A free-market energy blog
Random header image... Refresh for more!

Is Rail Really a Fuel Saver? (rethinking a rationale for Obama's National Transportation Plan)

[Editor Note: Transportation expert Randal O'Toole is a senior fellow of the Cato Institute and blogs at Antiplanner. His bio is at the end of this post.]

Amtrak, the American Public Transportation Association, and other passenger-rail advocates want everyone to believe that passenger trains are more energy efficient than driving. This helps them justify the hundreds of billions of tax subsidies they receive. But is this rationale true?

Comparing the Studies

A new study from the University of California (Davis) finds that the answer depends on such things as load factors: your auto carrying four people consumes a lot less energy per passenger mile than a subway (which on average is only one-sixth full) or Amtrak train (which on average is only half full).

The Department of Energy’s Transportation Energy Data Book says that, on average, cars consumed about 3,500 British thermal units (BTUs) per passenger mile in 2006, while Amtrak used 2,650 and rail transit (including light rail, subways, and elevateds) just under 2,800 (see tables 2.13 and 2.14). But this doesn’t mean your  hometown should build light rail or be connected to a national high-speed rail line.

For one thing, the national averages for rail transit are heavily weighted by the New York City subway, which by itself carries more than 55 percent of all rail transit passenger miles and which consumes only 2,150 BTUs per passenger mile. When rail transit lines are evaluated individually (as I did in a recent Cato Institute report), the vast majority consume more BTUs per passenger mile than the average auto.

Almost no rail transit lines in the country are as energy-efficient as a Toyota Prius, which consumes about 1,650 BTUs per passenger mile. This confirms something the late economist Charles Lave wrote in the Atlantic Monthly in 1979: efforts to save energy will be more effective if they focus on increasing the energy efficiency of the kind of transportation people use most — automobiles — rather than trying to get people to take transit instead of driving.

A different correction is needed when comparing driving with high-speed rail. The Transportation Energy Data Book assumes the average car carries 1.57 people. The Center for Clean Air Policy used an average of 1.6 people per car in its analysis concluding that high-speed trains will reduce CO2 emissions by 6 billion pounds per year.

The 1.6-people-per-car figure applies to all auto travel, two-thirds of which takes place in cities. But according to an independent analysis prepared for the California High-Speed Rail Authority, the average automobile in intercity travel carries 2.4 people, which means it consumes only 2,300 BTUs per passenger mile.

“Intercity auto trips tend to [have] higher-than-average vehicle occupancy rates,” said another Department of Energy report in 2001. “On average, they are as energy-efficient as rail intercity trips.” Moreover, the report added, “if passenger rail competes for modal share by moving to high speed service, its energy efficiency should be reduced somewhat — making overall energy savings even more problematic.”

Outside the Boston-to-Washington corridor and a few other places, most Amtrak trains have top speeds of 79 miles per hour. Boosting them to 110 mph, as the Obama high-speed rail plan proposes to do, will greatly increase their energy consumption. This means high-speed trains will be less energy efficient unless the higher speeds attract more riders per train.

The Center for Clean Air Policy’s report assumed that all high-speed trains would be 70 percent full. This is extremely optimistic: in 2008 — a banner year for Amtrak — the average Amtrak train was only 51 percent full (see p. C-1 and divide contribution per seat mile by contribution per passenger mile). The Acelas, Amtrak’s fastest trains, were 62 percent full, but Amtrak’s 110-mph Boston-to-Washington trains were only 48 percent full, while its 110-mph Philadelphia-Harrisburg trains were only 34 percent full. Only two Amtrak trains, neither very fast, were more than 70 percent full in 2008, and both were less than 70 percent full in 2007.

Some of the least energy-efficient vehicles in America are public transit buses, which the Transportation Energy Data Book says consume about 4,200 BTUs per passenger mile. On the other hand, some of the most-energy efficient motor vehicles are private intercity buses, which the American Bus Association says consume only about 750 BTUs per passenger mile (and if you don’t trust that source, DOE estimates are even lower).

What’s the difference? Private intercity bus companies have an incentive to fill up their coaches. But public transit agencies (along with Amtrak) are politically driven to serve all taxpaying areas, regardless of the demand or population density.


