Amtrak has gradually cut capacity on long distance trains between the Northeast and Florida and Chicago by 50% over the past few years through frequency and consist reductions. Perhaps if they themselves believed more in their own product, they would be as widely accepted and popular as they were in 1995, not now the case with for example the Silver Star running around with 2 sleepers and 3 coaches in the Florida winter travel season. What they offer now is the bare minimum that keeps red lines on the map and the US Senate believing they have a national system.

Comment by Joe M. Versaggi  on  02/11  at  01:10 PM

Scott Leonard even has a walk on role (literally)
Steve Musen

Comment by STEVEN H. MUSEN  on  02/14  at  09:36 PM

Reply to Paul J. Smith, Jr.’s comments:

You second paragraph is pretty much meaningless:
“If the subsidies are analyzed closely, they show that the average subsidy per passenger and per passenger mile for Amtrak riders is higher than the corresponding subsidies for airline and highway users.”

Of course, the subsidy for Amtrak is higher - the smaller the denominator, the larger the average (as should be known by every school student).

I could not read rest of your comments because your opening argument about averages is meaningless.  You give a great example of “How to lie with Statistics”.

Comment by Ronald W Degray  on  02/27  at  10:08 AM

Numbers may not lie, but their inference in the way they are stated does, which is directly related to the effort to which they were obtained, particularly when one is unable to get what is difficult or incalculable for other modes of transportation, then insist on making comparisons based on amateurish averages and incomplete data to start arguments.

They are stated to make inaccurate conclusions to suit the author’s preconceived notions, nothing less, nothing more. There is far more to highway subsidies than HTF cash flows. Perhaps I should request my township, county, and state send the FHWA a bill every time and everyday there’s a wreck on the Interstate, and levy the entire highway as a ratable to boot as we do freight railroads that charge Amtrak occupancy fees. Ditto for airlines, airports, the FAA, and airport authorities.

Subsidy per passenger is nothing more than a metric measuring typical distance traveled. Strange, isn’t it, that long distance trains like the Chief can have a high subsidy per passenger yet have one of the best farebox recovery ratios of any long distance train in the system. If we treated commuter trains like that, obviously longer distance commuter trains would get the axe first, regardless of usage or efficiency.

Averages across modes give no idea of subsidy because they cannot be put on a comparative basis, and are as nearly as meaningless as dividing by zero.

Comment by Joe M. Versaggi  on  02/27  at  08:27 PM

To claim to know the author’s preconceived notions, without knowing him, is remarkable. 

How someone can draw a conclusion regarding my views without reading them is unfathomable.

Dividing by zero is not nearly meaningless.  It is absolutely meaningless, since it is a non-operative.

Information regarding federal transfers to Amtrak, FAA, Highway Trust Fund, Homeland Security, etc. is available for someone who knows how to find them.  The same thing applies to information regarding the number of passengers carried by Amtrak, airlines, intercity buses, etc. 

The challenge comes in iterating the information when necessary.  For example, I relied on Amtrak’s September 30, 2007 Monthly Operating Report for most of the Amtrak data.  But in the case of airline passengers, as I made clear in my initial post, the reporting period ended on October 31, 2007.  A one month difference is not likely to have a significant impact on the comparative statistics.  Also, in the case of licensed motorists, I had to rely on 2004 data, since it is the latest available.  I used regression analysis to project the number to 2007.  Is it exact?  Of course not, but it is probably reasonable. 

Average passengers or passenger miles are comparative statistics.  Gross numbers are not except to the extent that dollars are a common currency unit. 

The average cost per passenger does not take into consideration the distance traveled.  But the cost per passenger mile, which I included in my initial posting, does consider it.  The cost per passenger mile is the cost incurred to move one passenger one mile.  It is distance correlated.  The more difficult comparison is equating the cost per passenger mile (train, plane, bus, etc.) with the cost per vehicle mile, which is a common benchmark used for highways.  Determining how many people are in a motor vehicle is iffy.  It is safe to say, however, that at least one person is in the vehicle.  Some studies show that the average occupancy rate for a vehicle in Texas is approximately 1.6 persons. 

For the fiscal year ended September 30, 2007, the Southwest Chief lost $44.3 million or 14.6 cents per passenger mile before interest and depreciation.  It had the third best loss per passenger mile of any of the long distance trains behind the Empire Builder and Auto Train.  If loss is how one defines a service, then it is a leader.

Overall Amtrak’s long distance trains lost $440.4 million or 17.6 cents per passenger mile in 2007.  They generated approximately 23 per cent of Amtrak’s total train revenues, but they racked up 142 per cent of the operating expenses before unallocated items, interest, and depreciation. 

Most businesses have shared costs and depreciation.  Most also have interest expense and ancillary charges.  Amtrak is no exception.  These costs are eventually assigned to a product or service line.  Generally Accepted Accounting Principles, which govern accrual accounting, require a reporting entity to do so. 

Depreciation is the amortization of expenditures that were capitalized.  It is expensed to the underlying activity over the expected life of the asset.  The cost of the locomotives and cars used for the long distance trains was capitalized when they were acquired.  Amtrak uses straight-line depreciation.  It depreciates locomotives and cars, the biggest long distance train asset, over 40 years.  The interest associated with the acquisition of these assets is also capitalized and included in the depreciation.  All the long distance locomotives and most if not all of the cars are still being depreciated.

