Animation Explains How Bad Planning Makes Car Ownership Compulsory


This is a pretty great animation explaining how American cities were undermined by a slavish dedication to storing and moving cars. It’s by comedian Adam Conover from TruTV’s “Adam Ruins Everything,” who also made this great video explaining the screwed up origin of the word “jaywalking.”

The best part may be the animated version of parking guru Donald Shoup, author of “The High Cost of Free Parking.”

Television comedians preaching urbanism? We hope this is part of a trend.

  • Joe R.

    I’m going by the number of people who use mass transit to enter Manhattan during the day, which is on the order of 8 million. If those trips were spread out evenly over 24 hours then I agree something like a few hundred thousand vehicles would be sufficient. The problem is they’re concentrated mostly in the space of a few hours. Hence you need a lot more vehicles. Mass transit incidentally suffers from exactly the same problem. You have a surplus of vehicles and track capacity for much of the day simply to cater to rush hour travelers.

    Note that even if it’s only 300,000, that’s still a huge capital investment.

  • catfink

    I’m going by the number of people who use mass transit to enter Manhattan during the day, which is on the order of 8 million.If those trips were spread out evenly over 24 hours then I agree something like a few hundred thousand vehicles would be sufficient.The problem is they’re concentrated mostly in the space of a few hours.

    Transit rides in Singapore are also mostly concentrated at rush hour. The researchers take this into account in their model. That’s why average wait times will be longer at peak travel times.

    You’re not doing any kind of serious, mathematically rigorous analysis. You’re just making uninformed assumptions and jumping to conclusions. Your comments are full of this kind of nonsense. You seem to think your back-of-an-envelope thought experiments are a serious substitute for academic research. They’re not.

    With 300,000 vehicles you’re still going to have wait times close to 15 minutes during peak periods. This is longer than typical mass transit wait times during the same period

    To illustrate what I said above about your uninformed assumptions: your comment here ignores the time required to get to and from train stations and bus stops, the delay on the transit vehicle as it stops at all the intermediate stops before the one where you need to get off, additional wait time if you need to make a transfer, and delays arising from indirect routing.

  • Dennis_Hindman

    ” A decrease in population-weighted density makes a city less conducive to walking, cycling and mass transit even if average density increases.”

    You’ve stated that at least twice in your comments.

    Your statement would also mean that metropolitan areas which have much less population weighted density should be less conducive to walking, cycling and mass transit.

    Here are the top five highest population weighted density metropolitan areas for 2010, according to the Census Bureau:

    12,113 Los Angeles
    12,144 San Francisco
    31,251 New York
    8,613 Chicago
    11,548 Honolulu

    The 2014 Census Bureau American Community Survey results for journey to work percentages for walking, driving and bicycling combined:

    9.3% Los Angeles
    23.6% San Francisco
    37.7% New York
    15.8% Chicago
    15.9% Honolulu

    The populated weighted density of those metropolitan areas above have little to do with how conducive they are to walking, cycling or mass transit.

    Just to reinforce that here are some other population weighted density results for other metropolitan areas:

    7,980 Boston
    7,773 Philadelphia
    2,990 Pittsburgh

    The 2014 American Community Survey results for commuting by walking, cycling or transit for these metropolitan areas are:

    19.2% Boston
    14.0% Philadelphia
    9.4% Pittsburgh

    Even though the Pittsburgh metropolitan area population weighted density is about a third of what it is for the Los Angeles Metropolitan area, the combine percentage of workers who primarily walked, cycled or used massed transit to commute is about the same.

    Boston, Philadelphia and Chicago metropolitan areas have about three quarters the population weighted density of the Los Angeles metropolitan area and yet their total average percentage of walking, cycling and mass transit commuting is at least 50% higher than it is for the Los Angeles metropolitan area.

    You simply do not understand how population density can make it conducive to walk, cycle and use mass transit.

  • catfink

    Your comment here is very confused. First, you refer to a combined market share for “walking, driving and bicycling” then purport to draw a conclusion regarding “walking, cycling or mass transit.” It’s apples to oranges.

    But you have inadvertently illustrated the stupidity of your obsession with average density. The average density of the New York and Los Angeles metro areas is very similar. But New York has a vastly higher share of transit commutes. This illustrates the absurdity of your assumption that a small increase in the average density of the city of Los Angeles tells us anything meaningful about how conducive it is to walking, cycling or mass transit.

