Why Your Local Transit System Sucks

Read the chart. And, no, highways don't pay for themselves. Also there are large negative externalities associated with driving that militate in favor of making drivers subsidize transit users (or pedestrians, cyclists, etc.) rather than the reverse.

UPDATE: Of course you can't bring this subject up without legions of people informing you that the gas tax pays for the highways. This simply isn't true. All the funds raised by the gas tax are spent on highways, and then a bunch of additional money is also spent on highways. It's exactly the same as with transit, financed by a mix of user fees and subsidies in order to encourage the economic benefits of infrastructure investment. But one kind of investment is better for the environment and for public health. So naturally we give the other kind radically more money.

Comments

trains suck. transit accounts for a minute 1% of total vehicle miles.

Posted by Blixer | July 21, 2008 4:35 PM

trains suck. transit accounts for a minute 1% of total vehicle miles.

Posted by Blixer | July 21, 2008 4:35 PM

I am on your side here, but using a cumulative graph really makes it look worse than it is for people who are not used to looking at cumulative graphs...

Posted by JWill | July 21, 2008 4:40 PM

The important metric is passenger-miles, not vehicle miles, if you're looking for relative cost-efficiency.

Now, I don't know what the relative passenger-miles amounts are, so I don't know if it actually supports the mass-transit case (though I'd suspect it does), but using vehicle miles is extremely misleading as it ignores the entire "mass" benefit of "mass transit".

Posted by Po-Mo Polymath | July 21, 2008 4:42 PM

But building transit would cost money, you see.

And this is central to my point.
.

Posted by Grand Moff Texan | July 21, 2008 4:43 PM

No linky, MY? Where'd you get the chart? The Magic Transit Fairy?

Posted by Andrew Fly | July 21, 2008 4:43 PM

And yet, the highway infrastrucutre here in Southern California is horrible. The highways are too narrow, they're in terrible condition, and they don't get you where you need to go. If half the people on the freeway started taking transit, the traffic would still be unacceptable. To make this place somewhat livable, you'd need huge invesetments in both transit and highways. I don't really know what the answer is. Tolls? Double-decker freeways?

Posted by too many steves | July 21, 2008 4:43 PM

I agree that we have likely underspent on transit, but I don't think this chart is really a great way to make that point. Cars and such really are quite useful (meaning they are the most efficient mode of transport for many transportation needs, even factoring in the negative externalities). So I am relatively confident that even if we had spent our public transportation dollars in the most efficient way, "highways" would still have come out ahead of "transit" in this graph. So I don't think this chart tells you much, if anything, about how much we should have spent on transit but have not, since if you can't use the highway amount as a baseline, you are left not really knowing where that transit line ought to have been.

Posted by DTM | July 21, 2008 4:45 PM

Whoa. I just noticed that "cumulitive investment" line. That makes no sense. Why would you graph cumulative spending? Even if the nation spent $1 this year on highways, that graph would make it look like an increase. Is that just idiocy, or an actual attempt to mislead?

Posted by too many steves | July 21, 2008 4:47 PM

Jwill,

If we were looking at operational expenditures I might agree, but since these are capital expenditures I think cumulative is indeed the right way to go to create a fair sense of the likely results of such spending behavior. Of course you would really want to depreciate those expenditures over time, but I don't think that would end up making much difference in terms of the visual impact of the graph.

Posted by DTM | July 21, 2008 4:54 PM

Drivers smell good. Transit riders smell bad.

Posted by ixnermay | July 21, 2008 4:55 PM

Highways more than pay for themselves if you count the taxes from jobs that they allow people to travel to.

Posted by Chris | July 21, 2008 4:59 PM

actually, the highway trust fund that is part of your purchase price of a gallon of gas pays for federal highways.

interestingly enough, there was a supreme court case on whether the funds in that trust fund were fungible or whether they had to be spent on their congressionally mandate purpose (i think the fight was, as a matter of fact, spending some highway trust fund money on mass transit, but i can't recall and don't have time to check), and the court ruled they did.

which, as a tangential point, is kind of important with respect to the social security trust fudn, but i digress.

Posted by howard | July 21, 2008 5:04 PM

Matthew writes,

And, no, highways don't pay for themselves.

As we determined in the "Externalities" thread from the BTS data on subsidies, highways very nearly pay for themselves. Transit doesn't even come close. The total public subsidy per thousand passenger-miles is about $5 for highways and about $405 for transit.

Also there are large negative externalities associated with driving that militate in favor of making drivers subsidize transit users (or pedestrians, cyclists, etc.) rather than the reverse.

How large? Show me your evidence that highways have externality costs $400 per thousand passenger-miles larger than transit's. Or even remotely close to that figure.

Posted by Mixner | July 21, 2008 5:08 PM

Um, Chris, where do you think the people that take mass transit systems are going? Communist Party meetings?

Posted by OhioBoy | July 21, 2008 5:09 PM

DTM,

I agree that we have likely underspent on transit,

Ah yes, DTM's faith-based public policy guesswork at work again.

Posted by Mixner | July 21, 2008 5:11 PM

too many steves,

Again, because it is specifically capital expenditures being graphed, it makes sense to treat them cumulatively. That is because the basic definition of a capital expenditure is that it creates future benefits, so capital expenditures in the past will still be creating benefits today (although as I noted above, it would make sense to depreciate those benefits over time when constructing this graph).

Posted by DTM | July 21, 2008 5:14 PM

Highways do more than enable rednecks to drive their SUV's 25 miles to Super Wal-Mart to buy pork rinds.

Eisenhower launched the interstate highway system after a cross-country military convoy took weeks.

Wanna try to evacuate an area about to get hit by a hurricane without highways?

Plus, could we as cheaply have a freight rail network to ship all those gadgets the hipsters want if we didn't have highways for trucks?

Posted by JohnMcG | July 21, 2008 5:16 PM

Honestly, though, what should we expect? Ever since the car and sprawl became the way we do things, a democratic process should favor highways over mass transit. Don't Congressmen vote their constituencies' interests?

Anyone know why Germany didn't go for a sprawl model the way we did? They've got cars, and like them, but they still prefer their housing concentrated

Posted by AlanC9 | July 21, 2008 5:20 PM

Sometimes a chart is just a chart, and that's especially true with a graph of cumulative expenditure (population growth, whatever), which is just an easy way to find out at any given point in time how much you've spent to date.

And this one really gets into numbers we can't comprehend- heck, George Bush has added over $4 trillion to the national debt and we're not seeing a dime of benefit, so that $3 trillion for the highways almost looks wise in retrospect.

Except for the fact that building highways did not cure congestion, has largely destroyed truck farming around our cities, fills our waterways with poisonous runoff, is now sticking suburbanites with fuel bills that would pay for an Ivy League education,, etc etc.

IOW, that chart just shows what we paid for highways, not what we paid and will continue to pay for our suburban lifestyle.

As for the idea that highways pay for themselves, the people of Minneapolis have a nice bridge to sell you if you believe that one...

Posted by serial catowner | July 21, 2008 5:21 PM

JohnMcG,

I live in a hurricane-prone city, and that's why I have a car. However, that's the only thing keeping me attached to that car. If there were high speed rail along the I-10 corridor, or multiple trains linking Chicago and New Orleans, and FEMA mandated evacuation plans that utilized transit for everyone, not just the disabled and infirm, I would sell my car (a Prius btw) in a heartbeat.

Everything in town is bikeable, except evacuating with a dog. Transit makes the 14 hour trek to Houston a little more bearable, since suburbanites aren't taking their 4 SUVs with them.

Posted by alli | July 21, 2008 5:21 PM

Mixner,

Shush. The adults are trying to have a conversation.

Posted by DTM | July 21, 2008 5:24 PM

As a former car owner, I know there is a good alternative: car cooperatives.

When my car was vandalized last year, resulting in nearly $2,000 in damages (over Blue Book), I chose not to replace it. Living in Chicago, just under a mile from the L, I'd used my car only once every 2-3 months. Now, when I need to make a Target/airport run, I reserve an I-Go car. My annual car expenses are about $100/year, all-inclusive.

Posted by JR | July 21, 2008 5:27 PM

By a cursory glance, the ratio of highway expenditure to transit expenditure appears stable. I think the cumulative graph is misleading, although I certainly support the development of superior mass transit.

Posted by Joel | July 21, 2008 5:30 PM

Mixner:
Show me your evidence that highways have externality costs $400 per thousand passenger-miles larger than transit's. Or even remotely close to that figure.

Well, the one most obvious externality that I don't often see mentioned is time. 1000 miles of driving at 60 mph is 16.7 hours of the driver's time. Even for a minimum wage worker, that's already $109.39 worth of time, and without going into details, I will say that my time is recompensed at a rate higher than $400/16.7 hours. If I was sitting on a train, I could be working, but since I'm driving, that time is just wasted. So, well, there's my evidence.

Posted by OhioBoy | July 21, 2008 5:32 PM

ohioboy,

Um, Chris, where do you think the people that take mass transit systems are going? Communist Party meetings?

Um, only a small fraction of American workers commute by public transportation (4.7% in 2000, down from 5.3% in 1990). The vast majority of workers commute by car, truck or van (87.9% in 2000, up from 86.5% in 1990).

Therefore, the positive externalities attributable to each mode of transportation from work-related tax revenues or economic productivity are vastly greater for highways than for transit. In fact, most people who commute to work on public transportation commute by bus. Buses run on highways. Therefore, spending on highways supports positive externalities attributable to bus commutes too.

Posted by Mixner | July 21, 2008 5:38 PM

Oh snap! Nice chart.

Note: The % of vehicle miles taken up by trains or other forms of mass transit is irrelevant. One train from NYC to DC covers the same amount of vehicle miles as one car. The difference is there are 304 full seats on an Acela train and 4 seats in your average car, 3 of them empty.

Posted by joejoejoe | July 21, 2008 5:40 PM

Holy crap. The cumulative capital investment on highways is very close to a constant slope line. Since this is a graph of cumulative investment, isn't a constant slope what you would get if your capital expenditures on highways were a fixed amount that never changed from year to year? Since our population, and the total amount of highway infrastructure, have grown during that time period, no wonder our roads and bridges are falling apart.

The mass transit investment, on the other hand, looks to be an exponential curve, indicating that annual amount of capital investment has been going up.

Posted by Dan Kervick | July 21, 2008 5:41 PM

DTM,

Hush, child. Put down your Big Book of Prayers to the Transit God, and try actually learning something about the subject.

