What's the Wisdom of "Profitable Above the Rail"?

[John Bobinyec]

When folks talk of how profitable the Acela trains or perhaps the NE Regionals are "above the rail", it makes me laugh. If only they could run those trains without the rail - but they can't. And the northeast corridor, with all its interlockings, bridges (especially moveable ones), tunnels and catenary, is the biggest money sinkhole there is.

So why do folks speak that way? Why don't we talk about the long distance trains that way? We could just separate out any costs that have anything to do with the rail, such as payments to the host freight railroads. If we could, then we could get a fair comparison between all trains. We might find that the non-NEC trains are doing better than we think.

Please elucidate.

Thanks,

jb

 

[neroden]

For reasons which are not entirely clear, Congress is happy to pay billions of general-fund money for construction and maintenance of roads, but nearly nothing to subsidize operations of intercity buses, trucks, or autos.

It seems that there are close parallels in railroads: it is much easier to get general-fund money for construction and maintenance and tracks than it is to get general-fund money for operations.

(Same is true in air travel, where there are vast subsidies for airports and air traffic control, but very little for operations.)

Don't ask me why this is the political alignment; I don't know. But it's been very consistent for a very long time, and it's true (to a lesser extent) in other countries too. Maybe it has to do with politicians being able to attend ribbon-cuttings and being able to have plaques with their names on them, and roads/tracks/station named after them?

If we were operating in a sensible business fashion, we would look strictly at marginal cost/marginal benefit of proposed changes. Congress seems to not want to do that, though.

 

[Anderson]

John,

We do talk about LD trains that way. The NEC just ends up with a ton of hard-asset overhead that no other route is really burdened with. The other issue at hand is the simple fact that said overhead is disputably allocated; excluding it from the calculations both avoids a fight over those numbers and puts all of the trains on a more-or-less comparable footing in terms of cost allocations.

Edit: To elaborate a bit more, Amtrak shares the NEC with a slew of commuter railroads. In theory, with a competent cost allocation model, the majority of the Class 5/6 maintenance costs should end up in the hands of the commuter roads (NJT has several hundred trains heading into NYP daily vs. Amtrak's 50 or so). a share with Norfolk Southern (which uses the NEC for freight ops), and anything which is incremental for Class 7 should be allocated to the Acela. Good luck with that.

The other option would be to create a synthetic version of the access fees charged on the non-NEC routes by the freights and apply that to the Regionals and Acela under operating expenses and then hold a separate capital account.

And of course, as Nathanael said, there is a standing expectation that trains will cover their operating costs while the government will cover a bunch of capital expenses.

 

[afigg]

When folks talk of how profitable the Acela trains or perhaps the NE Regionals are "above the rail", it makes me laugh. If only they could run those trains without the rail - but they can't. And the northeast corridor, with all its interlockings, bridges (especially moveable ones), tunnels and catenary, is the biggest money sinkhole there is.

So why do folks speak that way? Why don't we talk about the long distance trains that way? We could just separate out any costs that have anything to do with the rail, such as payments to the host freight railroads. If we could, then we could get a fair comparison between all trains. We might find that the non-NEC trains are doing better than we think.

Looking at the surplus or loss above the rail is useful as it separates the cost of operating the trains and maintenance/depreciation/acquisition of the rolling stock from the capital and maintenance cost of the railroad system it runs over. Generally, this is only applies to the discussions about the services that run on the NEC and the connecting corridors (Keystone East, SPG-NHV, SDY/ALB-NYP) . The NEC has a huge backlog of capital needs for modernization, capacity expansion, and to get it to a state of good repair that applies to Amtrak and the multiple commuter agencies that operate on it, which makes it quite complicated to determine what a fair capital share cost is.

The operating surplus "above the rails" generated by the Acela and NE Regionals can (and will) be used to acquire new equipment (ACS-64s, Acela replacements) and, I would expect starting in FY2015, directly contribute to maintenance and capital projects on the NEC.

When we talk about the operating loss or surplus shown in the Route Performance Reports for the services outside of the NEC and its corridors, the fees paid to the host railroads are included in the Total Costs (excl OPEB, APT Asset allocation, and IG costs). So the share of costs for "below the rail" is included in those discussions.

