Automated Vehicles

Like many industries, the automated vehicle industry faced setbacks due to this year’s many COVID-19 related local and regional lockdowns. In the spring, as the first wave of the virus spread, many companies had to stop testing to protect the human safety drivers and, in the case of Bay Area companies, follow local “shelter in place” orders. One partial exception to the rule was Waymo, which has been testing fully automated vehicles without safety drivers in Arizona, was able to keep some of those fully automated vehicles operating, since there were no drivers involved.

Beyond shutting down on-the-road testing, the AV industry has seen other COVID-related fallout. Due to the pandemic Ford delayed the launch of their robotaxi service to 2022, while GM’s Cruise unit laid off 8% of their staff in May. Yet desire to invest in the AV industry appears to remain strong. Zoox, which had (at least temporary) laid off its safety drivers in April, was bought by Amazon in June. Over the summer companies have begun to announce new testing sites – with Aurora testing automated semis and cars in Dallas-Fort Worth, and a Chinese AV developer, AutoX, launching a test around PayPal’s headquarters in San Jose, CA. Closer to home, Russian AV developer Yandex announced it would begin testing in Ann Arbor, their first test in the US, while May Mobility’s AV service in Grand Rapids will resume service at the end of August.

Indeed, two other developments in Michigan show that AV and mobility-related work is still an important focus, even during periods of major upheaval. In July the state of Michigan launched the Office of Future Mobility and Electrification, which is led by the “chief mobility officer.” The office’s structure and mission is based off work done by Detroit’s Office of Mobility Innovation – and hopes to recreate that office’s success at a state level. Part of the office’s mission will be to consolidate the work of 135 different councils, boards, and commissions spread out across 17 state agencies and departments – all of which have been working on some element of mobility. Earlier this month a second major announcement pointed to just how dedicated the state seems to be toward new mobility tech. On August 13th, a public-private partnership, named “CAVNUE,” was announced, with its goal being the creation of a 40-mile long testing corridor between Detroit and Ann Arbor. The corridor would be designed for testing both connected and automated vehicles as well as infrastructure. If this project is successful, it would be a major boon for the many companies in Southeast Michigan – and would signal a move to greater public testing of emerging mobility technology beyond more controlled environments like MCity.  

One lesson of the past year has been that the future can change very quickly, making rosy predictions of future AV successes harder to believe than in “the before time.” But these developments seem to show the AV industry finding its way forward. The future promise (and challenge) of AVs hasn’t diminished, even in our rapidly changing present.

Last week I discussed the California Superior Court decision that ruled that under California law Uber and Lyft must classify their ridesharing drivers as employees, rather than independent contractors. In response to that ruling, both companies had threatened to shut down service across the state. Yesterday, an appeals court issued a stay on that ruling, allowing both companies to continue operations, “pending resolution” of their appeal of the initial order. As I mentioned in my last blog post, the rideshare giant’s strategy currently appears to be “run out the clock,” until the November election, when California voters will decide on Proposition 22, which would establish a new classification for drivers. So for now those Californians who are willing to brave getting into a rideshare will be able to do so – while Uber and Lyft also explore more creative solutions, in case Prop 22 doesn’t pass.

Also on Thursday, another court case tied to Uber was just starting. Federal prosecutors in San Francisco filed criminal charges against Uber’s former security chief, Joe Sullivan. Sullivan is charged with two felony counts for failing to disclose a 2016 Uber data breach to federal investigators who were investigating similar earlier incidents that had occurred in 2014. In the 2016 incident, an outside hacker was paid $100,000 by Uber after the hacker revealed they had acquired access to the information of 57 million riders and drivers. Beyond the payment, Uber faced further criticism for failing to reveal the incident for a full year. Two of the hackers involved later plead guilty to charges related to the hack, and they are both awaiting federal sentencing. In 2018 Uber paid $148 million to settle a suit brought by state attorneys general related to the hack, while the FTC expanded a previous data breach settlement in reaction to the incident. Beyond the lack of transparency (to the public and law enforcement) Uber’s major misstep, at least in my view, is the payment itself. While many companies, Uber included, sponsor “bug bounties,” where outside security researchers are rewarded for reporting security flaws in a company’s products, this payment fell outside of that structure. Rather, it seems more like a ransom payment to less than scrupulous hackers. While Uber is far from the only company to have faced data breaches (or to have paid off hackers), this case should be a wake-up call for all mobility companies – a reminder that they have to be very careful with the customer data they are collecting, least they fall prey to a data breach, and, just as importantly, when a breach occurs, they have to face it with transparency, both to the public and investigators.  

The third Uber-related this month involves another former Uber employee, Anthony Levandowski, who was sentenced to 18 months in prison for stealing automated vehicle trade secrets from Google. In 2016, Levandowski left Google’s automated vehicle project to start his own AV tech company, which was in turn acquired by Uber. Levandowski was accused of downloading thousands of Google files related to AVs before he left, leading to a suit between Google’s Waymo and Uber, which was settled for roughly $250 million. There are a lot more details involved in the case, but it highlights some of the many challenges Uber, and the mobility industry at large, face.

Mobility and AVs are a huge business, with a lot of pressure to deliver products and receive high valuations in from investors and IPOs. That can incentivize misbehavior, whether it be stealing intellectual property or concealing data breaches. Given how central mobility technologies are to people’s daily lives, the public deserves to be able to trust the companies developing and deploying those technologies – something undermined by cases like these.

As the COVID-19 pandemic continues and our memories of the “before time” feel ever more distant, some have begun to wonder how this crisis and its aftermath could change how and where people live. Will people abandon expensive and dense major cities for smaller cities, suburbs or even small towns? On the one hand, I’ll admit that living in a small city like Ann Arbor has made weathering the lock down rather easy, which could lead credence to these ideas. Personally, I’ve had no issues finding supplies, or taking a walk without running into too many other people (though my apartment building’s shared laundry rooms are now a fraught location). Of course, Ann Arbor, a wealthy, educated college town with excellent access to medical care has a lot of resources other cities do not, so it may not be the best example.

Alternatively, there are those who argue our cities won’t actually change that much post-COVID-19, and there are even ways that the outbreak could make cities better (with the proper investment). Cities have survived disease outbreaks for millennia, and given that so much of our economy, culture, and infrastructure is built around cities it would be hard to seismically shift to some other model of living. Yet the economic upheaval that the pandemic has ushered in will no doubt influence where and how people live, and could last a good deal longer than the disease itself.

So what changes are well already seeing in cities, and what could that indicate about where we’re heading? In a number of cities, including New York, Seattle, and Oakland, are closing streets to open up more space for pedestrians and cyclists. Streets could also be closed to provide more outdoor space for restaurants, to help them reopen while preserving some measure of social distancing. New Zealand has gone as far as to make such street alterations national policy. Cities and towns in that nation are able to apply for funding to immediately expand sidewalks and modify streets, with the national government covering 90% of the cost. Some suggest these closures and modifications should be permanent – that we should take this opportunity to create more walkable and bikeable cities now, when we have the chance. In many ways these modified streets are similar to proposals for automated vehicle (“AV”) dominated cities. Supporters believe that wide adoption and deployment of AVs would mean more streets could have one lane of traffic in each direction, with the extra space turned over to alternative uses. The current demands of social distancing dovetail with those ideas – could cities use the current crisis to prepare themselves for an autonomous future? Given the difficulty of building new infrastructure, it may not be a bad idea to get ahead of the curve.

As noted by Phillip in a post earlier in the crisis, another effect of the global lockdown has been improved environmental conditions in cities around the globe. In India, for example, where cities have significant pollution problems, massive reductions in travel have led to clear skies. For the first time, we are seeing clear examples of what cleaner energy production could bring (pun intended). Such improvements could lead residents to demand continued reductions in emissions even after this crisis passes. These and other changes made to cities in the short term to cope with lockdowns and social distancing could dictate the future of urban design, but only if governments and citizens are willing to adopt them and protect them from being undone once the crisis passes.

P.S. Those of you who are interested in buying a bike to help navigate the new socially-distanced world may run into an issue – just like masks, cleaning supplies, and toilet paper, bikes are now becoming a scarce resource in some places.

Cite as: Bryant Walker Smith, How Reporters Can Evaluate Automated Driving Announcements, 2020 J. L. & MOB. 1.



This article identifies a series of specific questions that reporters can ask about claims made by developers of automated motor vehicles (“AVs”). Its immediate intent is to facilitate more critical, credible, and ultimately constructive reporting on progress toward automated driving. In turn, reporting of this kind advances three additional goals. First, it encourages AV developers to qualify and support their public claims. Second, it appropriately manages public expectations about these vehicles. Third, it fosters more technical accuracy and technological circumspection in legal and policy scholarship.

This third purpose goes to the core of this interdisciplinary journal. Legal and policy scholarship about emerging technologies often relies at least in part on popular reporting. On one hand, this reporting can provide timely and accessible insights into these technologies, particularly when the scientific literature cannot. On the other hand, this reporting can reflect misconceptions based on incomplete information supplied by self-interested developers—misconceptions that are then entrenched through legal citation. For example, I have pushed back against claims that automated driving will be a panacea, that its technical challenges have long been “solved,” and that nontechnical issues involving regulation, liability, popularity, and philosophy are therefore the paramount obstacles to deployment.