Posted by Mixner | July 21, 2008 5:42 PM

Mixner,

Shhh.

Posted by DTM | July 21, 2008 5:44 PM

The cumulative graph is so misleading that we could double our spending on transit every year and it would still look like a tiny blip compared to highways.

Posted by too many steves | July 21, 2008 5:45 PM

The cumulative graph is so misleading that we could double our spending on transit every year and it would still look like a tiny blip compared to highways.

We could even spend more on transit than on highways, and still, highways would look bigger on this graph.

I hope I didn't post this twice, but I think I did. Sorry.

Posted by too many steves | July 21, 2008 5:48 PM

Repeating things sarcastically makes me the coolest kid in tenth grade. No, you're a poopyhead.

Posted by Muxmer | July 21, 2008 5:50 PM

ohioboy,

Well, the one most obvious externality that I don't often see mentioned is time.

Positive externalities attributable to savings in time also favor subsidies for highways over subsidies for public transportation. The average commute by public transportation takes almost twice as long as the average commute by car.

New York has the longest average commute time in the country because it has the highest share of commutes by public transportation. New York also has the highest rate of "extreme commutes" in the country, defined by the Census Bureau as commutes taking 90 minutes or more.

Posted by Mixner | July 21, 2008 5:51 PM

too many steves: Where the hell are you going to build more freeways in Southern California? There's no room. And what do you mean, "they don't get you where you need to go?" Where *can't* you go on a freeway down there?

Southern California is the living proof that you can't build your way out of automobile traffic congestion. Because if you could, it would have happened there. SoCal has more freeway lane-miles than any other area in North America, maybe the world. Just look at a map.

Yet it is also one of the most horrendously traffic-congested areas in North America.

Are we going to sit here and pretend that's some sort of coincidence?

It didn't used to be that way. There was an extensive network of electric suburban rail service throughout the LA Basin, run by the Southern Pacific as the Pacific Electric Railway, along with a streetcar network called the Los Angeles Railway.

Of course, the oil and automotive companies bought it out and converted it to buses back in the 1950s.

What, you thought "Who Framed Roger Rabbit" was just a movie?

http://en.wikipedia.org/wiki/General_Motors_streetcar_conspiracy

Posted by Travis Mason-Bushman | July 21, 2008 5:53 PM

This stuff is really getting risible. I'm all for more mass transit. It's cleaner, more efficient, etc. However, it would be much more meaningful to look at dollars per passenger, or per passenger mile. The fact is that many, many more people use roads than use rail or bus, and for some completely understandable reasons.

For one, most of this country is just not set up to travel by foot, bus, or train, as Matt bemoans often enough. No matter how radically we change zoning or tax policy, and no matter how hard we punish drivers (sorry, internalize existing externalities), that isn't going to change soon. I would love to see comparative numbers, but I'd bet you'd see a much bigger reduction in carbon output by raising CAFE, subsidizing energy-efficient construction, or investing in sustainable energy. Generally, trying to modify the behavior of tens of millions of people via government investment policy involves some pretty heavy-duty dislocation and pain for the folks at the pointy end of the spear.

Posted by NAL | July 21, 2008 5:55 PM

If the gas tax completely covered highway expenditures, my state (PA) wouldn't be considering leasing the freaking Turnpike in order to fund road and bridge repairs.
http://www.forbes.com/feeds/ap/2008/06/26/ap5159695.html

If the gas tax completely covered highway expenditures, the Highway Trust Fund wouldn't be facing a $4 billion dollar shortfall. http://www.cqpolitics.com/wmspage.cfm?parm1=5&docID=cqmidday-000002913162

Can we get over this myth already?

Posted by Roman | July 21, 2008 5:56 PM

New York has the longest average commute time in the country because it has the highest share of commutes by public transportation.

Mixner, since you are capable of forming coherent sentences, I assume that you're not dumb enough to think that correlation equals causation. So you either have secret knowledge that you're not sharing, or you're being willfully dishonest.

Let me ask you this: are you proposing that we eliminate all rail service into NYC, in order to shorten commute times?

Posted by OhioBoy | July 21, 2008 6:01 PM

Mixner, did you just entirely miss the point of OhioBoy's post? (Not that I would be surprised...)

A person taking public transportation can get a substantive amount of work done while he or she is traveling - writing, teleconferencing, reading, etc. That is time which can be productive.

A person who is driving, cannot get this same substantive amount of work done, because the vast majority of his or her attention is (or should be) focused on making sure his vehicle gets from Point A to Point B. OK, sure, he can talk on the phone half-attentively, but that's hardly the same thing as being able to sit down in a rail car and e-mail memos from a laptop.

Posted by Travis Mason-Bushman | July 21, 2008 6:02 PM

I'm all for more mass transit. It's cleaner, more efficient, etc. However, it would be much more meaningful to look at dollars per passenger, or per passenger mile.

Transit is only slightly cleaner than current passenger cars, at least in terms of CO2 emissions (0.47 pounds of CO2 per passenger-mile for transit, versus 0.54 for passenger cars.). And transit is no more energy-efficient than current passenger cars (both use about 3,445 BTUs per passenger-mile).

The Toyota Prius beats the pants off transit in both CO2 emissions (0.26 pounds/p-m) and energy efficiency (1,659 BTUs/p-m).

Posted by Mixner | July 21, 2008 6:10 PM

O for heavens sakes. The chart shows quite clearly why the country (now) "is just not set up to travel by foot, bus or train"- we spent a heap of bucks making it not set up for travel by other modes.

Indeed, NAL, modifying the behavior of tens of millions of people by government investment policy did involve heavy duty dislocation and pain. When the freeways in Boston were built they just took the land in poor neighborhoods, in other cities, they paid a penny on the dollar. The whites were blockbusted out of the cities by fears that racially biased lending agencies would redline their neighborhoods, making it impossible for homebuyers to get mortgages. Most of youse guys are too young to remember that building the freeways was not a process of everyone cheering and waving flags- it was dirty politics in a time when the disappearance of Jimmy Hoffa barely rated a shrug.

And I guess NAL ain't been keeping up with current events, because it seems pretty obvious that something is going to change soon.

In fact, NAL's comment is a little strange because after stating "I'm all for mass transit" NAL goes on to think we'd see "a bigger reduction" from other efforts. Bigger than what, NAL? There is no proposal in Matt's post, just an observation.

We're not "punishing" drivers when we happen to notice that they're eating us out of house and home, and ask them to pay their full share. Drivers who think we are live in the mental world of a five-year-old, where people get what they want because "we've always done it this way", and when they don't get what they want, someone must be punishing them, probably out of pure envy and malice.

Probably the saddest part of the whole driving experience is that drivers get an hour a day of free time in which to reflect on and observe their world- and most of them turn on the radio and downgrade their intellects from a C to a D-.

Posted by serial catowner | July 21, 2008 6:23 PM

Mixner, since you are capable of forming coherent sentences, I assume that you're not dumb enough to think that correlation equals causation.

Ah, good one. We know from Census Bureau data that the average commute time by public transportation is almost twice as long as the average commute time by car. New York has the highest share of commutes by public transportation in the country. New York also has the longest average commute time in the country. And you're seriously claiming there is no causal relationship between these facts? To what do you attribute New York's exceptionally long average commute time, then?

It's pretty obvious why travelling by public transportation takes so long. It's because of the time costs associated with getting to and from train stations and bus stops, and waiting for buses and trains:

We found that public transportation appears to involve a fixed time cost of approximately 16-20 minutes, regardless of length. After this fixed time costs, cars appear to be about 50 percent faster than buses and roughly as fast as trains. It is this fixed time cost that makes public transportation so costly. The time spent walking (or driving) to the station or bus stop plus the time spent waiting for the bus or train plus the time spent walking or driving to the final destination appears to take up as much time as driving ten miles. As time has gotten more valuable, the time costs of public transportation have become more severe and the population has continued to move entirely towards the automobile.

Edward Glaeser, Sprawl and Urban Growth, Harvard Institute of Economic Research, 2003.

Posted by Mixner | July 21, 2008 6:31 PM

As we determined in the "Externalities" thread from the BTS data on subsidies, highways very nearly pay for themselves. Transit doesn't even come close. The total public subsidy per thousand passenger-miles is about $5 for highways and about $405 for transit.

Mixner doesn't seem to have actually understand the interesting things about the Texas data.

Number One Interesting Thing: Other states mostly don't do this (and Texas didn't either, until recently). Even the most basic cost/revenue analysis is apparently only rarely applied to new road construction, never mind detailed cost/benefit analysis. (Yet new transit projects are invariably gone through with a VERY fine tooth comb. Hmm...)

Number Two Interesting Thing: The BTS data that has triggered Mixner's latest round of goal post shifting is really only a minimum estimate for the roadway cost/revenue shortfall. Figures for the last couple decades of spending no doubt substantially understate the true costs because they do not reflect immense levels of infrastructure depreciation and neglect. And that bill will come due sooner or later.

However, the Texas data, by looking at the present value of expected lifetime costs, DOES include those costs. And they find that costs in Texas exceed revenues by a factor of two, often more.

I see no reason why Texas, with its mix of large cities and long stretches of open space, should be especially unrepresentative of the nation as a whole. So, our best estimate for roadway subsidies is a cost/revenue ratio in the neighborhood of at least 2:1, not the BTS numbers.

Posted by jack lecou | July 21, 2008 6:42 PM

Could it be that the New York metro area has the longest commutes because it is the largest, densest, most congested metro area in the country? I'd be willing to wager that New Yorkers with access to the subway system have substantially lower commute times then the commuters from beyond the city limits who choose to drive in. I'd bet it's not even close.

As far as Glaeser's analysis, that's pretty simple too. The reason that it takes so long to reach public transportation and why you have to wait so long for it, is primarily because it's underfunded. Because it is underfunded, the area it covers isn't as wide and the service isn't as frequent. Fund that bad boy and both of those problems are significantly amiliorated. Cause and effect!

Posted by Roman | July 21, 2008 6:45 PM

Mixner, I'll repeat the question that you chose to ignore: Are you proposing that we eliminate all rail service into NYC, in order to shorten commute times?

Posted by OhioBoy | July 21, 2008 6:47 PM

Mixner - Your BTU per passenger-mile stats are misleading at best.

