Solutions for congestion are never as simple and easy as armchair planners and engineers would like you to believe—especially those who suggest that congestion can be solved by forever building more roadway capacity.
This article is part two of three in a series on the role of the field of planning in studying, creating, and responding to congestion. Part one, published on April 13, 2022, dug into the causes of congestion. This article will examine the failed approaches in the ongoing search for a solution to this stubborn problem. The responses to congestion are just as likely to be misleading and misunderstood as the causes of congestion, as it turns out.
Congestion Reduction Fails
Blame for congestion is sometimes conveniently assigned to avoid admitting the facts about individual and collective responsibility for congestion. Congestion relief can often be more of a public relations talking point than actionable planning practice. One of the keys to understanding congestion mitigation, therefore, is understanding the vast terrain between the science and the politics of congestion.
Projects and concepts of all kinds are touted by planners, engineers, politicians, and the public as congestion mitigation tools despite centuries of evidence to the contrary. Not all of these false promises are created equally: some projects have additional benefits, aside from congestion relief. Other projects, however, manage something worse than failing to reduce congestion as promised: they also fail to account for negative externalities—the social, economic, and environmental problems they create.
Highway and Road Widening
"You can't build your way out of congestion," according to one of the most popular maxims of transportation planning. Any transportation engineer in charge of building the nation's roads for the last century or so clearly didn't get the message. While transportation and land use plans almost gratuitously mention the need to reduce automobile trips for the purposes of reducing congestion along with environmental impacts like greenhouse gas emissions and particulate air pollution, capital investment budgets almost uniformly revert to the default of building new road capacity.
If reducing congestion were as easy as adding roadway capacity, there would already be a lot less congestion. While other real world dynamics can be solved by increasing capacity, such as installing larger drains for stormwater runoff, the dynamics of traffic flow are stubbornly resistant to antecedents, as described in part one of this series. The results of so many attempts at adding roadway capacity to ease congestion are clear every day on roads that have been expanded for decades. For two exceedingly conspicuous recent examples of the failure of new capacity to relieve traffic congestion, consider the massive highway expansion projects on the Katy Freeway in Houston and the Interstate 405 in Los Angeles. Both projects invested billions of dollars toward new capacity for frequently traveled highway corridors, only to watch congestion return worse than ever before within months of project completion.
The failure of road widening projects to reduce traffic is most frequently blamed on a phenomenon known as induced demand, or the idea that new supplies of roadway capacity will inevitably be overwhelmed by latent demand for transportation options. According to induced demand, the counterintuitive consequence of building new capacity to relieve congestion is that traffic delays will only be worse after capacity is added to the system.
Induced demand is frequently invoked by opponents of the car-centric status quo, though the attention to induced demand has yet to achieve much success in changing the minds of either the voting public or the politicians with final say over how transportation funding is spent. Despite the stubbornness of the status quo, advocates, including many professional and academic planners, are currently, actively pushing back against the idea that new roadway capacity is an effective tool for reducing congestion. The nonprofit RMI and a consortium of partners, including Transportation for America and the National Resources Defense Council, recently published an induced demand calculator, known as the State Highway Induced Frequency of Travel (SHIFT) calculator, which connects new road construction to increased vehicle miles traveled (VMT) and greenhouse gas emissions.
An increasing number of U.S. cities are also rejecting the status quo of adding new road capacity by default, choosing instead to reduce roadway capacity by removing urban freeways and reducing car vehicle lanes (a street reconfiguration approach sometimes called a "road diet") in the hopes of enabling more efficient mode choices, like walking, biking, and public transit.
Despite public perception that removing roadway capacity will exacerbate congestion or create jams where traffic flowed freely, a steady stream of evidence suggests that removing vehicle lanes does not necessarily worsen congestion. Latent demand implies that people are finding other ways to get around.
There is a widespread belief among many in the media, the public, and, frankly, the planning profession, that self-driving cars will be a panacea for congestion. Like everything else with congestion, however, the idea of autonomously optimized vehicles moving in perfect harmony with each other and always in motion is likely too simple to solve such a complex challenge.
A growing collection of research presents evidence that autonomous vehicles will only worsen congestion, including a study by researchers at the University of Western Australia, published by Science Direct in February 2021, which showed self-driving cars creating the same kind of "shockwave" and "butterfly effects" traffic congestion as human drivers. (An article by Tom Stone for Traffic Technology Today described the study in less academic terms.) Another study, this one by researchers at the University of Adelaide, showed that attitudes toward self-driving cars would prevent the full benefits of the technology. "Our evidence suggests that as riders switch to autonomous vehicles, there will be an adverse impact on public transport. With most commuters not interested in ride sharing, this could increase peak period vehicle flows, which is likely to increase traffic congestion over the next 30 years or so," write the study's authors.
