Induced Traffic, Congestion, and Peak Spreading Redux

I’ve written often about induced traffic and its near cousin congestion. But they’re topics for which the lessons are non-intuitive and therefore bear frequent repetition. There were times in my life when I would have looked askance at the concept of induced traffic and been in personal need of recurring lessons. (I’ve learned to be more humble about my intuition. The world is often more complex and counterintuitive than my instincts tell me.)


I’ve had conversations with folks to whom I carefully explained the phenomenon of induced traffic, including how it makes congestion almost impossible to solve. The other parties nodded knowingly as I spoke, apparently following the thread of the argument just fine, and then, as their first question after I finished, asked “So what new roads must we build to solve congestion?”

Luckily, there are plenty of folks who are eagerly accepting the challenge of explaining induced traffic, including the answer that, within practical limits, there are no roads that will solve congestion. (For the benefit of those in Petaluma, the answer has significant implications for the on-going Rainier Crossing discussion.) Today, I’ll share the work of a couple of those folks.

Writing in Wired magazine, and a veteran of the Los Angeles freeways since early childhood, Stuart Dee reviews the current academic thinking on induced traffic, reporting the finding that traffic seems to increase in lockstep with newly constructed road miles.

As Dee summarizes the conundrum, “… we humans love moving around. And if you expand people’s ability to travel, they will do it more, living farther away from where they work and therefore being forced to drive into town. Making driving easier also means that people take more trips in the car than they otherwise would.”

Dee’s solution is one that I’ve previously endorsed, congestion pricing, which is a modulated approach to a vehicle mileage tax. Drivers, faced with a steep fee to drive through a particular location at a specific time, let’s say the financial district of San Francisco during the working day, will look for options to make their trip at another time. If even just a few drivers adjust their trips, traffic congestion would ease and the streets would be more efficiently used for more hours of the day.

EWashingtonStreet260Looking at the same problem but from a different angle, Chuck Marohn of StrongTowns, based in Brainerd, Minnesota, likens a hierarchical road system, with local streets feeding onto collectors feeding onto arterials, to a water basin, with brooks feeding into streams feeding into rivers.

Marohn notes that downstream flooding is almost always the result of too much water escaping the upper elevations of the basin and overtopping the riverbanks in the lower elevations. He then argues that the same is true of traffic systems, with too many trips congregating on the arterials where they become congestion.

The current regulatory approach to flooding is to require strict controls on the outflow from new development, typically through detention or retention.

Marohn contends that same approach would address congestion. He argues that we need to build neighborhoods from which car trips never reach the arterials or in which daily tasks can be accomplished without cars. In Marohn’s words, “For nearly seven decades, our national transportation obsession has been about maximizing the amount that you can drive. We now need to focus on minimizing the amount you are forced to drive.” Luckily, we know how to do the latter. It’s urbanism.

Two insightful writers with different life histories and different lifestyles looking at induced traffic and congestion and reaching similar and complementary solutions. I use find that a sign of a fundamental truth, which beats the heck out of flawed intuition.

Before closing, I should note a related concept I recently came across. At a public meeting of the Sonoma County Transit Agency, I asked a couple of SCTA planners and traffic engineers about their thoughts on induced traffic. They admitted that their task was more the building of roads, but they were nonetheless well aware of the induced traffic phenomenon in their work.

But rather than calling it induced traffic, they called it “peak spreading”. Their observation was the drivers who had time flexibility and an unwillingness to spend time in traffic adjust their travel times to drive or after the known peak time. As a result, the duration of peak traffic gets longer over time.

I can confirm the description. When my wife and I have a late afternoon appointment in Marin County, we make plans for dinner and perhaps shopping afterwards, preferring to burn a couple of hours rather than sitting in traffic. We are induced not to travel during peak congestion and instead induced to travel at a slightly off-peak time.

And of course, the SCTA observation of peak spreading gives a window on the future when the third lane of 101 is built between Novato and Petaluma. Rather than peak congestion diminishing, the principal effect will be reduction in its duration. And then the duration will re-expand as induced traffic takes advantage of the new lane.

It’s a painful, but quite real, truth. We’re not going to build our way out of congestion. We’ve had a nearly hundred years of trying and it just doesn’t work. Not only is the failure empirically evident, but we now have a theoretical basis that confirms the empiricism. And we have solutions, in congestion pricing and urbanism, to better manage congestion.

Do I think that all readers now fully grasp induced traffic and the resulting congestion? Nope, I know that counterintuitive realities take time to gain a foothold. But I’ll keep sharing the work of engaging writers like Dee and Marohn and hope to someday have worn away the granite of flawed intuition.

Your questions or comments will be appreciated. You can email me. And thanks for reading.

This blog entry was originally published here.

Check out Dave Alden’s blog: Where Do We Go from Here?