Do Long Distance Trains Really Lose Money

[henryj]

I don't know about you guys, but my personal calculation for the operating cost of the LSL each years puts it around $50,000,000-$60,000,000.

I only got around $31m but I could have missed something. It's only a one night out train and under 1000 miles. Amtrak shows revenue for this train of $33m so it makes an operating profit. lol. If you believe that. Amtrak still shows fully allocated costs for this train as $72m regardless. If you look at the Capitol's PIP for similar route and costs it has operating costs about the same and an operating loss of around $8m because it's revenues are only $22m. Amtrak list this trains fully allocated costs as $48m

 

[johnny.menhennet]

I don't know about you guys, but my personal calculation for the operating cost of the LSL each years puts it around $50,000,000-$60,000,000.

I only got around $31m but I could have missed something. It's only a one night out train and under 1000 miles. Amtrak shows revenue for this train of $33m so it makes an operating profit. lol. If you believe that. Amtrak still shows fully allocated costs for this train as $72m regardless. If you look at the Capitol's PIP for similar route and costs it has operating costs about the same and an operating loss of around $8m because it's revenues are only $22m. Amtrak list this trains fully allocated costs as $48m

Yeah, you know, I believe Swadian's figures too. Throw in some amount of money from fuel, add a sprinkle of OBS costs, maybe a pinch of food service, bake it, and pretty soon you have a cake of very accurate cost estimates!

 

[Trogdor]

The FY2011 performance reports lists the Sunsets revenue as $12.7m and total costs as $52.5m. If the train is only running three times a week and you add 4 more to that then the incremental costs would be more than the current operation.

Why do you say that? The costs are not linear. 3x/week operation is inefficient, and requires a lot more relative to the amount of service provided than daily service. So, the incremental costs of running daily service would not have to be more than the (then) current operation.

I think the OBS number is just mis-stated.

Based on what evidence?

If you actually look at the report from which you're quoting, you'd note a few things:

-First, the numbers that you cite are actually net system-wide numbers that account for not only changes to the Sunset/Eagle, but also follow-on changes that would impact other routes under that scenario (such as adding a sleeper to the Capitol Limited). The combined Eagle/Sunset actually had a net savings in OBS costs under that scenario.

-Second, the numbers even for the Sunset are mixed in with the Texas Eagle, because the whole basis of the report is to have the Texas Eagle run as one solid consist Chicago to Los Angeles, and have the Sunset Limited operate as a stub train New Orleans-San Antonio.

When you add up the costs associated with running the Texas Eagle daily to San Antonion, 3x/week (with associated crew costs for the sleeper attendant) to LA (I don't know offhand if the 421/422 sleeper has a single attendant all the way through, or if they change off; that has gone both ways over the years), and the inefficient crew turns that existed with the Los Angeles to New Orleans 3x/week schedule that existed at the time the report was generated, and compare that to the OBS costs of a daily Texas Eagle on a rescheduled LA-San Antonio time, plus the reduced OBS needed for the San Antonio-New Orleans portion, then it absolutely seems within the realm of possibility that the reconfigured service would be, essentially, neutral in OBS costs (a net savings of $90,000 implies that they're probably cutting a total of one job in the entire combination).

Again, all of those numbers were based on the way the Sunset and Eagle operated in 2010, when the report was published. It would be a bit different today with the new Sunset schedule saving OBS crew costs.

If I rework the numbers then the train may lose $4m a year in avoidable operating costs.

If I use the very same document you're sourcing your numbers from, and go about three pages later (for those trying to follow along at home, it's the PRIIA Sunset Limited/Texas Eagle report, page 40 on the document which is actually page 46 of the PDF), then I see that the Sunset Limited in FY09 had revenues of $9.8 million and avoidable costs of $26.2 million, resulting in a net loss of $16.5 million.

While that particular page does not have a breakdown of avoidable costs, we see from the prior chart (page 37 of the document, 43 of the PDF) that avoidable costs include things such as Host Railroad costs, fuel, T&E crews, OBS crews, food & beverage/commissary costs, mechanical costs, station costs, and remaining direct and shared costs. The footnote on the bottom of page 39/45 says that the latter does not include non-avoidable costs such as marketing, advertising, police, environmental, general and administrative, support, etc. So, in other words, no overhead costs.

The chart on 40/46 says that the Sunset Limited's avoidable cost recovery was 37% in FY09.

I challenge you to provide some concrete data (not selectively including certain data, and making up other data) that the Sunset's loss is really only 1/4 that stated in the report.

