Building highways doesn't reduce gridlock, thanks to "The fundamental law of road congestion."
Toronto writer Edward Keenan is tired of hearing politicians promise that they are going to end gridlock, whether by building more highways, subways, light rail or magic monorail. But he claims that " the hard truth is that none of these ideas will reduce traffic congestion. At all." He points to a study by Gilles Duranton and Matthew A. Turner titled The fundamental law of road congestion which concludes that traffic volume, measured in vehicle-kilometers-travelled increases in exact proportion to lane-kilometers of road built. In any successful city in America this rule holds. Keenan writes:
The only thing that has historically worked in practice to reduce congestion is not something anyone’s likely to propose. Since traffic is caused by a city’s prosperity, it can be eliminated by what you could call the “rust belt” approach—have the economy totally tank. If there’s mass unemployment and the office towers downtown become vacant, then congestion will be significantly reduced. But no one wants that.
The only approach that seems to work anywhere (not including Detroit and the rust belt tactic) is a congestion charge, but few politicians in North America have the nerve to try that. Instead they promise to spend billions on roads and subways. Keenan continues:
But as of right now, no prominent politician is willing to support road pricing to fight congestion. Instead, they focus on things we know—or should know—will not work. Which is not to say we shouldn’t build transit and fix traffic lights and do the other things politicians propose. Adding capacity to roads allows more vehicles to travel on them—even if they don’t wind up going any faster—so more people get served. Adding new bike lanes encourages more people to ride, and lets them do so more quickly and safely.
But we’ll still have just as much traffic on the roads when we’re done. More in the Grid.