The Potential Effects of CAVs on GHG Emissions: A Research Synthesis

The advent of connected and automated vehicles (CAVs) will revolutionize the way we get around more than any technological advance since the invention of the automobile. And, because transportation accounts for 27% of America’s greenhouse gas emissions[i], this upcoming revolution will have immense effects on our country’s emissions levels. It’s not immediately clear, however, whether these effects will be positive or negative. Limited research and broad variability in potential future behaviors and adoption patterns lead to uncertainty on the effects of CAVs on our greenhouse gas emissions.

What follows is a synthesis of current research on topics that are closely linked to CAVs’ effect on greenhouse gas emissions. CAVs’ potential effects on vehicle miles traveled (VMT) and fuel efficiency have been studied, but further research is needed for a more robust understanding of these topics. Only with a firm understanding of the expected effects of CAVs and a clear identification of our values and goals can we enact sound policy to pursue desired outcomes.

Vehicle Miles Traveled

A variety of research has been conducted on the effects of CAVs on VMT, and while there are many factors at play, most experts feel that CAVs portend a rise in VMT. Because VMT and greenhouse gas emissions are closely linked, this trend could lead to a similar increase in our country’s emissions. As researchers at Argonne National Laboratory state, “Changes in travel behavior, in effect lowering the cost of driving, have the potential to greatly increase the utilization of the transportation systems with concurrent negative externalities such as congestion, energy use, emissions, and so on, working against the positive effects on the transportation system due to increased capacity.”[ii]

With a new freedom to work, read, sleep, or watch TV during a long drive, the burden of driving will decrease and travel demand will increase.

Almost all experts agree with the synopsis that changes brought about by CAVs will lower the cost of driving. With less stress due to defensive driving and a new freedom to work, read, sleep, or watch TV during a long drive, the burden of driving will decrease and travel demand will increase. Commute distances and sprawl will likely increase in kind as people are willing to be in a car more often and for longer periods of time. This increase in sprawl has the potential to lead to a variety of negative social impacts. As people and businesses spread out further, cities with already tight budgets will be obligated to spend more money on public works to support the infrastructure for this growth. Additionally, it becomes more difficult for our public transportation systems to serve those who can’t afford a vehicle, and creating effective walking and biking routes becomes more difficult.

In addition to enabling existing drivers to accept longer and more frequent travel, automated vehicles (AVs) will also open the door for new users to access the roadway. The young, elderly, and disabled, many of whom are incapable of driving, will now have an additional travel mode choice. Some children may be able to ride home from after-school activities by themselves, and the elderly and disabled will be able to ride to health care appointments they previously had limited access to. KPMG, a financial firm with a business advisory sector, estimated that in the United States, “small changes among the oldest and youngest demographic groups will likely produce large increases in personal miles traveled (PMT) by 2050: approximately 500 billion more PMT annually.”[iii] While transit and taxi services currently provide transportation options to these types of users in many locations, CAVs have the potential to be an easier option to use and to reach users who live in locations that are not covered by these existing services.

Trips made for errands will also increase as a result of AVs, with many of them likely being zero-person trips. The personal burden of sending a car home to retrieve a forgotten item is substantially less than that of making the round-trip yourself, so this type of trip will see increased demand. The Center for Automotive Research (CAR) in Ann Arbor, Michigan,[iv] states that “[i]f automated vehicles perform many empty trucking backhauls – return trips without cargo or passengers – VMT could increase due to empty vehicles traveling between a drop-off and the next pick-up.” Trip-chaining, or combining multiple trips into one, will also likely decrease. Instead of dropping each child off at his or her school in succession, for example, parents may send children to school separately using AVs.

Experts agree that CAVs will make driving easier and less burdensome and remove barriers to access. In this sense, CAVs will almost certainly lead to an increase in VMT. This isn’t the only way CAVs will affect VMT, though. To fully understand CAVs’ cumulative effects, a variety of other factors need to be considered.

Some experts envision a future of shared autonomous vehicles (SAVs), where personal car ownership is obsolete.

