Observation on sleeper revenue

[Anderson]

I just decided to pull some numbers for fun.

-An LD Superliner coach has 74 seats (62 upstairs, 12 downstairs).

-An LD Superliner sleeper has 28 "roomette slots" (14 roomettes and 7 total bedrooms), or 27 accounting for the SCA's roomette.

Taking the Capitol Limited's endpoint buckets, a full coach at low bucket would net $6,660 in revenue. A full coach at high bucket would net $13,172. Now, let's slide over to the sleeper...

Compared to low bucket, a sleeper "should" generate $237.86 per "slot" (i.e. that for a roomette or $475.71 for a bedroom). A roomette with one rail fare at low bucket will generate $294 (+$56.14) while a bedroom with two rail fares at low bucket will generate $450 (-$25.71).

Note: Taking into account one roomette given over to an SCA, the numbers are $246.67/$493.33. The roomette therefore shows +$47.33 and the bedroom shows -43.33.

Comparing both to high bucket, the roomette should generate $470.43 while the bedroom should generate $940.86. The actual numbers are a roomette/one rail fare at $458 (-$12.43) and the bedroom/two rail fares at $816 (-$124.86).

Note: Again, taking into account one "lost slot", you get $487.85/975.70. This gives -$20.85 for the roomette and -$159.80 for the bedroom.

What does this say? At least on the Capitol Limited, roomette occupants are likely to be paying for "their share" of the car (and then some, considering that a non-negligble number of roomettes will have two occupants instead of one), and are likely putting in enough to cover the added meal costs as well. Bedroom occupants, however, aren't. Though I can't quite peg the picture, roomette occupants may even be putting in enough to also cover their fraction of extra SCAs' pay versus the cost of having coach attendants (i.e. the additional per-car staff dictated by the nature of a sleeper versus a coach). The situation does get a bit dodgier at high bucket versus low bucket, but that end of things is probably a bit more academic (since Amtrak isn't yet to the point that they can sell multiple cars full at high bucket on either end of the train).

 

[AlanB]

-An LD Superliner sleeper has 28 "roomette slots" (14 roomettes and 7 total bedrooms).

While the Superliner sleeper has 14 roomettes, only 13 are revenue rooms. Room #1 is given over to the attendant and generates no revenue.

 

[pldenc44]

Any info on how many staff per sleeper vs staff on coach? I was surprised to see that a full sleeper grosses so much more than coach. It debunks the idea that Amtrak's poor performance is a result of running so many sleepers.

 

[Anderson]

I think it's one extra per two cars (i.e. one coach attendant per two cars vs. one SCA per car), but I'm not sure.

Alan: Good catch. Let me redo the figures for 27 instead of 28.

Edit: I accounted for 27 roomettes and put in new numbers. Really, the difference is almost immaterial. At worst, this suggests that roomettes more or less break even against a similarly-loaded coach (and probably go to a higher bucket at a lower cost than do coaches; I should probably be weighing low-bucket coach against second-bucket roomettes considering that I think I've generally had to look way ahead to find lowest-bucket sleepers) and the bedrooms come in a little bit behind. The reason for coach seats being available at a lower bucket (on average) is that, I would suspect, Amtrak has a lot fewer roomettes to sell (maybe 36 on the Capitol Limited) than they do coach seats (216)...so there's plenty of incentive to "start high" on the roomettes while on the coach side...there's probably going to be a little bit more slack in the system that it's easier to fill with a lower bucket.

Of course, I could also be sharply sarcastic and blame the ADA for some of the losses (and I'll probably make an appropriate note for any analysis on a Viewliner sleeper...on the Superliner, you just eliminate part of a dead end hallway, but on a Viewliner you lose a "slot" to the Accessible Bedroom).

 

[AlanB]

I was surprised to see that a full sleeper grosses so much more than coach. It debunks the idea that Amtrak's poor performance is a result of running so many sleepers.

Actually NARP debunked that myth several years ago with a paper that showed the Fed spends less per passenger mile for those in sleepers than it spends for those in coach.

 

[Anderson]

I was surprised to see that a full sleeper grosses so much more than coach. It debunks the idea that Amtrak's poor performance is a result of running so many sleepers.

Actually NARP debunked that myth several years ago with a paper that showed the Fed spends less per passenger mile for those in sleepers than it spends for those in coach.

That was a good paper...but I'd also been wondering about changes in the interim (especially considering things like the bucket jump on the Silvers, for example). This is a large part of why I'm trying to keep track of a lot of these things over the longer term...to track relative performance and figure out the why of the what, so to speak.

Two more observations:

1) During the week in August, I frequently find the $90 coach fare facing off against the $311 sleeper bucket, if not the $368 bucket.

2) A mea culpa: I found a $141 sleeper bucket on one two days in September (9/9/12 and, on a second look, 9/23/12). However, this borders on being a "for show" bucket, as I found that on two trains in a complete search of the rest of the booking timeframe (I think). In any meaningful sense, $255 is the low bucket. I'm going to amend my "bucket list" to note this, but I'm also going to note that above, $255 should be treated as low bucket (since that's where fares seem to open up).

