More Roads, Same Congestion

Induced Demand.

For a long time, the prevailing strategy for dealing with congested roads has been to build new roads or lanes.

“I gotta fever and the only prescription is MORE ROADS!!!”

This typically has an unexpectedly short lasting impact. Often within a year, traffic seems as congested as it was before the “improvements.” At this point people start to wonder what kind of planner thought this would work, if the engineers knew anything about what capacity they truly needed, or if government bodies really deserve all that tax revenue.

Expectation. Photo by Blair Fraser on Unsplash
Reality. Photo by Jens Herrndorff on Unsplash

This problem arises for many reasons. One being that roads are most cost-effective when they are carrying a certain number of cars. It’s a high enough number that when this many cars show up, it results in decreased in driving enjoyment for all. In other words, roads are designed to be pretty full. Travel speeds are slightly decreased by this optimum traffic volume and free movement is restricted between lanes. It’s fairly intuitive when you think of roads as corridors that move money around in the form of products and labor. Barely used roads aren’t moving a lot of money around, despite costing a lot to build. Ultimately, roads aren’t designed for easy movement of drivers, they’re designed to balance between trade-offs of driving pleasure and efficient land use.

Another cause of the short effective life of increased road capacity is an idea that is getting more attention these days, Induced Demand. It might not currently be a mainstream idea, but it is intuitive when you get down to it. It’s originally an economists’ term, but is now also used in transportation planning and design. It is a situation where increased supply is matched proportionally by increased demand. Put simply, big roads CAUSE big traffic. You don’t need to get very technical to explain this:

  • Some road improvement happens: A lane is added, overpasses are built to replace traffic lights, a two-way road is replaced by a one-way, etc.
  • Trips are now shorter; what used to take 50 minutes may now take 20 minutes.
  • Congestion is relieved and traffic flows more freely for a few months.
  • People who avoided this route before, now reconsider and come back to it due to its increased performance.
  • Suddenly workers that have discovered or re-discovered this improved route tell one or two of their friends at work, they tell a few others, and so on.
  • A few people realize that despite a little back-tracking, this is now the fastest way to Ikea (or other weekend destination). They mention it in passing to friends and family.
  • There is a commute time threshold. It’s the concept behind the “drive until you qualify” expression. Most people will choose 45 minutes of driving for much cheaper real estate. With the road improvements 45 minutes of driving now takes a person farther from their destinations.
  • Suddenly developments further afield can say “Only 25 minutes to downtown!” instead of the former 50 minutes. This results in a development boom on the edge of town, cheap land is now slightly profitable to build upon. It’s all about quick-to-market volume for developers.
  • In some cases, former park-and-ride commuters decide to just drive now.
  • In a matter of months, all of these additional drivers meet or exceed the new capacity and the road feels just as congested as it did prior to the expensive “improvements.”

A lot of time, roads are planned and built using past traffic count data with estimations of future volumes based on past trends. They do not account for Induced Demand. In almost all cases, projected volumes fall quite short of what is actually observed when the effects of Induced Demand are ignored. This is fairly intuitive, but until recently, Induced Demand wasn’t a major consideration in transportation planning and it is still often overlooked. Induced Demand applies differently to all cases, but the effect is always significant. It is not only logical, but it has also been demonstrated in research.

While the ideal commute found by Redmond & Mokhtarian in 2001 was 15 minutes, 45 minutes seems to be where people start to reconsider their options.

And of course, Wikipedia has some information that might be useful for those looking to learn more on the subject.

On the flip side, we have Disappearing Traffic and solutions to this problem of Induced Demand.