I ran the numbers again, using the data from the FY 08 report, which reports separate number sets, one including all direct costs, and one excluding allocated shared costs. The same conclusion about long-distance trains holds: in fact, excluding shared costs the non-NEC short-distance services are profitable.

The shared and indirect costs do more to bring down the NEC, which has the fourth highest shared-to-avoidable cost ratio, trailing Keystone, Hiawatha, and Piedmont, and the highest indirect-to-direct cost ratio. Non-NEC short-distance routes in general have a higher shared-to-avoidable cost ratio than long-distance routes and a marginally higher indirect-to-direct cost ratio. The track access fees are there on all routes except Keystone, Wolverine, and the NEC, so they’re not what makes the difference. So regardless of which numbers you choose to include, non-NEC short-distance routes outperform long-distance routes.

What you say about the cost of a small town stop is true, but there’s a caveat. If Amtrak has to run a Chicago-Minneapolis-Seattle train, it has no reason not to stop at Fargo and Grand Forks. However, just running that train incurs a huge operating cost, which ticket revenues don’t recoup. Running the train only as far as Minneapolis means lower track access fees, lower labor costs,

This is sort of the agglomeration of a bunch of different plans.
(1) New Mexico wants to link Albuquerque/Santa Fe up with the most logical intercity connections — namely El Paso and Denver. This is understandable for NM but probably shouldn’t be considered high priority by anyone else.

(2) Colorado wants to establish fast rail in its north-south corridor (Pueblo-Colorado Springs-Denver-Boulder-Fort Collins-Cheyenne, Wyoming). This makes sense and *should* be a high priority.

(3) El Paso wants better connections. A very El Paso-specific desire.

If you add that up and analyze it, you find that the only trouble spot is Pueblo-Albuquerque. South of Albuquerque demands very modest upgrades and would run quite decent speeds with quite decent numbers of riders; similarly north of Pueblo. Unfortunately Pueblo-Albuquerque is mountain railroading *and* the lowest-demand section of the entire proposed route.

Oh, and Alon: “In truth, half of Amtrak’s operating losses in FY 2008 came from just four routes: California Zephyr, Empire Builder, Southwest Chief, and Silver Star. ”

100% False, and I’ll tell you why. You’re depending on one of the bogosities of Amtrak accounting: “fully allocated costs”. The allocation of “costs” in Amtrak accounting is not realistic — essentially it ends up allocating a lot of costs by number of miles travelled, making longer routes look worse.

These numbers are just not true. If you look at *avoidable* costs — real numbers — you find that most of the long-distance trains run very modest operating losses. Amtrak’s operating losses come largely due to an enormous set of fixed, *unavoidable* costs, and not enough trains to spread that cost around on. Amtrak fails to gain economies of scale. Even looking at avoidable costs, the LD trains look worse than they should because track access fees convert what are “capital” costs on Amtrak-owned lines into “operating” costs on freight-owned lines.

And again even some of the truly avoidable costs for LD trains *shouldn’t* be avoidable. The large cost of running a station in Minneapolis/St.Paul is “avoidable” by removing the Empire Builder, but I think you’d agree that that’s stupid; Minneapolis needs train service. If we were running, say, Midwest HSR, then that cost *wouldn’t* be an avoidable cost of running the Empire Builder.

Looking only at *avoidable* costs, the deadbeat trains are (1) the three-times-a-week trains, which really are unusually unpopular (so run it daily or not at all); (2) the Silver Star, which has an outrageous and super-slow routing in Florida (so don’t run trains on STUPID routes).

There is a lot of historic evidence that long, multi-day train routes really need to run daily, and rarely benefit from running more than daily. If you’re thinking of running a second “Silver Meteor” frequency you’re likely better off running a “Silver Star” going to different intermediate towns — provided you don’t end up with the silly routing it currently has. Likewise, if the Empire Builder fills up, the North Coast Hiawatha seems to be a better idea than a second Empire Builder.

Note that small town stops, nowadays, cost very little to operate. (Unlike big city stops like Minneapolis.)