Supporters of the idea of self-driving cars as a congestion solution say connected cars, shared vehicle fleets, "smart intersections," and a reduced need for parking can mitigate the inevitable backups. Skeptics predict that the complexity of transportation—with so many vehicles, headed in so many different directions, starting and stopping for so many reasons—will quickly interrupt the efficiencies of self-driving cars.
Moreover, self-driving cars will still take up a lot of space, even if parking is suddenly rendered obsolete. Transit consultant Jarrett Walker says self-driving cars are limited by a problem of geometry: "It doesn’t matter if they’re electric or driverless. Where will they all fit in the urban street? And when they take over, what room will be left for wider sidewalks, bike lanes, pocket parks, or indeed anything but a vast river of vehicles?"
Also harmful to the potential of autonomous vehicles to reduce congestion are the number of vehicle trips induced by the perceived convenience of hopping in a self-driving car for a trip that otherwise would have been avoided. Researchers at the University of California, Berkeley found evidence in 2018 that autonomous vehicles will create so many new trips that VMT will increase well beyond the capacity created by the efficiency of self-driving technology.
Self-driving cars could produce numerous benefits compared to the current system of automobile-dependent sprawl and rampant reckless driving. Improved traffic safety from taking the keys out of the hands of drunk drivers is just one example. The potential for millions of acres of parking lots to suddenly become available for more useful purposes like parks and open space or affordable housing, is another. But while those with a vested interest in autonomous vehicle technology want to believe it will be a silver bullet to end all traffic congestion, we have been warned, repeatedly, by the world's leading experts, that congestion just isn't that easy.
Elon Musk founded the Boring Co. in December 2017 with big promises to solve congestion by building tunnels. A test track in Los Angeles provided the first demonstration of the concept before the Boring Co. worked on projects, to varying degrees of completion, in Las Vegas, Chicago, San Antonio, Miami, and Fort Lauderdale. Numerous critics and transit experts have weighed in, promising that the concept wouldn't solve congestion above ground and would likely only create congestion underground as well.
Other cities have back out on the idea before they could offer more evidence. Chicago has backed off its plan for the Boring Co to build an "Express Loop" to O'Hare International Airport, and Los Angeles has also dropped plans for the Boring Co to build a two-mile tunnel on the city's Westside.
Sometimes Congestion Relief
While adding new road capacity, above ground or below, has been shown repeatedly to be fool's gold for the purposes of reducing congestion, so too are many of the alternatives—even the alternatives favored by opponents of the car-centric status quo.
Even the most ardent public transit supporters must begrudgingly concede that the evidence for the utility of public transit as a congestion mitigation tool is decidedly mixed. The limitations of public transit are similar to that of added roadway capacity: latent demand for mobility options quickly overwhelms the new capacity. It's induced demand on a fresh set of wheels.
Various studies over the years have shown evidence that new public transit service does not reduce automobile congestion, including a World Bank working paper published in 2003 and a paper written by David R. Heres Del Valle, a PhD candidate at the University of California, Davis, in 2009. Moreover, the 2011 research credited with popularizing the "fundamental law of congestion," referenced in Part One of this series, also found "no evidence that the provision of public transportation affects [Vehicle Kilometers Traveled]."
There is contrary evidence, however, that a lack of public transit worsens automobile congestion—especially on roads that closely follow transit routes. A study published in 2013 crunched transportation data from a 2003 transit workers strike in Los Angeles to find that average highway delay increases by 47 percent when public transit service ceases. The delays varied depending on how closely automobile corridors mirrored public transit routes. For example, the researchers found that traffic delays on U.S. Highway 101, which parallels the route of the B Line (then known as the Red Line) subway, increased by 90 percent during the strike.
Despite their common lack of utility in solving congestion, the differences between public transit and new roadway capacity are immense. The latter enshrines congestion as a perpetual consequence of the system. The former provides an alternative to the system and, if done well, a way over, around, or under congestion. The solution for the fallacy of transit as a congestion solution is simple: don't oversell public transit as a solution for congestion. Public transit is still a low-cost alternative to the congestion and pollution effects of automobile dependency with immense benefits to transit riders in terms of economic and social mobility. Or, as stated in a 2020 article by Daniel Herriges for Strong Towns, don't count on public transit to reduce congestion, but build it anyway to ensure that people have an alternative to congestion.