 

[Met113]

I just priced Amtrak vs. Greyhound for Chicago to Los Angeles on Aug. 14.

Amtrak - Texas Eagle (65 hours) / Southwest Chief (43 hours)

$195

Greyhound (around 48 hours; 2 or 3 transfers depending on route selection)

$139 advance

$185 web only

$206 standard

$229 refundable

I would never wish the Greyhound experience on my worst enemy. RUDE staff, seedy stations (Not every Amtrak station is like CUS or GCT but I have never feared for my life in an Amtrak station up to this point), dirty busses, etc. Spending the extra $56 on Amtrak is WELL worth it.
Another aspect of Amtrak finances I found interesting is labor costs. Compared to similar occupations in the airline industry Amtrak pays its employees almost twice as much as the airlines pay. For example; an Amtrak Ticketing/Customer Service Agent starts out at $17.50/HR. An airline Customer Service Agent starts out at around $9/hr. A flight attendant starts out around making $20K/yr (This figure may actually be a little generous). An Amtrak conductor (Yes, it is a fair comparison. Flight Attendants have significant safety and operation duties as well) starts out around $20/hr which comes out to around $40,000 a year. Amtrak may realize a great deal of cost savings if it slashed employee pay to that of the airline industry.

 

[Trogdor]

Another aspect of Amtrak finances I found interesting is labor costs. Compared to similar occupations in the airline industry Amtrak pays its employees almost twice as much as the airlines pay. For example; an Amtrak Ticketing/Customer Service Agent starts out at $17.50/HR. An airline Customer Service Agent starts out at around $9/hr. A flight attendant starts out around making $20K/yr (This figure may actually be a little generous). An Amtrak conductor (Yes, it is a fair comparison. Flight Attendants have significant safety and operation duties as well) starts out around $20/hr which comes out to around $40,000 a year. Amtrak may realize a great deal of cost savings if it slashed employee pay to that of the airline industry.

I'm not going to address the station ticket agent pay, but as for a conductor, it is not fair to compare the conductor to a flight attendant. Train attendant vs. flight attendant, maybe.

But, for one, conductors have to be rules and territory qualified and are responsible for the operation of the train. They are also responsible for en-route mechanical issues that may crop up.

A more fair comparison would be what a freight railroad pays its conductors, not what an airline pays flight attendants.

 

[jis]

Another aspect of Amtrak finances I found interesting is labor costs. Compared to similar occupations in the airline industry Amtrak pays its employees almost twice as much as the airlines pay. For example; an Amtrak Ticketing/Customer Service Agent starts out at $17.50/HR. An airline Customer Service Agent starts out at around $9/hr. A flight attendant starts out around making $20K/yr (This figure may actually be a little generous). An Amtrak conductor (Yes, it is a fair comparison. Flight Attendants have significant safety and operation duties as well) starts out around $20/hr which comes out to around $40,000 a year. Amtrak may realize a great deal of cost savings if it slashed employee pay to that of the airline industry.

The equation of a flight attendant to a railroad conductor is pretty dubious. The train conductor has every responsibility of a pilot in command of a flight except for the responsibility to actually drive the train. A flight attendant has no such responsibility. In particular they have no operations duty in the sense that the conductor does. A flight attendant does not have the route and the route coverage safety procedures and rules. They have zero navigational or similar responsibilities that requires any route familiarization and knowledge.

 

[AlanB]

Amtrak over the last couple of years has changed how they report things, based in part on changes by Congress. But if we roll the clock back to 2009, here's how a few things looked.

     Revenue    FRA   Direct   Non
LSL    $25.7   $36.9   $21.6   $10.5
CZ     $43.3   $76.9   $19.8   $14.8
Sun     $9.4   $29.1    $9.5    $6.0
SM     $32.6   $43.4   $22.9   $14.1

Now the FRA Defined Costs represents Host Railroad MofW and Performance Incentives, Fuel and

Power, T&E Crew, OBS and Commissary costs, Car and Locomotive maint. and Turnaround

Costs, Commissions, Reservations, Call Centers, Psgr Inconvenience, and Route Stations.

The Direct column represents Total Remaining Direct Costs include Shared Stations, MoE Supervision and Training,

Maintenance of Way, Yard Ops, Marketing and Distribution, Insurance, Terminal Payments,

Procurement/Purchasing, Police/Environmental and Safety, and T&E Overhead.