For the LD trains, the fees paid to the host railroads is a small part of their total operating costs. The FY2010 round of PRIIA PIP reports provided useful financial tables with breakdowns of the revenues and expenses for the LD trains analyzed that year. For the CL in FY10, the fees paid to the host railroads were $1.6 million against a total operating cost of $40.1 million (4%). For the CZ, the fees paid to the host railroads were $7.6 million out of a total cost of $101.1 million (7.5%). For the 3 day a week Cardinal, the fees were $1.1 million out of a total cost of $21.9 million (5%). The cost of the crews and maintenance of the equipment are a much larger piece of the total operating cost than are the fees paid to the host railroads.

 

[jis]

It should also be noted that the trackage fees paid to the freight railroads is arguably not one that everyone is happy with, and to some extent it is no less a conjured up number than it would be if/when NEC goes to slot based pricing, which is incidentally how some in the NEC Commission would apparently like to go. Even though it may be an obscurely derived number one can at least in principle negotiate it to something reasonable and agreeable. The situation on the NEC at present is such an odd mish-mash that it is (a) not transparent at all and (b) not sustainable in its present form.

Sent from my iPhone using Amtrak Forum

 

[beautifulplanet]

Thank you for your post.

Some might think it's not such a good wording to talk of profits "above the rail", as in many cases those might just be called operational profits. The operational profits of Amtrak in the Northeast Corridor (or of other railroads in other corridors) might matter to a lot of people.

When folks talk of how profitable the Acela trains or perhaps the NE Regionals are "above the rail", it makes me laugh.

Many people might not laugh when airlines make profits "above the ground" (as they don't pay for the construction of airports either, it's the public doing that). Still the operational profits airlines make, including usage fees for airports, just like many railroads pay a usage fee for tracks and stations, might matter to many people. Many people might not laugh when bus companies make profits "above the road" (as they don't pay for the construction of roads either, it's the public doing that). Still the operational profits bus companies make, even though they don't even pay usage fees for most of the roads they use, might matter to many people.

If only they could run those trains without the rail - but they can't.

Somebody might say, "if only they could fly those planes without the airports - but they can't". Somebody might say, "if only they could run those buses without the roads - but they can't". Some might come to the conclusion that nearly all modes of transportation rely on investments and costs that are not taken care of by the operator of the transportation service.

And the northeast corridor, with all its interlockings, bridges (especially moveable ones), tunnels and catenary, is the biggest money sinkhole there is.

Some might think, possibly there might be some disagreement on what kind of spending in general constitutes a "money sinkhole". Some might think, investing into rail transportation infrastructure that sees more than 11 million passengers annually through Acela and Northeast Regionals alone, plus 39 million annual passengers through MTA Metro-North Railroad's commuter rail New Haven Line, plus the ridership of the other commuter, inter-city and long-distance rail operated on that corridor, is a worthwhile endeavor.

Looking at Amtrak's FY 2015 federal grant proposal offered the following numbers:

All Northeast Corridor Amtrak operations combined (Acela and Regionals) were estimated to make $290 million of a profit.

The proposed investment for the Northeast Corridor was supposed to be $735 million. At least, of those $735 million of investment, $290 million of them were to come from the operational profits, and only $445 million were requested to come from as a federal grant.

Long-distance in comparison was estimated to make a $618 million of an operational loss. Consequently, all $295 million in planned capital investments in long-distance were supposed to come from federal grants. So combined the total federal grants requested for long-distance were $913 million.

(State-supported rail services are yet another category.)

While looking at these numbers, many still think that long-distance services are vital and federal funding should not only be continued, but expanded to improve services. While many think that long-distance rail shouldn't even have it as its aim to make a profit, because it is a public service and it's not the goal of public services to be profitable but to be a service to the people, still many think that increased investments into long-distance and improved services probably would also lead to decreasing the losses in long-distance operating costs.