From the Bureau of Transportation Statistics, Energy Intensity of Passenger Modes
Btu per passenger-mile, 2001 (last year available with data for all modes):

Passenger car - 3,597 BTU per passenger-mile
Amtrak - 2,100 BTU per passenger-mile

Since light rail and subways uses mass-produced electricity and not diesel engines to drive their trains I'd guess their BTU per passenger-mile is even lower.

Note: Current BTS data shows commercial air travel uses slightly less BTUs per mile than driving a passenger car so I apologize for calling all the Aspen Ideapaloozafest attendees wankers for flying in for the conference. You are all still wankers for going [for reasons too numerous to state] but you didn't add to your wanker quotient by flying instead of driving.

Posted by joejoejoe | July 21, 2008 6:47 PM

serial catowner:
As for the idea that highways pay for themselves, the people of Minneapolis have a nice bridge to sell you if you believe that one...

No argument with the broader point (and I'm about to get on my bike and pedal 11.3 miles back home for the evening), but we Minneapolitans will in fact have a nice new bridge to show off very soon ... for example, at some convention or other that I hear is coming to these parts.

Just under a year after the bridge collapse, the new one appears to be almost finished, if my inexpert inspection on a walk across a parallel bridge this past weekend is any indication. There appears to be quite a lot of work left to put together the road surface, but it looks like all of the piers and superstructure are up, with a little over a month left to go. Several months of 24/7 construction shifts clearly will put up a bridge in a hurry.

Of course, doing all of this--and finishing it in time for the RNC--cost a bazillion dollars. Good thing the Pawlenty administration didn't see fit to spend pennies on serious bridge inspections before the old bridge collapsed. That would have been fiscally irresponsible.

Posted by Rieux | July 21, 2008 6:50 PM

DMonteith. Sorry, I mean "jack lecou"

Number Two Interesting Thing: The BTS data that has triggered Mixner's latest round of goal post shifting ...

What "goal post shifting?"

is really only a minimum estimate for the roadway cost/revenue shortfall.

No it isn't. It's the sum of federal, state and local public subsidies by transportation mode.

Figures for the last couple decades of spending no doubt substantially understate the true costs because they do not reflect immense levels of infrastructure depreciation and neglect.

What "spending figures" would those be? Where is your evidence that they "understate the true costs?" Stop with the faith-based wishful thinking, D, and produce some evidence for your claims.

Posted by Mixner | July 21, 2008 6:50 PM

The problem is with pretending there's an "average" commute in the United States, Mixner. Because, for all your whining... people keep making those long commutes via public transit into New York. How on Earth can that be?!?!

Maybe, because it is impossible to make a rational comparison between someone who is, say, commuting from a Des Moines suburb to an office-park in another Des Moines suburb, to someone commuting from a Long Island suburb to Manhattan.

Any commute to Manhattan will inherently be lengthier and costlier, as well as favor mass transit, for a number of reasons.

Being an island, points of access such as bridges and tunnels are limited in number and extremely expensive to maintain. Transit lines are a much more efficient use of these expensive, space-limited access points because they can carry far more passengers in a given space than highways can. Hence, the economics of commuting prefer public transit.

Being an island, and a fairly small one at that, internal street networks are of limited capacity. You could build more bridges into the city, but there's nowhere for those cars to go. There is no feasible way of increasing this capacity, so congestion on city streets is almost perpetual. Hence, the economics of commuting prefer public transit.

Being an island, and a fairly small one at that, land is extremely expensive. Capitalism being what it is, there are much higher-value uses for land than simply paving it and parking automobiles on it. You could build more bridges and bulldoze city blocks to expand the streets, but there's nowhere for those cars to park. Parking in Manhattan is, and always will be, severely limited and astronomically expensive. Hence, the economics of commuting prefer public transit.

Posted by Travis Mason-Bushman | July 21, 2008 6:52 PM

We know from Census Bureau data that the average commute time by public transportation is almost twice as long as the average commute time by car.

This is actually (with caveats) true, so it doesn't make transit advocates look especially well-informed to claim otherwise. Just saying.

I'm leaving work now, but maybe I'll remember to write more about that, and the problems with that Glaeser paper Mixner is so fond of, later tonight.

Posted by jack lecou | July 21, 2008 6:53 PM

Mixner's reading comprehension is so atrociously bad I'm disinclined to engage him directly, but before I leave maybe I'll just repeat the key phrases "infrastructure depreciation" and "deferred mainentence".

Posted by jack lecou | July 21, 2008 6:58 PM

Mixner's reading comprehension is so atrociously bad I'm disinclined to engage him directly, but before I leave maybe I'll just repeat the key phrases "infrastructure depreciation" and "deferred maintenance".

Posted by jack lecou | July 21, 2008 6:59 PM

Transit is only slightly cleaner than current passenger cars, at least in terms of CO2 emissions (0.47 pounds of CO2 per passenger-mile for transit, versus 0.54 for passenger cars.). And transit is no more energy-efficient than current passenger cars (both use about 3,445 BTUs per passenger-mile).

Yuh, maybe because so much of what you're calling "transit" is actually buses, which we should not expect to be cleaner or more energy efficient than passenger cars, after all, it's pretty much the same technology. Trains, on the other hand, are much more energy-efficient. Steel on steel is better than rubber on road.

Posted by James | July 21, 2008 7:04 PM

joejoejoe,

Mixner - Your BTU per passenger-mile stats are misleading at best. From the Bureau of Transportation Statistics, Energy Intensity of Passenger Modes
Btu per passenger-mile, 2001 (last year available with data for all modes):
Passenger car - 3,597 BTU per passenger-mile
Amtrak - 2,100 BTU per passenger-mile

Huh? How is energy efficiency data for "Amtrak" less "misleading" than the data for "Transit?" The claim I was addressing was a claim about autos versus transit, not autos vs. Amtrak. Only a small fraction of transit is provided by Amtrak. And the Amtrak figure you state above includes all Amtrak services (intercity rail as well as commuter rail), which makes it even less relevant to a discussion about transit. Try to follow the issues being discussed.

Since light rail and subways uses mass-produced electricity and not diesel engines to drive their trains I'd guess their BTU per passenger-mile is even lower.

Light rail provides only a small fraction of all transit, and subways serve only small commuting areas in just a few very dense cities. In fact, the vast majority of subway commutes take place in just one city--New York. Most transit consists of buses. And even many rail transit systems are less energy efficient than hybrid cars.

Posted by Mixner | July 21, 2008 7:05 PM

What "spending figures" would those be? Where is your evidence that they "understate the true costs?" Stop with the faith-based wishful thinking, D, and produce some evidence for your claims.

shifting standards of evidence

DMonteith. Sorry, I mean "jack lecou"

totally unsubstantiated claims based solely in his own imagination

Mixner, I'll repeat the question that you chose to ignore

cherry-picking minutae while ignoring the larger, substantive point

Hush, child. Put down your Big Book of Prayers to the Transit God

quickly devolve into "I-know-you-are-but-what-am-I" ad hominem attacks

Full effect, bitches! All I need is the sockpuppetry and I take the thread.

Posted by scythia | July 21, 2008 7:07 PM

Roman, the pennsylvania turnpike is not a federal highway, so how the state chooses to deal with it is neither here nor there to the highway trust fund.

as for your "shortfall" link, what it says is that revenues are declining and therefore, there is a shortfall between needs and revenues; thus far, the general fund hasn't filled that need, and there's no reason for it to....

Posted by howard | July 21, 2008 7:11 PM

"James"

Yuh, maybe because so much of what you're calling "transit" is actually buses,

What do you mean "what you're calling transit?" Transit buses are a form of transit. In fact, buses are the most common form of transit. You can't just pretend buses don't count as transit.

Trains, on the other hand, are much more energy-efficient. Steel on steel is better than rubber on road.

No, trains can be, but are not necessarily, more energy-efficient than cars. It depends on the type of fuel and engine used to power the train, and the average load factor of the service (proportion of seats that are actually occupied by passengers), among other things.

Posted by Mixner | July 21, 2008 7:15 PM

Good catch by James at 7:04pm. Mixner's arguments about the alleged superiority of the automobile in terms of energy efficiency and C02 emissions are essentially arguments in favor of cars vs buses. He doesn't attempt to make such arguments with respect to cars vs. rail transport, because no such arguments exist. Rail transport is vastly superior for the environment, and vastly superior in terms of energy efficiency. And development of rail transport doesn't turn the countryside into the sprawl that Mixner (and only Mixner) finds so appealing.

Posted by Juillet | July 21, 2008 7:22 PM

What do you mean "what you're calling transit?" Transit buses are a form of transit. In fact, buses are the most common form of transit. You can't just pretend buses don't count as transit.

Um, nice straw man, Mix. Absolutely nobody is "pretending" buses "don't count as transit." What most of us are claiming is that buses are an inferior form of transit, and trains are superior.

Posted by Fineburg | July 21, 2008 7:27 PM

"Roman",

Could it be that the New York metro area has the longest commutes because it is the largest, densest, most congested metro area in the country?

Population and density is the root cause, yes. In particular, it is the hyperdense concentration of jobs in midtown and lower Manhattan. This requires the daily transportation of vast numbers of workers from the outer New York boroughs and surrounding counties and states (Long Island, New Jersey, Connecticut, etc.) into and out of a very small geographical area. That means most of those workers cannot effectively drive to work, and must rely on mass transit instead. For the reasons explained by Glaeser, mass transit is slow. Hence New York's exceptionally long commute times.

As far as Glaeser's analysis, that's pretty simple too. The reason that it takes so long to reach public transportation and why you have to wait so long for it, is primarily because it's underfunded. Because it is underfunded, the area it covers isn't as wide and the service isn't as frequent.

Huh? How would covering a wider area reduce the average commute time? And service is already frequent at rush hour. Increasing it would do little or nothing to reduce average commute times, even assuming the fixed infrastructure (highways, bus stops, rail stations, rail track, signalling systems, etc.) could support a significant increase in peak frequency, and the money was available to fund it.

Posted by Mixner | July 21, 2008 7:30 PM

"No, trains can be, but are not necessarily, more energy-efficient than cars. It depends on the type of fuel and engine used to power the train, and the average load factor of the service (proportion of seats that are actually occupied by passengers), among other things."

So, then, maybe if we started making investments in trains that use these technologies, like the shinkasen trains that use regenerative braking, and encouraging people to use the train instead of the car, to improve the load factor efficiencies, we'd have a solution that was actually better than everyone running out and buying a Prius?

