Like many libertarians attracted to transportation policy – Reason founder Robert Poole comes to mind – I love trains, especially passenger trains. So Amtrak is a particularly painful subject. On one hand, I know that government subsidies are wasteful and likely to be abused. On the other hand, I am thrilled by the idea of taking a train from Seattle to Los Angeles or Chicago to San Francisco (actually Oakland).
Joseph Vranich, who helped create Amtrak back in 1970 and later worked fora pro-Amtrak lobby group, gives train lovers some reason for hope. Working with the Amtrak bureaucracy for years has convinced Vranich that the best way to protect passenger trains is to get rid of Amtrak. His 1997 book, “Derailed: What Went Wrong and What to Do About America’s Passenger Trains,” argued that killing Amtrak would lead many railroads or other private entrepreneurs to get back into passenger service. His more recent book, “End of the Line: The Failure of Amtrak Reform,” showed why Amtrak reforms approved by Congress in 1997 were never put into place.
Amtrak is truly a failure. After nearly $30 billion of public investments, Amtrak logged fewer passenger miles of travel last year than the railroads did in 1970, the last full year of private rail·operation. From 1975 to 2000, airline passenger miles grew by 260% and highway passenger miles by 83%, but Amtrak passenger miles increased by less than 40%.
Amtrak cannot blame its failure on greater subsidies to competing modes of transportation. According to the U.S. Department of Transportation, subsidies to highways in 2000 averaged less than four-tenths of a penny per passenger mile, while subsidies to airlines were just one-tenth of a
penny per passenger mile. At close to 20 cents per passenger mile, subsidies to Amtrak were 50 times highway subsidies and nearly 200 times airline subsidies. As the Washington Post wrote earlier this year, “every 10,000 miles that a train passenger traveled in 2002 cost.federal taxpayers $200 in subsidies, compared with $6 for passenger jets and $4 for long-distance buses.”
YetI can’t help feeling that it didn’t have to be this way. If it had been properly designed from the start, Amtrak could have operated and possibly expanded passenger service without any operating subsidies. While it might have needed federal or state grants for capital purchases of new rail cars, it is even possible that passenger revenues could have paid for part or all of the cost of such new cars.
Was Amtrak doomed to fail from the start? Are passenger trains truly obsolete outside of a few busy corridors? Or was Amtrak’s failure due to poor management, ill-advised business models, congressional meddling, or any other factor within our control? With a different history, could Amtrak have done better? And if so, can we still save it today?
Agreeing on the Mission
America’s freight railroads have proven that rail transport is not dead. Between 1970 and 2000, railroads increased their freight movement (measured in tons carried times miles traveled) by 92%. During this time, overall freight movement increased by only 71%, so rail’s market share increased.
How was this gain accomplished? The simple answer is that Congress deregulated the railroads in 1979, allowing them to be much more responsive to their customers. The more complicated answer is that the railroads took advantage of deregulation to create a new business model that made more effective use of both labor and equipment.
Congressional deregulation of the airlines was also the key to their huge gain in passengers over the past two-and-a-half decades. It is especially worthwhile to compare Amtrak with Southwest Airlines, which, like Amtrak, began operating in 1971. After the airlines were deregulated, Southwest followed a different business model than its competitors, making far more effective use of its airplanes and workers.
Years of working with the Amtrak bureaucracy convinced Vranich that the best way to protect passenger trains is to get rid ofAmtrak.
This model has made it the most profitable, most imitated, and, by some measures, the largest domestic airline in the nation.
By comparison, Amtrak followed the same tired business model for passenger rail that the railroads had used for decades. This model required high labor costs and high capital costs that made it unable to compete with highways and deregulated airlines.
Could a government entity such as Amtrak be as innovative as Southwest Airlines? Libertarians would say no. As Great Northern Railway builder James J. Hill once observed, government operation is “slow, cumbrous and costly.” But for the sake of argument, let’s assume we are in 1970 and Congress is writing legislation to take over the nation’s private passenger trains. What would we include in that legislation?
The first thing government agencies need is a clearly defined mission. “Government will may perform if an agency is under pressure to satisfy different constituencies with different values and different demands,” says management expert Peter Drucker. “Performance requires concentration on one goal.”
At first glance, Amtrak’s mission would seem to be simple: attract and carry passengers. But for many powerful members of Congress, that goal has been secondary to the mission of providing pork to their states and districts. West Virginia Sen. Robert Byrd, for example, was famous for diverting Amtrak’s high-speed turbo trains to low-speed service through his state. For unions, the goal of carrying passengers has been secondary to keeping labor costs high. For many urban mayors, Amtrak’s chief goalwould seem to be to restore oversized, but little used, historical train stations. These and other side goals detract from the supposed chief mission of carrying passengers.
