Why doesn't adding lanes cure traffic jams?

Adding lanes rarely cures traffic because a faster road invites more driving until it is full again. That effect is called induced demand: a variety of interconnected effects that cause new roads to quickly fill to capacity. When the road gets freer, some people switch from other routes or other times of day, some switch from the train or bus, some take trips they would have skipped, and some even move farther away.

It is an old observation. In 1930 an executive of a St. Louis electric railway told a commission that widening streets simply produces more traffic. Later, in Robert Caro's account, planners watched New York's Robert Moses open one bridge to relieve another, and then another to relieve that one, with all three as jammed as the first had been. Engineers still joke that adding capacity to cure congestion is like loosening your belt to cure obesity.

The numbers are striking. In a 2004 review of dozens of studies, a 10 percent increase in lane miles produced an immediate 4 percent increase in vehicle miles travelled, and the increase climbed to 10 percent within a few years, the entire new capacity. Planners often miss this because they forecast traffic from car ownership and economic activity, and do not count the traffic a new road creates.

The effect works backwards too. When road space is taken away, traffic does not just move elsewhere: across dozens of case studies about a quarter of it vanished.

Jeff Speck called induced demand the one professional certainty that every thoughtful person acknowledges, yet almost no one acts upon.