mixner says:
    June 10, 2011 at 1:29 am

    low-tech cyclist,

    You seem to have missed the point about capacity, density and demand. Mass transit can provide greater capacity than cars, and can therefore support higher densities, but there simply isn’t enough demand for high densities to consume more than a small fraction of the capacity provided by transit. As I said, on average, around 7 out of 10 seats on transit vehicles are empty.

    Autonomous cars would allow new urban development to be more high-density and walkable because you could get rid of parking requirements.

    We could already get rid of parking requirements, and allow high-density, with existing transportation technologies — mass transit and bicycles, instead of cars. But few people want to live that way. That’s why mass transit and bicycles have such a tiny share of the urban transportation market, and why most post-war urban development has been low density and car-oriented. Self-driving cars will significantly increase urban mobility. They will allow virtually everyone to enjoy cheap urban car travel, and will greatly reduce or even eliminate road congestion. The need for urban buses and trains will go away almost completely. Greater mobility generally means lower densities. The further people can travel in a given period of time, the less incentive they have to build things close together. Self-driving cars will be the final nail in the coffin for New Urbanism.

  15. Mixner says:
    June 10, 2011 at 1:39 am

    And there will still be plenty of demand for more mass transit in a world of autonomous cars,

    You’ve got to be kidding. Autonomous cars could double road capacity simply by reducing the spacing between vehicles. And increase it even more by eliminating stops at intersections. And increase it even more by more efficient routing. Those effects alone would virtually eliminate all current road congestion. And the vehicles themselves will likely shrink significantly also. The typical driverless urban taxi is likely to be about the size of a Smart Car. You could fit two or more of them into the amount of road space required by today’s typical taxi — a Ford Crown Victoria or similar vehicle. Manhattan could support at least twice as many taxis with no increase in congestion just from this size reduction.

  16. low-tech cyclist says:
    June 10, 2011 at 10:23 am

    Alex,
    I think it does solve the problem, either way.

    If we keep individual cars, it’s rarely far from the core business district to some industrial zone where enormous car lots could easily be situated. If someone’s commuting to 16th and K from Reston, sending their car a few more miles to a lot off New York Ave. NE or across the Douglass Bridge isn’t going to be a big deal.

    And while the car won’t be there instantaneously at the end of the day, most people know when they’re starting to wrap up their work for the day, and can summon the car 10 or 15 minutes before they head out the door, and that’ll probably be as simple as clicking an icon on one’s iPad or Android.

    On the flip side, if the world of autonomous cars is a world of shared cars (probably more efficient!), some of that customization is probably going to be easy to do. Give each car a USB port where you plug in your thumb drive with your A/C settings, seat adjustment and lighting (for reading, natch!) settings, and of course your playlist, and voila! instant customization.

  17. low-tech cyclist says:
    June 10, 2011 at 1:24 pm

    Mixner: You seem to have missed the point about capacity, density and demand. Mass transit can provide greater capacity than cars, and can therefore support higher densities, but there simply isn’t enough demand for high densities to consume more than a small fraction of the capacity provided by transit. As I said, on average, around 7 out of 10 seats on transit vehicles are empty.

    To which my answer is: and so what? It’s like saying there isn’t enough demand in my household to consume more than a small fraction of the capacity provided by our cars, because they’re either sitting in our driveway or in a parking garage near the office for 20+ hours a day.

    Similarly, almost every room in my house goes unused most of the time.

    These statements are true, but meaningless.

    Quite frankly, I’m not sure what your point is.

    We could already get rid of parking requirements, and allow high-density, with existing transportation technologies — mass transit and bicycles, instead of cars. But few people want to live that way. That’s why mass transit and bicycles have such a tiny share of the urban transportation market, and why most post-war urban development has been low density and car-oriented.

    Um, this is factually wrong. The demand for urban living is demonstrably higher than the supply, that’s why urban real estate prices are so damned high. Supply and demand, and all that. Lots of people who would love to live in walkable urban neighborhoods are simply priced out of the market.

    A huge reason “why most post-war urban development has been low density and car-oriented” is that governments at all levels have been subsidizing that sort of development and throwing up regulatory roadblocks to dense development.