In the Northeast Corridor, many might have the impression that the situation is a different one, as there already is an operating profit, and more investments here could possibly not only provide a better service but provide even bigger operating profits as well. Amtrak's 2012 Vision for the Northeast Corridor states that if capacity was increased by laying additional tracks, significant investments in infrastructure and rolling stock were made, and additional 220mph high-speed services were introduced connecting Boston and New York in 94 minutes, New York and Washington in 94 minutes, New York and Philadelphia in 37 minutes, Philadelphia and Washington in 54 minutes, then by 2040 there would be a projected annual operating surplus of $1,650 million dollars. So $1.36 billion dollars in additional profits, year after year, compared to the profits estimated for FY2015 in above numbers. That is why many may think, in addition to investing in long-distance rail, which many might think is very needed, that it might also be worth it to invest significantly in the Northeast Corridor, because the return on investment would be so huge with the additional operational profits, besides of course the service to the public, increase in productivity and economic activity that would be consequence and huge benefit as well.

And the northeast corridor, with all its interlockings, bridges (especially moveable ones), tunnels and catenary, is the biggest money sinkhole there is.

On the interlockings, bridges, tunnels and catenery, so the capital investment in the Northeast Corridor, the numbers above showed that the FY2015 requested investment in federal grants are higher than the federal grants for long-distance, but that's also because much of the corridor is owned by Amtrak. The tracks the long-distance services are run on are owned by freight railraods, and still there those $295 million in the FY2015 requested federal grants. And in addition, often there are federally funded improvements for freight railroads that also benefit the long-distance passenger rail operations. So what conclusion one takes away from all of this is probably different from person to person, many might think that both significant investment in the Northeast Corridor is necessary, as well as investments into the vital long-distance passenger rail services.

So why do folks speak that way?

Some might think, the reason people talk about "above the rail" profits or operational profits is because these operational profits seem to matter to them. In regards to rail f.e. these "surplus" operational profits can be used for other investments, as seen in the FY2015 federal grant request proposal, so some might think the operational profits are another source of funding to be able to make investments that one would not make without them. Also for many it might be important that the sheer fact that there are operational profits, means that there are no operational losses which one then has to find a source of funding for, so that might be important as well. Last but not least, for some the fact that there are operational profits is another important indicator that rail is heavily used (just like all other recent ridership records are), that it is an expression of the wish of residents and visitors alike to use rail services, and so it's a strong reason those rail services should not only continue to exist, but be expanded.

Why don't we talk about the long distance trains that way? We could just separate out any costs that have anything to do with the rail, such as payments to the host freight railroads. If we could, then we could get a fair comparison between all trains.

As already noted in this thread, the payments to host freight railroads are small. In addition, Amtrak also paid $15 million annually in rent for the CTDOT and MetroNorth owned tracks on the Northeast Corridor. Capital investments are taking place both in the Northeast Corridor as well as in long-distance rail. Many might think there already is a fair comparison between the lines.

We might find that the non-NEC trains are doing better than we think.

Finally some might think, in case anybody thinks the non-NEC trains are doing better than one thinks, probably depends on what one thought of how the non-NEC trains are doing previously.

While non-NEC trains do not show an operational profit, many might think they are a valuable service that should continue to be funded by the public, and the cost recovery might be better than for a lot of other transportation services that are funded by the public. Just in order to provide one other, very concrete example (despite many more that could be mentioned): many might think that it is surprising how little public discussion there is about the Essential Air Service program that is funded by the United States Department of Transportation. While the Essential Air Service budget still was $131.5 million in 2011, it increased to $241 million in 2014, and already in 2006 a New York Times article stated that average subsidy per passenger is approximately $74 (excluding Alaska flights), while the American Bus Association in 2011 claimed the subsidy can be as high as $801 per passenger. With the recent increases in the program's budget, it might seem to some that data about the current subsidy levels is not publicly available. Still many might note the difference, in the debate about the public funding provided towards the operations of different modes of transportation.

 

[neroden]

I just did a whole crapload of analysis in response to something over in another thread, and concluded that the "long distance" Viewliner trains are being unreasonably loaded up with NEC operations costs from "shared stations". In the "above the rail" department.

There's all kinds of ways to look at things. The one which makes the most sense depends on what question you're asking. It is true that the majority of NEC maintenance should be charged to NJT.

If you really dig through it, though, you find that the Acela income probably does cover most of the NEC's ongoing maintenance costs -- but does not cover the need to catch up with deferred maintenance from the Pennsylvania Railroad's underinvestment in the 1940s through 1970s.

