In my last piece, in which I discussed the relative costs of riding world high-speed rail systems, I don’t believe I adequately expressed the need to provide lower fares. In the process, I confused my argument. By directly opposing rail and air prices, I implied that the main point of comparison was between those two modes. In reality, and as was suggested by several individuals in the comments section, the major issue here is attracting people away from cars and buses.
That’s because true high-speed rail, offering services at fast average speeds, will demolish any airline competition in virtually every corridor of less than 600 miles where new trunk lines are built — as long as tickets are somewhere near air prices. Trains are simply more convenient, more comfortable, and more reliable than airplanes; that’s been proven time and time again in Europe and explains why Amtrak has its larger-than-airline market share between New York and Washington even though its tickets can often be as expensive or even more expensive than those on airplanes. Success in dramatically reducing air traffic along the Los Angeles-San Francisco corridor is assured if California High-Speed Rail is built as planned — even if tickets prices are at 100% of air fares.
Rather, the aim of reducing ticket prices should be to attract people away from buses and cars that clog the roadways and that are not particularly effective in providing travel between cities more than 100 miles apart.
There are two ways to encourage people currently relying on road-based transportation to travel by trains: one, lower ticket prices; two, increase speeds. Both actions would provide a substantial motivation for highway users to reconsider their options. The first would put train travel back into the sphere of the economical. Plenty of bus companies market service at less than $20 between New York and Washington. At $2/gallon, a 26 mpg car could be driven between the cities for less than $20 in gas. That number doesn’t account for maintenance and ownership costs, but drivers rarely consider those factors when making decisions about how to get from one place to the next.
By increasing speeds, train travel’s time benefit multiplies significantly. While buses get into traffic, they can still make the trip from the capital to Gotham in five hours — versus the three hours required by rail. This is an advantage for train users, but increasing speeds to allow for a 1h40 trip would make it nearly impossible to justify riding the bus or driving, even at a lower cost.
There is no magic fare that will increase American rail ridership while maintaining financial stability in the short-term, but it is clear based on an international comparison that Amtrak’s costs per distance and per speed in the Northeast Corridor are far higher than those charged by foreign operators of true high-speed trains. Fares seem hardly justifiable and the majority market share of car and bus commuters along the corridor shows the inevitable consequence of such pricey tickets. Rail ought to command a pure majority of traffic between Washington and New York as it does on similar corridor around the world.
All that said, while faster, cheaper Amtrak service would be a welcome improvement in the American transport landscape, Amtrak has little motivation today to lower fares. Its Acela and Regional trains run at a high occupancy rate and are sometimes sold out — because its train capacity is too low. Any increase in service would require more frequent, longer trains. It’s too bad such operations expansions are nowhere in Amtrak’s future plans. Even worse, not one state submitted an application for stimulus funds to finance a serious upgrade of service along the Northeast Corridor. We won’t be getting lower fares or faster trains any time soon — at least between New York and D.C.
24 replies on “Reframing the Fare Debate”
How about offering “family discounts”, or other discounted fares where a group of people are travelling together? What about price discrimination in general?
Part of the airline’s business model is based on charging a lot more to business travelers, on the theory that they are willing to pay more. Some of the ways this is done involving offering better service to such travellers–use of airport lounges, a nicer seat, refundable/changeable tickets. But many airline rules, such as discount fares requiring Saturday stays, are designed for no purpose other than nailing the business traveler.
As others have pointed out, Acela is right now a business-class service; given the contraints Amtrak runs under (capacity in the NE corridor; the need to fund all sorts of unprofitable routes, no rail competition), their present pricing strategy is certainly rational.
But at some point, HSR will need to attract riders other than business travelers.
“This is an advantage for train users, but increasing speeds to allow for a 1h40 trip would make it nearly impossible to justify riding the bus or driving, even at a lower cost.”
While I don’t disagree with your overall thesis, I think “nearly impossible” is overstating things. There will always be people whose time is less valuable than cash, or who simply don’t have an extra $250 but do have an extra 3 hours.
Yonah, both of your analysis are very well thought out. You laid out really clearly the subtleties between the trade-offs of speed and cost. Whenever I have ridden on the NEC, I always choose the regional trains over the acela because I can’t justify the cost difference for the little bit of difference in speed. The analysis really makes it clear why this is the choice so many make. I think capacity is largely the problem as you have said as the trains aren’t really bigger than in other parts of the country. I was shocked as well that no real requests for stimulus funds were made for the NEC. I mean it has the most chance to be the most profitable and get people to switch to rail. It seems something has to be done soon as everyone keeps on raving about Bolt Bus and Mega Bus. I’ve even known a few people that have sworn of taking Amtrak for good after going on them but they tend to be the more price sensitive types.
