Is transit as green as it should be?

Cato Institute "scholar" Randal O'Toole recently published a report claiming to show that public transit produces about the same amount of carbon emissions per passenger mile as autos, and far more than hybrid cars. O'Toole argues that it would make far more sense to subsidize the purchase of hybrids than to keep investing in expensive new light rail systems, as many cities are doing now.

Libertarians, of course, hate public transit because it involves government planning in pursuit of the public good. So it's not surprising to find that O'Toole has deliberately distorted his data to make transit look bad. Even so, O'Toole has compiled a lot of very useful information comparing transit systems in different cities and raised an important question: are transit agencies doing as much as they should to reduce their carbon emissions? But before we get to that, let's look at why O'Toole's political agenda is so badly flawed.

O'Toole's data show that the average auto uses 3885 BTUs per passenger mile (only 3445 if you leave out SUVs and pick-ups), compared with 4365 for buses, 3465 for light rail, and 2600 for heavy rail. A Toyota Prius uses 1659 BTUs. When you average all this out, passenger cars use only one more BTU per passenger mile than transit does. And O'Toole argues that, since the fuel efficiency of cars is likely to rise rapidly in coming years and the number of light trucks will decline (probably fair assumptions if gas prices stay high), cars will soon be clearly better from a global warming perspective.

What O'Toole does not tell us (I had to email him to find out) is that the fuel efficiency numbers he's using for cars come from both urban and highway driving. This makes his comparison between cars and transit completely invalid. Cars are far more efficient on the highway, but the only kind of driving that transit replaces is the kind that involves frequently sitting in traffic and lots of stops and starts - urban driving. So O'Toole is comparing apples and oranges in an attempt to smear transit, meaning that despite the many problems with public transit, it is still far greener than driving.

Second, the biggest reason transit's already good numbers aren't higher is low ridership (as O'Toole acknowledges, most of the energy expended by buses and trains is used to move the vehicle so adding transit riders dramatically increases energy efficiency). It's not at all surprising that, with fuel prices held artificially low by a number of indirect subsidies, potential riders have chosen to drive rather than take transit (ridership figures and thus efficiency numbers have been steadily rising with the price of gas). So transit's efficiency numbers have in part been dragged down by decades of public policies favoring cars.

One big exception is the New York subway system, which is highly energy efficient because it has so many riders. And why are there so many riders? One of the main reasons is that the system is so big that it'll get you almost anywhere you want to go. Those cities only now building urban rail systems have low ridership in part because having only one or two lines is not very useful. It will take many years of investment before these systems are complete enough to attract enough riders to make them highly energy efficient. Yet O'Toole would end the process now, permanently crippling transit in these cities.

Finally, O'Toole ignores what may be the most valuable thing about transit from an environmental perspective - it makes possible a certain kind of very low impact lifestyle, namely urban living. People who live in urban areas not only drive less, they also walk more, have smaller living spaces to heat and cool, and take up less land. All of these cause far less greenhouse gas emissions than suburban-style living, but none of them show up in O'Toole's figures. Yet that kind of living is just not possible without a good rapid transit network, including rail. In Hyde Park or Cambridge I make most of my trips by bike or on foot, producing no greenhouse gases at all. That's just not possible in the suburbs, yet O'Toole's figures are incapable of reflecting the smaller environmental impact that dense development, facilitated by transit, makes possible.

Even without these considerations, it's not just the amount of energy used per passenger mile - it's also the total number of miles you need to move. People who take transit not only require less energy per mile, they also travel fewer miles overall because everything's closer together in the city. O'Toole's numbers don't include this either.

Okay, so O'Toole has both misrepresented the efficiency comparison of transit and autos, and he's ignored the key role that transit plays in making urban living possible. But if we just forget about his political agenda, he still gives transit advocates some useful information and ideas.

First, he points out that there's actually major variations in the efficiency of different cities' transit systems. The El is in the middle of the pack among heavy rail systems, a lot less efficient than New York (and Atlanta?!) but better than LA and some smaller systems like Cleveland. O'Toole doesn't explain these variations, but it's worth looking into.

Second, rail is far more efficient than buses. There are many reasons for this - rail technology is inherently more efficient, but the political imperative to extend bus service to all parts of cities even when ridership is very low is also a big factor. O'Toole makes some good suggestions for reducing buses' environmental impact, including switching to hybrid buses and using smaller buses on less popular routes (the CTA is doing both of these to a certain extent). We should also seriously look into establishing bus-only lanes on key corridors like Western and Ashland and giving buses signal priority, both of which would reduce the traffic problems that eat up so much gas.

The CTA could also follow New York's example by using green building and station design and getting actively involved in transit-oriented development. It's not enough that transit is already far greener than driving - there's a lot more progress to be made.