  • Dennis_Hindman

    Alright, to clear up your supposed confusion here are ACS journey to work percentages broken down into walking, cycling and transit for each metropolitan area:

    Walk___Transit___Bike

    2.5%____5.8%___1.0%___Los Angeles

    4.7%___16.7%___2.2%___San Francisco

    6.0%___31.1%___0.6%___New York

    3.2%___11.9%___0.7%___Chicago

    5.4%____9.3%___1.2%___Honolulu

    5.3%___12.9%___1.0%___Boston

    3.7%____9.7%___0.6%___Philadelphia

    5.6%____3.4%___0.4%___Pittsburgh

    “But you have inadvertently illustrated the stupidity of your obsession with average density. The average density of the New York and Los Angeles metro areas is very similar.”

    Per the 2010 Census, persons per square mile:

    27,012 New York City
    _8,092 City of Los Angeles

    The population density per square mile in New York City is over three times more than the city of Los Angeles. Not even close to being similar.

  • Joe R.

    The paper doesn’t mention the total number of trips into Singapore so it’s difficult to extrapolate their data to NYC. I do however have some data on how many people enter Manhattan by subway and commuter rail during rush hour: https://wagner.nyu.edu/files/rudincenter/dynamic_pop_manhattan.pdf

    During the peak AM hour the report says 389,000 people enter Manhattan by subway. If all those people were in self-driven taxis spaced nearly bumper to bumper (i.e. 25 feet apart) moving at 25 mph you would need about 75 lanes to accommodate them. In practice you’ll need more. Bumper to bumper flow isn’t sustainable except on highways where you don’t need to stop. On local streets figure at least half the time vehicles will be stopped waiting for people to cross streets. That implies you’ll need twice as many lanes, or about 150 lanes, to get all those taxis into Manhattan. Last I checked you had 30 something lanes going into Manhattan from the other boroughs.

    Bottom line—it’s mathematically impossible to replace the subway with self-driven taxis unless you either make them 20% the size of existing taxis or put at least 5 passengers in each taxi. Even then, you still need to account for the fact some traffic lanes into Manhattan will be used for delivery vehicles or other types of vehicles. It might be a nice mental exercise to envision such an idea, but when you look at it empirically the idea fails big time. It also fails socially and politically. The people in NYC don’t want a bunch of extra vehicles flooding their already congested streets. If anything, we want to reduce car traffic so there’s more space for pedestrians, buses, and bikes. And there’s little evidence this will be a superior way to get around in terms of either speed or cost. It’s an unknown, which is why the cost of the vehicles, regardless of the actual number you’ll need, matters. No bank will loan a business some tens of billions of dollars on a largely unproven business model. Add to that the fact is all it will take are a few regulations to entirely derail such a business. It might be that a terrorist puts bombs in some of the cars, or maybe they kill a few children. You’ll have legislators banning their use in NYC probably within a week. NYC is great at banning things people think are even remotely dangerous.

    I think self-driven car advocates should instead focus on what these vehicles can do—namely greatly reduce the total number of vehicles and the travel times in suburban and rural areas. Here there are obviously no constraints on space since nearly everyone in these areas is already getting around by car. In the final analysis, the major benefit of self-driving cars might be to radically reduce road costs in the suburbs by allowing one lane to move as many vehicles as three lanes of human-driven cars. They will radically reduce parking requirements, and also eliminate the need to even own a car.

  • ahwr

    You count trips. catfink counts miles.

  • ahwr

    When a kid runs out in front of a self-driving car going 30 or 40 mph that car might not be able to stop or turn to avoid them. On the other hand, if the car was only going 10 mph it could probably avoid injuring that kid every single time.

    Do you think that’s what the public outside of dense walkable areas will support if it means a loss of automobility?

    Why do you think we won’t end up with something more modeled on the way trains operate. NYC only has a handful of at grade rail crossings, but they’re common elsewhere. Nobody expects a train to stop for them if they want to cross the street. Nobody expects trains to slow to a crawl so they can stop if a kid darts across. The train has a defined operating procedure, and everyone is expected to work around it to keep themselves safe. People not on the train are often poorly accommodated, and so take risks.

    http://www.charlotteobserver.com/news/local/dr-traffic-blog/article41965557.html

    But longer delays at crossings lead to greater frustration. Some pedestrians and bikers will wait only so long and then they take a risk, Helms said.

    “Anytime a train is blocking the area for more than 20 or 30 minutes, people start to climb,” he said. “They climbed over the train.”

    Helms has been commuting through the area for 16 years for his job at Armature Winding Co. He sometimes sees children standing between the train cars, passing bikes from one side to the other. He even has seen a baby passed over a train.

    “A boy climbed up on the train – 20 something,” he said, describing the male. “His girlfriend or wife handed him the baby while she climbed down. It’s very common.”

