Will the Transportation Revolutions Improve Our Lives—or Make Them Worse?

Abstract

We love our cars. Or at least we love the freedom, flexibility, convenience, and comfort they offer. That love affair has been clear and unchallenged since the advent of the Model T a century ago. No longer. Now the privately owned, human-driven, gasoline-powered automobile is being attacked from many directions, with change threatening to upend travel and transportation as we know it. The businesses of car making and transit supply—never mind taxis, road building, and highway funding—are about to be disrupted. And with this disruption will come a transformation of our lifestyles. The signs are all around us.

We must steer oncoming innovations toward the public interest—toward shared, electric, automated vehicles. If we don’t, we risk creating a nightmare.

We love our cars. Or at least we love the freedom, flexibility, convenience, and comfort they offer. That love affair has been clear and unchallenged since the advent of the Model T a century ago. No longer. Now the privately owned, human-driven, gasoline-powered automobile is being attacked from many directions, with change threatening to upend travel and transportation as we know it. The businesses of car making and transit supply—never mind taxis, road building, and highway funding—are about to be disrupted. And with this disruption will come a transformation of our lifestyles. The signs are all around us.

Maybe you use Zipcar, Lyft, or Uber or know someone who does. You’ve probably seen a few electric vehicles (EVs) on the streets, mostly Nissan Leafs, Chevy Volts and Bolts, Teslas, and occasionally others. And you’ve undoubtedly heard and read stories about self-driving cars coming soon and changing everything. But how fast are the three revolutions in electric, shared, and automated vehicles happening, and will they converge? Will EVs become more affordable and serve the needs of most drivers? Will many of us really be willing to discard our cars and share rides and vehicles with others? Will we trust robots to drive our cars?

We’re at a fork in the road.

Over the past half century, transportation has barely changed. Yes, cars are safer and more reliable and more comfortable, but they still travel at the same speed, still have the same carrying capacity, and still guzzle gasoline with an internal combustion engine. Public transit hasn’t changed much either, though modern urban rail services have appeared in some cities since the 1970s. Likewise, roads are essentially unchanged, still made with asphalt and concrete and still funded mostly by gasoline and diesel taxes. We have a system in which our personal vehicles serve all purposes, and all roads serve all vehicles (except bicycles). It is incredibly expensive, inefficient, and resource intensive.

But it’s even worse than that. Most cars usually carry only one person and, most wasteful of all, sit unused about 95 percent of the time.¹ As wasteful and inefficient as they are, cars have largely vanquished public transit in most places. Buses and rail transit now account for only 1 percent of passenger miles in the United States.² Those who can’t drive because they’re too young, too poor, or too physically diminished are dependent on others for access to basic goods and services in all but a few dense cities.

Starting in Los Angeles, the United States built this incredibly expensive car monoculture, and it is being imitated around the world. Cars provide unequaled freedom and flexibility for many but at a very high cost. Owners of new cars in the United States spend on average about $8,500 per vehicle per year, accounting for 17 percent of their household budgets.³ On top of that is the cost to society of overbuilt roads, deaths and injuries, air pollution, carbon emissions, oil wars, and unhealthy lifestyles. The statistics are mind-numbing. For the United States alone, consider that nearly 40,000 people were killed and 4.6 million seriously injured in 2016 in car, motorcycle, and truck accidents.⁴ Nearly ten million barrels of oil are burned every day in the United States by our vehicles.⁵ Transportation accounts for a greater proportion of greenhouse gases than any other sector.⁶ Farther afield, in Singapore, 12 percent of the island nation’s scarce land is devoted to car infrastructure.⁷ In Delhi, 4.4 million children have irreversible lung damage because of poor air quality, mostly due to motor vehicles.⁸ We have created an unsustainable and highly inequitable transportation system.

But change is afoot, finally. For the first time since the advent of the Model T one hundred years ago, we have new options. The information technology revolution, which transformed how we communicate, do research, buy books, listen to music, and find a date, has finally come to transportation. We now have the potential to transform how we get around—to create a dream transportation system of shared, electric, automated vehicles that provides access for everyone and eliminates traffic congestion at far less cost than our current system. Or not. It could go awry. It could turn out to be a nightmare.

Let’s take a minute to imagine two different scenarios set in the year 2040.