Non represents Total Non-Direct Costs includes Amtrak Infrastructure Maintenance and System costs.

Just for comparison, the Total Remaining Direct Costs for all of Amtrak breakout like this:

NEC Total $611.9

State Total $627.6

LD's Total $831.4

And total Non-Direct breakout:

NEC Total $253.1

State Total $134.6

LD's Total $172.6

All numbers of course are Millions.

 

[Swadian Hardcore]

I don't know about you guys, but my personal calculation for the operating cost of the LSL each years puts it around $50,000,000-$60,000,000.

I only got around $31m but I could have missed something. It's only a one night out train and under 1000 miles. Amtrak shows revenue for this train of $33m so it makes an operating profit. lol. If you believe that. Amtrak still shows fully allocated costs for this train as $72m regardless. If you look at the Capitol's PIP for similar route and costs it has operating costs about the same and an operating loss of around $8m because it's revenues are only $22m. Amtrak list this trains fully allocated costs as $48m

I'm sorry, but I seriously forgot what I calclated, I just have the final numbers. Still, my numbers are quite a bit below $72,000,000.

Maybe you only calculated the costs per direction, not for both directions. My calculations were for the total yearly cost Amtrak has to pay for this train.

edit: more info

 

[Anderson]

Alan,

Thanks for reminding me of that other format that Amtrak used for showing costs, etc. at one point.

Though a share of the overhead should probably be allocated when bidding out an operation (i.e. a state-sponsored route), I'm wondering if some overhead simply shouldn't be ignored with some trains, if just because those costs wouldn't be saved if the train was axed. An example of this would be that, in the '12-'16 financial plan, there's a "Non-Allocated Capital" line. At least for the sake of analyzing train performance, there probably should be a "Non-Allocated Overhead" line dealing with some of the professional fees, salaries, etc.

Mind you, I say "some" and not "all" for a reason. Amtrak has its "other" category (for example, their "other revenue" category consists of about $480-520 million in income per year), so I think that some of this revenue along with some overhead, benefits, etc. should be allocated "to the system" rather than any given train. For a serious example, Joe Boardman's salary...let's face it, axing the Sunset or the Cardinal wouldn't likely affect his compensation (heck, axing the whole LD system wouldn't) or do much to the allocations to his office, so I don't see why the LD trains should be burdened with those expenses when analyzing their performance.

 

[jis]

Mind you, I say "some" and not "all" for a reason. Amtrak has its "other" category (for example, their "other revenue" category consists of about $480-520 million in income per year), so I think that some of this revenue along with some overhead, benefits, etc. should be allocated "to the system" rather than any given train. For a serious example, Joe Boardman's salary...let's face it, axing the Sunset or the Cardinal wouldn't likely affect his compensation (heck, axing the whole LD system wouldn't) or do much to the allocations to his office, so I don't see why the LD trains should be burdened with those expenses when analyzing their performance.

If Joe Boardmans salary is not going to be paid out of revenues, where the heck else is it going to be paid out of. And if it is going to be paid out of revenues should it not be accounted for in the cost incurred to earn said revenue, and hence apportioned to individual cost line items? It is exactly this sort of smoke and mirrors analysis that we need to get away from IMHO.

 

[Anderson]

Mind you, I say "some" and not "all" for a reason. Amtrak has its "other" category (for example, their "other revenue" category consists of about $480-520 million in income per year), so I think that some of this revenue along with some overhead, benefits, etc. should be allocated "to the system" rather than any given train. For a serious example, Joe Boardman's salary...let's face it, axing the Sunset or the Cardinal wouldn't likely affect his compensation (heck, axing the whole LD system wouldn't) or do much to the allocations to his office, so I don't see why the LD trains should be burdened with those expenses when analyzing their performance.

That's interesting coming from a CPA. If Joe Boardmans salary is not going to be paid out of revenues, where the heck else is it going to be paid out of. And if it is going to be paid out of revenues should it not be accounted for in the cost incurred to earn said revenue, and hence apportioned to individual cost line items? It is exactly this sort of smoke and mirrors analysis that we need to get away from IMHO.

1) I am not a CPA and have never claimed to be one. Please correct that post. (Edit: Henry is the CPA here.)

2) The term here is "gross profit". Basically, the cost of actually running the train is your cost of goods sold (i.e. the tickets). Net profit is what throws in overhead, etc.

