Railroad Industry Subsidies of/by Amtrak

From the discussion thus far, it appears that in a universe of under used track capacity, such as was the case industry-wide on A-Day, Amtrak IS paying "fair share' for their track access, train dispatch, and supervision. The small Class II roads, which to me are the Central Vermont and the Rutland (OK they have different names but darned if I can think of 'em), holding Amtrak contracts are probably happy to have 'em, their traffic is light, they can run 'em on time, and in the process pick up a little "pin money'.

Likewise, considering that the Chief, operates some one third of its route over a BNSF secondary line, the same analogy applies.

However, anywhere else Amtrak operates, Ms. Bly's earlier observation regarding where Amtrak goes is 100% on mark, and Amtrak's requirement to operate on schedule often at times against the flow traffic on a road that has adopted a "pipeline' means of operation and interfering with scheduled MofW "work windows', the opportunity cost parameter, or how could we put the 'slot' to our best use, comes into play.

Since "dereg", railroad rates have become proprietary, and are not discussed in venues such as this Forum. Occasionally, a 'sliver' surfaces such as did in Fred Frailey's Nov 2007 TRAINS piece on the Sunset Route:

What makes the Sunset Corridor so valuable are trains like IHIDI. It left Hanjin's terminal at the Port of LA late in the day before, fully loaded - 254 double stacked containers stretching almost 8000 feet - destined for Dallas Intermodal Terminal in Willmer. To run this one train, Union Pacific is probably being paid about $200,000

Since it is a very safe assumption that Mr. Frailey's manuscript was 'signed off' upon by Omaha, I'm surprised that little sliver even found its way into the edited copy.

But from such and with a review of a "handy" SP PTT, we find that LA-Dallas via Sunset and T&P is 1502 miles, or UP is grossing ($200K/1502) $133.16 a train mile.

Obviously IF, and I am in no position to say myself and likely anyone who is wears a big strong muzzle, UP's contractual obligation to host the Sunset has caused an additional train earning that kind of revenue to be turned away, or even delayed to the extent that any performance component of the revenue is affected, then there are clearly opportunity costs associated with handling Amtrak.

Just something to chomp upon with your Starbucks this morning.

To make a valid comparison of Amtrak fees to freight revenues, it's probably worthwhile to start using the operating ratio figures for us to extract profit from revenue. It's often accepted that 80% is a pretty good operating ratio for a railroad so if a train grosses $9/mile, it's safe to make some kind of judgment that points toward $1.80/mile in operating margin. That's less than what Amtrak is paying even after you subtract out estimates for maintenance and dispatching already referenced in the thread.

Now if the revenue figure is more like $100/mile, then the margin is obviously much more than Amtrak is paying.

With regard to the whole concept of "opportunity cost", I'll submit that the figure is "zero" unless it can be conclusively demonstrated otherwise. When a rail line utilized by Amtrak is at capacity, the opportunity cost is borne by Amtrak via the railroad delaying them. It will take a lot for me to believe that any railroad turned down a chance to operate a freight train so they could move the Amtrak train. If that were the case, Amtrak would almost never be late.

Steve touched on a very important point. I have never seen anyone attempt to put a number to the unfunded pension liability the railroads "donated" to Amtrak, but I think it could be much larger than many people realize. It's very possible that it might be so large that most of the freight railroads would have been forced to attempt to discharge it through a bankruptcy proceeding -- assuming, of course, that some measure had been enacted to end the practice of unfunded and unreported liabilities.

courtesy of Mr. Norman and perhaps Starbuck's wi-fi:

Since it is a very safe assumption that Mr. Frailey's manuscript was 'signed off' upon by Omaha, I'm surprised that little sliver even found its way into the edited copy.

But from such and with a review of a "handy" SP PTT, we find that LA-Dallas via Sunset and T&P is 1502 miles, or UP is grossing ($200K/1502) $133.16 a train mile.

Obviously IF, and I am in no position to say myself and likely anyone who is wears a big strong muzzle, UP's contractual obligation to host the Sunset has caused an additional train earning that kind of revenue to be turned away, or even delayed to the extent that any performance component of the revenue is affected, then there are clearly opportunity costs associated with handling Amtrak.

Note that UP has decided to make a major investment (over $1 billion) in double tracking the Sunset Corridor, some of which I witnessed last month on my trip to LA. So rather than wring their hands about Amtrak's silly little Sunset Limited slowing down their money making machine, Union Pacific has decided to make the investment to provide the services the customers desire. And judging by what I saw on the ride to LA , the investment should pay off handsomely.

I can also watch the growth of the Cascade Corridor here in the Pacific Northwest. BNSF seems to be encouraging the growth of both Amtrak and Sound Transit passenger services on its property as long as the needed capacity improvements are made and paid for by the taxpayers. I don't think BNSF is doing it just for the access fees, they're getting a refurbished railroad that can more efficiently handle cargo and passengers in the Pacific Northwest.

