Donate Menu

Does America’s Transportation Future Really Need More and Bigger Roads?

One of the bigger debates in Congress last summer centered on where to find more money to put into the Highway Trust Fund (HTF), the federal government’s mechanism to fund transportation infrastructure projects—usually new road construction or road expansion. The HTF is funded by the gasoline tax, which hasn’t been adjusted since the 1980s. In those 30 years, improved fuel economy and alternative powertrains like hybrids, electric vehicles (EVs), and extended-range EVs have eroded gas tax revenues. Gasoline consumption is just no longer a reliable proxy for a vehicle’s impact on highway infrastructure. With decreasing revenues, the HTF has been spending more than it collects and was on its way to insolvency.

Yet finding more money for the HTF overlooks the more fundamental questions: will we need more HTF revenue and will we need to build more road capacity in the years and decades ahead? How could we spend this money better?

Congressional funding for the HTF was to run out at the end of this month, and Congress is currently debating how to keep money flowing into the HTF, even as a meager two-month extension puts a tiny bandage on a gushing wound. Several states have increased their gasoline tax this year, including Iowa ($0.10/gal), Utah ($0.05/gal), and Idaho ($0.07/gal). Another possible solution is to decouple the HTF from gasoline sales, using a vehicle-miles-traveled (VMT) tax weighted by gross vehicle weight to better approximate real impact on highway infrastructure. Oregon is planning to test this approach, and will launching a pilot program in July.

And this pattern of searching for funds for the HTF looks to continue. The Obama administration released its 2015 budget proposal, including $317 billion for roads and bridges, paid for partially through repatriation of overseas investments. Congress recently passed its 2015 budget resolutions, providing a framework for appropriations bill set to pass this month, including funding for the HTF set to expire on May 31. The budget frameworks set by the House and Senate cut transportation infrastructure spending by 28 percent and 22 percent respectively through 2025 and make no attempt to find new sources of revenue. Once again, as we close in on the deadline to authorize HTF funding with no long-term solutions in sight, Congress is floating a two-month extension for HTF funding.

Four Solutions That Could Change How We Drive—and How We Think About Roads

The only argument during the debate last summer was for more roads. If we assume we’ll drive tomorrow the same way we drive today then as U.S. population grows, everyone will buy cars and drive, causing more congestion. To reduce congestion, the argument goes, we need to build more and bigger and wider roads.

Yet several studies debunk that last notion. While adding road capacity temporarily reduces congestion, the capacity quickly fills up again. But, what about the first part of the argument: does an increasing U.S. population necessarily mean more cars and more driving?

To answer, I modeled independently the adoption of and impact on vehicle miles traveled (VMT) of four emerging mobility solutions—new business models or technologies poised to transform the U.S. transportation system—with a business-as-usual (BAU) and accelerated case. Each modeled solution is market-driven with profit potential, and all would increase affordability, convenience, and mobility for consumers. U.S. annual VMT (3 trillion miles averaging out to roughly 14,000 VMT per licensed driver) has leveled off since 2005, with VMT per person decreasing every year. This trend alone points to reconsidering the basic assumptions of this debate. If these four solutions can further reduce VMT significantly, then we won’t need more roads or more drilling and oil.

1. Smart Growth

Smart growth principles guide planning in urban and suburban neighborhoods toward more compact developments with less sprawl. Two key principles are walkability and mixed-use zoning.

As the name implies, walkable, mixed-use neighborhood means that most daily activities are within walking distance. One popular method to quantify walkability uses Walkscore, which maps out walking routes to nearby destinations (e.g., restaurants, grocery stores, schools), measures the quantity of destinations in the area, and outputs a score between 0 and 100.

Since daily activities make up more than 75 percent of trips in the U.S., living in a walkable, mixed-use neighborhood can dramatically transform mobility, eliminating most driving needs. A comprehensive Victoria Transport Policy Institute study found it could reduce residents’ VMT up to 20 percent.

A 2013 survey found 50–60 percent surveyed prefer to live in a walkable neighborhood, often found in dense urban centers. But, it’s not just for urbanites: in Washington, D.C.—one of the U.S.’s most walkable cities—half of the walkable neighborhoods are in the suburbs.

Developers and investors also see the value in these neighborhoods, which command higher rents and housing prices than those in sprawling suburbs. Right now, about 5 percent of the U.S. population lives in these neighborhoods.

2. Car Sharing

Car sharing includes many different business models. I focused on station-based models where users pay annual subscription and hourly use fees. This is conservative since I don’t consider other mobility services—peer-to-peer, point-to-point, and bike- and ride-sharing—that could one day be integrated into a single-payment on-demand mobility system not unlike smartphone packages today that offer phone minutes, texting, and data bundled together.

