In the last Sunday Train, I talked about the study on Keystone West improvements commissioned by the PennDOT. This study finds that upgrades are expensive, and benefits are modest, in terms of allowing for one or two additional services per day, but at a substantially higher subsidy per passenger mile.

However, this study had a quite peculiar "hole" in the range of options: even though the Keystone East is a Rapid Passenger Rail corridor, electrified and upgraded to 110mph to allow the successful upgrade in frequency and transit speed of the Keystone service between Harrisburg and NYC via Philadelphia ... Rapid Passenger Rail was completely ignored as an option.

This meant that the only speed upgrade that was considered was an Express HSR corridor that was "designed to fail" under the designated criteria, since it would be on a different alignment, and so not pass through the communities between Harrisburg and Pittsburgh currently served by the Keystone West.

While "back of the envelope" calculations suggested that filling in this hole would offer some advantages, it would still give an intercity service requiring operating subsidized for a decade or more.

However, this was all under "status quo" assumptions. What I look at this week is what changes for the Keystone West if we were able to start building out a Steel Interstate system for this country, to shift some of the petroleum-dependent, carbon-emitting pavement-destroying heavy diesel truck long-hail freight onto sustainable powered electrified Rapid Rail Freight. Join me for this much more promising future ... below the fold.

Why Doesn't a Pittsburgh / Philadelphia Rapid Passenger Rail Corridor Break Even

I have reported on a number of different feasibility studies on Rapid Passenger Rail over the years. I have looked at the Chicago to Saint Louis corridor, the Chicago to Iowa "Rock Island" corridor, the Chicago to Columbus via Northern Indiana corridor, the Ohio Triple C corridor, and the Cleveland to Pittsburgh corridor, among others. And a common characteristic of all of them was that, while they required a capital subsidy, they would be able to generate an operating surplus.

This is a key strategic feature to increase the willingness of people to fighting for investment in an intercity passenger rail. So long as a passenger rail service requires an operating subsidy, there is always an appreciable risk that a future budgetary decision will lead to a curtailment or abandonment of the service. We see that this year with the Hoosier State, where IndianaDOT wants to cobble together a passenger rail service by contract with two different companies, Amtrak and a private rail operator, but wants to have its cake and eat it to by avoiding being considered to be the railway, with all of the additional work and liability rules that this entails. The Hoosier State was extended for another month, on a promise of the FRA to have another look at its ruling, but there is no guarantee that the FRA will agree to ignore the actual merits of the argument, and so the extension is no guarantee that the Hoosier State will survive.

We see it with the current budget fight in Oregon, where the Eugene to Portland leg of the (conventional speed) Cascade Corridor is at risk if the Oregon legislature does not come up with sufficient funds to cover the required subsidy.

Now, using the "elasticities" from the Keystone West improvements feasibility study, increases in transit speed on the Keystone West could have a substantial impact on ridership:

So, everything else equal, an additional 15 minute time savings on a four and a half hour schedule would be predicted 6.6% to 7.7% increase in ridership, on top of the ridership gains from the speed improvements and second daily train in Alternative Two, so would push the ridership gain in 2020 from about 59% to about 70%, and in 2035 from 76% to about 88%. A half hour time savings wold push the ridership gain up to about 80% in 2020, and 100% in 2035.

... but for a service that generates 60%-70% of its operating costs from passenger fares, even a 100% gain in ridership in going from one, slower, service to two, faster, services only maintains the status quo as far as subsidy per passenger mile ... and takes about two decades to get there.

Why doesn't the upgrade to Rapid Rail deliver better results? Because of the population distribution along the corridor. The best improvement in ridership occurs when we bring a trip that was over four hours down to one that is under three hours, since trips that are under three hours open up much more opportunities for same-day trips. And same-day trips have strong demand because there is no need to include the cost of lodging as part of the trip.

But the two ends of the Keystone West corridor are Pittsburgh and Harrisburg, PA. In the list of US urbanized areas, Pittsburgh (#27 in 2010) is an excellent Rapid Passenger Rail anchor, with an urbanized population of 1.7m. But Harrisburg (#86) is much smaller, with an urbanized population of 444,000. Even though it is a State Capital, and so would be expected to punch above its weight, even performing 50% better than expected at average would make it like a 700K urbanized area.

And even with the half hour improvement envisioned above, the Keystone West transit would still be four hours, with additional time required to reach the larger population centers of Philadelphia and NYC on the Keystone East.

So compared two Pittsburgh Rapid Passenger Rail corridors that would operate at a surplus, according to the 2000's era Ohio Hub studies (Pittsburgh / Cleveland, OH and Pittsburgh / Columbus, OH), Pittsburgh / Harrisburg is an appreciably longer transit time to a smaller population center.

Looking at Chicago / St. Louis, Chicago / Columbus via Northern Indiana or Chicago / Des Moines, which are similar transit times to Pittsburgh / Harrisburg ... the most obvious difference is the size of the intervening population centers. Champaign-Urbana has an urbanized population of 145K, Fort Wayne has an urbanized population of 313K, Davenport has an urbanized population of 280K. Meanwhile, Altoona PA has an urbanized population of 80K, and all of the other communities served have urbanized populations of under 50K.

