FY2013 PERFORMANCE REPORT

[henryj]

It's out. The final performance report for the fiscal year. Looking at revenues, Auto Train is king with 75.4 million followed closely by the EB at 72.9. Third thru sixth are CZ at 55.7, SWC at 49.1, Starlight at 47.7 and Meteor at 42.4. After that they decline into the 30's and on down to the Cardinal at only 8.6 on the bottom.

Operating profit/loss. Three trains end the year in the black, Auto Train of course is tops followed by the CONO and the Palmetto. Biggest losers are CZ at 26.8, SWC at 17.9 and Crescent at 12.1. After that you have the Sunset, Capitol, Star and LSL all in the 10-11 million range. Then in progression there is the Starlight, EB, Meteor, Cardinal and Eagle. Total operating loses are around 108 million for an 84% recovery. Actually not too bad.

By Amtrak standards after they load up everything with overhead and who knows what else, the LD trains lose 627 million. Biggest losers according to Amtrak are the CZ, SWC, Starlight and EB. The SWC by their standards loses 69 million which may explain the funny actions they are taking regarding the re-route. Note if they were to discontinue the SWC they lose 49 million in revenue and probably save less than that in avoidable costs.

Anyway, that's my read on the FY2013 numbers. Enjoy.

 

[Paulus]

For all those coming into this thread unfamiliar with henryj: Please disregard all his numbers and claims of profit or loss as they are his own imaginings. Here is the actual report

Edit:

First observations: Texas Eagle, Sunset Limited, and Auto Train are only long distance trains to financially improve over FY2012 (loss per passenger-mile).

The Acela has almost as much revenue as all the long distance trains combined and a per passenger-mile profit 50% higher than long distance sleeper fares tend.

 

[PRR 60]

For all those coming into this thread unfamiliar with henryj: Please disregard all his numbers and claims of profit or loss as they are his own imaginings. Here is the actual report

You beat me to it. I looked at the report, and no long distance trains came close to break even.

 

[Ryan]

Thanks for posting the link Paulus.

Looking at the ridership by month graph on A-2.4, it looks like there's a pretty hard cap on ridership at the 2.8-2.9M riders per month (it's flat there March-August). That seems to suggest that without more capacity added, getting above 33.5-34.5M annually is going to be real tough.

Also interesting: Looking at the ticket yield ($/psngr mile), it does best in April/May and tanks in the summertime when ridership is highest. Intuitively, I would think that summer would do better with sold out trains meaning high bucket tickets and better yield. Wonder why that isn't the case.

 

[henryj]

For all those coming into this thread unfamiliar with henryj: Please disregard all his numbers and claims of profit or loss as they are his own imaginings. Here is the actual report

Edit:

First observations: Texas Eagle, Sunset Limited, and Auto Train are only long distance trains to financially improve over FY2012 (loss per passenger-mile).

The Acela has almost as much revenue as all the long distance trains combined and a per passenger-mile profit 50% higher than long distance sleeper fares tend.

Some people on here don't like me. lol. And some just don't like the numbers, but can't generate any of their own. All of this comes from Amtrak own 'actual reports'.

 

[Ryan]

Ridership up 1% and revenue up 4.2% is a good sign that revenue management works.

Henry, your numbers do not come from 'actual reports'. It's 'actual reports' + a much of made up stuff.

 

[Anderson]

Thanks for posting the link Paulus.

Looking at the ridership by month graph on A-2.4, it looks like there's a pretty hard cap on ridership at the 2.8-2.9M riders per month (it's flat there March-August). That seems to suggest that without more capacity added, getting above 33.5-34.5M annually is going to be real tough.

Also interesting: Looking at the ticket yield ($/psngr mile), it does best in April/May and tanks in the summertime when ridership is highest. Intuitively, I would think that summer would do better with sold out trains meaning high bucket tickets and better yield. Wonder why that isn't the case.

I disagree with there being an obvious cap:

June, FY13 showed increases in both the NEC and in state corridors and an overall increase in ridership of about 43,000.

July, FY13 showed increases in all three areas and an overall increase of about 133,000.

August, FY13 showed increases in state corridors and long distance trains, and an overall increase in ridership of about 82,000.

