By Request: Density and Intercity Rail

Nicholas Beaudrot asks:

Are American metropolitan areas outside of the Northeast Corridor dense enough or well layed out enough to support inter-city rail? Beyond putting light- or medium-rail in those cities, how much would have to change before inter-city rail made sense as a way to travel from, say, Milwaukee to Indianapolis?

As a general matter, rail works better as an alternative to driving when the destination is someplace with a good walking/transit network. That said, for a lot of trips it's not really necessary to have a car at your destination even if your destination is a very car-dependent area. I went to Forth Worth for a conference once and both the hotel and the convention center are in Fort Worth's smallish walkable downtown. Combined with a cab ride to the Fort Worth Cultural District to see the Kimbell Museum and the Fort Worth Modern that contained plenty of things to do for a few days without the expense of renting a car and that kind of thing is reasonably common for business travel.

Second rail is not only an alternative to driving, but also an alternative to flying. There are a lot of flights between Portland and Seattle, for example or between Chicago and Detroit. Any time you have two cities that are pretty close by and serviced by a lot of flights, you have a situation where a good inter-city rail option would attract customers notwithstanding any issues related to the density of the destination city. Replacing air trips with train trips is good for the environment, and any time you have a viable rail option that'll displace some of the intercity car travel which is also good. Meanwhile, a passenger rail hub can become a focal point for neighborhood development and a node on a growing urban transit network.

Comments

Another advantage of rail over flying for shortish distances is that w/ rail you usually end up in the middle of the city or at least close to it while with flying you are inevitably out in some far away part having to deal with an expensive cab ride or some other way in to town. A fast train between down-town portland and Down-town Seattle might even be faster than flying once you take into account driving from Sea-Tac.

Posted by matt (not the famous one) | June 24, 2008 3:33 PM

why are you writing a blog instead of, say, running for office?

Posted by tnko | June 24, 2008 3:34 PM

For what it's worth, just a couple of months a go I took Amtrak between Portland and Seattle. It worked just fine and was real cheap.

The thing is though, that the fact that everything went smoothly was just dumb luck. My understanding is that the train is often subject to major delays by the freight lines needing to use the tracks. So that uncertainty is a major impediment to the system's use.

Posted by ResumeMan | June 24, 2008 3:35 PM

For what it's worth, just a couple of months a go I took Amtrak between Portland and Seattle. It worked just fine and was real cheap.

The thing is though, that the fact that everything went smoothly was just dumb luck. My understanding is that the train is often subject to major delays by the freight lines needing to use the tracks. So that uncertainty is a major impediment to the system's use.

Posted by ResumeMan | June 24, 2008 3:38 PM

One other advantage for rail is the lack of a need for security theater. You are supposed to arrive at the airport well before your flight to allow time to shepherd you and everybody else through all the searches/removal of shoes etc. Not yet doing any of this crap to rail passengers (not to mention no no-fly lists for trains either). Plus I think they let you have more carry-ons.

Posted by jimBOB | June 24, 2008 3:42 PM

Matt,
Interesting that you use Fort Worth as an example ... rail service there is not as out there as you might imagine. Currently is rail service - Trinity Rail Express or TRE - between Fort Worth and Dallas - linking the two the the DFW airport. In addition there is a proposal for light rail in the city - an ultra-light rail street car type system. There's been some talk about the plan over the past year and the Star Telegram recently covered the issue .. http://www.star-telegram.com/metro_news/story/706843.html

Posted by Alex in DC | June 24, 2008 3:42 PM

ResumeMan - Amtrak's reputation for major delays comes mostly from the really long-haul trains (Chicago-SF, LA-Seattle, etc). My experience in many, many trips on the line between Seattle and Portland is that the Cascades corridor trains, which run only in BC, WA and OR, are pretty reliable.

Posted by cdc | June 24, 2008 3:50 PM

By ResumeMan:

> My understanding is that the train is often
> subject to major delays by the freight lines
> needing to use the tracks.

That's what I have heard too- that freight has right of way over passengers. In Europe, I understand that it is the opposite- passenger trains always take precedence over freight. Yet another thing to add to the list of things to change about rail policy.

Speaking also of the Portland-Seattle train - I had a friend who would commute back and forth fairly often to see her boyfriend (about twice a month), and she found it to be a reasonable way to go - that particular route is in decent shape. And with gas at about $4.25 here, it's about on par with driving (cheaper if you have a gas guzzler).

cheers-
Eric

Posted by Eric | June 24, 2008 3:55 PM

Any time you have two cities that are pretty close by and serviced by a lot of flights, you have a situation where a good inter-city rail option would attract customers notwithstanding any issues related to the density of the destination city.

I think a Pittsburgh-Philadelphia line would be a great example of this, plus it would have the added benefit of extending westward the existing rail infrastructure of the Northeast Corridor.

Posted by Mike T | June 24, 2008 4:00 PM

How many of the flights between close cities (such as Portland and Seatle) are connecting flights? Because if say Seatle is just a transit hub, not a final destination, flying there might make more sense than taking the train.

Posted by MS | June 24, 2008 4:01 PM

The hypothetical rail line between Portland and Seattle goes all the way to Vancouver, BC. It goes through some really beautiful areas, offers nice seats and decent beer/food, and is relatively affordable.

Posted by ben | June 24, 2008 4:04 PM

Though transit is obviously a nice benefit for any rail destination, let's not forget that by this point, most cities have some sort of carshare. So even if the city is auto-dependant, a service like ZipCar would let you get wherever you need to go.

Posted by Roman | June 24, 2008 4:04 PM

That said, for a lot of trips it's not really necessary to have a car at your destination even if your destination is a very car-dependent area.

It may not be "necessary," but the vast majority of American cities are not dense enough to make walking or transit a practical alternative to car travel for most trips. This is especially true for business trips. Business travellers are even less likely to be willing to put up with the inconvenience, discomfort and lost time involved in using transit than are leisure travellers.

Second rail is not only an alternative to driving, but also an alternative to flying. There are a lot of flights between Portland and Seattle, for example or between Chicago and Detroit. Any time you have two cities that are pretty close by and serviced by a lot of flights, you have a situation where a good inter-city rail option would attract customers notwithstanding any issues related to the density of the destination city.

