Braess’ Paradox and Building Roads
In the linked article from CityLab, the writer discusses different state departments of transportation and their propensity to add or widen roads in order to reduce traffic congestion. More specifically, and against this trend, California’s DOT actually recently acknowledged that increased road capacity does not necessarily mean less traffic, and can even have the opposite effect. Essentially, while most states’ DOTs spend most of their budget on building new roads, these new roads aren’t actually solving the traffic problems they were intended to target. This is the result of induced demand. Added road capacity, according to California’s DOT and the policy brief they cited, increases traffic by the same amount, if not more. The added, or induced, traffic is a combination of redirected traffic from now-slower alternative routes, and new traffic, from people who otherwise would have used another form of transportation, or would not have traveled at all. Because of this induced traffic, roads end up just as congested as they were before, and in fact total environment pollution may even increase due to the decision of some travelers to use cars over more efficient methods of transportation.
What the article describes is Braess’ paradox, which we discussed in class while covering network traffic, whereby adding a new path in a directed network can actually increase the average cost of traveling from source to sink. Just as with the examples in class, a state widening a highway or building a new road can attract more drivers to that route by (originally) reducing the time cost of that route versus the other alternative routes with the original traffic situation. But, once drivers adjust and a new equilibrium is reached, the average time cost is higher than it was before without the added capacity. This happens by redirecting drivers from more efficient routes that better scale with more people, or by pushing more people to drive at peak times.
In being a paradox, of course, this reality is difficult for most DOTs to understand, or at least, incorporate into their thinking. The article quotes Connecticut’s chief of planning for their DOT as saying “You can’t build your way out of congestion,” even though the state has multiple highway widening projects planned and currently under way. If they were to properly digest the evidence for Braess’ paradox, states would better spend their money on true improvements that do indeed reduce congestion, such as public transportation.