Taken together, that's an increase of around 250k over the course of the peak season.

Also: I'd need to get a look at more detailed data, but from what I can tell, March-August has generally been a fairly flat peak: If you look at an old press release about the consistent monthly records, you had similarly flat peaks in FY10 and FY11, albeit at lower levels. March/April just gets mangled sometimes because Easter moves.

Link: http://www.amtrak.com/ccurl/63/236/ATK-11-100%20Amtrak%20to%20Exceed%2030%20Millions%20Passengers.pdf

================================================

MOD HAT ON: I know we disagree on a lot of stuff, but let's try to stay civil.

 

[Nathanael]

Here is the actual report

First observations:

- What caused the 10% drop in September ridership on the Quincy trains? Just a fluke?

- There isn't much of an overall trend in full-year ridership; it seems to be roughly flat with random fluctuations. Of course, there are the usual swings caused by 2012 and/or 2013 weather disruptions, and the Newport News results are affected by the introduction of the Norfolk trains.

- Chicago-St. Louis has the best year-over-year percentage growth in ridership;

- San Joaquins have the second best;

- Coast Starlight has the third best (Californians like their trains, I think);

- the Heartland Flyer has the worst percentage drop, not counting special trains.

- The revenue numbers which were most off-from-budget were "Buses", which had much more revenue than Amtrak expected. Does this include train replacement buses or just connecting buses? Anyone know?

Interesting tidbit from the sleeper ridership section:

Over the full year, the Silver Meteor has a higher sleeper load factor than the Lake Shore Limited. But in September, the Lake Shore Limited has a much higher sleeper load factor. Looking back, the LSL has more sleeper passengers from June through October, while the Meteor has more sleeper passengers from November through May. Very interesting seasonal pattern. The LSL is a *lot* weaker in the winter, with a very sharp drop from September to October; the Meteor is not quite as much weaker in the summer, which is why the Meteor does better overall. I wonder if the LSL ridership pattern has something to do with school schedules.

 

[Nathanael]

Also worth noting:

- to a first approximation, the sleepers on the single-level trains are always full. Most sleeper compartments are sold several times per trip, with (yearly passengers / yearly capacity) being roughly 1.7 on the Silver Meteor and roughly 1.4 on the LSL -- remembering that that's counting all the roomettes as double occupancy. Perhaps the numbers are just as high on the bilevel trains too but I haven't bothered to compute it. Buying new sleepers is sensible.

- Mechanical is still complaining about lack of availability of Amfleet II coaches. Also, management cut down the Amfleet II Cafe refurbishment program (they're listed as Amfleet II "Diners" FWIW). I think Amtrak is serious about trying to replace the Amfleet IIs next.

- Amtrak loses money on NEC freight service, on a "fully allocated" basis. But less than last year.

- Amtrak *also* loses money on "ancillary customers" on a "fully allocated" basis this year, unlike last year. I'm not sure what to make of this.

- Allocations are sufficiently meaningless than I'm not going to pay further attention to the "fully allocated cost" charts.

 

[Anderson]

Also worth noting:

- to a first approximation, the sleepers on the single-level trains are always full. Most sleeper compartments are sold several times per trip, with (yearly passengers / yearly capacity) being roughly 1.7 on the Silver Meteor and roughly 1.4 on the LSL -- remembering that that's counting all the roomettes as double occupancy. Perhaps the numbers are just as high on the bilevel trains too but I haven't bothered to compute it. Buying new sleepers is sensible.

- Mechanical is still complaining about lack of availability of Amfleet II coaches. Also, management cut down the Amfleet II Cafe refurbishment program (they're listed as Amfleet II "Diners" FWIW). I think Amtrak is serious about trying to replace the Amfleet IIs next.

- Amtrak loses money on NEC freight service, on a "fully allocated" basis. But less than last year.

- Amtrak *also* loses money on "ancillary customers" on a "fully allocated" basis this year, unlike last year. I'm not sure what to make of this.

- Allocations are sufficiently meaningless than I'm not going to pay further attention to the "fully allocated cost" charts.

The "allocated cost" basis basically means that once you put a portion of overhead, etc. to those lines, they lose money. However, it seems likely that Amtrak is at least making some money vs. directly attributable costs and putting something towards overhead...just not enough to cover all the allocated overhead.