PS. Earlier, I did not mean to imply that buses are not transit (maybe the quotes gave this impression). However, it seems that the argument you are making is bolstered by the fact that buses make up such a large proportion of American public transit, and buses are superficially just larger cars, so we should not expect them to have much better efficiency numbers. What I am saying is we would be better off trying to take cars off the road in favor of trains, improving those transit numbers as the relative train share increases and stop wasting so much gasoline on long city-to-city trips.

Posted by James | July 21, 2008 7:30 PM

Gosh, the duplicate handles are really multiplying now. "Fineburg" = "Juillet" = "Roman" = "James". And doubtless many more. But let's play along for now.

"Juillet" writes,

Mixner's arguments about the alleged superiority of the automobile in terms of energy efficiency and C02 emissions are essentially arguments in favor of cars vs buses.

No, they're more arguments against the claim that transit (not just "buses" but transit in general) is more energy-efficient and less polluting than cars. It's mostly a myth.

He doesn't attempt to make such arguments with respect to cars vs. rail transport, because no such arguments exist. Rail transport is vastly superior for the environment, and vastly superior in terms of energy efficiency.

You do of course have evidence to support this assertion.....right? So where is that evidence? Let's see it.

Posted by Mixner | July 21, 2008 7:42 PM

"James",

So, then, maybe if we started making investments in trains that use these technologies, like the shinkasen trains that use regenerative braking, and encouraging people to use the train instead of the car, to improve the load factor efficiencies, we'd have a solution that was actually better than everyone running out and buying a Prius?

Huh? Shinkansen trains for commutes? Do you have any kind of serious cost-benefit analysis to support your proposal? Remember, any energy-efficiency analysis must include the energy costs of construction as well as operating energy. And what actual policies do you propose to "encourage people to use the train instead of the car?"

Obama wants to spend $150 billion to speed up the commercialization of plug-in hybrid cars, develop more efficient batteries for hybrid and electric cars, and update the national electrical grid to support large-scale recharging of electric cars at night. Does that sound to you like it's going to encourage people to stop driving and take the train instead?

Posted by Mixner | July 21, 2008 7:56 PM

Of course transit commuting in New York is going to be relatively slow. Commuting into New York by any means is going to be relatively slow because of the limited throughput of the area's transportation network as a whole. Being on an island tends to make it difficult and expensive to get in and out.

So Mixner, what would happen if we just shut down all the transit systems in New York tomorrow? Is it your argument that suddenly commute times in the New York area would speed up because, gee, all these people aren't riding those slow subway systems, they're on these wonderful newfangled super-fast freeways?

Posted by Travis Mason-Bushman | July 21, 2008 8:14 PM

My actual policy is offering a choice of transit systems instead of reinforcing our all-eggs-in-one-basket system that we have now. Maybe you would like to continue supporting the driving system we have now. I guess that's your prerogative. However it is my experience that quality-of-life WRT transportation is better in New York (trains everywhere) than Detroit (no trains at all). It's true that we've completely bought into the highway system, but that's no reason for us to continue accelerating blindly down that path.

Posted by James | July 21, 2008 8:18 PM

to return to the gas tax question, wrt matthew's update: i personally misread the table. i assumed it was federal highway spending, not government highway spending.

yes, if we are looking at total highway spending, then the gas tax doesn't do it, but on the federal level (as i noted above), it does.

Posted by howard | July 21, 2008 8:57 PM

TMB,

Of course transit commuting in New York is going to be relatively slow. Commuting into New York by any means is going to be relatively slow because of the limited throughput of the area's transportation network as a whole.

As I said, the fundamental problem with commuting in New York is the hyperdense concentration of jobs. If New York were less dense, and more of its workers were able to drive to work instead of having to rely on mass transit, then commute times in New York wouldn't be so long. That's one reason why we're not building any more New Yorks. New jobs tend to be created in suburban and exurban developments where densities are much lower and people are able to reap the benefits of shorter commute times from driving.

So Mixner, what would happen if we just shut down all the transit systems in New York tomorrow?

Chaos, probably. But since I haven't advocated shutting down all the transit systems in New York tomorrow, or anything like that, the question doesn't seem terribly relevant. The long-term solution to New York's exceptionally long commute times is to move jobs outside the city, so that there is no longer this mad scramble of millions of people converging on lower Manhattan every morning, and trying to get out of it in the evening.

In fact, this is the long-term trend throughout the country. Jobs are moving out of inner-city central business districts and into surrounding suburban office parks (and suburban factories, suburban shopping centers, etc.).

Posted by Mixner | July 21, 2008 8:58 PM

"James",

My actual policy is offering a choice of transit systems instead of reinforcing our all-eggs-in-one-basket system that we have now.

Well, I'm still waiting for your concrete proposals. You say you want to "encourage people to use the train instead of the car." Barack Obama's proposal to spend $150 billion subsidizing the development and commercialization of more fuel-efficient cars would tend to have the opposite effect of what you say you want. It'll make driving more attractive, not less.

Posted by Mixner | July 21, 2008 9:07 PM

The businesses don't want to relocate out of Manhattan. They pay an absurd premium for their office space specifically because it's in Manhattan, where the conglomeration of other businesses and workers boosts their business.

The spread of jobs to the suburbs has been going for a half-century of more. We're there, dude. The jobs left in major urban centers are located there because they want to be there. More people bumping into more people means greater productivity, greater creativity, greater synergy, and a more dynamic culture and economy.

"We" aren't building more New Yorks, btw, because "we" have made such things illegal through sprawl zoning, which "we" had to adopt to prevent developers and businessowners from building in a manner that you don't like, market be damned.

Posted by joe from Lowell | July 21, 2008 9:19 PM

As for that job spread, it's notable that it only happened after a massive investment of public dollars into the construction of a highway system, and even then, it took a few decades, until the suburban-to-urban commute subsidized by the construction of that highway system led enough people to move into the outer suburbs that it became plausible to locate businesses there.

More people take cars than transit because a combination of subsidies and regulation has led to most of the housing stock being built so that its residents don't really have a choice.

Posted by joe from Lowell | July 21, 2008 9:23 PM

Shorter commute times from driving?

Bwahahahaha.

Tell that to the people who bought homes in San Francisco Bay Area exurbs like Brentwood, Tracy and Salinas, so they could commute to their jobs in the Silicon Valley, East Bay and San Francisco peninsula.

Hint, Mixner: Businesses *choose* to locate in hyperdense areas because it provides superior efficiency. You can easily and frequently meet, face to face, your colleagues, suppliers and customers when you're located near them. When you need to travel, there is a critical mass of demand which provides for superior road, rail and air transportation networks. Communications, supply chain, prestige and more - all militate toward density.

Mixner, if you were planning to open a DC lobbying firm, would you put your money where your mouth is and open your office in Reston, Virginia, instead of K Street?

Posted by Travis Mason-Bushman | July 21, 2008 9:49 PM

joe from lowell,

The businesses don't want to relocate out of Manhattan.

They will if they can no longer attract the employees they need to run their business, because those employees get fed up with all the time, hassle, crowds, dirt, noise, stress, etc. having to commute daily into Manhattan.

But New York is probably a special case. It's the national center of finance, fashion, publishing, the arts and other important areas of American commerce. It attracts many ambitious and talented people who are willing to put up with long commutes and other unpleasantnesses for the sake of their careers. And it's a huge magnet for new immigrants, who are in demand for all the grunt work that needs to be done to support those higher up on the food chain (taxi drivers, hotel maids, restaurant workers, garbage collectors, etc.) So New York is probably somewhat insulated from the decentralizing influences at work everywhere else in the country. At least for now.

The spread of jobs to the suburbs has been going for a half-century of more. We're there, dude. The jobs left in major urban centers are located there because they want to be there.

It's an on-going process, dude. City centers are continuing to lose population and jobs to the suburbs. But yes, we're already a massively suburbanized nation. That's why we spend so much on highways and so little on transit. If you think this is likely to change much in the future, you're probably going to be disappointed.

More people bumping into more people means greater productivity, greater creativity, greater synergy, and a more dynamic culture and economy.

And you do of course have evidence to support this claim.....right?

Most often, I hear people complaining that their productivity suffers because they keep getting interrupted by coworkers. Too much physical proximity, rather than the reverse.

Posted by Mixner | July 21, 2008 9:49 PM

TMB,

Shorter commute times from driving? Bwahahahaha. Tell that to the people who bought homes in San Francisco Bay Area exurbs like Brentwood, Tracy and Salinas, so they could commute to their jobs in the Silicon Valley, East Bay and San Francisco peninsula.

As I said, the average commute time by car is only about half the average commute time by public transportation. You can choose to ignore this fact, but that won't make it any less true. Of course, cars also provide many other advantages over public transportation in terms of comfort, convenience and flexibility. But the time savings alone from driving would probably be enough to doom transit to niche markets and a small share of total commutes, even without all the other benefits driving has to offer.

Hint, Mixner: Businesses *choose* to locate in hyperdense areas because it provides superior efficiency.

Yes, that must be why jobs have been moving out of the inner cities and into the suburbs for decade after decade. Even Joe lowell seems to recognize that.

Posted by Mixner | July 21, 2008 10:01 PM

More people bumping into more people means greater productivity, greater creativity, greater synergy, and a more dynamic culture and economy.

And you do of course have evidence to support this claim.....right?

Highest density US states / rank of household income
1) NJ - #1 - $66,752
2) RI - #15 - $52,541
3) MA - #7 - $56,592
4) CT - #4 - $60,551
5) MD - #2 - $63,082

Mixner - See any correlation?

Posted by joejoejoe | July 21, 2008 10:20 PM

"Fineburg",

What most of us are claiming is that buses are an inferior form of transit, and trains are superior.

I'm not sure I've seen anyone claim that except you, let alone "most" other people here. Have you really thought it through very carefully?

Let's take Los Angeles as an example. The LA Metro Local bus system operates 189 bus routes. The routes have an average of almost 100 bus stops each, for a total of about 18,500 stops. There's also about 30 routes in the Metro Rapid system, and two Metro Express routes.

And you propose to replace this with a "superior" rail system, do you? Light-rail running along existing streets? Subway? Elevated rapid rail? Have you figured out how much that would cost, even roughly? How many stations you would need to build? (I somehow doubt you'd be able to afford 18,500 of them.) How you're going to raise the money? How incredibly disruptive the construction would be to existing transit services and traffic? How long it take to get the necessary environmental impact approval? Whether it would even save energy and pollution overall, taking into account the total consumption of energy and pollution emissions that it would cause? (Digging subway tunnels, for example, uses vast amounts of energy).