    Sometimes people die doing this. Is the railroad responsible for that death?

    Will all streets be full of cars limited to 5-10 mph for safety? Will the loss of automobility that entails be supported? Or will there be different defined operating procedures for different roads. So on a minor side street the car would be limited to 5 mph at all times. But most travel would be on arterials like union turnpike, where pedestrians would be required to yield to traffic outside of crosswalks and when they don’t have a walk sign. Even if line of sight at an intersection is low an automated vehicle would cruise down at 30 mph. If someone darts out, the car will try to stop, but if it’s unable to a legal framework will be in place to remove liability from the car. More people would die, and it would be harder to cross the street on foot than you want it to be. But if a self driving car tries to be absolutely safe and that means it goes down every street at 5 mph, will anyone buy one? Might the choice come down to having people drive cars, sometimes irresponsibly. Or automated cars that sometimes operate in a dangerous manner, but are perfectly consistent affording people the chance to keep themselves safe, which would you support?

    If the current users of 164 or other big streets near you were given the choice, would they choose to pedestrianize it, where cars are permitted but limited to 5 mph and operating as ‘guests’? Or would that only be supported on minor streets? Would it even then? Would there be support for some streets to act as high speed arterial corridors, even if prohibits other uses of the public space and leads to some deaths and injuries? If a street is designated a travel corridor, what is the state’s response to someone darting out instead of waiting 30 seconds for a light? Say they did it in a manner than allowed cars to stop, but only barely. Nobody died. Nobody was hurt. But other people had to slow down so they could go faster. Would there be enforcement operations to prevent this behavior? Does the state support the use of the street as a travel corridor, or abandon it if there’s even the slightest opposition? If self driving cars are ubiquitous the calculus on how public space is apportioned changes somewhat. The technology doesn’t create a pedestrian utopia on its own.

  • ahwr

    Also worth a mention here is one HUGE advantage of rail is it’s a lot more immune to weather.

    It’s more an advantage of subways than rail over pavement.

    http://www.oregonlive.com/commuting/index.ssf/2015/06/max_trains_slowed_by_summer_he.html

    Even if they could move fast enough to manage 2 or 3 round trips during rush hour from the outer boroughs to Manhattan that’s still a fleet size on the order of 2 or 3 million

    Automated cars don’t change geometry, but your numbers are probably a bit off. 8-9am inflows into the Manhattan CBD are about 622k. 60k are by cars/taxis/trucks. Do you think a fleet of 2-3 million might be overkill?

    Automated cars might increase road capacity, but not by a factor of ten. Heavy rail will still be used to move people around during the day. But midnight to 1am when the majority of people entering or leaving the CBD do so in a car and there is surplus road capacity, will the subways still be run? Maintenance could get a lot more convenient if you shut down the trains.

  • ahwr

    I’m going by the number of people who use mass transit to enter Manhattan during the day, which is on the order of 8 million.

    8 million is total travel two ways. Total each way is ~3.8 million, 25% by auto/taxi/van/truck.

  • Joe R.

    I can envision wide public support for self-driven buses operating the way trains do, perhaps even going highway speeds on surface streets. That would still allow pedestrians a large percentage of time when they can cross intersections unimpeded. When a bus is coming, the lights would go red in advance so people wouldn’t start crossing the street without being able to finish before the bus arrives. Since don’t walk signals would be relatively rare outside of peak hours, it’s likely they would be obeyed a bit more than they are now.

    I can’t envision huge fleets of self-driven taxis operating the same way. It would mean half the time people couldn’t cross the street. And it would also cut the average speed of the taxis by half while they waited for people to cross, negating a lot of the speed/capacity advantage of self-driving cars in the first place. The concept of self-driving vehicles can be a boon if we get the actual volume of vehicles way down from where it is now. We can do that by utilizing vehicles more efficiently.

    Thinking things through a little further, the MTA could run somewhat more buses given that getting rid of the driver removes the major cost. That combined with the higher running speeds could put bus travel on par with where private car travel is today, at least during off-peak hours. Furthermore, you might be able to have the buses divert a block or two off fixed routes to pick up or drop off people without incurring huge time penalties. Again, this makes bus travel nearly as attractive as a fleet of self-driven taxis.

    Self-driven taxis might exist also, perhaps primarily as a mode of travel for the disabled. Since their numbers wouldn’t need to be very high, they could probably operate under similar rules as buses, basically at high speeds with signal priority on arterials, perhaps 10 or 15 mph off arterials.

    You’ll always have highways where private self-driven cars and other vehicles can operate at high speeds. That should give reasonable trip times overall even if restrictive rules are in place on a lot of surface streets.

