Weekend Links

» This week’s big news. Open thread in the comments.

Follow my Twitter account (@ttpolitic) to get news in real time.

On The Transport Politic:


  • An animated version of George Gershwin’s Rhapsody in Blue, set to the tune of day in New York City. (via Spacing)
  • New York Subway train operator Dennis Boyd answers questions about his job in a series of questions and answers: One, Two, Three.

Not Fun:

  • The American Association of State Highway and Transportation Officials (AASHTO) reveals its nominees for this year’s American Transportation Awards. Despite claiming to representall five transportation modes: air, highways, public transportation, rail, and water,” the organization’s nominees include only one non-road project: the District of Columbia’s Union Station Bike Center. Depressing.
  • Richland Hills, Texas — a suburb of Fort Worth — threatens to leave local transit district. But the transit district strikes back, threatening to charge the city for expenditures made there over the years.

Building Better Subways:

Intercity Rail:


Image above: Los Angeles’ proposed Westside Subway, from Metro

29 replies on “Weekend Links”

I’m not sure why Virginia is worried about paying for local services. Through April, Washington-Newport News has lost $1.2M (fully allocated) and Washington-Lynchburg has made $1.4M. So Amtrak should owe Virginia $200K, no?

I didn’t know Amtrak is going to install wi-fi on the Regional trains. I said this about the Acela and I’m going to say it again about the Regional: it’s not quite a game changer, but it’s an excellent way of competing with the premium buses, which are now Amtrak’s chief competition on the routes.

NARP is worried most about the legacy corridors.

“… sharp increases in state payments for intercity passenger trains under Section 209 of the Passenger Rail Investment and Improvement Act (PRIIA).

Services for which states do not currently pay will be the toughest to save as PRIIA takes effect. These include about 1/3 of California’s Pacific Surfliners; Michigan’s Chicago-Detroit-Pontiac line; New York Empire Corridor; most but not all trains on Pennsylvania’s Keystone Corridor; one Seattle-Portland train and some of the Chicago-St. Louis trains.”

I assume that California will cover the Pacific Surfliners; it covers a lot of contract trains. Michigan will pay for Chicago-Detroit. Pennsylvania will pay for the Keystone trains and the Pennsylvanian. They’ll grumble, but they’ll pay.

The big bill will be New York. Currently it pays about $4M a year for the Adirondack. If it covers the losses of all the current Empire Service, that bill will go to $4M a MONTH. And the state is not in good financial shape. So there’s a real possibility of service cuts there.

The one they don’t mention is the Hoosier State. That loses $5M a year under the old accounting, perhaps more under the new. Indiana is not known as a hotbed of rail enthusiasm. I very much doubt the Hoosier State will survive past Oct 16th 2013. If the Cardinal route is shortened to make it viable (New York-Cincinnati is an 18 hour run which could be done daily with two trainsets and rescheduled so as to create a connection with Triple-C, if both trains could somehow get into the same station!), then there might be no Indianapolis-Chicago service left.

The current Cardinal is not the optimal route for NY-Cincinnati. It’s almost certainly faster to go through Pittsburgh than through West Virginia, and the intermediate markets are stronger. Even going through Cleveland on the LSL and 3C would be marginally faster than the Cardinal.

On the other hand, if you kill service to West Virginia, it’s probably more useful to redistribute the trains to a good frequency on some Midwestern corridor that’s not horrendously slow.

No easy or obvious solution to the Cardinal, or the half-ass Hoosier. The Amtrak team reportedly is looking at everything, including taking the train to St. Louis! But the Cardinal is not a NYC-Cincinnati train. The NARP Fact Sheet lists its 10 largest city pairs by revenue, and 9 out of 10 are Chicago-someplace. Only NYC-Charlottesville cracks the top 10 by revenue, though D.C.-Charlottesville and D.C.-Cincy also make the top 10 by total passengers.

The Cardinal line is really two parts, NYC-Charlottesville on one end doing fine, and the sorry main section Charleston, WVa-Cincy-Indianapolis-Lafayette-Chicago at the other.

They are joined by a stunningly beautiful section through the New River Gorge National River. See pix at

When the President of Amtrak called the l.d. trains a “national treasure” and compared then to the National Parks, some folks laughed. The Cardinal passes through nearly 50 miles of National Park Service protected lands, alongside white water and forested canyons. I’m not gonna defend Amtrak having 5 or 6 land cruises out West, but I think I’m OK with one land cruise in the East, within a day’s drive of more than half the US population. But it sure needs better marketing.