We could just separate out any costs that have anything to do with the rail, such as payments to the host freight railroads. If we could, then we could get a fair comparison between all trains. We might find that the non-NEC trains are doing better than we think.

As for the "long-distance trains", there's east and there's west.

East: once you strip out the overhead (most of which is NEC station costs, I suspect), the east-of-Mississippi network cost less than $28 million to operate in 2013 (estimated off the bar graph by Boardman). This has already improved by 6.5 million due to extra cars on the Cardinal and Auto Train, should improve by around 9.7 million with the new Viewliner sleepers, could be improved by over $1.9 million just by changing the LSL schedule as suggested in the PIP, and will be improved by an unknown amount (maybe $2 million) by retiring the Heritage cars. Adding the Pennsy-CL through cars should be slightly profitable (maybe $0.2 million). A daily Cardinal could improve things by another $4 million and Atlanta cutoff cars by another $1.9 million or more. Add the benefits of various improvements to the network, from the Miami station to the Schenectady station, and the Eastern network should be breaking even by 2017. (Before overhead.) Add in natural growth in demand and it should cover its own equipment costs ("capital charges") too. There are yet more potential improvements to be made.

In short, the entire eastern network can pay its incremental costs, and will do so soon. Amtrak was correct to order new Viewliners, they will fill up and pay for themselves, and Amtrak should order more.

West: The west-of-Mississippi trains are doing a *lot* worse. It really is costing $30 million per year (revenues minus avoidable costs) to run the California Zephyr. I remain suspicious that Denver-Chicago performs a lot better than that -- because it's a corridor with characteristics similar to the LSL or NY-Florida -- but it's never broken out separately so there's no way to tell. There are not as many potential improvements to be made, because on the middle of most of these routes there simply isn't that much demand. The Empire Builder was costing $11 million (revenues minus avoidable costs) in 2012, and that was the best of the bunch. The Texas Eagle (costing maybe $13 million) probably has the most potential for improvements.

The total operational cost (no overhead) for the western network is $133 million a year (compare to the $28 million for the eastern network) and there's no improvements coming in the near future. Extra cars won't help on most of the western routes because there simply isn't the demand for them (except on Denver-Chicago, the Texas Eagle, and maybe the Coast Starlight).

There's just not much you can do with Salt Lake to Reno or San Antonio to El Paso -- they're bad train routes. The Western Senators need to understand that their states are the ones getting the Amtrak subsidies.

 

[Phil S]

It'll take me a while to digest these really great posts but, in the meantime, could someone tell where you all got these $ numbers? I've tried Googling and plowing through Amtrak.com with little success. I did scan through the PRIIA PIP report for 2011 and found essentially no budget $ info. Thanks.

 

[neroden]

Monthly reports, mostly. Some people who were more forward-thinking than I have saved years worth of archives, since Amtrak only keeps one year up on their website at a time. Occasionally annual reports help too. The PIPs vary wildly on what financial information they have, some have a lot, some have almost none.

Then there's a very interesting presentation from Boardman in 2013: http://www.amtrak.com/ccurl/778/373/Amtrak-Covers-88-Percent-of-Operating-Costs-ATK-13-022.pdf

There's a long discussion on some of this stuff at this thread:

http://discuss.amtraktrains.com/index.php?/topic/58411-thoughts-sparked-by-boardman-presentation/

A few of us have been digging into this for waaaaay too long, just out of, I don't know, personal obsession. There are some long-running arguments which are quite impossible to settle definitively from public information.

Some things which do come through clearly:

- the NEC spins off operating money, but has a huge backlog of capital costs

- the Virginia corridors spin off operating money which is used for capital improvements

- the other corridor routes generally have substantial operating losses paid for by the states, *and* have large capital needs

- the eastern long-distance trains come close to breaking even on operations -- though the routes they run on also have a huge backlog of capital costs

- the western long-distance trains have substantial operating losses paid for by the federal government, and have large capital needs too, though it seems like most of those capital needs will simply not get paid for at all

- overhead is massive -- this is just a railroad thing in general -- most costs are on the fixed costs side, rather than the variable costs side (though there are several levels of this: fixed cost if you're running any trains at all, fixed cost if you're running any trains on this route, fixed cost if you're running this train, fixed cost if you're running this car on this train).

