NEC Upgrades: Andrew Selden responds to Progressive Railroading
In reference to the story in Progressive Railroading (“Amtrak’s Northeast Corridor upgrades hinge on overcoming political, funding challenges,” August 2013, by Angela Cotey, associate editor)
Dear Ms. Cotey,
I read with interest your story on the Northeast Corridor in the August issue of Progressive Railroading.
It fairly represented Amtrak’s long-running propaganda campaign to extract tens of billions of new taxpayer dollars to lavish on the NEC infrastructure, but did so without offering any critical analysis of Amtrak’s claims, or any opposing views or data. Even the views attributed to Mr. Schank from Eno Foundation offered no additional perspective.
It is what this story didn’t report that leaves the message highly misleading in terms of its public policy implications.
Among these omitted points, these stand out:
Amtrak has “invested” into its NEC fixed facilities something in excess of $40 billion in cost-free federal capital subsidies (cost free in the sense of a zero cost of capital, but not in the sense of opportunity cost associated with diverting that capital away from other, more productive, investment opportunities). As a consequence of that investment, Amtrak’s financial condition has not improved; instead, Amtrak’s annual losses on its audited financial statements have climbed steadily, suggesting rather plainly that Amtrak’s investments have produced a negative rate of return on capital. That is quite an accomplishment.
In return for the $40 billion, Amtrak has built a transportation facility that–according to published data of the US Bureau of Transportation Statistics–carries a market share IN THE NEC for intercity travel (non-recurring trips greater than 100 miles in distance) of less than 2% and probably 1 1/2% or less. Motorcycles produce more transportation (passenger miles) than does Amtrak, according to the BTS data.
Amtrak often touts its modal split between Washington, D.C. and the New York area, but that is a largely meaningless number. Let’s stipulate that Amtrak’s modal split against air carriers in any NEC submarket (e.g., Baltimore-Philadelphia) is 100%. So what? It is the carrier’s contribution to overall non-commuter mobility in its travelshed that matters, and on that measure Amtrak has wasted $40 billion to create an irrelevant service. Indeed, close analysis of Amtrak’s own “ridership” data show clearly that, outside of purely local traffic occurring between Philadelphia and New Haven, ALL of Amtrak’s other riders could easily be accomdated in existing air and road capacity in the NEC.
“Ridership” by the way measures only transaction volume, not output or the actual production or delivery of transportation. Amtrak touts it, but the number is also largely meaningless, because of point 3, below.
Amtrak trains today are not particularly crowded. Amtrak’s own published data indicate that its NEC load factors are steady at about 40%. The branch lines (Harrisburg and Springfield) are closer to 30% and Acela is slightly higher, closer to 50%. Amtrak cannot sell more than half (and on many trains, two-thirds) of its existing inventory. Under that circumstance, why does it make any sense at all, financially or otherwise, to invest anything to increase capacity?
But even these numbers overstate Amtrak’s actual traffic flows in the NEC, because they omit segmentation analysis. We know that Amtrak trains are consistently full or nearly so in the quasi-commuter territory between PHL, NYP and NHV. That being the case, as a matter of simple arithmetic it must follow that to get overall NEC load factors of less than 50% Amtrak’s trains have to be very lightly loaded outside of PHL-NYP-NHV. Again, given this fact, why does it make sense to put any new public money into a facility that has not only a sub-2% share but also grossly under-utilized trains now (despite the “investment” of the $40 billion)?
A load factor that is stuck in the 40% range (with a declining sub-2% share) is plain and unequivocal proof that Amtrak is already over-invested in its NEC facility.
The trends for Amtrak in the NEC are not favorable. That is to say, the dismal results obtained over four decades using a torrent of free federal capital have not improved. Amtrak’s financial losses have worsened steadily, and in fact increased even in the years following the introduction of the Acela service 13 years ago. If Acela worsened Amtrak’s financial results, why in the world does it make sense to be planning still grander HSR schemes?
Amtrak’s NEC market share has eroded steadily even as population growth and mobility have increased. And its load factors have not improved. Even its vaunted ridership data show that Amtrak is underperforming the Pennsylvania Railroad in post-war decades. Growth in transactions as compared to Amtrak’s ridership nadir of the late 1970s proves nothing.
Train traffic congestion in the NEC reflects mostly the overload of regional commuter trains taking advantage of the severely under-priced federal infrastructure of the NEC. (Amtrak’s capital subsidies include somewhere in the range of $200 million a year in direct subsidy to the regional agencies for facilities and capacity that are of no use to Amtrak.) Bring SEPTA’s and NJT’s track rents up to true market rates and the train congestion will abate in a big hurry. And to the extent that Amtrak itself contributes to train congestion, it is largely its own operating practices that are responsible. Amtrak does not need to send four or more short (5-6 car), more than half-empty, NEC trains up the corridor in peak hours when two longer (10-car) trains would suffice, and STILL have sub-50% load factors.
Congestion in the PHL-NYP-NHV commuter markets is not Amtrak’s issue.
I mentioned earlier the challenge of opportunity cost. If you look at Amtrak’s own published data (in its oft-delayed Annual Reports), you will see that of Amtrak’s three operating divisions, measured by actual transportation output, not simply transaction volume, the NEC is the smallest, weakest, and poorest performing of the three. It also costs by far the most to subsidize (your story mentions a figure of $400 million, but the actual annual cost is probably half-again that much when one includes the subsidized cost of all of the NEC’s facilities and operations, and its proportionate share of G&A, headquarters and insurance costs, plus depreciation if you care to include that).
The regional corridors elsewhere (e.g., the Chicago hub, California, and the Pacific Northwest) as a group produce slightly more output than does the entire NEC, and at a trivial fraction of the cost (which means the federal subsidy per passenger or per passenger mile in these markets is lower to much lower than in the NEC).
And the standout is the long distance group, which produces 160% of the output of the entire NEC at less than half the cost (again proving that the federal cost or subsidy per passenger and per passenger mile in the long distance markets is lower to much lower than in the NEC).
The regional corridors have received a modest capital allocation, but the long distance trains–Amtrak’s biggest and most productive, and lowest cost, segment–have received next to no new capital in two decades.
What is the opportunity cost to Amtrak of lavishing 95% of its free annual capital subsidies on its smallest and weakest market, while starving the larger and more productive segments?
It is also worth noting in this context that the long distance trains’ collective load factor is bumping up against the statistical limit of their carrying capacity. Real growth is not physically possible in these markets until new capacity is added, but Amtrak stubbornly refuses to do that. It is also worth noting that most of the long distance network operates as a series of largely discontinuous individual routes rather than as an interconnected matrix or newtwork of origin/destination travel opportunities. (Amtrak’s obsolete business model in the NEC afflicts even that segment with a lack of travel opportunities.) Interconnecting any of these routes (only at constant levels of market penetration) would drive huge increases in demand.
When you consider that the long distance trains have load factors and–in their respective “corridors”–market shares that are nearly double Amtrak’s LF and share in the NEC, the lost opportunity is very large. Amtrak could even earn positive rates of return on capital by investing in growth in its biggest, strongest, best-performing and yet most constrained services.
Based on these data and performance factors, investing anything beyond minimal upkeep into the NEC is irrational. It can be explained solely by purely political factors unrelated to the actual transportation services or opportunities presented by Amtrak’s NEC facility.
If you are at all interested in pursuing these ideas, I would welcome the opportunity to discuss them with you at your convenience.
Andrew Selden, President
Minnesota Association of Railroad Passengers
Vice President, Law and Policy
United Rail Passenger Alliance