I’m not the one claiming the market will respond to an externality problem. I think it’s you who’s looking like an idiot.
It makes no sense to use HSR for freight - existing trains do just fine at that for less cost. Using HSR for evacuations only works when the crisis is such that your rail links don’t go down. And having one high-speed line out of Dodge is not nearly as useful as having dozens or hundreds of roads allowing egress in many directions.
I can just see it now: “Tornadoes are coming! Quick, everyone in the city head for the rail terminal!” That’ll work.
Reducing volume of other transport systems is a positive externality? How so? Airlines also need high duty cycles to maintain profitability. Reducing the use of an existing transportation infrastructure is not a positive externality. I think you’re over-using the term.
The point is that an under-utilized road is not the same as an under-utilized rail line, because the trains still have to run. An under-utilized road just sits there. It has some maintenance costs, but they go down because a lot of road damage comes from the vehicles, and because under-utilized roads tend to be less well maintained.
But a rail line doesn’t move anything at all unless a train is running on it. Trains are very expensive to run. If you have a rail line that’s under-utilized, you start losing money fast.
You can know by looking at the number of them currently running in the midwest (i.e. none). And “Hey, let’s build a train and see what happens!” is a bad way to plan to spend billions of dollars.
There aren’t that many big unknowns here. Lots of HSR proposals have come and gone. Lots of governments have been highly interested in building them - they’re exactly the kind of big legacy projects politicians love to sponsor. But they are extremely difficult to make profitable in North America, because A) they have to compete against a very good existing road and air network, and B) the population is too spread out and rely too much on other modes of transportation at each end of a rail link.
So do hybrid-electric cars. I’m a big fan of decoupling energy production from energy consumption by using an abstraction layer like electricity or hydrogen or some other intermediate. But there are a lot of ways to do this that don’t involve trains. And you keep missing the point that trains will never have but a tiny effect on the energy infrastructure. A wildly optimistic estimate for the number of HSR links we could plausibly run in North America might make a difference of a percentage point or two in our energy consumption patterns, and that’s about it. A difference that amounts to a few months of economic growth and population growth. Almost a rounding error.
You do not want to compare the flexibility of cars vs trains. The average lifespan of the auto fleet is about 10 years. If a new energy source came along, you could turn over half the country’s transportation in that time. What if the economy or energy conditions change in a way that’s not advantageous to trains? How long to change you HSR infrastructure? What if you build that nice link out to Vegas, and economic conditions cause Vegas to go into a decline like Detroit did?
And you’re missing a big way in which autos can be flexible - if energy prices rise, auto use can decline fairly linearly. Have a look at this chart of miles driven vs gasoline prices. When gas prices go up, people drive less. Since last November, drivers in the U.S. have driven 53.2 billion miles less than they did over the same period a year ago.
But you can’t do that with trains. The train either runs or it doesn’t. If high costs cause people to not ride the train, all that does is drive up the marginal cost for the rest of them.
I don’t think HSR is an answer to either one, for reasons I’ve already given. First, the overall effect of them is small. Second, given the amount of energy and labor required to build out an HSR link, we probably wouldn’t even get energy payback on the ones we build for decades. Third, the energy cost per passenger mile is only advantageous if your train has a high fill rate for passengers. Run it 50% empty, and it uses more energy than cars to get those passengers to their destination.
This is why I asked what problem you’re trying to solve. Let me answer it: HSR, where it makes sense, would be valuable because it widens bottlenecks, makes travel easier and safer, and is a big convenience over air travel for shorter distances. It’s not a replacement for cars. Where it might make sense is to move traffic from the air to the ground on heavily traveled routes between major centers on the east coast, or between to large, economically entwined cities elsewhere. If the airports are overcrowded, flight delays are growing, the freeways are jammed, and there is high demand for additional capacity, an HSR link might provide it. Note that the end result could be even more travel and even more energy being used.
HSR’s ‘niche’ is in the range where travel by car is painful but travel by air has too much overhead. Distances on the order of 200-300 miles, probably.