Not at a price they would be willing to pay. Even the Acela, with its massive subsidies, is still so expensive that most of its riders are business travelers. Most people who have to pay their own way use cars or buses or discount air fares instead.

Replacing air trips with train trips is good for the environment, and any time you have a viable rail option that'll displace some of the intercity car travel which is also good.

Rail already has such a minuscule share of total intercity travel that even a large increase would have a negligible environmental benefit. And there isn't going to be a large increase, because the subsidies required to make intercity rail ticket prices affordable to the general population are so huge that Congress would never provide it. The only real question is how much longer it will be willing to provide the existing level of subsidy.

Posted by Mixner | June 24, 2008 4:06 PM

Also the Seattle train station lets you out in downtown, right across from Qwest Field (home of the Seahawks) and Safeco Park (home of the shell of the Mariners). It's a great alternative to trying to drive to the game and beating the traffic.

Posted by ben | June 24, 2008 4:08 PM

By the way, Milwaukee to Indianapolis is about 275 miles. Even accounting for a possible train change in Chicago, a high speed network could make that about a 3 hour trip, easily competitive with flying. In fact, I just checked Orbitz and only found one airline flying that route direct--Midwest (which is Milwaukee based).

Posted by DTM | June 24, 2008 4:10 PM

Southwest Airlines helped destroy an intra-city rail system that would have linked Dallas, Houston and San Antonio.

Posted by UN Plaza | June 24, 2008 4:15 PM

Another consideration is how miserable the drive is. What other parts of the country have a route as unpleasant to drive as I-95 in the northeast? When I take the Northeast Corridor Amtrak trains, that's the drive I'm trying to avoid. The same miles on an open road in the midwest would be another thing altogether.

Posted by M.C. | June 24, 2008 4:23 PM

Mike T,

The major problem with Philly to Pittsburgh is that it is rough terrain, so getting high speed service along that route would be expensive. But it might be worth it, since as you note the real benefits come when you can create networks, and that Philly-Pittsburgh link would be the logical connection between the Northeast Corridor and a Midwest network based in Chicago. Plus, central PA is one of the places losing a lot of air service in the cutbacks, so people in those areas would likely supply additional demand even for what added up to pretty long routes.

Posted by DTM | June 24, 2008 4:42 PM

The issue you haven't mentioned goes beyond density: rights of way and rails that aren't also used by freight. Amtrak owns the northeast rails. They have the right to use freight rails elsewhere, but rails that get flattened by 90-100 tons for each freight car aren't exactly the best for passenger cars.

Posted by Sherman Dorn | June 24, 2008 4:43 PM

The frequent, crowded flights between Dallas and Houston (4 hours driving; 45 minutes flying; e.g., perfect rail distance) are evidence of the demand for such a service even when the cities have only marginal local mass transit services. (business people not being willing to put up with that much mass transit, but being more than willing to pay for the taxi between the airport and downtown...)

Posted by Vermando | June 24, 2008 4:43 PM

There's a really simple solution to the dilemma of taking a train to a car friendly destination: take the car with you:

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

Posted by Chris Fabri | June 24, 2008 4:45 PM

Because if say Seatle is just a transit hub, not a final destination, flying there might make more sense than taking the train.

Or you could have some sort of connection to SeaTac from the train station. I wouldn't suggest using it as a model, but the LIRR connects to the AirTrain in order to get people to JFK, and I think that NJ Transit has a station at Newark.

Posted by daveNYC | June 24, 2008 4:51 PM

"Replacing air trips with train trips is good for the environment, and any time you have a viable rail option that'll displace some of the intercity car travel which is also good."

What Matt forgets here is the multi-year battle that would ensue over rights of way for each such line. Environmental impact statements (beloved by the left), property fights, etc, etc, etc.

Meanwhile, the planes Matt thinks are inconvenient can fly without needing to get a right of way organized first; there are a very small number of restricted airspaces in the US.

Rail is a good alternative in Matt's head. Back here in the real world, things are way, way more complex. Sure, there are similar issues with runway expansion - but the amount of land involved there is way, way smaller.

Posted by James Robertson | June 24, 2008 4:55 PM

"The frequent, crowded flights between Dallas and Houston (4 hours driving; 45 minutes flying; e.g., perfect rail distance) are evidence of the demand for such a service even when the cities have only marginal local mass transit services. (business people not being willing to put up with that much mass transit, but being more than willing to pay for the taxi between the airport and downtown...)"

Dallas-Houston is a bad example. Southwest has such a dominant share of this route that they can afford a huge number of flights, and they don't fly from far-from-downtown DFW or Houston Intercontinental -- instead, they fly from the near-downtown Love to the sort-of-near-downtown Hobby. So you essentially have flights every 30 minutes or so to and from airports within 20 minutes of the city center, and you normally only need to show up 20-30 minutes before the flight leaves. It's sort of an absolute best case for flying.

Now, if you needed to arrive 45-60 minutes before the flight, and the flights were more like 2 hours apart, AND the airports were more like 35-45 minutes from the city center, then I can almost guarantee that you would see express train service by now.

Posted by Joe | June 24, 2008 5:00 PM

vermando,

The frequent, crowded flights between Dallas and Houston (4 hours driving; 45 minutes flying; e.g., perfect rail distance) are evidence of the demand for such a service even when the cities have only marginal local mass transit services.

If "frequent, crowded" flights are evidence of anything, it's that air service may need to be increased. It doesn't tell you anything about the demand for rail service. How much would the rail tickets cost?

Posted by Mixner | June 24, 2008 5:11 PM

Re: Best case train routes.

Going back to the Dallas-Houston traffic, I think this illustrates where we should initially concentrate -- cities that have a lot of back-and-forth business traffic that (1) are too far to realistically drive and (2) have inefficient plane service when travel to and from airports is considered. Some non-East Coast ideas that come to mind are: Chicago-(Minneapolis, St. Louis, Milwaukee, Cleveland, Indianapolis), LA-SF, Seattle-Portland, Denver-Colorado Springs.

Posted by Joe | June 24, 2008 5:11 PM

The amount of land involved in runway expansion is irrelevant. Iwo Jima is only 20 square kilometers. It took three Marine divisions 38 days to take it away from the Japanese. The opposition to airport expansion is only marginally less fierce. When was the last significant expansion of a major US airport -- anybody know?

