North Coast Hiawatha - Big Sky Passenger Rail Authority (BSPRA)

Gilbert B Norman wrote:The very point you note, Ron, confronted the Incorporators. During 1969-70 when they were developing the Basic System, they of course had the option to choose the Northern Pacific over the Great Northern (both were BN). While the prospect of a greater on-line traffic base from the likes of Billings, Bozeman, Helena, and Missoula, should have been a factor, they (or their consultants Booz Allen and Arthur Andersen) were "fixated" by end point mentality and the Buider handled more end to end traffic than did the Limited.

Mr. Norman has repeated this theory about why the Empire Builder route was chosen over the North Coast Limited route many times in this forum. As has always been the case, choosing any route on endpoint-to-endpoint ridership has no validity whatsoever because since almost all the city pairs that Amtrak designated for service in 1971 had only one train between them. Therefore, with the exception of New York to Florida, there was only one way of getting from endpoint to endpoint (on a through train), and that train, regardless of routing, would logically capture 100 percent of the endpoint-to-endpoint travel. From a post I made earlier this year in response to Mr. Norman's insistence that endpoint-to-endpoint travel has anything to do with the intermediate routing, I offered these actual reasons that the Empire Builder route was chosen:

From the Great Falls (Montana) Tribune on March 23, 1971 reporting on why the route through Northern Montana was selected as the lone Chicago-Seattle Amtrak service:

“The federal agency made its decision, it said, on six points:
-Ridership, defined as the current number of rail passengers using a route;
-Population, most of which in Montana is located in the south;
-Competition, transportation available over a particular route;
-Existing railroad facilities;
-Profitability, and
-Need.”

The June 1971 issue of TRAINS magazine, in an article entitled, “The Riddle of Railpax” states, “ Railpax selected its routes and schedules on the following criteria: current ridership and trains per week; operating costs; adequacy of other travel modes; on-line population; and physical characteristics of track and equipment.”

The book “Amtrak in the Heartland” by Craig Sanders (2006) states, “Railpax incorporators were to choose routes using the following criteria:

-Current patronage and number of trains per week
-Current operating costs
-Adequacy of other travel modes
-Population along the route
-Physical characteristics of track and equipment”

The Great Falls Tribune version varies a bit from the other two, but in no case is “endpoint” ridership mentioned.

Frankly, I’m at a loss as to why this would a consideration, because doing do seems illogical. In 1971, much like today, most of the ridership of long distance trains was not from endpoint to endpoint. Why would it make any difference to endpoint riders how they got there? Would they not simply ride the one remaining train? (And, in the case of Amtrak on May 1, 1971, there was only one long distance train between any two endpoints, with the exception of New York-Florida service.) If the endpoint-to-endpoint ridership did prefer a specific train, then why choose the Empire Builder route over the North Coast Limited route (or any other)? The Empire Builder and North Coast Limited both offered similar services. The North Coast Limited actually traversed a more scenic route. The Empire Builder was a bit faster, but not much in the end of its pre-Amtrak years as it was consolidated with the North Coast Limited between Chicago and St. Paul and between Pasco and Portland which increased the overall running time so that its time advantage was minimal.

If you were going to choose a route that supposedly was chosen for “endpoint” ridership, then by far the worst of any of the May 1, 1971 Amtrak long distance routes one could pick would the Empire Builder. The Empire Builder route had the biggest change from its pre-Amtrak days of any long distance train. If the Empire Builder was chosen for its endpoint ridership as Mr. Norman suggests, then why change its routing? Route changes between Chicago and St. Paul, Fargo and Minot, and Spokane and Seattle only lengthened the running time over the pre-Amtrak timing. He seems to be saying that while a routing between Chicago and Minneapolis via Milwaukee was chosen on the basis of population, the routing west of Minneapolis was based on “endpoint” travel. The fact that the route through Milwaukee was chosen proves that other criteria than endpoint travel (if that was used at all, and I’ve seen no evidence to support that it was) was indeed used.