If you’re thinking, “oh, look at those polluting SUVs driving around the midwest - we should push those people all into clean, electric trains to save the planet”, then you’re barking up the wrong tree. HSR is not even remotely a solution to that problem.
But the road system is very flexible. Rail isn’t. A mesh of roads allows traffic patterns to change dramatically. A single rail link does not.
Again, transportation makes up only 25% of our fossil fuel use. And if we stopped using fossil fuels for transportation, we’d use more of it elsewhere because the price would drop when the demand dropped. I hope you understand that there are no conservation methods that will result in less oil being burned. Conserve in America, and you lower the price for Chinese factories. America’s energy costs go up, China’s go down, and the same amount of oil gets burned and the bad people in the Middle East still get the money. Only now it’s Chinese money, which means China gets to influence Middle East politics instead of the U.S.
(cont’d)
The externalities of roads and air are greatly over-estimated by the environmental lobby. People pay road taxes. They pay gasoline taxes to pay for road maintenance. The externalities due to CO2 might amount to 50 cents a gallon, maybe. Gas has already gone up four times that amount, and people still drive. The fact is, cars on paved roads are a pretty efficient way to move the population around.
Why not? Are you claiming that people who have studied the feasibility of HSR links are incapable of figuring out how much it costs? Or that there is some massive hidden cost in road use that’s impossible to determine?
None of this stuff is mysterious.
A lot of new cars being made today are low-emission vehicles. Quite a few PZEV (Partial Zero-Emissions Vehicle) models are available, which have miniscule pollutant outputs. A lot of the pollution emitted from cars today comes from the old beaters that are still on the road. In any event, we certainly have the technology to make cars very, very clean, and a lot of of the new ones are (every Ford made today produces less than half the pollutants than what the law mandates, and the PZEV models are 10% of that.)
You lost me. Of course GDP deals with externalities. An externality is a cost imposed on someone other than the people involved in a transaction. But if it’s a real cost, it shows up in GDP numbers. Again, I think you’re using the word externality in a way that is incorrect.
The impact of CO2 emissions is a complex topic. Suffice it to say that estimates of the damage done by CO2 vary dramatically, and the IPCC’s best estimate model shows no real economic damage at all to the world as a whole. Certainly global warming will damage equatorial regions and certain other areas that are particularly susceptible to that kind of change, but it will also bring benefits - lower heating costs in the populous northern regions, longer growing seasons, etc. The current models show that if the overall warming is greater than 2.5 degree C, there will be a net economic loss to world GDP. For warming below that number, there may be a net economic benefit. We don’t know enough yet to know exactly how much warming there will be.
In addition, if there is net economic damage, it won’t be for a long time. You need to apply a proper discount rate to figure out the net present value of the damage, and when you do you find that even large numbers for the amount of damage 100 years from now translate into surprisingly low cost today.
There doesn’t seem to be one for HSR, either. But that’s not stopping you from lobbying for it.
Most of the time, congestion pricing is seen by municipalities as a way to raise extra revenue, which makes it a tax hike. Also, it’s difficult to apply it to existing commuters, because it punishes some people over others. So it’s a tough problem, no doubt. One answer is to make sure it’s revenue neutral. Another might be to find a way to give the money back to the people who paid it, but in a way that doesn’t reward travel. For example, if congestion charges cause a bedroom community to see their travel costs skyrocket, the money could be given back to the community in the form of property tax cuts. You maintain the incentive to use the roads more efficiently, while keeping the money in the community.
Really? Congestion pricing is a solution to congestion. So is HSR and more freeways. If you can solve your congestion problem by making your current infrastructure more efficient, why wouldn’t you do that instead of building more infrastructure?
So implement congestion pricing, and give the money back to the poor as tax credits. They still have an incentive to optimize their use of the public space, and it costs them nothing. Hell, if they start taking public transit they might even make more money than they get now. And if congestion pricing causes the price of car travel to grow vastly more expensive, that will be the incentive business needs to invest in mass transit.