Transportation 2040: The Dream

In one vision of the future, the government has managed to steer the three revolutions toward the common good with forward-thinking strategies and policies. Citizens have the freedom to choose from many clean transportation options. They can spend their time with family and friends rather than in traffic thanks to pooled automated cars. They breathe cleaner air, worry less about greenhouse gas emissions, and trust that transportation is safer, more efficient, and more accessible than ever before. The search for parking is an inconvenience of the past. Worries about Grandma being homebound have evaporated. No longer must parents devote hours to ferrying their kids everywhere. Transportation innovations have made it easy for people to meet all their transportation needs conveniently and at a reasonable cost.

On a typical day in this optimistic scenario, Patricia Mathews and Roberto Ruiz eat breakfast at home with their two children before Pat is picked up by an electric automated vehicle (AV) owned by a mobility company. The AV is dispatched from a mobility hub, where trains come and go, bikes are available, and AVs pick up and drop off passengers.

Like most homes in the neighborhood, the Mathews-Ruiz home has a small pickup area and vegetable garden in front, replacing what had been a large driveway. The garage has been converted to a guest room. Parks and public gardens are connected in a greenbelt that runs behind the homes. Children scamper around without parents worrying about traffic.

As Pat approaches the dispatched AV, it recognizes her and opens a door. Her unique scan authorizes a secure payment mediated through blockchain from her family’s mobility subscription account, which also pays for transit, bikeshare, and other transportation services. For a small monthly fee, plus a per-mile charge, the family gains access to a variety of shared vehicles and services, including AVs, electric scooters, and intercity trains. Discounts are also available for special services like air travel. The account isn’t connected to a traceable bank account, and travel data are erased every two months.

As Pat settles in for the short commute, the AV is notified to pick up one more passenger along the way, a neighbor Pat knows. On the way into the city, they chat about the upcoming neighborhood block party. The AV picks up another passenger and heads to the city center, where it is routed onto a broad boulevard with two lanes for auto travel, a reserved lane in the middle for trucks and buses, and bike lanes on each side, flanked by wide pedestrian walkways.

The rest of the Mathews-Ruiz family heads out on shared-use bicycles to the children’s school (with AVs available as backup on rainy days). Roberto continues on to the fitness center where he works. It takes him about twenty minutes. At lunchtime, Roberto will hop on a shared electric bike to meet his mother for lunch on the other side of town. She lives in a little neighborhood with a dense mix of shops and residences. Street parking was removed years ago and replaced with a sprinkling of passenger-loading and goods-delivery spaces, extensive bike pathways, wide sidewalks, outdoor seating, and pocket parks.

After lunch, Roberto helps his mother arrange a ride to a nearby medical center in an AV specially designed for physically limited passengers. On her way home, she will be dropped off to visit one of her friends. Her retirement income easily covers her mobility subscription and gives her and other low-income citizens many options for travel. For those with less income, subscription subsidies are available. The subsidies go further if the travelers use AVs during off-peak hours, when many AVs are being parked and recharged. AV dispatching is optimized to match shifts in demand, and travel is priced accordingly.

Back at work, Pat calls Roberto to make plans for the evening. The kids will return from school with a bicycle group and meet their babysitter. Roberto and Pat will hail an AV to go to dinner at one of the pop-up restaurants in the neighborhood park—knowing that if they drink too much, they won’t need to worry about driving themselves home.

Transportation 2040: The Nightmare

Now imagine the very different future that could come about if our community is unprepared for the three revolutions. Instead of adopting policies and incentives to encourage pooling of rides, the city allows the private desires of individuals and the competitive instincts of automotive companies to prevail. Traffic congestion gets worse as people who can afford AVs indulge themselves and send their cars out empty on errands. Most AVs are not electrified, and greenhouse gas emissions increase as people travel more. Time to spend with children and engage in community service becomes scarce. Transit services diminish as rich commuters abandon buses and rail and withdraw their support for transit. Those without driver’s licenses and cars continue to be marginalized as the divide between mobility haves and have-nots becomes a chasm. Meanwhile, suburbs sprawl as people seek affordable homes farther and farther out, opting for long commutes and cheap mortgages over proximity and more expensive real estate.