The claim of “endpoint ridership” (but only on certain routes, evidently) also seems to suggest that the routes chosen could have not have otherwise merited retaining service, but still doesn’t explain why the choices specifically were made in 1971, as do criteria as stated in the Great Falls Tribune and TRAINS magazine articles. Numerous books written on Amtrak also mention these criteria with mentioning endpoint ridership.

Indeed, there was/is merit to the route chosen. The book “Amtrak in the Heartland” states that about 60 percent of the patronage of the Empire Builder and North Coast Limited boarded between Minneapolis and Spokane and that Great Northern carried 15 percent more patrons than the Northern Pacific. The book “Amtrak” by Rodger Bradley states that the choice to run the train “along this northern route (were made) as much to gain traffic as to secure an adequate quality of ride for the passenger.”

The reality is that there were reasons the Chicago-Seattle route was chosen back in 1971, and most, if not all, had to do with the route between the endpoints. While the Empire Builder route has changed from 1971 to today, the reasons it still runs are the same today back then: Little alternate transportation, tourism, and strong online support with ridership. And that’s why today, more people use Amtrak in Minot (population 34,984) than in Omaha (population 414,521), or in Havre (population 9,390) than in Lincoln, NE (population 239,213), or in Whitefish (population 7,067) just about the same number of people (72,000) use Amtrak than in Albuquerque (population 494,236), or that the Empire Builder has been Amtrak’s single most-ridden Long Distance train for the past five years.

kmillard wrote:

vermontanan wrote:[quote="wigwagfan"}

By NYC standards, yes, there would be "no new large cities". But by Montana standards - the EB doesn't serve ANY of Montana's "large cities" but an NCH would serve virtually all of them.

I just need to clarify... Are you basically saying that you support the idea of the new train service but that Southern Montanans have advocated it the wrong way?? My main point was the new service would be one of Amtrak's strongest in terms of ridership and revenue and ridership right out of the gate according to Amtrak's own feasibility study which tend to be a little on the pessimistic side. So I'm supporting the train because that's where riders are. I'm not sure that the new train would come at the "expense" of the Empire Builder at all. Sure, they may take SOME of the Builder's riders at first, but I also think they would take some riders from the California Zephyr as well as from other modes of transport as well as generating trips that might not otherwise be taken at all.

That's exactly what I'm saying. This isn't the first time there have been studies about reviving this route, but that's about all that's happened. It gets studied, and that's it, except one time Amtrak did actually run a test train over the route. While I don't think the $1 billion price tag is correct, it's hard to know how far off it is. As stated earlier, if the train was run on a slower schedule, the cost of infrastructure improvements wouldn't have to be so high. On the other hand, the cost of stations is understated, so who knows?

All we know for sure is that for 30 years, those (mostly in Montana) who want a train have been fantastically unsuccessful. Since they expect everyone else in the country to pay for it, the cost for just this one train is too high. I think a different approach is in order. Instead of shelling out $1 billion or whatever for one long distance train, what if passenger train advocates could get $500 million allocated and invest in 4 sets of 4 trains of about 10 cars each (at about $2.5 million per car)? This would be more politically palatable to the nation overall than asking for a billion dollars for a train through Montana (where?) and North Dakota (what?). Then instead of adding a route, add a second train on, for example, the Coast Starlight, Lake Shore Limited, and Silver Meteor routes, as well as restarting a Chicago-Cleveland-Pittsburgh-Philadelphia-New York train (via existing Amtrak routes). It's well known that multiple frequencies on the same route reduce the cost of the service because infrastructure cost for things like track and stations is about the same whether there is one train daily each way or two, but revenue would more than double as travel options increase along the route. This would make the cost of long distance trains look more favorable, and it would also expose more people to the service. After this was done, request a similar amount of money, and make the Cardinal and Sunset Limited daily, and add a second Chicago-Los Angeles and Chicago-New Orleans train. Amtrak's long distance ridership would soar, but its corresponding subsidy would not. I think that multiple frequencies on such routes (not to mention state-supported corridor trains) would increase everyone's familiarity with long distance trains so that maybe some day when a North Coast Hiawatha is proposed, the relative price tag won't seem so high since more will see the value of long distance trains and a greater equipment pool might even lessen the cost all the more. It would be a long haul way of going about it, and it might not even work, but we know that studies that include numbers with lots of zeros has done nothing....for over three decades now.