Higher Gas Prices
Concurrent with this writing, the United States is facing a historic spike in gas prices, adding fuel to historic inflation rates as war rages in Ukraine and the world deals with the long-term effects of the Covid-19 pandemic. Opponents have long argued that gas is too cheap in the United States—too cheap to pay for the cost of building and maintaining road infrastructure, too cheap to pay for the numerous negative externalities of driving, and too cheap to be a disincentive for the poor choices of consumers. High gas prices, on the other hand, are such a political liability that ostensibly liberal politicians in California, New York, and Washington, D.C. have suggested gas tax holidays to lower the price of a gallon of gas at the pump. President Joe Biden even allowed a release from the country's Strategic Petroleum Reserve after his political opponents blamed the high cost of gas on his administration.
In proposing these courses of action in response to rising gas prices, the country is effectively avoiding an experiment into the effects of gasoline prices on congestion and, potentially, an opportunity to let prices settle in line with the outcomes produced by automobile dependency.
The only problem is that like so many other potential policy responses to congestion, high gas prices don't always produce expected outcomes. Previous price spikes have produced an immediate drop in driving, so high gas prices will reduce the number of cars on the road, but not at a rate commensurate with the increase in gas prices—due in part to a lack of viable alternatives. Increased public transit ridership, for example, isn't guaranteed by high gas prices (although the evidence is mixed).
Still, if gas prices were cheaper, there would be more cars on the road.
Gas prices don't happen in a vacuum of social consequence either. High gas prices affect consumers disproportionately depending on income levels and geographic location. Low-income residents of automobile dependent communities can be priced out of mobility entirely when gas prices rise, which is not an outcome any responsible city planner would suggest as a congestion solution. What's a minor inconvenience or a chance to virtue signal for some can have tragic consequences for others.
Improved Traffic Flows
Transportation engineers have numerous methods for redesigning streets to improve traffic flow and speed up traffic. For over a century, the configuration of the entire public realm has favored the practices and facilities favored by transportation engineers. But the status quo of transportation engineering comes with tremendous costs and often limited benefits to congestion. Like so many other of the proposed solutions for congestion, the mixed bag of benefits and consequences created by traffic flow improvements makes it impossible to pronounce this approach as a definite solution for congestion.
To find evidence of traffic flow improvements as a fundamental congestion mitigation strategy, look to almost every road in the United States. The list of common street features designed to improve traffic flow are too long to list completely here, but they include lane widths (12 feet in most places), lane straightening, intersection designs (including slip lanes), interchange designs, and one way streets. Despite almost every road and street in the United States having been pushed and pulled to the absolute maximum to allow vehicles more space to move quickly, even beyond the boundaries of private property, evidence suggests that urban road expansion and traffic flow improvements achieve only small congestion improvements. See the "Roadway Capacity Expansion" section of a 2022 report published by the Victoria Transport Policy Institute and authored by Planetizen blogger Todd Litman for evidence of this counterintuitive fact of life.
While straightening and widening roads can provide short-term relief from congestion during peak commute hours and increase vehicle speeds during off-peak hours, the improvements on average aren't enough to overcome induced demand or the risks introduced by high-speed drivers.
Traffic flow is also impacted by traffic controls—including intersections, stop signs, stop lights, yield signs, lane merges, and toll collections, to name a few examples. Traffic controls can create congestion by creating traffic bottlenecks and resulting shockwave effects. Conversely, it stands to reason that traffic congestion can also be mitigated by improving, removing, or adding traffic controls. Again, if it were only that simple.
In a contemporary twist on the age-old theme of traffic control innovations, new technology promises to offer traffic controls that are more responsive to the real-time conditions on the street or that can predict and anticipate the times when poorly timed traffic controls could create a backup. In addition to the improvements for traffic flow, some of these technology-enhanced traffic control systems—referred to as advanced transportation controllers or ATCs—also promise traffic safety improvements.
Given the ongoing existence of traffic congestion, even in the most technologically savvy cities, it might be hard to imagine that these advanced traffic control technologies are living up to the promise. Los Angeles, famous for its traffic congestion, is often credited with leading the nation in the deployment of advanced transportation controllers, including traffic light synchronization and real time traffic sensors. On April 1 (coincidence), 2013, the New York Times credited Los Angeles as "the first major metropolis in the world" to synchronize every single one of its traffic signals—4,500 in all, covering 469 square miles. As bluntly summarized in a 2020 report by University of California, Los Angeles researchers, these technological advancements failed to solve congestion in the city.
"Authorities relayed updates about traffic conditions to drivers, and engineers adjusted synchronized systems of street light to optimize flows. Information technology was used to match travelers by their work locations and commute times so they could share rides or replace work trips by telecommuting. Even these cutting-edge innovations did not bring an end to recurring traffic jams."
On the other side of the issue of technological innovations for traffic controls, however, are automated toll collection facilities. Manual toll collection facilities, another form of traffic control, are also notorious for causing congestion as drivers line up to physically hand over the payment required to cross a bridge or enter a toll road. In recent years, however, many toll collection facilities have switched to electronic toll collection (ETC) technology, which has demonstrated improvements for travel delays while also reducing emissions and air pollution from vehicles. Here we see evidence that the removal or improvement of traffic controls can improve traffic congestion.