In this future, the middle-class Mathews-Ruiz family owns their own AV, which they’ve named Hal after the all-powerful computer in the movie 2001: A Space Odyssey. They live in the outer reaches of suburbia in a modest home. With vehicles still personal property (a residue of the twentieth century) and public transportation minimal, the family is trapped into spending nearly half their income on their beloved AV. They pay not only the high cost of the vehicle but also substantial expenses for remote parking (when not at home), required software updates, safety checks for software and hardware, and access to special AV lanes.

Their eighty-mile commute to jobs in the city center consumes about an hour and a half each way. When they pay to use a special AV lane, Hal can cruise at eighty miles per hour, but the trip is slowed down by Hal’s having to drive on mixed-use lanes to and from the high-speed freeway. And on some days, they opt for the mixed-use lanes on their way to and from work to reduce their toll costs, increasing their commute time to two hours each way.

The family buys an AV lane pass that allows them three thousand miles per month, after which they pay $0.40 per mile. With their eighty-mile commute, they use up the entire allocation each month just getting to and from work, forcing them to pay the higher off-subscription rates for the rest of the miles they log.

On days when one of them works late, they send Hal back empty to fetch that other person, or whoever finishes work earlier bides his or her time meandering in Hal along the streets in the crowded, congested city center. But the car is comfortable and the rider can use the time productively (even napping!).

Hal accrues still more miles traveling to pick up the children in the afternoon and running errands during the day. These errands might include a trip to pick up packages at a warehouse, where Hal is recognized with scanning technology and robots load and unload boxes. With all this travel, plus weekend recreation, Hal’s mileage generally exceeds five thousand miles per month.

The Fork in the Road

Will either of these futures materialize? Will the three revolutions usher in more vehicle use, increased urban sprawl, more marginalization of mobility have-nots, more expensive transportation, and higher greenhouse gas emissions? Or will they lead to reduced congestion and environmental impacts, safer communities, and easier and cheaper access for all (see figure 1.1)? The answer is unknown and will likely vary greatly across regions and countries. The three revolutions will unfold at different speeds in different places, creating waves of unintended—or at least unanticipated—consequences. We do have a say, but only if we wake up now to the speed and scope of change and how the coming revolutions will impact mobility and cities. Decisions made now about infrastructure and vehicle technologies will strongly influence the path and speed of change (a concept known as path dependence).

Figure 1.1

The fork in the road: Will transportation in 2040 be a dream or a nightmare?

Two of these revolutions—vehicle electrification and automation—are inevitable. The third, pooling, is less certain but in many ways most critical, especially as AVs come into being. All three have the potential to offer large benefits. EVs, including those powered by hydrogen, will decrease the use of fossil fuels and lower greenhouse gas emissions. Shared use of vehicles will reduce the number of cars on the road and thus congestion and emissions. Automation, in most cases, will reduce crashes. These benefits will be fully realized—indeed, enhanced—only if the three revolutions are integrated. Integration means EVs carrying multiple occupants under automated control. The result will be low-cost, low-carbon, equitable transportation.

Will this integration—the dream scenario—come about? As Automotive News, the principal trade magazine of the auto industry, put it, “There are millions of ways that flawed, messy, sometimes inconsiderate humans can mess up a perfectly good utopian scenario.”⁹ Many factors will shape the future, including how willing consumers are to accept new services and to share rides, how open transit operators are to embracing new mobility services as complements to their bus and rail services, and how inclined automakers are to become mobility companies instead of just car companies. Equally important will be the myriad policy, regulatory, and tax decisions made by local, state, and national governments—and how willing local politicians are to promote pooled services.

In the chapters that follow, we take a deep dive into the three revolutions and the potential synergies among them. We aim to inform and elevate the discussion about what needs to be done to unlock the enormous upside potential of vehicle electrification, automation, and ridesharing—and to avoid the downsides. Choices will be made by consumers, taxpayers, governments, transit operators, start-up mobility companies, and automotive and energy companies. With this book, we provide insight and knowledge that we hope will lead to wiser choices by all. We remain hopeful that more informed decisions will steer these revolutions toward the public interest and a better quality of life for everyone.

Overview of the Three Revolutions

Before looking at what might lie ahead, let’s briefly take stock of where we are now with the three revolutions. We elaborate in the following chapters.