Common to many of these misconceptions is the question of whether automated driving is finally here. AVs were 20 years away from the late 1930s until the early 2010s and have been about five years away ever since. This is clearly a long history of misplaced optimism, but more recent predictions, while still moving targets, are now proximate enough to realistically drive decisions about investment, planning, and production. Indeed, of the companies that claim to be even closer, some really are—at least to automated driving of some kind.

The “what” of these predictions matters as much as the “when,” and the leading definitions document for automated driving—SAE J3016—is helpful for understanding this what. The figure below offers a gloss on these definitions, including the widely (mis)referenced levels of driving automation. No developer has credibly promised level 5 (full automation) anytime soon. But many are working toward various applications of level 4 (high automation), which could, depending on their implementation, include everything from low-speed shuttles and delivery robots to traffic jam automation features and automated long-haul trucks. When anything approaching level 5 does becomes a reality, it will likely be an afterthought in a world that has already been revolutionized in a hundred other ways.

Figure: A Gloss on SAE J3016

Your role in driving automation

Driving involves paying attention to the vehicle, the road, and the environment so that you can steer, brake, and accelerate as needed. If you’re expected to pay attention, you’re still driving — even when a vehicle feature is assisting you with steering, braking, and/or accelerating. (Driving may have an even broader legal meaning.)

Types of trips

  • You must drive for the entire trip
  • You will need to drive if prompted in order to maintain safety
  • You will need to drive if prompted in order to reach your destination
  • You will not need to drive for any reason, but you may drive if you want
  • You will not need to drive for any reason, and you may not drive

Types of vehicles

  • Vehicles you can drive
  • Vehicles you can’t drive

Types of vehicle features

These are the levels of driving automation. They describe features in vehicles rather than the vehicles themselves. This is because a vehicle’s feature or features may not always be engaged or even available.

The operational design domain (“ODD”) describes when and where a feature is specifically designed to function. For example, one feature may be designed for freeway traffic jams, while another may be designed for a particular neighborhood in good weather.

By describing a feature’s level of automation and operational design domain, the feature’s developer makes a promise to the public about that feature’s capabilities.

Assisted driving features

  • L0: You’re driving
  • L1: You’re driving, but you’re assisted with either steering or speed
  • L2: You’re driving, but you’re assisted with both steering and speed

Automated driving features

  • L3: You’re not driving, but you will need to drive if prompted in order to maintain safety
  • L4: You’re not driving, but either a) you will need to drive if prompted in order to reach your destination (in a vehicle you can drive) or b) you will not be able to reach every destination (in a vehicle you can’t drive)
  • L5: You’re not driving, and you can reach any destination

As the following questions for reporters make clear, automated driving is much more than just a level of automation. The questions, which fall into five overlapping categories (human monitoring, technical definitions, deployment, safety, and reevaluation), are:

1. Human monitoring

1.1. Is a person monitoring the AV from inside the vehicle? Why? Are they always paying attention? How can they intervene? How often do they intervene? How are they supervised?

1.2. Is a person monitoring the AV from outside the vehicle? Why? Are they always paying attention? How can they intervene? How often do they intervene? How are they supervised?

1.3. Is a person monitoring the AV from a remote center? Why? Are they always paying attention? How can they intervene? How often do they intervene? How are they supervised?

1.4. What are specific examples of difficult scenarios in which a person did not intervene? In which a person unnecessarily intervened? In which a person necessarily intervened? What form did this intervention take?

1.5. At any moment, what is the ratio between the number of people who are monitoring and the number of AVs that are deployed?

2. Technical definitions

2.1. What level of automation corresponds to the design intent for the AV? What level of automation corresponds to how the AV is actually being operated?

2.2. In what environment is the AV operating? On roads open to other motor vehicles? To bicyclists? To pedestrians?

2.3. What infrastructure, if any, has been changed or added to support the AV in this environment?

2.4. If the AV perceives that its path is obstructed, what does it do? For example, does it wait for the obstruction to clear, wait for a person to intervene, or plan and follow a new path?

3. Deployment

3.1. What is the AV’s deployment timeline? For how long will it be deployed? Is this a temporary or permanent service?

3.2. Who can buy the AV or its automated driving feature? Under what conditions?

3.3. Who can ride in, receive products or services from, or otherwise use the AV? Under what conditions?

3.4. As part of the deployment, who is paying whom? For what?

3.5. What promises or commitments has the developer of the AV made to governments and other project partners?

3.6. What previous promises, commitments, and announcements has the developer made about their AVs? Have they met them? Do they still stand by them? What has changed, and what have they learned? Why should we believe them now?

4. Safety

4.1. Why do the developer of the AV and any companies or governments involved in its deployment think that the deployment is reasonably safe? Why should we believe them?

4.2. What will the developer of the AV and any companies or governments involved in its deployment do in the event of a crash or other incident?

5. Reevaluation

5.1. Might the answers to any of these questions change during the deployment of the AV? How and why? What will trigger that change?

The remainder of this article explores these questions with a view toward assessing the reality behind a given automated driving announcement or activity. To this end, it is important to understand that a vehicle that requires an attentive safety driver is not truly an automated vehicle. Aspirational, yes. But actual, no. This point underlies many of the questions that follow.

Human Monitoring

Is a person monitoring the AV from inside the vehicle? Why? Are they always paying attention? How can they intervene? How often do they intervene? How are they supervised?

Imagine that as you are boarding a plane, the captain announces that “I’ll be using autopilot today. We’ll be pushing off shortly. Have a nice flight.” How do you feel?

Now imagine that the captain instead announces that “You’ll be using autopilot today, because I’m getting off. You’ll be pushing off shortly. Have a nice flight.” How do you feel now?

Just as there is a significant difference between these two scenarios, automated driving under the supervision of a safety driver is not the same as automated driving without this supervision. Yet news headlines, ledes, and even entire articles often describe only “driverless” vehicles—even when those vehicles are supervised by at least one trained safety driver who is physically present for every trip.

This confusion has consequences. Casual readers (and even reporters) may believe that an automated driving project is far more technically advanced or economically feasible than it really is. They may therefore be more likely to look for nontechnical explanations for the seemingly slow rollout of automated vehicles. Ironically, they may also discount truly significant news, such as Waymo’s recent decision to remove safety drivers from some of its vehicles.

Reporters should therefore ask whether an automated vehicle is being operated with or without a safety driver inside it, and they should include the answer to this question in the first rather than the final paragraph of their stories. Related questions can then provide further context. Is the safety driver seated in the traditional driver’s seat (if there is one) or elsewhere in the vehicle? Can they immediately brake, steer, and accelerate the vehicle? And, in the interest of safety, how are they supervised? As Uber’s 2018 fatal crash tragically demonstrated, a system’s machine and human elements can both be fallible.

For the most part, an AV developer that uses safety drivers is not yet confident that its vehicles can reliably achieve an acceptable level of safety on their own. This is still true even if a vehicle completes a drive without any actual intervention by that safety driver. At least in the United States, alternative explanations for retaining the safety driver—to comply with ostensible legal requirements, to reassure passengers, or to perform nondriving functions—are generally lacking.

At the same time, AV developers might reach different conclusions about the requisite level of safety or the requisite level of confidence in that safety. To use a very limited analogy: A rock climber’s rejection of ropes and harnesses probably says more about the climber’s confidence than about their skill.

Is a person monitoring the AV from outside the vehicle? Why? Are they always paying attention? How can they intervene? How often do they intervene? How are they supervised?

A safety driver might be present near rather than inside a vehicle. For example, a demonstration of a small delivery vehicle that is not designed to carry people may nonetheless involve a safety driver seated in a car that trails the delivery vehicle. Reliance on such a safety driver places a significant technical and economic asterisk on claims about the capabilities of these delivery vehicles. Because reliance on safety drivers also involves reliance on a robust communications system, reliance on them also introduces an additional issue of safety.

Tesla’s recent introduction of its Smart Summon feature also shows why unoccupied does not necessarily mean driverless. This feature does not reach the threshold for automated driving—and certainly not “full self-driving”—because it is designed with the expectation that there will be a human driver who will supervise the vehicle from the outside and intervene to prevent harm. Emphasizing that the user is still a driver may help to temper claims and assumptions that could lead to the dangerous misuse of this driver assistance feature.

Is a person monitoring the AV from a remote center? Why? Are they always paying attention? How can they intervene? How often do they intervene? How are they supervised?

For years, one of the more contentious issues in the automated driving community has involved what might be neutrally termed “remote facilitation of the driving task.” This phrase encompasses a broad spectrum of potential roles performed by actors outside the vehicle—roles that are important to understanding the technical and safety claims made by developers of automotive technologies.

On one side of the spectrum lies remote driving, in which a human driver who may be many miles away from a vehicle uses a communications system to perceive the vehicle’s driving environment and to steer, accelerate, and brake in real time—what SAE J3016 calls “performance of the dynamic driving task.” This remote driving is orthogonal to automated driving (in other words, neither its synonym nor its antonym). Indeed, some automated driving developers skeptical of remote driving are eager to differentiate the two in both language and law.