If environmental groups really wanted to save energy, they would advocate strategies to privatize transit, privatize Amtrak, and sell highways to private entrepreneurs who would have an incentive to relieve the congestion that wastes nearly a billion gallons of fuel each year. The real goal of at least some environmentalists, however, is not to save energy, but to reshape American lifestyles. They just can’t stand to see people enjoying the freedom of going where they want to go, when they want to get there.

This contradiction should inspire internal debate and external reform.


Randal O’Toole specializes in the areas of urban growth, public land, and transportation issues. His research on national forest management culminated in Reforming the Forest Service (1988), which had a major influence on Forest Service policy and on-the-ground management. His analysis of urban land-use and transportation issues, brought together in his The Vanishing Automobile and Other Urban Myths (2001) has influenced decisions in cities across the country.

His most recent book, The Best-Laid Plans, calls for repealing federal, state, and local planning laws and proposes reforms that can help solve social and environmental problems without heavy-handed government regulation.

O’Toole travels extensively and has spoken about free-market environmental issues in dozens of cities. An Oregon native, O’Toole was educated in forestry at Oregon State University and in economics at the University of Oregon.


1 Richard W. Fulmer { 06.12.09 at 8:04 am }

Perhaps H. L. Mencken had public transportation in mind when he stated: “There is always an easy solution to every human problem – neat, plausible, and wrong.”

2 Andrew { 06.12.09 at 10:11 am }

And one might add, Not all human problems are…

(Okay, I’m not the one who first said that, but I can’t remember where I heard it)

3 Sid Burgess { 06.20.09 at 1:11 pm }

I never thought that the train was an alternative to the vehicle solely based on BTU consumption. Nor do I believe that most towns considering rail are using that as a talking point.

Rail, as it has for nations all around the world, provides an alternative to the rivers of steel and concrete a growing population of cars demand. This isn’t an issue about whether one may get better gas millage than the other, it is an issue of, should the government subsidize highway construction at such astronomical rates thereby forcing communities to pick between no money or highways. As long as the government is in charge of handing out transportation funding (which I don’t support in the first place), I applaud cities for considering the alternative or rail.

Granted, not all projects will be viable, which is why the money should come from local sources. Accountability and local peer review would scrutinize any project doomed to failure.

But do we need to list the countless highway projects that have equaled nothing but a complete waste of taxpayer dollars.

Free market principles demand that we first object to the Federal government choosing our infrastructure for us, and secondly, that we protect our cities from building infrastructure that isn’t sustainable.

I am curious if you would, or have done, a post outlining the life-cycle costs of both forms of transportation. It is my understanding that the train has much better efficiencies over a long period. What have you discovered?

I enjoy your work and appreciate you working hard to promote free market principles.



4 Do Passenger Trains Save Energy? Another Look — MasterResource { 07.09.09 at 1:01 am }

[...] response to my recent post about whether passenger trains save energy, a reader asked about the “life-cycle costs of [...]

5 BCassidy { 07.09.09 at 4:03 pm }

Even 110 mph is pretty slow. I wouldn’t be inticed to take that over just driving from Houston to Austin or Dallas myslef. Not when you consider you have to park, get in line, and then get transportation at your destination.

I’d be more interested in the European capacity numbers. Do they hit 70+ %? I would be MUCH more likely to ride knowing I could get to Dallas in about 2 hours.

It seems obvious, but each area/region should evaluate its needs and then decide if they need rail. I live in Houston (used to live in Austin), and I think we need a high speed (200+ mph) rail system. The distances between San Antonio, Dallas, Austin, and Houston are pretty big. I’d also connect up to Baton Rouge. There are a lot of people traveling between Houston and Baton Rouge because of the industry here. I think it would be very popular, decrease energy use, and more importantly, probably save quite a few lives by taking people off the road.

6 BCassidy { 07.09.09 at 4:06 pm }

Not to mention, I’d much rather read a book or take a nap for a couple of hours rather than drive the hellish nightmare of I-10, I-45, or I-35 (one of the deadliest roads in the country) for 5 to 6 hours.

Leave a Comment