    Sure, lots of people want to live in the ‘burbs, myself included. But a lot of people are out here in no man’s land because there simply wasn’t an affordable alternative in town, and that is at least partly due to height limitations and parking requirements that reduce the supply of both office and residential space in the urban core.

  18. Mixner says:
    June 10, 2011 at 3:13 pm

    To which my answer is: and so what?

    I already told you, twice. The density-enabling capacity of transit is irrelevant, because there isn’t enough demand to use more than a small fraction of that capacity.

    Um, this is factually wrong. The demand for urban living is demonstrably higher than the supply, that’s why urban real estate prices are so damned high.

    Assuming “urban real estate” is supposed to mean real estate in dense/transit-oriented/walkable forms of urban development, this is just factually wrong. Real estate of that kind is expensive to buy because it’s expensive to supply. Land prices are higher, because more people are competing for each square foot, and construction costs are higher, because it costs more to build vertically than horizontally.

    A huge reason “why most post-war urban development has been low density and car-oriented” is that governments at all levels have been subsidizing that sort of development and throwing up regulatory roadblocks to dense development.

    If you seriously think you can make a serious case that low density, car-oriented development is subsidized at a higher rate than compact/walkable/transit-oriented development, then do so. Mass transit is subsidized something like 70 cents per passenger-mile in direct subsidies alone. This dwarfs the subsidies provided to car users.

    The latest Census data shows a continued dispersion of the population to lower-density, car-oriented forms of urban development. Suburbs have continued to grow faster than central cities, and the outlying, most car-oriented suburbs have grown fastest of all.

  19. low-tech cyclist says:
    June 10, 2011 at 4:16 pm

    I already told you, twice. The density-enabling capacity of transit is irrelevant, because there isn’t enough demand to use more than a small fraction of that capacity.

    You say that like it’s supposed to mean something.

    Real estate of that kind is expensive to buy because it’s expensive to supply.

    And if few people wanted to use it, it wouldn’t be expensive to supply, now would it?

    Land prices are higher, because more people are competing for each square foot

    Exactly.

    and construction costs are higher, because it costs more to build vertically than horizontally.

    And there wouldn’t be a market for building vertically unless a lot of people wanted to live and work there in the first place.

    Nonetheless, parking requirements reduce the amount of this very sought-after land that can be built on, and height limitations prevent building vertically even where there’s buckets of money to be made from it.

    If you seriously think you can make a serious case that low density, car-oriented development is subsidized at a higher rate than compact/walkable/transit-oriented development, then do so.

    Every year, more is spent on highway construction in this country than the entire DC-area Metro subway system cost to build. And suffice it to say we aren’t building one Metro system per year, or anything remotely close.

  20. Mixner says:
    June 10, 2011 at 5:02 pm

    You say that like it’s supposed to mean something.

    I don’t know why you think it’s meaningless to point out that the higher capacity of transit is irrelevant because there isn’t enough demand to use more than a small fraction of that capacity.

    And if few people wanted to use it, it wouldn’t be expensive to supply, now would it?

    Few people do want to use it. That’s why it’s so rare compared to low-density, car-oriented development. And part of the reason few people want to use it is because it’s so expensive, for the reasons I just explained.

    And there wouldn’t be a market for building vertically unless a lot of people wanted to live and work there in the first place.

    Whatever you mean by “a lot of people” here, it is far fewer than the number of people who want low-density, car-oriented urban development. That’s why almost all new development is of the car-oriented variety. The point is not that there’s no market at all for dense, walkable urbanism, but that the market for that kind of lifestyle is much, much smaller than the market for car-oriented suburbanism.

    Every year, more is spent on highway construction in this country than the entire DC-area Metro subway system cost to build.

    But the highway system provides vastly more travel for vastly more people than the DC-area metro system, so I’m not sure why you think this is a meaningful comparison. Per passenger-mile, mass transit users, including users of the DC metro, receive vastly higher subsidies than car users.

  21. low-tech cyclist says:
    June 14, 2011 at 2:22 pm

    I don’t know why you think it’s meaningless to point out that the higher capacity of transit is irrelevant because there isn’t enough demand to use more than a small fraction of that capacity.

    Because the difference between peak and off-peak matters. It’s not like you can build a subway system at one-third the cost that only exists between 6-10am and 3-7pm.