A further question for J. Robertson (or Mixner or whoever): How are costs of right-of-way acquisition for new rail lines different from those for new highways? what is your position on the desirability of new highways?

Posted by roac | June 24, 2008 5:12 PM

Matt: "Replacing air trips with train trips is good for the environment, and any time you have a viable rail option that'll displace some of the intercity car travel which is also good."

James Robertson: What Matt forgets here is the multi-year battle that would ensue over rights of way for each such line. Environmental impact statements (beloved by the left), property fights, etc, etc, etc.

James, you're being unresponsive. Matt didn't "forget" any of that, it's just not relevant to his point. Matt is saying that rail infrastructure would be good. You're saying that it would be difficult. But the two are hardly mutually exclusive! But in order to enact a difficult change (and ALL change is difficult, there's always a status quo bias), the first task is to convince people that the change in question is worthwhile. Are you saying that, in fact, substituting rail travel for air/car travel would be a bad thing?

Posted by OhioBoy | June 24, 2008 5:13 PM

I have heard rumors that Southwest Airline is the primary opponent of high speed rail in California. I'll be sad if that's true.

Posted by Megan | June 24, 2008 5:19 PM

Or you could have some sort of connection to SeaTac from the train station. I wouldn't suggest using it as a model, but the LIRR connects to the AirTrain in order to get people to JFK, and I think that NJ Transit has a station at Newark.

Posted by daveNYC | June 24, 2008 4:51 PM

Although it's not official policy, groups in the UK conducting feasibility studies for a UK-wide high-speed rail network have argued that congestion at Heathrow could be greatly lessened by making Heathrow a direct stop on a high-speed line that connects with the Eurostar train.

In other words, Eurostar trains that dock at London's St. Pancras station would continue on to Heathrow and then into Northern England and into Scotland.

Heathrow would actually become a second major train station in the process. And you could eliminate a lot of short flights from Heathrow were a good high-speed rail line developed. For example, someone in Birmingham trying to get to Los Angeles could take a high-speed train directly to Heathrow airport and then fly out of the airport to LA.

Posted by ASI | June 24, 2008 5:20 PM

Joe,

You might find this interesting:

http://www.fra.dot.gov/us/content/203

It more or less confirms your suggestions (minus Colorado), and makes a few more.

roac,

Land costs tend to favor rail over new highways. That is because the capacity per width of rail is significantly higher (around three times, sometimes more).

Posted by DTM | June 24, 2008 5:25 PM

By the way, I think it is somewhat interesting to compare that map of possible high speed rail corridors to this map of current airline service cutbacks:

http://www.usatoday.com/travel/news/2008-05-30-airline-cutbacks_N.htm

Posted by DTM | June 24, 2008 5:37 PM

About 1.5 million passengers fly between Dallas and Houston each year.

The distance between Dallas and Houston is about 240 miles.

The federal government subsidizes rail travel at $0.2 per passenger-mile more than air travel.

So for rail to take half the market (750,000 passengers), assuming that that many people would even be willing to pay rail ticket prices comparable to the current Amtrak average for that distance, would require something like $36,000,000 a year in federal subsidies (750,000 * 0.2 * 240). About $50 per passenger per trip. That's just for regular-speed Amtrak service (less than 100 mph). High-speed service, requiring special track, trains and more energy, would require much more.

That's about 40% of the cost of a walk-up ticket on Southwest ($122). I just checked.

Next time I fly between Dallas and Houston, will the government pay 40% of the price of my ticket?

Posted by Mixner | June 24, 2008 5:39 PM

I love it when proponents of sprawl come on here to tell us how the world "really is".

Just for fun, I did a real world analysis of getting to Pittsburgh from Philly. The train ride that exists now takes about 7 hours and includes 11 (eleven!!!) stops inbetween. The cost is $45 each way. So, 7 hours each way, $90.

Driving takes 308 miles, or approximately 5 hours (according to google) assuming no traffic. Let's assume your cars get 30 mpg highway and gas prices are $4 a gallon. That's $40 each way for gas, plus $16.75 toll for the turnpike. So, that's 5 hours each way, and total $113.50.

Southwest had flights for $160 roundtrip, 1.5 hours each way. Add another 1.5 hours for the airports and that's 3 hour each way, for a total of 160. Not bad, but by far the most expensive option.

So you actually save money if you go by rail, but all the stops slow you down. I imagine an express train would more or less even out the time difference.

Posted by Roman | June 24, 2008 5:41 PM

DTM,

Land costs tend to favor rail over new highways. That is because the capacity per width of rail is significantly higher (around three times, sometimes more).

Wrong yet again. What matters is not capacity, but volume. Your train may have the capacity to carry three times as many passengers per unit width of land as a highway, but if nine out of ten seats on the train are usually empty, the capacity isn't terribly relevant.

Posted by Mixner | June 24, 2008 5:46 PM

"It more or less confirms your suggestions (minus Colorado), and makes a few more."

The problem with the proposed San Antonio-Dallas-Oklahome/Arkansas corridor is that you run into the same SW problem -- air service between these cities is already plentiful, and since the Dallas airport is close to the city center, you eliminate a lot of the downtown-to-downtown benefit of a regional train.

As for Colorado, it's actually interesting. With several biggish (100,000-600,000) Front Range cities a few hours apart stretching (really) from Albuquerque to Cheyenne, it would seem to be a prime rail corridor -- except the intercity traffic (except for Colorado Springs-Denver) is fairly minimal. An extensive heavy commuter rail hub-and-spoke system with trains running from Colorado Springs (maybe extended to Pueblo), Boulder, Fort Collins and maybe Greeley would be a smart idea (and indeed, I think that's what is currently planned).

Posted by Joe | June 24, 2008 5:46 PM

Another advantage cars have is the ability to take second and third passengers at zero extra cost. The price of an Amtrak ticket vs. the cost of driving alone seems to be about the same on the routes I have checked, but with multiple passengers the advantage is all with the car. Amtrak could establish a policy of second and third passengers in a group riding free or at very low cost. Since their marginal cost is zero, they would make money on the deal and make more rail converts.