Perhaps it was this confusion over goals that led early Amtrak president Paul Reistrup to tell a rail-enthusiast mag-
azine in the mid-1970s that “we can’t afford to subsidize sightseeing.” That might have been true if Amtrak’s goal was to provide pork or support unions. But if Reistrup thought that Amtrak’s goal was attracting passengers, he would have realized that sightseers were part of his core market and that sightseeing was Amtrak’s main competitive advantage over the airlines.
Once everyone agrees that Amtrak’s sole goal is to carry passengers efficiently, the company must be insulated from congressional pressure to do otherwise. Amtrak must be made as independent as possible, especially for operating costs.
Partnerships to Minimize Costs
The best way to lower operating costs is to create incen- tives to reduce costs so that minimal subsidies, and ideally no operating subsidies, are needed. The first step would be to rely on competition and private enterprise rather than government monopoly to carry out core passenger rail ser- vices: train operations, food services, interior cleaning, and railcar maintenance.
Urban transit agencies that contract out bus services to private operators spend 40-50% less per bus mile than agen- cies which operate buses themselves. The agencies purchase buses, then lease them to Laidlaw or other bus operators who maintain them and hire the drivers to run them. Transit agencies in Boston, Los Angeles, Ft. Lauderdale, and other cities also contract out their commuter rail operations. to Herzog Transit Services and other companies.
Amtrak could start by contracting out dining car services. Many passenger trains have both a full-service diner and a grill car providing fast food. Rather than operate these ser- vices itself, Amtrak could accept bids from restaurants to provide them, in the same way that the Fred Harvey Company long provided meal services to passengers on the Santa Fe Railway. McDonalds, Wendy’s, Pizza Hut, and Taco Bell might bid on the fast-food services. Chili’s, Applebees, or other sit-down restaurant chains might bid on the dining car services. Operators would not have to be national chains, and local restaurants might be particularly interested in serving short-distance corridor trains that pass through their home cities.
With hundreds of daily trains, many different companies could provide meals on different trains. Such competition would encourage each operator to provide better service at lower costs. Ideally, Amtrak would require contractees to charge the same prices on the trains as they do in stationary restaurants. Except possibly on the busiest routes, private operators might not be able to provide Amtrak food services at a profit, but the bids they would submit would offer to provide those services at the lowest possible subsidy.
In the same way, Amtrak could experiment with contracting out first-class sleeping car operations to hotels, just as the railroads once contracted sleeping car service to the Pullman Company. For that matter, Amtrak could contract out actual train operations to private operators, possibly the railroads themselves. Railroads that had proven particularly friendly to passengers before 1971, such as the Santa Fe, might become the preferred operators even off of their own lines.
Contracting various rail services to different operators has been used with varying degrees of success in Britain since that nation privatized most of its railroads. One result is that Britain is providing rail services as good as or better than those on the European continent at a much lower cost.
Maximize Labor Productivity
Amtrak can save even more on labor. The railroads’ business model of large station crews, one attendant for each revenue car, huge dining car crews, plus a five-person train crew was unsustainable by 1970. Today, train crews have been reduced to three, but Amtrak still pays too many onboard attendants, station employees, and baggage handlers.
Americans who marvel at European trains should note that they use significantly fewer workers. Passengers handle their own baggage, there are no coach attendants, on-board crews are held to a minimum, and many stations have no ticket agents or other staff.
Three out of five dollars spent on Amtrak operations go to salaries, wages, and benefits. Cutting labor costs in half would eliminate Amtrak’s operating deficit. Reducing labor costs would allow Amtrak to operate more trains, and the revenue from those trains would similarly reduce operating losses.
A revised business model could considerably reduce Amtrak labor requirements. First, ticketing should be handled mostly by travel agents or machines. Rather than employ ticket agents at every station, Amtrak could rent out part of its stations to travel agents who would sell all forms of travel including rail travel. Ticket machines and on-board sales by conductors would suffice to serve towns too small to support a travel agent.
For baggage handling, Amtrak could contract with UPS or another shipping company to move goods that wouldn’t take up an entire train car. As a part of the contract, the company would also handle passengers’ baggage. This might work only in larger cities that generate lots of business less than a carload. In smaller towns, Amtrak might offer no bag-
Subsidies to Amtrak have been 50 times greater than highway subsidies and nearly 200 times airline subsidies.
gage service, but would encourage people bringing baggage that is too large to fit in overhead luggage racks to stow it in lockers provided for them on selected train cars.