    I know self-driven cars and buses won’t create a pedestrian utopia on their own but done right they can make things much better than they are now. Streets where cyclists or pedestrians rarely need to wait for red lights, and where they know vehicles will never be operated recklessly or incompetently, is a huge step up from what we have now.

  • Joe R.

    If all those people are each in their own self-driven taxi as catfink envisions then saying you’ll need upwards of 1 million might not be out of line. If peak inflows into Manhattan are 622,000 and each vehicle takes an hour to make a round trip then you need at least 622,000. If it takes longer then you might need over 1 million. Quite a few Manhattan commuters are going long distances which will easily take upwards of 30 minutes, plus another 30 to go out of Manhattan to pick up another fare. Then you also have to account for enough extra vehicles to give reasonable waiting times. In the Singapore example you could technically service all trips with only 100,000 vehicles but the wait times were often unacceptably long. To get them down to acceptable levels you needed three times as many vehicles. That might put the total around my 2 to 3 million estimate. I certainly wouldn’t discount that until someone actually does a study, which I’m sure will happen sooner or later.

    Automated cars might increase road capacity, but not by a factor of ten. Heavy rail will still be used to move people around during the day. But midnight to 1am when the majority of people entering or leaving the CBD do so in a car and there is surplus road capacity, will the subways still be run? Maintenance could get a lot more convenient if you shut down the trains.

    This is actually a great use of self-driving vehicles. Wait times during overnight hours are often long and the roads are mostly empty anyway, so it makes perfect sense to just shut down the subways for maintenance from maybe midnight to 5 or 6 AM and have on demand self-driving provide service. The MTA could run a fleet of self-driving taxis and small vans for exactly that purpose, charging the regular fare.

  • Joe R.

    I feel self-driving cars will be a radically disruptive technology but at the same time they won’t be the answer to all our transportation problems. In fact, we made the mistake in the 1950s of thinking cars and planes were the answer to everything, to the point we abandoned a lot of rail systems which may well have remained viable. We shouldn’t make that mistake again. Self-driving cars promise to be most disruptive for the scenarios where human driven vehicles make sense now, and public transit doesn’t. That’s mostly in rural and exurban areas. Self-driving taxis and buses will doubtless serve well in urban and suburban areas. Perhaps in the suburbs self-driving taxis can largely replace all other forms of transportation other than commuter rail into major cities. In urban areas though you’ll need some combination of self-driven taxis, self-driven buses, and heavy rail. Unlike the suburban or rural areas, cities are designed first and foremost for people. Even if there were no space constraints preventing everyone from getting around by self-driven taxis, we wouldn’t want them to because the streets would be constantly teeming with cars. If anything, we can and should use self-driving vehicles to dramatically reduce the volume of motor vehicles on city streets in order to make cities more livable. We can only do that by keeping rail systems intact, then using the self-driving vehicles mainly to feed passengers to them.

  • catfink

    But to clear up your confusion

    It’s not my confusion. It’s your confusion. As I told you, you’re mixing up two different sets of modes, “walking, driving and bicycling” and “walking, cycling or mass transit.” I don’t know what you think this comparison is supposed to mean.

    Per the 2010 Census, persons per square mile:

    You’re confused again. The statement of mine you’re responding to here addresses metro area densities. The numbers you cite in response are central city densities. These are not the same thing. I don’t know what you think the city densities you cite have to do with my statement about metro area densities.

  • catfink

    During the peak AM hour the report says 389,000 people enter Manhattan by subway. If all those people were in self-driven taxis spaced nearly bumper to bumper (i.e. 25 feet apart) moving at 25 mph you would need about 75 lanes to accommodate them

    No, the report does not say that. It says that 389,000 people enter the Manhattan central business district, not Manhattan itself. Since the rest of your “analysis” is based on this false assumption, it’s worthless.

    This is yet another example of the errors and confusions that plague your comments.

  • Joe R.

    The exact wording in the report is as follows:

    According to New York Metropolitan Transportation Council (NYMTC) estimates, Manhattan-bound subways carried 389,000 passengers during the peak hour of the morning commute (8:00 AM) on an average fall work day into the Manhattan central business district.

    Note the two words “Manhattan-bound”, which implies these subway trips originate outside the borough of Manhattan. This report doesn’t mention figures for the number of people entering the Manhattan CBD by subway who originate in other parts of Manhattan. The only mention is that “58% of Manhattan residents travel to work regularly by mass transit” but there is no actual number given.

  • Dennis_Hindman

    I’ve corrected my inadvertent use of the word driving instead of transit in my first comment, that wasn’t confusion on my part, it was a mistake.