I'd love to see your cost-benefit analysis.

Posted by Mixner | July 21, 2008 10:35 PM

I think it is worth carefully considering this chart:

http://www.bts.gov/publications/national_transportation_statistics/html/table_04_20.html

As I have noted before, both gasoline and diesel prices started increasing in 2000. And from 2000 to 2005 the BTU per passenger mile for cars went down from 3589 to 3458, a reduction of about 3.7%. But transit buses went down from 4147 to 3393, a reduction of about 18.2%. Meanwhile, other four-wheel, two-axle vehicles (your SUVs, vans, pickup trucks, and such, making up around 40% of the passenger vehicle fleet, and increasing--see beloow) went from 4509 to 4452, a reduction of only 1.3%.

So these relative efficiencies have been changing, and I think the data supports the hypothesis that transit authorities have the ability to more quickly respond to high fuel prices with efficiency gains than the private passenger vehicle fleet.

And although Mixner isn't really worth a response, I do think it is worth running the numbers on new vehicles sales to see if the popularity of more efficient vehicles could dramatically speed up the rate at which private passenger vehicles are gaining efficiency.

Roughly speaking, there are about 250 million cars in the U.S. private passenger vehicle fleet. New sales are about 8 million a year. So, new vehicles make up only about 3% per year, which is part of why any efficiency gains for the total fleet from shifts in buying habits is going to be very slow. Indeed, the U.S. population is growing around 1% a year, so a sizable portion of those 3% new cars aren't actually replacing old cars, but just providing cars for new drivers.

In fact, with around 100 million vehicles in the SUV et al category, at 8 million new vehicles a year, it would take over 12 years just to swap out that segment of the fleet. But that isn't happening anyway. The latest relevant data I have seen for vehicle sales is in this document:

http://www.epa.gov/otaq/cert/mpg/fetrends/420r07008.pdf

For MY2007, the SUV et al category still made up a total of 49% of new sales (for comparison, hybrids made up 2.2% of sales in MY2007), so the SUV et al category is continuing to increase its overall share of the US fleet (again, it is about 40% right now). As a result of all this, although the EPA did report some modest efficiency gains in recent model years, the MPG of MY2007 was still below the peak MPG back in MY1987, despite all the technological advances in the last 20 years. Again, as the EPA notes, that is largely effect of SUVs et al steadily gaining share of the total fleet.

In short, the US private passenger fleet is gigantic in relation to new sales, and furthermore new sales are only very gradually trending to more efficient vehicles, and thanks to SUVs et al are still below the efficiency rate of 20 years ago. So, I think it will remain true that the US fleet will only very, very gradually become more efficient through new sales, and we will need to look to somewhere else for rapid efficiency gains in transportation.

Posted by DTM | July 21, 2008 10:49 PM

Mixner - See any correlation?

No, not really. D.C. has a higher density than every single state, but is #26 in income--right in the middle. Alaska is #50 in density, but #6 in income. RI is #2 in density, but #15 in income. Nevada is #43 in density, but #17 in income. Florida is #8 in density, but #33 in income. And so on.

In short, I don't see any clear correlation between density and income at all. And state-level population density is a very poor proxy for "bumping into people at work," anyway. If high population density is so important to productivity and wages, how come incomes are so high in, say, Silicon Valley--a classic sprawling suburb? Or west Los Angeles? Or Scottsdale, Arizona?

Posted by Mixner | July 21, 2008 10:53 PM

Trains and public transportation are caught in a prisoner's dilemma, just like cabs in SF during the early 1990s.

Mixner will never understand this argument.

Posted by mike | July 21, 2008 11:12 PM

Some more data:

Here is a comparison of urban area populations between the 1990 and 2000 Censuses:

http://www.demographia.com/db-ua2000compare.htm

As noted there, "The Census Bureau's method for classifying urbanized land changed in 2000, to correct previous over-estimates of urbanized land. As such the density and land area comparisons in this page should be used with caution."

However, I think the total population numbers are useful. To summarize, the vast majority of urban areas increased in population. Overall, about 21 million more people lived in the 69 urban areas in this chart, which was about 2/3 of the total population increase in the U.S. during that time (about 32.7 million). That also represents an increase of about 17.5% for these urban areas, while the U.S. population as a whole grew about 13.2% in this period.

So, the notion that urban areas in the U.S. are in the process of withering away turns out not to be accurate.

Posted by DTM | July 21, 2008 11:16 PM

DTM,

So these relative efficiencies have been changing, and I think the data supports the hypothesis that transit authorities have the ability to more quickly respond to high fuel prices with efficiency gains than the private passenger vehicle fleet.

The data doesn't remotely support your hypothesis. Your BTS table doesn't provide any data on energy efficiency for the "private passenger vehicle fleet" overall. Just for "passenger cars" and "other 2-axle 4-tire vehicle" as separate categories.

In case you don't know this, when gas prices change, many people switch between classes of vehicles. Low gas prices tend to promote switching from passenger cars to trucks, minivans and SUVs. High gas prices tend to promote the reverse. Your table doesn't capture the effect of these switches on overall "private passenger vehicle fleet" energy efficiency, because it only provides data for each class of vehicle, instead of for all vehicles.

And there are other problems with it, as I've told you before.

Posted by Mixner | July 21, 2008 11:22 PM

Mixner,

Shhh.

Posted by DTM | July 21, 2008 11:29 PM

DTM,

In short, the US private passenger fleet is gigantic in relation to new sales, and furthermore new sales are only very gradually trending to more efficient vehicles, and thanks to SUVs et al are still below the efficiency rate of 20 years ago. So, I think it will remain true that the US fleet will only very, very gradually become more efficient through new sales, and we will need to look to somewhere else for rapid efficiency gains in transportation.

The BTS publishes age data on private motor vehicles and transit vehicles. The average age of cars and light trucks in the U.S. is about the same as the average age of transit buses. The nation's transit bus fleet doesn't turn over any more quickly than the nation's private motor vehicle fleet.

Therefore, any hope you might have that transit is likely to attract market share from autos by being able to adapt more quickly to rising fuel costs is probably misguided.

And rail transit, of course, has much longer startup and vehicle turnover times than buses. A new rail system or route approved tomorrow probably wouldn't enter service for at least another decade (because of all the time required for approval, funding, environmental impact assessment, construction, etc.). And then it would need to operate for decades to accumulate benefits commensurate with its huge startup costs. This means that any new rail systems aren't going to be competing merely with the typical car of 2008, but with the typical car of 2018 and beyond. At it's virtually certain that the typical car of ten or more years from now will be much cleaner and much more fuel-efficient than the typical car of today.

Posted by Mixner | July 21, 2008 11:45 PM

Mixner,

Shhh.

Posted by DTM | July 21, 2008 11:48 PM

Mixner - Riiiight. Because the burden is on me to refute your cherry-picked list full of outliers and not for you to address my orderly list of the top five states or the obvious correlation between the top half of high density states to the top half of top income states.

See page 39 of this study on the correlation for a graph showing the correlation of population density and income (or more likely...reply off topic with new strawmen).

http://www.syedbasher.org/wp/pap28sept.pdf

Posted by joejoejoe | July 21, 2008 11:54 PM

DTM,

However, I think the total population numbers are useful. To summarize, the vast majority of urban areas increased in population.

Er, I'm not sure why you're citing data on population change between 1990 and 2000 from an unofficial source when the Census Bureau itself provides much more recent data. Not that your own source supports your position, anyway. But here's the Census Bureau Data on the 50 Fastest-Growing U.S. Metro Areas from 2000 to 2006.

As the headline notes, the fastest growing metro areas are concentrated in the sprawling new cities of the south and west, both numerically and, especially, in percentage growth. In first place was Atlanta. Followed by Dallas, Houston, Phoenix, Riverside-San Bernadino and Los Angeles.

Have you ever visited any of these places, DTM? They are mile after mile after mile of sprawling, low-density development. McMansions and malls. Freeways and 6-lane surface streets. Very limited public transportation. And this huge growth of sprawling suburbia occurred during a period of rising oil prices. So much for your theory.

Equally interesting, as the document notes, is that "of the 50 fastest-growing metro areas, none were in the Northeast."

In fact, for most of the old, dense cities in the Northeast, it has been a disaster The rust belt is disintegrating. Pittsburgh (which you hilariously thought had gained population) was the second-biggest loser of all (after New Orleans). Pittsburgh lost 60,000 people. Cleveland was next, losing 34,000 people.

Posted by Mixner | July 22, 2008 12:20 AM

More data:

As I related above, SUVs et al are continuing to increase their share of the total passenger vehicle fleet thanks to new sales.

But one might ask if maybe people could be switching passenger miles to passenger cars in response to higher fuel prices. It turns out that through 2005 (the last year in the BTS tables), the answer is no:

http://www.bts.gov/publications/national_transportation_statistics/html/table_01_37.html

From 2000 to 2005, passenger cars' share of highway passenger miles went from 58.0% to 54.7%. The SUV et al category's share of passenger miles went from 33.4% to 37.6%. So in this period of rising fuel prices, passenger-mile share continued to switch to SUVs et al and away from passenger cars.

Posted by DTM | July 22, 2008 12:22 AM

DTM,

Hush, child, hush.

Posted by Mixner | July 22, 2008 12:23 AM

Mixner,

I recognize that imitation is the sincerest form of flattery. Nonetheless, shush now.

Posted by DTM | July 22, 2008 12:37 AM

DTM,

From 2000 to 2005, passenger cars' share of highway passenger miles went from 58.0% to 54.7%. The SUV et al category's share of passenger miles went from 33.4% to 37.6%. So in this period of rising fuel prices, passenger-mile share continued to switch to SUVs et al and away from passenger cars.

Which completely contradicts your hypothesis that people wanted to switch to more energy-efficient vehicles in response to historical increases in gas prices. Since people were shifting from more-efficient cars to less-efficient SUVs and trucks even as the price of gas was increasing, other factors must have swamped the influence of energy-efficiency on vehicle purchasing decisions.

It only seems to be within the last year or two that gas prices have risen to the point at which fuel-efficiency has become a significant factor in vehicle sales.

Posted by Mixner | July 22, 2008 12:45 AM

Incidentally, for those not familiar with how the Census defined urban areas in 2000, this is the relevant link:

http://www.census.gov/geo/www/ua/ua_2k.html

The upshot is that if you are interested in what is happening with urban populations specifically, this is the most objective approach of which I am aware.