Alon, you also haven’t read the Empire Corridor schedules recently. It’s really two parts: NYC-Albany, with very high frequencies, and consistently packed, and Albany-Niagara Falls, which is twice a day each way (less than Chicago-St. Louis). You have to run the numbers for the two parts *SEPARATELY* in order to get a fair comparison with Chicago-St. Louis. It’s like lumping St. Louis-Kansas City in with Chicago-St. Louis. Financially the results of “Empire Corridor West” and “Empire Corridor South” are very different from each other.

In truth, half of Amtrak’s operating losses in FY 2008 came from just four routes: California Zephyr, Empire Builder, Southwest Chief, and Silver Star. The other long-distance routes supply the other half of the losses; without the long-distance routes, Amtrak would operationally break even in good years. There’s a lot of nostalgia around those trains, reflected in their streamliner-era names, but the only reason they’re around is because they go through states that are overrepresented in the Senate due to their small population, and are a necessary evil as long as the Northeast Corridor doesn’t consistently make enough profit to cover interest and depreciation.

What you say about running hourly trains doesn’t make sense in light of ridership demand. Running empty trains is not a good use of money. Illinois’ increase in ridership is unique: Chicago-St. Louis already performs much better financially than the higher-frequency Empire, Keystone, Pacific Surfliner, and Capitol Corridor routes. Other lines that could support more trains are Heartland Flyer and Piedmont. The long-distance lines are a waste – even in the golden age of rail, the railroads ran their long-distance trains daily. Amtrak tried running multiple long-distance trains – there used to be 6 Chicago-West Coast routes – but discovered it cost so much it cut them down.

What you say about development isn’t true. Putting hourly trains on a route doesn’t automatically revitalize anything. Just look at the failure of BART and the Maryland portions of Metro to create much TOD – and those are electrified lines where trains run every 15 minutes at least. Amtrak lines are even worse. When I took Empire to Buffalo-Depew, which has service to New York every 2 hours, all I saw was a sad little waiting room by a railyard with nothing in sight except parking. For what it’s worth, HSR isn’t a good development tool, either; it’s a transportation tool, shortening trip times and inducing new trips, as well as a profitable business investment.

In general, throwing around infrastructure money doesn’t revitalize run-down areas. This goes back to the New Deal, when the federal government threw billions on building hydroelectric dams in the Tennessee valley, hoping to improve the region’s standard of living. The infrastructure spending didn’t draw any economic activity to the area, and as late as the 1970s, when the government opened a coal mine in the Mississippi portion of the TVA, there were 40,000 people applying for 1,400 jobs. Even today the region remains dirt poor, with none of the rapid development that characterizes much of the rest of the South. While Knoxville may have hydroelectric dams, Raleigh and Charlotte have large knowledge bases, which helps explain why they’re much more successful. If the government wants to help poor regions, it should push for more income support for the poor and more funding for education rather than for money-losing megaprojects.

If adding a new route, like a Front Range line from Cheyenne down to El Paso, adds a lot of connections, it can add many more passengers to the system. The Front Range line is especially ripe with potential connections off its north-south route to three different east-west routes at Denver, Albuquerque, and El Paso.

If that route, and those connections, are supplied, they will draw forth demand from passengers who want to go, say, from Salt Lake City to San Antonio by train, but don’t want to go by way of Chicago to do it. Cumulatively, these new connecting passengers from many different cities and towns can help to fill Amtrak’s trains.

Andrew says Dr Herzog also emphasized the importance of having three or four daily frequencies on the long-distance Amtrak routes, in part to ensure that the wait to transfer from one train to another won’t take more than a few hours. Otherwise it would be impossible to synchronize the schedules to make close connections possible all along the main routes. And they advocate having two trains running express or skip-stop service with at least one local train making frequent stops.

There are other very good reasons to want three or four trains a day on each Amtrak route. Depending on the season, at a minimum one third and up to almost two-thirds of the arrivals and departures on the schedules occur in the dark. For example, Amtrak’s arrivals and departures in Cincinnati and Cleveland are all post-midnight or pre-dawn, and who calls that service?