Maybe the interstate highway system shouldn’t have been built. I don’t know. I’m leery of any massive government intrusions in the economy. Maybe if they hadn’t been built, it would have stimulated production of more efficient modes of travel. But the fact is, the interstate system was built. It exists. Given that, it seems to make more sense to try to make it more efficient before trying to build an alternative infrastructure.
Except that I would say that you can tell where it doesn’t make sense - in those places where it wasn’t built. And before you say that there are other obstacles towards construction of HSR, you might like to know that Canada had a trans-Canada HSR link all the way back in the 60’s. We had a train that could do 200 kmh (in practice, it ran at 160). It failed because it operated at huge losses, and no longer exists.
Now, I’m all in favor of leveling the playing field - one way to pay for the externalities of auto use would be to levy an additional gasoline tax and use the money to fund mass transit, I suppose. But rather than government mandating the construction of such projects, I’d simply make the money available as tax breaks for companies willing to invest. If you still can’t get investment, well… That’s a pretty good sign that it doesn’t make sense. What I would demand is that general tax revenues not be used to prop up such a system. If you can’t make a go of it with subsidies derived from honestly-calculated carbon taxes, you’re out of luck.
Airports most certainly can be profitable. Some of them are very profitable. Auckland International in New Zealand makes an annual after-tax profit over $100 million dollars. Calgary International Airport pays over $300 million dollars in taxes TO the government. It receives zero government dollars for upgrades and renovations. Some airports are run as profitable enterprises, some are run by not-for-profit corporations, and some are subsidized by muncipalities because the amount of traffic they bring in doesn’t allow them to be profitable but they are still necessary.
The same is true for roads. There are tons of profitable toll roads scattered throughout the world. Other roads are paid for through gasoline taxes or property taxes.
If you can make an HSR link profitable, go for it. Hell, if you can even build one that pays for itself through mandatory ticket taxes, go for it.
Are you of the opinion that they haven’t been studied for HSR and other modes of transportation? These studies are being done all the time, all around the country. We’ve had numerous ones here in Alberta. Ultimately, people can’t make the math work out. The Federal Railroad Administration has done a number of feasibility studies for different HSR links. The most feasible one they could come up with was the Washington-Boston-New York line, and it could only recoup 55% of its costs through revenue generation, requiring 45% of its cost to be subsidized.
Here’s an abstract for a paper that examined the feasibility of an HSR link between Los Angeles and San Franscisco:
So even accounting for all externalities of all modes of transport, HSR still lost out to air travel and cars.
To be fair, other HSR feasibility studies have had better results, with some showing a potential profit. Like I said, I’ve got no problem with it where it’s feasible. The U.S. Department of Transportation has done a major study of HSR feasibility, and they’ve publshed a map of potentially profitable links - all of them are tween the major populous cities along the U.S. coastlines. None go across the midwest.
One thing to be aware of - HSR is heavily pushed by a lot of vested interests. For example, it’s hard to find an HSR study in Canada that did not have major input from Bombardier - a huge conglomerate that stands to make billions from such projects. Often these big corporate interests will get in bed with government and produce wildly optimistic studies that are intended to win contracts. So it’s hard to sift through the chaff.
Well, I’m sure you do. Anyone else who is actually following along however can see that you are making up strawmen like we are going over the rainbow soon. At any rate, I’m pretty much done with you. I give you responses and you wave your hands in the air and claim I simply don’t understand how economics works. Why exactly should I take you seriously? Frankly, I think Sam is wasting his time on you, but I’ll leave it in his more than capable hands.
Ado.
-XT
No, anyone who is following along can see that you refused to read the wiki article on externalities, so you could understand how and why markets routinely fail to deal with pollution (and an ideal market should fail to deal with pollution entirely). Nobody should take you seriously because you have demonstrated very clearly in this thread that you are unwilling to learn about how markets work.
The reason I’m not dismissing Sam the way I’m doing to you is because it’s clear that he has some background in economics, even if I disagree with his analysis.