Meanwhile the population centers, the potential, and the problems radiate from Chicago. The problem is not the hour or two getting from White Sulfur Springs and down through the New River Gorge. The problem is the 8 hours to go 320 miles from Cincinnati-Chicago, and the 4 hours to go 200 miles Indianapolis-Chicago. And on top of that problem, the train leaves Cincy an hour after midnight and leaves Indy at a quarter of 5 in the morning. With a schedule like this, it’s surprising the Hoosier was getting 150 passengers a day.

But Indiana doesn’t get all the blame. I’m comfortable doing average Amtrak speeds though wild river and mountain scenery. But WTF does it take 1 hour 25 minutes to go 50 miles from Charleston to Huntington? Then 3 hours 28 minutes to go another 160 miles to Cincinnati?

Charleston, WVa to Cincinnati, 209 miles, in nearly 5 hours? Take out an hour from that slow zone and the Cardinal could arrive in Cincy before midnight, and in Chicago by 9 a.m. That could help ridership. Well, I’m sure the upgrade would cost a zillion dollars and even Sen Byrd can’t easily hustle up that kind of scratch in these hard times.

Honestly, I read the national treasure bit as saying, “Yeah, we lose money, I don’t care.”

The Chicago-Indy time is terrible, but it’s still low enough that the current equipment is enough for a couple of runs per day. Chicago-Cincy is harder – you’d get one roundtrip per day per train, in which case you might as well bite the bullet and drop the route until there’s HSR there.

The problem is that there’s no good way of serving West Virginia. With Kentucky, you could extend Hoosier State runs to Louisville; the track between Indy and Louisville is fairly straight, which should allow decent speeds. With West Virginia, the only thing that can serve Charleston well is lightweight tilting trains; even a low-speed train like the KiHa 283 could do Chicago-Charleston in about 8 hours, without any infrastructure upgrade.

Out west, the ‘land cruises’ which are left are practically all justified as ‘essential rail service’ to major cities or vital connections between otherwise disconnected networks.
1 – Empire Builder
2 – California Zephyr
3 – Southwest Chief
4 – Sunset Limited/Texas Eagle
5 – Coast Starlight

First, accept that it makes sense to have at least one Midwest-West Coast link, so you have a system rather than two disconnected hunks. Then you quickly find that you can’t cut any of the four, because they each form the only rail link for huge cities (1 – Minneapolis, 2 – Denver & Salt Lake City 3 – Albequerque 4 – Dallas/Fort Worth) and tourist attractions into the West Coast and into the Midwest networks. The Starlight links the Pacific Northwest and the California networks.

When there are more rail lines serving these cities and interconnecting these networks, I expect these trains will go away, but they can’t until then.

Since when is rail service “essential”? On routes like these (given the state of the infrastructure today) buses can provide better service for less cost. Rail service is of course more comfortable, but that’s a luxury, not “essential”.

How high would Amtrak have to raise fares for the long-distance routes to cover costs? If people want a land cruise, can’t they pay for it?

Nathanael, Amtrak is already a disconnected network. The Downeaster isn’t connected to the rest of the system. In that case, there’s a real market that’s being foregone – namely, Maine and New Hampshire to southern New England and New York. With a direct connection, the Downeaster would probably perform as well as the Virginia services.

The above benefit of connection doesn’t occur in the Interior West. All those LD trains are money drains. Many could be redistributed to better service on the corridors they overlap – Chicago-MSP, the Texas Triangle, LA-Phoenix, Chicago-Omaha-Denver. Those could be run at decent frequency with the same equipment that’s now wasted on sleepers, in which case Amtrak could plausibly compete with other modes of transportation while keeping losses to a minimum. With LA-Phoenix and Houston-Dallas this would offer a massive improvement regardless of frequency – Houston-Dallas has no service, and LA-Phoenix serves the Valley of the Sun in Maricopa, the ultimate beet field station.

There’s no rule that says every city has to be connected to the national rail network. It would be folly to connect e.g. Anchorage; the cost is too high. The difference between Anchorage and Salt Lake City or Albuquerque is one of degree, not kind. It’s okay to serve some areas with buses rather than trains, just like, at higher investment levels, it’s okay to serve some areas with low-speed rather than high-speed trains.