    You build it for those times, because that’s when you need it most. And if a reduced number of people are riding it the rest of the time, that’s a bonus, not a failure.

    If you don’t have enough capacity at peak hours, and if in fact the peak hours are spreading out because of the limits on capacity, then you don’t have enough capacity, rather than having too much.

    If you can’t get your mind around that, then Baravelli, you’ve got the brain of a four year old child, and I bet he was glad to get rid of it.

  22. Mixner says:
    June 14, 2011 at 5:58 pm

    You still don’t get it. It’s not a matter of peak vs. off-peak demand. It’s a problem of total demand. There simply isn’t enough total demand to use more than a small fraction of transit’s capacity. So the fact that transit has higher capacity than cars, and can therefore support higher densities, is irrelevant. A car-based transportation system provides more than enough capacity to meet demand in most places at most times, and driverless cars will greatly increase the capacity of car-based transportation.

  23. EngineerScotty says:
    June 14, 2011 at 9:22 pm

    At risk of feeding trolls…

    You still don’t get it. It’s not a matter of peak vs. off-peak demand. It’s a problem of total demand. There simply isn’t enough total demand to use more than a small fraction of transit’s capacity.

    You’re argument is confused. On one hand, you’re saying that demand for transit is lower than its capacity, therefore transit is bad. On the other hand, you’re saying that capacity of the road network is higher than its demand, therefore cars are good; and driverless cars (which can increase capacity) are better.

    You’re attempting to imply two completely opposite conclusions with the same premise.

    What I suspect you’re trying to say–but looking ridiculous when you juxtapose things this way, is that transit is inefficient because it has unused capacity, but the unused capacity of roads is a good thing, as it permits a higher quality of service. That argument can be reversed; of course–the ample reserve capacity of transit demonstrates its flexibility and scalability; whereas the existence of empty streets at certain times of the day prove that the roads are inefficient boondoggles.

    And that would be just as stupid of an argument as yours.

  24. Mixner says:
    June 15, 2011 at 12:31 pm

    You’re argument is confused. On one hand, you’re saying that demand for transit is lower than its capacity, therefore transit is bad.

    I haven’t said that at all. I said, for the umpteenth time, that the “raw density-enabling capacity” of transit is irrelevant because there’s only enough demand to use a small fraction of it.

  25. EngineerScotty says:
    June 15, 2011 at 1:07 pm

    I haven’t said that at all. I said, for the umpteenth time, that the “raw density-enabling capacity” of transit is irrelevant because there’s only enough demand to use a small fraction of it.

    And in many cases, that would be wrong. The busses and trains around here are routinely crushloaded at rush hour–but that’s nothing compared to the impact on the streets were all these folks to drive instead.

    If what you are arguing is that there’s no demand for density–and thus no need for infrastructure geared to it–that’s rather simply hogwash. Much varies depending on where you are, and there are plenty of places where the vast majority of the population prefers suburbia, but there’s plenty of other places where high-density real estate is in high demand.

    And has been noted countless times; in many places high-density development is restricted by zoning codes–a key point made in this blog. If and when urbanists use the force of law to try and mandate their land-use preferences, you would scream bloody murder; don’t pretend that legal regimes essentially mandating suburbanism are simply “reflecting the wishes of the market/people”; if something is truly the wish of the market it need not be codified in law.

  26. Mixner says:
    June 15, 2011 at 6:16 pm

    And in many cases, that would be wrong. The busses and trains around here are routinely crushloaded at rush hour–but that’s nothing compared to the impact on the streets were all these folks to drive instead.

    No, it’s not wrong. The vast majority of the time, on the vast majority of routes, transit vehicles do not run remotely close to capacity. They run mostly empty. There isn’t enough demand for transit to support a transit-based transportation system. That’s why our transportation system is based on cars, and transit has only a tiny niche role in the system. Driverless cars will virtually eliminate even that tiny niche role. There might still be few rush-hour trains in Manhattan, where even driverless cars might not be able to fully meet rush-hour travel demand, but virtually all other conventional mass transit will disappear. No one’s going to bother with buses and trains when they can get a door-to-door taxi ride for less than they pay today for a bus ticket.

    And has been noted countless times; in many places high-density development is restricted by zoning codes–a key point made in this blog.