Shared Autonomous Vehicles

CAVs have the potential to revolutionize not only our travel patterns, but our car ownership patterns as well. Some experts envision a future of shared autonomous vehicles (SAVs), where personal car ownership is obsolete. Services like Zipcar and Uber would become the norm for travel, as a typical trip would start with a smartphone app instead of reaching for the car keys. Mobility as a service – the concept of consuming trips individually instead of through an owned vehicle – could rapidly become more affordable as technology eliminates the need for drivers and streamlines the processes of services such as Zipcar and Uber. The Center for American Progress (CAP)[v] states that “[b]y deploying self-driving cars, ride-sharing companies can become more efficient and profitable. AVs allow ride-sharing companies the opportunity to mechanize their service and eliminate the need to pay human drivers.”

Indeed, ride-sharing companies have already begun to go down the path toward SAVs. CAR states that “companies like Uber and Lyft that are developing self-driving taxis are already proposing services for shared rides, such as UberPOOL and Lyft Line, that will become even more attractive with automated vehicles.” However, the popularity of these services, even at a reduced cost, is not a foregone conclusion. Safety concerns remain prevalent, as users may prefer not to take rides with strangers without a third-party driver present.

Many experts foresee mobility as a service increasing the number of shared trips taken. A shift in culture from most trips being made in single occupancy vehicles to most trips being made in concert with other trips would dramatically reduce vehicle miles traveled. CAP states that “a system of shared autonomous vehicles could discourage individual car ownership and use technology to plan efficient routes to transport people from point to point.”

Secondary effects of this culture shift also could play a major part in vehicle miles traveled. According to the Sustainable Cities Initiative,[vi] parking is currently the largest land use in U.S. cities. Researchers at the International Transport Forum[vii] showed that SAVs could decrease the need for parking by up to 90%, leading to infill opportunities, cheaper land costs and less expensive housing. With this infill, more people could live closer to the places that they travel to, increasing the viability of walking and transit trips and decreasing the distances of vehicle trips made. If this shift in living patterns does occur, VMT could decrease dramatically.

Some aspects of SAVs do have the potential to raise VMT, though. If market penetration is low, vehicles won’t always be near customers when needed, leading to longer zero-passenger trips to pick customers up or longer wait times (a downside to current transit services for some people). Additionally, SAVs could choose to drive idly without occupants while waiting for their next customer rather than paying to park (if parking is even available). Even with these potential drawbacks accounted for, however, experts tend to agree that a future of shared autonomous vehicles would lead to a lower VMT than a future of personally owned vehicles.

Connected vehicles can make our transportation systems more efficient by lowering following distance, connecting to infrastructure such as signals, and communicating with surrounding vehicles.


One of the most heralded advantages of the dawn of CAVs is the expected decrease in congestion. With a high enough market penetration, connected vehicles can make our transportation systems more efficient by lowering following distance, connecting to infrastructure such as signals, and communicating with surrounding vehicles. A report by Atkins Global[viii] states that “impacts caused by congestion will be reduced as CAVs will be able to drive closer together, increasing roadway capacity without impeding safety.”

Additionally, CAVs can be expected to decrease congestion by reducing crashes. Automation could reduce almost all crashes caused by driver impairment, and a report by KPMG and CAR[ix] notes that CAVs could also be expected to eliminate as many as 80% of unimpaired crashes. Because traffic incidents such as crashes cause 25% of congestion,[x] this potential crash reduction would lead to more efficient operations on our transportation systems. These safety benefits are not guaranteed, however. Studies showing large crash reductions often assume proper maintenance of CAVs, but policy ensuring this is not yet enacted. The responsible party for ensuring that all sensors and software are up to date is not immediately clear and will likely be the result of future policy decisions.

Lastly, a shift toward SAVs would reduce the number of cars on the road, again lessening congestion. Research from the University of California at Berkeley[xi] shows that each shared vehicle in North America reduces the need for nine to 13 personal cars. Although SAVs would be expected to drive more miles than a personally owned vehicle, this shift would likely still cause a reduction in congestion.

The extra capacity added by the improved operations of CAVs will be filled quickly. Capacity induces demand, and the capacity freed up by CAVs surely will be no different. The reduction in congestion offered by CAVs will again lower the opportunity cost of driving, leading more users to choose to live farther away and travel longer distances for commutes, ultimately causing VMT to increase. Researchers at the University of Texas at Austin explain that “additional VMT increases may be realized from induced demand as travel costs and congestion fall.” The extra capacity CAVs add to our streets will likely turn quickly into extra volume and VMT.