Of course, multiple trains daily allow return trips between nearby towns and cities. If Amtrak can take you there in the morning and bring you back in the evening, you might take the train. Sadly, few routes allow this now. Think Kalamazoo-Niles/South Bend, for example, but with only three trains a day, or Ann Arbor-Niles/South Bend but only two trains, it’s just not enough. Even with an overnight stay, return trip schedules can be difficult.

Frequent departures give the passenger flexibility too. Airlines are well aware of the importance of this factor, and most try to arrange their flights so that if you miss one, there’ll be another one in a couple of hours, or less. The heaviest routes, like NYC-ATL, offer hourly flights.

I guess it is a given now in thinking about transportation routes that more frequent service will always attract more riders — whether it’s bus routes in Portland, or Southwest Airlines planes between city pairs. or trains on California’s Capital Corridor. The British are aiming for service so frequent on their main lines that the passenger need only “turn up and go.”

Isn’t this also another way of thinking about Boardman’s provocative remark that there are other ways of reducing trip times besides high speed?

A few years ago, a lawyer in Springfield could hope to catch one of three trains to Chicago. Let’s say, making up the schedule, one at 8 a.m., one at 1 p.m., and one at 6 p.m. If he finished his business mid-afternoon, he had to await until 6 p.m. for the next train — or drive or fly. But Illinois added two train to the existing three, and ridership almost doubled. Maybe a new departure added at, say, 4 p.m. made the difference to our lawyer. After all, his TRIP TIME was not station to station, but from the courtroom to his living room. Hanging around the station waiting for the next departure is “trip time” to him.

Looking at it this way, the quickest easiest way to reduce trip times on every Amtrak route in the country is to add more frequencies. On the basis of the Illinois experience, where 67% more trains gained 90%+ more riders, we’d see a remarkable increase in total passengers — before spending billions on HSR lines.

There is enormous pent-up demand for passenger rail. Of course we’d all like faster trains. But not to lose sight of the need for more trains. Many more trains on Amtrak’s existing routes, as well as added or restored routes like the Pioneer and the Front Range.

Sadly, few besides unitedrail.org seem to be beating the drum for tripling or quadrupling the frequencies on Amtrak’s current long distance routes.

At this site, there’s a lot of attention given to the notion that HSR can help the movement to livable cities and lowering the need for autos in our cities, thereby reducing demand for energy and lowering pollution of various types.

No doubt it is true that HSR could have a dramatic impact on center cities it serves. So to me, one powerful reason to support HSR, even watered down to 110 mph, on routes like Chicago-Detroit, Cleveland-Columbus-Cincinnati, Chicago-St Louis, Buffalo-Rochester-Syracuse-Utica-Albany-NYC, and Philly-Harrisburg-Pittsburgh-Youngstown-Cleveland, to name a few, is to help revitalize these rust belt cities.

But standard speed trains can also have a great impact. Let’s get back to that proposed Front Lines train. It does not have to be HSR to have a big impact on livability, energy use, and urban revitalization. Imagine starting service with three trains daily from ELP to ABQ and beyond. Like the Sunset Limited, it would use the same grand 100-year-old station in El Paso that today sees more parked buses than Amtrak passengers. If both these trains ran three times a day each way, the station that now greets two trains a day would see 12 trains. It wouldn’t be underused any more, it would be a lively place.

And those three trains a day between El Paso (and the University of Texas at El Paso) and Las Cruces (and New Mexico State) about 45 miles away? For the price of two more trainsets you could have hourly service, say, 18 trains a day, all feeding into the station on the edge of El Paso’s downtown. The railroad line would transform the somnolent district, now home to offices, courthouses, and museums — emptied out by 6 p.m. The commuters and other passengers could support new retail, restaurants and bars, as well as the streetcar line always in the city’s dreams.

Nowadays there must be dozens of restored and preserved train stations in the US, representing usually some federal funds and a lot of local funds. Unfortunately, in Amtrak’s bookkeeping, and in Congress’ books, these stations are not counted for what they are: an enormous waste of taxpayers’ money — because they are so little used. But I think that many taxpayers and voters do see the waste, and don’t like it.