EngineerScotty: price discrimination may work for the airlines and for SNCF, but I tend to trust JR’s strategy more. In Japan, as well as South Korea, there’s a table where you look up your departure and arrival station and the train class you’re taking, and get your price. In Taiwan there’s no such table, but the fares are the same no matter how you order, and the closest thing there is to price discrimination is a 5% discount for group travelers.
Jfruh: the Acela doesn’t cost $250. For travel from NY to DC, it costs $100. If it takes 1:40, versus 4:00 for the bus, it’s not that expensive. At Regional (=coach class) fares of $50, it’s even less expensive.
Yonah, you write, “There are two ways to encourage people currently relying on road-based transportation to travel by trains: one, lower ticket prices; two, increase speeds.” You’re forgetting the most important one: make driving more expensive or less convenient.
That may be a pipe dream, but it should be acknowledged that the NJ Turnpike widening, the I-95/I-276 interchange, and the Q-Bridge replacement will all draw passengers away from Amtrak and the commuter railroads.
Yonah, the reasonable way to drive from Metro New York to Metro DC involves tolls, lots of tolls. I’d have to go look it – add 30 bucks round trip for tolls.
The states didn’t submit any proposals for the NEC because they don’t own it. Amtrak does. There’s separate funding in the stimulus, outside of the HSR funding, specifically for the NEC.
Sold out trains are not uncommon. Amtrak overbooks just like airlines do. Fortunately or unfortunately depending on how you want to look at it, Amtrak doesn’t bump people, they get to stand from Baltimore to NY or Philadelphia to DC. http://www.youtube.com/watch?v=nTepfymeQp4
Alon, which Amtrak fares are you looking at? I’m talking round trip here, and I’m also talking Baltimore-NYP, which is the route I usually travel. Acela Expres fares can be as much as $420 round-trip, depending on time of day; even the NE Regional trains (remember, these aren’t branded as Acela anymore) can be as expensive as $180 round-trip if you want to travel at peak times. The *cheapest* Baltimore-NYC regional trains are $64 each way.
In the larger sense, I don’t think you can just hand-wave price sensitivity away for all riders, no matter how much faster the train gets. There will always be people for whom even an extra $100 for a round trip is an insurmountable obstacle, or at least enough to make them think twice. This goes double when it comes to leisure travel. It’s easy to make the time=money calculus if you’re travelling for work, but folks just going up to NY for a weekend trip — with the activities in NYC themselves likely to soak up their disposable income — will be more likely to say, “Well, an extra couple of hours each way isn’t so bad if it’s less than half the cost.” It’s easier to make that decision because, even though trip by bus might be twice as long as a trip by train, the absolute numbers involved — three and half to four hours — aren’t terrible. (Obviously something like, say, a three-and-a-half-hour train ride to Raleigh, NC, vs. an eight-hour bus ride to Raleigh would be a different story.)
This is sort of hitting home for me because my wife and I are going to NYC for the weekend in a couple of weeks — and we’re taking Bolt Bus. Yes, I’m a huge train buff, and yes the train ride is significantly more pleasant, but with the current price matrix, it’s just hard to justify the price difference: to go coach on NE Regional trains, the total cost of tickets for both of us would be $404 at the times we want; to go by Bolt Bus, it fares were $61.50. When you’re discussing a better than sixfold difference in price, you need to be a LOT more convenient, and a 2:30 ride vs. a 3:45 ride isn’t that much more convenient, even with the extra leg room and cafe card in the mix; I don’t know that a 1:20 ride would even be able to justify the cost.
I’m not trying to be anti-train here by any means. I love trains, and I wish the NE corridor had more capacity. I’m just saying that if *I*, a huge train nerd, am making these kinds of calculations, you can bet that other people will be doing so, even if the trains become significantly faster.
As the Capt says above, the major argument left out of your ways to even the gap is making driving more expensive. This is a major part of things. Until such time as this becomes part of the plan, the plan will be less than fully effective.
And again, there is that often misunderstood bit about how higher average speed is more important than top speed. Yes they can both work together, but increasing the slower points will increase the average speed faster.
The other issue with all of this is that I really have a problem with all of these debates are us against them, 100% or 0%. It makes me think bush is still in office.
The goal shouldnt be to bash driving, air, or any other mode to get more trains. It should be how having trains will compliment things. How if you offer the traveling public another viable option of the train it can reduce congestion, both in the air and on the road.
This is an excellent site and I am fairly sure my last two comments have come off harsh, but I think if we’re more upfront with our arguments and present a solution instead of pointing out a problem in another mode, we’ll be better off.
Amtrak fares are cheaper if you book 2 weeks or more in advance. It’s their crude mechanism of price discrimination. If you book in advance, then NYP-BAL is $48 each way, making the total price for you $192.