In my assumptions I assigned only 10 per cent of the interest and depreciation to the long distance trains.  However, even if they did not wear any of the interest and deprecation, they would still be money losers no many how the numbers are sliced and diced.

The only recent train off that I can think of was the Three Rivers.  Any expenses associated with discontinuing this train were probably minimal.  Most of the crew members would have been given other assignments.  The equipment would have been moved to other trains.  The stations, for the most part, are owned by the communities that the train served. 

Reducing the number of units of production (passenger seat miles in the case of Amtrak) means that the shared costs are spread over fewer units.  Thus, the remaining units attract a higher amount of the shared costs unless there is a corresponding reduction in the total. 

Adding more trains to reduce the amount of overheads spread to them is not a good strategy.  The current fleet, which is used by a tiny percentage of the traveling public, is underutilized.  And there is little indication that people in significant numbers will ever use them.

Some people believe that Amtrak cooks its books.  They further believe that Amtrak engages in cost shifting to hide the true cost of a segment of it operations.  I frequently hear people say that a portion of the NEC fixed costs (interest and depreciation) and shared expenses (IT, HR, Procurement, etc.) are shifted to the long distance trains.  Those who make this assertion have never shown any proof for it.  It is unlikely as per the next paragraph. 

Amtrak’s financial statements are audited by KPMG, which is one of the largest and most reputable accounting firms in the world.  The KPMG auditors examine the accounting for and allocation of all costs including shared expenses.  If they find that there has been a shift of fixed costs and overheads that cannot be supported by the underlying drivers, they would report an audit exception.  By the same token, they pay careful attention to the classification of accounting items.  If an item that should have been capitalized is expensed or vice versa, they would report an audit finding.  If it was significant or a deliberate attempt to hide the true cost of a service, they would issue a qualified opinion.  In this hypersensitive post Enron era, Amtrak management does not want to go there. 

Highway users receive subsidies or benefits beyond the federal intergovernmental transfers to the Highway Trust Fund.  Determining the amount of them is a challenge. 

In Texas, where I live, motorists pay a state gasoline tax that is used to fund state highways.  However, 25 per cent of the Texas motor fuels tax is allocated to the State Education Fund, so here we have a case of the motorists subsiding public education.  The Department of Public Safety (DPS), which is Texas’ state police, is responsible for traffic control and accident investigations outside of incorporated communities.  The monies to fund the DPS traffic division comes from the motor fuels tax.

County roads are funded by bond issues, for the most part; and most city streets are funded with property taxes.  On the surface one could argue that these financial arrangements are a subsidy for motorists.  But are they?  There are an estimated 204 million licensed motorists in the United States.  Most of them pay the property taxes that are used to fund county roads and city streets.  Thus, they are paying a user fee for the roads and streets, but they are paying it indirectly. 

I have argued elsewhere that a better way to fund all highways and roadways would be through direct user fees (fuel taxes, mileage assessments, tolls, etc.), but given the politics of transport in the United States, it is not likely to happen. 

Far more people fly on the airlines (678 million domestic passengers in 2007) than take the train (25.8 million).  And most adults in the United States drive.  Most airline passengers and motorists pay directly or indirectly federal income taxes.  In the case of airline passengers and motorists, the user base is relatively large, so they are paying most of the subsidy that benefits them.  But in the case of Amtrak, which has a much smaller user base, a greater portion of the subsidy is being underwritten by non-users, many of whom live where there is little or no train service.

Comment by Paul J. Smith, Jr.  on  02/29  at  12:23 PM

In regards to what Paul Smith wrote, the over all the funding for rail/transit, isn’t as great compared to other modes. So even if the figure cited are for only Amtrak, they’re not off by any means in the big picture of things.

Comment by Andrew Dawson  on  03/11  at  11:46 AM

Unfortunately Amtrak is running at cross-purposes through misuse of Revenue Management and on self-fulfilling prophecies. They will state that most growth occurs only on the state-supported corridors, yet they make virtually no attempt to grow their business on the rest of the national system, while consist sizes and capacity on many routes are at record lows. Besides the Sunset Limited, they could recapture much of the Three Rivers business that they killed (that train’s headcount ridership was 89% of the Capitol Limited’s) with a longer Pennsylvanian consist, checked baggage, and interchanging a couple of coaches at Pittsburgh to run through to Chicago. A larger Capitol Limited could then capture some the Ohio and Indiana business lost when the eastbound Lake Shore Limited schedule was made later to lower Chicago hotel bills for misconnects. But Amtrak will simply scream “Yield Management”, claim impossible logistics, and do nothing. Maybe if they focused on metrics like Incremental Accounting rather than Revenue per Train Mile, they could grow their business and reduce their operating subsidy. They do occasionally run the 2 Hoosier State coaches on the Cardinal west of Indy and sell them, but that is the exception, not the rule with the way they operate. Their farebox recovery ratio of about 67% is no better now than it was in the 1980’s when we had more routes, larger consists, and lower fares.

Comment by Joe M. Versaggi  on  03/27  at  07:01 AM