Vehicle Electrification

It’s been a slow and circuitous journey. In 1900, about a quarter of all the cars in the United States were electric. They were quickly overtaken, though, by internal combustion vehicles and didn’t mount a comeback until a hundred years later. In 1990, California took an important first step: it adopted a zero-emissions vehicle (ZEV) mandate intended to curb the air pollution blanketing Los Angeles. The mandate suffered a tortured life for two decades, as industry launched lawsuits and technology evolved more slowly than anticipated. General Motors (GM) leased a thousand sporty electric cars in the 1990s, to some acclaim, but crushed them a few years later.

The breakthrough came in 2008, building on decades of research and development, largely government funded, on batteries and power electronics. That year, Tesla jolted the automotive world with a racy, high-performance electric sports car and followed in 2012 with the sleek, elegant, powerful Model S sedan. In 2010, Nissan introduced its Leaf, the first mass-market EV in almost one hundred years, followed quickly by the plug-in hybrid Volt from GM. Observing these impressive technological advances, California rejuvenated its ZEV mandate in 2012, requiring automakers to ramp up EV sales to 15 percent market penetration by 2025. Nine other states followed suit—and as this book goes to print, China and the European Union are poised to follow with similar mandates.

China soon surpassed California in EV sales, with a half million electric cars, trucks, and buses sold in 2016.¹⁰ China’s motivation was to eradicate local air pollution and create a domestic motor vehicle industry that could leapfrog to global dominance. Cumulative global sales of all EVs—a broad category that includes those powered by batteries and by hydrogen fuel cells—reached two million in 2016.¹¹ Overall market penetration was still just 1 percent globally, but there were big success stories in smaller markets, such as Norway, which has only 5.2 million people but saw market penetration of EVs approach 35 percent in 2017.¹²

By 2017, every major automaker was making massive investments in EVs. More than thirty-five different models were for sale in the United States, and many more elsewhere, especially in China, with every automaker planning to expand its offerings. Battery costs were dropping faster than anyone had anticipated, and nations around the world were implementing aggressive policies in support of EVs. Where subsidies were massive, as in Norway and parts of China, sales were soaring, but where subsidies were more modest, progress was slower. In California, EVs captured 5 percent of the new-car market in 2016. But why weren’t sales higher? In many regions of the state, consumers could find high-quality EVs for prices lower than comparable gasoline cars, thanks to generous incentives. The promise was great, but progress was measured.

Pooling and Sharing

The sharing economy and information technology have finally come to transportation, enabled by the release of the iPhone in 2007. Early champions Lyft and Uber first used smartphone apps to enlist private car owners in providing rides on demand. These initial offerings were glorified taxi services—glorified in the sense that they were superior to traditional taxi services in cost and convenience but still functioned like taxis. They were a first step toward transformational change.

Lyft’s introduction of Lyft Line in 2014 was a truly game-changing event for sustainable transportation, making it possible for two strangers going in the same direction to seamlessly share the trip. Uber quickly copied Lyft with its own version of pooling, UberPool. Riders pay about half the normal price in exchange for sharing the car with other riders and accepting a detour of a few minutes to pick up and drop off a second (or third) passenger. By 2016, 50 percent of Lyft and Uber riders in San Francisco—where Lyft Line and UberPool were first launched—were opting to share rides with strangers.

The next frontier for Lyft, Uber, and similar companies is incorporating AVs. In 2016, Lyft partnered with GM to begin developing an on-demand network of AVs across the United States, with Lyft cofounder John Zimmer predicting that by 2021, the majority of Lyft rides will be provided by AVs.¹³ In the same year, Uber launched its first fleet of technician-assisted self-driving cars on the streets of Pittsburgh and announced a partnership with Volvo to use AVs to offer shared rides.

But will large numbers of people be willing to share rides and get into robot cars with strangers? Given the small and dwindling number of conventional carpoolers, we should not assume people will forgo car ownership and readily share rides just because apps and smartphones make it easier.

Vehicle Automation

Self-driving cars once seemed like they belonged in some distant sci-fi future. First featured in the GM exhibit at the 1939 New York World’s Fair and then demonstrated in the real world by GM and Honda in 1997, they are now nearing commercialization. In 2010, Google announced it had a car that was safely self-driving around San Francisco—with no special roadside infrastructure or city retrofitting. This was just six short years after not a single automated car had been able to complete a course in the middle of the desert set up by the Defense Advanced Research Projects Agency (DARPA), the US military agency that financed the Internet.