Johnstown, PA. Photo by Greg Hume

In short, in any system of competitive grants for capital subsidies to stand-alone Rapid Passenger Rail corridors, the Keystone West would stand behind a large number of corridors that would yield higher economic benefit for the capital investment required, and would not require ongoing operating subsidies to provide three to six{+} Rapid Passenger Rail services per day.
{+ Note that a corridor that can sustain hourly Rapid Passenger Rail service could well justify a much higher cost and much higher potential benefit per mile Express HSR corridor alignment.}

 
Network Economics and the Keystone West as a "gap filling" service

As noted above, these are "back of the envelope" calculations based on assumed response to reduced transit time ("elasticity technically refers to percentage change in impact for a one percent change in cause of the impact). That is why the values are given as a range rather than as a single estimate ... more detailed study would be required to pin down the most likely response, and so the estimates are based on a range of responses that line up with observed responses of passengers and with more detailed ridership models.

The strongest assumption in these estimates is the status quo assumption. The baseline "no improvement" demand for these services is based on population estimates, so the baseline assumption is that the underlying demand for rail travel on this corridor will be constant on a per-capita basis.

Now, most of passenger demand along this corridor is generated by populations along this corridor. Even for trips that do not begin and end on the Keystone West, a large proportion of these are trips that begin and end somewhere between Pittsburgh and NYC on the combined Keystone East and West and the Philadelphia / NYC segment of the Northeast Corridor.

If the trip is outbound from a Keystone West community, that is obviously tied to the populations along the corridor, because those are the people taking the trips. But even if the trip is inbound into a Keystone West community, that is driven by the attractive destinations along the Keystone West, and whether for business or personal travel, those are also largely driven by the populations of the communities that people are traveling to.

However, not all passenger demand is anchored to Keystone West in this way. While there is no longer a direct Chicago service over the Keystone West, there is a Pittsburgh connection with the Capital Ltd between Chicago and DC, so passengers for Philadelphia, New Jersey or NYC from Chicago, northern Indiana, Toledo, and Cleveland can catch the Capital Ltd and change to the Pennsylvanian. For sleeping car passengers, this can be a "single seat" trip, since the eastbound Capital Ltd drops off one of its sleeping cars in Pittsburgh to be attached to the eastbound Pennsylvanian, while the westbound Pennsylvanian returns the sleeping car to Pittsburgh to be picked up by the westbound Capital Ltd.

With connections of 5:05am arrive / 7:30am depart eastbound and 8:05pm arrive / 11:59pm depart westbound, the connection adds an average of ~3hrs to a round trip journey. The connection for these two services cannot be tightened substantially, since the Capital Ltd. is limited by connections with services in Chicago and DC, while the Pennsylvanian cannot depart Pittsburgh much earlier or return to Pittsburgh much later without losing substantial Pittsburgh patronage.

However, if a Rapid Passenger Rail upgrade was meeting Rapid Passenger Rail services from Cleveland and Columbus, the greater travel time reliability on the Keystone West corridor and the opportunity for multiple daily Columbus and Cleveland connections would allow for much tighter connections, as well as allowing some Keystone West services to run through to Cleveland and/or Columbus. For both Central and Northern Ohio traffic, this would make the best rail connection through to Philadelphia and northern New Jersey, and for Columbus it would make for the best rail connection through to NYC.

If at the same time the Northeast Corridor was improved along the lines originally described in Sunday Train in July 2012, repeated here August 2014, there would be an increase in the appeal of rail transport along the Keystone West to Philadelphia, where there would be more rapid connections to points along the NYC than there is today.

However, while these changes from the status quo would drive up demand for rail travel on the Keystone West, the likely impact would be more like 10% to 20% improvement than 100% to 200% improvement. And its improvements in the range of 100% to 200% that would be required for operating break-even for two to three services per day on the Keystone West.

 
The Steel Interstate Alternative

The frame for the study commissioned by the PennDOT is service improvements within the State of Pennsylvania to provide for two to three passenger rail services daily between Harrisburg and Pittsburgh PA, with connections through to Philadelphia and NYC via the Keystone East Corridor.

The frame for a Steel Interstate is, by its nature, much broader in geographic scope. The primarily target of a speed upgrade in a Rapid Freight Rail corridor is to be able to deliver improved reliability and speed of delivery. There will still be a time overhead on most Rapid Rail Freight for collecting the freight onto a train at the origin railhead, and distributing the freight from the train at the destination railhead. The upgrade to (single stack container and light goods train) freight speeds of 80mph-90mph allows this overhead to be more than offset for long-haul freight, allowing faster average transit time from loading dock to loading dock.

This faster average transit time from loading dock to loading dock is what allows Rapid Freight Rail can deliver faster delivery times at the same reliability or better reliability at the same delivery times as competing long haul diesel heavy trucking, which will allow Freight Rail to once more compete in markets that they have been largely locked out of by the subsidies to trucking accompanying the build-out of the Interstate Highway system.