Is this true? Is there a cite that confirms this prediction? I don’t know what the projected passenger capacity on a HSR is but I’m assuming it’s well into the hundreds, probably larger than your average jumbo jet. 50% of that is still probably going to be 100 passengers and I find it difficult to believe that a HSR is less efficient than 100 automobiles (or 50 automobiles if you assume double occupancy).
First off, thanks for your very thorough and informative reply.
I understand the geographic need to use the Central Valley route for the HSR lines, I didn’t appreciate how big of an impediment the turns and hills of a more direct coastal route would present.
I see that I under-estimated the population density in and around the Fresno and Bakersfield areas and understand the desire to serve those communities.
However, I’m still a little skeptical of the travel times being estimated along those routes. The Fresno to Los Angeles route is projected to take 84 minutes according to the website you linked to, however that route has 5 stops on it. If we assume that a stop to negotiate the station and unload/load passengers takes 10 minutes that 50 minutes worth of time in the stations leaving only 34 minutes to travel the 255 mile distance. Even if the 10 minute estimate is high those stops still seem to dramatically inflate the actual travel times.
You describe express bypasses through stations but the stops themselves in each of those locations are not even the most pressing issue. Even if those stops I mentioned above were all bypassed via a Fresno-LA express train they still would require the train to negotiate an urban area and pass through a lot of crossings and in close proximity to populations and pedestrians, not to mention entering and leaving the station itself. Those issues would obviously require the train to travel at well below its maximum speed. Each city center and station it passes through would greatly increase the overall time of the journey whether it stops or not.
I just find the travel times being quoted to be completely unbelievable, it seems they are predicting the train to travel at 200mph consistently across the entirety of the route regardless of the population it passes close to, even through the stations.
All this raises the question of what the real goal would be. If the goal is to offer the most time and cost effective transit between SF and LA to reduce reliance on air travel then the route along I-5 with a minimum of stops seems like a necessity. If the vast majority of passengers are expected to be LA-SF travelers opting away from air travel then every single stop along the way will drive away passengers. In order to make those stops worthwhile more people have to get on at each station than are driven off but the extended trip duration. While the Central Valley population is large, is there a great need for them to travel to LA and SF? How many people going from LA to SF will take a train that is direct along I-5 taking a legit 2hr 30min and how many less will take a LA-SF train that makes 10 stops and takes closer to 3hr 30min? Is the difference more than those boarding at those 10 stops?
All in all, the real issue I envision for this proposed CA-HSR is the utter disaster that local public transportation is throughout much of the urban centers. SF is so-so and is dense enough to be a good walking city but the rest of the Bay Area is basically unnavigable without a car. LA is a joke. SD does a pretty solid job. What good is it to take the HSR from SD or SF to LA’s city center if you are stranded there is a city that’s impossible to navigate without a car.
An interesting read is the Transportation Energy Data Book put out by the DOE. Of interest to this discussion are tables 2.13 and 2.14, which show the energy intensity in BTU per passenger mile of the various current transportation options. Cars today average about 3,500 BTU per passenger mile. Rail travel comes in at around 2,700. So already you can see that it’s not *that efficient. Per vehicle mile, the car is at around 5,500. This means the average car trip has less than 2 passengers in it. If you’re driving a car with 2 or more people in it, you’re already more efficient than rail. Or, if you’re driving a car that gets better than average gas mileage, you’re doing just slightly worse than rail if you’re driving by yourself, and much better if you have a passenger. Rail currently gets around 40-50 MPG per passenger. So if you’re driving a Prius, you’re already more efficient than a train, no matter how many people are in it.
The San Jose light rail system currently gets worse mileage per passenger than does a car with a single person in it - it uses about 7,000 BTU per passenger mile. How come? Low utilization. As I said, one problem with trains is that they burn about the same energy empty as they do full of passengers, and scheduled trains have to run whether they are full or have five people in them. If you build a train and it’s not used very much, you’re doing more damage to the environment than if you hadn’t built it at all. Not only did you expend a whole lot of energy building the infrastructure, but a half-utilized train will burn more energy moving those people across the country than would have been burned if you’d just let them drive their cars. A lose-lose proposition for everyone. This is why even environmentalists should be very careful about new rail lines - build it in a poor place, and you’re spending billions of dollars to damage the planet.
