PNAS Plus has published my co-authored new paper. Â For an excellent overview of our main ideas read Eric Jaffe’s Atlantic Cities piece about my new China bullet train paper.
14 thoughts on “China’s Bullet Trains”
Comments are closed.
Everyone is entitled to his own opinion, but not his own facts. Founded by Mark Kleiman (1951-2019)
PNAS Plus has published my co-authored new paper. Â For an excellent overview of our main ideas read Eric Jaffe’s Atlantic Cities piece about my new China bullet train paper.
Comments are closed.
The end of that spectrum will very likely change in a few years when the taxpayers are on the hook for the money spent on a system that’s never finished. Or, if it is ever finished, a state that is already having severe financial problems has to maintain a system that has a ridership far below the optimistic projections.
If self driving vehicles become a reality, that niche distance can only get shorter.
Ah, the traditional strength of a command economy: The ability to do one or two great things, by throwing resources at them regardless of economic sanity. Which is coupled to the traditional weakness of a command economy: The resources thrown at those great things aren’t available for not so “great”, but much more necessary, things.
Brett,
Did you read Matthew’s link? He was arguing that bullet trains are economically sane: a high-valued use of resources. His argument is not obviously self-refuting, and you did nothing to refute it. (What’s more-even you can’t call Matthew some kind of socialist.) You’re a smart guy. Could you produce some arguments? Or do you just take it as a matter of faith that there is no such thing as an externality, either positive or negative?
Yes, I did follow the link. It notes that building bullet trains in China is rather cheaper than in the US for a variety of reasons, (Little in the way of enforcible property rights, few regulatory hurdles or safety requirements, and, yes, command economies not looking at projects’
economics with a gimlet eye.) and then goes on to mention the benefits without a close look at the costs.
No question, if a proposed bullet train route happens to duplicate your commute, or is built with a stop in your town, and you can get somebody else to foot the bill, it’s going to benefit you even if it’s a net loss to the economy. A lot of things carry benefits to identifiable people even though they’re net losses to the economy, that’s why the government has grown so bloated.
I’m not buying that. The major high speed train operators in democratic countries that are using their own technologies (France, Germany, and Japan) are quite profitable on their own, regardless of any indirect benefit for their respective economies. In particular, SCNF and Deutsche Bahn had operating profits of around 1.1 billion and 1.9 billion Euro, respectively; it’s a bit harder to do a breakdown for all of the Japanese railway companies, but they all appear to be profitable, too. And while I am not familiar with Japanese law, both French and German law (via their constitutions) have eminent domain requirements (for the public good, just compensation) similar to the Takings Clause (which has created the occasional problem, e.g. the Stuttgart 21 kerfuffle). Safety requirements similarly are not being given a pass.
That is not to say that the respective governments do not impose regulatory hurdles (such as the requirement to cover routes that would be a net loss for the operators). But it doesn’t require a command economy to run a profitable high-speed train.
Now, that does not say anything about whether high-speed trains could become a self-sustaining part of US public transport — it is quite possible to bungle such things, and there are many things that are quite different in the US compared to the aforementioned countries — but it is demonstrably not the case that you require a command economy to have profitable, working high-speed trains.
Things are, in fact, quite different in the US: We have a considerably lower population density, for one thing, and less concentrated city centers. And most of the time, Kelo not withstanding, more respect for the property rights of individuals.
There might very well be a few places in the US where bullet trains would make sense, setting aside the ruinous cost of assembling the land. I wouldn’t count on a government built on kickbacks to identify them, and more importantly, identify places that aren’t them.
. . . and socialism!
(Brett, the US does not have a substantially lower population density than most of Europe, once you back off the wasteland and Buffalo commons. Generic US suburbs are damned dense. Most of the population of the flyover states lives in cities and suburbs, just like everybody else.)
As Ebenezer pointed out, the lower population density is a bit misleading. Nobody is really talking about high-speed trains connecting, say, New York and Los Angeles.
But it is worth remembering that the Acela Express is in fact highly profitable, operating in a densely populated region of the country, and capturing 54% of the combined air/train market between New York and Boston.
It is also worth noting that more and more people have been starting to drive during holidays instead of flying (up to a certain distance). Trains are used for largely the same reasons (more comfort) and high-speed trains increase the distance up to which that is viable.
Generally, while a train trip will take longer than a flight, counting only actual travel time, this is not all that matters. Airplanes have cumbersome check-in and security procedures that trains can dispense with, and the trip itself is generally far more comfortable. For example, the distance between Hamburg and Munich is almost 400 miles. Pure travel time will be about 80 minutes (give or take) by plane and some 6 hours by train (because not all of the route is high-speed rail). Prices are roughly comparable. This sounds like a clear win for flying, but in practice there are a number of countervailing considerations:
* Airports are outside the cities, so you have to generally add more time to travel to/from the airport.
* You have to be at the airport much earlier to check in, and getting off the plane and out of the airport also takes considerably longer.
* Traveling by plane, you have to go through a cumbersome check-in and security process.
* You have considerably more leg room on a train, not to mention table space (either stationary or fold-out) that’s actually useful.
* Modern high-speed trains have WiFi (at reasonable speeds) and electric outlets if you are a business traveler using a laptop. You will also have constant access to your mobile phone or mobile broadband if desired. You can actually work fairly productively for the duration (been there, done that).
* You can bring your own food and drinks on board without having to get them through security or having to buy them at usurious prices at the airport; if you are willing to overpay, you can have a nice lunch or dinner at the train’s restaurant.