3) This reminds me of the depreciation debate. On the one hand, Amtrak depreciates a lot of equipment on its balance sheet and posts a very large loss because of that. On the other hand, new equipment has tended to be provided via direct, dedicated grants and so forth, making at least some argument that either:

A) The grant should be kept on the books as a liability in some form, and drawn down as the equipment is "used up"; or

B) The equipment should be considered to have been acquired for salvage value and the grant/purchase process handled independent of the balance sheet on the grounds that in many cases, Amtrak isn't likely to actually bear the cost of replacement (witness how they're replacing the electric engines).

Edit: Though it doesn't break things down all the way and give me an idea of what goes into gross profit specifically, I pulled Southwest Airlines' financials from Morningstar. Can anybody tell me what shows up as "cost of revenue" in a case like this versus "other operating expenses"?

 

[jis]

1) I am not a CPA and have never claimed to be one. Please correct that post.

Corrected. Sorry about that.

2) The term here is "gross profit". Basically, the cost of actually running the train is your cost of goods sold (i.e. the tickets). Net profit is what throws in overhead, etc.

At least as far as I am concerned at the end of the day it is net profit loss that matters, since that is what determines how much subsidy goes in.

3) This reminds me of the depreciation debate. On the one hand, Amtrak depreciates a lot of equipment on its balance sheet and posts a very large loss because of that. On the other hand, new equipment has tended to be provided via direct, dedicated grants and so forth, making at least some argument that either:

A) The grant should be kept on the books as a liability in some form, and drawn down as the equipment is "used up"; or

B) The equipment should be considered to have been acquired for salvage value and the grant/purchase process handled independent of the balance sheet on the grounds that in many cases, Amtrak isn't likely to actually bear the cost of replacement (witness how they're replacing the electric engines).

What does depreciation have to do with whether Amtrak will pay for replacement or not beats me. You either account for the cost as a one shot deal charged against the year in which it is accrued, or you decide to accrue it over a number of years typically the lifetime of the asset by using depreciation. I see no connection between this and whether Boardman's salary is apportioned or not. It has to do with how you account for the asset in your financial account as opposed to in your cash account.

I also don;t get the point about electric engines. Amtrak is taking a loan and paying for it over a period of time. So? They will be paying down the loan over a period of time. That has nothing to do with depreciation.

 

[afigg]

A) The grant should be kept on the books as a liability in some form, and drawn down as the equipment is "used up"; or

B) The equipment should be considered to have been acquired for salvage value and the grant/purchase process handled independent of the balance sheet on the grounds that in many cases, Amtrak isn't likely to actually bear the cost of replacement (witness how they're replacing the electric engines).

Amtrak will be paying for the replacement electric locomotives and the upgrades to the maintenance facilities out of NEC revenue. Amtrak took out a RRIF loan, but that loan has to be paid back over the next 25 years. So the ACS-64s should be fully subject to depreciation.

As for Boardman's salary and, for that matter, the rest of upper management and the Amtrak Board, of course those costs should be part of the overhead for the LD trains. If the LD train system were to be spun off into a separate operating company, that company would have to have a president, upper management, and a board of directors.

 

[dlagrua]

Some facts to consider:

Amtrak's federal grant, constitutes just 0.05% of federal spending. That's a minuscule number.

Amtrak takes in $1.9 billion in ticket revenue.

The number of visitors, commuters and vacationers that Amtrak brings to the major cities is undetermined but in fiscal year 2011, Amtrak served 30.2 million passengers. Of those 3.8 million alone passed through Washington's Union Station. Point is that Amtrak transports travelers to cities and they spend money on hotels,restaurants,cab fares and it boosts tourism.

Amtrak employs more than 20,000 people. Each of these employees pays Federal Income tax that goes back to the US Treasury. Perhaps 25% of the salary cost is sent back to the government. This is a number that is never discussed.

Many independent contractors are employed to serve Amtrak to supply food, fuel and other services.

Long distance trains carry about 4.8 million passengers 2.9 billion passenger miles ― 44 percent of total Amtrak passenger miles.

Amtrak's loss ( depending on the source) is claimed to average between $12 to $32 per passenger.

Of planes, trains, and automobiles, only trains can accommodate America’s growing need for urban and intercity transportation. The US is poised for a passenger rail boom.

There is upward of 4 million people who depend solely on rail for public intercity travel. Many more million live far from airports and rely on Amtrak as their only source of LD travel.

We can argue about Amtrak accounting methods and the fact that it losses money but do the roads make money, do the airports make money, how about the buses?

Amtrak serves the public interest, is energy efficient transportation and after all what are we paying taxes for?