If you want to see what happens when trains get faster, then look at the modal share in South Korea. The KTX takes you from Seoul to Busan in 2:43 and costs 47,900 Won, while the bus takes 4:20 and costs 19,800. The exchange rate is about 1,000 Won per dollar, but incomes in South Korea are about half as much as in the US. Despite this bigger-than-a-factor-of-2 price difference, the KTX dominates the market, with a 63% market share versus 7% for express buses. Mind you, South Korea has never had the express bus culture of the Northeastern US – the KTX poached riders from slow trains and the airlines – but it does show that you can create a successful line even with relatively high prices.
On the DC to NYC drive, you’re not including the tolls
– Baltimore tunnel – 2
– out of Maryland – 4
– Delaware 5?
– New Jersey – ~8
– The tunnel into Manhattan. ~8
Both ways, that’s 54 bucks. I only get 18 mpg, so that tank and a half of gas is 75. I’m typically north of 120 if I drive. That’s the cost of a planned Delta round trip but not including the time and public transportation fare.
It’s also 20 bucks more than a planned regular Amtrak.
If I get advance notice, and i’m going by myself, I take the train.
The only time the drive makes sense, is if I have more than one person with me.
There are many other factors in the NEC debate between local and Acela. One, I find the Acela trains much more pleasant, and to have more room, about the equivalent of Business Class on the regional. Also, (and if you travel on weekends or holidays you’ll know what I mean), the Acela trains do not over-sell their seats! I have been on way too many friday night and holiday trains where the conductors spend the whole trip rudely reminding passengers who have to STAND for 4 hours that “your ticket only buys you passage, not a guaranteed seat.”
There is a lot more to the price equation than just dollars and hours. Look at airline price structures – Business Class is 5x more and First Class is 10x more than coach, but you’re on the same plane taking the same trip in the same time! People ARE willing to pay for certain qualitative things, not just quantitative.
You’re right, the economics work for one person. But add another, and the equation changes dramatically. Make that a family of four, and there’s no way a train can compete. Add a dog, or any sort of cargo, etc, etc.
I think trains need to appeal to drivers for qualitative reasons more than sheer dollars. It would great if they are faster, door-to-door, taking traffic into account, provide quality time for families, etc…
If you think about it, it’s possible to drive anywhere in the contiguous 48, but I would bet that most people who choose to fly make the decision based on time, not money.
A big reason people drive to another city is that they need the car when they get there. We need to make urban mass transit more convenient and frequent. The best path to this goal is through elimination of the fare.
fpteditors, are you suggesting that the reason Amtrak doesn’t get more ridership is that New York and Washington don’t have enough mass transit?
Alon, I really invite you to actually plug dates into the Amtrak ticketing app. I just did so for the weekend of Oct 4-6. I’m seeing some $48 fares going to NYC on Friday (along with some $92 and some $110), but all the Regional tickets returning on Sunday afternoon are $92 or even more. To get tickets at more or less equivalent time to our $61.50 Bolt Bus tickets would still cost $280 — a fourfold increase. Even $192 is triple the cost.
I’m not sure why I’m going back and forth with this so much with you as I do believe your central point — that faster train travel will result in increased market share even if the tickets remain similarly priced. If it gets fast enough it might even make me consider paying $280 for our weekend getaway. I just think don’t like rhetoric that implies that practically everybody could cough up an extra $220 or $130 dollars if only the trip were faster. Many will, and it’s the aggregate numbers that are important for making policy, but others won’t and many can’t.
The capacity issue on the NEC seems to be lack of equipment. When I was a regular PRR rider, the equivalent of Regional trains were 10-14 cars long, now they are 6, maybe 8. Pax per car were similar so current trains carry fewer riders wasting an operation slot. ATK does demonstrate what should be daily service levels in a few holiday rush situations, they just don’t do it often enough. The other pricing problem is 535 low iq employees in DC.
I rode a $64 (one way) Keystone on Sunday to NYC from Philly, EMPTY. As I exited the station, Megabus pulled up to the corner and it was FULL. $64 one way plus $45 back for 90 miles? ouch. The buses are $20-45 RT. If amtrak wants to carry more people, it will lower prices on the Keystone trains for commuters and others for the Philly-NY trains to free up seats for through Philly service. Keystones should be around $50 RT and maybe $600/mo for commuters
it is clear based on an international comparison that Amtrak’s costs per distance and per speed in the Northeast Corridor are far higher than those charged by foreign operators of true high-speed trains.
Of course. Foreign high-speed trains are even more outrageously subsidized than Acela.