The reality is that most new cars in Europe, the United States, Korea, and Japan are already partly automated. Many have adaptive cruise control, which speeds them up or slows them down based on the speed of the car in front of them. Many also have emergency braking, whereby the car takes over control and stops when it detects an imminent crash. And many also have lane-keeping and blind-spot assistance, whereby the car detects when it is crossing a lane (without a turn signal on) and when another car is in the driver’s blind spot. Much of the hardware and software needed to automate cars was already in place and commercial by 2015.

Today, a handful of companies are leading the way in developing AVs, with Tesla and Google (now Waymo) joining major automakers. By 2016, Tesla had sold fifty thousand vehicles to the public with Autopilot technology built in. Those cars were essentially capable of full autonomy, at least on freeways, where there are few obstructions and no traffic signals. Experience with partial automation and many test fleets gives companies like Tesla, GM, Ford, Toyota, BMW, Mercedes, Volvo, and Nissan the confidence to promise commercial sales of self-driving cars (but not driverless, as we explain in chapter 3) by 2020. Companies such as Uber say they will switch to automated cars as soon as they can in order to shed one of their most significant costs: drivers. There will also be big savings for the trucking industry, so it’s no surprise that start-ups like Otto (founded by ex-Googlers and purchased by Uber in 2016) are testing long-haul trucks that drive themselves.

These hopeful visions are premised mostly on technical progress. But in major transitions like this, success often requires more than technological superiority. There will be regulatory battles over vehicle licensing, corporate battles as car companies transition from manufacturing to mobility services, and public debates over fascinating but remote dilemmas like self-driving cars being forced to choose between either holding their course and hitting Grandma or swerving into a troop of Boy Scouts.

The Road Ahead

There is much uncertainty about how these three revolutions will play out. We know for sure that cars are here for the foreseeable future and that most cars will eventually become electrified and automated. Once cars are automated and the driver has been removed from the car, an array of new uses will become economical. Some changes will happen more slowly than others. Some changes will be undeniably positive for consumers, the economy, and society, while others might be disastrous—in terms of overall cost, environmental impact, and social equity.

The Elusive Promise of Safer Vehicles and More Equitable Access

Robot cars will undoubtedly be far safer than ones piloted by humans—eventually. Robots won’t drink and drive, won’t get tired, won’t be distracted by texting and children, and will have lightning-fast reflexes. In a world where all vehicles are self-driving, perhaps thirty thousand lives will be saved per year in the United States and millions of injuries avoided. But if humans are still driving or want to drive themselves, and if the transition from partially automated to fully driverless cars is delayed by safety regulators and local governments, as well as human reticence to cede control, we might not see changes in safety outcomes for many, many decades. In fact, deaths might increase if the technology is frozen one step short—with cars automated but requiring (unreliable) human intervention.

Likewise, the shared-use revolution, with the promise of low-cost travel, greater access for mobility-disadvantaged travelers, and less vehicle use, is also elusive. Personal cars are very expensive to own and operate, but most people have no alternative. Taking into account depreciation, insurance, fuel, registration, and maintenance, the full cost of owning and operating a new internal combustion engine car in the United States is about $0.57 per mile when the car is driven fifteen thousand miles per year. Used cars are less expensive, but not much, since older cars require more maintenance and more repairs. Where offered, bus and rail transit is much cheaper for travelers, but only because it is heavily subsidized. In the United States, riders pay only $0.25 per mile on average, but the true cost is five times that.

Travel by automated cars would be far cheaper if provided commercially, meaning the cars would be used more intensively. With the large up-front depreciation costs spread over many more miles, the cost could be reduced to about $0.20 per mile for an automated car logging one hundred thousand miles annually (with the vehicle assumed to cost $10,000 extra).¹⁴ And if the rides were pooled, as with UberPool and Lyft Line, with no driver, the costs would be pushed down to $0.10 or less per passenger mile. These costs are summarized in figure 1.2.¹⁵ This very low cost would assure more equitable access and also be highly disruptive in ways we are only beginning to imagine. But there’s no guarantee that the pooling services being offered by Lyft, Uber, and others will flourish, especially outside dense urban areas. Indeed, the dominant shared vehicles in 2017 were still Uber and Lyft cars carrying one passenger at a time.

