Any American who has traveled in Europe or Asia is in awe of the train service there. Train service is cheap, clean, and efficient. It is remarkably easy to travel from country to country by train, even if traveling with small children.
If you wish to travel around the American Southwest, however, you’re obliged to go by car. Depending on the time of year and your destination, it can take all day to cross a hot desert. It is almost impossible to make that trip without wishing for a European-style high-speed rail system.
Well, that is not going to happen anywhere in the Southwest, except possibly in California.
While there are no hard and fast rules for predicting when rapid intercity rail transport becomes feasible, there is a set of guidelines, and using those guidelines, the American Southwest doesn’t even come close to making such infrastructure profitable. Let’s look at the possible routes, step by step.
The first requirement for building high-speed rail is corridor distance—the number of miles between the starting and ending points of the line. Using the existing, successful high-speed rail lines as a guide, the sweet spot for profitable lines is more than 150 miles and less than 600 miles. For shorter routes, it is hard for a railroad to compete with cars, and for longer routes, it is generally cheaper and faster to go by air.
Possible routes that are far too long for high-speed rail include those from any city in Texas or New Mexico to any city in California. Routes that are too short to be feasible include Tucson to Phoenix, Las Cruces to El Paso, and Albuquerque to Santa Fe. With the exception of California, the distances between many towns in the Southwest are multiples of 35-50 miles—the distance a steam locomotive could travel before needing to stop and take on water. Longer stops—at roughly 150 miles apart—accommodated crew changes and maintenance needs. For most cities in the desert Southwest, neighboring cities in the same state are simply too close together to make high speed rail feasible.
The next criterion is terminus to terminus population. As a general rule, each end of the route should have a population between one and three million people, with a total population along the route between 8-10 million people. If we follow this rule strictly, we eliminate any high-speed rail in New Mexico, Utah, or Nevada. Even the Los Angeles to Las Vegas route does not meet this guideline because the population of Las Vegas is too low. The only remaining possible routes are Phoenix to Los Angeles and the already-proposed California High-Speed Rail (CAHSR) system that links Los Angeles and San Francisco.
Both of the remaining possible routes meet the next guideline hurdle: Competitive travel time. High-speed rail should be at least 20–30% faster, door-to-door, than driving and should be competitive with flying (including airport delays). This usually means train speeds of 150–220 mph with minimal intermediate stops. The Phoenix to Los Angeles route meets this easily, while the proposed California route only passes because California freeway traffic could only be slower if every car were driving in reverse. As it is, on just the Los Angeles to San Francisco route, California is estimating a time of 3 hours, give or take 15 minutes (depending on whether the train is an express or stops at any of the ten intermediate stations). The same trip by car would take twice as long, even if the train were to stop at every intermediate stop.
The next guideline is load capacity. For a high-speed rail project to operate profitably—ignoring initial construction costs—the train needs an average daytime load above 70% (not just at peak hours). Even the most generous estimate eliminates the Phoenix to Los Angeles route. (Nationwide, existing passenger rail lines average only 40%)
The last guideline is estimated ridership. The break even on operating costs is approximately 35 million riders per year. According to the State of California, that number of riders for the future CAHSR is absolutely assured. On the other hand, if you listen to several independent analysts (by such entities as the California High-Speed Rail Authority or third-party analysts) the proposed California system will be lucky to have half that number of riders.
There are a few other minor considerations. If the town to be served by high-speed rail is also served by existing slower rail facilities, that is a plus. In the case of the American Southwest, Amtrak connects to Albuquerque, Tucson, El Paso, and Los Angeles. Amtrak does not go to San Francisco, Las Vegas, or Phoenix. It may surprise you to learn that the Atchison, Topeka, and Santa Fe Railroad was started without a connection to Santa Fe. Eventually, there was a connecting spur line, but Amtrak does not connect to Santa Fe. None of above make the success of high-speed rail in the Southwest any more likely.
There is a theory that building a commuter railroad will foster development, and that ridership will gradually increase until the new railroad is profitable. This was exactly the reasoning behind New Mexico’s building the Railrunner Express that links Albuquerque and Santa Fe. Setting up the line cost the state $635 million. Since the line opened in 2006, ridership has actually decreased to approximately 1,900 riders a week. Fares currently account for almost 4% of the annual operating costs, which do not include interest on the debt. Another 4% of the operating cost is donated by the federal government. At the current level of ridership, the New Mexico taxpayers would save approximately $40 million a year if the state paid for each of the riders to take an Uber instead of the train. The riders would also arrive at their destinations faster.
If you are interested in math, economists actually have a formula (at right) for evaluating whether a transportation project is fiscally sound. NPV is the net present value and r is the interest rate, which is normally between 3-7% for transportation projects. C stands for Cost and B is for benefits. Setting T for the time period of 40 years, if the NPV is greater than 0, the proposed project is fiscally sound, but if the NPV is below 0, the project is unsound and will be a drag on the taxpayers.
I started to work out the NPV for the New Mexico Railrunner Express, but the results were just too tragically depressing, so I decided to run the numbers for Phase I of the California High Speed Rail using 4% for the interest rate. As you might expect, the answer was a little under 0. Actually, it came out to a negative $65 billion. And that was if the Federal Government continued to help fund the project at the present rate, something that seems increasingly unlikely.
What about all the other Southwestern states that, defying all logic, are still talking about building their own high speed rail systems? Let me give you some advice: Call Uber instead.