Posted by David L. Todd | June 24, 2008 5:46 PM

American cities are built for horses! Any transportation model that doesn't understand that the horse is the basis of personal freedom is foolhardy. That's why horseshit is the future!

Posted by Granpa Dixner, c. 1900 | June 24, 2008 5:47 PM

Another advantage cars have is the ability to take second and third passengers at zero extra cost. The price of an Amtrak ticket vs. the cost of driving alone seems to be about the same on the routes I have checked, but with multiple passengers the advantage is all with the car. Amtrak could establish a policy of second and third passengers in a group riding free or at very low cost. Since their marginal cost is zero, they would make money on the deal and make more rail converts.

Posted by David L. Todd | June 24, 2008 5:51 PM

Doesn't San Francisco - San Jose (actually Gilroy) and now Sacramento count as a useful rail link outside the northeast? And has been functioning continuously as such since 1868? (I know, the station's not on Valencia anymore, but still.)

And could easily be put back into Monterey, and with a bit more effort into Santa Cruz?

By the way, is there any in service rail line with a longer continuous history of uninterrupted service?

Posted by Gene OGrady | June 24, 2008 5:53 PM

roman,

Driving takes 308 miles, or approximately 5 hours (according to google) assuming no traffic. Let's assume your cars get 30 mpg highway and gas prices are $4 a gallon. That's $40 each way for gas, plus $16.75 toll for the turnpike. So, that's 5 hours each way, and total $113.50.

But if there are two passengers in the car, the cost drops to $56 per passenger. Only two-thirds the cost of the (massively-subsidized) Amtrak tickets. And you save 2 hours each way. And you have a car to use when you get to Philly, so you don't have to pay for a rental car or taxis/transit. If there are more than two passengers in the car, you save even more. It's no wonder Amtrak can't compete, even with its massive subsidies.

Posted by Mixner | June 24, 2008 5:55 PM

The ability to actually build rail lines is relevant - if it will take 30 years to argue through the building (see the IPCC in MD. While it's a road, the land requirements are the same).

Runways are easier to allocate, and in fact, have been built out in the last few years in more than one airport

Like LAX and O'Hare, for instance. Rail will take - based on how actual opposition tends to play out - take about as long as the IPCC has here in MD.

Not what I'd call useful.

Posted by James Robertson | June 24, 2008 5:57 PM

Wrong yet again. What matters is not capacity, but volume. Your train may have the capacity to carry three times as many passengers per unit width of land as a highway, but if nine out of ten seats on the train are usually empty, the capacity isn't terribly relevant.

Actually Amtrak's capacity is strained. Acela is 90% capacity during peak hours. Now with the high fuel prices, other routes are selling out:

Many of the long-distance trains are already sold out for some days this summer. Want to take Amtrak’s daily Crescent train from New York to New Orleans? It is sold out on July 5, 6, 7 and 8. Seattle to Vancouver, British Columbia, on July 5? The train is sold out, but Amtrak will sell you a bus ticket.
http://travel.nytimes.com/2008/06/21/business/21amtrak.html?scp=3&sq=Amtrak&st=nyt

Fuel prices aren't going down. This is the new reality. Either we adapt with an eye to the future, or else.

Posted by roman | June 24, 2008 5:59 PM

And to emphasize the point, here's why rail doesn't really work in the US:

arithmetic

Sure, there are areas outside the northeast where intercity rail could conceivably work - assuming you could get past the rights of way issue, and the years of NIMBY delays in construction that would happen. While those fights are ongoing, air travel will continue to work for people, because it's there already, and it's flexible between short haul and long haul trips.

Posted by James Robertson | June 24, 2008 6:05 PM

I have ridden Amtrak between Eugene, OR to Portland,OR and between Portland and Seattle many times over the last 30 years. I enjoy train travel. That being said, I would never, ever ride any of the trains on this corridor if I needed to be somewhere at a certain time. The train is more often late (up to 4-6 hours) than not. It would be wonderful if the NW system could be a reliable method of travel. Sadly, it is not.

Posted by Laurie | June 24, 2008 6:07 PM

roman,

Actually Amtrak's capacity is strained. Acela is 90% capacity during peak hours.

Peak load factor is irrelevant with respect to the cost-benefit comparison. What matters is average load factor. The more variation there is in demand between peak and off-peak periods, the more empty seats there will be at off-peak times if peak demand is to be accommodated. This also applies to variations in demand at different points along the train's route.

Fuel prices aren't going down. This is the new reality.

But Amtrak is also subject to fuel price increases. Especially since most of its locomotives are diesel. It's going to be burning through those federal subsidies even faster. Which means it will run out of money sooner. Or it'll have to raise ticket prices. Or cut back services.

Posted by Mixner | June 24, 2008 6:09 PM

I have ridden Amtrak between Eugene, OR to Portland,OR and between Portland and Seattle many times over the last 30 years. I enjoy train travel and the route is beautiful. That being said, I would never, ever ride any of the trains on this corridor if I needed to be somewhere at a certain time. The train is more often late (up to 4-6 hours) than not. It would be wonderful if the NW system could be a reliable method of travel. Sadly, it is not.

Posted by Laurie | June 24, 2008 6:12 PM

But if there are two passengers in the car, the cost drops to $56 per passenger. Only two-thirds the cost of the (massively-subsidized) Amtrak tickets. And you save 2 hours each way. And you have a car to use when you get to Philly, so you don't have to pay for a rental car or taxis/transit. If there are more than two passengers in the car, you save even more. It's no wonder Amtrak can't compete, even with its massive subsidies.

You think you're really clever with the whole "massively subsidized" shtick. Like roads that we drive on aren't "massively subsidized". Or the gas we fill our cars with isn't "massively subsidized". Fact is, right now, today, it's cheaper for one person to take the train to Pittsburgh than it is to drive. Not to mention that train travel for some is a much more pleasant experience. If you want to break it down to two people fine. Than let's include costs like car payments, parking, car maintenance and car insurance.

Posted by Roman | June 24, 2008 6:15 PM

You think you're really clever with the whole "massively subsidized" shtick. Like roads that we drive on aren't "massively subsidized".

Drivers pay more in gas taxes to the federal government than it spends on roads and highways.