On the other side of the spectrum lies network monitoring. An automated driving company might maintain a facility in which human agents collectively monitor its AVs, communicate with the users of those vehicles, and coordinate with emergency responders. While stressing that their human agents are not performing the dynamic driving task, some AV developers have been vague about what specifically these agents are and are otherwise not doing.

Journalists, however, can be concrete in their questioning. They can ask whether there is a remote person assigned to or available for each vehicle, what that person does during the vehicle’s normal operation, and what that person does in less common situations. For example, imagine that an AV approaches a crash scene and concludes that it cannot confidently navigate by itself. What role might a remote agent play? Might this person give the vehicle permission to proceed? Might they manually identify roadway objects that the AV could not confidently classify? Might they sketch a rough travel path for the AV to follow if the AV agrees? Might they direct the AV to follow the path even if the AV would otherwise reject it? Or might they actually relay specific steering, accelerating, and braking commands to the AV?

How a company answers these questions can provide insight into the maturity of its automated driving program. If the company uses physically present safety drivers in its deployments (as most still do), then these questions are largely speculative. But if the company plans to remove these safety drivers, then it should have careful and concrete answers. And if the company declines to share these answers, one might reasonably inquire why.

What are specific examples of difficult scenarios in which a person did not intervene? In which a person unnecessarily intervened? In which a person necessarily intervened? What form did this intervention take?

While anecdotes alone are not enough to establish reasonable safety, they can be helpful in measuring progress. An automated driving developer that has been testing its vehicles will have stories about unusual situations that those vehicles (and their safety drivers) encountered. Many of these developers may be happy to share situations that the automated vehicle handled or could have handled without intervention. But pairing these with situations in which human intervention was necessary provides important context. And a company’s willingness to share these more challenging stories demonstrates its trustworthiness.

At any moment, what is the ratio between the number of people who are monitoring and the number of AVs that are deployed?

Economic feasibility offers another metric for automated driving—and one that is intertwined with technical feasibility. Economically, automated driving is both attractive and controversial in large part because, true to its name, it promises to reduce the need for human drivers. Asking whether this is in fact happening—that is, whether the ratio of human monitors to automated vehicles is less than 1.0—is another way to assess the technical progress of an automated driving program.

This may be especially helpful with respect to pilot projects involving specialized vehicles traveling at low speeds in limited areas such as airports, downtowns, and shopping malls. There have been and will likely continue to be numerous announcements about these projects across the country. But so long as these vehicles are deployed with at least one safety driver on board, their economic viability is unclear. After all, their hosts could have achieved (and could still achieve) the same functional benefits by simply deploying conventional fleets.

Technical definitions

What level of automation corresponds to the design intent for the AV? What level of automation corresponds to how the AV is actually being operated?

Automated driving developers are almost certainly familiar, though not necessarily proficient, with the levels of driving automation defined in SAE J3016. They may even reference these levels in their announcements—correctly or not. Understanding the levels may help to assess the claims.

Most automated driving development is focused on levels 3 and 4. On one side, levels 0, 1, and 2 are in fact driver assistance rather than automated driving, and a credible developer should not suggest otherwise. After all, features at these levels only work unless and until they don’t, which is why a human driver is still needed to supervise them. On the other side, level 5 describes a feature that can operate everywhere that humans can drive today. But while this is the hope of many automated driving developers, it remains a distant one.

A confusing quirk in the levels of automation is the difference between what I call an aspirational level and what I call a functional level. The aspirational level describes what an automated driving developer hopes its system can achieve, whereas the functional level describes what the automated driving developer assumes its system can currently achieve. For example, most developers of low-speed automated shuttles envision level 4 automated driving, which would not require a human driver for safe operation. But most of these developers still keep their systems under the supervision of human safety drivers who are expected to pay attention, which corresponds to level 2 rather than level 4. Nonetheless, because SAE J3016 focuses on design intent, developers of these systems correctly characterize them as level 4 (the aspirational level) rather than level 2 (the functional level).

Similarly, California’s Department of Motor Vehicles considers automated vehicles that are merely being tested to be “autonomous” even though their safe operation still requires a human safety driver. Otherwise, rules requiring a safety driver absent specific permission otherwise would apply to a null set. Because of this interpretation, companies that are testing or deploying automated driving features in California must comply with these rules, while companies that are testing or deploying mere driver assistance features need not. This is why Uber needed permission to test its automated vehicles in California, but Tesla did not need permission to make its Autopilot or Smart Summon driver assistance features available in that state. Yet, as these examples suggest, testing an automated driving feature is in many ways technically indistinguishable from using a driver assistance feature.

Asking about the aspirational level of automation invites a company to make a public characterization that has marketing and regulatory implications. And asking about the functional level of automation invites a company to temper its aspirations with the current limitations of its technologies.

References to the levels of automation may be helpful in discussions with companies but are generally not necessary or even helpful when reporting to the public. Instead, key phrases can more clearly communicate the current state of a given technology. Three of the most important are:

  • “A driver assistance feature that still requires a human driver to pay attention to the road” (levels 1 and 2)
  • “A vehicle that is designed to drive itself but needs a safety driver until it can reliably do so” (aspirational level 4)
  • “A vehicle that drives itself without the need for a safety driver” (functional level 4)

In what environment is the AV operating? On roads open to other motor vehicles? To bicyclists? To pedestrians?

Automated vehicles have been a reality for decades: They are called elevators, escalators, people movers, and automated trains. But whereas these vehicles operate in highly controlled environments, automated motor vehicles are particularly challenging in large part because the driving environments they will face are so challenging.

Below level 5, however, these driving conditions are limited. SAE J3016 terms these driving conditions the operational design domain, and this ODD is essential to defining an AV’s capabilities. For example, some automated driving features may operate only on freeways, and some AVs may be restricted to certain low-speed routes within certain neighborhoods. Indeed, early automation activities are generally characterized by some combination of slow speeds, simple environments, and supervised operations.

Developers should be upfront about these limitations in their announcements—and if they are not, reporters should ask whether and how the AVs mix with other road users, including pedestrians, bicyclists, and conventional drivers. There is a big difference, for example, between deploying in complex mixed traffic and deploying on a dedicated route with no other traffic.

As an aside: State vehicle codes apply to public roads, and they may also apply to private facilities such as parking garages and private roads that are nonetheless open to the public. For this reason, AVs that are deployed only in privately controlled areas may still have to comply with state laws generally applicable to motor vehicles as well as state laws specific to AVs. Similarly, these laws may (or may not) also apply to delivery robots that travel on sidewalks and crosswalks. Developers that suggest otherwise can be asked to explain the basis for their legal conclusion.

What infrastructure, if any, has been changed or added to support the AV in this environment?

Many AV announcements involve specific tests, pilots, or demonstrations that may or may not be easily replicated in another location and scaled to many more locations. An AV that can accept today’s roads as they are—inconsistently designed, marked, maintained, and operated—will be much easier to scale than one that requires the addition or standardization of physical infrastructure. Even if they would be beneficial and practical, infrastructure changes are nonetheless important considerations in evaluating scalability. For this reason, automated driving developers should be asked to identify them.

If the AV perceives that its path is obstructed, what does it do? For example, does it wait for the obstruction to clear, wait for a person to intervene, or plan and follow a new path?

Even infrastructure that is well maintained will still present surprises, and how an AV is designed to deal with these surprises provides some insight into its sophistication. Many early automated vehicles would simply stop and wait if a pedestrian stepped into their path (or a drop of rain confused their sensors). Even today, many AVs rely on frequent human intervention of some kind. This question accordingly invites a developer to describe the true capabilities of its system.

Deployment

What is the AV’s deployment timeline? For how long will it be deployed? Is this a temporary or permanent service?

Many recent AV announcements have focused less on technical capabilities and more on actual applications, from shuttling real people to delivering real products. These specific applications often involve partnerships with governments, airports, retailers, shippers, or property managers. But it can be unclear whether these applications are one-time demonstrations, short-term pilots, or long-term deployments. Querying—and, in the case of public authorities, requesting records about—the duration of these projects helps to understand their significance.

Who can buy the AV or its automated driving feature? Under what conditions?

There is an important difference between an automated driving developer that is marketing its actual system and a developer that is merely marketing itself. Yet automated driving announcements tend to conflate actual designs, promises of designs, and mere visions of designs. Automakers previewing new vehicle features, shuttle developers announcing new collaborations, and hardware manufacturers touting new breakthroughs all invite the question, “Can I actually buy this vehicle now?”

Who can ride in, receive products or services from, or otherwise use the AV? Under what conditions?

This same logic applies to announcements about services that purportedly involve automated driving. The launch of an automated pizza delivery service open to everyone in a city is much more significant than the staged delivery of a single pizza by a single AV. So too with the automation of long-haul shipping, low-speed shuttles, and taxis. Services that at least part of the public can actually and regularly use are far more significant than one-off demonstrations.

As part of the deployment, who is paying whom? For what?

For the reasons already discussed, the economics of early deployments can be hazy. Why are automated shuttles, each with its own safety driver, more cost-effective than conventional shuttles? Why are automated trucks, each with its own safety driver, more cost-effective than conventional trucks? The financial arrangements with project partners—especially public authorities subject to open records laws—can offer some insight into whether these early deployments provide tangible benefits or are instead largely exploratory or promotional.

What promises or commitments has the developer of the AV made to governments and other project partners?