    A decrease in population-weighted density in a metropolitan area has little bearing on how conducive one city within that metropolitan area is to walking, cycling and mass transit. Nor does the population-weighted density of a metropolitan area necessarily indicate how conducive it is for walking, cycling or mass transit compared to another metropolitan area as my listing of population-weighted densities of metropolitan areas and their corresponding share of bicycling, walking and transit indicate. Average densities over a wide area are useless for determining public transit outcomes. What matters to public transit is the density right around its stations and stops, and this can be quite unrelated to the average density of a whole urban area. The attractiveness of bicycling and walking is dependent on how close destinations are in areas within a metropolitan region. Population weighted densities do not give an indication of that.

  • catfink

    The exact wording in the report is as follows

    Yes, it says “carried 389,000 passengers … into the Manhattan central business district.” Not “into Manhattan island.” Did you miss that?

    You would still need at least 150 lanes of traffic going into the Manhattan CBD to accommodate them in self-driving taxis

    No you wouldn’t. Once again, you are just making up numbers out of thin air.

    In any case, you’re quibbling over irrelevancies. Even if self-driving cars will not be able to completely substitute for the subway on all routes at all times, they will at the very least massively reduce the total demand for the subway. Total subway ridership will fall dramatically. That means the huge fixed costs of the subway will be spread across many fewer passenger-miles of travel. The cost per passenger-mile will skyrocket. Subway users will resist paying enormously higher fares. Taxpayers will resist paying enormously higher subsidies for each subway passenger. The economics of the subway will become insane. There will be an enormous economic incentive to reduce the maximum demand for travel to the level that can be satisfied by self-driving cars.

  • Joe R.

    Apparently you neither understand English, math, nor simple spatial concepts. Nobody other than you seriously suggested the idea of self-driving cars to replace mass transit in NYC. You know why? Because we don’t have enough traffic lanes for it, even putting aside the fact that it’s an idiotic idea even if we did.

  • catfink

    So you’re back to Argument by Personal Incredulity. It wasn’t persuasive the first time you said it, and it’s not persuasive now. In any case, you apparently completely missed the last paragraph of my last comment. Either self-driving cars will completely eliminate mass transit in NYC (and everywhere else), or they will almost completely eliminate it. Either way, it’s very bad news for mass transit.

  • Joe R.

    They won’t “almost eliminate it” at all. The math works against it. At best I can see self-driving vehicles replacing the subway during the midnight to 5 or 6 AM period when ridership is very light. And I think that might not be a horrible thing because we’ll be able to do maintenance much more rapidly if we can completely shut down the system.

    About 600,000 people enter the Manhattan CBD during the peak hour. You don’t have the physical room for them to all be in cars, or even buses. Besides all that, part of what makes NYC unique is its mass transit system. We can’t and wouldn’t want to be just like everyone else, getting around in isolated pods. If I wanted to go everywhere by car, I would live in the suburbs. The problem is you don’t have a clue why cities exist or how they work. To you they’re just centers of congestion and nothing else.

  • catfink

    If that’s the best you can see, you’re blind. The subway serves large areas of four New York boroughs. There is already a lot of unused road capacity in these areas at most times of the day and night, and self-driving cars will expand the capacity enormously. Self-driving cars will be used to the maximum extent possible to meet the demand for travel. If there are any times and places where demand for travel exhausts all available road capacity in self-driving cars, the subway will get the excess. This will be just a small fraction of the current demand for the subway. Hence, the cost of the subway per passenger-mile of travel will skyrocket, leading to the economic death spiral I described in previous comments.

  • Joe R.

    I’m not blind. I can easily see self-driving buses and vans supplementing the subway in areas which currently aren’t served by it. They would perform much the same function feeder buses do now—namely feed passengers into the subway system. They’ll make travel to the subway faster and more convenient than our current system of slow buses on fixed routes. Long term as these outer borough areas densify there will be enough ridership to justify building subways to them, if there isn’t already. You’ll still want to get the vast majority of road users on larger, multi-passenger vehicles in order to decrease road congestion. If you have too many vehicles, people will have difficulty crossing streets and the vehicles themselves will suffer delay. A self-driving bus which goes 50 or 60 mph between stops because the road is mostly empty can offer as good or better travel times than hordes of self-driving taxis which end up needing to stop often just to let people cross streets. The heart of the problem here is self-driving car proponents never seem to account for the fact people walk and bike in cities. When large numbers do, as in NYC, then much of the efficiency of self-driving vehicles is lost, unless of course you only have relatively few of them.