Posted by DTM | July 22, 2008 12:49 AM

Mixner at 11:18:

"In case you don't know this, when gas prices change, many people switch between classes of vehicles. Low gas prices tend to promote switching from passenger cars to trucks, minivans and SUVs. High gas prices tend to promote the reverse."

Mixner at 12:45:

"Which completely contradicts YOUR hypothesis that people wanted to switch to more energy-efficient vehicles in response to historical increases in gas prices." (emphasis added)

Aha. Pronoun trouble.

Posted by DTM | July 22, 2008 12:55 AM

joejoejoe,

See page 39 of this study on the correlation for a graph showing the correlation of population density and income

Yes, there's a correlation. A very, very, very weak one. In case you hadn't noticed, the states are all over that chart. Only a few of them are even close to the trendline. What's the standard deviation?

Posted by Mixner | July 22, 2008 12:56 AM

DTM,

Aha. Pronoun trouble.

No, just another case of "DTM managing to completely confuse himself" trouble.

So what's your position now, DTM? Is it:

(1) Between 2000 and 2005, Americans wanted to increase the fuel-efficiency of their private passenger vehicles to a greater degree than they actually did, but for some reason were not able to do so. Or,

(2) Between 2000 and 2005, other factors were more important than fuel-efficiency in private passenger vehicle purchasing decisions.

(1) is inconsistent with your post about passenger car vs. SUV market share. And (2) is inconsistent with your hypothesis in (1).

Posted by Mixner | July 22, 2008 1:12 AM

Mixner,

Shhh.

Posted by DTM | July 22, 2008 1:47 AM

DTM,

Hush, child, hush.

Posted by Mixner | July 22, 2008 2:01 AM

Drivers choose cars over trucks as auto sales plunge Updated 6/6/2008 1:21 PM

By Chris Woodyard, USA TODAY

Cars outsold the top-selling Ford F-series truck in May for the first time since 1992, a sign of the rapid shift in customers' preferences from trucks and SUVs to small cars that is forcing painful production cuts and plant closures at General Motors and Ford Motor .

For a generation, pickups and SUVs have symbolized a rugged, oversized, no-holds-barred American lifestyle.

Tuesday, automakers made it clear that consumers are hitting the brakes on their love affair with the hardiest, roomiest — and thirstiest — vehicles.

Thus the increase in popularity of SUVs and larger sedans is not continuing.

About efficiency of buses: probably it could be improved a lot, for example, by adding a fleet of vans to be used in off-peak hours, rather than running empty-ish buses, and vans cost only a fraction of what the buses use (and the transit authorities would save fuel). Buses can run on natural gas which is less scarce, and on electricity (this is less frequent, but technologically easy). With reasonable investments, buses -- and other forms of transit -- could be faster, more efficient and more attractive. Reserving some lanes for buses and high occupancy vehicles and adding express lines could improve the advantages of buses -- and carpooling.

Carpooling probably has a larger potential to reduce traffic and energy use than transit, but both should be encouraged.

One bad externality of individual driving is enormity of parking lots, another -- community institutions centered around parking lots, or surrounded by them, are hard for walking, leading to ridiculous lifestyle of driving and stopping for every single shop, library, office etc. With time so saved we can also make a separate ride to the gym. Haha!

Another externality of public transit and carpooling is lesser congestion on roads and in parking lots, which is a boon for drivers. It is not a war of civilizations but the question of the optimum mix of investments.

Posted by piotr | July 22, 2008 7:11 AM

If the BLS numbers from the last thread are correct and complete, then Mixner is right that public transportation is more heavily subsidized per passenger mile than private automotive transportation.

$400 per 1000 passenger miles (public transportation) v $5 per 1000 passenger miles (driving).

An externality of $2 per gallon would end up as less than $100 per 1000 passenger miles.

From BLS:

"Federal, state and local expenditures on highways runs about $110 billion per year, revenues from gas taxes, fees, fines and tolls etc. runs about $86 billion per year."

The Texas DOT says, at least for Texas this is not plausible:

"The Asset Value Index is the ratio of the total expected revenues divided by the total expected costs. If the ratio is 0.60, the road will produce revenues to meet 60 percent of its costs; it would be “paid for” only if the ratio were 1.00, when the revenues met 100 percent of costs. Another way of describing this is to do a “tax gap” analysis, which shows how much the state fuel tax would have to be on that given corridor for the ratio for revenues to match costs.

"Applying this methodology, revealed that no road pays for itself in gas taxes and fees. For example, in Houston, the 15 miles of SH 99 from I-10 to US 290 will cost $1 billion to build and maintain over its lifetime, while only generating $162 million in gas taxes. That gives a tax gap ratio of .16, which means that the real gas tax rate people would need to pay on this segment of road to completely pay for it would be $2.22 per gallon. This is just one example, but there is not one road in Texas that pays for itself based on the tax system of today. Some roads pay for about half their true cost, but most roads we have analyzed pay for considerably less."

The BLS says roads pay for over three quarters of their expenses. Texas DOT says some roads almost pay half, but most pay much less. Their example road, a freeway in Texas' most populous city would seem to skew toward a higher ratio, but its revenue comes in at about one sixth its expenses.

Mixner is right if BLS is right. Mixner is wrong if T-DOT is right.

Posted by Arnold Evans | July 22, 2008 7:24 AM

Car versus light truck sales of 5 largest companies in USA, April, 2007 versus 2008:

cars, sales from 417 to 453 k units,
truc, sales from 506 to 413 k units.

Whatever the motivation of the public in the era before high gas prices, it seems that many people regret their purchases of trucks.

Currently, there is a huge shortage of Prius, due to some difficulties in manufacturing. Using the opportunity, Honda sells Civic Hybrid for the same price -- even though it uses ca. 5-10% more fuel.

Posted by piotr | July 22, 2008 7:29 AM

Mixner is right if BLS is right. Mixner is wrong if T-DOT is right.

Wow. Looks like T-DOT just got a serious credibility boost!

Posted by DMonteith | July 22, 2008 9:48 AM

The BLS says roads pay for over three quarters of their expenses. Texas DOT says some roads almost pay half, but most pay much less. Their example road, a freeway in Texas' most populous city would seem to skew toward a higher ratio, but its revenue comes in at about one sixth its expenses.

Mixner is right if BLS is right. Mixner is wrong if T-DOT is right.

And for the reasons I mentioned above, BTS is at best a lower bound on costs. T-DOT is likely to be more representative of a cost-comprehensive figure.

Posted by jack lecou | July 22, 2008 10:26 AM

Mixner writes, They will (relocate from Manhattan) if they can no longer attract the employees they need to run their business, because those employees get fed up with all the time, hassle, crowds, dirt, noise, stress, etc. having to commute daily into Manhattan.

Except this hasn't happened. At all. Businesses still pay a premium to located in the central business districts of major cities.

You ask And you do of course have evidence to support this claim.....right? Yes, I do. I have the ridiculous differential between the cost of office space in Lower Manhattan vs. office space in the burbs. Ditto for Boston/Cambridge vs. 128. And, of course, there is your own argument, which states the comparative advantage of an urban location quite effectively: (New York is) the national center of finance, fashion, publishing, the arts and other important areas of American commerce. It attracts many ambitious and talented people who are willing to put up with long commutes and other unpleasantnesses for the sake of their careers. And it's a huge magnet for new immigrants, who are in demand for all the grunt work that needs to be done to support those higher up on the food chain (taxi drivers, hotel maids, restaurant workers, garbage collectors, etc.) Now, repeat that for Boston, San Francisco, Philly, Washington, Atlanta, Houston, Chicago, LA, San Diego, Las Vegas, Seattle, San Francisco, Oakland, and Dalla, and you're getting somewhere. Cities have a great natural advantage, because they generate dynamic interactions which make them centers of commerce and culture - a situation made possible specifically because of their density of population and businesses, which itself is made possible by a strong public transit system.

BTW, transit usage is up 10-20% over last year.

Posted by joe from Lowell | July 22, 2008 10:29 AM

Mixner, business that move out of Manhattan do so almost invariably to move closer to the home of the CEO.

Nevertheless, if Manhattan were so unpopular a place to locate offices, then it would be cheaper. That doesn't seem to be happening, so I thus infer that the place is fairly popular.

Posted by Tyro | July 22, 2008 11:01 AM

Mixner's problem is that he only sees transportation mode choice as a driver of public spending decision, without recognizing that the relationship is a cycle.

He looks at the vast disparity in funding, then looks at the split between transit and automobile usership, and fails to understand that spending more money on highways and less on rail will lead individuals to choose highway over rail, as well as leading developers to build in a manner that depends on highways.

Posted by joe from Lowell | July 22, 2008 11:28 AM

No, Mixner's problem is that he hates other people, and rationalizes it by saying that everything in America lines up behind his splendid basement isolation.

Posted by Muxmer | July 22, 2008 12:20 PM

Trying to identify Mixner's worst problem is a little like trying to identify the single tree which makes a small grove into a large forest.

But anyway, I will chime in that one of his many problems is that he doesn't understand basic economic concepts and their implications. So, for example, people who understand pricing automatically understand that relatively high prices for urban property indicate relatively high demand. But Mixner just thinks that means rich people are crazy.

Similarly, he has an obsession with using national averages when discussing transportation policy issues. But of course if he actually understood markets, he would understand that the transportation industry is made up of a very large number of distinct markets, and so each market has to be addressed separately.

And he doesn't understand what an externality is or how externalities relate to subsidies. And so on.

In short, these days you can't discuss public policy in a meaningful fashion without understanding basic economic concepts. And Mixner doesn't understand basic economics.

Posted by DTM | July 22, 2008 12:48 PM

piotr,

About efficiency of buses: probably it could be improved a lot, for example, by adding a fleet of vans to be used in off-peak hours, rather than running empty-ish buses, and vans cost only a fraction of what the buses use (and the transit authorities would save fuel).

We've been over this before. If you only run your regular buses at peak hours, rather than throughout the day, and leave them parked at the bus depot at all other times, your return on the capital cost of those buses goes down dramatically. You'd also have the additional capital cost of buying the parallel fleet of vans. You'd also have the additional costs associated with storing, scheduling, maintaining, etc., twice the number of transit vehicles. Your seat-mile fuel costs would also likely increase. It seems unlikely that the fuel-cost savings from a higher average load factor would offset all these other costs, which is probably one reason why no transit authority has adopted your proposal.