The main reason Amtrak has an attendant in each car is to open the doors at stations and help passengers on and off the trains. Instead, Amtrak should equip all rail stations with high-level platforms and all passenger equipment with automatic doors that can be centrally controlled by the conductor. The labor savings would quickly repay the cost of these changes. At each stop, the conductor opens all the doors, people get on or off, the conductor closes the doors, and the train goes on. This virtually eliminates the need for coach attendants and greatly reduces the need for sleeping car attendants. This could also reduce station dwell time and thereby speed up timetables.
Sleeping car attendants might be completely eliminated by contracting for cleaning crews to clean coaches and sleep-
ing cars on every train each day. Such cleanings, including the replacement of all linens in sleeping rooms that had been occupied the night before, might take place during long layovers or while the train is traveling between two cities. When
For many members of Congress, the goal of carrying passengers has been secondary to the mission of providing pork to their electorate.
the crew is done, they would get off the train and get on the next train back, cleaning that train while returning to their home city.
Aside from the engine crew and food service workers employed by restaurant-contractees, the train crew effectively becomes no more than a conductor and a trainman, themselves perhaps employed by railroad-contractees. Ultimately, Amtrak itself might have no employees other than the managers who oversee all of the contractees.
Maximize Use of Equipment
Reinventing passenger rail service also means making more effective use of equipment. The old business model had trains travel for anywhere from several hours to two days, then sit for as long as 24 hours before making a return trip. The result was that cars would be in service for as little as 65-75% of the time. Eliminating this downtime would allow Amtrak to operate more trains with the same number of cars.
One of Amtrak’s early discoveries was that a continuous Seattle-to-Los Angeles train was much more popular than the Seattle-to-Portland, Portland-to-Oakland, and Oakland- to-Los Angeles trains that the railroads had been running. Why not extend this lesson and run a train from, say, Florida to Chicago to Seattle to Los Angeles to New Orleans and back to Florida?
A more complex routing would have a train weave back-and-forth across the country for several weeks at a time. After three to four weeks, the train would be taken out of service for a day or two for major maintenance. Passengers from, say, Oakland to Albuquerque could make the trip without changing trains. The “weave” would also be designed so that popular routes, such as San Francisco-to- Los Angeles, would get more than one daily train.
Naturally, schedules would include padding to allow for late trains, including long layovers at ‘1 corner” cities such as Seattle and Los Angeles. But through passengers might appreciate these layovers as an opportunity to get a quick tour of the city. Even with such padding, Amtrak could get 90% use of its equipment instead of 70% or less, while the increased connectivity offered by such schedules might actually boost ridership.
Make Trains Interesting
In addition to cutting costs, Amtrak needs to attract passengers. Depending on the route, Amtrak serves three different markets. First are business travelers who appreciate high-speed downtown-to-downtown service. Second are students and other low-income travelers who don’t have a car but want to travel by something more comfortable than a bus. Finally there are vacationers who believe that getting there should be half the fun.
Amtrak can compete for business travelers in the Boston- to-Washington corridor based on speed and convenience. Elsewhere, Amtrak is often slower than driving for short-distance trips and certainly slower than flying for long-distance trips. To compete for passengers, especially on long-
To compete for passengers, especially on long-distance trains, Amtrak has to realize that it is really in the entertainment business.
distance trains, Amtrak has to realize that it is really in the entertainment business. This means it has to make its trains interesting places to be.
Some of the railroads realized this in the closing years of private rail passenger operations. The Great Northern Empire Builder of the late 1950s, for example, had three non- revenue seats, in eight different locations, for every four revenue seats. The non-revenue seats were located in grill cars, diners, and various observation cars and lounges, and were for use by any ticketed passenger.
While Amtrak inherited the equipment of the Empire Builder and other fascinating trains, most of this equipment was near the end of its expected service life. When private bus companies were taken over by public transit agencies in the 1970s, the first thing the public agencies did was obtain federal grants to replace their aging fleets. But Amtrak made little effort to replace its long-distance passenger cars for several years, leading to a reputation for being unreliable. When it did replace the trains, the double-decker cars it purchased, known as Superliners, were a step backwards from the Santa Fe passenger cars that inspired them.