 

[Anderson]

Ok, you're right...that was my bad on the ACS-64s; I thought it was strictly the Acela-IIs that were obtained with a loan. For some reason, I thought that had been a stimulus appropriation (much like the Viewliner Is back in the 90s were, for example); the talk of doing an equipment order if the CA HSR funding blew up left me a bit backwards. If it's a loan, then I'd attach the two to one another.

As to equipment...yes, I'm arguing that the purchases should probably be handled as a one-shot if Amtrak got a grant (or as a mix if the order is partly a result of a grant and then supplemented with a loan or other cash source). It would be a mess, yes, and it would arguably be a violation of the matching principle of accounting...but when you have cash grants being made by Congress, I believe that allows for things to be handled differently.

I would also note that there is no real attempt to split depreciation costs over routes, even where Amtrak could kick out a ballpark figure (i.e. the Meteor runs with four sets of 3 sleepers, 4 coaches, a diner, and a cafe most of the time, so the train could have depreciation costs for 12 sleepers, 16 coaches, 4 diners, and 4 cafes assigned to it). The stand-in capital charges seem to get slammed around way too much on a year-to-year basis to fill this role, and I don't see any other attempt to assign those costs.

 

[jis]

Ok, you're right...that was my bad on the ACS-64s; I thought it was strictly the Acela-IIs that were obtained with a loan. For some reason, I thought that had been a stimulus appropriation (much like the Viewliner Is back in the 90s were, for example); the talk of doing an equipment order if the CA HSR funding blew up left me a bit backwards. If it's a loan, then I'd attach the two to one another.

No problem.

As to equipment...yes, I'm arguing that the purchases should probably be handled as a one-shot if Amtrak got a grant (or as a mix if the order is partly a result of a grant and then supplemented with a loan or other cash source). It would be a mess, yes, and it would arguably be a violation of the matching principle of accounting...but when you have cash grants being made by Congress, I believe that allows for things to be handled differently.

That is exactly how I feel. Indeed railroads traditionally did not do capital accounting and they accrued entire capital costs in the year incurred. But that got the taxmen in a huff since that effectively reduced the taxable income in the year in which capital work was done. So they went about forcing railroads to do depreciation based capital accounting. Of course for a railroad that has no taxable income to speak of the whole thing is a bit hoaky. Then throw in the politicians and one has an opportunity to grandstand about it, never mind what the net net effect of doing it this way or that is or is not.

I would also note that there is no real attempt to split depreciation costs over routes, even where Amtrak could kick out a ballpark figure (i.e. the Meteor runs with four sets of 3 sleepers, 4 coaches, a diner, and a cafe most of the time, so the train could have depreciation costs for 12 sleepers, 16 coaches, 4 diners, and 4 cafes assigned to it). The stand-in capital charges seem to get slammed around way too much on a year-to-year basis to fill this role, and I don't see any other attempt to assign those costs.

The only logical way IMHO is to split such using proportionate number of hours that an asset sits captive to one train's service and then allocate in same proportion the down time associated with out of service maintenance time etc. The other way to do it is to account for things as if the equipment side was a separate company and the operating side was leasing the equipment from the equipment side. If done right the two ways of doing it should come out with close to similar numbers.

 

[Anderson]

I actually like the idea of treating the equipment operation as a wholly-owned subsidiary that "leases" the equipment to Amtrak and/or the states (as the situation dictates) in the broad vein of the ROSCO setup. In an ideal world, this would mean that just about any sort of passenger rail startup could go to the leasing company and lease a set of equipment (rather than either having to order brand new equipment or trying to shake some old cars free from somewhere...a good example is that I believe at one point the VRE leased some cars from the Sounder, which were in turn acquired from Metra, and which ultimately were purchased from Pullman in the 60s; also, the single-level VRE cars that I've seen pictures of scream "Heritage fleet" to me).