On the Meteor/LSL: The LSL behaves like a lot of the other trains. The Meteor goes to Florida, so it gets a lot more off-season business than a number of other trains do.

 

[Anderson]

I figure I can go ahead and elaborate on the general pattern of a year for Amtrak:
-The "good" months, ridership-wise, tend to be March-August. March/April have Easter and spring break to boost ridership, while May-August tend to get lots of vacationing traffic from the warmer months. July has often peaked above the others, but the peak isn't that massive (around 100k over most of the other months in this period) and August doesn't trail too far behind.

-The "bad" months tend to be January, February, and to a lesser extent September and October. January and February have awful ridership due to seasonal issues, and February gets slammed with a shortage of days to boot. September and October would be worse off except that September has Labor Day to help, while October has fall break, Columbus Day, and some other miscellaneous effects to help.

-November and December fall in the middle: You have 2-3 weeks of "bad month" ridership patterns mixed with 1-2 weeks of peak-of-the-peak ridership patterns.

 

[jis]

What is funny is that Amtrak is so used to reporting loss on everything that even in the rare occasion where a subaccount has a positive number the title of the line calls it a loss, instead of using a title like "gain or (loss)". See the bottom line of table A 4.5 for example.

 

[Paulus]

Anyone know why the Wolverine expenses exploded? They're up 21.3% from FY2012, 46.7M vs 38.5M including OPEBs. Expenses majorly up on the Maple Leaf and Ethan Allen Express as well. Hoosier State expenses almost doubled.

 

[PRR 60]

One interesting item from the "Delays" section (graph Page E-2). For the 12 months ending September, 2013, the host railroad with the best performance as measured by the fewest delay minutes per (10K) train miles is..... (wait for it).... (drum roll)....

UNION PACIFIC!

 

[jis]

One interesting item from the "Delays" section (graph Page E-2). For the 12 months ending September, 2013, the host railroad with best performance as measured by the fewest delay minutes per (10K) train miles is..... (wait for it).... (drum roll)....

UNION PACIFIC!

You mean the same much maligned in AU the ultimate thorn in the side of all passenger operations the US according to some vocal elements on AU .... UNION PACIFIC? Who'd have thunk? It is probably a good thing they do not include Amtrak as a host railroad in there. Specially it gets pretty ugly when you look at what it does to Commuter trains at times.
I just noticed that Amtrak does not even report what it does to its own trains on the NEC. How clever?

 

[Nathanael]

I figure I can go ahead and elaborate on the general pattern of a year for Amtrak:

-The "good" months, ridership-wise, tend to be March-August. March/April have Easter and spring break to boost ridership, while May-August tend to get lots of vacationing traffic from the warmer months. July has often peaked above the others, but the peak isn't that massive (around 100k over most of the other months in this period) and August doesn't trail too far behind.

-The "bad" months tend to be January, February, and to a lesser extent September and October. January and February have awful ridership due to seasonal issues, and February gets slammed with a shortage of days to boot. September and October would be worse off except that September has Labor Day to help, while October has fall break, Columbus Day, and some other miscellaneous effects to help.

-November and December fall in the middle: You have 2-3 weeks of "bad month" ridership patterns mixed with 1-2 weeks of peak-of-the-peak ridership patterns.

Thanks.

Given this, the LSL deviates from that pattern in that it's weak in March and April (no spring break ridership, I'd guess) and quite strong in September and October. What I found interesting is that the LSL actually outperforms the Meteor in the summer, though the Meteor substantially outperforms the LSL in the winter. There's a strong case for the "rotating sleeper".

Regarding OTP.

The Empire Service specifically is still suffering a lot of Amtrak-induced delays; first look at the very high number for "crew related delays / delays in block", and then add some percentage of the "passenger holds" to that, as those can be due to bad crew management of boarding. Something needs to be done in that specific operational division.

As others noted, UP has had excellent OTP since the Amtrak lawsuit a few years back. Perhaps someone in management decided it was worth the small extra cost to run the trains on time, in terms of public relations; this is what BNSF decided when Matt Rose took over (he was quite public about it). NS is also doing quite well.