    Another thing you’re ignoring is the same benefits automation gives to self-driving cars will also accrue to buses, trains, really every facet of life. Robotics promises to be even more revolutionary than self-driving cars, although they’re really two sides to the same coin. The primary cost of public transit systems is the labor to build and run them. It will become less expensive to build infrastructure like subways, and certainly less expensive to run them, than at present. Subways then might not need public subsidies at all to charge a reasonable fare.

  • catfink

    I can easily see self-driving buses and vans supplementing the subway in areas which currently aren’t served by it.

    No one’s going to bother with buses and vans of any type when they can get a cheap taxi ride instead.

    Long term as these outer borough areas densify

    They’re not going to densify. The long-term trend is already toward lower densities, and self-driving cars will accelerate that trend by making inexpensive car travel universally available.

    A self-driving bus which goes 50 or 60 mph between stops because the road is mostly empty can offer as good or better travel times than hordes of self-driving taxis which end up needing to stop often just to let people cross streets.

    The road won’t be mostly empty. It’ll have as many self-driving cars as needed to meet the demand for travel. Buses are much slower than taxis. Making the bus self-driving wouldn’t change that.

  • Joe R.

    The reason the buses would be able to go highway speeds on surface streets is because they are relatively few of them, so you can run them under rules similar to railroading, where intersections would be treated like railroad crossings and the bus will always have priority. Pedestrian signals would go red shortly before a bus comes, the bus passes, and they go green. Probably a pedestrian could cross a street 90+ % of the time without needing to wait. And the bus only needs to stop to pick up passengers. It might turn out even including stops the bus can average over 30 mph.

    Now if you have hordes of self-driving cars, you’re back to what we have now, which is timed signals. Half the time pedestrians will be forced to wait for a green light to cross, the other half of the time self-driving cars will be waiting for people to cross. That cuts the average travel speed to no more than half the cruising speed, assuming the vehicles pick up their passenger and go straight to their destination. And the cruising speeds will need to be closer to present urban speed limits because there will be no guarantees pedestrians won’t be crossing midblock or against the light. So you have 30 mph cruising speeds, stopping half the time. The end result is 15 mph average speeds. And on top of that, every inch of the street is crowded with self-driving cars. It may seem strange to you, but most city people don’t like streets congested with cars.

  • catfink

    A decrease in population-weighted density in a metropolitan area has little bearing on how conducive one city within that metropolitan area is to walking, cycling and mass transit.

    Yes it does. A decrease in the population-weighted density of an area (whether it’s a metro area, a city or any other kind of subdivision) generally means that the area has become less conducive to walking, cycling and mass transit, and more conducive to driving, because it means that typical distances between people and destinations in the area have become longer.

    Nor does the population-weighted density of a metropolitan area necessarily indicate how conducive it is for walking, cycling or mass transit compared to another metropolitan area as my listing of population-weighted densities of metropolitan areas and their corresponding share of bicycling, walking and transit indicate.

    Your comparisons aren’t very meaningful because you’re comparing population-weighted density against transportation mode share for one type of trip only — commutes. Commute mode shares are strongly influenced by job distribution. A city or metro area in which jobs are highly centralized in a central business district may have a relatively large transit commute mode share even if it has a relatively low population-weighted density. This is why Los Angeles has a lower transit commute mode share than Boston, for example, even though LA has a higher population-weighted density. Boston has a much higher centralization of jobs than LA, even though LA has a higher population-weighted density, and a higher job density.

    Average densities over a wide area are useless for determining public transit outcomes.

    Yes, that’s what I’ve been telling you. But you seem obsessed with changes in the average density of the city of Los Angeles, as if that tells us anything useful about whether the city has become more transit/walking/biking-oriented or less. It doesn’t.

  • Dennis_Hindman

    “A decrease in the population-weighted density of an area (whether it’s a metro area, a city or any other kind of subdivision) generally means that the area has become less conducive to walking, cycling and mass transit, and more conducive to driving, because it means that typical distances between people and destinations in the area have become longer. It doesn’t necessarily mean that every subset of the area has become less conducive to transit/cycling/walking, but it does mean that the area overall has become less conducive”

    People are not going to be walking, bicycling or for that matter using mass transit to make trips over a distance of 4,850 square miles in the metropolitan area of Los Angeles or the 13,318 square miles of the New York metropolitan area. The distances to most destinations for those living in the city of Los Angeles or New York City have not gotten longer, the geographical boundaries have not changed. Due to increased population density which makes products and services within those cities in closer proximity, average trip distances to destinations have become shorter. Moving further away from New York City or the city of Los Angeles and then commuting to those cities would make trips less conducive to walking, biking or taking mass transit but that does not change the fact that within those two cities walking, biking and mass transit are increasing becoming more attractive overall . Biking and walking are short distances forms of transportation. Mass transportation is most effective when there is higher population density and its generally not as effective over very long distances.

