By Christopher M. DiPrima, NMA Board Member
Editor’s Note: We touched on induced demand—the theory that as highway capacity increases, demand does also, thereby creating a never-ending need for more capacity—in NMA E-Newsletter #599 last year. In light of the current infrastructure negotiations in Washington, and the ongoing effort to reshape urban land use by decommissioning freeways, we asked Chris, who has a Masters degree in City Planning from the University of Pennsylvania, to provide a more detailed look at the topic that he says is “ripe for myth-busting.”
While the concept of induced demand has been understood for decades, if not centuries, the seminal work in induced demand was conducted by Mark Hansen and Robert Cervero of the University of California. In their initial papers, they studied the results of lane-mile additions on freeways in California, finding that over the course of around a decade, most of the capacity added by various freeway construction projects ends up being used up, sapping the rush hour travel time benefits of the new capacity.
Cervero later published an excellent short summary on induced demand, entitled “Are Induced Travel Studies Inducing Bad Investments?” Written for a broad audience, this is a great starting point to explain the underlying relationships between new highway capacity and travel times, and bust some myths. In Figures 2 and 3 of that article, Cervero illustrates the relationships between capacity, land use, and travel time. There are a few key observations:
Twenty percent of added capacity is preserved over a six to eight-year period.
This is crucial. Twenty percent may not sound like a lot, but even a one to two percent difference in the volume-to-capacity ratio on freeways can have a huge impact because traffic congestion is a “threshold effect.” In other words, traffic congestion does not increase linearly with traffic, but rather exhibits an S-curve. We know this to be true intuitively: You can add a lot of traffic to an empty road without any change on travel speeds, but a few additional cars during rush hour can grind an entire freeway to a halt. That is why rush hour traffic is often noticeably better in the summer. Even though a very small proportion of the traveling public are on vacation in any one summer week, that small reduction in demand has a huge effect on congestion.
In short, a 20% capacity preservation after the better part of a decade is a huge success story, not a failure.
The most prevalent induced-demand behavior is “behavioral shift.”
Behavioral shift means that when people are confronted with new highway capacity, i.e., their travel time is reduced, they respond by traveling more. People might go and try out a new restaurant or see a friend on the opposite side of town which would have been prohibitively distant before the capacity was added. I would argue that this is a net societal benefit since it gives more people the opportunity to do more things.
In the most extreme case, people might move to the outskirts of an urban area because it has become more convenient to commute into the urban core. This is the typical argument wielded against adding capacity: that it will cause urban sprawl. While this is theoretically true, it is facile to suggest that people make a choice as important is their housing location as a reaction to a freeway expansion. People in the U.S. have generally been become less mobile over time, and the purely transactional economic argument of “less travel time = sprawl” ignores the many social variables which go into one’s choice of housing.
One of the other sources of behavioral shift is a change in the demand pattern over the course of a day. If peak hour travel times improve, people who would have waited until after rush hour to take a discretionary trip instead decide to take it during rush hour. Again, I would consider this to be a net benefit, with more people traveling when they want rather than waiting for rush hour to end.
The next-most prevalent induced-demand behavior is “land use shift.”
This pattern is most easily explained by example, where an outlet mall sprouts around a new freeway interchange. This should not be a surprise because it is often part of the reason why new capacity is added: to increase the value of the surrounding land. This is exactly the same reason why developers in the late 1800s built streetcars: they built the loss-leading mass transit system to increase the value of their own land. It is also for exactly this reason why transit systems in the 1800s are wildly inefficient by modern standards – they followed the paths that the developers chose, often in direct competition with neighboring developers. To this day, the New York City subway system operates as a legacy of the competing IRT, IND, and BMT systems.
Personally, I interpret both forms of induced demand to be parts of the same story. Just like it is a choice to up-zone the area around a freeway interchange for commercial use, it is a choice to build housing on the periphery of a region because the freeway provides access. If we believe that we can limit freeway expansion, then we should be equally capable of limiting urban sprawl without using congestion as a weapon. Indeed, the green belt development limits first put into organized practice by disciples of Ebenezer Howard’s “garden city” movement predate freeways by almost 50 years.
The bottom line is that while induced demand is real under certain conditions, it is neither infinite, uncontrollable, nor inevitable.
Part 2 of A Primer on Induced Demand will appear in next week’s NMA E-Newsletter.