 

[afigg]

It'll take me a while to digest these really great posts but, in the meantime, could someone tell where you all got these $ numbers? I've tried Googling and plowing through Amtrak.com with little success. I did scan through the PRIIA PIP report for 2011 and found essentially no budget $ info. Thanks.

The first round of PRIAA PIP reports for FY2010 had financial data tables for revenue and cost components for the LD trains that was generally not provided in the later PIP reports. Those tables were a snapshot of a single FY prior to a new set of cost allocation formulas agreed to by Amtrak, the FRA, and the states for the state subsidies, so some of the components such as equipment maintenance might have changed by a lot,

If you want to see the bottom line numbers for each train service, look up the Route Performance tables and the ridership & revenue reports in the monthly reports. The September monthly reports provide the totals for the complete fiscal year. In a month or so, the September 2014 report should be posted, at which point we can see how each train did for the year. Who knows, maybe Amtrak will post the FY2013 annual and audited financial reports by then.

 

[neroden]

The PIPs most useful numbers are the numbers where they say "What would be the effect on revenue, and what would be the effect on costs, of this change". You can do some projections to see what the effect might be nowadays (what with ticket prices being *way* up since then).

 

[neroden]

The main problem with the monthly reports is "allocated costs".

As I noted, something like $30 million in costs is "allocated" to the Lake Shore Limited, probably mostly for shared stations which would be operating whether or not the LSL was running, as well as a portion of the cost of the reservations system, the Amtrak Police and so on.

This is kind of misleading, because it makes it look as if the LSL is losing money... but if you cancelled the LSL, you might well find that the change was bad for the bottom line, because the costs would remain. I therefore find that it is useful to try to figure out what the losses based on avoidable costs are; the last report which gave us that information was in Boardman's presentation which I linked.

 

[Philly Amtrak Fan]

Monthly reports, mostly. Some people who were more forward-thinking than I have saved years worth of archives, since Amtrak only keeps one year up on their website at a time. Occasionally annual reports help too. The PIPs vary wildly on what financial information they have, some have a lot, some have almost none.

Then there's a very interesting presentation from Boardman in 2013: http://www.amtrak.com/ccurl/778/373/Amtrak-Covers-88-Percent-of-Operating-Costs-ATK-13-022.pdf

There's a long discussion on some of this stuff at this thread:

http://discuss.amtraktrains.com/index.php?/topic/58411-thoughts-sparked-by-boardman-presentation/

A few of us have been digging into this for waaaaay too long, just out of, I don't know, personal obsession. There are some long-running arguments which are quite impossible to settle definitively from public information.

Some things which do come through clearly:

- the NEC spins off operating money, but has a huge backlog of capital costs

- the Virginia corridors spin off operating money which is used for capital improvements

- the other corridor routes generally have substantial operating losses paid for by the states, *and* have large capital needs

- the eastern long-distance trains come close to breaking even on operations -- though the routes they run on also have a huge backlog of capital costs

- the western long-distance trains have substantial operating losses paid for by the federal government, and have large capital needs too, though it seems like most of those capital needs will simply not get paid for at all

- overhead is massive -- this is just a railroad thing in general -- most costs are on the fixed costs side, rather than the variable costs side (though there are several levels of this: fixed cost if you're running any trains at all, fixed cost if you're running any trains on this route, fixed cost if you're running this train, fixed cost if you're running this car on this train).

Nate,

I was looking for data of trains being profitable above the rails and finally came to this presentation. Thank you!

The Palmetto has low R & R but because it does not run overnight and doesn't use sleepers its costs must be very low. According to FY 2014, the Palmetto's "Total Costs excl. OPEB's, APT Asset Allocation and IG Costs" was $28.2 million compared with over $70 million for the Florida trains.

http://www.amtrak.com/ccurl/243/158/Monthly%20Performance%20Report%20-%20September%202014%20(Preliminary%20and%20Unaudited).pdf


I don't see the "profits" listed in this report. Can these numbers be calculated from this report or are there other data available? Each of these trains have the following:

"Contribution / (Loss) excl. OPEB's, APT Asset Allocation & IG"
"Contribution / (Loss) before APT Asset Allocation"
"Fully Allocated Contribution / (Loss)"

None of these show a profit for any of the LD trains. So where does the "profit" come from? Which category is the most accurate?
The Palmetto and Silver Meteor are the two "profitable" trains although I did not fund this in the monthly performance report. They both serve the same route. But the Silver Star which detours into Raleigh "loses money". According to FY 2014, the Meteor brought in more revenue ($44.3M to $41.2M) and "cost" less ($72.9M to $79.9M).