    As has been noted countless times in response, zoning codes are created and sustained by the political process. People support laws that limit density because density creates negative externalities, such as congestion, crowding, noise, pollution, loss of privacy, loss of mobility and in general a degradation in the quality of urbanized life. Unless you can find a way to internalize these costs, people aren’t likely to give up their zoning laws.

  27. EngineerScotty says:
    June 15, 2011 at 6:41 pm

    No, it’s not wrong. The vast majority of the time, on the vast majority of routes, transit vehicles do not run remotely close to capacity. They run mostly empty. There isn’t enough demand for transit to support a transit-based transportation system.

    And as has been pointed out to you, much is the same with roads; at most hours of the day, they have ample capacity. And as noted, driverless autos will increase the capacity (so would driverless transit vehicles, BTW–and we have those today). You’re still suggesting that a capacity/demand balance is a virtue for one mode and a demerit for the other.

    That’s why our transportation system is based on cars, and transit has only a tiny niche role in the system. Driverless cars will virtually eliminate even that tiny niche role. There might still be few rush-hour trains in Manhattan, where even driverless cars might not be able to fully meet rush-hour travel demand, but virtually all other conventional mass transit will disappear. No one’s going to bother with buses and trains when they can get a door-to-door taxi ride for less than they pay today for a bus ticket.

    The word is “if”, not “when”. Autonomous driverless vehicle technology (as opposed to current driverless metros, which are not autonomous) will reduce the cost of operating an auto by the cost of time lost due to driving; but autonomous vehicles will still need fuel and maintenance and the other expenses associated with car ownership. Transit, particularly well-used transit, does far better on these things–and driverless transit vehicles eliminate the key expense of transit systems–the need to pay for a driver. Current driverless systems like SkyTrain are expensive to build; but if the technology Google and others are developing can make ordinary cars driveable in ordinary traffic; it can do the same for busses and surface rail.

    Certainly, at some point in the future, it is conceivable that a fleet of autonomous taxis might become part of a region’s transit system (whether public or private, I don’t much care). Many consider taxis today to be part of the transit system, even if not government-operated. And such might eliminate or reduce the need for the “social service” bus–the runs that you complain about are always empty. But in dense areas, there will be a need for larger-capacity vehicles; and depending on what improvements in vehicle power technology occur, larger-capacity vehicles may have a significant advantage in operational cost.

    As has been noted countless times in response, zoning codes are created and sustained by the political process. People support laws that limit density because density creates negative externalities, such as congestion, crowding, noise, pollution, loss of privacy, loss of mobility and in general a degradation in the quality of urbanized life. Unless you can find a way to internalize these costs, people aren’t likely to give up their zoning laws.

    Density doesn’t create congestion or pollution; cars do. The cars that cause most of the problems are generally those travelling into the city from somewhere else. Density has, generally, a positive impact on mobility because it shortens most trips. Some people don’t like crowds, I’ll condede; but others prefer the urban lifestyle to the suburban or rural.

    And a major reason for the current state of zoning laws, even it isn’t ever stated openly by officials, is exclusion of the poor, by making housing too expensive. (And often times, zoning codes are supported by powerful landowners who seek to prevent competing properties from being built).

  28. Mixner says:
    June 15, 2011 at 7:28 pm

    And as has been pointed out to you, much is the same with roads; at most hours of the day, they have ample capacity.

    Exactly. Which is why the “raw density-enabling capacity of transit” is irrelevant. We don’t need that extra capacity because there isn’t the demand for it. We only need extra capacity on a few travel routes during a few hours of the day, and driverless vehicles will virtually eliminate even that small need by increasing the effective capacity of roads. In a world of driverless cars, the need for transit will go from “small” to “virtually non-existent.”

    Autonomous driverless vehicle technology (as opposed to current driverless metros, which are not autonomous) will reduce the cost of operating an auto by the cost of time lost due to driving; but autonomous vehicles will still need fuel and maintenance and the other expenses associated with car ownership.

    So what? Cars already overwhelmingly dominate the urban transportation market, and driverless cars will make car travel even easier and available to even more people. Without the need to pay a driver (or provide a seat for him), the cost of taxi rides will plummet. In addition, driverless urban taxis will most likely be electric — with much lower running costs than today’s vehicles. I have no idea how you think buses and trains will be able to compete.