In short, you’re trading 2-3 hours of concentrated stress for six hours of relaxed travel. This can be beneficial both for families (especially with children) and business travelers. And, of course, that’s going 400 miles, or close to the maximum where it’s viable. For shorter trips, trains become ever more attractive.
This trade-off is not fundamentally different in America. There may be fewer opportunities (whether per capita or per square mile), because a lot of trips are too long for such a trade-off to be worthwhile, but that’s not the same as saying that they don’t exist. The average domestic flight in 2012 had a length just shy of 1000 miles, so it is not unreasonable to expect that there is a sufficiently large subset of trips for which high-speed train travel would be competitive.
As to respect for property rights, I’d be careful to assume that US law is more protective. After all, a German state government was only recently toppled over the Stuttgart 21 project, which involved eminent domain issues (though the project was thereafter confirmed by referendum); the original ICE tracks in Germany could only be laid after 10,700 administrative appeals and 360 lawsuits had been dealt with. As I mentioned above, both the French and the German constitution have essentially the same protections as the Takings Clause (public use, just compensation), and at least in the case of Germany, with a less elastic interpretation of “public use”; in particular, under German law, a taking must be necessary for the public interest or the public welfare, not just beneficial, and must be the least intrusive means to accomplish the underlying goal [1].
This is not to say that high-speed trains would be an unalloyed good or that state governments would not screw up the implementation (there’s plenty of justified criticism for the California high-speed rail project, for example). I am just saying that principal objections to high-speed rail on the grounds of “America is too different” need to be substantiated in more depth.
[1] This is a specific application of the proportionality principle.
Self driving vehicles will have many of the positives you list for high speed rail without a number of the negatives such as inflexible routes and schedules.
It would be unfortunate to spend billions on trains only to have them undercut by auto-driving vehicles.
Yes, self-driving vehicles will be an interesting alternative — at some point in the future.
Right now, they’re still in the proof of concept stage, and even thereafter they will not necessarily be affordable. In short, driverless cars are still pie-in-the-sky technology (though exciting, at least for a geek such as myself), while high-speed trains have already had their bugs worked out over the past decades.
Long-term, there’s the question of the environment (trains have a considerably lower carbon footprint per person mile) and congestion (especially in densely populated areas, especially where high-speed trains would be attractive). There’s also the question of cost: gas prices are likely to go up over the coming decades, not down.
Even if/when driverless cars become a viable alternative, there will be pros and cons to both cars and trains.
BMW is predicting cars will be driverless by 2025 assuming there’s no unforeseen difficult or insurmountable obstacles. Once the technology has matured, it may be a relatively small part of a vehicle’s overall cost.
If gas prices continue to go up, it will be due to taxes, regulations and other causes, not the supply of oil.
Charles, I see BMW’s statement (or Volvo’s similar statement) as little more than ad copy right now. The industry has known for almost decades how to make an almost completely self-driving car work; what has been eluding them is to make that 100% instead of “almost”. And I’m skeptical of anybody in R&D who offers a concrete time table for an open research question. While I expect that we will see more technologies to assist driving (such as Lane Assist) in the coming years, and that they will hopefully have very beneficial effects on road safety, there is still a fundamental difference between still having to be at the wheel, even in a very limited role, and actually being able to read a book or write on a laptop while driving.
As for the cost, right now Google’s self-driving cars require $100,000-$150,000 in additional technology. Obviously, mass production will bring that price down, but that may not make it sufficiently low for the mass market (especially the $15,000-$25,000 price range for small cars).
Overall, even so, self-driving cars will narrow the comfort gap, they won’t eliminate it (you still have more room on a train, you can stand up and stretch your legs without having to stop, you can go to the bathroom anytime you need).
Even disregarding comfort, there are tradeoffs between high-speed trains and cars: Cars give you greater flexibility, more privacy, and the ability to take more baggage with you. Trains, on the other hand, are faster (you won’t be driving 150+ mph on any Interstate); you won’t be able to take your car with you, but you also don’t have to take your car with you if that’s inconvenient (congested cities, combining train and flight trips, such as a Boston-NY-Orlando round trip).
Mind you, pragmatically, I don’t expect high speed trains to take off in America. What I’m arguing is that they would be a viable alternative, not that in practice we’ll actually go that way. There are significant practical obstacles: for example, one practical difference in Europe and Japan is that high-speed trains are generally much better integrated with the rest of the public transport system. The upfront investment needed also creates obstacles (both political and economical) that are difficult to overcome.
Regarding cost of gas: As I recall it, the world energy outlook report by the IEA expects crude oil prices to slowly go up over the next two decades, which would also increase average gas prices (unconventional oil, of course, is already more expensive). The one scenario where they see a potential reduction in crude oil prices is a lowered demand due to taxation and such (which would also drive up gas prices at the consumer level).
Am I alone in my distaste for the gee-whizz metaphor of “bullet train”? Handgun bullets travel at over 600 mph, two or three times the velocity of actual trains; rifle ones at five times that. A more important point is that almost all operating high-speed trains use a conservative, incremental technology, with steel wheels running on steel rails at standard gauge. Maglevs and monorails are restricted to toy showoff airport links.
What you say is true, although high-speed trains have a number of neat technological gimmicks, such as cars that lean into the turns. But it’s irrelevant. Just ride the Shinkansen. You’ll have a goofy grin on your face and no SUPERTRAIN metaphor will ever again seem strained.