It seems to make sense that the west coast trains are the biggest money losers. Yet the Empire Builder loses about half what the two Chicago to California trains do and less than the Coast Starlight. The EB brings in the second most revenue behind the Auto Train. The three CHI-West Coast trains have the highest "costs".

 

[neroden]

With a background in finance, the only "profit" number which matters to me for an individual train is the profit on variable or avoidable costs. Why? Because a train which is making a profit on variable costs is contributing to overhead and cancelling it would be bad for the bottom line, while a train which loses money on variable costs is a drain on the treasury and cancelling it might be good for the bottom line (but you have to look at network effects). Railroading is a business of extremely high fixed costs and very large overhead; running a railroad economically means taking advantage of those economies of scale.

Boardman's 2013 Viewgraph presentation is the last time we've seen a correct presentation of the profit based on avoidable costs for the so-called long-distance trains.

Every other number you look at has "allocation" involved. The trains are being charged for a percentage of costs of upgrading ARROW, and a percentage of Boardman's salary, for example. A completely senseless and irrational percentage. "Allocation" of overhead is black magic and simply causes confusion. It is much more informative to see the overhead as separate line-items, which explains where the money is really going. We've been spending a lot of time figuring out how to back the overhead out of the numbers Amtrak gives (the *total* overhead quantities themselves show up occasionally in annual reports).

I analyzed another capital-intensive business recently. The profits on an avoidable-cost basis for the rollout in each area are massive, but the overhead is a large lump -- CFO, CIO, Marketing, Sales remain the same regardless of how many areas they roll out to. Therefore in order to get to bottom-line profitability, they have to get big -- they have to have enough areas operating to generate cash to cover that overhead. "Allocating" overhead to individual areas simply obscures the picture.

Anyway, Boardman's 2013 Viewgraph presenation is the most accurate set of numbers published by Amtrak to date. These "allocations" are why a lot of people have accused Amtrak of having suspicious accounting. It *is* misleading accounting. I could assign overhead to make *any* train look unprofitable if I had an ax to grind, and Amtrak's overhead allocation is equally arbitrary. You can't doctor the direct-costs numbers that way.

I have been estimating what the numbers look like for subsequent years by doing math based on the differences between the "fully allocated" numbers reported in annual reports for 2012 (the year used in the Viewgraph presentation) and the "fully allocated" numbers reported in annual reports for recent years, together with the difference in the sum total of overhead (also reported in annual reports). I can't be sure that this is correct, because Amtrak may have arbitrarily moved overhead from one train to another for political purposes, but if Amtrak has just "done what it did last year", it'll be close to correct.

When you look at true, avoidable-costs profitability the so-called long-distance trains continue to line up in roughly the same order as in the Viewgraph presentation, though not quite.

From most to least profitable based on direct costs in FY2012:

Palmetto

Silver Meteor -- (small profit)

Auto Train -- (at breakeven)

Lake Shore Limited -- (small loss)

City of New Orleans

Capitol Limited

Silver Star

Cardinal

Crescent

Empire Builder

Texas Eagle

Coast Starlight

Sunset Limited

Southwest Chief

California Zephyr

From most to least profitable based on direct costs in FY2015, my estimates:

Auto Train

Silver Meteor

Palmetto

Silver Star

Lake Shore Limited

Empire Builder -- (small profit)

Cardinal -- (small loss)

Capitol Limited

Crescent

City of New Orleans

Texas Eagle

Coast Starlight

Sunset Limited

Southwest Chief

California Zephyr

(I've got a whole spreadsheet for doing these estimates, available upon request if I can figure out how to upload it)

So you can see they've shuffled around a bit but not much. In particular, the five with the largest losses (based on direct costs) have remained in exactly the same order.

You are correct that the Meteor is more profitable than the Star.