    But in dense areas, there will be a need for larger-capacity vehicles;

    Very unlikely, except perhaps on a few routes, for a few hours of the day, where demand is so high that even driverless cars might not be able to fully meet it. But even in Manhattan, driverless cars will decimate demand for mass transit.

  29. Mixner says:
    June 15, 2011 at 7:47 pm

    Density doesn’t create congestion or pollution.

    Yes it does. Pollution is measured as the concentration of contaminants in the environment. Density increases pollution. The more people, cars, trucks, buses, houses, machinery, etc. you pack into a given area or volume, the higher the level of pollution. That’s why Manhattan has some of the worst air pollution in the country. “The solution to pollution is dilution.” That means lower density. The same is true for congestion. The more vehicles per lane-mile of roadway, the greater the chance or intensity of congestion.

    The cars that cause most of the problems are generally those travelling into the city from somewhere else.

    The worst congestion tends to occur where lots of commuters need to converge on a small geographic area to get to their jobs. Hence the terrible rush hour congestion in lower Manhattan. The solution is to disperse jobs over a larger area, which is what we’re doing. The traditional hub-and-spoke model of suburb-to-inner-city commuting is increasingly giving way to suburb-to-suburb or intra-suburb commutes.

    Density has, generally, a positive impact on mobility because it shortens most trips.

    No, density has an overwhelmingly negative impact on mobility. Congestion and travel times tend to increase, not decrease, with density. Car travel is more difficult, so more people have to rely on mass transit. The latest NHTS survey found that the average commute time by mass transit is more than double the average commute time by car. Car-oriented urban environments increase mobility.

    And a major reason for the current state of zoning laws, even it isn’t ever stated openly by officials, is exclusion of the poor,

    I seriously doubt that.

  30. Alex B. says:
    June 16, 2011 at 1:56 pm

    Exactly. Which is why the “raw density-enabling capacity of transit” is irrelevant. We don’t need that extra capacity because there isn’t the demand for it. We only need extra capacity on a few travel routes during a few hours of the day, and driverless vehicles will virtually eliminate even that small need by increasing the effective capacity of roads. In a world of driverless cars, the need for transit will go from “small” to “virtually non-existent.”

    Transit has higher capacity than roads. We have demand that will make use of that capacity at peak hours. I’m not sure what your argument is – and the things I am sure that you’re arguing are empirically false.

  31. Mixner says:
    June 17, 2011 at 3:11 pm

    We have demand that will make use of that capacity at peak hours. I’m not sure what your argument is

    Only on particular routes in certain places. Most transit vehicles aren’t full even during peak hours. There may be a legitimate need for transit on certain routes at peak hours where current road capacity is insufficient, but driverless vehicles, by greatly increasing the capacity of roads, will virtually eliminate even that limited need.

  32. EngineerScotty says:
    June 19, 2011 at 1:35 am

    >>Density doesn’t create congestion or pollution.

    >Yes it does. Pollution is measured as the concentration of contaminants in the environment. Density increases pollution. The more people, cars, trucks, buses, houses, machinery, etc. you pack into a given area or volume, the higher the level of pollution.

    There’s localized pollution, and there’s global pollution. The former is affected by density; the latter, though, is only affected by total pollutants released into the atmosphere. And while density decreases the denominator in the local pollution index (by virtue of a lower volume of air), it also decreases the numerator as well. Denser land uses reduce pollution per capita, as people in urban areas need to drive less, and also consume less energy and produce less pollution for their heating and cooling needs. Much downtown pollution is caused by suburban commuters, not by local residents.

    > That’s why Manhattan has some of the worst air pollution in the country. “The solution to pollution is dilution.” That means lower density. The same is true for congestion. The more vehicles per lane-mile of roadway, the greater the chance or intensity of congestion.

    No, the solution to density is to use alternate means of transportation that don’t involve each commuter hauling around their own internal combustion engine.

    >[future electric driveless taxis will make transit obsolete]

    Were future driverless electric taxis to come into existence–vehicles which a) did not burn fossil fuels, b) could navigate more safely and with fewer negative impacts than human-powered autos, c) did not require valuable urban real estate to be consumed for parking/storage–then most of my objections to the SOV would vanish. (Though transit too would become driverless, and not disappear completely).