wigwagfan wrote:
As a former Kalispell resident - quite a few people live outside of the city limits; in fact Evergreen (east of Kalispell) has a significant population outside of city limits; as does the Many Lakes area, the area in and around Bigfork, Lakeside (another unincorporated area), Somers (another unincorporated area), Hungry Horse (roughly a bit closer to the West Glacier station than Whitefish), Martin City, Coram... While certainly not as "dense" as downtown Kalispell, there are many year-round residents living in outlying areas.

I was merely commenting on your comment, "The largest Amtrak station (by ridership) the EB serves is Whitefish, with its population of only 8,000. Fortunately, it's not too far away from Kalispell which finally hit 20,000 in population. Surrounding area's population (including Evergreen, Columbia Falls, and several smaller communities) is under 45,000."

Clearly, your intent was that Kalispell/Whitefish/Columbia Falls and "surrounding area" (which would logically include people that lived just outside these communities or in between), had a population of under 45,000. Therefore, you must believe that other areas of the county have the remaining 43,000, which I don't, even if you happen to think that Somers, and Bigfork and the like aren't considered the "surrounding area" which I think most Montanans would.

vermontanan wrote:Then instead of adding a route, add a second train on, for example, the Coast Starlight, Lake Shore Limited, and Silver Meteor routes, as well as restarting a Chicago-Cleveland-Pittsburgh-Philadelphia-New York train (via existing Amtrak routes). It's well known that multiple frequencies on the same route reduce the cost of the service because infrastructure cost for things like track and stations is about the same whether there is one train daily each way or two, but revenue would more than double as travel options increase along the route. This would make the cost of long distance trains look more favorable, and it would also expose more people to the service. After this was done, request a similar amount of money, and make the Cardinal and Sunset Limited daily, and add a second Chicago-Los Angeles and Chicago-New Orleans train. Amtrak's long distance ridership would soar, but its corresponding subsidy would not. I think that multiple frequencies on such routes (not to mention state-supported corridor trains) would increase everyone's familiarity with long distance trains so that maybe some day when a North Coast Hiawatha is proposed, the relative price tag won't seem so high since more will see the value of long distance trains and a greater equipment pool might even lessen the cost all the more. It would be a long haul way of going about it, and it might not even work, but we know that studies that include numbers with lots of zeros has done nothing....for over three decades now.

Interesting... Another Shasta Daylight/Cascade, a Pacemaker, another Champion, another Three Rivers (Broadway??), an El Capitan and another Panama Limited!! Don't for get a Western Star for the busiest route of all and a new Denver Zephyr to boot along with a separate New England States for Chicago to Boston!! I like your thinking!!

wigwagfan wrote: It should also be noted that Flathead County is served by THREE Amtrak stations (Whitefish, West Glacier, Essex) - and so to use Flathead County's population as the base for Whitefish's service area would be tripling up on the ridership potential of Amtrak in the county.

Three completely different Amtrak stations. Whitefish is a full service staffed station. West Glacier is unstaffed and 90 percent of ridership occurs during Glacier Park's summer season. The station isn't even open (open only when the Glacier Park Historical Association which occupies the structure is open) for most trains during the off-season. 99 percent of the ridership in and out of Essex is for the Izaak Walton Inn. While West Glacier is mostly a ghost town during the off-season, even those from year-round communities such as Martin City and Hungry Horse tend to most often use the Whitefish stop rather than West Glacier due to the amenities offered. (The West Glacier stop, for instance, doesn't even have a caretaker during the winter to do things like shovel the platform.)

