Internet Explorer 11 is not supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

Searching for Ways to Limit Induced Demand in a Car-Loving Society

Reducing congestion and its problems of pollution and carbon emissions won’t be easy or cheap. But transportation experts continue to search for answers.

A high occupancy vehicle (HOV) lane near Seattle.
Editor’s note: This is the second in a two-part series about induced demand. Read the first article here.

American transportation planning is all about getting people to travel more and ensuring that experience is as frictionless as possible. Congestion mitigation is the name of the game.

For much of the post-World War II period that meant building roads, and not much else. In the wake of the 1956 Federal Aid Highway Act, the federal government began pouring $20 billion to $25 billion a year into expanding and maintaining the highway system. All that extra capacity, however, did not make driving a pleasure — especially during peak hours.

As the well-documented concept of “induced demand” or “induced travel” demonstrates, building more highways does not ease congestion. Instead it attracts more drivers, creating more deadly air pollution and more carbon emissions that contribute to climate change.

“For a long time, transportation policy’s goal was to get people to travel more, haul goods, go drive those interstates,” says Amy Lee, researcher at the University of California, Davis. “That was before we understood a lot of the air pollution and greenhouse gas externalities of driving. We're trying to think differently in an age of climate change, but it's a hard ship to turn around.”

In the last few decades, the concept of induced travel has become widely accepted in academic and planning circles. But its implications have not often been acted on. Partially that’s because of divisions within state departments of transportation, and because politicians are incentivized to endorse what seems like an easy answer: Surely traffic would ease if we widened this highway a little?

In recent years, transportation departments in California, Colorado and other states have incorporated the concept more thoroughly into their policy and planning. The U.S. Department of Transportation under Pete Buttigieg has acknowledged the concept, and in theory has more ability to influence state and local decisions due to the discretionary funds in 2021’s infrastructure law.

But in a society where mass car ownership has been the norm for at least three-quarters of a century, how do you solve a problem like traffic congestion?

Public Transit Alone Is Not the Answer

In urban centers where traffic congestion is especially bad, and mass transit is available, it is often argued that more train or bus service will make driving easier. If you get more people out of their cars, the vehicles that remain will have a better time of it. This idea has been used to sell mass transit initiatives to the public from Nashville to Los Angeles.

But evidence suggests that building more public transportation alone doesn’t solve congestion. New transit lines may initially succeed in luring drivers off the road and easing traffic. But in the medium-to-long term, the effect would presumably be similar to building another lane of highway (although without the same environmental harms). As driving becomes easier, people will notice and take advantage of that fact.

“To the extent that transit draws drivers away, it may induce other people to start driving more,” says Jamey Volker, researcher at the National Center for Sustainable Transportation. “It might not be a net increase in induced travel: You may just have the people who left to take transit replaced by people who still drive, but are driving more.”

The research on this question isn’t as strong as the studies on the induced demand effects of new highways, but what there is suggests outcomes would be similar. One of the more influential studies that quantitatively calculated induced travel demand found that transit wasn’t a great boon for easing congestion on its own.

“If you divert a trip from a car to a bus, then you're just creating highway space that someone else will fill up just as if you create a new highway,” says Matthew Turner, professor of economics at Brown University.

In 2009, he published a study with Gilles Duranton that showed a close to perfect correlation between lane miles constructed and vehicle miles driven. But less well known is that their report found little evidence that new public transit service would ease traffic.

“The empirical evidence for that claim is not as strong as for the claim that adding one percent of lane miles causes a one percent increase in vehicle miles traveled,” says Turner. “But it's suggestive that transit doesn't lead to a decrease in miles traveled in a city.”

Turner says there is some evidence that if you build a subway line next to a particular length of highway, that specific road will see less traffic. But it still doesn’t reduce vehicle miles traveled in the city overall.

For economists who study induced travel and link it to basic questions of supply and demand, the solution to traffic is to appropriately price driving during peak hours. People are incentivized to drive because there are no upfront costs to getting behind the wheel, alone, during rush hour.

Many policy experts herald the concept of congestion pricing, which can either operate as a cordon — where those who enter a particularly busy area are charged — or as fees on particular roadways at set times.

“If you want to manage congestion, you want to think about pricing roads,” says Turner. “People are resistant because they don't like paying for things that they're used to having for free.”