“Since when is rail service “essential”? ”

Since the government started funding “essential air service”.

Alon wrote:
“Many could be redistributed to better service on the corridors they overlap – Chicago-MSP, the Texas Triangle, LA-Phoenix, Chicago-Omaha-Denver.”

The thing is, as long as you’re running Chicago-Omaha-Denver, Denver-Salt Lake, and Salt Lake-California, but you don’t have enough equipment for many runs on each, why *not* stick them together?

The main function of the Zephyr is trips to Denver (from either end). It has over 50% turnover at Denver last I checked. The distances from Denver are such that you’re going to want sleepers for the trips from Denver in either direction. You add this stuff up, and eventually you conclude that either you drop all intercity passenger rail service to Denver, or you provide something like the Zephyr.

Dropping all passenger rail service to Denver is stupid; there’s plenty of demand for it. It’s not a tiny little town like Anchorage, it’s *Denver*.

I will agree that the Sunset has serious design problems (especially since it stopped serving Phoenix directly) and doesn’t seem to have the fairly strong justifications and constituencies which the Empire Builder and Zephyr do. The Southwest Chief has similar design problems east of Albuquerque and west of Kansas City.

As for the Downeaster disconnection, it is a problem, but it’s not a major disconnection (what with all the local cross-Boston transportation available). Losing any connection to Denver would be a major disconnection, especially since almost none of the airports people would instead fly to are connected to the passenger train network.

What this blog needs is a post where us commentators can hash over the l.d. trains. But since we are here …

Sorry if my half-joking reference to “land cruises” suggested that I oppose the Amtrak l.d. trains. I think we have to keep them all. In fact, we should be pushing for a few more. No hurry, it will take a few years to get even one back on the tracks. But if we give up a route, it seems very difficult to get it restored.

I happen to really like Denver-C Springs-Pueblo-Trinidad to meet the Chief-Albuquerque-Las Cruces-El Paso. That one FrontRange line would link three of the transcontinental routes, and three or four of the fastest growing metros in the country. It would also solve a lot of Amtrak’s ‘you-can’t-get-there-from-here’ problems. Notably, all of Texas could reach Colorado at long last.

We fall too easily into arguments of ld lines vs short corridors. The correct answer is BOTH. With the right planning, they should overlap and complement each other.

That FrontRange proposal overlaps a potentially very strong corridor Ft Collins-Denver-C Springs-Pueblo, as well as the southern section of the New Mexico RailRunner and a similar potentially strong commuter route Las Cruces-El Paso. The rest of it is wide open desert, much of the land government-owned and much of it flat, without sprawl or many grade crossings to deal with.

On existing ld lines, taking 2 hours out of Richmond-Raleigh will take 2 hours out of the Carolinian and the Silver Star. The work being done with NC’s share of the first HSR grants will take almost another hour out of the Carolinian and maybe half an hour out of the Crescent.

The Upstate portion of the Empire Corridor NYC-Albany-Syracuse-Rochester-Buffalo will get lots of passengers going only one or two segments or the route. But taking an hour or two out of this corridor service will speed the Lake Shore Limited, as well as any new train that might come out of the 3-Cs, say, Cincinnati-Cleveland-NYC/Boston, or the much-needed Detroit-Cleveland-NYC service. Those routes along the Great Lakes and the Erie Canal are truly corridors put end to end.

In Dixie, the Crescent runs Atlanta-Birmingham-Meridian-New Orleans daily. If you overlay a new almost ld train Atlanta-Birmingham-Montgomery-Mobile-Biloxi-New Orleans, you have the beginnings of corridor service Atlanta-Birmingham and Mobile-Biloxi-New Orleans. Restore Sunset Limited service New Orleans-Florida and the three trains a day in this corridor would equal today’s service Raleigh-Charlotte.

There’s costly work around Chicago, where one big project may cut 20-30 minutes out of the Chicago-Detroit corridor timetable. That will invite another train or two running East Lansing-Battle Creek-Kalamazoo-Chicago, or Ann Arbor-Chicago, or just Kalamazoo-Chicago. But the same project will also improve the long distance Lake Shore Limited and the Capitol Limited and any new train that could also run through Cleveland, perhaps to Pittsburgh.