In other words three stations could only "triple up the ridership" in Flathead County if they were all equally accessible and equally desirable. And they're not.

kmillard wrote:

vermontanan wrote:Then instead of adding a route, add a second train on, for example, the Coast Starlight, Lake Shore Limited, and Silver Meteor routes, as well as restarting a Chicago-Cleveland-Pittsburgh-Philadelphia-New York train (via existing Amtrak routes). It's well known that multiple frequencies on the same route reduce the cost of the service because infrastructure cost for things like track and stations is about the same whether there is one train daily each way or two, but revenue would more than double as travel options increase along the route. This would make the cost of long distance trains look more favorable, and it would also expose more people to the service. After this was done, request a similar amount of money, and make the Cardinal and Sunset Limited daily, and add a second Chicago-Los Angeles and Chicago-New Orleans train. Amtrak's long distance ridership would soar, but its corresponding subsidy would not. I think that multiple frequencies on such routes (not to mention state-supported corridor trains) would increase everyone's familiarity with long distance trains so that maybe some day when a North Coast Hiawatha is proposed, the relative price tag won't seem so high since more will see the value of long distance trains and a greater equipment pool might even lessen the cost all the more. It would be a long haul way of going about it, and it might not even work, but we know that studies that include numbers with lots of zeros has done nothing....for over three decades now.

Interesting... Another Shasta Daylight/Cascade, a Pacemaker, another Champion, another Three Rivers (Broadway??), an El Capitan and another Panama Limited!! Don't for get a Western Star for the busiest route of all and a new Denver Zephyr to boot along with a separate New England States for Chicago to Boston!! I like your thinking!!

Bring back the Montrealer overnight on the Conn River line, and the Morning Zephyr on the Twin Cities-Chicago run!

Seriously, though, I think there actually might be a decent market for a new New England States or something else running 10-12 hours off the schedule of the Lake Shore, at least on the Boston-Albany leg.

At this point, it seems that reinstating a Seattle to Chicago daily LD train at a cost of over $1 billion is a bit of a dream; but I've gone back and studied the proposal again to see what it would cost to start service on the 3 segments of the NCH proposal that would likely be the most promising: Seattle to Spokane, Missoula to Billings, MSP to Chicago.

For the Chicago to MSP segment: the cost to improve the tracks and signals is estimated at $44 million, add 2 more Talgo trainsets to the Hiawatha order for $50 million, plus 3 more locomotives and 3 cab cars. Another daily roundtrip between Chicago and MSP could be initiated for about $100 million.

For the segment between Billings and Missoula: tracks and signal improvements are estimated at about $53 million, 2 trainsets would be needed and stations would need to be rehabilitated. If the Siemens DMU proves to be sufficient and station re-hab costs are reasonable, this service might start for under $100 million.

For the segment between Seattle and Spokane via Stampede Pass: track and signal improvements are estimated at about $200 million, 2 trainsets would be needed and several stations would need to be repaired. This service could be started for about $300 million.

So added altogether, 3 new, short corridor, daytime services could be established along the proposed NCH route at about half the cost of a full service endpoint-to-endpoint LD train. The operating costs of the corridor services would also be considerably lower than the NCH. Daily train miles of the 3 corridors would be about 55% of the NCH's, so most operating costs would be about one-half of the NCH's (considering the reduction in dining car and sleeping car staffing expenses). I'm not an authority on predicting ridership, but if the 3 corridors are able to set the train schedules to best serve their enroute stations and passengers, then the trains could be maximizing their revenue potentials. The study predicts that the NCH would have a farebox recovery of about 58% but increase Amtrak's annual operating loss by $39.1 million, so I'm not visualizing a Congressional stampede to approve the NCH. I think the various States along the route should have their DOTs looking at alternatives to the NCH proposal.