Fuel Efficiency

In the fight against rising greenhouse gas emissions, VMT is only one side of the coin. The other important metric is fuel efficiency, which can be measured by greenhouse gases emitted per mile. CAVs, and advances in vehicle technology in general, have the potential to cause major changes in this area.

First, CAVs can affect fuel efficiency in a variety of minor ways. Because AVs, and especially SAVs, are likely to be running much more frequently than traditional personally owned vehicles, engines will stay warmer. CAP states that “each shared AV would require fewer cold starts. Emissions are higher when the engine is cold than when the catalytic converter has warmed up, so keeping the engine warm would reduce emissions.”

Additionally, once CAVs reach high market penetration, crashes could decrease precipitously and the need for additional bulk for safety’s sake could decrease. Vehicle size could also decrease as mobility as a service takes hold. Single vehicle trips can be paired with smaller vehicles, whereas in the past all trips were made with the same personal vehicle, regardless of the number of travelers. These decreases in vehicle size could lead to less greenhouse gas emitted per mile. Platooning, or decreasing the distances between vehicles using electronic communication, can reduce aerodynamic drag and increase fuel efficiency, especially among trucks.

The ease of use of CAVs could reduce the demand for transit, however, which is often more fuel efficient than single-occupancy vehicles. Many of the advantages of transit, such as people not needing to park and being able to multi-task during the trip, will be present in AVs and thus could reduce the demand for transit, especially among choice riders. While advantages to mass transit remain, especially in dense cities where fixed routes are the most efficient, this decrease in transit ridership could result in less money being invested in transit and ultimately could be disastrous for our country’s public transportation systems.

Electrification of vehicles has the potential to markedly change the greenhouse effects of each vehicle mile traveled.

The most important variable in the greenhouse gas equation, however, is the electrification of vehicles. Carbon dioxide emissions from electricity generation have been decreasing since 2007.[xii] Assuming the continuation of this trend, electrification of vehicles has the potential to markedly change the greenhouse effects of each vehicle mile traveled. CAP states that “the best way to minimize emissions from the transportation sector is to combine vehicle automation with electrification. To achieve the nation’s climate goals, the United States cannot remain dependent on gasoline-powered vehicles.”

While electric vehicles and CAVs are not intrinsically linked, many new CAVs are electric for a variety of reasons. According to Greg Gardner of the Detroit Free Press,[xiii] “Electric vehicles are easier for computers to drive”¦ and it’s easier, cheaper, and safer to recharge an unmanned car than to gas one up.” Reliability and lower fuel costs are additional factors that will encourage ride-sharing companies to adopt electric instead of gas-powered vehicles. “We see that as very complimentary, technology-wise,” Harry Lightsey, executive director of emerging technologies policy at GM said of CAVs and electrification.[xiv]

Electric vehicles are complementary to shared CAVs. Research from the Boston Consulting Group[xv] shows that the convergence of ride sharing, autonomous driving, and vehicle electrification create a more compelling economic case for adoption of shared autonomous electric vehicles (SAEVs) than any of these factors would in isolation. BCG states that “[d]ue to their ability to cut travel costs by 60%, SAEVs could shift about 25% of miles traveled from private automobiles.”

Planning and Policy Implications

Many policy measures have already been proposed that are intended to mitigate some of the negative effects of CAVs. The Upstate Transportation Association in New York State has urged the state of New York to ban self-driving cars from the state’s roads for 50 years in an effort to eliminate driver job loss from vehicle automation.[xvi] Truck driver unions surely won’t be far behind. A bill proposed in Massachusetts and frequently echoed elsewhere would tax self-driving vehicles on a per-mile basis in an effort to eliminate zero-passenger rides.[xvii] Without more robust research, however, the full effects of these kinds of policies can’t be known for certain.

Research has yet to reach a conclusion on a variety of factors that will determine the effects of CAVs on greenhouse gas emissions. Whether VMT increases or decreases could be dictated by whether shared autonomous vehicles thoroughly saturate the market or personally owned vehicles continue to be the norm. The fuel efficiency conversation hinges largely on the adaptation or lack thereof of electric-powered CAVs. With the future still cloudy, additional research into these and other trends is still necessary to determine the expected outcome of CAVs.