Figure 1.2

Comparative costs of travel by different means in the United States. Cross-hatched bars indicate the range of costs due to length of trips (longer is cheaper per mile) and variation across cities.

Congestion and Environmental Impacts: Down or Up?

The effects on traffic congestion could go either way. In the most positive scenario, if all AVs are shared rather than privately owned, the congestion problem evaporates. Vehicle use would drop significantly (thanks to poolings), and road space utilization would improve dramatically. On-street and much off-street parking, including parking lots and garages, could be repurposed as public space—including wider sidewalks, more trees, bike lanes, and street furniture—and used for affordable housing and parks. On the other hand, if self-driving cars are privately owned by individuals, many of those expensive cars will spend considerable time circling the block endlessly and returning to remote parking lots instead of paying for parking. Think of what this will do for congestion. Similarly, some have suggested that private ownership of AVs will cause cities to sprawl into a new ring of “exurbs” as drivers forgo their distaste for car travel, in some cases abetted by being able to travel at higher speeds in AV-only lanes, though this sprawl could be restrained by land use rules, water and utility policies already in place, and people preferring urban lifestyles and walkable neighborhoods.

In terms of energy use and emissions, the potential synergies from combining the three revolutions are huge. Studies from independent groups such as the US Department of Energy,¹⁶ the International Transport Forum,¹⁷ the University of Washington,¹⁸ and the University of Texas¹⁹ suggest that shared, electric, automated vehicles could dramatically reduce greenhouse gas emissions. But all those studies are based on a few key assumptions. They assume travelers will share rides and make choices based solely on time and cost savings.

Perhaps. It’s well documented that travelers and consumers do not act as purely rational economic beings. As for the vehicles, if AVs run on petroleum and are individually owned, sometimes cruising empty between drop-offs and pickups, emissions and energy use will increase. figure 1.3 highlights the many ways in which automated cars could affect energy use and greenhouse gas emissions. Increases would be due to higher speeds, which greatly increase energy use, and increased travel, since people would not see time in a vehicle as wasted and because the young, the physically disabled, and others would be able to access vehicles more easily. Of course, if the vehicle were carrying multiple riders, the energy and greenhouse gas impact per trip (or per person mile) would be sharply reduced.

Figure 1.3

Projected changes in energy consumption due to vehicle automation. For each contributing factor, a range of values is estimated. Source: Zia Wadud, Don Mackenzie, and Paul N. Leiby, “Help or Hindrance? The Travel, Energy, and Carbon Impacts of Highly Automated Vehicles,” Transportation Research Part A 86 (April 2016): 1–18.

In a study of urban passenger travel worldwide, researchers at the University of California, Davis, estimated that with driverless cars but little pooling and electrification, greenhouse gas emissions would increase 50 percent and vehicle use 15 to 20 percent between now and 2050. In contrast, in a dream scenario where driverless cars are pooled and electrified, vehicle use would drop by 60 percent compared to business as usual, greenhouse gas emissions would drop by 80 percent, and overall costs of vehicles, fuel use, and infrastructure would drop by more than 40 percent—representing a savings of $5 trillion per year. Though these scenarios would play out in very different ways from one country to another, the researchers estimated that the scale of these reductions would be similar across most of the world.²⁰

Impacts on Road Financing and Jobs

Another uncertain and potentially problematic outcome is the issue of who pays for roads. The motor fuels tax, imposed by the federal government and states, was originally intended as a user fee for road construction and upkeep that people pay in rough proportion to the distance they drive. This simple and sound concept began to be undermined as vehicles became more efficient. Users of energy-efficient vehicles use much less fuel and thus pay much less per mile, while owners of fully electric vehicles do not pay any tax at all. This undermines federal and state funding for roads and transit; it also means that as more people purchase EVs, even less gas tax is paid. Because EVs today are disproportionately owned by higher-income households, and older cars, often gas guzzlers, are owned by less affluent people, the burden of paying for roads via the fuel tax is gradually being shifted onto lower-income drivers.

In our dream scenario of shared electric AVs, today’s deteriorating user-based financing system completely collapses as vehicle use shrinks and most vehicles don’t consume gasoline. This scenario is also a disaster for local governments, which depend on a mix of parking fees and fines, speeding tickets, and other car-related fees and fines to finance local roads, parking garages, and transit.