Fragmentation on a long-distance route, especially a piece served just in the middle of it, is not likely to work. The economics of such routes depends on strong end point "anchors" (large traveling population living there, plus potential passengers from connecting trains). Getting the passenger volume up to an acceptable level requires that many overlapping markets be served. If one piece of the route is missing, much of the overlapping ridership disappears.

On the cost side, the freight railroads usually charge by the train mile for Amtrak's access; so a doodlebug costs just as much as a ten-car train for that cost element.

So it's all or nothing. Unfortunately, this is perhaps the longest possible LD route between population anchors that is conceivable in the U.S. Even a scheme for a split-off from the EB west of Fargo results in a very long added-service run.

Vincent wrote:At this point, it seems that reinstating a Seattle to Chicago daily LD train at a cost of over $1 billion is a bit of a dream; but I've gone back and studied the proposal again to see what it would cost to start service on the 3 segments of the NCH proposal that would likely be the most promising: Seattle to Spokane, Missoula to Billings, MSP to Chicago.

For the Chicago to MSP segment: the cost to improve the tracks and signals is estimated at $44 million, add 2 more Talgo trainsets to the Hiawatha order for $50 million, plus 3 more locomotives and 3 cab cars. Another daily roundtrip between Chicago and MSP could be initiated for about $100 million.

For the segment between Billings and Missoula: tracks and signal improvements are estimated at about $53 million, 2 trainsets would be needed and stations would need to be rehabilitated. If the Siemens DMU proves to be sufficient and station re-hab costs are reasonable, this service might start for under $100 million.

For the segment between Seattle and Spokane via Stampede Pass: track and signal improvements are estimated at about $200 million, 2 trainsets would be needed and several stations would need to be repaired. This service could be started for about $300 million.

That's what Amtrak would want. Let the states put up the money. Let the states pick up the subsidy. Let Amtrak run the service without taking any risk. A Chicago to Seattle train would be a national train and Amtrak would have to subidize the service. The Seattle-Spokane service would do respectably. The Chicago-St. Paul service would do well. Billings-Missoula doesn't have strong endpoints.

Washington state has set the Cascades as it's priority, so WA isn't likely to initiate anything on it's end. Meanwhile, Minnesota is involved in an internal struggle to determine whether future Chicago-St. Paul service should be routed through Rochester or continue to follow the curves of the Mississippi River. And of course, all the states would rather see the money come from Washington DC than their own capitols.

As far as the rolling stock is concerned, DMUs seem better suited to commuter operations with frequent stops where they alternate between accelerating and braking.

Fragmentation on a long-distance route, especially a piece served just in the middle of it, is not likely to work. The economics of such routes depends on strong end point "anchors" (large traveling population living there, plus potential passengers from connecting trains). Getting the passenger volume up to an acceptable level requires that many overlapping markets be served. If one piece of the route is missing, much of the overlapping ridership disappears.

On the cost side, the freight railroads usually charge by the train mile for Amtrak's access; so a doodlebug costs just as much as a ten-car train for that cost element.

So it's all or nothing. Unfortunately, this is perhaps the longest possible LD route between population anchors that is conceivable in the U.S. Even a scheme for a split-off from the EB west of Fargo results in a very long added-service run.

I think Amtrak has been evolving away from the LD model towards the short corridor concept for quite some time. In WA we haven't been waiting for the restoration of the Pioneer, we established the Cascades. VA has funded new service because the James Whitcomb Riley isn't coming back. The Desert Wind won't blow any more, but I'll bet that an HSR train runs between Los Angeles and Las Vegas long before any traveler catches a passenger train in Bismarck ND again. The FL legislature just voted to fund passenger rail, not to restore The Seminole or The Champion, but to give Floridians local service that will serve the local population. I'm not one who advocates eliminating the current LD network, but I'm not in favor of adding new LDs at the expense of shorter, less costly and likely more heavily patronized corridors.