In contrast to the 1955 Empire Builder, Superliner non- revenue areas were limited to the diner, the Sightseer Lounge with its inexplicably small number of seats, and a cafe below the lounge. The non-revenue spaces were inadequate, and the revenue spaces were boring. Coaches combined the monotony of Greyhound buses with the discomfort of airline seating. With four different room configurations, sleeping cars’ were slightly better but were priced out of reach of most travelers.
Instead of developing its own, unimaginative specifications, Amtrak should have invited railcar manufacturers to propose their own plans for medium- and long-distance train cars. Here is how I envision. long-distance, double-decker Superliners.
Instead of a mid-train lounge car, I would put lounges at the beginning and end of every long-distance train. One half of each car would be covered with wraparound windows, giving passengers a tremendous,view of the passing scenery.
These windows would face forward on the’first car on the train and provide viewing for 30 to 40 theater-style seats. The back half of this car would be the train’s fast-food or grill car.
On the last car of the train the windows would face backward, with 30 or so seats facing in different directions, parlor-car style. The front half of this car would have tables and seating, for 40 to 50 dining car patrons. When demand was high enough, diners could also eat in the lounge.
Coaches would have two different configurations downstairs. One would have room for handicapped passengers or other riders who don’t want to climb stairs to their seats. The other would have a large’self-serve baggage area with lockers, ski racks, bicycle hooks, and other storage areas.
Upstairs, coaches would have seating in various configurations and might be divided into more than one compartment for variety’s sake. Comfortable seats would face forward in some compartments, but seats in other compartments might face to the side or circle around tables.’Long-distance trains would have a children’s car with toys, games, and playsets. “Quiet cars” might serve business travelers by providing power outlets and wireless Internet. Of course, laptops and the Internet were not known in 1970, but the idea of keeping some cars quiet, while encouraging noisy children to use another car is timeless.
Another style of coach could provide revenue seats in the off-season but quickly be converted into a mid-train feature car during high-demand seasons or on popular routes. The car could include a small stage for musicians and other family entertainers to perform on board in exchange for tips and travel discounts. Movies could be offered when no live entertainment was available. The lower section of the car might hold vending machines with snacks and beverages.
Sleeping cars would come in first-class and second-class configurations. Second-class sleepers would, be divided into compartments with seats for four that fold’ down into two double bunks at night. The bunks would have mattresses but passengers would bring their own sleeping bags or linens, and parties of fewer than four people would expect to share
Ultimately, Amtrak itself might have no employees other than the managers who oversee all of the contractees.
compartments with other passengers. First-class sleepers would be private rooms similar to or better than those found on current Superliners.
Amtrak’s red-grey-and-blue pointless-arrow, exteriors and Betsy-Ross Moderne interiors promised passengers a journey devoid of excitement. Each car should instead be individually decorated, with a strong regional emphasis. Exterior colors would give potential passengers a preview of the the.forests, fields, mountains, and streams they would see on their journeys. Interior decorations might include regional Native American art, scenic photographs, and, in the feature cars, the type of “found” decorations seen in many restaurants and bistros. Every train, and every car on the train, would offer passengers something new and exciting.
Would It Work?
If this new business model had been implemented in 1971, would it have prevented the disaster that Amtrak has become? As previously noted, reducing Amtrak’s operating costs by 30% would eliminate its operating deficit. If passenger revenues covered all operating costs, federal and state governments would be limited to providing capital grants for rail equipment. If better-designed trains further increase revenues and better-utilized equipment further reduces costs, Amtrak might even be able to cover some or all of those capital costs. Reducing Amtrak’s financial dependence on taxpayers would protect it from pork-barrel Congressmen.
Amtrak would benefit by making such improvements today, though resistance to change is now greater and it will take several years for the benefits to be seen. The 1997 Amtrak Reform Act allowed Amtrak to contract out services and cave less often to unions. However, as Vranich detailed in “End of the Line,” Amtrak failed to take advantage of this law.
Amtrak could start by contracting out dining car services, experimenting with through schedules that better utilize equipment, and refitting Superliners and other cars with automatic doors. Once most trains were refitted, Amtrak could build platforms at all stations on selected routes to see if it could reduce the size of on-board crews.
Would these changes allow Amtrak to become a significant player in American travel? Probably not. Despite billions in subsidies, European trains have a small and declining share of intercity traveL But a new business model that minimizes or eliminates subsidies would allow passenger rail to stay on the tracks in the United States without the periodic political threats and continuing fiscal crises that characterize it today. That would be enough for most every- one to call it a success.