Toll facilities can play another key role in the reduction of congestion by helping manage the level of demand for the use of a road by adding the disincentive of extra cost, but I digress. (Part three of this series will revisit the idea of transportation demand management and the related idea of congestion pricing—one of the most proven and effective, though controversial, tools of congestion relief—in more detail.)
What has been accomplished by redesigning streets to accommodate higher expectations about travel times from the general public, flashy new traffic control technology, and faster, more powerful cars? A century's worth of traffic flow improvements has not ended collisions. Collisions create bottlenecks and rubbernecks, as described in the "Collisions" section of part one of this series, for example. There were 12 million collisions on U.S. streets and roads in 2019.
Improving traffic flow to lessen traffic congestion also comes with this significant and alarming caveat regarding all of those millions of collisions: High-speed drivers are more likely to create collisions that kill—other drivers as well as people on bikes and pedestrians. Traffic fatalities on U.S. roads increased significantly in 2020 and 2021—reversing decades of progress, during which safety features like seat belts and airbags and blind spot assist were added to cars. It's impossible to talk about street design and traffic controls for congestion mitigation without the context of the carnage on U.S. roads.
Looking Beyond the Status Quo
As documented in the book Confessions of a Recovering Engineer, by Charles Marohn, the status quo of transportation engineering has heavily favored designing for traffic flows for more than a century, but based largely on a pseudoscience of arbitrary and biased thinking. The result? Congestion never seems to improve and these improvements come at the expense of 40,000 lives a year, public space, private property, and the freedom to move around communities outside of a car.
An increasing number of cities, counties, and even a few states are beginning to design new and existing streets with traffic safety as the primary concern instead of traffic flow, prioritizing "traffic calming" measures like speed bumps, bulbouts (i.e., curb extensions), high-visibility crosswalks, protected bike lanes, and road diets (i.e., removing vehicle lanes to make space for alternative modes and traffic calming features). The shift from prioritizing traffic safety over traffic flow is an ongoing experiment with inconsistent levels of adoption, but it's gaining traction in the planning profession.
A document published by the Federal Highway Administration (FHWA) called the "Manual on Uniform Traffic Control Devices" (MUTCD) defines the standards used by road managers nationwide to install and maintain traffic control devices on all public streets, highways, bikeways, and private roads open to public travel. As of this writing, the FHWA is updating the MUTCD for its 11th edition, already collecting 35,000 individual comments in public feedback. Among those comments, advocates from the planning community are pushing for reforms that would make it easier for these federal standards to allow traffic safety features, rather than solely focusing on traffic flow (see articles from Streetsblog USA, Bloomberg CityLab, and.the National Association of City Transportation Officials (NACTO) for more on how this advocacy movement is centering around the update to the MUTCD for the benefit of numerous policy goals).
In another counterintuitive moment, by enabling alternative forms of transportation and lowering the number of delay-causing collisions, the shift toward traffic safety and walkable, complete streets might still prove to reduce traffic congestion, unlike the status quo of trying, and failing to build our way out of congestion.
Planetizen will continue its search for congestion solutions in part three of this series, which will be posted next week. Part one, "Planning and the Complicated Causes and Effects of Congestion," was published on April 13, 2022.
The Right to Mobility
As we consider how to decarbonize transportation, preserving mobility, especially for lower- and middle-income people, must be a priority.
America’s Best New Bike Lanes
PeopleForBikes highlights some of the most exciting new bike infrastructure projects completed in 2022.
Early Sharrow Booster: ‘I Was Wrong’
The lane marking was meant to raise awareness and instill shared respect among drivers and cyclists. But their inefficiency has led supporters to denounce sharrows, pushing instead for more robust bike infrastructure that truly protects riders.
L.A. County Towns Clash Over Homelessness Policies
Local governments often come to different conclusions about how to address homelessness within their respective borders, but varying approaches only exacerbate the problem.
A Mixed-Use Vision for Houston Landfill Site
A local nonprofit is urging the city to consider adding mixed-use development to the site, which city officials plan to turn into a stormwater detention facility.
Milwaukee County Makes Substantial Progress on Homelessness
In 2022, the county’s point-in-time count of unhoused people reflected just 18 individuals, the lowest in the country.
HUD's Office of Policy Development and Research
HUD's Office of Policy Development and Research
Harvard GSD Executive Education
City of Fitchburg, WI
City of Culver City
Sonoma County Transportation Authority
This six-course series explores essential urban design concepts using open source software and equips planners with the tools they need to participate fully in the urban design process.