Some of the corridors within the ld lines are very short, like Chicago-Milwaukee. Alon suggests cutting back the Empire Builder to Chicago-Milwaukee-Twin Cities. I’d like instead to get a new medium distance route, Chicago-Milwaukee-Twin Cities-Fargo. The Empire Builder now passes through Fargo in the dark, but with an early departure from Chicago and an overnight in Fargo, you could have daylight arrivals and departures there, and a second frequency Chicago-Twin Cities.

On the other hand, some corridors would be as difficult to operate now as any ld train. The Texas Triangle? San Antonio-Austin-Killeen (Fort Hood)-Waco-Ft Worth-Dallas looks like rich turf. But it ain’t easy at all. That north-south NAFTA mainline is full of freights, full, full, full. The plans to put commuter trains San Antonio-New Braunfels-San Marcos-Austin involve completely new right of way some miles east of the current route, where you’ve already got I-35 and a new toll road under construction. Not easy ROW to get at this late date. It’s been built. So I don’t see speeding up passenger trains Dallas-San Antonio, much less making them run hourly or even every two hours.

In short, ha ha ha, I believe you rarely shrink your way into success or prosperity. Amtrak has tried shrinking for 40 years. It is long past time for Amtrak to grow, ld trains and all.

Nathanael gives three justifications for the ld trains: they provide connection across the system, they provide essential train service, analogous to essential air service and they’re really conglomerations of corridors which happen to abut and can therefore be served by a single train.

Alon has already rebutted the connection argument, but let me just add that the connections are really random; there’s lots of trips that simply can’t be made and there doesn’t seem to be any rhyme or reason why some trips are enabled and others not.

The joined-up corridors argument seems more plausible. If you’re going to run Chicago-Omaha-Denver and Denver-Salt Lake-Bay Area, then you might as well run one train and call it the California Zephyr. But if there wasn’t a California Zephyr the last two segments you’d choose to run a corridor train on are Omaha-Denver or Salt Lake-Reno. 500-600 miles with no intermediate stops to speak of. Denver-Salt Lake is made mildly plausible by passing through Grand Junction, but it would make more sense to run a corridor train just between Denver and Grand Junction than to extend it to Salt Lake. Chicago-Omaha is a sensible corridor, but not along the route the Zephyr actually takes. If you were designing a Chicago-Omaha corridor, you’d either run further north to run through Des Moines and Iowa City or further south through St. Louis and Kansas City. Nathanael says Denver is Denver and should have intercity trains. Perhaps, but a north-south corridor (taking in Pueblo and Colorado Springs) makes much more sense than a corridor east from Denver. The truth is that Amtrak (and its Congressional masters) want to run trains between Chicago and the Pacific Coast and these are the least unjustifiable routes available.

I am somewhat sympathetic to subsidizing train service to transportation-poor communities. But this method of doing it is very much a lottery — if a community happens to lie on a least unjustifiable transcontinental route, it gets a subsidized train; if not, not — and highly inefficient — the trains that are provided are high cost overnight trains. Surely we can spend half a billion dollars a year in subsidy to better effect.

I’m not opposed to Amtrak running land cruises. I do object to subsidizing them. Overnight trains ought to be all sleepers and fares computed to break even. If at those fares there’s not enough demand, then run them less frequently.

Nathanael: whoever talked about SLC-Denver? That corridor is marginal: the route is slow and capacity-constrained, has little upgrade potential, and connects two medium-sized cities with nothing in between.

Woody: the strongest potential Texas corridor, Houston-Dallas, is rather marginal for freight. BNSF has actually proposed to sell it to the state.

Within the Western LD routes, some sub-corridors are good: LA-Phoenix, Chicago-MSP, Chicago-Omaha-Denver, New Orleans-Houston-San Antonio, Dallas-Austin-San Antonio, Dallas-Houston. Others suck: Denver-SLC, SLC-Sac-SF, LA-Albuquerque, MSP-Fargo. (Fargo is a metro area of 200,000. At current national service levels, it should get one daily train, stopping at a reasonable hour. It’s Chicago-MSP that deserves a 2- or 3-hour takt.) The fact that some of those can be done during the daylight doesn’t make them good performers.

This is different from the issue of some Eastern LD routes, which could be improved by improving the corridor services that they overlie. The entire section north of Raleigh is ripe for incremental cuts in trip times, which should marginally improve LD performance. However, the keyword is “marginally.” Modern tilting diesel trains could do Raleigh-Jax in about 7:30, down from 9:40 today; more powerful trains, running at 110 mph (which may be prohibitively expensive for the route), could do it in about 6:30. Given full HSR at both ends it’s viable as a day train, but the mode share is still going to suck. It’d be what New York-Niagara Falls is today.