You are also correct that the Empire Builder is *far* more profitable than the other Western trains, really in a class with the worse Eastern trains and better than the Texas Eagle. It is notable that the EB's financial performance has improved while ridership, revenue, and on-time-performance have dropped: apparently the incentive payments to BNSF for OTP are a big deal, and not making those payments improves the train's financial performance.

Also worth noting: if you add up all the direct-costs losses for *all* the loss-making LD trains *other than* the Sunset Limited, Southwest Chief, and California Zephyr -- leaving out the profit-making trains entirely -- you get roughly $36 million/year. I believe (though I can't prove it, no evidence) that these trains together probably generate more revenue on connecting trains (from passengers switching to other trains at Chicago, Seattle, DC, NY, etc.) than that.

Coincidentally, $36 million is about equal to the sum of the profit (on a direct-costs basis) of the trains other than the Auto Train.

More coincidentally, Auto Train seems to make about $36 million in profit (on a direct-costs basis) by itself.

By contrast, each of the SL, SWC, and CZ costs roughly $18 million/year on a direct-costs basis. There's your Congressional "above the rail" subsidy. It's not much is it? When you add everything together it's about $18 million / year.

I'm not saying there isn't a large federal subsidy for Amtrak -- it just isn't for what most people think it's for. Most of the half-billion dollar Congressional "operating subsidy" is for overhead which can't reasonably be assigned to individual trains. Stuff like rewriting ARROW, keeping the lights on at Beech Grove, etc. It's the equivalent of subsidizing the Air Traffic Control system, airports, airport security, the FAA, etc. etc. It is very different from the subsidies for "Essential Air Service" which go to particular routes.

 

[Anderson]

I've seen past-year numbers from you; I shudder to ask, but...I know there's some strange cost allocation surrounding the Auto Train (it's sui generis within Amtrak for a host of reasons) but how do you come up with a $36m direct operating profit on the Auto Train? The numbers from a few years ago gave it a <$5m profit IIRC...and station renovations notwithstanding, a $30m swing is a lot to think of in the last three years.

 

[neroden]

For Auto Train, that's Amtrak's swing in numbers from 2012 to 2014 -- the fully allocated numbers go from roughly $36 million down to roughly $0. It's quite possible that Amtrak just started allocating less overhead to Auto Train and more to other trains, in which case my assumption that this is a change in the *Auto Train's* direct-costs performance numbers is incorrect.

Amtrak is claiming that Auto Train "almost" breaks even on a fully-allocated basis now, and Amtrak's numbers do say this. Amtrak's annual report specified how much allocated expenses were allocated to the long-distance trains as a group in 2014 (and it's way up from 2012). If the allocated expenses are allocated among the so-called long distance trains in the same proportions as they were in 2012, that means the Auto Train is ahead by about $36 million, because it was being allocated roughly $36 milllion back in 2012.

The other possibility, as I say, is that the expenses formerly allocated to the Auto Train were dumped on some other long-distance train, in which case *those* trains are doing better than I think. (The expenses can't have been allocated anywhere else and can't have disappeared, because as I said, total non-direct expenses allocated to the long-distance trains as a group are up from 2012 to 2014.)

I'd *like* to believe that Amtrak managed to expand capacity on the Auto Train enough to raise revenue to make the difference, and that is clearly part of it. Also it looks like fares went up (priority unloading fees at least). But a large chunk of the difference seems to be on the cost side. I'd like to believe that Amtrak found an actual way to cut direct costs significantly on the Auto Train, and maybe they did, though I couldn't say what it would be. Losing one lounge car attendant doesn't seem like enough, but maybe it was, though I think the timing was wrong. (Perhaps there are more efficient operations at the two stations, both of which were upgraded recently? Perhaps extremely senior people retired and were replaced by brand new people who are paid less? Perhaps CSX is getting paid less? Perhaps maintenance was switched to LCPM? Who knows.)

 

[Alex M.]

Given how expensive the NEC is, I wonder if it was a mistake for Amtrak to take it over. I have seen some opinion pieces that advocate that Amtrak divest itself of this infrastructure to an arm of the FRA that would treat it like the FAA does air traffic control. All user fees and state appropriations in that region would go to maintaining and upgrading this line. Amtrak could take the proceeds it gets for this system that the FRA or whoever paying for it and invest it in the rest of the system.