The access costs for any Amtrak train is so small that it isn't even listed as a line item expense in the NCH proposal. However the infrastructure improvement costs are tremendous (about $619 million) and about half of that money would have to be spent between Billings and Fargo. WA is already studying a Seattle-Stampede Pass-Spokane train, Montanans have been pushing for the Missoula-Billings service for years and the existing Empire Builder between Chicago and MSP is already heavily used. The "all or nothing" attitude is most likely to leave us with nothing.

A Missoula-Billings service, by both location and length, would hardly qualify as a long-distance service. It would be a rather light-density corridor. As such, it isn't primarily Amtrak's business; and initiative and funding would have primarily to come from Montana.

That would seem to be even less likely to happen than some kind of through train linking the Amtrak system at both ends or even at just one end.

Gilbert B Norman wrote:Mr. Ultra, were you able to review the piece from The Times archive?

Unless a subscriber (possibly you are one such yourself), access is "hit or miss".

I've clicked it several times in the hope that 'hit or miss' meant that they sometimes don't protect it behind the firewall, but haven't been able to read it so far. I purchase the paper frequently, but am not a subscriber nor on-line subscriber.

I admit I completely overlooked the distinction you were making between the two trains when I first replied, so my reply must have seen pretty obtuse. I'm a knucklehead.

If I remember correctly, Montana Rail Link [MRL] owns/operates the lines from Billings to Sandpoint, at least the ones that would be most useful for a commuter type project. Until about five years ago they were running a tourist train from Livingston to Sandpoint 2 - 3 times a week. That's now gone by the board - didn't keep making money. However, I wonder how difficult either a LD daily or shorter commuter sets would be to implement in the face of potential opposition from the rail owner. They run a hefty amount of freight over those lines.

Additionally, wouldn't even a short line be more effective if it tied in with Amtrak's LD system at least at one end? I don't see the value of Missoula to Billings in terms of external revenue (particularlly tourist dollars) if visitors have to get to Missoula or Billings first. Where, hypothetically, would such a connection be made?

Bus companies (I know I'm not supposed to mention them, but they give the best current comparison) in MT have managed to stay afloat, but is there enough Missoula to Billings ridership to also justify a rail system?

It would be great to see such a system come into being, but I think there are some good reasons why nothing has come of past proposals.

AmtrakRider wrote:Additionally, wouldn't even a short line be more effective if it tied in with Amtrak's LD system at least at one end? I don't see the value of Missoula to Billings in terms of external revenue (particularlly tourist dollars) if visitors have to get to Missoula or Billings first. Where, hypothetically, would such a connection be made?

Who says that the "connections" must be another Amtrak train?

I know a LOT of tourist trains that are doing quite, quite well, and are nowhere near an Amtrak line. In fact, every tourist railroad in Washington and Oregon is detached from an Amtrak route. And VIA's Malahat is detached from any other VIA service; not to mention the Alaska Railroad from any other railroad...

There is far, far too much emphasis on ensuring that every dot is connected. Then again, maybe an intra-Montana passenger train makes sense -- but not as an Amtrak train, possibly one operated by Montana RailLink and sponsored by the state/federal government, where the train simply has an Amtrak train number and Amtrak can sell tickets for (like Oregon's Lewis & Clark Explorer Train in its first year of operation.)

vermontanan wrote:In other words three stations could only "triple up the ridership" in Flathead County if they were all equally accessible and equally desirable. And they're not.

What you are saying is that the permanent residents of towns like West Glacier, Coram, and Hungry Horse - DESPITE having access to Amtrak in West Glacier, are going to drive all the way to Whitefish to board a train, just because it's "full-service" (whatever that means)?

Unless there is some need to have the service of the station agent (buy a ticket, and that one can do from the Conductor on the train at the other stations)...basically the argument is that there's a waiting room in Whitefish. And I can park my car right next to the tracks in West Glacier with the heat running until the train shows up...

I'm wondering if the argument is really that there's no reason for Amtrak to stop at three locations in one county.