A different and better taxation system will need to be created—preferably one that imposes the full cost of car travel on drivers (and passengers), including not only the cost of building and maintaining roads but also pollution, congestion, sprawl, and climate change. Many observers envision a gradual, longer-term transition from fuel taxes to some form of mileage-based user fee that charges vehicles directly for their use of roads. Such fee systems raise their own set of equity considerations,²¹ but they do lend themselves to refinements that take into account whether vehicles are used for shared multipassenger service and whether AVs are owned by individuals or commercial mobility providers.

One other pivotal concern is job loss. What of those 3.5 million freight and delivery truck drivers and 665,000 bus drivers in the United States? And the taxi and livery drivers—90,000 of them registered in New York City alone, not counting Uber and Lyft drivers? All these jobs are presumably at risk. But the story is more complicated. A massive shift to shared mobility, say 10 to 40 percent or more of travel, would convert a private, personal activity into a commercial activity. Many new jobs would be created for those managing the new businesses, cleaning and maintaining the vehicles, designing and updating the computer systems, dealing with disgruntled customers, and so on. Similarly, truck drivers might find new work managing aspects of automated freight hauling that haven’t even been imagined yet. The job impact of pooling and automation is highly uncertain. With an integration of the three revolutions, the net effect may well be positive.

Steering toward the Public Interest

The three revolutions are part of larger societal and economic transformations. Globalization, industry automation, and part-time service jobs threaten social cohesion and have contributed to income inequality. Will the mobility sector be part of the problem or part of the solution?

Left to the market and individual choice, the likely outcome is more vehicles, more driving, and a slow transition to electric cars—reflecting many elements of our nightmare scenario. It is the tragedy of the commons, where no one is championing the public interest. Yes, it is possible to imagine a sustainable AV future without pooling premised on individual ownership—at least in rich countries such as the United States—but only if the vehicles are electrified, road pricing is enacted judiciously, conventional transit continues to be funded, and transport finance is transformed so that those with low incomes and physical disabilities still have access to work, school, health care, and other activities. But that is a lot of “ifs.” If all these contingencies don’t materialize, the alternative to pooling is the nightmare scenario.

The future is likely to be far more positive if government intervenes and guides the invisible hand of the marketplace to avoid the excesses of self-interested behaviors. Antitrust laws, along with regulations focused on pollution prevention and worker protections, are all aimed at minimizing the direst excesses of capitalism. In a dream scenario, they guide powerful stakeholders toward the public interest.

That is the challenge here. For the first time in many decades, mobility is on the cusp of not just one transformation but several. The dream scenario depends critically on the merging of electrification, pooling, and automation.

In practice, how does this dream scenario unfold? The key first step seems to be the creation of travel choices that increase convenience and comfort and reduce personal travel costs. Not until people have choices will they be willing to give up their personal cars—the linchpin for a cascade of changes. And not until they have choices will they accept pricing policies—policies that encourage pooling and discourage ownership of personal vehicles. Pooling and pricing are key to setting us on a path to less costly, less resource-intensive, more widely accessible, and more sustainable transportation.

Change is going to happen. And it will be transformational. Many will try to slow it, while others will try to accelerate it. We can be ostriches with buried heads and hope that disinterest, disengagement, and the normal workings of the marketplace will somehow turn out well. Or we can apply our best thinking to harness vehicle electrification, mobility sharing, and automation to create better cities, a livable planet, and a future that serves us all. Read on.

Key Policy Goals

• These are the overarching goals in transitioning to a safer and more economical, environmentally benign, and equitable transportation future:

  • User incentives: Incentivize travelers to choose pooled mobility services over individual ownership of vehicles now, and more so when driverless vehicles become available.

  • Pooling and EVs: Encourage mobility service companies to embrace pooling and EVs (including hydrogen fuel cell vehicles). Motivate automakers to design AVs for pooling that are powered by electricity or hydrogen.

  • Equity and transit: Encourage transit operators and mobility service companies to collaborate in providing more access and service at lower cost.

  • Land use: Begin redesigning cities for a transportation system that uses less parking and fewer roads and is more conducive to pooling, walking, biking, and affordable living.

• In the following chapters, we elaborate on specific policies that could be adopted to achieve these broad goals.