Or the gas we fill our cars with isn't "massively subsidized".

If it is, then the diesel fuel Amtrak pumps into its trains also benefits from that subsidy, so it doesn't help your case.

Fact is, right now, today, it's cheaper for one person to take the train to Pittsburgh than it is to drive.

Only because the government pays so much of the true cost of providing the rail service. If the person is travelling for pleasure, he's probably not going to be alone. He'll probably be travelling with spouse/family/friends, and driving is a complete no-brainer. Even if he is alone, the car saves time (2 hours each way) and provides a means of getting around at his destination. If he's travelling alone on business, he's probably much more sensitive to differences in time than differences in price. So he'll most likely fly.

Like I said, it's no wonder Amtrak can barely compete, even with its massive subsidies.

Posted by Mixner | June 24, 2008 6:27 PM

Thank goodness at least one other person noted the existence of Caltrain-that must mean there's at least one other Californian here. There is already pretty decent commuter rail service between San Jose and San Francisco, as well as the Altamone Commuter Express that connects San Jose and Stockton.

What slows Caltrain down is the large number of stops it makes-there are mini bullet trains during peak commuter hours that make fewer stops.

San Jose-San Francisco may be an outlier in the Western US, though-lots and lots of people commute between SF and SJ, whereas even for close routes like Eugene-Portland or Seattle-Portland, yeah, there are business people who have to make those trips,but not nearly anywhere near the density of the Bay Area.

Posted by Amanda in San Jose | June 24, 2008 6:41 PM

Thank goodness at least one other person noted the existence of Caltrain-that must mean there's at least one other Californian here. There is already pretty decent commuter rail service between San Jose and San Francisco, as well as the Altamone Commuter Express that connects San Jose and Stockton.

What slows Caltrain down is the large number of stops it makes-there are mini bullet trains during peak commuter hours that make fewer stops.

San Jose-San Francisco may be an outlier in the Western US, though-lots and lots of people commute between SF and SJ, whereas even for close routes like Eugene-Portland or Seattle-Portland, yeah, there are business people who have to make those trips,but not nearly anywhere near the density of the Bay Area.

Posted by Amanda in San Jose | June 24, 2008 6:43 PM

Roman,

I heartily concur that automobiles are subsidized - it is one part of the national conversation regarding our national transportation costs that is missing.

Another part of our transportation structure that is subsidized is the airlines (by hundreds of billions of dollars) in airport subsidizes, air traffic control and security, tax preferences, and special treatment from the government including protection from foreign competition.

Mixner - I'd really love to see numbers to believe your claim that Amtrack's subsidizes are out of line in comparison to automobiles and airplanes.

Posted by Alex in DC | June 24, 2008 6:44 PM

Drivers pay more in gas taxes to the federal government than it spends on roads and highways.
That's blatantly untrue, unless you can back it up with hard numbers.

In a June 12, 2008 interview with Reuters[37], current Amtrak President Alex Kummant made specific observations: $10 billion per year is transferred from the general fund to the Highway Trust Fund; $2.7 billion is granted to the FAA; $8 billion goes to "security and life safety for cruise ships." Overall, Kummant claims that Amtrak receives $40 in federal funds per passenger, while highways are subsidized at a rate of $500-$700 per automobile. Moreover, Amtrak provides all of its own security, while airport security is a separate federal subsidy. Kummant added: "Let's not even get into airport construction which is a miasma of state, federal and local tax breaks and tax refinancing and God knows what."

Critics claim that gasoline taxes amount to use fees that entirely pay for the government subsidies to the highway system and aviation. In fact this is not true: gas taxes cover little if any of the costs for "local" highways and in many states little of the cost for state highways.[52][53] They don't cover the property taxes foregone by building tax-exempt roads. They also don't cover policing costs: Amtrak, like all U.S. railroads, pays for its own security, the Amtrak Police; road policing and the Transportation Security Administration are paid for out of general taxation.

In Pa, there was talk of selling the turnpike and/or tolling I-80 to make up for the lack in highway funding.

If it is, then the diesel fuel Amtrak pumps into its trains also benefits from that subsidy, so it doesn't help your case.
Pretty sure diesel is used mostly for freight, passenger trains use electric power.
Only because the government pays so much of the true cost of providing the rail service. If the person is travelling for pleasure, he's probably not going to be alone. He'll probably be travelling with spouse/family/friends, and driving is a complete no-brainer. Even if he is alone, the car saves time (2 hours each way) and provides a means of getting around at his destination. If he's travelling alone on business, he's probably much more sensitive to differences in time than differences in price. So he'll most likely fly.

Like I said, it's no wonder Amtrak can barely compete, even with its massive subsidies.
I doubt any businessmen are taking the train to Pittsburgh from Philly these days. It's not set up for that yet, but it certainly could, as the fuel prices contine rising. However, plenty of businessmen are taking Acela, that's why it has 90% capacity during peak hours. My whole point about the Pittsburgh train is that, even though, line has been completely neglected, it's really not far from being competitive. In fact an express line from Harrisburg to Philly (that saw $145 million in renovation) takes 90 minutes. If you extend that to Pittsburgh, it'll be quicker than driving.

Infrastructure needs constant investment. Rail, highway and air. We shell out billions of dollars on highway and air, but rail gets ignored and people cry about it being subsidized. Again, it's all subsidized. Rail might've made less sense in the time of cheap and abundant energy, but that time is ending. So it seems to makes sense to focus on the most energy efficient mode of transportation.

Posted by roman | June 24, 2008 7:03 PM

Alex in DC,

Mixner - I'd really love to see numbers to believe your claim that Amtrack's subsidizes are out of line in comparison to automobiles and airplanes.

From the Bureau of Transportation Statistics:

Net Federal Subsidies Per Thousand Passenger-Miles By Mode, 2002

Highway: -$1.00 (yes, for highways it's a negative subsidy. The government makes a profit).

Commercial Aviation: $6.18

Railroad: $210.31

Care to explain why rail transportation deserves this vastly disproportionate subsidy?

Posted by Mixner | June 24, 2008 7:23 PM

Care to explain why rail transportation deserves this vastly disproportionate subsidy?

Because the study is comparing apples to oranges?