 

[henryj]

Sorry I have been away from the discussion so long, but I had to mow the grass and do some other errands and chores. Even a pensioner has work to do. lol. Thanks to all your good feedback I worked up a spread sheet model for the LD network. Sadly, the Sunset now loses

$10m, but nothing like what Amtrak reports. Something like $103 per rider. All the 14 LD trains post an operating loss of around $119m with revenues of $450m and operating expenses of $569. Amtrak then adds their allocation of $463m. A rather large sum for just 14 trains, in my opinion. I do have further questions for the 'experts' on here. What is the typical T&E crew now days? Is it 3 or 4 or 5?. People have told me the Sunset sometimes leaves Houston with only one person in the engine. Is that true? What is Amtrak using for OBS? Is the dining crew 4? Do they assign an attendant for each superliner coach? Is there an OBS for each lounge car? What about single level coaches? I assume each sleeper has an attendant regardless except for the trans-dorm space. Why don't they use baggage-dorms for the single level trains rather than using up marketable space in the viewliner sleepers? I don't recall the private railroads providing any OBS for coaches(this would be in the 1960's) even on the luxury trains. The T&E took care of that. They did have to have a Pullman conductor for sleeping cars in addition to the attendants. Meals weren't provided. Appreciate any information you can add. Thanks.

I know the railroads experimented with on board nurses and things for the luxury LD trains, but I never saw one. They quit this when profits starting going south. I know the western railroads considered a train successful if it covered it's operating costs and contributed something to overhead. By the time of Amtrak of course these trains were just costing more and more and contributing nothing. However trains crews were still on the old 150miles equals a day thing then. Now days I believe they work an 8 hour day. So there has been improvement on the labor cost front. Also most rural stations are not manned.

 

[Tom]

Go to the monthly performance reports from the beginning of 2011 (At Amtrak.com: About Amtrak, reports & documents and the scroll down to monthly reports). There are excellent GAAP conforming numbers for every train. The only service with distinctly positive earning before depreciation, interest and allocated expenses is Acela. The NE regionals are about break on that basis. The long distance trains (of which I am a huge fan) lose tremendous amounts with just direct expenses. The best performer is Auto-train, whose revenues cover 90% or so of direct expenses. (The detailed monthly numbers by train were suspended in the middle of 2011 as a function of a systems change, but are supposed to be coming back.)

Amtrak publishes a level of detail on a monthly basis that you would never get from a public company. The idea that they keep their numbers secret doesn't hold water. While monthly numbers, like all interim numbers, are unaudited, their conversion into annual, audited numbers does not seem to have been a big problem. The idea that Amtrak uses a fraudulent "creative accounting" again holds no water. If someone want to make that assertion, they would do well to discuss it with the auditors first. Amtrak's accounting system may well have room for improvement, but the basic story is very clear- passenger trains require subsidies.

TRAC, the group here in CA, argues that long distance trains actually make money and that the NEC is the problem. Their explanations sound like a group of 4th graders wading through an annual report. They embarrass themselves.

Amtrak's numbers are readily available, very detailed and conform to GAAP. The allocation of overhead is always an argued about judgment call. Amtrak is not unique in that regard.

David Gunn made a very good point that should be made more often: The NEC does fairly well on a direct costs basis, but requires a great deal of capital investment, which when charged back makes the corridor a big money loser. The long distance trains are a big money loser on a direct costs basis, but require relatively little capital investment and therefore don't look all that bad on a bottom-line basis compared to the corridor trains. The "other short distance corridors/state supported trains" look pretty darn good when you first look at them. However, state support is included in revenues, so the actual operating performance is pretty poor.

The above is clearly my take on the situation, but factual assertions are just that, factual assertions. I have worked as a senior manager in corporate finance for 30 years.

Amtrak is a political animal. While there is an argument that the long distance trains are not essential, if they went away, political support for Amtrak would go away and therefore the NEC would be in big trouble. Fifteen years ago George Will, the conservative columnist, wrote a piece on why the Repubs should stop making Amtrak their whipping boy. His reason was that year after year, in good times and bad, through red and blue administrations, people like their trains and want them funded. So, he said stop rehashing the same argument every year with the same guaranteed result.

 

[Trogdor]

Sorry I have been away from the discussion so long, but I had to mow the grass and do some other errands and chores. Even a pensioner has work to do. lol. Thanks to all your good feedback I worked up a spread sheet model for the LD network. Sadly, the Sunset now loses

$10m, but nothing like what Amtrak reports. Something like $103 per rider. All the 14 LD trains post an operating loss of around $119m with revenues of $450m and operating expenses of $569. Amtrak then adds their allocation of $463m. A rather large sum for just 14 trains, in my opinion. I do have further questions for the 'experts' on here. What is the typical T&E crew now days?

:excl:

My first question is, how do you have a new spreadsheet where you claim to know have a more accurate view than Amtrak about how much LD trains cost to operate, yet you don't even know how many crew members are on board the trains?

Is it 3 or 4 or 5?.

