Why cutting passenger trains doesn’t save money


By Noel T. Braymer

The simple fact is most intercity passenger trains make money, on their direct operating costs. The problem at Amtrak is it doesn’t run enough trains to bring in enough money to cover all its overhead costs. Amtrak has some large overhead costs, centered mostly in the Midwest and East Coast. Yet Amtrak usually tries to save money by cutting back on passenger service.

Back in the middle 1960’s the Western Pacific Railroad was losing a lot of money. It looked at cutting back the original California Zephyr to less than daily service. The WP discovered that would not save money because it wouldn’t reduce the train’s fixed cost but would reduce income from the days the train didn’t run. Every time Amtrak has cut back or eliminated routes on long distance trains, Amtrak has lost money not saved money. Back in 1995 Amtrak tried to cut back service west of St. Paul on the Empire Building to four times a day. This resulted in a loss of 100,000 riders and 25 million dollars.

Why does this continue? Amtrak has a very strange accounting system. Amtrak knows how much money it brings in and how much money it spends. What it doesn’t know is what individual trains cost to run. Amtrak arbitrarily assigns costs including system overhead to trains on a train mile basis. The more miles a train runs the more it “costs” Amtrak. Amtrak basically assumes that when a train is standing still it doesn’t cost them money. Where this gets really silly is if you want to add a second train to a route. Amtrak will assume that the second train will cost about as much as the first. It doesn’t matter if no additional equipment or more employees are needed to run additional service. What gets really interesting is Amtrak will charge the same amount of system overhead for both trains. In the real world the overhead should be charged by the route not the train. In other words two trains on a route should be charged the same as only one train for Amtrak’s overhead. The inverse of this is Amtrak assumes that by cutting back service or eliminating a route the overhead assigned to that train magically disappears as service is cut. Of course it doesn’t, the system overhead stays the same, but fewer trains have to cover the overheard so that on paper the remaining trains costs go up and Amtrak doesn’t save money, it loses money.

Back in June 1999 the late Dr. Adrian Herzog wrote in this publication that Amtrak then needed an additional 1800 rail passenger cars in order to carry enough passenger miles to cover its current overhead. He also warned in 1999 that the Acela program would add to Amtrak’s overhead and not greatly increase income. “The bottom line is that they will increase the deficit dramatically between now and 2002.” Adrian was right; Amtrak almost went bankrupt in 2002 and requires much more subsidy now than in 1999.

What did Dr. Herzog suggest Amtrak do with these additional 1800 passenger cars to break even? The key is to increase passenger miles while not increasing system overhead. In other words no new maintenance facilities, major new stations, crew bases etc. The first thing to do is add more cars to existing trains so they can carry more passengers. Adding one sleeping car to a Superliner train requires 7 cars to cover every consist and backup equipment for maintenance. Typical Superliner trains run about 9 cars each. In the past in America, 18 car passenger trains were common.

The second thing that can be done is to add more frequent service on existing trains. As the WP discovered less than daily train service doesn’t save money. Not only should there be daily trains, but Dr. Herzog’s plan for the 1800 cars called for three trains a day on most long distance routes. The reality is when there is more frequent rail service ridership increases more than the level of service. Back in the 1970’s service between Los Angeles to San Diego doubled from 3 to 6 daily trains, but ridership more than tripled from about 300,000 to a million annually. These additional trains will need more cars than the trains of today.

The third part of Dr. Herzog’s plan is to add extensions and improve connections to existing services. A simple example of this would be to extend Coast Starlight trains to Vancouver. Amtrak already has service to Vancouver so there is no increase in overhead costs. Additional Starlights would be able to connect with other San Francisco Zephrys and Sunsets. A service highly recommended by Dr. Herzog was to run a South West Chief train from the Bay Area down the San Joaquin Valley and continue east from Barstow. Using his computer model for ridership, Dr. Herzog found adding the Valley and Bay Area to the Chief raised ridership dramatically. Dr. Herzog also recommended splitting the Chief Trains at Kansas City with a section running to St Louis and then up to Chicago. There is no shortage of good service extensions or routes that were wrongly eliminated that should have passenger trains. What’s needed are more rail cars and locomotives. For that we need the support of congress.