Car sharing is growing as consumers, millennials especially, find it more economical to use a shared mobility service than to own a car. Car sharing has already expanded rapidly from 50,000 to nearly 1.5 million U.S. members between 2004 and 2014. And this growth reduces vehicle ownership—every car-share vehicle eliminates 9–13 personal vehicles, with users selling their car or deferring buying one—and reduces total number of vehicles on the road by using cars that are on the road more efficiently via higher utilization rates. Also, because hourly rates and subscriptions better reflect the total cost of the vehicle (insurance, maintenance, capital, and fuel), drivers of shared cars are more aware of the true costs of car ownership and actually drive less and make more-efficient trips. More-efficient and reduced trip-making combined with lower vehicle ownership points to 27–44 percent fewer VMT for car-share users.

3. Smart Parking

Despite roughly three non-residential parking spaces per vehicle in the U.S., we’ve all cruised around—burning fuel, wasting time—looking for parking. In fact, drivers searching for parking represent about 30 percent of vehicles on the road in cities. Smart parking solves this by equipping parking spaces with sensors detecting occupancy, then integrating this data with smartphone apps directing users to empty spots.

Smart parking transforms driving in two ways. First, individual trips are more efficient, with cruising for parking eliminated. While this has minimal impact on VMT— circling the block only adds up to 0.5 miles per leg of a trip—it can have significant fuel and time savings. Second, there are also system-wide impacts that my model does not account for. As cruising vehicles are removed from the road, congestion eases for cars traveling through. And the collected data on parking usage combined with smart growth means cities can reduce parking requirements and repurpose existing parking.

Adoption of smart parking is just under way, with pilot programs in San Francisco, New York City, Kansas City, and Boulder, Colo., to name a few. Despite the hurdles, smart parking’s future looks promising: as pilot programs scale, the convenience far outweighs the frustration of finding parking.

4. Connected, self-driving vehicles

Self-driving vehicles are defined by level. Levels 1–2 (adaptive cruise control, lane change assist, collision avoidance) require full driver engagement and are available on current vehicle models. For levels 3–4, the car drives itself with minimal driver intervention. Connected refers to vehicle-to-vehicle and vehicle-to-infrastructure communications (V2X): cars talk to other cars, dynamically reroute around traffic, and “call ahead” to traffic signals to avoid sitting at empty intersections.

First, some 90 percent of crashes are caused by human error and these could be nearly eliminated by self-driving cars, saving thousands of lives. As a secondary benefit, crash-related congestion—accounting for approximately 25 percent of congestion events—would be avoided.

Second, recurring stop-and-go traffic (think rush hour) could be eased through platooning/highway driving with reduced headway, dynamic rerouting, and traffic flow smoothing. Self-driving vehicles can thus double to quadruple highway capacity without expanding the highway, making for smoother, faster trips—although this will require significant market penetration.

Because these technologies require consumer familiarity, comfort, and acceptance, and have significant technical hurdles to overcome, I assume their adoption will be similar to other in-vehicle technologies. This means 25–30 years before they represent 90 percent of vehicle sales (with stock turnover, a smaller percentage of total U.S. vehicles). Level 3–5 automation may see its first commercial iteration in 2020 if Google succeeds; for broader availability beyond luxury markets, costs need to decrease and volumes increase, as with any other emerging technology.

A Transformed Transportation System

Putting it all together, the business-as-usual model output shows that these solutions reduce VMT 15 percent by 2040, stopping VMT growth at 2030 levels (using the U.S. Energy Information Administration’s projections as the baseline even though these projections—that VMT will continue to increase despite a decade of flat growth—are generally poor at predicting future traffic trends):

In the accelerated case, where infrastructure spending focuses not on building more roads but enabling these solutions, VMT growth stalls by 2020 and then decreases. Mobility as a service, the sharing economy, and connectivity offer greater affordability, convenience, and mobility than car ownership—and businesses, developers, and investors profit.

The Federal Highway Administration predicts federal spending of $28.7 to $54.4 billion annually (in constant 2010 dollars) through 2040, assuming the federal share of spending stays constant. Using this range of spending, we can save $47 to $89 billion on new road construction or expansion through 2040 in the BAU case, and $79 to $150 billion in the aggressive case. Now, imagine taking those savings from not building new roads and instead deploying sensors and software to enable a smarter, more-efficient transportation system.

In the 1950s, Eisenhower had a vision for a national highway system that would serve this nation’s mobility needs for decades. We’re already driving fewer miles than we thought we would, and with a robust suite of on-demand, shared, self-driving mobility services available will more likely than not drive even less in the years to come. This future has very different implications for how we use the Highway Trust Fund: there is ample opportunity to make smaller investments in better data for smarter infrastructure.

A major shift is under way, changing the way we will travel over the next 25 years, and the debate needs to shift with it. We should be talking more about maintaining the roads we have and using them (and the vehicles on them) better, instead of how to fund our steady march to more paved land.

Image courtesy of Shutterstock.