Now, High-speed rail is slightly better than average trains, because the trains are more efficient. I’ve seen numbers as low as 1200 BTU/passenger mile for typical load factors. I’d guess that this low number is partly greater efficiency, and partly because HSR is expensive and so only built on the most densely-packed corridors, assuring a higher utilization rate.
Pro-rail activists quote numbers as high as 300-500 mpg per passenger for high-speed rail, but I’ve never seen a number anywhere near that in serious published literature. I imagine they are taking optimal conditions and assuming a 100% fill rate for passengers to get to that number. There’s no doubt that you can come up with some amazing efficiency numbers for a train if you pack it full of passengers. But in the real world, trains rarely run at anywhere near full capacity. Even if trains were more popular, you can’t run them full, because you have to size your route for peak loads, and trains still run off-peak. Hell, even airlines can’t run completely full, and they go to great pains to fill every seat they can.
The fact is, the actual, measured numbers for the current rail system show that it’s somewhat more efficient than cars, but not hugely so. In fact, now that gas prices are higher and the market pushes us into more fuel efficient vehicles and into more shared trips, it may be that the auto fleet as a whole in 5-10 years will be more efficient at moving people around than will current rail systems.
If you took BrightNShiny’s advice and built an HSR link across the midwest, just to see how well you could do, you could easily wind up with a hundred billion dollar boondoggle that runs 80% empty and results in higher overall energy use for all the travel on the route.
If you want to save the environment, you need to think like an engineer. You need to be hard-headed about the tradeoffs, and take into account things like where your best bang for the buck is. Rail is a very difficult solution to the problem. It would have a very small impact, would cost a fortune, and if done wrong could be worse than useless. And because it’s inflexible, you could find that even if you size it right and it’s efficient today, in 20 years changing economic conditions could make the rail system as a whole an energy loser over autos simply because passenger flows have changed and trains are running at a lower passenger fill rate.
An all-electric HSR link on a currently congested route might make sense. If you can keep it reasonably full, you could probably save a significant amount of energy - especially if traffic on that corridor is often gridlocked and therefore very expensive energy/wise. But even there, the prime benefit to HSR would be to alleviate congestion on the roads, not to save the planet.
As an anecdote, last year my wife and daughter took a passenger rail trip from Vancouver to Edmonton. I met them at the rail station and watched as the train came in. The first passenger car was completely empty - shut down and dark. And the next. And the next. And the next. I can’t remember exactly how many passenger cars were empty but it was quite a few. How many had passengers? The last two or three. I’d guess that train was running at maybe 10-20% capacity. That gigantic huge train went from Vancouver and back and carried maybe 50-100 people. I’ll bet it would have been more energy efficient to drive there by SUV.
[Firstly, my apologize to Sam Stone for not yet responding to his challenge back in post #91. This has been an unexpectedly busy last couple of weeks for me, and I hope that Sam takes it as a compliment that I’m not prepared to toss an easy setup for him to hit out of the ballpark. I’ll try to finish up and post my still-building "case for California-HSR " as soon as I’m able to.]
An additional stop on a HSR should take 2-3 minutes (more if it’s an airport stop due to passengers with heavy luggage), including deceleration and acceleration.
Look at the current NorthEast Corridor schedule (PDF). The Acela Express trains, the closest thing the US has to true HSR at the moment, have run numbers beginning with ‘X’ and shaded grey; on page 4, we can see that X2109 leaves New York at 08:00, and leaves Philadelphia at 09:13, with stops at Newark and Metropark NJ. An hour later, X2151 leaves New York at 09:00 and leaves Philadelphia at 10:10. Same day of the week, similar time of day, but 3 minutes less because it didn’t stop at Metropark. The pattern is repeated for other trains in the schedule, although one should be careful if comparing weekday times with weekend times. An HSR using European or Japanese off-the-shelf trains would have better acceleration / deceleration profiles than Amtrak’s Acela, the latter being a bit like an SUV body on a Lamborghini chassis.