An understanding of our collective values and goals is also crucial to moving forward. Many of the effects of CAVs discussed above have come from the perspective of vehicle miles traveled or fuel efficiency. However, these effects surely should be weighted and considered beyond their performance in these metrics. Crash reductions are most significant, not for the reduction in congestion but for the lives saved. The restoration of an elderly person’s access to hospitals, community centers, and food choices is important to his or her life and well-being, not just to the VMT on our transportation system. Sarah Hunter, a policy director for Google, states that “what we’re going to have is people who are locked at home able to get around for the first time. That’s a great outcome from an equity perspective. From a number of cars on the road perspective, not so great.”[xviii]

With this in mind, we need to fully understand both the potential trade-offs involved in the shift to CAVs and the relative importance of each of these trade-offs to our society. Only once we’ve done this can we hope to enact fair, meaningful, and effective policy that will maximize the benefits and mitigate the negative externalities of the upcoming transportation revolution.



[i] “Inventory of U.S. Greenhouse Gas Emissions and Sinks.” EPA, Environmental Protection Agency, 14 Apr. 2017,

[ii] Auld, Joshua, et al. “Analysis of the Impacts of CAV Technologies on Travel Demand.” Argonne National Laboratory.

[iii] Silberg, Gary, et al. The Clockspeed Dilemma. KPMG, Nov. 2015,

[iv] Dennis, Eric, et al. Planning for Connected and Automated Vehicles. Center for Automotive Research, Mar. 2017,

[v] Alexander-Kearns, Myriam, et al. The Impact of Vehicle Automation on Carbon Emissions. Center for American Progress, Nov. 2016,

[vi] Larco, Nico. “AVs and Real Estate ”“ A Guide to Potential Impacts.” Urbanism Next, Sustainable Cities Initiative, 21 Aug. 2017,

[vii] “Urban Mobility System Upgrade: How Shared Self-Driving Cars Could Change City Traffic.” OECD, International Transport Forum, 16 May 2015,

[viii] Bradburn, Josh, et al. “Connected & Autonomous Vehicles.” Connected & Autonomous Vehicles ”“ Atkins, Atkins Global,

[ix] Silberg, Gary, and Richard Wallace. Self-Driving Cars: The next Revolution. KPMG and CAR,

[x] Daniel J. Fagnant and Kara Kockelman, “Preparing a Nation for Autonomous Vehicles: Opportunities, Barriers and Policy Recommendations for Capitalizing on Self-Driven Vehicles,” Transportation Research Part A 77 (2015): 167”“181,

[xi] Martin, Elliot, et al. The Impact of Carsharing on Household Vehicle Holdings: Results from a North American Shared-Use Vehicle Survey. Transportation Sustainability Research Center, Mar. 2010,

[xii] Wirman, Channele. “U.S. Energy Information Administration – EIA – Independent Statistics and Analysis.” Carbon Dioxide Emissions from Electricity Generation in 2015 Were Lowest since 1993, 13 May 2016,

[xiii] Gardner, Greg. “The Self-Driving Revolution Will Be Mostly Electric.” Detroit Free Press, Detroit Free Press, 21 Sept. 2016,

[xiv] McCauley, Ryan. “Why Autonomous and Electric Vehicles Are Inextricably Linked.” Government Technology: State & Local Government News Articles, 15 Mar. 2017,

[xv] The Boston Consulting Group, 10 Apr. 2017,

[xvi] McFarland, Matt. “The Backlash against Self-Driving Cars Is Officially Beginning in New York.” CNNMoney, Cable News Network, 10 Jan. 2017,

[xvii] Muoio, Danielle. “Massachusetts Might Tax Self-Driving Cars to Prevent the Rise of ‘Zombie Cars’.” Business Insider, Business Insider, 4 Mar. 2017,

[xviii] Hepler, Lauren. “Apple, Google, Tesla and the Race to Electric Self-Driving Cars.” GreenBiz, GreenBiz Group Inc., 22 Sept. 2015,