Building more roads to reduce traffic congestion doesn’t work. A comprehensive examination of all American cities shows that the amount of driving increases in lock-step with the expansion of the road system.
University of Toronto economics professor Dr Matthew Turner told a conference last week that stunningly consistent findings suggest cities can’t build their way out of congestion and that building more road capacity simply results in more driving. The Solving Gridlock Forum organized by Transport Futures heard from experts from across Canada and the United States and also hosted a transit debate between the four main Toronto mayoralty candidates.
Turner’s team examined publicly-available data over several decades for every American metropolitan area with a population over 50,000. It first looked at amount of travel of the interstate highways within those cities and compared it to the amount of driving on that system, and found a consistent relationship.
“If you have a city with 10% more roads there’s 10% more driving,” Turner told the conference. “If you take two cities which otherwise are about the same, one has 10% more miles of interstate lane kilometres, you will see that there are about 10% more driving on those interstate lane kilometres.”
The review also found the same relationship “for other classes of roads within those cities”. And the pattern also held when changes to road capacity were examined. For this part of the analysis the team compared data from 1980, 1990 and 2000.
“If you add 10 percent lane kilometres to any given city then on average the amount of driving increases by 10 percent,” he reported. “So both of those things line up in a way that you would expect if roads were causing driving.”
A third test looking just at cities over 5 million got the same results and a replication using data from Japanese cities generated the same conclusions.
“This is as good as it gets in social sciences,” Turner said. “This is as close as you’re going to get to a fact.”
Where the additional traffic comes from Turner said is more difficult to determine with existing data, but most of it appears to come from changes in individual driving behaviour.
“When you add roads you get more traffic primarily because households change their individual behaviour: they drive more. So they will live a little further from work, they will drive further to get to a restaurant; they will be more likely to go out for shopping trips.”
He also found about a quarter of increased traffic on the interstate highway system came from long-haul trucking – echoing a complaint by Hamilton councillors over heavy truck use on the Red Hill Parkway.
“Long haul truck traffic seems to be very opportunistic. If you put more capacity through your city, the long-haul truckers are there to use it.”
Turner’s studies also found that when you compare time spent driving to road capacity, a doubling of road space generally resulted in a 10 percent decrease in average speed of travel.
“So if you build roads in response to traffic congestion, you will elicit this extra demand, you will suck more users into their cars and scale up your city and things will get slower,” he reported. “So you can’t build your way out of traffic congestion and this says you can’t even stay even.”
Turner cautioned that his findings indicate expanding transit is unlikely to reduce congestion because any resulting road capacity will quickly fill up. This isn’t a reason to not expand transit services, but he argues that should be done where the cost is justified by the resulting growth in transit use.
Turner suggests that road pricing is a good option given that his studies show it takes very little to change driver behaviour because so much of it is optional or possible to shift to less busy periods of the day.