Most highway miles are suburban commuting which are best compared to mass transit NOT interurban rail. (Looked on google for stats, couldn't find them, but ~70M commuters with an average ~20 mile commute just dominates all other highway activity).

Those empty four lane highways in Wyoming don't come close to paying for themselves. Hence the massive Eisenhower interstate highway system subsidy.

Posted by theo | June 24, 2008 7:46 PM

"Actually Amtrak's capacity is strained. Acela is 90% capacity during peak hours.

'Peak load factor is irrelevant with respect to the cost-benefit comparison. What matters is average load factor. The more variation there is in demand between peak and off-peak periods, the more empty seats there will be at off-peak times if peak demand is to be accommodated. This also applies to variations in demand at different points along the train's route.'"

Mixner - if you think peak load is irrelevant then you understand nothing about transportation economics. The economics of any capacity increase is the marginal cost of increased capacity. You know marginal cost from Economics 101. Supply and demand.

If you only use average load factor then NO road in America is congested. We can stop all road projects tomorrow, as all road's average capacity utilization is under 100%!

Back in the real world capacity increases are, or should be cost-benefited based on the marginal cost of additional capacity as one of the factors.

By the way you have not responded to my point that transit uses less that one percent 1% of the land for parking that private autos do and offers flexibility in parking placement, rather than wasting 25%-75% of the destination land area for parking, as autos do. This is a huge productivity boost for the entire economy.

Posted by Brian S. | June 24, 2008 7:57 PM

Brian,

Transit has those advantages, but it has disadvantages as well - it doesn't necessarily go where you want it to. Transit is great when you are trying to move lots of people from point A to point B (and then back again later).

It works less well as that becomes less true. If you have a typical set of suburbs - and the typical set of jobs spread amongst them - transit becomes very, very hard to set up, as there's no critical mass of people to move to/from any given set of points. In such a situation, cars - and their ability to drive "anywhere" - are far more useful.

The trouble for transit advocates is that this latter scenario is exactly what we have in the US, and the trend in that direction has been steady to accelerating for decades now.

Posted by James Robertson | June 24, 2008 8:09 PM

The US will spend somewhere around a trillion dollars on the Iraq war, which was fought mainly to secure long-term oil interests - that's some subsidy.

Posted by novakant | June 24, 2008 8:10 PM

The bottom line is that the debate over subsidies to existing infrastructure is really beside the point -- the question is, what is the most efficient use for a marginal subsidy dollar.

In many areas, the marginal benefit from building a new highway lane as significantly less than the train alternative, at roughly the same price per mile.

This is particularly true in California: there are a number of areas in California where it is impossible to add highway capacity at any feasible cost -- mountain passes, or urban areas where extensive demolition would be required. Same is true of the urbanized Northeast.

Not to mention that the benefit of a marginal subsidy dollar to a transportation mode depends on demand.

From the BTS report Mixner linked:

Economic theory argues that the economically efficient price, which maximizes consumer welfare, is the price that just covers the marginal costs of transportation usage. If a transportation mode has high fixed costs, but low variable costs of operation, charging a fare that covers all of the fixed costs may discourage usage to the point that the infrastructure is underused and consumer benefits are reduced.

It's pretty obvious to anyone who's ever ridden Amtrak that we're currently barely covering the fixed costs of running an intercity rail service. Acela proves the demand for rail is there. The variable cost of operation (i.e. the cost of running additional trains) is low.

So spending an extra dollar on highways gives you a linear return, but spending an extra dollar on certain rail improvements give you a nonlinear return.

Posted by theo | June 24, 2008 8:10 PM

Umm,

For areas where it's too expensive to add highway lanes, I'm not sure where you would stick additional rail lines....

Posted by Jam | June 24, 2008 8:15 PM

Theo,

Because the study is comparing apples to oranges? Most highway miles are suburban commuting which are best compared to mass transit NOT interurban rail.

Er, transit is subsidized almost as much as railroad ($159 per thousand passenger-miles), so this doesn't help the case for railroad subsidies. Try again.

Posted by Mixner | June 24, 2008 8:25 PM

James Robertson,

you have a typical set of suburbs - and the typical set of jobs spread amongst them...this latter scenario is exactly what we have in the US, and the trend in that direction has been steady to accelerating for decades now.

I would hardly call the trend "accelerating" right now. More like "hit the wall, and going in reverse."

To point out the obvious, oil is increasingly expensive, and cities are increasingly liveable.

Right now we have an enormous glut of suburban sprawl and no one is building anything profitable in the exurbs. Right now, transit oriented development is practically the only residential construction happening.

Posted by theo | June 24, 2008 8:28 PM

Jam,

For areas where it's too expensive to add highway lanes, I'm not sure where you would stick additional rail lines....

Thanks to 50 years of highway construction, a lot of highways have already maxed out their rights-of-way and can't get any wider.

There are a lot of unused (or underused) rail rights-of-way, many of which have capacity for an extra line or two. That's the plan for the California high speed rail system.

Posted by theo | June 24, 2008 8:33 PM

Umm, Jam - here in Maryland, as you head west out of the DC/Baltimore urban area, there are many, many exurbs still going up. maybe that's not true in other areas, but here in MD, there's plenty of that going on. And DC and Baltimore are still losing population.

Posted by James Robertson | June 24, 2008 8:36 PM

Mixner,

The topic is intercity rail, remember?

Er, transit is subsidized almost as much as railroad...so this doesn't help the case for railroad subsidies. Try again.

My point is that intercity highway travel is already highly subsidized. Since most highway travel is commuting, your comparison of all highway travel to intercity rail travel is bogus.

The bigger problem is that you're still stuck in Econ 1/2 mode. We're talking about infrastructure investment; existing subsidies are basically irrelevant. The question is, what is the most efficient use of a marginal subsidy dollar, and the answer ain't more highways.

Posted by theo | June 24, 2008 8:47 PM

Theo,

The bottom line is that the debate over subsidies to existing infrastructure is really beside the point -- the question is, what is the most efficient use for a marginal subsidy dollar.

No, they are both relevant questions: (1) What is the most efficient use of current funds used to subsidize existing infrastructure? And (2) What is the most efficient use of additional funds used to subsidize new infrastructure? If you think there's a good case that rail's hugely disproportionate share of subsidies is the "most efficient" use of that money, then make it.

