Building more roads will not cause more congestion

Our new report “Twin Cities Traffic Congestion: It’s No Accident” lays bare the scale and causes of congestion in the Twin Cities. The amount of time the average Twin Cities driver spends stuck in traffic quadrupled between 1982 and 2014, from 12 hours to 47 hours. Annually, congestion costs commuters and businesses in the area close to $4 billion. The report also sets out some solutions, including road pricing and more road building.

But won’t building more roads be a waste of time? Doesn’t increasing road capacity simply bring more cars onto the roads and lead to more congestion?

That is the logic of ‘induced demand.’ This says that after the supply of a good increases, such as miles of freeway, more of it is consumed. The basis of this argument is a 2009 paper called ‘The fundamental law of road congestion: Evidence from US cities’ by economists Gilles Duranton and Matthew A. Turner. The argument is a non-starter.

How do we measure congestion?

Duranton and Turner’s measure of congestion is something called “vehicle kilometers traveled (VKT)”. This is the total kilometers traveled by motor vehicles on the highway system during a given period of time.

Clearly, this is not a measure of congestion. Congestion is characterized by slower speeds, longer journey times, and increased vehicular queueing. To quantify congestion we would look at average speeds or journey times, not how many miles people drove. It is possible, after all, that longer journey times resulting from congestion might encourage people to drive less and so reduce VKT. Indeed, some have used exactly that argument in favor of less road building. Conversely, if journey times are lower, motorists might be encouraged to drive more and VKT would increase.

More driving =/= more congestion

Even so, Duranton and Turner claim to have discovered the “‘fundamental law of road congestion’ where the extension of most major roads is met with a proportional increase in traffic.”

In fact, Duranton and Turner find that in the average metropolitan area, between 1983 and 1993, lane miles grew by 32 percent, while driving grew by 77 percent. Between 1993 and 2003, lane miles grew by 18 percent and driving grew by 46 percent. This is what they describe as “proportional.”

Either way, in the latter period, the amount of driving increased by nearly half. Assuming that people drive because they see some value in doing so – and they wouldn’t do it if they didn’t — these drivers were better off. And, assuming, as Duranton and Turner do, that this increase of road capacity was the cause of the increase in diving, that increase in road capacity was also a direct cause of that increase in value.

Is ‘induced demand’ even a thing? 

Induced demand is a curious thing. Some have elevated it to the status of “economic principle.” Yet it is a phenomenon manifesting itself solely in urban traffic debates. Nobody ever says it is pointless to build more houses or baseball bats because people will only want more of them.

Demand is not infinite. If Duranton and Turner were right, the stretch of 316 south of Hastings would be bumper to bumper 24/7. It isn’t. Just because there is a road outside your house at 3am doesn’t mean you will drive on it. Instead, people drive because the benefits of doing so outweigh the costs of doing so. Building more roads allows people to do more of something they value — driving. Despite the spin, even this research shows that.

John Phelan is an economist at the Center of the American Experiment.