What makes one car more efficient may not work when you apply it to a city.
For years, progressive urbanists and environmentalists have advocated for synchronized traffic lights. Syncing lights, theoretically, makes everyone's lives easier: They promote a sense of flow and easiness on the road, and they reduce pollution, because a car running smoothly runs cleaner than a car stopping-and-starting.
So, the need to sync traffic lights has become somewhat well-known. The Baltimore Sun's transportation reporter wrote in 2010 that the "most common source of complaints" he heard from readers was out-of-sync lights. In 2011, a libertarian think tank praised Georgia's effort to syncronize lights, citing statistics about reduced drive times and gasoline usage. And sometime this year, according to an LA Daily News report, Los Angeles will complete its three decade-long effort to syncronize traffic signals across the city.
Except synchronizing traffic lights may not actually work. Todd Litman, of the Victoria Transport Policy Institute, writes in a recent post that efforts which increase flow, like signal-syncing or expanding-road-capacity, only theoretically reduce emissions:
[R]esearch suggests that at best these provide short-term reductions in energy use and emissions which are offset over the long-run due to Induced Travel. Field tests indicate that shifting from congested to uncongested traffic conditions significantly reduces pollution emissions, but traffic signal synchronization on congested roads provides little measurable benefit, and can increase emissions in some situations.
(I've stripped that quote of its citations.)
Why? Because if you make it easier and cheaper and faster for people to drive, more people drive. It's Jevons paradox, applied to the city -- if you make it more efficient to use a resource, more of that resource will get used.
There's a fun fallacy at work here. Changing the conditions of traffic flow changes the entire environment of the city. And re-creating conditions across the scope of a city which make one car more productive fail to account for the fact that, in a city, you're never just dealing with one car. You're not even dealing with one car many times. You're dealing with a whole environment of cars, and traffic-light-syncing, while it leads to many single cars having a faster trip, doesn't account -- as a policy -- for changes to the whole environment.