Minimum of one engineer and two conductors (though they can legally go down to a single conductor, my vague understanding is that they actually have to pay that conductor double because he's essentially doing the job of two people). Busier trains can have extra conductors. If the engineer's run is scheduled over six hours in the locomotive, then a second engineer is required.

People have told me the Sunset sometimes leaves Houston with only one person in the engine. Is that true?

Quite possibly. HOS-SAS is only 5 hours, so it would fall under the six-hour rule.

What is Amtrak using for OBS? Is the dining crew 4? Do they assign an attendant for each superliner coach? Is there an OBS for each lounge car? What about single level coaches?

Dining crew varies depending on the train and time of year. I think 4 is the minimum (possibly 3 in a diner-lounge), some trains will have more. Coach attendants, from what I've observed, are typically one for every two coaches. One LSA for the lounge car, except when the Empire Builder has the second LSA serving a limited menu upstairs. I don't pay too close attention to dining car staffing, so I don't have exact numbers on hand.

I assume each sleeper has an attendant regardless except for the trans-dorm space.

Correct.

Why don't they use baggage-dorms for the single level trains rather than using up marketable space in the viewliner sleepers?

Because they don't own any. Turn the clock forward a year or two, and the answer will be, they do.

 

[Nathanael]

A few points:

(1) Transportation accounting is nightmarishly hard to "get right". You really want to look at things totally differently for different purposes.

(2) The most useful thing you can do when looking at operations is to look at the incremental change to the bottom line from making a change, as the PIPs do, but this doesn't provide any way of "comparing trains".

(3) Because trains are massively capital-intensive, your best results almost always come from adding extra people to a train, adding extra cars to a train, and adding more trains per day to the same route. This gets more revenue for given capital costs and for given operating costs.

(4) Which routes exist at all is clearly a political matter, and has been since the 1830s.

(5) Because of the above, and connecting passengers, network effects, the nonlinear effect of more frequencies per day, etc., trying to compare one route to another (unless they run between the same endpoints or something like that) is practically meaningless and a fool's errand. You should instead compare the operations of a particular route to the alternatives to that route which satisfy similar purposes.

Guest_Tom_* made the best generalization which can be made about the "types of trains":

"David Gunn made a very good point that should be made more often: The NEC does fairly well on a direct costs basis, but requires a great deal of capital investment, which when charged back makes the corridor a big money loser. The long distance trains are a big money loser on a direct costs basis, but require relatively little capital investment and therefore don't look all that bad on a bottom-line basis compared to the corridor trains. The "other short distance corridors/state supported trains" look pretty darn good when you first look at them. However, state support is included in revenues, so the actual operating performance is pretty poor."

 

[Green Maned Lion]

On Amtrak's LD trains, with the exception of the Auto Train (obviously) and maybe the CL, the majority of coach passengers do not travel from endpoint to endpoint, but travel between the various other stations. Perhaps the same could be true for Greyhound.

.

I'm sure the same is true for Greyhound, because whenever I've priced a long distance trip, you have to changes buses a lot (and at inconvenient times, too.)

If they had a lot of passsenger demand for longer trips, they'd have through buses.

Greyhound isn't in the passenger transportation business. Expecting them to operate on that basis is an exercise in faulty logic. Most of their income comes from fast-delivery, LTTL cargo.

Now, to the discussion at hand. Like HenryJ I have quite a bit of accounting training. In fact, I am 3 credits and a CPA exam away from being able to call myself a CPA. So I have some idea from where I speak.

In every business, one has different kinds of costs. I know most of you know this, but bear with me. Imagine you are Sony. On one assembly line in a factory, you build headphones. They cost a buck or two a piece to produce, lets say. The other assembly line produces plasma televisions, 65", the cost $900 or so each to make. Now lets simplify this and assume that Sony has two plants, each of which produce two items, and one group of executives. There are, of course, the direct costs of assembling each unit. Those are easy to track.

BUT there are costs to operating the factory- its A/C units, its hallway lights, its bathrooms, its lunch rooms, its managerial offices, its shipping department, and so on- which are shared between the two assembly lines. Beyond that, there are costs associated with senior management, internal infrastructure- for instance, Sony probably owns (or leases, whatever) its fleet of trucks- which it also probably maintains in its own garages using its own mechanics- and so on.

These costs exist. You can't ignore them. And thus you need to allocate these costs to each individual product produced by those factories in order to be able to figure out in realistic terms how much the units cost you- and allow you to price them accordingly.