I'll be disclaimering any such list when I post it...aside from highly "directional" routes (i.e. the FL trains in the spring and fall), finding the low bucket on some of these routes involves Amsnagging a given route into the fall and then trying to make sure that I've noted all five buckets lest I be missing one.

For the Cap, these buckets are:

Sleeper:

$368 (very common over the summer)

$311 (common shorter-range fare)

$255 (most common long-range fare)

$198 (rare, really only shows up in September from what I find)

$141 (borderline extinct, only found on two occasions in the next eleven months)

Coach:

$176

$141

$113

$90 (common farther out [i.e. Sept. and beyond], rare over the summer)

 

[Anderson]

Well, a serious question does come out of this. With coach buckets, there's a pretty decent distribution of fares across all four buckets...higher buckets during the summer, with $90 being universal at extreme time ranges and $113 only intervening during peak season right now (i.e. Thanksgiving week). Obviously this is likely to change as time goes on and seats fill up. However, Amtrak seems to basically keep the $141 and $198 buckets on the books for the sake of having them. Why does Amtrak maintain a pair of "gag" buckets like that now? One rarely-used bucket makes at least some sense, but it seems odd to have two of them...especially as it would seem that having a low one at $141 is acting to "tie down" the top bucket more than it is doing anything to sell space on the train. Maybe it's there to encourage on-board upgrades or something like that (which seems silly considering that Amtrak would effectively be selling those for $114 less than the "normal" fare and $57 less than the lowest fare someone seems likely to come across). This might also be instructive on "sleeper gambling" in the off-season if you want to sort the odds on that.

Also, a note: The roomettes and bedroom do seem to apply the "top bucket shall not be more than double bottom bucket" rule..just once one applies a single railfare in addition to the room charge. I'd wondered about this. Still, given its non-use, this really begs an inquiry as to why they don't just sack the $141 bucket and scatter the spare bucket either in the middle of the existing ones...or toss it in at the top. I'll try to remember to revisit this in a month or two to see if more roomettes start coming down into at least second bucket range. Likewise, the sleeper buckets move in increments of $57-$57-$56-$57...while coach buckets move in increments of $13-28-$35.

(Note: I pulled this out of the last post since it was really becoming a post in its own right)

 

[gatelouse]

The top coach bucket and bottom sleeper bucket are ephemeral--you'll see both closer to departure.

Check out 30 of May 29--coach is selling at high bucket. 30(28) is sold out.

I've seen low bucket sleeper pop up within a week or two of departure when people cancel reservations. Perhaps a revenue management scheme of some sort.

 

[The Davy Crockett]

I was surprised to see that a full sleeper grosses so much more than coach. It debunks the idea that Amtrak's poor performance is a result of running so many sleepers.

Actually NARP debunked that myth several years ago with a paper that showed the Fed spends less per passenger mile for those in sleepers than it spends for those in coach.

Which, really, is exactly as it should be. It makes sense that those who can afford to pay for more of their share.

 

[Anderson]

I was surprised to see that a full sleeper grosses so much more than coach. It debunks the idea that Amtrak's poor performance is a result of running so many sleepers.

Actually NARP debunked that myth several years ago with a paper that showed the Fed spends less per passenger mile for those in sleepers than it spends for those in coach.

Which, really, is exactly as it should be. It makes sense that those who can afford to pay for more of their share.

Really, I don't think anyone disputes that. It's more that:

A) A lot of people have misconceptions about the revenue picture;

B) Remember, 50 years ago this wasn't the case;

C) Because the sleepers back up so much about the LD trains' picture (i.e. the dining car operations) and they require things like the diner to attract passengers (let's face it, I don't think anyone wants to travel from Chicago to Los Angeles on cafe food alone), there's a very delicate ecosystem that has to be defended from top to bottom. We saw what happened when it wasn't (for example, I'm advised that under Warrington when the food service situation went to hell in a handbasket); and

D) Finally, there's the question of how much over coach fares Amtrak can extract before it "prices out" a market. Obviously, it is capable of commanding some rather substantial premiums, but there's also obviously a point where folks will start looking at other ways of traveling and/or question the wisdom of taxpayer subsidies to Amtrak. Let's be honest, regardless of the merits of rail travel, it's easier to attack if a lot of those subsidies are going to cover $500-a-night roomettes and/or if the coach fares are high enough that nobody below the upper middle class can afford them.

 

[Guest]

Keep in mind that there are fares in every bucket for every city pair along a route and, for shorter distances, it is not uncommon to use both highest and lowest buckets to optimize revenue and capacity by charging higher buckets for short-distance high-demand markets, and for filling in low demand portions of a route.

 

[Rlschur]

For what it's worth, I managed to get that $141 bucket on this past March 17 about a month before departure. The fare showed for several days before I bought it, too.