When project partners are involved for long-term rather than near-term benefit, it can be helpful to query their expectations. Imagine, for example, that an airport or retirement community announces its intent to host automated shuttles that are supervised by safety drivers. When has the developer of these shuttles suggested or promised that safety drivers will no longer be necessary? And who bears the cost of paying these drivers in the interim?

What previous promises, commitments, and announcements has the developer made about their AVs? Have they met them? Do they still stand by them? What has changed, and what have they learned? Why should we believe them now?

Because innovation is unpredictable, claims about deployment timelines may turn out to be incorrect even if they are made in good faith. However, the companies (or people) responsible for these claims should acknowledge that they were wrong, explain why, and temper their new claims accordingly. Reporters should demand this context from their subjects and report it to their audience. Of course, a commercial emphasis on speed and controversy can make this especially challenging, in which case the headline “Company X makes another claim” could at least be used for the more egregious offenders.

Safety

Why do the developer of the AV and any companies or governments involved in its deployment think that the deployment is reasonably safe? Why should we believe them?

While the broader topic of AV safety is beyond the scope of this article, it should occupy a prominent place in any automated driving announcement. For years, I have encouraged companies that are developing new technologies to publicly share their safety philosophies—in other words, to explain what they are doing, why they think it is reasonably safe, and why we should believe them. Journalists can pose these same questions and push for concrete answers.

The phrasing of these questions matters. For example, a company might explain that its AV testing is reasonably safe because it uses safety drivers. But it should also go further by explaining why it believes that the presence of safety drivers is sufficient for reasonable safety. Conversely, if a company does not use safety drivers, it should explain why it believes that they are not necessary for reasonable safety. And in answering these questions, the company may also have to detail its own view of what reasonable safety means.

In this regard, it is important to recognize that safety is not just a single test. Instead, it includes a wide range of considerations over the entire product lifecycle, including management philosophy, design philosophy, hiring and supervision, standards integration, technological monitoring and updating, communication and disclosure, and even strategies for managing inevitable technological obsolescence. In this way, safety is a marriage rather than just a wedding: a lifelong commitment rather than a one-time event.

What will the developer of the AV and any companies or governments involved in its deployment do in the event of a crash or other incident?

Safety is not absolute. Indeed, just because an AV is involved in a crash does not mean that the vehicle is unsafe. Regardless, an AV developer should have a “break-the-glass” plan to document its preparation for and guide its response to incidents involving its AVs. (So too should governments.) How will it recognize and manage a crash? How will it coordinate with first responders and investigators? A developer that has such a plan—and is willing to discuss the safety-relevant portions of it—signals that it understands that deployment is about more than just the state of the technologies.

Reevaluation

Might the answers to any of these questions change during the deployment of the AV? How and why? What will trigger that change?

This article ends where it began: Automated driving is complex, dynamic, and difficult to predict. For these reasons, many of an AV developer’s answers to the questions identified here could evolve over the course of a deployment. On one hand, the realties of testing or deployment may demand a more cautious approach or frustrate the fulfilment of some promises. On the other hand, developers still hope to remove their safety drivers and to expand their operational design domain at some point. How—and on what basis—will they decide when to take these steps? Their answers can help to shift discussions from vague and speculative predictions to meaningful and credible roadmaps.

Up to now, the way forward for roadways-based, commercial automated mobility remained somewhat of a mystery. Surely, we would not see AVs in the hand of individual owners anytime soon – too expensive. “Robotaxi” fleets commanded by the likes of Uber and Lyft seemed the most plausible option. There was, at least in appearance, a business case and that most industry players seemed to be putting their efforts towards an automated version of common passenger cars.

Over the course of 2019, the landscape slowly but steadily changed: public authorities started to worry more about safety and the prospects of seeing fleets of “robotaxis” beyond the roads of Arizona, Nevada or California seemed remote. This is how automated shuttles found their way to the front of the race towards a viable business model and a large-scale commercial deployment.

Many now mock these slow-moving “bread loafs,” ridiculing their low speed and unenviable looks. However, some of these comments appear slightly disingenuous. The point of the shuttles is not “to persuade people to abandon traditional cars with steering wheels and the freedom to ride solo.” I don’t see any of these shuttles driving me back home to Montreal from Ann Arbor (a 600 miles/1000km straight line). But I see them strolling around campuses or across airport terminals. The kind of places where I don’t quite care about the good looks of whatever is carrying me around, and also the kind of place where I wouldn’t take my car to anyway. There might be much to say about how certain electric vehicles marketed directly to the end-user failed because of their unappealing design, but I don’t plan to buy a shuttle anytime soon.

Looks aside, these automated “turtles” have a major upside that the “hare” of, say, Tesla (looking at you, Model 3!) may not dispose of. Something which happens to be at the top of the agenda these days: safety. While notoriously hard to define in the automated mobility context (what does safety actually imply? When would an AV be safe?) removing speed from the equation immediately takes us into a safer territory; public authorities become less concerned, and more collaborative, agreeing to fund early deployment projects. Conversely, scooters irked a lot of municipal governments because they go too fast (among other things). As a result, there was little public appetite for scooters and operators were forced to withdraw, losing their license or failing to become commercially viable.

As a result, it is the safe vein that various industry players decided to tap. Our turtles are indeed slow, with a top speed of 25mph, usually staying in the range of 15 to 20 mph. This is no surprise: that is the speed after which braking means moving forward several dozen if not hundreds of feet. Within that lower bracket, however, a vehicle can stop in a distance of about two cars (not counting reaction time) and avoid transforming a collision into a fatality. Hence, it goes without saying that such shuttles are only suitable for local transportation. But why phrase that as an only? Local transportation is equally important. Such shuttles are also suitable for pedestrian environments. Outside of the US, pedestrians have their place on the road – and many, many roads, across the globe, are mostly pedestrian. Finally, they can also be usefully deployed in certain closed environments, notably airports. In many places, however, deployment of such shuttles on roadways might require some additional work – creation of lanes or changes to existing lanes – in order to accommodate their presence. Yet the same observation can also be made for “robotaxis,” however, and the adaptations required there may be much more substantial. The limited applications of automated shuttles may be what, ultimately, makes them less appealing than our Tesla Model 3 and its promises of freedom.

Overall, turtle shuttles appear closer to a marginal development from widely used rail-based automated driving systems, rather than a paradigm shift. That might precisely be what makes them a good gateway towards more automation in our mobility systems; there is wisdom in believing that we will have a better grasp of the challenges of automated mobility by actually deploying and using such systems, but it is not written anywhere that we need to break things to do so.

The California DMV recently released several 2019 reports from companies piloting self-driving vehicles in California. Under state law, all companies actively testing autonomous vehicles on California public roads must disclose the number of miles driven and how often human drivers were required to retake control from the autonomous vehicle. Retaking control is known as “disengagement.” The DMV defines disengagements as:

“[D]eactivation of the autonomous mode when a failure of the autonomous technology is detected or when the safe operation of the vehicle requires that the autonomous vehicle test driver disengage the autonomous mode and take immediate manual control of the vehicle.”

Because of the proprietary nature of autonomous vehicle testing, data is not often publicly released;  this is one of the few areas where progress data is made publicly available. The 60 companies actively testing in California cumulatively traveled 2.88 million miles in 2019. The table below reports the various figures for some of the major testers in California.

Company Vehicles Active in CA Miles Driven in 2019 Engagements Engagements per 1,000 miles Average Miles Between Engagements
Waymo 153 1.45 Million 110 0.076 13,219
GM Cruise 233 831,040 68 0.082 12,221
Apple 66 7,544 64 8.48 118
Lyft 20 42,930 1,667 38.83 26
Aurora ? 13,429 142 10.57 95
Nuro 33 68,762 34 0.494 2,024
Pony.ai 22 174,845 27 0.154 6,493
Baidu 4 108,300 6 0.055 18,181
Tesla 0 0 0 0 0

What these numbers make clear is that there are several contenders who have made significant progress in the autonomous vehicle space, and there are some contenders which are not yet so competitive. Companies like Waymo, GM Cruise, and Baidu (which also tests extensively in China) have made incredible progress in decreasing the frequency at which a driver must engage with an automated vehicle. Others, like Apple, Lyft, and Aurora, while making progress, are nowhere near as sophisticated in avoiding engagements yet. Noticeably Tesla, the manufacturer frequently in the news for its “Autopilot” feature, does not test on public roads in California. The company says it conducts tests via simulation, on private test tracks, public roads around the world, and “shadow-tests” by collecting anonymized data from its customers during normal driving operations.

What these numbers seem to illustrate is that the autonomous vehicle industry is not all on par, as many often believe. It is often said that Henry Ford did not conceive the idea of an automobile; he perfected it. Similarly, companies like Waymo or GM may be the first to perfect autonomous vehicles, and gain an incredible market advantage once they do so. They are striving to be the Ford’s in this space, while others look like they’re still manufacturing carriages. However, despite these impressive numbers from a select few, the companies themselves think these metrics “do[] not provide relevant insights” (per Waymo) and that the idea that they give any “meaningful insight . . . is a myth” (per GM Cruise).

Why are the head and shoulder leaders on these metrics saying that they provide very little indication of progress on the technology? Disengagement reports may not be the best way for these companies to build trust and credibility in their products. They are only transparent in that they provide some data with no detail or context.