But even if the economics favored your proposal, practical and political considerations would probably kill it anyway. Buses are specifically designed for accessibility--wide, automatic doors to facilitate entry, exit, the collection of fares, and driver inquiries, special seating for the elderly and disabled at the front of the bus, a central aisle to facilitate access to regular seating, sufficient space to allow for bags, shopping, etc. For these reasons and others, vans just wouldn't be a realistic alternative to regular buses in most transit markets.

Buses can run on natural gas which is less scarce, and on electricity (this is less frequent, but technologically easy). With reasonable investments, buses -- and other forms of transit -- could be faster, more efficient and more attractive.

But of course these alternative fuel and engine technologies can also be applied to cars and trucks, so they won't increase the relative efficiency of buses.

Reserving some lanes for buses and high occupancy vehicles and adding express lines could improve the advantages of buses -- and carpooling.

Of course we already do this in many places. But the more highway space you reserve for special uses (buses, HOV vehicles, etc.), the less you have for regular traffic, which tends to increase congestion. It's a balancing act. If authorities go too far in restricting highway access to regular drivers, political pressure will force a change. Thus, for example, HOV 3+ lanes (cars using the lane must have at least three occupants) have been changed to HOV 2+ lanes, in response to pressure from drivers that the 3+ policy was wasteful because there were so few cars that met the requirement.

One bad externality of individual driving is enormity of parking lots,

How is that an externality?

Another externality of public transit and carpooling is lesser congestion on roads

That goes both ways. More roads means less congestion on public transit.

Posted by Mixner | July 22, 2008 1:01 PM

One bad externality of individual driving is enormity of parking lots,

How is that an externality?

You can't charge rent on a parking lot.

Another externality of public transit and carpooling is lesser congestion on roads

That goes both ways. More roads means less congestion on public transit.

Oh my god, you are in third grade.

Oh my god, you are in second grade.

[taking care of that so you don't have to]

Posted by scythia | July 22, 2008 1:21 PM

scythia,

And with respect to parking lots, not only is the landowner adversely affected, but so are the potential users of the land who would be paying that rent, since they would presumably expect to nonetheless net a benefit.

Posted by DTM | July 22, 2008 1:31 PM

Except this hasn't happened. At all. Businesses still pay a premium to located in the central business districts of major cities.

Yes, it has happened. It's been happening for decades. People and jobs have been moving out of high-density core urban areas and into low-density suburbs and exurbs. The daily suburb-to-inner-city commute is dying out. More and more, commuting means travelling from a home in the suburbs to a workplace in the suburbs.

For various reasons, New York has been something of an exception to this trend. It still has a huge concentration of jobs in a small geographical area. Hence the need for extensive mass transit. Hence the exceptionally long average commute time in New York (and the exceptionally high rate of "extreme commutes" taking 90 minutes or more). In newer cities, where most of the growth is occurring, jobs are much more decentralized. As a result, most commuters in these cities are able to drive to work, and the average commute time is therefore much shorter than it is in New York.

Posted by Mixner | July 22, 2008 1:48 PM

"People and jobs have been moving out of high-density core urban areas and into low-density suburbs and exurbs."

Again, this is why looking at the Census-defined urbanized areas is useful, since their methodology focuses on defining urban areas through density, not political boundaries. And also again, it turns out that from 1990 to 2000, the population of the largest urban areas in the United States grew at a faster rate than the nation as a whole, and indeed about 2/3s of the increase of population in the United States was in those large urban areas.

So, it turns out that among Mixner's many problems is that he is also simply wrong on key factual issues.

Posted by DTM | July 22, 2008 2:00 PM

Shorter Mixner: New York City is icky.

(Let's give him another chance to avoid OhioBoy's question: Are you proposing that we eliminate all rail service into NYC, in order to shorten commute times?)

Posted by Muxmer | July 22, 2008 2:04 PM

Mixner, there are a lot more businesses today than there were 50 years ago. They have to go somewhere, and city centers, by their very nature, have limited space in them. Thus you have a situation where, simultaneously, rents in city centers remain high because of demand while many businesses also locate themselves in more spread out, more distant areas.

In other news, the population of the original 13 colonies makes up less and less of the overall percentage of the US population.

Posted by Tyro | July 22, 2008 2:12 PM

He looks at the vast disparity in funding, then looks at the split between transit and automobile usership, and fails to understand that spending more money on highways and less on rail will lead individuals to choose highway over rail, as well as leading developers to build in a manner that depends on highways.

We spend vastly more on highways than on rail because there is vastly more demand for highways than for rail. Given the huge advantages of car travel over rail travel, and the fact that highways also serve transit in the form of buses, this isn't terribly surprising. Rail will always be relegated to niche markets in our overall surface transportation system. Rail is especially impractical and inefficient in low-density, suburban and exurban areas, where most of the growth in housing and jobs in the United States is occurring. That is why rail is so rare in such areas. The vast majority of travel in such areas made is by private motor vehicles. And most transit in such areas consists of transit buses, because they make much more sense than transit rail.

Posted by Mixner | July 22, 2008 2:20 PM

DTM,

But anyway, I will chime in that one of his many problems is that he doesn't understand basic economic concepts and their implications. So, for example, people who understand pricing automatically understand that relatively high prices for urban property indicate relatively high demand.

DTM is once again exposing his complete ignorance of even basic economic principles. Market prices are a function of both demand and supply. The fact that the price of a commodity is relatively high tells you nothing about the relative level of demand in the absence of information about supply. The supply of "urban property" is necessarily limited by the geographical constraint implicit in the concept of "urban." This is the fundamental reason why the price of "urban property" is "relatively high."

Similarly, he has an obsession with using national averages when discussing transportation policy issues.

National averages are obviously more relevant to discussions of national transportation policy than numbers that apply to only specific and unrepresentative local or regional markets.

I've pretty much given up trying to figure out what changes you seek in our national transportation policy, because your various statements and arguments on the issue are so inconsistent and incoherent. You don't seem to have any clear idea what you think on the matter.

But Matthew and other transit-boosters here clearly favor a shift in national transportation policy and priorities away from cars and towards more transit. National transportation averages are obviously relevant to that position.

Posted by Mixner | July 22, 2008 2:42 PM

Here are some references from the Victoria Transportation Policy Institute for anyone who wants to dig deeper into this topic:

Todd Litman (2004), Rail Transit In America: Comprehensive Evaluation of Benefits, VTPI (www.vtpi.org); atwww .vtpi.org/railben.pdf .

Todd Litman (2005), Evaluating Public Transit Benefits and Costs, VTPI (www.vtpi.org); atwww .vtpi.org/tranben.pdf. See in particular the discussion of comparisions between automobile and transit costs.

Todd Litman (2005c), Evaluating Rail Transit Criticism, Victoria Transport Policy Institute (www.vtpi.org); atwww .vtpi.org/railcrit.pdf.

Todd Litman (2006), Smart Congestion Reductions II: Reevaluating The Role Of Public Transit For Improving Urban Transportation, VTPI (www.vtpi.org); atwww .vtpi.org/cong_reliefII.pdf.

Posted by bobskis | July 22, 2008 3:22 PM

Mixner,

"National transportation policy" can't ignore the fact that transportation is actually made up of a large number of distinct markets with highly variable characteristics, not just one big undifferentiated national market.

Anyway, another little dramatic play:

Normal Person: OK, Nixmer, as instructed we are putting your life savings (all $512) into the "Trains are Icky!" T-Shirt line. We plan to offer XS, S, M, L, and XL, but do you want to add petite and big sizes?

Nixmer: No! In fact, we will only sell mediums!

Normal Person: Um, only a small segment of the T-Shirt market is actually medium ...

Nixmer: But what is the AVERAGE size?

Normal Person: Well, I guess that would be medium.

Nixmer: Aha! That is why all my T-Shirts will be medium! Nixmer wins! Nixmer wins!

Posted by DTM | July 23, 2008 10:18 AM

About vans for public transit: in some places in the world, private vans (or mini-buses) dominate over public buses, it is not an experimental mode of transportation. I happen to use a bus line that in off-peak hours usually has fewer than 10 passengers in the bus.

I think that a mini bus costs proportionally less than a bus, something like 1/4, and of course uses much less fuel. Augmenting a fleet of 50 buses with 25 vans would not increase capital expenses a lot. Smallish transit authorities are not run in the most economic or user-friendly manner, which decreases ridership. The minibus model could use them on more sparse routes, with reserve minibuses showing up if there is a sudden increase in demand.

Now, size of parking lots is a very negative aspect of cars as mode of transportation. These parking lots increase distances between office buildings and businesses. Larger shopping malls have parking lots more complicated than those of major airports, once I spent 30 minutes trying to get out of a shopping mall in Paramus, NJ. The jumble of wide roads and huge parking lots converted bucolic expanses of NJ to some circle of hell, in my humble opinion.

My employer charges 35 bucks per month to use a parking lot next to the building where I work. Somewhat idiotically, it invested huge millions to build a number of multi-level parking structures, rather then to encourage people to use more transit and carpool, and have more productive use for the precious space. (True, my employer has 6000 acres to waste, including several decent horse pastures. Hence my dream: have a number of horse path for people to use horses as alternative to cars; you ride to the university, leave your horse of the pasture to "refuel" and take the bus to your office. Actually, horses graze close to a commuter parking lot so buses already run there. While not a mass model for commuting, it would be very cute.)

Recently, with energy cost rising, there is growing popularity of living in more compact communities like downtowns and inner suburbs, with outer suburbs being more heavily affected by declining prices and flight of inhabitants. Mixner's stats do not reflect these changes.

Posted by piotr | July 23, 2008 6:59 PM

DTM,

National transportation policy" can't ignore the fact that transportation is actually made up of a large number of distinct markets with highly variable characteristics, not just one big undifferentiated national market.

You're still not making any sense, DTM. Regardless of the way in which any particular "distinct market" contributes to our national transportation statistics, it is the national statistics that are most relevant to debates about national transportation policy. Not the statistics for any particular market or region. You seem to be under the false impression that Matthew has argued only for an expansion of transit in a few "distinct markets" rather than in the United States as a whole.

You yourself seem to have no clear idea what changes you seek in our national transportation policy. You frequently make generic cheerleading noises in favor of more transit and intercity rail, and against highways and private autos, but you also deny that you have any general policy preference for one mode of transportation over another, which is why your writing is so confused and incoherent.

