Here’s how American Airlines explains why they do it:
“Because we compete with other airlines with different route structures, we sometimes find it necessary to give a traveler who is traveling beyond a connecting point a better price than travelers who are just traveling to the connecting point. For example, a passenger who is traveling to Austin, Texas from Los Angeles can go on one airline via Phoenix for a price that is lower than the cost of traveling on American between Los Angeles and Dallas. If we want to offer the same price to Austin as the other airline, but the only way we can get travelers there is via Dallas, we find ourselves charging the Austin passengers less than the Dallas passengers.”

(http://www.aa.com/content/agency/Booking_Ticketing/Ticketing/hidden_city_ltr.jhtml)

It makes economic sense, however, to price seats that would otherwise be empty at a lower price. The poster who suggested the Springfield Line has heavier passenger loadings because of speed and frequency has the idea. The train is less likely to have empty seats south (timetable west) of Springfield. Because Amtrak doesn’t cut cars north (timetable west) of Springfield, and because the train goes the great way around (east to Palmer, northwest via Amherst to Bellows Falls) there are plenty of opportunities to sell otherwise empty seats — the term of art is “buckets” at what look like low prices.

You might recall people buying two weekend-stay airline tickets to make a midweek trip at a lower price than they’d pay for the weekday round trip.

I live in Boston. It is much closer to Stowe than New York City. Here’s the rub - I either have to take the ONCE per day train between Boston and Springfield, and then wait for the Vermonter OR I have to take a train to New Haven (Acela or regional), and then go North. How f%&*ed up is that?!?

Okay, I’ll answer my own question. It is very f%&*ed up, as I have in my office an old poster (probably from the 50s) beckoning me to take the “Snow Train” from Boston’s north station to the ski areas in NH for a glorious “Sun-day”. There are exactly ZERO trains serving that route now.

The second was a case similar to yours where local evening service (in this case from blackburn to manchester via bolton but i will use US cities and currency to simplify). An evening return from LA to new york was $5 because the city of new york had 50% subsidised the travel in the evening to encourage people to come out and see a show and leave the car behind and also to curb drink driving. but you had to go all the way to new york. A ticket from LA to philadelphia (for example) whilst on the same route and same train would cost $9 because philadelphia didn’t subsidise travel there, stay on one more stop and you’d pay half as much.

Full points on style. But, on substance, I disagree. At least partly. The world might be more functional if scientists thought a bit more like journalists–particularly on the matter of climate change.

Then again: if it is the small thing that leads to big breakthrough–then, yes, I see your point.

I really would love to have an economist weigh in though.

An interesting insightful piece, and yes “as a journalist” means “like a scientist” — the important thing is to know why.

Seems like there’d be times when people wanting to leave Denver would be told that plane was full, so either they’d have to wait or the airline would have to put up another plane, even though the first plane was “full” only because of the screwed-up pricing.

And airlines and Amtrak wonder why they’re going broke!

Wonderful and fascinating slice of life, Jebediah, and a GREAT idea for a website keeping us up to date on what’s going on with our crumbling infrastructure.

I’ll definitely be double-checking destination prices after reading this.

Robert

Thanks for the thoughts. I mean that explanation is interesting, but it doesn’t really answer the question. I would think one of the most basic lines of code in any program to figure out fares would ensure that the fare from A to C can never be lower than the from A to B.

Because, you know, every rational person going to B then just buys a ticket to C. You could say they might not know about about the difference in fares, but I don’t really buy that. As I say, it’s seems like a law of a nature not a guideline in pricing things — something we all understand instinctively.

And I’m not sure the airlines really do this (direct analogy — not just a longer flight cheaper than a shorter one)… I’d be curious to see evidence. But even if true, it’s a lot harder to work the logistics of hopping of halfway when dealing with the security hassles of flying these days.

What makes this amusing to me–and I’m really not a government basher–is that it seems like the kind of mistake (or decision) that only a government agency could make. It involves a certain tolerance of irrationality.

Those tracks up north are a mess, by the way. Like riding over a dirt road.

Jeb

One reason for this is that Amtrak has more pricing power between New York and Springfield where it is faster (80 and 90 mph) than on the rest of the leg to Stowe (59 mph with some slow patches and a detour via Palmer).

The other thing that’s going on is that it costs more to take the Vermonter from New York to Springfield than regular Northeast Regional trains. I just checked: the Vermonter is $63 tomorrow, but other trains are $53. That makes sense. They don’t want seats on a train going all the way down the length of Vermont taken up by people just going from NY to Springfield and preventing people from going New York - Stowe. This may be deliberate or may just be the result of the way the yield management system automatically increases the fare when there are more riders booked.

Here’s how that works: There are four fare levels. Suppose on an off-peak time it’s $56 to go from New York to Stowe. When a certain number of early seats are sold, the system automatically raises the fare. Then when a certain number of seats at that fare are sold, the fare goes up and so on. That way the fare for peak periods is higher than for off-peak. And when someone knows it will be peak (like around the holiday) the fares start out in the higher bucket. But here’s where it can get screwy: if there are separate processes for NY-Springfield and NY-Vermont (as it appears there are), then the NY-Springfield tickets can get kicked into the higher buckets before the NY-Vermont tickets. I suspect that’s what happened. It makes sense that the train would be fuller on that part - besides more population down your way, the train also fills up at New York and just gradually empties the further it goes. Most likely this is a completly automatic process and nobody noticed at Amtrak that there was this funny discrepancy. (Not that I know, since I don’t work for them).

Another funny thing is that the club car also has it’s own independent process. The one time previously that it came up it resulted in a much higher fare for a club car seat - even though it was only a $12 upgrade, there are less seats in the club car, meaning the base fare was in a higher category than in coach because the certain percentage of seats had already been sold. I suspect it could go the other way too, although maybe there is something built into the system to prevent that.