Yet if frequencies were increased on all Amtrak routes, these many stations, as I suggest for El Paso, would no longer stand as monuments to public waste. They would become new centers of life and business in downtowns all across the land. Huge increases in the number of passengers using the stations each day — in the El Paso example, from two trains to 12 a day, before adding another dozen commuter trains — would slash the cost-per-passenger spent to rebuild them. So almost paradoxically, larger Amtrak subsidies to provide for additional trains would end the waste involved in these many stations that are owned by local governments and transit agencies.

Let me add a political viewpoint. Fort Collins-Denver-Colorado Springs already has the needed concentration of population. No brainer. Pueblo is close to Colorado Springs, and economically depressed, so add it to lock in a majority vote for rail in the Colorado Legislature. Add Fort Collins-Cheyenne and we maybe pick up the support of two WY Senators otherwise likely to oppose any and all HSR or Amtrak funding.

Now, Albuquerque-Santa Fe has a rail success, and Gov Richardson and other leading politicians are ready to support more rail. But ABQ is remote, far from anywhere. The nearest city is El Paso, roughly 300 miles and also remote, like islands in the desert sea. The next closest is Denver, roughly 400 miles (but only 300 miles to Pueblo).

And the second largest city in New Mexico is Las Cruces, near the border and very near El Paso. Commuter or regional rail between Las Cruces and El Paso makes as much sense as the RailRunner between Santa Fe and ABQ. But by itself it will be a harder sale to the public and the Lege — it doesn’t link the state capital, it seems to benefit a city in another state, it won’t grab tourists. But fold the needed right-of-way improvements into a more glamorous Front Line route reaching all the way up to Cheyenne and you can put stars in their eyes.

So parts of this line make a compelling case for passenger rail, while other stretches seem empty. What’s the problem? Fast trains and plenty of them running Cheyenne-Denver-Colorado Springs-Pueblo, plenty trains running ALQ-Santa Fe and ELP-Las Cruces. Three or four trains a day through-running ELP-Cheyenne at speeds somewhat better than Amtrak now offers on similar east-west trains like the Sunset Limited. That train obviously goes through w i d e o p e n s p a c e s, but it’s worth having. Even in its slowest sections, a Front Range line will also be a tourist train, attracting riders from afar to the spectacular views of deserts and mountains.

And the Sunset Limited and Front Range trains would feed passengers to each other, for example, L.A.-ELP-Santa Fe-DEN-L.A for tourists, or ABQ-San Antonio-Houston/Austin/Dallas for those committed rail passengers who currently can’t get there from here.

But no politician will get any bonus points these days for proposing a few trains a day running an Amtrak-type service. No, if you don’t talk about HSR you will sound like a fuddy duddy.

So let the New Mexico and El Paso Congressmen talk about this new route as HSR. By the time any money gets spent, it will be for Sunset Limited type service or a little better, and much better within Colorado.

And what’s wrong with that? Everybody here wants to see daily service restored on the Sunset Limited. We were all cheered to hear that the proposals include noticeably speeding up the timetable. So let’s do that first. Then do a similar Front Range train. Meanwhile do HSR somewhere else.

Maybe in 10 or 20 years we return to this subject. Population will have increased in every city from El Paso (home of Fort Bliss — the desert warfare training center of the U.S. Army — ya think that’s gonna go out of fashion?) to Denver, Cheyenne, and in between. We’ll have good stations in place all along the route, some with local transit connections. By that time we’ll have built a constituency of passengers using the route and have a realistic sense of the market.

Then we’ll find that building much faster lines on this route will be cheaper than almost any place in the U.S. Much of the right of way will be on federally owned land, so no eminent domain troubles. And most of it is flat, flat, flat — the Raton Pass a noted exception.

Now, Albequerque (already connected to Santa Fe) yearns for a Denver connection, hence the rather questionable link from there north. The track’s in place to El Paso and fairly straight, so….

Variations of this come up all the time.

‘Course if we don’t get a handle on global warming, and fast, nothing south of Denver will be inhabitable by humans anyway.

Meh, not so much really. Keep in mind that the NYC subways were built to relieve the overcrowding on the Els from the same era as the Chicago ones.