Sunday, August 14, 2016

The Pacific Surfliner continues to impress

LOSSAN's marketing folks are tooting their horn, and quite deservedly so. While they've managed a quite nice 3.2% year over year increase in ridership for the first half of 2016, in an overall relatively gloomy period for Amtrak, where they really stand out is is the double digit increase in ridership for special events. The Del Mar races posted a 12.3% increase over last year with 33,674 passengers in the opening weekend; Comic-Con had a 15.7% increase to 47,319 total passengers over its four days of operation.

While those numbers are for all trains and passengers, not merely those going to the Del Mar races and Comic-Con, they also show the major impact on the service that those events have. From the daily average of 7,983 riders, Del Mar's opening weekend boosted that to 11,225 and Comic-Con 11,830. It's no wonder that the Surfliners, despite their large individual capacity (specially increased as well for these events), have crowding problems during these events when overall ridership is 50% higher! Thanks to the delays with Nippon-Sharyo's new bilevel cars, the Surfliner stands a strong chance of being hit with the same curse as the recent Expo Line extension: Too many riders and not enough cars for them.

Of course, there's also the cynical view: The Surfliner is only getting 8,000 riders a day along a 351 mile corridor with substantial sections of automobile traffic seeing 300,000 AADT and only 16,000 riders out of a 130,000 attendee event. Granted, many of those trips aren't really replaceable by intercity rail and many of the attendees are either too local or out of state for Amtrak to be useful, but as the double-digit growth shows, there is still room to snag more of them. For that, increased frequencies and speed are necessary: Traffic is bad, but the Surfliner is usually worse, especially with consideration of time delay from when the train leaves to when you'd prefer to leave, and increased speed also broadens the market itself by enlarging the catchment area and enticing riders who would not have made the trip, even by car, but would take a train that is car-competitive or faster.

Luckily, it looks like we will have an additional Surfliner frequency later this year thanks to more efficient use of existing equipment. Given the cost reductions already evident against what was budgeted for this year, this run will be cost neutral even if it didn't pick up a single passenger. Obviously we wouldn't run such a train and based both on LOSSAN's estimates and my own, it should be marginally profitable and help contribute to reducing the need for state subsidy. As the Fullerton triple track project, new equipment, and other capacity increasing projects come online, the potential of the Surfliner posting a net profit for the state of California, which may then be invested back into the local infrastructure for a virtuous cycle, grows ever closer.

Saturday, April 23, 2016

Pacific Surfliner budget breakdown

LOSSAN has released a budget presentation from Amtrak which, for the first time that I’m aware of, publicly shows a breakdown of the Surfliner’s budgeted revenues and expenses. For us transit geeks, this is a welcome insight into the busiest corridor service outside of the NEC.

Using the provided Train Fuel and Power figures for the previous fiscal year (FY15) and the approximate number of train miles ran by the Surfliner (1.56 million), we have a fuel expense of $7.55 per train mile, which divides by Amtrak’s budgeted $3.23/gallon (September 2015 Monthly Performance Report) to give us an average fuel consumption of 2.34 gallons per train mile. That’s right where we’d expect it to be based on Metrolink/Coaster performance (which is a bit higher consumption level) and Amtrak’s own reported fleet average fuel consumption of 2.3 gallons per train mile. This does, of course, assume that Amtrak’s average fuel price is what the Surfliner’s fuel is bought at.

While there’s nothing too major about a confirmation of average figures, it’s nice to have because it validates all calculations and critiques based upon those figures. There's always the worry in the back of one’s mind that things are actually quite off from the average figures and that the Surfliner actually has a significantly higher fuel consumption (if single level sets dropped the average down) or significantly lower (if long distance trains brought it up). As it is, we can say with confidence that the Surfliner would, for example, trim a significant amount of time from the schedule while also burning nearly half a million gallons less diesel by using modern rolling stock like the Class 222 Meridian (though I believe EMD and Siemens both claim that they will have similar fuel consumption with the F125 and Charger).

The host railroad fees are a bit interesting. At an average of $6.62 per train mile, I’m wondering who is overcharging Amtrak. The overall system average is somewhere just south of $5 per train mile and Metrolink is receiving $4.33 per train mile. Or rather, they would be if we ignored revenues from the long distance trains, though only the Coast Starlight should contribute much of anything. That means that we’re looking at an average of $8.72 for the remaining miles. My suspicion is that NCTD is responsible for the high charges; per the NCTD budget they receive approximately $9.5 million a year in revenue on the Coastal Rail right of way while expending $4.3 million to maintain it. Some of this money comes from Metrolink, some from BNSF, and a probably minuscule amount from Waco’s Pacific Sun Railway, but a decent chunk will be coming from Amtrak.

Based on the motor coach figures presented in this budget, and comparing with the connecting bus schedules in the Pacific Surfliner timetable, we get an average cost of $3.19 per scheduled bus mile. Since that seemed a bit low at first, and I thought possibly due to some cost-sharing between the Capitol Corridor and Surfliner that I was unaware of for the Santa Barbara to Oakland buses, I also looked at the Capitol Corridor’s budget and scheduled bus connections: $3.23 per mile and a combined average expense of $3.21 per scheduled bus mile. Some additional research shows that I needn’t have fretted: It’s quite comparable to the figures reported by Rimrock, Capital Trailways, and Greyhound.

As I have said repeatedly, bus connections are the best and most cost-effective ways of feeding additional riders into the system and demonstrating support for expansion. After all, how much more cost-effective can you get than completely profitable? Yet there is a dearth of bus routes outside of California and it is rare to see a route study which includes any bus connection, much less a wealth of them such as California has. That being said, while I doubt anyone at Amtrak is reading my blog or twitter feed, there have been a number of new bus routes added over the past year, including one announced the other week connecting the Empire Builder to Rochester, MN, which may indicate that they’re beginning to understand just how useful, and profitable, bus connections are.