Posted by Mixner | July 23, 2008 9:11 PM

piotr,

About vans for public transit: in some places in the world, private vans (or mini-buses) dominate over public buses, it is not an experimental mode of transportation.

In some places in the world, most people live on less than five dollars a day. Just because vans or minibuses dominate buses in some places in the world (if that's even really true), obviously doesn't mean your proposal for using them here in the United States makes sense.

I happen to use a bus line that in off-peak hours usually has fewer than 10 passengers in the bus. I think that a mini bus costs proportionally less than a bus, something like 1/4, and of course uses much less fuel. Augmenting a fleet of 50 buses with 25 vans would not increase capital expenses a lot.

And if the transit authority's current fleet is only 50 buses, it wouldn't save much in fuel costs either. It's the relative cost that matters, not the absolute one. If you think you have a serious analysis showing that your proposal is feasible, not only in terms of economics but with respect to the kind of practical and political considerations I described in my previous post that transit authorities must also consider (accessibility for the disabled, etc.), then produce it.

Now, size of parking lots is a very negative aspect of cars as mode of transportation. These parking lots increase distances between office buildings and businesses.

Parking lots are part of the total cost of the car-and-highway transportation system. Those costs are obviously not high enough to induce more than a small fraction of trips to be made by public transportation rather than by car.

My employer charges 35 bucks per month to use a parking lot next to the building where I work. Somewhat idiotically, it invested huge millions to build a number of multi-level parking structures, rather then to encourage people to use more transit and carpool, and have more productive use for the precious space.

You'll have to produce a serious cost-benefit analysis of using the space for the parking structure vs. whatever "more productive use" you have in mind before this claim can be taken seriously. Guesses and wishful thinking are not a substitute for facts and figures, piotr.

Recently, with energy cost rising, there is growing popularity of living in more compact communities like downtowns and inner suburbs, with outer suburbs being more heavily affected by declining prices and flight of inhabitants.

Really? Where is your evidence of this "growing popularity?" There may have been a (modest) recent increase in the absolute number of people living in "compact communities" and "inner suburbs," but I've seen no evidence of any shift in market share towards "compact communities." Do you have any such evidence?

Posted by Mixner | July 23, 2008 9:35 PM

Guesses and wishful thinking are not a substitute for facts and figures, piotr.

See above.

Also, for the fifth time of asking: are you proposing that we eliminate all rail service into NYC, in order to shorten commute times?

Posted by Muxmer | July 23, 2008 9:40 PM

Mixner,

Again, your failure to understand that "national policy" ultimately has to reflect the actual structure of transportation markets just proves you don't understand the basic concepts we are discussing.

Anyway, you say: "You frequently make generic cheerleading noises in favor of more transit and intercity rail, and against highways and private autos, but you also deny that you have any general policy preference for one mode of transportation over another, which is why your writing is so confused and incoherent."

As I have informed you many times before, I am not "against highways and private autos." That is why you are so confused--unlike you, I view the very idea of being "against" modes of transport silly. Indeed, up above I wrote:

"Cars and such really are quite useful (meaning they are the most efficient mode of transport for many transportation needs, even factoring in the negative externalities). So I am relatively confident that even if we had spent our public transportation dollars in the most efficient way, 'highways' would still have come out ahead of 'transit' in this graph."

Again, among your many problems is that you view transportation as a zero-sum game, which is why you assume that to say anything in favor of one mode of transport in one context is necessarily to be "against" other modes of transport. But I simply don't share your childish outlook on these matters, and your inability to think like an adult is why you find me confusing.

Posted by DTM | July 23, 2008 9:54 PM

DTM,

Again, your failure to understand that "national policy" ultimately has to reflect the actual structure of transportation markets

This bizarre statement is typical of your incoherent utterances about transportation policy.

As I have informed you many times before, I am not "against highways and private autos."

Right. That must be why every time Matthew writes a new post consisting of some variation on his "We need to shift away from cars and highways and in favor of buses and trains, dammit!" meme, you make mindless cheerleading noises in support.

Despite repeated requests, you haven't even attempted to provide any serious argument in favor of the massive subsidies currently provided to transit. You seem to think a serious argument consists of uttering the word "Externalities!" as if it were a magical incantation that can justify any amount of transit subsidy, and then denying that any further analysis is necessary to support your position.

Again, among your many problems is that you view transportation as a zero-sum game,

Transportation market share is obviously a zero-sum game. While the absolute levels of both transit and private motor vehicles can increase or decrease in response to population changes or total travel volume, any significant increase in the share of total passenger-miles or total trips provided by one mode means a significant decrease in the share provided by the other. As car travel has grown to provide an increasingly large share of our total travel needs, transit has shrunk to provide an increasingly smaller share. The massive exodus of jobs and people out of the old dense cities in the northeast and into the new sprawling low-density cities of the south and west has fueled this massive growth of car travel and massive decline of transit.

Posted by Mixner | July 23, 2008 10:51 PM

Mixner,

I have linked studies discussing the externalities associated with specific cases. But you have claimed they don't discuss externalities. What that proves is that you don't know what an externality is.

Anyway, as I have noted before many times now, I don't care about market shares. Again, though, you care deeply about market shares because you are a child who has his favorite modes of transport and the ones he loves to hate. But adults care about things like efficiency, not whether or not particular modes of transport are "winning" in some made-up contest.

Posted by DTM | July 23, 2008 11:31 PM

DTM,

I have linked studies discussing the externalities associated with specific cases.

You have produced no studies whatsoever that even vaguely suggest, let alone establish to any reasonable degree of confidence, that the externality costs of transit vs. highways are such as to justify subsidizing transit on the order of $400 per thousand passenger-miles more than highways. If you think a transit subsidy of this magnitude is justified, then produce your evidence that it is justified.

Anyway, as I have noted before many times now, I don't care about market shares.

Then why do you keep going on about energy efficiency, CO2 emissions, carrying capacity, etc. of transit vs cars? If you're not trying to suggest that transit is a superior alternative to cars (that is, that we should shift market share from cars to transit), what is your point in making these comparisons at all?

Posted by Mixner | July 23, 2008 11:48 PM

Mixner,

First, as I have explained to you many times now, one cannot justify subsidies on anything but a case-by-case basis. So what I linked for you was studies for specific cases. And as I have also noted to you before, I think some transportation subsidies today are being wasted on uneconomic routes and services. So, I don't consider myself a defender of the status quo when it comes to subsidies.

"Then why do you keep going on about energy efficiency, CO2 emissions, carrying capacity, etc. of transit vs cars? If you're not trying to suggest that transit is a superior alternative to cars (that is, that we should shift market share from cars to transit), what is your point in making these comparisons at all?"

Because to understand when various modes of transportation make economic sense, you have to understand their various characteristics.

And that is my actual interest--not in deciding which mode of transport should be my favorite and which mode of transport I should detest, but rather in identifying the conditions and cases in which various modes of transport (all of them in turn) make the most economic sense. Which, again, is the sort of thing adults care about, once they get past the "Cars good! Trains bad!" phase of intellectual development.

And that is why I don't care about market shares. As long as we are doing what makes the most economic sense in particular cases, I don't see any reason to care about market shares.

Posted by DTM | July 23, 2008 11:58 PM

DTM,

one cannot justify subsidies on anything but a case-by-case basis. So what I linked for you was studies for specific cases.

Another lie. You haven't produced any study that even provides an estimate of the value of the subsidy for any specific transit system at all. Not even one. Let alone a study that compares this transit subsidy to the corresponding subsidy for highways and makes a case that the transit subsidy is justified.

Because to understand when various modes of transportation make economic sense, you have to understand their various characteristics.

More nonsense. You cannot determine whether a mode of transportation "makes economic sense" without comparing it to other modes. The mode that makes economic sense, if any, is the one with the highest ratio of benefits to costs. To favor transit over cars in any "specific case" or in the totality of cases is to seek a shift in market share from cars to transit. It's a zero-sum game.

Posted by Mixner | July 24, 2008 12:33 AM

Mixner,

You write:

"You haven't produced any study that even provides an estimate of the value of the subsidy for any specific transit system at all."

You only think that because you don't understand the relationship between the justification for subsidies and externalities, or for that matter what an externality is.

"Let alone a study that compares this transit subsidy to the corresponding subsidy for highways and makes a case that the transit subsidy is justified."

Well, some studies do in fact consider competing project proposals involving different modes of transport. But there won't always be a competing project proposal, so that is why not all studies include such a comparison.

"You cannot determine whether a mode of transportation 'makes economic sense' without comparing it to other modes. The mode that makes economic sense, if any, is the one with the highest ratio of benefits to costs."

Right, I am interested in identifying the most efficient modes for specific cases. But I have no interest in overall shares--just the specific cases.

"To favor transit over cars in any 'specific case' or in the totality of cases is to seek a shift in market share from cars to transit. It's a zero-sum game."

This is truly nonsensical.

Suppose we studied one thousand total transportation cases. Suppose in Case #526, I favored a transit solution over a car solution. Would that indicate that I had a particular goal in mind for market shares for transit and cars once we dealt with all 1000 cases? Of course not.

Now, it is true that once we dealt with all 1000 cases, there would be some sort of implied market shares as a result. But I'm not saying there is no such thing as market shares. I'm just saying I don't care about them--I just want to get the best possible answer case-by-case, and the market shares can be whatever they end up being.

Posted by DTM | July 24, 2008 1:05 AM

DTM,

You only think that because you don't understand the relationship between the justification for subsidies and externalities, or for that matter what an externality is.

No, I think it because you haven't produced any such study. Where is your study that even provides an estimate of the value of the subsidy for any specific transit system at all? Where is your study that compares this transit subsidy to the corresponding subsidy for highways and makes a case that the transit subsidy is justified? You don't have any such studies, do you, DTM? It's all just more of your same old faith-based bullshit.

Right, I am interested in identifying the most efficient modes for specific cases. But I have no interest in overall shares--just the specific cases.

You just don't get it, do you? In any case where you favor transit over highway, you are necessarily favoring a shift in market share from highway to transit.

Posted by Mixner | July 24, 2008 1:23 AM

Mixner,

(1) You clearly don't understand the studies I have linked;

(2) Studies only deal with actual proposed alternatives;

and (3) in various individual cases, I will alternatively support the use of all the modes of transport. So what I think in any given case tells you nothing about what overall market shares I would favor. And as it turns out, I just don't care about market shares.

Posted by DTM | July 24, 2008 9:10 AM