The worst two Class Is are now CN (improving) and CP (getting worse). Therefore I blame Hunter Harrison, who is the common element, and who moved from CN to CP. He's from the generation which most commonly hates passenger trains, FWIW. Things may get better when that generation retires from Class I management.

 

[Nathanael]

Anyone know why the Wolverine expenses exploded? They're up 21.3% from FY2012, 46.7M vs 38.5M including OPEBs. Expenses majorly up on the Maple Leaf and Ethan Allen Express as well. Hoosier State expenses almost doubled.

No idea but I'm going to guess it's overhead allocation procedures. It seems really unlikely that it's actual expenses. Remember that Auto Train had a major change in its cost allocations not long ago.

 

[Paulus]

Anyone know why the Wolverine expenses exploded? They're up 21.3% from FY2012, 46.7M vs 38.5M including OPEBs. Expenses majorly up on the Maple Leaf and Ethan Allen Express as well. Hoosier State expenses almost doubled.

No idea but I'm going to guess it's overhead allocation procedures. It seems really unlikely that it's actual expenses. Remember that Auto Train had a major change in its cost allocations not long ago.

Auto Train expenses were actually down for the year.

Here's some of the major information related to cost recovery, color coded for your convenience. The Hoosier State expenses just make no sense whatsoever now. It's more expensive than several other routes.

 

[henryj]

Ridership up 1% and revenue up 4.2% is a good sign that revenue management works.

Henry, your numbers do not come from 'actual reports'. It's 'actual reports' + a much of made up stuff.

Yes Ryan, they come from Amtrak's own 'actual reports'. They are not made up. I take it they just don't tell the story you want them to tell. lol.

 

[Anderson]

Well, and we probably lost another 1% of ridership to the incessant NEC disruptions over the course of the year.

 

[Ryan]

Ridership up 1% and revenue up 4.2% is a good sign that revenue management works.

Henry, your numbers do not come from 'actual reports'. It's 'actual reports' + a much of made up stuff.

Yes Ryan, they come from Amtrak's own 'actual reports'. They are not made up. I take it they just don't tell the story you want them to tell. lol.

What report tells us that the Auto Train made a profit this year?
Care to post your spreadsheet?

 

[Paulus]

Slight oddity: Washington-Newport News is credited with 32.9M in ticket revenue, yet in Table C, only 31M in total revenue.

 

[Anderson]

For the moment, I will assume that's an error. It's possible that something odd happened with splitting off the NFK service.

 

[afigg]

I figure I can go ahead and elaborate on the general pattern of a year for Amtrak:

-The "good" months, ridership-wise, tend to be March-August. March/April have Easter and spring break to boost ridership, while May-August tend to get lots of vacationing traffic from the warmer months. July has often peaked above the others, but the peak isn't that massive (around 100k over most of the other months in this period) and August doesn't trail too far behind.

-The "bad" months tend to be January, February, and to a lesser extent September and October. January and February have awful ridership due to seasonal issues, and February gets slammed with a shortage of days to boot. September and October would be worse off except that September has Labor Day to help, while October has fall break, Columbus Day, and some other miscellaneous effects to help.

-November and December fall in the middle: You have 2-3 weeks of "bad month" ridership patterns mixed with 1-2 weeks of peak-of-the-peak ridership patterns.

The Amtrak ridership numbers by month follows the same pattern as the domestic commercial air passenger numbers. January & February are the slow months, 46 to 47 million commercial air passengers in 2012 while July & August are the peak months with 58 to 59 million in 2012. The broad peak for air travel is the same as Amtrak from March to August, then lower numbers from September to December, then the lowest numbers in January to February. What we see in the Amtrak monthly ridership numbers charted over the course of the year are American domestic travel patterns.

 

[Bob Dylan]

Looks like the Dreaded UP will Win the OTP Contest Again this Year with the Problems on the Hi-Line Dragging Down BNSF's Record! Down here in Texas BNSF still does a Better Job with the Eagles than UP (Damn Coal/Rock Trains and Long Freights) but I'll hand it to UP for their West of El Paso to LAX Dispatching of the Sunset Ltd. Of course with Winter Coming the OTP Performance Record of all Trains except the Soiuthern Ones tend to Trend Down!