Fees and tolls are easy political targets, especially in the context of the United States where New York City is the only jurisdiction that’s approved a cordon-type approach. (Even in that case, the policy’s implementation has been delayed time and again.) There are roads where such charges are utilized in Minnesota, Northern Virginia and California, where the state DOT has experimented with high occupancy vehicle (HOV) and high occupancy toll (HOT) lanes.

Turner believes that more American regions will move in this direction eventually, because congestion is a big problem and this is the only available policy fix. But he admits that it may be hard to sell, especially in today’s easily weaponized political environment.

“This is a market-based solution, it should be what the Republicans are all about,” says Turner. “This is not government handouts of infrastructure. I would think that if anybody objected to it, it would be Democrats. But that's not what's happening. Democrats all live in cities where they have to deal with congestion. They have to actually solve this problem.”

Road pricing isn’t only controversial in the United States. Efforts to institute congestion pricing in the British cities of Manchester and Edinburgh were defeated by popular vote. Stockholm, Sweden, did approve congestion pricing by popular vote, but only after first experiencing its benefits.
A sign notifying drivers about a congestion charging zone in London.

Once and Future Congestion 

There is plenty of research that suggests tolling roads more heavily and cordon-style fee systems both ease congestion. But it is not without complexities and there is some evidence that absent truly radical action, congestion is just a price we pay for a car-centric society.

In California, the Department of Transportation incorporated induced travel into their policymaking in recent years and is trying to change their roadways to ease congestion. But a 2021 study of HOV and HOT lanes, which create incentives for carpooling, did not change the basic logic of induced travel. Traffic flows, on net, actually increased.

The basic logic of induced travel still holds, it turns out, even if the lanes added are HOV or HOT. Toll lanes make it easier to drive for those who can afford it, which increases driving. General-purpose lanes will then see an ease in congestion, making them easier to navigate and resulting in an eventual return to earlier levels of traffic flow. Even adding high occupancy lanes increases the total amount of traffic, vehicle miles driven and emissions generated.

“We reach a new equilibrium, where the total traffic flows increase: You add the new flows on the toll lane to the flows on the general purpose lanes that have returned to the same pre-expansion level,” writes Volker in an email message.

There are also limits to the cordon policy, Volker points out. London’s famous congestion pricing scheme resulted in short-term reductions in congestion, but long-term journey times returned to previous levels. A 2018 paper by David Metz reported this finding: “While it is difficult to mitigate congestion by means of congestion charging, it is possible to raise revenue.”

Metz’s findings can also be read this way: It’s not that cordon-style congestion pricing doesn’t work at all, but that continual adjustments are needed to reflect macro transportation shifts like the rise of ride-hailing, economic trends like inflation or micro changes in driver behavior to evade the policies. In Singapore, for example, prices are adjusted every three months in response to changes in driver speeds: If congestion is picking up again, the fee is increased.

Maybe that’s part of why Metz only found congestion pricing continually successful in Singapore, where car ownership is also made extremely expensive. But policy solutions that revolve around making driving very expensive are unlikely to be adopted in the U.S. Similarly radical solutions like Jakarta, Indonesia,’s conversion of all general-purpose lanes on two major roads to HOV lanes during peak hours are not being considered here. (The Jakarta law successfully eased travel times substantially, but when the policy was suddenly eliminated, congestion came roaring back.)

Are we just doomed to live with traffic congestion? The answer in high personal car-ownership societies seems to be a qualified yes. In the long term, electric cars could ease the harms brought on by air pollution and the climate harms of greenhouse gas emissions. Adding public transit and facilities for walking or biking are good in their own right, especially if paired with urban design that enable a happy life without a personal car. Reducing road capacity and parking options in places where there are alternatives to car transport is an option as well.

But in locales where development patterns are auto-centric and require people to drive to do anything, congestion will always be a problem.

“A lot of people who've been in the field longer than I have come to the conclusion that it's an inevitable part of urban life,” says Calvin Thigpen, director of policy research with Lime, the micromobility company. “If your goal is to alleviate congestion, it will be hard to achieve. If your goal is environmentally responsible accessibility, endlessly expanding highways is a losing proposition and the alternatives are much more attractive.”
Jake Blumgart is a senior writer for Governing and covers transportation and infrastructure. He lives in Philadelphia. Follow him on Twitter at @jblumgart.