It's pretty obvious to anyone who's ever ridden Amtrak that we're currently barely covering the fixed costs of running an intercity rail service. Acela proves the demand for rail is there. The variable cost of operation (i.e. the cost of running additional trains) is low.

You can create a demand for anything if you price it low enough. The "demand" for Acela is a consequence of its massively subsidized ticket prices, not any inherent advantages it may have over alternative modes of transportation. The point is that Amtrak's huge subsidies massively distort the supply and demand for its services, and so obscure the relative value of those services to consumers in comparison to road and air alternatives. If rail had to compete with air and road on a remotely level playing field its already-tiny market share would shrink to virtually nothing, because whatever benefits it may offer to consumers over those alternatives are not remotely justified by its costs.

Posted by Mixner | June 24, 2008 8:51 PM

Theo,

My point is that intercity highway travel is already highly subsidized.

And my point is that given the vastly higher subsidies for both intracity and intercity rail in comparison to highways, whatever subsidies intercity highways may receive, intercity rail receives vastly more. What is the justification for this vastly higher subsidy for rail?

We're talking about infrastructure investment; existing subsidies are basically irrelevant.

This is just utter nonsense. Existing subsidies are what keep Amtrak going. Billions of dollars in subsidies every year. You can't just wave your hand and declare the question of whether those subsidies are justified to be irrelevant. Of course it's relevant. Why should that money be spent to subsidized Amtrak rather than in some other way?

Posted by Mixner | June 24, 2008 9:01 PM

Brian S,

Mixner - if you think peak load is irrelevant then you understand nothing about transportation economics.

You seem to be confused. The assertion at issue is DTM's claim that "land costs tend to favor rail over new highways [because] the capacity per width of rail is significantly higher." That assertion is false, because the cost-benefit analysis depends primarily on how many people each mode of transportation would actually carry in practise (passenger volume), not how many they could theoretically carry if they always operated at full capacity. And volume is a matter of average load, not peak load.

Posted by Mixner | June 24, 2008 9:22 PM

Mixner, go live in Japan for a few years and then come back and pontificate, ok?

Trains are a damn sight more efficient at moving around people than cars are, provided you have enough density and enough network. Don't forget to add in the cost of building parking garages downtown and how much people have to pay, daily, for parking!

We're not pricing in all the externalities of car/roads with our "costs"--why aren't we throwing in a sizable amount of the cost of the Iraq war, hmmm? And the cost of building and maintaining the highways/roads? Stuff that into your calculations before you yowl again about "subsidies" for Amtrak. I bet that if some disinterested observer (which you have shown yourself NOT to be) were to carry out the calculations, travel by auto would prove to be a heck of a lot more expensive than you claim it is.

Posted by grumpy realist | June 24, 2008 9:58 PM

I still think there are a lot of issues making inter-city rail viable in the US outside of a few areas. Density is part of it, maybe.

Having a good public transportation in the city makes it easier to expand rail. You have more people not using a car, or have short range transportation. It should be easy to get to the rail station without using a car and preferably taxi so you can avoid parking charges. Being able to get around the destination city without a car matters some too. Another advantage should be the ability to buy tickets at the last minute. Your plans fall through for the weekend and want to visit your parents? No problem. It's harder with airlines. You need enough riders to keep the trains running on a pretty regular basis.

Highspeed lines are also pretty important too. Especially with todays technology we should be able to get lines running faster then Japan's shinkansen. Having trains significantly slower then cars just isn't acceptable.

Ft Worth's downtown is really nice. On the other side of the metroplex Dallas has a growing light rail system. They have plans to expand it to the airport and more major destinations in Dallas. Service is also planned to Denton, not a major distance, but a good start and with a lot of students it might be some good use. More people are already riding it to and from work. Around a few stations small urban areas have been popping up. Many stations just don't have the land area to do that which I find disappointing (poor planning?). The future expansions should offer more opportunity for transit oriented development and possibility of carless living in Dallas (hopefully expanding to Ft Worth and Denton).

Posted by Charles | June 24, 2008 10:02 PM

grumpy,

Trains are a damn sight more efficient at moving around people than cars are, provided you have enough density and enough network.

We don't have enough density. We don't have enough "network." We don't want those things. We don't want to live like the Japanese. Or even the French, for that matter. There are things Americans value besides "efficiency," you know, like big houses, and the convenience of cars. In any case, the efficiency benefits of rail are generally greatly exaggerated by its proponents, who typically ignore fundamental questions of demand and cost when comparing rail to other modes of transportation.

We're not pricing in all the externalities of car/roads with our "costs"--

Really? Then show me your calculations for the externality costs of road vs. rail that justify the hugely disproportionate subsidies rail receives. A serious, quantitative analysis. Not a handwave.

Posted by Mixner | June 24, 2008 10:16 PM

Glad that Amanda in San Jose picked up my point on the SF Bay Area rail. But I'm now an ex-Californian (Oregon, for better or worse). I did commute on CalTrain from Palo Alto/Mountain View to San Francisco for nine years, and was proud of the fact that I never took the Muni bus to work, walking the mile or two come rain or shine. Guess that makes me a conflicted commuter.

I remember a study, but not the details, that showed that the towns on the Peninsula that grew up around rail commuting had consistently higher property values than the later ones that grew after the switch to the automobile.

And I am old enough to have ridden the CalTrain extension to Monterey (actually it was the SP then) -- now that was a fun train! But not old enough to remember the SunTan special to Santa Cruz, which was shut down to avoid Japanese terrorist attacks in 1942.

Posted by Gene O'Grady | June 24, 2008 10:34 PM

"We don't have enough density. We don't have enough "network." We don't want those things. We don't want to live like the Japanese. Or even the French, for that matter. There are things Americans value besides "efficiency," you know, like big houses, and the convenience of cars."

Your fundamental error is confusing "we" with "me". You don't like density. Fine. Americans are starting to crave density. Remember the link I sent you last week? Right now you're just screaming at the wind. The winds of change are already in motion, and your low density point of view is the loser. It was a 50 year trend that is coming to an end. In the long run it was a blip on the radar, an anomaly, probably mostly spurred by racism. You can troll this site as long as you want, but your argument are fallacious because of your confusion over nouns. Are you on the payroll of realtors or highway construction? I'll bet you work for some road contractors don't you?