As for shrinkage: you can improve performance by shrinking where you’re weak and growing where you’re strong. This is in line with SEPTA, which lost ridership with the strike and the cutting of service beyond electrification, but has since regained it and then some by focusing on suburban regional rail. Amtrak could in principle do the same, running 2-hour takts on potentially strong corridors and zero service on weak ones.

I expect at this point that the “states must fund short routes” requirement will be repealed or altered before implementation.

Empire Service would be a *very* serious loss, and I wouldn’t be surprised if Amtrak and/or New York came up with *something* to keep it going. First, I’ll lay bets the lawmakers in Albany are not going to let their trains to New York City go away.

The Western New York trains are a more serious problem. I expect political pressure will get at least one of the three funded (possibly through some gimmick like extending it to be more than 750 miles — we need a second train to Chicago anyway), but I’m not sure about the second one.

The *real* problem is the Maple Leaf — PRIIA managed to put this into the state-supported basket, but it runs through Ontario, and NY is certainly not going to pay Ontario’s share, so who is? It’s going to be really easy to cut this, unfortunately; cross-border trains have proved to be really problematic thanks to customs and immigration.

Connecticut is already planning to take over the Springfield branch, and despite being *really* bankrupt Michigan will probably dig up funds for Detroit service (though possibly not as many per day).

Agree, BTW, that the Hoosier State is very likely to get cut. It’s the single worst-performing corridor train, and by a huge, huge margin. Indiana simply doesn’t care. The runs to Beech Grove can be managed on the Cardinal, or by more likely by deadheading.

The Cardinal may go daily however, so….

I expect at this point that the “states must fund short routes” requirement will be repealed or altered before implementation.

I wonder if Kevin Page was actually preemptively staking out a claim that if one state gets relief from PRIIA 209 then all (including Virginia) should.

Nevermind. I see you went off on a tangential mention of Caterpillar’s purchase of the ElectroMotive Diesel operations. I don’t read that brief news item as “talking up” anybody, or failing to mention anybody else. And if you are grouching against NARP, I can’t think of any public voice that has been talking up the need for more and better passenger rolling stock more consistently than NARP, not even URPA. ;-)

To me, the problem with the EMD news item is that it concedes EMD as a passenger rolling stock manufacturer. It’s not; it’s a freight locomotive manufacturer that occasionally modifies its products to act as substandard passenger locomotives.

And you’d expect that an organization that wants better rolling stock would at least lobby for making better rolling stock legal on US tracks.

I read it differently. EMD was a step-child at GM; the parent corporation was all about cars. Then it was an orphan, bought out by investors probably looking to break the union, plunder the pension fund, lay off all the employees over age 50, and then flip it. Now Caterpillar has adopted EMD, and it may at last have a good home.

The new parent, with expertise in diesels, wants to be a player in the rail market. Somehow it’s a good sign for passenger rail that Cat sees freight rail as a growth market. And who knows where it will go with EMD now.

Reply to Alon. I know Amtrak doesn’t compete with airlines, except perhaps the shuttles. But comparisons may help to shape perception of value. To use Wi-Fi on a Delta flight, I have to pay $14 or so for a service that is WORTHLESS on the ground in Atlanta. To me that’s a RAW deal. If Amtrak provides free Wi-Fi on the Regionals and other trains, that will help passengers feel much better about Amtrak’s ticket prices.

At the New York-Atlanta distances, wi-fi would help, yes, but not that much. Business travelers would still fly, because it’s faster and more frequent. Others would probably still be repelled by the overnight service even if the fares were competitive.

However, services where Amtrak can give the airlines, the buses, and the car a run for their money – Chicago-St. Louis, Surfliner, CC, Empire – would benefit a lot.

Not quite my point. The product perception extends beyond any single route, air or rail. Free WiFi will now become one of the standard product advantages of trains of whatever distance, like wider seats and aisles, the freedom to get up and walk around, and on-board food service. WiFi becomes part of the product feature that you can get some work done on the train when you aren’t driving.

Of course none of these features beat consideration of trip times and frequencies. But it’s good to have them part of the Amtrak brand, even when people are only thinking of how to get to Harrisburg or Boston.

Leave a Reply