I was having a conversation about these disengagement numbers with a colleague* this week, and the topic of driver distraction arose. In the CA tests, the driver is constantly alert. Once these vehicles are in use for the general public, a notification to engage may not be effective if the driver is distracted. One reason these numbers do not provide particularly useful information is that for the metrics to be useful, at least two things must be true:

  • If the vehicle does not indicate it needs to disengage, no technical errors have been made; and
  • The driver is paying attention and can quickly engage when necessary.

In California testing, the drivers behind the vehicle are always alert and ready to take over. They may take over when the vehicle indicates they must, because of a malfunction or poor conditions. The driver can also engage when the vehicle has done something incorrectly, yet does not indicate that the driver needs to take over. This could include veering into a lane or failing to recognize a pedestrian.

One of the allures of autonomous vehicles is that a driver may not need to be 100 percent engaged for the vehicle to function correctly. However, current technology has not yet achieved this  result, as reiterated this past week by the National Transportation Safety Board (NTSB). The NTSB is an independent federal agency, which lacks enforcement power, but makes recommendations which are considered thorough and are taken seriously by policymakers.

The NTSB put forward many findings on Tuesday, February 25th regarding a Tesla crash that killed a California driver in March 2018. (A synopsis of the NTSB report and findings can be found here.) The facts of the crash involved driver of a Tesla in Autopilot mode, which struck a barrier between the highway and a left exit lane. NTSB found that the Tesla briefly lost sight of the lines marking the highway lane, and started to follow the right-most lane marker of the exit lane (because of fading on the highway lines) caused the vehicle to enter the “gore area.” This same action had apparently occurred several times in this exact vehicle, but the driver on previous trips was paying attention and was able to correct the vehicle. This time, the driver was playing a mobile game and did not correct the vehicle, causing the crash. Here was how NTSB presented three of their findings:

The Tesla’s Autopilot lane-keeping assist system steered the sport utility vehicle to the left into the neutral area of the gore, without providing an alert to the driver, due to limitations of the Tesla Autopilot vision system’s processing software to accurately maintain the appropriate lane of travel. (emphasis added)

The driver did not take corrective action when the Tesla’s Autopilot lane-keeping assist system steered the vehicle into the gore area, nor did he take evasive action to avoid the collision with the crash attenuator, most likely due to distraction by a cell phone game application. (emphasis added)

The Tesla Autopilot system did not provide an effective means of monitoring the driver’s level of engagement with the driving task.

Here we see a combined failure of both (1) and (2) presented above, combined with an inability to adequately monitor driver engagement. The vehicle took an action which it assumed to be correct, and thus did not notify the driver to take over. This combined with the driver not paying attention, failing to notice the need to disengage, and resulted in the crash. This tragic accident highlights that the AV industry still has many areas to improve before higher SAE level vehicles are ready for mass adoption. (The ADAS on the Tesla was SAE Level 2)

As I discussed last week, the federal Department of Transportation has taken a rather hands-off approach to regulation of automated vehicles, preferring to issue guidance rather than mandatory regulations. The  National Transportation Safety Board (NTSB) criticized this approach in their Tesla crash findings. The NTSB wrote that there has been “ Insufficient Federal Oversight of Partial Driving Automation Systems.”

The US Department of Transportation and the National Highway Traffic Safety Administration (NHTSA) have taken a nonregulatory approach to automated vehicle safety. NHTSA plans to address the safety of partial driving automation systems through enforcement and a surveillance program that identifies safety-related defect trends in design or performance. This strategy must address the risk of foreseeable misuse of automation and include a forward-looking risk analysis.

Because the NTSB lacks enforcement power, it cannot compel industry actors or other government agencies to take any action. It can only perform investigations and make recommendations. NTSB Chairman Robert Sumwalt had much to say regarding distracted driving, the AV industry, and the lack of government regulations in the hearing on Tuesday, February 25th.

“In this crash we saw an over-reliance on technology, we saw distraction, we saw a lack of policy prohibiting cell phone use while driving, and we saw infrastructure failures, which, when combined, led to this tragic loss,”

“Industry keeps implementing technology in such a way that people can get injured or killed . . . [I]f you own a car with partial automation, you do not own a self-driving car. Don’t pretend that you do.”

“This kind of points out two things to me. These semi-autonomous vehicles can lead drivers to be complacent, highly complacent, about their systems. And it also points out that smartphones manipulating them can be so addictive that people aren’t going to put them down,”

Chairman Sumwalt is right to be frustrated. The DOT and NHTSA have not regulated the AV industry, or ADAS as they should. Tragic accidents like this can be avoided through a variety of solutions; better monitors of driver engagement than torque-sensing steering wheels, lock-out functions for cell-phones when driving, stricter advertising and warning regulation by companies offering ADAS. Progress is being made in the AV industry, and automated vehicles are getting smarter and safer every day. But incidents like this that combine a failure of technology, regulation, and consumer use, do not instill public confidence in this incredible technology that will be beneficial to society. It only highlights how much farther we still have to go.

*I would like to thank Fiona Mulroe for the inspiration to take this approach to the disengagement report

In January of this year, the United States Department of Transportation and the National Science & Technology Council released Automated Vehicles 4.0: Ensuring American Leadership in Automated Vehicles Technologies (“AV 4.0”). The report is intended to act as a set of unifying principles across 38 federal departments, agencies, commissions, and Executive offices. It offers guidance and an overarching vision to state and local government agencies, as well as technical experts and industry participants. AV 4.0 builds on AV 3.0, which was released in 2018, and AV 2.0, which was released in 2017.

Consistent across the three iterations of Automated Vehicle reports produced under the Trump administration has been the wholly restrained voluntary/guidance approach, without mandates or true regulation laid down. This “light touch” approach recognizes that much of the regulatory action is taking place at the state level, and voluntarily by industry and other AV stakeholders. The lack of federal enforceability removes the hope of any near-term consensus or consistency to the coordination of states and industry approach to automated vehicles.

As in AV 3.0, AV 4.0 begins with a discussion of guidelines and broad overarching principles that the federal government will recognize when developing AV technology. The three Principles and associated sub-areas are:

  1. Protect Users and Communities
    • (a) Prioritize Safety
    • (b) Emphasize Security and Cyber Security
    • (c) Ensure Privacy and Data Security
    • (d) Enhance Mobility and Accessibility
  2. Promote Efficient Markets
    • (a) Remain Technology Neutral
    • (b) Protect American Innovation and Creativity
    • (c) Modernize Regulations
  3. Facilitate Coordinated Efforts
    • (a) Promote Consistent Standards and Practices
    • (b) Ensure a Consistent Federal Approach
    • (c) Improve Transportation System-Level Effects

While these broader principles are new to AV 3.0, the sub-areas within each principle are a mixture of old principles from AV 3.0, combined with some new government focuses. (The new components are italicized  in the above list) These new components center mainly around the Trump administration’s desire to “Buy American, Hire American,” and bring in new focuses for security, cybersecurity, and privacy. The report does indicate that DOT will establish manufacturing, performance, and operational standards to increase safety in AV testing and integration. Still, the parameters of these standards remain unclear.

However, the majority of the report is devoted to Section III: U.S. Government Activities and Opportunities for Collaboration. The  driving purpose of the report is to provide general descriptions of the vast array of government agencies that are responsible (or will be responsible) for some level of AV-related policies or subjects. The report detail both the big and small ways that federal agencies, departments, offices, etc. will play a role in the development and commercialization of AVs. However, there is no real substantive or specific policy discussion.

The survey of government agency activity (and an appendix with links to government websites and contacts for agencies responsible for AV-related policies) are the real substance added. There is no description or plan of how interagency cooperation will occur, nor is there an overarching plan for the government’s approach to implementation. AV 4.0 is useful as a catalog for the various ways the federal government could interact with AV-policy but gives no further direction to how industry and states should regulate AVs. The DOT continues to kick the can down the road but has thankfully provided a clearinghouse of information on which agencies may be responsible for what activities, and the current efforts underway at the federal level, particularly on research and funding.

This restrained approach could be a good thing, given the rapid pace at which the AV industry is developing. However, there is a great deal more work that needs to be done at the federal level before any of the 10 core principles articulated can be realized. The Department of Transportation has requested public comment on AV 4.0.

Cars are getting smarter and safer. And yet this new breed of automobile remains inaccessible to large parts of the consumer base due to high costs. Some of these costs are a natural result of technological advancements in the automobile industry. Others however may be a product of inefficient market dynamics among car manufacturers, insurers and technology companies – which ultimately contribute to a reduced state of safety on our roads.

Automated Driver Assistance Systems (ADAS) that equip cars with services like autonomous braking systems, parking assistance, and blind spot detection are growing at an exponential rate. The global ADAS market size was estimated to be around $14.15 billion in 2016. Since then, it has witnessed a high rate of growth and is expected to reach $67 billion by 2025. Not only is this good news for ADAS developers, it can also significantly increase road safety. The Insurance Institute for Highway Safety estimates that the deployment of automatic emergency braking in most cars on the road, for instance, can prevent 28,000 crashes and 12,000 injuries by 2025.