    There is lots of data to get results that would come to different preconceived conclusions. You choose to use only average subsidies per passenger mile for transit, or rail compared to cars because that gives the select data that supports your bias. Transit and passenger rail are a lower percentage of all trips for users of that service compared to car owners. Therefore subsidies are for more miles and more trips for car users than transit users. All of the operating costs of transit and passenger rail are included for those self contained systems and yet for driving it only includes the subsidies for just movement. The data for subsidy per mile for driving does not include the costs of constructing and maintaining parking (which the costs of transit does include, along with maintenance yards), law enforcement (transit pays for most of its own security), all costs of damage to property or personal injury (included in the operating cost of transit) and the fact that car owners only pay for part of the cost of building and maintaining the roads (the total costs of building and maintaining exclusive transit right-of-ways of transit is included in its operating costs).

    Your use of population-weighted density for metropolitan areas as a measurement of how conducive an area is for walking, biking or transit is the same tactic of picking data that supports your bias.

    The city of Los Angeles has added 234,000 more residents from the year 2000 to 2014. New York City increased by 482,000 people in that amount of time and by 596,217 people since 1970. These two cities are not being hollowed out. Their populations are increasing, not decreasing.

  • catfink

    People are not going to be walking, bicycling or for that matter using mass transit to make trips over a distance of 4,850 square miles in the metropolitan area of Los Angeles or the 13,318 square miles of the New York metropolitan area.

    No, people use those modes to make much shorter trips within those metro areas. As the metro area population-weighted density declines, there are likely to be fewer such trips compared to trips by car, because average distances between people and destinations will increase.

    The distances to most destinations for those living in the city of Los Angeles or New York City have not gotten longer.

    And your evidence for this claim is…?

    Due to increased population density which makes products and services within those cities in closer proximity, average trip distances to destinations have become shorter.

    An increase in the average density of an area tells us nothing about average trip distances within that area. You’re still confused about the difference between average density and population-weighted density.

    You choose to use only average subsidies per passenger mile for transit, or rail compared to cars

    What supposedly superior alternative subsidy measure do you propose, and why do you think your proposed alternative is superior?

    Transit and passenger rail are a lower percentage of all trips for users of that service compared to car owners. Therefore subsidies are for more miles and more trips for car users than transit users.

    It’s hard to make sense of these two sentences. If you’re trying to say that people who travel primarily by car tend to travel more miles than people who travel primarily by transit, that’s probably true. But so what? Each passenger-mile of travel by transit is subsidized at vastly higher rate than each passenger-mile of travel by car. How do you justify this enormously higher rate of subsidy for transit? Are you saying that you think that each person should receive an equal dollar amount in transportation subsidies, regardless of how much they travel by each mode and regardless of their number of trips or passenger-miles? Or what? If you think you have an argument to justify transit’s massive subsidies, then make it.

    All of the operating costs of transit and passenger rail are included for those self contained systems and yet for driving it only includes the subsidies for just movement. The data for subsidy per mile for driving does not include the costs of constructing and maintaining parking

    We’ve been over this already.The cost of street parking is included in the cost of building and maintaining streets. The cost of private off-street parking is paid for by its users, either through a dedicated fee or through the voluntary pricing policy of bundling parking costs into the prices of other goods and services purchased by those users.

    law enforcement (transit pays for most of its own security),

    Drivers also pay for most of their own security.

    all costs of damage to property or personal injury (included in the operating cost of transit)

    No, the operating costs of transit do not include “all costs of damage to property or personal injury.”

    and the fact that car owners only pay for part of the cost of building and maintaining the roads

    Roads provide essential public services for everyone, so it doesn’t make sense to expect car owners to pay the entire cost of roads. Transit users pay only a small fraction of the costs of building, maintaining and operating transit.

    Your use of population-weighted density for metropolitan areas as a measurement of how conducive an area is for walking, biking or transit is the same tactic of picking data that supports your bias.

    No it doesn’t. The only bias here is your enormous bias in favor of transit. A typical bus or train ride for which the transit user pays $1-2 costs more like $4-8 to provide. Why should transit users get this enormous public subsidy? I keep asking and you keep ignoring the question.

    The city of Los Angeles has added 234,000 more residents from the year 2000 to 2014. New York City increased by 482,000 people in that amount of time and by 596,217 people since 1970. These two cities are not being hollowed out. Their populations are increasing, not decreasing.