Now, with a target market including 500, 1,000 and 2,000 mile freight shipments, a Steel Interstate is naturally targeting rail corridors that extend beyond the boundaries of a single state. So in the Backbone Campaign Solutionary Rail project, we are proposing establishing a Steel Interstate along the BNSF Northern Transcon, which stretches from Seattle to Chicago through Washington state, the Idaho panhandle, Montana, North Dakota, Minnesota and Wisconsin.

The Northern Transcon is one of two main trunk freight rail corridors between Chicago and the Pacific Northwest, with a mix of single track and double track sections. This corridor is a main source of both oil and coal freight traffic, and so one part of our proposal is to look to find a way so that the Northern Transcon can "use its powers for good, rather than evil".

The Northern Transcon is a good corridor to start out a stand-alone Steel Interstate. However, network economies apply to Rapid Freight Rail as they do to passenger rail, and once we start a Steel Interstate connecting the western Great Lakes to the Pacific Northwest, that will increase the benefit available from extending that Steel Interstate toward the east.

In the Northern Transcon proposal, we envision building the Steel Interstate in two phases. First comes the electrification of the existing rail corridor. Second comes the progressive improvement of freight speeds to establish the Rapid Freight Rail paths over the corridor. This two phase approach is proposed because if the rail electrification is pursued with a not for profit public interest authority, able to borrow at the rate of state revenue bonds, and the electrification infrastructure is publicly owned infrastructure that is exempted from property tax (similar to Interstate Highways), then the rail electrification phase can be self-funding.

Once a Steel Interstate is already in operation, the Rapid Freight Rail corridors can be extended a segment at a time, with electrification and speed upgrade provided in tandem ... similar to the combined electrification and speed upgrade of the Keystone East. And this would open up the possibility of placing the Rapid Freight Rail in the kind of secondary, more lightly used freight rail corridors that are favored for stand-alone Rapid Passenger Rail proposals ... like the proposed Chicago to Columbus passenger rail corridor (September 2013). In Columbus, that corridor could extend through to the idle alignment from Columbus to Pittsburgh ... where it could connect to the Keystone West.

And in Harrisburg, we are now at the starting point for the Steel Interstate from Pennsylvania to Knoxville via the Shenendoah River Valley rail alignments as proposed by the RAIL Solution advocacy group.

Now, the cost of the Rapid Freight Rail upgrade to the Keystone West is higher than the cost of the "third track" alternative proposed by the feasibility study commissioned by PennDOT. The reason, of course, is that in addition to the third track, the entire corridor would be electrified.

But the average upgrade cost for freight running between Chicago and Harrisburg, NW Ohio and Harrisburg, or Columbus and Harrisburg would be substantially lower. And while the costs of the upgrade are higher in the hilly and mountainous terrain of western and central PA ... that is the direct route to the freight destination and origin rich areas of Philadelphia, New Jersey, and downstate New York.

Passenger train operating costs on a Steel Interstate would be 20%-30% lower. At the same time, since the corridor would be electrified, Active Tilt Trains could be used on the corridor, so that instead of trimming fifteen minutes to half an hour off of the trip, it ought to be possible to trim 45 minutes or more. And since the passenger service staff work on an hourly basis rather than per train-mile, the faster trip also means a reduction in costs of staffing a passenger train, so it would be reasonable to expect the operating costs of a single trip to drop by 20% or more.

Drop the operating costs of the service by 20% through a speed upgrade and less expensive electric traction, and the Pennsylvanian at today's speed would have 75%-90% farebox recovery ... and given the upgraded speeds, would operate at a surplus. Combine the network economies of at least one of those services connecting to Columbus and Chicago over a Rapid Rail corridor, and the demand response to an increase number of frequencies, and it would be reasonable to expect two services a day to be able to run without operating subsidy, and three to run at a total subsidy similar to the funding required for the once-a-day Pennsylvanian at present.

 
Conclusions & Conversations

Although the Steel Interstate concept is focused upon freight rail, operating Rapid Passenger Rail on a Rapid Freight Rail corridor through the Keystone West seems to me that it may solve the core problem of this corridor, which is the relatively small populations of the communities that lie along the corridor between Harrisburg and Pittsburgh.

I do wish to stress that this is partly because of the long-standing benefits of a central Pennsylvania alignment for moving freight between the East Coast and the Great Lakes. The main target of a Steel Interstate is to move freight from long-haul heavy diesel trucks to electrified freight rail powered by renewable, sustainable, electricity. Where there is an opportunity to take advantage of the 90mph-110mph Rapid Freight Rail paths to offer an appealing Rapid Passenger Rail service, we should take it.

There still will be a need to fund the construction of Rapid Passenger Rail corridors that have a strong economic case, entirely independent of whether or not the alignment is a prospective Steel Interstate corridor. But when the opportunity presents itself to take advantage of a Steel Interstate Corridor for Rapid Passenger Rail service, I'd argue that we should take it.