The biggest roadblock to the easy adoption of ADAS-equipped cars remains its prohibitive cost. Lower rates of adoption not only reduce the overall safety of cars on the road, but also disproportionately affect poorer people. Unsurprisingly, a study in Maryland found that individuals at the upper end of the socioeconomic spectrum have greater access to vehicle safety features leaving those at the lower end at higher risk.

A significant contributing factor to the continued high cost of automated vehicles is the high rate of car insurance. This seems rather counter intuitive. The technological evolution of safety systems reduces the risk of car crashes and other incidents. Consequently, this was expected to cause a decline in insurance premiums. And yet, costs remain high. Insurance companies have resisted the demands for lowering the cost of premiums claiming that the data about ADAS systems and their efficacy in reducing risk is just not conclusive. Moreover, the industry claims, that even if ADAS systems can cause a reduction in the number of vehicular incidents, each incident involving an automated car costs more because of the sophisticated and often delicate hardware such as sensors and cameras installed in these cars. As the executive vice president of Hanover Insurance Group puts it, “There’s no such thing as a $300 bumper anymore. It’s closer to $1,500 in repair costs nowadays.”

There is no doubt that these are legitimate concerns. An industry whose entire business model involves pricing risk can hardly be blamed for seeking more accurate data for quantifying said risk. Unfortunately, none of the actors involved in the automated vehicle industry are particularly forthcoming with their data. At a relatively nascent stage, the AV industry is still highly competitive with large parts of operations shrouded in secrecy. Car manufacturers that operate fleets of automated vehicles and no doubt gather substantial data around crash reports are loathe to share it with insurers out of fears of giving away proprietary information and losing their competitive edge. The consequence of this lack of open exchange is that AVs continue to remain expensive and perhaps improperly priced from a risk standpoint.

There are some new attempts to work around this problem. Swiss Re, for example, is developing a global ADAS risk score that encourages car manufacturers to share data with them that they in turn would use to recommend discounts to insurers. Continental AG has similarly developed a Data Monetization Platform that seemingly allows fleet operators to sell data in a secure and transparent manner to city authorities, insurers and other interested parties. These are early days so whether these initiatives will be able to overcome the insecurities around trade secrets and proprietary data remains to be seen.

It is however clear that along with the evolution of cars and technologies the insurance industry too will need to change. As a recent Harvard Business Review article points out, automated vehicles will fundamentally alter the private car insurance market by shifting car ownership from an individual-centric model to a fleet-centric one, at least in the short to medium term. This shift itself could cost auto insurers nearly $25 billion (or 1/8th of the global market) in revenue from premiums. It is imperative therefore that the insurance industry devise new innovative approaches to price the risk associated with AVs. Hopefully they can do this without further driving up costs and while making safer technologies accessible to those that need it the most.

As I wrote about last time, the Uniform Law Commission recently passed the Uniform Automated Operation of Vehicles Act. Today, I want to focus on Sections 5, 6, and 7 of that Act, which are titled, respectively, “Vehicle Registration,” “Automated-Driving Provider,” and “Associated Automated Vehicle.” The three sections are meant to complement each other and the generally applicable rules regarding motor vehicle registration in a state. The Comments to Section 7 give a nice synopsis of the way these three sections interact:

Existing state law generally requires the registration of a motor vehicle that is operated on a public road. If an automated vehicle qualifies as such a motor vehicle, it too must be registered. The person seeking that registration—typically the vehicle owner—must comply with all conditions of registration under existing law. Section 5 of this act adds a further condition: For the owner of an automated vehicle to register the vehicle, an automated driving provider must have designated that vehicle as an associated automated vehicle. Section 6 specifies how an entity declares that it is an automated driving provider, and Section 7 specifies how that entity then designates its associated automated vehicles. These three sections work together with existing law to ensure that a properly registered automated vehicle has a legal driver when it is under automated operation. In general, only if an automated vehicle is associated with an automated driving provider may it be registered and operated on public roads.

The Act’s comments are fairly dense, but we can work through them section by section. Under current state law, the owner of a motor vehicle must generally register that vehicle with the state according to state registration rules. The Act retains that requirement for the owner of an automated vehicles, but also adds a new condition of registration. Under Section 5, an automated vehicle may be registered only if an entity has:

(1) declared itself to be an automated driving provider (ADP) (explained in Section 6) and

(2) designated that particular automated vehicles as one of its associated automated vehicles (explained in Section 7).

The vehicle owner and the ADP do not necessarily have to be the same legal person. The vehicle owner could be an individual, and the ADP could be an original equipment manufacturer (OEM) like Ford, Honda, or Tesla. The manufacturer, or some other entity like an insurer or fleet operator, would declare themselves to be the ADP to the state, and declare the automated vehicles to be one of its associated vehicles, but the individual would own and register the car. This has the effect of “compelling” vehicle manufacturers, or some other entity, to declare themselves to be the entity legally defined as the driver for any consequences that arise from the vehicles actions on public roads. The comments to Section 6 clarify:

To become an automated driving provider, an entity must make an affirmative declaration that includes specific representations. This means that, first, an entity does not become an automated driving provider against its will and, second, not every entity can become an automated driving provider. Subsection (a) identifies three basic qualifications, at least one of which a provider must satisfy, and subsection (c) identifies four key requirements, all of which the provider must satisfy.

To qualify as an ADP, an entity must have either participated substantially in the development of the system, submitted safety self-assessments with NHTSA, or be a registered manufacturer with NHTSA. The purpose of these sections is to require registration with the state, ensuring that every automated vehicle on a state’s roads has an entity associated with it, against whom the state can credibly enforce relevant provisions of the state vehicle code.

Manufacturers are not required to register as an ADP. But they will be incentivized to declare themselves as ADP’s for the simple reason that if they do not, their customers will be unable to register or use their vehicles in a state that has adopted the Act. If customers in one state were unable to register Ford vehicles but could register Honda vehicles, then everyone in that state would buy Honda automated vehicles and nobody would buy Ford. Under Section 7, once an ADP has designated an associated automated vehicle, the association remains until the ADP is not recognized by the state agency, ceases to exist under principles of corporate law, or affirmatively withdraws the designation.

This approach is a great way to allow manufacturers of automated vehicles to select the states in which they wish to be responsible for their vehicles. If they register as an ADP in Arizona, but not New Mexico, then their customers will be able to register and drive their vehicles on the public roads in Arizona, but not New Mexico. This can allow manufacturers to choose where they accept liability for the automated features of their vehicles.

However, this could cause problems. Assuming uniform adoption of the Act (which is unlikely), if manufacturers are selective with the states where they register as ADP’s then there could be adjacent states where a manufacturer is an ADP in state X, and not in state Y. If customers in state X drive their automated vehicle across the border into state Y, there could be legal questions if the manufacturer is liable for accidents that occur in state Y, especially if they specifically chose not to register there. This could lead to geofencing at state borders, requirements that shift control back to the human driver as they cross state borders, or a whole host of other potential solutions. These solutions could also cause problems. What if a driver is asleep as the vehicle crosses into a state where that manufacturer has not registered? What if the driver overrides and continues allowing the vehicle to drive? Has liability shifted from the manufacturer to the owner given the owner’s conscious choice?

Questions of tort liability, jurisdiction over manufacturers, and technological work-arounds could abound if OEM’s are selective with their registration as ADPs. But they should be allowed to select where they want to sell their automated vehicles if they will be required to legally be identified as the responsible entity. Sections 5, 6, and 7 of the Automated Vehicle Act will likely cause much debate in states that consider adopting the Act.

The Uniform Law Commission (“ULC”) is a non-governmental body composed of state-selected lawyers who oversee the preparation of “Uniform Laws” to be proposed to the states for adoption. The group’s most well-known body of law will be familiar to any lawyer or law student who paid attention in first-year contracts: the Uniform Commercial Code (UCC). Not all projects of the ULC are as successful as the UCC. In fact, many are never adopted by any state.

The ULC appointed a Drafting Committee on Highly Automated Vehicles in 2017.  The Committee recently completed an Automated Vehicles Act, titled “The Uniform Automated Operation of Vehicles Act,” which is a “uniform law covering the deployment of automated driving systems (SAE levels 3 through 5).” The Act is intended to cover a vast array of issues likely to be faced by states in the coming decades as autonomous vehicles become more ubiquitous. The ULC description of the Automated Vehicles Act states:

The Uniform Automated Operation of Vehicles Act regulates important aspects of the operation of automated vehicles.  This act covers the deployment of automated vehicles on roads held open to the public by reconciling automated driving with a typical state motor vehicle code.  Many of the act’s sections – including definitions, driver licensing, vehicle registration, equipment, and rules of the road – correspond to, refer to, and can be incorporated into existing sections of a typical vehicle code.  This act also introduces the concept of automated driving providers (ADPs) as a legal entity that must declare itself to the state and designate the automated vehicles for which it will act as the legal driver when the vehicle is in automated operation.  The ADP might be an automated driving system developer, a vehicle manufacturer, a fleet operator, an insurer, or another kind of market participant that has yet to emerge.  Only an automated vehicle that is associated with an ADP may be registered.  In this way, the Automated Operation of Vehicles Act uses the motor vehicle registration framework that already exists in states – and that applies to both conventional and automated vehicles – to incentivize self-identification by ADPs.  By harnessing an existing framework, the act also seeks to respect and empower state motor vehicle agencies.