    We’ve been over this too. An increase in population is not the same thing as an increase in population-weighted density. If the population-weighted density of those cities has decreased, then those cities have been “hollowed out,” because the typical distance between people and destinations in those cities has increased. Population-weighted density has definitely decreased in the New York and Los Angeles metro areas.

  • gneiss

    It’s nonsense using percentage passenger-miles as a measurement of what mode share dominates in European cities. You are comparing apples and oranges. Most trips in a European city are going to be less than 2 miles, and people there in general don’t have to travel as far as they do in the United States, so longer trips are going to show up more dramatically in the data then they do in the US. The Anti-Planner admits that people here are required to travel far more for every trip (to the tune of 9,100 miles auto travel per capita) then they do in Europe which completely distorts the data.

    In any evident, the numbers quoted for passenger-miles in the Anti-Planners blog are for all trips in Europe, not just those in cities, so you cannot use that as a reference to your claim that “cars dominate travel in all major European cities” Give us numbers for passenger-miles only in cities, and then we might have a reasonable discussion.

  • catfink

    It’s nonsense using percentage passenger-miles as a measurement of what mode share dominates in European cities. You are comparing apples and oranges. Most trips in a European city are going to be less than 2 miles, and people there in general don’t have to travel as far as they do in the United States, so longer trips are going to show up more dramatically in the data then they do in the US.

    I don’t know what “longer trips are going to show up more dramatically in the data” is supposed to mean, or how you think it’s relevant to the issue of mode shares. What is “more dramatic” supposed to mean in clear empirical terms? Why is it relevant?

    The Anti-Planner admits that people here are required to travel far more for every trip (to the tune of 9,100 miles auto travel per capita) then they do in Europe which completely distorts the data.

    No one here has “admitted” any such thing. The fact that people travel more miles obviously doesn’t mean they are “required” to do so.

    In any evident, the numbers quoted for passenger-miles in the Anti-Planners blog

    I don’t know what “the Anti-Planners blog” is supposed to mean. I haven’t mentioned any blog. The passenger-miles data I referred to comes from Eurostat, the EU statistics agency.

  • Emily68

    I don’t understand why self-driving cars are going to be so low in cost. A taxi fare without paying the driver would still be more than a bus ride.

    And as an aside, I lived in Seattle for 10 years without a car. I walked, bused, and once in a while rented a car. The savings over car ownership was significant.

ALSO ON STREETSBLOG

El Paso Congressman Holds Town Hall Meetings on the Bus

|
Congressman Beto O’Rourke takes constituent outreach seriously. Not satisfied that he was reaching a broad enough segment of his El Paso constituents with traditional town-hall meetings, the freshman Democrat has taken up an unorthodox approach: holding discussions on local buses. Americans for Transit’s Rider’s Blog caught up with O’Rourke recently and asked him to explain: […]

El Paso Congressman Holds Town Hall Meetings on the Bus

|
Congressman Beto O’Rourke takes constituent outreach seriously. Not satisfied that he was reaching a broad enough segment of his El Paso constituents with traditional town-hall meetings, the freshman Democrat has taken up an unorthodox approach: holding discussions on local buses. Americans for Transit’s Rider’s Blog caught up with O’Rourke recently and asked him to explain: […]

2-Minute Video: Why Parking Minimums Are the Worst

|
“Minimum parking requirements act like a fertility drug for cars,” Donald Shoup wrote in his celebrated investigation of parking economics, The High Cost of Free Parking. The above video, from the city of Ottawa, does a good job explaining exactly why that is and the problems it causes. Ottawa commissioned the animation to explain why its 1960s-era parking […]

There Will Never Be "Enough" Parking

|
Employees at the Mayo Clinic in Rochester, Minnesota, have to accumulate 13 years of service time before they get an on-site parking permit. To get a sense of how much employees become invested in this system, check out this YouTube video of one man’s elation the day he gets his parking privileges (and notice how towering parking garages dominate […]

There Will Never Be “Enough” Parking

|
Employees at the Mayo Clinic in Rochester, Minnesota, have to accumulate 13 years of service time before they get an on-site parking permit. To get a sense of how much employees become invested in this system, check out this YouTube video of one man’s elation the day he gets his parking privileges (and notice how towering parking garages dominate […]

As Tuition Prices Rise, Fresno State Spends $4 Million on Parking

|
Perhaps you’ve seen the headlines recently about how higher education costs have been ballooning, leaving a generation shackled with crushing debt? In its recent series on college costs and student debt, the New York Times noted that one factor, according to some analysts, is the host of expenses colleges and universities have assumed that are […]