The expected 84 minutes between Fresno and LA would be for the fastest scheduled train, which might only stop at Bakersfield and one suburban LA station. All 5 stops would probably be 93 minutes.
It’s a given for HSR lines worldwide that there are no grade crossings and the track is fenced off, so no pedestrians. This is the main reason that HSR travel has proven to be so safe; note that when the French TGVs run on the regular SNCF lines (lignes classiques) they are as liable to hit something at a grade crossing as any normal train. California HSR is planned to be grade-separated throughout its entire length (even close-in to SF and LA where speeds will be ~100mph max), but will definitely be grade-separated wherever new high-speed track is built, which is the great majority of the route length.
The Central Valley cities such as Fresno and Bakersfield don’t have dense cores (the high populations are due to sprawl), and the current rail right-of-way cuts a pretty straight, wide path through them. It’s pancake-flat there, and the railroad was in place before the towns grew. If the stations are built with outer platforms and inner express tracks, the non-stopping trains will go through at ~200mph, just like elsewhere in the world.
They quote 2h38min for the 432 track miles between SF and LA, which comes to an average of 164 mph. That would be 100mph running near SF and LA, and up to 218mph elsewhere.
If you’re going all the way from SF to LA at any of the peak times, you won’t take a train that stops ten times. If you look again at the Amtrak NEC schedule I linked to above, you’ll notice that at almost any time that trains are running, there’s a mix of Acela, limited-stop, and stopping trains. CA-HSR would be similar, except that there would also be non-stop (or one-stop) SF-LA service (which is difficult to do in the NEC for political reasons; every intermediate state wants an Acela stop, and Joe Biden’s importance to Amtrak guarantees that no train goes through Wilmington DE without stopping).
The population of California is predicted to grow from 36 million to 50 million by 2030, and most of that growth is expected to take place in the Central Valley where land is cheaper than the SF Bay Area or the LA Basin. However, the Central Valley towns only have low-frequency, high-cost air service to SF and LA, and this is not likely to improve much because airlines want to get out of the short-haul business wherever possible. Amtrak rail service there is slow (although quite popular), but doesn’t run from the Valley to the LA basin. Fresno and Bakersfield stand to gain a huge amount from HSR, and IMHO it would be foolish to overlook them (and their votes…). Running the HSR through them really shouldn’t add more than a few minutes to a nonstop SF-LA run, and they add a great deal of potential passengers to the mix.
I’d say that the SF Bay Area is much better than your impression of it – BART and Caltrain do their job pretty effectively, although everyone still loves to moan. The main problem with Bay Area public transit is its multitude of jurisdictions; if one knows the system it’s actually easy to get around without a car. I would expect high-priofile HSR stations to attract local transit service like moths to a flame.
A certain amount of the HSR money (S1billion I think – I’ll check on that) will be set aside for local transit feeder improvements. So, Caltrain and the relevant LA commuter rail tracks will be upgraded.
Of course, the “needing a car at the other end” is true for air travel, as well. I’d expect good car-rental business at the HSR stations in the LA region. However, a high percentage of business travelers could get away with taxicab rides from HSR to their destinations.
Heavy freight trains and HSR tracks don’t go well together. HSR trains can go up much steeper grades than freight, so one tends to build HSR tracks straight and over shallow hills, rather than flat and twisting around them which is how most freight railroads are built. Also, HSR trains at >200mph don’t like track that’s been chewed up by heavy freight cars that may have uneven wear. There are also FRA weight regulations for passenger trains that run on the same tracks as freight trains, which is why Amtrak’s Acela is about twice the weight of equivalent European trains (a restriction that has caused Acela no end of problems). CA-HSR would use off-the-shelf European or Japanese trains, rather than building an expensive chimera like the Acela.
Please note this is a thread that started in August 2008.