The final version of the act can be downloaded here.

This Act is a step in the right direction. It does much of the leg-work for state legislatures to exempt autonomous vehicles from a variety of state laws by providing language which can be easily inserted into various state vehicle codes. States can choose to enact certain parts of the Uniform Act, picking and choosing the sections or phrases they want and discarding the rest. This is beneficial because it will likely mean more states will enact some form of AV exemption. However, it also means there could be substantial variation between states that adopt some but not all of the Act. The passage of a Uniform Act by the ULC does not ensure there will be uniform adoption.

The act is not very long, only 28 pages including all the comments and legislative notes. There are many sections that deserve a more extensive dive, but I want to begin with a subsection that relates to a topic I’ve written about before: Platooning. The Act does not include a provision that would legalize platooning, but it does contain a single provision that addresses state laws regarding minimum following distance: Section 9 (h). Section 9 covers “Rules of the Road.” Subsection (h) states:

A provision of [this state’s vehicle code] imposing a minimum following distance other than a reasonable and prudent distance does not apply to the automated operation of an automated vehicle.

The comment to the section clarifies subsection h:

[T]his section provides that a numerical minimal following-distance requirement does not apply to the automated operation of automated vehicles. These numerical minimums may be unnecessarily large for automated vehicles that react faster than human drivers. However, the common “reasonable and prudent” following-distance requirement continues to apply. This bracketed subsection (h) differs in scope from following-distance legislation enacted in some states to facilitate the platooning of vehicles, particularly commercial trucks, that use advanced technologies but may not necessarily qualify as automated vehicles.

As I’ve written about before, platooning vehicles that follow at incredibly close distances could be considered “reasonable and prudent” given the connected nature and quick response times of the technology. If the Uniform Act were adopted in some states, it could present the opportunity to argue that there is, or should be, a reasonable car standard applied to autonomous vehicles. The act also solves the problems of states with 300-500-foot following distance requirements for trucks.

The passage of the Act is exciting for many reasons. It shows that the legal world is taking autonomous vehicles seriously, and is taking fundamental steps to create a legal framework within which these vehicles can operate. It also provides a baseline for states to modify their existing laws to allow autonomous vehicles to be exempted from many requirements that need not apply to autonomous vehicles. For example, there is no need for a steering wheel or gas pedals in an AV. There may be a need for a large touchscreen like in the various Tesla models, which would be distracting in traditional vehicles. The Act will hopefully spark discussions about the proper way to regulate autonomous vehicles at the state level, and may even spark debate over the merits of varied state or uniform federal regulation.

Several major OEMs have recently announced scaling back of their shared or automated mobility ventures. Ford and Volkswagen are giving up investments in “robotaxis” – the CEO of their software partner, Argo, was quoted saying he “hates the word” anyway – and similar services operated by German automakers are withdrawing from various markets or shutting down altogether, after overextending themselves during the last 18 months.

Two separate trends seem to contribute to that movement. The first one, car ownership is still growing worldwide, albeit modestly – roughly 1% per year over the last ten years in Germany, for example – while sales of new cars is slumping. It is important to differentiate these two: while new car sales affect the revenues of OEMs, and may indicate changes in consumption patterns, car ownership rates indicate people’s attitude vis-à-vis car ownership better. In that sense, we see a continued attachment to personal car ownership, a cultural phenomenon that is much more difficult to displace or even disrupt than what some may have thought previously. Hence, the dreaded “peak car” that will relegate the iconic 20th century consumer good to museums may not materialize for a while.

The second trend has to do with an observation made time and again: OEMs are not naturally good at running mobility services: their business is making cars. As one bank analyst put it, no one expects Airbus or Boeing to run an airline. Why should it be any different with car OEMs? Thinking about the prospects of automation, it became commonplace for large industrial players to partner with specialized software developers to develop the automated driving system. That may result in a great product, but it does not give create a market and a business plan when it comes to the AVs themselves. As it turned out, the main business plan, which was to use these cars as part of large car-sharing services or sell them to existing mobility operators, ran into a some roadblocks: OEMs found themselves competing with already existing mobility operators in a difficult market; and putting an AV safely on the road is a much more daunting task than once thought. As 2019 comes to a close, we have yet to see an actual commercial “robotaxi” deployment outside of test runs.

This second trend puts a large question mark on the short and medium term financial viability of investments in “robotaxis” and automated mobility operations, generally. OEMs and their partners, looking for ways to put all those vehicle automation efforts to profitable use, look at other markets, such as heavy, non-passenger road and industrial vehicles. Nevertheless, no one seems poised to completely exit the automated passenger mobility market; they all keep a foot in the door, continuing their tests and “gathering more data,” in order to allegedly understand the mobility needs of road users. Beyond these noble intentions, however, there is an exit plan: if all else fails, they can monetize their data sets to data hungry software developers.

In the end, this comes back to a point frequently addressed on this blog, that of safety. Technological advances in automation (broadly speaking) are bringing increased safety to existing cars, and they will continue to do so. We might have become overly fixated by the golden goose of the “Level 5” robotaxi (or even Level 3), which may or may not come in the next ten years, neglecting the low-hanging fruit. While laugh at our ancestors dreaming about flying cars for the year 2000, our future selves scoff at us for chasing robotaxis by 2020.

Over the last few years, emerging mobility technologies from CAVs to e-scooters have become the targets of malicious hackers. CAVs, for example, are complicated machines with many different components, which opens up many avenues for attack. Hackers can reprogram key fobs and keyless ignition systems. Fleet management software used worldwide can be used to kill vehicle engines. CAV systems can be confused with things as simple a sticker on a stop sign. Even the diagnostic systems within a vehicle, which are required to be accessible, can be weaponized against a vehicle by way of a $10 piece of tech.

For mobility-as-a-service (“MaaS”) companies, the security of their networks and user accounts is also at threat. In 2015 a number of Uber accounts were found for sale on the “dark web,” and this year a similar market for Lime scooter accounts popped up. Hacking is not even required in some cases. Car2Go paused service in Chicago after 100 vehicles were stolen by people exploiting the company’s app (the company is now ending service in the city, though they say it’s for business reasons).

The wireless systems used for vehicle connectivity are also a target. On faction in the current battle over radio spectrum is pushing cellular technology, especially 5G tech as the future of vehicle-to-vehicle communication. While 5G is more secure than older wireless networks, it is not widespread in the U.S., leaving vulnerabilities. As some companies push for “over-the-air” updates, where vehicle software is wirelessly updated, unsecure wireless networks could lead to serious vehicle safety issues.

So what can be done to deal with these cybersecurity threats? For a start, there are standard-setting discussions underway, and there have been proposals for the government to step up cybersecurity regulation for vehicles. A California bill on the security of the “internet-of-things” could also influence vehicle security. Auto suppliers are putting cybersecurity into their development process. Government researchers, like those Argonne National Labs outside Chicago, are looking for vulnerabilities up and down the supply chain, including threats involving public car chargers. Given the ever-changing nature of cybersecurity threats, the real solution is “all of the above.” Laws and regulations can spark efforts, but they’ll likely never be able to keep up with evolving threats, meaning companies and researchers will always have to be watchful.

P.S. – Here is a good example of how cybersecurity threats are always changing. In 2018, security researchers were able to hack into a smartphone’s microphone and use it to steal user’s passwords, using the acoustic signature of the password. In other words, they could figure out your password by listening to you type it in.

On November 19, the NTSB held a public board meeting on the 2018 Uber accident in Tempe, Arizona, involving an “automated” (actually level 3) Uber-operated Volvo SUV. One woman, Elaine Herzberg, a pedestrian, died in the accident. In the wake of the report, it is now a good time to come back to level 3 cars and the question of “safety drivers.”

Given that the purpose of the meeting was to put the blame on someone, media outlets were quick to pick up a culprit for their headlines: the “safety driver” who kept looking at her phone? The sensors who detected all kinds of stuff but never a person? Uber, who deactivated the OEM’s emergency braking? Or maybe, Uber’s “safety culture”? A whole industry’s?

The Board actually blames all of them, steering clear of singling out one event or actor. It is probably the safest and most reasonable course of action for a regulator, and it has relevant implications for how law enforcement will handle accidents involving AVs in the future. But because we are humans, we may stick more strongly with the human part of the story, that of the safety driver.

She was allegedly looking at her phone, “watching TV” as one article put it; following the latest episode of The Voice. The Board determined that she looked at the road one second before the impact. That is short, but under more normal circumstances, enough to smash the brakes. Maybe her foot was far from the pedal; maybe she just did not react because she was not in an “aware” state of mind (“automation complacency,” the report calls it). In any case, it was her job to look on the road, and she was violating Uber’s policy by using her phone while working as a safety driver.

At the time of the accident, the Tempe police released footage from the dash cam, a few seconds up to the impact, showing a poorly-lit street. The relevance of this footage was then disputed in an Ars Technica article which aims to demonstrate how actually well lit the street is, and how just the front lights of the car should have made the victim visible on time. Yet, I think it is too easy to put the blame on the safety driver. She was not doing her job, but what kind of job was it? Humans drive reasonably well, but that’s when we’re actually driving, not sitting in the driver seat with nothing else to do but to wait for something to jump out of the roadside. Even if she had been paying attention, injury was reasonably foreseeable. And even if she would have been driving in broad daylight, there remains a more fundamental problem besides safety driver distraction.