    What this has to do with today, though, I’ve no idea–you seem to be suggesting that transit today is a bad idea because of speculative technologies which may make it obsolete in a few decades.

  33. Mixner says:
    June 19, 2011 at 2:06 am

    There’s localized pollution, and there’s global pollution. The former is affected by density; the latter, though, is only affected by total pollutants released into the atmosphere. And while density decreases the denominator in the local pollution index (by virtue of a lower volume of air), it also decreases the numerator as well. Denser land uses reduce pollution per capita, as people in urban areas need to drive less, and also consume less energy and produce less pollution for their heating and cooling needs. Much downtown pollution is caused by suburban commuters, not by local residents.

    The evidence clearly shows that urban air pollution, measured as the concentration of a pollutant in the air, tends to increase with density, for the reasons I explained. So does noise, litter and general dirtiness, all of which are themselves forms of pollution. That’s why Manhattan, with its huge concentration of pollution-emitting people, homes, offices, machines and motor vehicles, is so noisy and dirty, and has some of the highest levels of airborne carcinogens in the country. It’s why low-density suburbs tend to be so much quieter and cleaner than inner cities. As well as less congested, less crowded, and more private. One of the basic reasons people have been moving from cities to suburbs throughout the developed world for the past 50 years is to escape the noise, crowds, congestion, dirt, lack of privacy and other negative externalities of dense urban life.

    What this has to do with today, though, I’ve no idea–you seem to be suggesting that transit today is a bad idea because of speculative technologies which may make it obsolete in a few decades.

    There’s nothing speculative about it. The technology already exists, and will only get better, cheaper and more widespread in the future. The only real question is how fast we’re going to adopt it. Hugely expensive new urban rail transit lines, which don’t even make economic sense given current levels of car travel, are going to be even bigger boondoggles 20 or 30 years from now when car travel is even more attractive by comparison than it is today.

  34. Mixner says:
    June 19, 2011 at 2:06 am

    There’s localized pollution, and there’s global pollution. The former is affected by density; the latter, though, is only affected by total pollutants released into the atmosphere. And while density decreases the denominator in the local pollution index (by virtue of a lower volume of air), it also decreases the numerator as well. Denser land uses reduce pollution per capita, as people in urban areas need to drive less, and also consume less energy and produce less pollution for their heating and cooling needs. Much downtown pollution is caused by suburban commuters, not by local residents.

    The evidence clearly shows that urban air pollution, measured as the concentration of a pollutant in the air, tends to increase with density, for the reasons I explained. So does noise, litter and general dirtiness, all of which are themselves forms of pollution. That’s why Manhattan, with its huge concentration of pollution-emitting people, homes, offices, machines and motor vehicles, is so noisy and dirty, and has some of the highest levels of airborne carcinogens in the country. It’s why low-density suburbs tend to be so much quieter and cleaner than inner cities. As well as less congested, less crowded, and more private. One of the basic reasons people have been moving from cities to suburbs throughout the developed world for the past 50 years is to escape the noise, crowds, congestion, dirt, lack of privacy and other negative externalities of dense urban life.

    What this has to do with today, though, I’ve no idea–you seem to be suggesting that transit today is a bad idea because of speculative technologies which may make it obsolete in a few decades.

    There’s nothing speculative about it. The technology already exists, and will only get better, cheaper and more widespread in the future. The only real question is how fast we’re going to adopt it. Hugely expensive new urban rail transit lines, which don’t even make economic sense given current levels of car travel, are going to be even bigger boondoggles 20 or 30 years from now when car travel is even more attractive by comparison than it is today.

  35. Mixner says:
    June 19, 2011 at 2:21 am

    No, the solution to density is to use alternate means of transportation that don’t involve each commuter hauling around their own internal combustion engine.

    Alternate means of transportation are too slow, uncomfortable, inflexible, and inconvenient to compete effectively with cars. That’s why alternatives have only a small share of the market. If Americans in general were living to live like Manhattanites, in small, cramped, expensive high-rise apartments, then alternate means of transportation would be a viable option, though only for the near-term future. But most people are not willing to live like that. Once again, the fundamental reality here that you simply refuse to accept is that the vast majority of people simply do not want to live the way you want them to live.

  36. EngineerScotty says:
    July 10, 2011 at 12:16 am

    More on the subject here.