Posted by freddiemac | June 24, 2008 10:54 PM

Your fundamental error is confusing "we" with "me".

Yes, that's right Freddie. Americans would actually prefer to live like the Japanese. They don't really want all those big houses and cars they keep buying. They don't really prefer the suburbs to the inner city. They've forced to live that way against their will. Thank God we have people like you to refute these vicious lies.

Posted by Mixner | June 24, 2008 11:04 PM

Shorter Mixner: Real Americans want to live in the largest house they can afford, on the largest lot they can afford, at the greatest possible difference from the nearest poor and/or dark-skinned person. And drive the largest possible vehicle. Anyone who does not so choose must not be a real American, but a hippie with a trust fund. And probably gay.

Posted by roac | June 24, 2008 11:06 PM

The problem with blogs is that most of the users are folks like Freddiemac, who live a rather unique lifestyle that they extrapolate to Americans as a whole. Despite watching Seinfeld, Friends, and the like, most Americans do not live on a 4th floor apartment in a major city, probably don't want to, and can't afford it, especially given that the same folks seem to find a way to suppress continued development in cities, in order to "preserve their quality of life"

The US and Europe simply cannot be compared in this respect. Germany, where I have lived for most of the last 8 years, is a country of 80 million people inside a space one could drive across in a day. Intercity rail is no longer subsidized and has become extremely expensive. Real estate is extremely small and very expensive for the size provided. Most people rent at extremely high prices for tiny apartments.

Interestingly enough, Germany spent hundreds of millions building a new rail station and hub in Berlin, praised by Paul Krugman, despite the fact that Berlin has been rapidly shedding population and business for years. Maybe government doesn't get it right.

Is this "better"? I don't know--but I don't know what business people have imposing an answer on others.

The intercity rail question leaves out a key question. Although most airports are in the suburbs, in many metro areas business development occurs near the airport. Thus it makes little sense to take the train downtown only to have to embark on a 45 minute drive or mass transit trip out of town. This is what I think of when I read of a "rapid" line between Pittsburgh and Philly? Why???

Posted by GDub | June 24, 2008 11:31 PM

The problem with blogs is that most of the users are folks like Freddiemac, who live a rather unique lifestyle that they extrapolate to Americans as a whole. Despite watching Seinfeld, Friends, and the like, most Americans do not live on a 4th floor apartment in a major city, probably don't want to, and can't afford it, especially given that the same folks seem to find a way to suppress continued development in cities, in order to "preserve their quality of life"

The US and Europe simply cannot be compared in this respect. Germany, where I have lived for most of the last 8 years, is a country of 80 million people inside a space one could drive across in a day. Intercity rail is no longer subsidized and has become extremely expensive. Real estate is extremely small and very expensive for the size provided. Most people rent at extremely high prices for tiny apartments.

Interestingly enough, Germany spent hundreds of millions building a new rail station and hub in Berlin, praised by Paul Krugman, despite the fact that Berlin has been rapidly shedding population and business for years. Maybe government doesn't get it right.

Is this "better"? I don't know--but I don't know what business people have imposing an answer on others.

The intercity rail question leaves out a key question. Although most airports are in the suburbs, in many metro areas business development occurs near the airport. Thus it makes little sense to take the train downtown only to have to embark on a 45 minute drive or mass transit trip out of town. This is what I think of when I read of a "rapid" line between Pittsburgh and Philly? Why???

Posted by GDub | June 24, 2008 11:33 PM

Passenger rail has a formal priority in the US ... however, when running over single track without a large number of passing sidings, having formal priority does not always help if a passenger train misses its slot, since the rail operator might be able to a passenger train going the other way that needs the freight train to clear the route.

Duplication of rail and increased frequency of passing sidings would not only increase the reliability of interurban passenger rail service ... bringing rail up to the grade required for the 100mph tilt trains in normal rail corridors also brings it up to the grade required for 80mph express container freight.

The other side of the issue is electrification. With electrification of passenger rail, the train station to connect to the airport can be underneath the terminal, just as there are Sydney Cityrail station is directly underneath the Sydney International and Domestic terminals ... there, its the reliance on diesels that interferes with underground stations.

And, of course, most carbon neutral sources of energy are electricity generators, whether wind turbines, CSP solar, or nuclear power. And unlike the famous prize winning battery, electric rail is a mature, existing technology for turning electricity directly into transport of both people and freight.

Posted by BruceMcF | June 24, 2008 11:48 PM

Fundamentally, people who think its about "population density" are confusing inter-urban rail with mass transit and subway systems.

A "Express" class inter-urban train ... which is classed as High Speed Rail in the US, if not in Europe ... shrinks distance. Unlike a city bus, it does not get caught in car traffic, and does not bog down as car travel speeds bog down in the endless cycle of congestion leading to roadwork leading to development at the outer end of the roadwork leading to congestion.

Now, there is a section of the country where the density is not "high enough", in the sense of not enough small and large cities close enough together to support a healthy mix of one to three hour inter-urban trips.

At the "moderately high" 110mph speeds, its "only" all the US east of the Mississippi, the next tier of population centers as far west as Dallas, and then Southern California, Northern California, and the Pacific Northwest that have density "high enough" for interurban rail. Indeed, the travel time between the Northern California cluster and the Southern California cluster is one reason why there is so much impetus in California to leapfrog the "Express" HSR level to true, fully grade separated 220mph High Speed Rail ... despite the substantial increase in cost per mile.

On the other hand, that is the bulk of the population of the United States that lives in regions that have the population density required to support regular inter-urban rail transport.

The population density question is about the local transport choice ... at the number of trips per route that would be seen in most US cities, heavy mass transit and subways are not justified .... most urban transit projects over the next decade are going to involve light rail of one sort or another or bus priority measures to reduce delays on bus routes to increase the effective capacity of a given bus fleet.

However, we would already be investing in substantial numbers of new inter-urban rail systems today if they were competing head to head on a full cost-benefit basis against road transport.

Posted by BruceMcF | June 25, 2008 12:11 AM