Mixner, out of curiosity, you never did admit that you were wrong about New York’s congestion numbers. Or about the fact that the Tokaido and Sanyo Shinkansen and the LGV Sud-Est paid off their construction debts, as explained in New Departures (and, one might add, future lines were built mostly from those lines’ operating profits). Or about 1,001 other refuted Reason talking points you spouted off on the Austin Contrarian.
Or maybe you just think “outrageously subsidized” means “outrageously milked for profits.” I’m not sure.
(For people who don’t know what I’m talking about: Mixner claimed on some thread that New York’s density has made it the second most congested urban area in the US. I countered with statistics showing that per capita its congestion ranks in the middle of large urban areas. Mixner then responded by saying that if you multiply the per capita number by the population, then New York is second.)
Mixner, out of curiosity, you never did admit that you were wrong about New York’s congestion numbers.Or about the fact that the Tokaido and Sanyo Shinkansen and the LGV Sud-Est paid off their construction debts, as explained in New Departures (and, one might add, future lines were built mostly from those lines’ operating profits). Or about 1,001 other refuted Reason talking points you spouted off on the Austin Contrarian.
What are you talking about? You haven’t “refuted” anything I’ve written on those subjects, or anything else. Your own writings are full of false and unsupported claims of fact and gross misrepresentations of studies and reports.
(For people who don’t know what I’m talking about: Mixner claimed on some thread that New York’s density has made it the second most congested urban area in the US.
New York is the second most congested metro area in the U.S. The data are here: http://scorecard.inrix.com/scorecard/Top100Metros.asp
I countered with statistics showing that per capita its congestion ranks in the middle of large urban areas. Mixner then responded by saying that if you multiply the per capita number by the population, then New York is second
More nonsense. First, I didn’t respond to your “counter” at all. It was too stupid to bother with. You’re now attributing to me statements that exist only in your fevered imagination.
If you had bothered reading how the congestion ranks were calculated, you might have realized that they are independent of population. I tire of correcting your endless falsehoods and misrepresentations. Even when your errors are explained to you, you simply ignore them. Remember your recent confusion of “energy” with “residential electricity?” Or your hilarious claim that you know the TGV isn’t subsidized because the wikipedia entry for the TGV doesn’t mention subsidies? I’m not sure why I even bother with you at all.
You bother with me because you’re a troll. Trolls love to play with regulars. That’s why your congestion data come from a derivative source of the TTI. The metric you use is total wasted hours, rather than total wasted hours per capita.
And in related news, China has the largest number of person life-years in the world. It also has the highest number of deaths per year. So is it the healthiest or least healthy country in the world?
The questions I have are
1. What is the Direct Operating Cost (DOC) for operating a high speed train that travel at an average of 150 MPH or 360 KMPH? For instance you have a train with 7 sets; the first and last trains where the controls are and 5 other bogie sets if that’s what it’s called. The controller train sets have 125 seats each x 2 and the other 5 have 150 seats bring it to a total of 1,000 seats. I guess my question is what is the DOC for operating each controller set and an added bogie set?
2. I was looking at an article comparing modes of transportation to their CASM, and Amtrak was operating about 1.5x more than what the standard airline was charging, so I will play it safe and assume that if that be the case, a standard bogie train set with 150 seats (equivalent to a B737-700 from SWA operating at $0.25c CASM) is operating at $0.38 CASM. If that’s the case, irrespective of the number of train sets added, the cost remains the same unless if the sets carry more seats to offset the balance. I mean I’m really curios to know how their cost, expense and profit are calculated. So the question is; are the associated cost more fixed than variable or marginal? More so that cost is fixed regardless of the number of train sets, or they increase as the trains sets increase or irrelevant after a certain number of train sets are added? At $0.38 per CASM, and at 225 mile from DC to NY, an average ticket will cost about $86; however, you are suggesting @ $49; that is a 75% increase from $49 but if that be the case, then CASM would be @ $0.21. I used this calculation like it’s used for airlines because most airlines (DOC) factor in a huge amount of other variable components.
3. Also I wanted to know if it’s true that no company can compete with Amtrak for better service. I read from a website (I don’t know how valid the website is) that it is federal law that no other service will compete with Amtrak. If that is true, it would make valid sense why there isn’t other private rail service and explains why Amtrak is struggling rather than focusing on niche routes such as NEC.
A huge proportion of rail costs are fixed regardless of the number of trainsets. A larger proportion are fixed regardless of the number of trains per hour. An even larger amount is fixed regardless of the number of cars per train.
Trains have high fixed costs, low variable costs, it’s why they excel at high volume applications.
To answer your last question, the fact is that any company can try to compete with Amtrak. It just has to get access to the railroad tracks — and only Amtrak has preferential access to railroad tracks, and it isn’t required to share it.