If Amtrak is paying $1.00/train mile for access and the Class I is paying $20000/year/mile for MOW then Amtrak is covering about 3.6% of the annual MOW bill. So without considering any external factors like opportunity cost, if Amtrak uses more than 3.6% of the track capacity, the host is subsidizing Amtrak. If Amtrak uses less than 3.6% of the track capacity, then Amtrak is subsidizing the host.

Amtrak's MPR used to report the amount of payments made to host railroads for On Time Performance, (the info hasn't been in any of the last year's reports). BNSF used to receive just under $1,000,000 a month for On Time bonuses. So if we figure that BNSF was getting about $10 million a year for On Time handling and that BNSF was spending about $20,000/mile/year for MOW, then BNSF could maintain about 500 miles of track every year from the On Time bonuses.

Vincent wrote: Note that UP has decided to make a major investment (over $1 billion) in double tracking the Sunset Corridor, some of which I witnessed last month on my trip to LA. So rather than wring their hands about Amtrak's silly little Sunset Limited slowing down their money making machine, Union Pacific has decided to make the investment to provide the services the customers desire. And judging by what I saw on the ride to LA , the investment should pay off handsomely.

for what it's worth, there's an article in today's WSJ that called "a New Era" about rail building in teh US. It notes that the current Sunset hosts 5,000 cars/day but that when this project is finished in 2010 that will double. It would appear that they already have plans for that corridor. It would also appear that if the opportunity costs applied anywhere, it applies to this corridor. Also, the listed the sunset investement at $2 bn.

NellieBly wrote:Okay, this is a complicated subject, but it's one I know a bit about, since I helped Conrail win a compensation case against Amtrak in 1996.

At the time, Amtrak paid less than $1 per *train mile* for track maintenance. THere were other payments for dispatching, plus incentives in some contracts. But let's leave those aside for the moment.

Amtrak argued that they should pay CR only $.030 per train mile for track maintenance. My cost model indicated Amtrak should owe about $1.15. The ICC accepted my conclusion.

Now, railroads generally pay each other a car mile fee for trackage rights, and it's typically around 40 cents a car mile, which suggests that an eight car Amtrak train should pay about $3.20 per TRAIN mile if they are to pay what freight railroads pay each other. Based on Mr. Norman's calculation, I'd guess that's about what they do pay (the difference is probably dispatching cost).

Now, is there an "opportunity cost" to handle Amtrak? Yes, I'd say so, since most of the freight network (certainly where Amtrak runs) is at or near capacity. So Amtrak should probably yield railroads about the same revenue they get from freights, less the costs of loco maintenance, fuel, and crews. Based on some work I've done in the past, I'll opine that track costs are about a quarter of the full direct cost of train operations. So you can work out from there what Amtrak might pay.

I think, at this point, that Amtrak is paying approximately a market rate for access, possibly not including the opportunity cost.

Two caveats -- one is that an Amtrak car shouldn't be as heavy and therefore cause as much wear as a freight car, particularly at speed, as friction, etc. should be reduced at speed.

On the other side of the balance sheet, the logistical difficulties are different for, say, running a single 100 car train vs. 17 amtrak 6-car trains that would amount to 100 cars or so. And the difficulties caused by hosting trains running faster than your trains would be very different depending on the array of track at your disposal.

Gilbert B Norman wrote:

What makes the Sunset Corridor so valuable are trains like IHIDI. It left Hanjin's terminal at the Port of LA late in the day before, fully loaded - 254 double stacked containers stretching almost 8000 feet - destined for Dallas Intermodal Terminal in Willmer. To run this one train, Union Pacific is probably being paid about $200,000

Obviously IF, and I am in no position to say myself and likely anyone who is wears a big strong muzzle, UP's contractual obligation to host the Sunset has caused an additional train earning that kind of revenue to be turned away, or even delayed to the extent that any performance component of the revenue is affected, then there are clearly opportunity costs associated with handling Amtrak.

Just something to chomp upon with your Starbucks this morning.

But of course you are in a position to answer you're own question, as are we all, because the implication of his words is that trains like the IHIDI are not normal. He doesn't write "what makes the Sunset so valuable is that it regularly runs trains like the IHIDI ..."

And this stands to reason, since most trains aren't 8,000 feet, and they aren't double-stacked containers (containers typically being non-commodity, with timing being much more important than on bulk goods like grain, coal, etc. make up so much of the freight being carried.

So we actually know fairly certainly that the freights typically don't get anything like the kind of return in your post.

The Florida Fun Train was charged $20 per train mile and limited to 50MPH on CSX tracks. The slow speeds eventually killed the project, among other things. This was in the mid 1990s.