Now lets say that in a year, this factory turns out 20,000 TV sets and 2,000,000 headsets. Let us assume the units cost $900 and $2.50 in direct costs to make. Lets assume the factory has $4,000,000 in overhead cost, and the company has $8,000,000 in overhead costs. Lets assume the other factory also, by chance, produces 2,020,000 units as well.

There is the divide each section equally method, then divide by units produces. So you allocate to this factory $4,000,000 in management overhead costs. You have $8,000,000 for the factory, and you allocate half to each product production team, so that you add $4,000,000 in allocated costs to the TVs, $4,000,000 to the headphones. Now divide by the number of units produced. The "allocated overhead costs" add $2 to the cost of the headsets, and $200 to the cost of the TV sets, using this metric. That nearly doubles the cost of producing the headsets, and adds 22% to the cost of the TVs.

But let us use a different metric- also accounting wise proper. I have 8,000,000 in management expenses, and we produce 4,040,000 units companywide, so management expense per unit is $1.98. I have 4,000,000 in factory expenses and 2,020,000 in production, or also $1.98. So my headsets now cost $6.46 to produce and the TVs cost, uh, $904.46.

By changing my allocations, I have changed the cost of the headsets massively upward- $4.50 to $6.46. And amazingly dropped my TV costs- $1100 to $904.46. But here's the thing- both methods are absolutely legitimate ways to divide those expenses. As Sony, I would indubitably chose the first option, since it most realistically allocates the cost- and figuring out sensible ways to operate my business is in my best interests. But Amtrak... they need to justify to a bunch of morons- Congress- that they deserve to exist. Their accounting is set up to put costs in certain places.

Do the LD trains make sense financially? I have no idea. I assume so- I suspect that outside of the SL and Cardinal, which provide very limited economic benefit, Amtrak more than covers its costs in increased economic benefits- and its resultant revenue to the IRS. But the fact of the matter is, looking at what I've seen, Henry J is actually right- the LD trains do have "bloated overhead"- far too much of the overhead is attributed to them relative to what they actually take.

For example- you can divide overhead by many different metrics- by train, by Passenger mile, by train mile, by passenger count- and more. Which metric you use to allocate costs greatly influences what money goes where.

 

[henryj]

Because they don't own any. Turn the clock forward a year or two, and the answer will be, they do.

Thank you for the informative reply. I used T&E count of 4 and 4 for the diners. I appreciate all the other informative replies also. This has been a very useful discussion for me and I hope for others also. Sorry if I got off on the wrong foot with some of you initially.

jf

 

[TWA904]

The eight levels of management can be reduced to six. I vote for eliminating the

Assistant Superintendents and the General Superintendents.

If Amtrak is still paying commissions to travel agents, that is a cost that can be

totally eliminated. The airline industry no longer pays commissions.

When it comes to the dining car, for those passengers who book sleeping accomadations, a percentage of the fare should be allocated to the dining car to cover cost. Of course, those who only book a coach seat and use the dining car will pay up front as usual. Dining should

always show a profit, so that is a cost center that can be eliminated.

I haven't seen a paper timetable lately, but if they still print one, that is a cost that

can be eliminated. Airlines don't print them any more.

Marketing cost is another area that can be eliminated or reduced. There are plenty of travel

related companies that would pay to advertise in the Amtrak magazine and other areas in

stations and maybe even tastefully done on the trains.

 

[Anderson]

1) I would agree in principle with eliminating a level or two of management. Mind you, there might be some cases where a department does need an additional level for some reason (maybe size, maybe geographic issues), but in general I don't see an issue with this...of course, with the disclaimer that

2) Actually, this one raises a good question: How many tickets does Amtrak sell through agents?

3) I'm pretty sure that the dining operations do get some sort of cost allocation off of the sleepers. As to the dining showing a profit, that depends on your definition of "profit"...do you count the cost of the food and the OBS? Do you add in the cost of having the dining car? Etc.

4) Amtrak does print a paper timetable, and they give away a lot of "pocket" timetables for different routes. Adjusting what is available on this front might not be a bad idea, but I'm a bit iffy about completely axing them both because of ease of reference and because a substantial minority of Amtrak's passengers seems to be old enough to still be somewhat computer illiterate. A longer-term phase-out might be in order over the next 5-10 years, though, as this segment declines.

5) As to advertising in the Amtrak magazine(s), there's plenty of advertising there...and I don't think you want USAir or Megabus doing ads in something distributed on the NEC, for example.