If Amtrak could get their trains running with an OTP of 95% or better, they could afford to at least double their payments for access fee. Most people would pay an extra 5-10 cents per mile to be on time, all of the time. Over 1000 miles, this is $50.

Steve touched on a very important point. I have never seen anyone attempt to put a number to the unfunded pension liability the railroads "donated" to Amtrak, but I think it could be much larger than many people realize. It's very possible that it might be so large that most of the freight railroads would have been forced to attempt to discharge it through a bankruptcy proceeding -- assuming, of course, that some measure had been enacted to end the practice of unfunded and unreported liabilities.

Didn't these liabilities finally get taken off of Amtrak and handed straight to the federal government last year, or maybe the year before? I seem to remember something about that.

Some information I found at BNSF's website (I couldn't find anything similar on UP's) regarding freight tariffs.

In 2006 the "Average Revenue Per Car" is reported as follows:
Consumer Products (Intermodal/Automotive) $1,017
Industrial Products (Building & Construction Products, Petroleum, Chemicals) $2129
Coal $1186
Agriculture Products $2494

So we can see that an intermodal train generates the lowest revenue on a per car basis, but intermodal is probably also the easiest to handle, justifying the tariff.

those revenue fiqures look like they relate to how well trucking can compete with each catergory . improving speeds for amtrak would also improve speeds for frieght competing with trucks .

One issue that has not received much attention in this discussion is, that as is explained in any entry-level college economics course, an entrepreneur is guided primarily by marginal revenues and costs, that is, the additional cost of producing one more unit of output, and the additional revenue arising from it.

In the days before the industry's revival began in the mid-1980's , the railroads were the carier of last resort, pretty much prohibited from abandoning unprofitable segments of their operations by their public-utility status, and glad to get any additional revenue that came their way.

But when that business showed up, as in the Vietnam "boomlet" of the mid-1960's, a lot of that revenue went to the bottom line very quickly. The other side of the coin was that although adding a few more cars to the daily freight didn't strain the facilities very much, eventually that equipment did wear out and required replacement. The slow strangulation of the eastern trunk lines during the 1950's-60's is the best known example.

But in the light of regulatory reforms and the new energy realities, the freight roads now actually offer a service shippers want to buy, and in some instances, are constrained by limited capacity.

in response to icgsteve's comment:

To my mind the railroads made an assumption, a bet, which like just about every other assumption they have made over the course of my lifetime has proven to be wrong. There are consequences to habitually being wrong, you make deals that turn out to be not good for you.

It should be recalled that the railroads "made their bet" against an opponent which was free to change the rules in the name of political expediency.

The first attempts at rate regulation in the 1890's (in which, by their own admission, the railroads were allowed much influence), "captive" shippers (who had access to only one line) picked up a disproportionate share of the costs. Once this was corrected, lines in highly-competitive territories, usually in the overbuilt Midwest, sometimes had difficulty recovering their costs.

This was eventually "corrected" by the Transportation Act of 1920, which imposed a uniform tariff in which the higher-valued manufactured/miscellaneous traffic was expected to cover the overhead....

at precisely the time in which a new form of competition, the motor carrier, arrived on the scene, able and eager to offer lower, cost-based rates because the right-of-way was government-funded.

Many of the regulars at this forum are aware that the economics of transportation, and even among the various modes and players, are far different from those in a perfectly-competitive economy, and the prospect of increased participation by a public which has previously paid little attention to the subject, and politicians who seldom plan beyond the next election, simply serves to perpetuate a culture of short-sight. In this scenario, the freight roads can hardly be blamed for holding their cards close to the vest, and driving a hard bargain.

David Benton wrote:those revenue fiqures look like they relate to how well trucking can compete with each catergory . improving speeds for amtrak would also improve speeds for frieght competing with trucks .

I think they also relate to the degree of difficulty of handling -- petrochemicals and agriculture require special cars with special degrees of purity -- obviously the food products can't get contaminated, but it may be less obvious that the petrochem stuff requires similar handling Because you can't have chemicals mixing with the wrong things, you've got to clean the cars to a high standard. They also require specialty cars.

BNSF's 4th quarter financial presentation showed that locomotive average "velocity" is just under 300 miles per day and for cars, right at 200 miles per day. My perception is that the industry averages are lower than that and maybe to a large degree (my speculation only). It strikes me that improving those figures depend much less on raising track speed than on other operational optimizations. Besides, how many main rail lines does Amtrak use, anyway, out of the total in the USA?

Which shows increased speed would benefit both passenger and freight .
Amtrak and the freights should be working together for more funding to raise speed and service everywhere .
Freights that are been good hosts should be rewarded with cxapital improvements , those that arent should get a big fat zero .