The [NTSB] also found that Uber’s autonomous vehicles were not properly programmed to react to pedestrians crossing the street outside of designated crosswalksone article writes. I find that finding somewhat more appalling than that of a safety driver being distracted. Call that human bias; still I do not expect machines to be perfect. But what this tells us is that stricter monitoring of cellphone usage of safety drivers will not cut it either, if the sensors keep failing. The sensors need to be able to handle this kind of situation. A car whose sensors cannot recognize a slowly crossing pedestrian (anywhere, even in the middle of the highway) does not have its place on a 45-mph road, period.

If there is one thing this accident has shown, it is that “safety drivers” add little to the safety of AVs. It’s a coin flip: the reactivity and skill of the driver makes up for the sensor failure; in other cases, a distracted, “complacent” driver (for any reason, phone or other) does not make up for the sensor failure. It is safe to say that the overall effect on safety is at best neutral. And even worse: it may provide a false sense of safety to the operator, as it apparently did here. This, in turn, prompts us to think about level 3 altogether.

While Uber has stated that it has “significantly improved its safety culture” since the accident, the question of the overall safety of these level 3 cars remains. And beyond everything Uber can do, one may wonder if such accidents are not bound to repeat themselves should level 3 cars see mass commercial deployments. Humans are not reliable “safety drivers.” And in a scenario that involves such drivers, it takes much less than the deadly laundry list of failures we had here to have such an accident happen. Being complacent may also mean that your foot is not close to the pedals, or that your hands are not “hovering above the steering wheel” as they should (apparently) be. That half second extra it takes to smash the brakes or grip the wheel is time enough to transform serious injury into death.

The paramount error here was to integrate a human, a person Uber should have known would be distracted or less responsive than an average driver, as a final safety for sensor failure. Not a long time ago, many industry players were concerned about early standardization. Now that some companies are out there, going fast and literally breaking people (not even things, mind you!), time has come to seriously discuss safety and testing standards, at the US federal and, why not, international level.

A University of Michigan Law School Problem Solving Initiative class on AV standardization will take place during the Winter semester of 2020, with deliverables in April. Stay tuned!

Developments in technology have led to an increased reliance on artificial intelligence and autonomy in various vehicles such as cars, planes, helicopters and trains. The latest vehicles to implement autonomous technology into their operations are shipping vessels. Autonomous ships will transform the industry and current regulations are being reassessed to determine the best way to include this futuristic way of shipping.

he shipping industry is regulated on a global level and it remains one of the most heavily regulated industries today. International shipping is principally regulated by the International Maritime Organization, a United Nations agency responsible for the safety of life at sea and the protection of the marine environment. The International Maritime Organization (IMO) developed a comprehensive framework of global maritime safety regulations that was adopted from international conventions. In order to be proactive, IMO initiated a regulatory scoping exercise on Maritime Autonomous Surface Ships (MASS). The scoping exercise is led by IMO’s Maritime Safety Committee and is expected to be completed by 2020. The goal of the exercise is to determine how autonomous ships may be implemented into regulations and will touch on issues such as safety, security, liability, the marine environment and the human element.

In order to assess the scope of differing levels of autonomous ships, IMO defined four degrees of autonomy. The lowest degree of autonomy involves automated processes that can control the ship at times. Seafarers will remain in charge of operating and controlling the ship when the automated system is not activated. The second degree is a remotely controlled ship with seafarers still on board. The ship will be controlled from another location but the seafarers on board will be able to take control if necessary. The next degree is a remotely controlled ship without any seafarers on board. Lastly, the highest degree of autonomy is a fully autonomous, unmanned ship that is equipped with the ability to make decisions and take action by itself.

Several companies have already begun implementing autonomous capabilities into their ships and the technology is rapidly developing. While the scoping exercise is underway, the Maritime Safety Committee approved interim guidelines for trials to be completed on existing and emerging autonomous ships. The trials should be generic and goal-based and take a precautionary approach to ensure the operations are safe, secure, and environmentally sound. In 2018, Rolls-Royce conducted its first test of an autonomous ferry named Falco. To demonstrate two degrees of autonomy, the ferry was fully autonomous on its outward voyage and then switched to a remotely controlled operation on its return to port. The controller was in a command center 30 miles away and he successfully took over operations of the ship and guided it to the dock.

Autonomous ships are expected to improve safety, reduce operating costs, increase efficiency and minimize the effects of shipping on the environment. An increased reliance on autonomy will reduce the chance for human error thereby improving safety. Human error accounts for 75-96% of marine accidents and accounted for $1.6 billion in losses between 2011 and 2016. Operational costs are also expected to decrease as there will be little to no crew on board. Crew costs can constitute up to 42% of a ship’s operating costs. If there is no crew then accommodations such as living quarters, air conditioning and cooking facilities can be eliminated. Further, a ship free from crew accommodations and seafarers will make voyages more efficient because the ship will have an alternate design and an increased carrying capacity. Lastly, autonomous ships may prove to be better for the environment than current vessels. The ships are expected to operate with alternative fuel sources, zero-emissions technologies and no ballast. 

As we have seen in other transportation industries, regulation for autonomous vehicles falls far behind the technological innovation. By taking a proactive approach in the case of autonomous shipping, IMO may be ready to create regulations that better reflect the future of shipping within the next decade. 

Last time I wrote about platooning, and the potential economic savings that could benefit the commercial trucking sector if heavy duty trucks were to implement the technology. This week, I’m writing about one of the current barriers to implementing platooning both as a commercial method, and in the larger scheme of highway driving.

One of the most readily identifiable barriers to the widespread implementation of truck platooning is the ‘Following Too Close’ (“FTC”) laws enforced by almost every state. There is currently a patchwork of state legislation which prevents vehicles from following too closely behind another vehicle. Violating these laws is negligence per se.

For those who don’t quite remember 1L torts, negligence per se essentially means “if you violate this statute, that proves an element of negligence.” Therefore, if one vehicle is following too closely behind another vehicle in violation of an FTC statute, that satisfies the breach element of negligence and is likely enough to be fined for negligent driving.

These laws are typically meant to prevent vehicles from following dangerously close or tailgating other vehicles. The state laws that regulate this conduct can be divided into roughly four categories. Some states prescribe the distance or time a driver must remain behind the vehicle in front of them; others impose a more subjective standard. The subjective standards are far more common than the objective standards.

Subjective Categories

  • Reasonable and Prudent” requires enough space between vehicles for a safe stop in case of an emergency. This FTC rule is the most common for cars and seems to be a mere codification of common-law rules of ordinary care.
  • “Sufficient space to enter and occupy without danger” requires trucks and vehicles with trailers to leave enough space that another vehicle may “enter and occupy such space without danger.” This is the most common rule for trucks.

Objective Categories

  • Distance-Based: Some states prescribe the distance at which a vehicle may follow another vehicle; others identify a proportionate interval based on distance and speed. These are the most common rules for heavy trucks and frequently set the minimum distance between 300 and 500 feet.
  • Time: Timing is the least common FTC, but the two jurisdictions that impose this rule require drivers to travel “at least two seconds behind the vehicle being followed.”

It is easy to see how, given the close distance at which vehicles need to follow to benefit from platooning, any of these laws would on their face prohibit platooning within their borders. However, several states have already enacted legislation which exempts the trailing truck in a platoon from their “Following Too Close” laws. As of April 2019, 15 states had enacted legislation to that effect. Additional states have passed legislation to allow platoon testing or pilot programs within their states.

However, despite some states enacting this legislation, a non-uniform regulatory scheme does not provide  the level of certainty that will incentivize investment in platooning technology. Uncertain state regulation can disincentivize interstate carriers from investing in platooning, and could lead to a system where platooning trucks only operate within single state boundaries.

Although the exemptions are a step in the right direction, non-uniformity will likely result in an overall lower platooning usage rate, limiting the wide-spread fuel efficiency and safety benefits that are derived when platooning is implemented on a large, interstate scale. Without uniform legislation that allows platooning to be operated consistently across all the states, the need for different systems will hinder the technology’s development, and the rate at which trucking companies begin to adopt it.

However, even if not all states pass legislation exempting platooning vehicles from their FTC laws, there could be a way around the subjective elements. The most common subjective law, “Reasonable and Prudent” requires only enough space that the vehicles can safely stop in case of an emergency. When considering a human driver this distance is likely dozens of feet, given the speed at which cars travel on the interstate. However, recall from last week that platooning vehicles are synchronized in their acceleration, deceleration, and braking.

If the vehicles travel in tandem, and brake at the same time and speed, any distance of greater than several feet would be considered “reasonable and prudent.” Perhaps what needs to be developed is a “reasonable platooning vehicle” standard, rather than a “reasonable driver” standard, when it comes to autonomous vehicle technology. Then again, considering the ever-looming potential for technological failure, it could be argued that following that close behind another heavy vehicle is never reasonable and prudent, once again requiring an exemption rather than an interpretive legal argument for a new “reasonableness” standard.

Either way, to ensure certainty for businesses, more states should exempt platooning vehicles from their “Following Too Close” laws. Otherwise, the technology may never achieve a scale that makes it worth the early investment.

One of the most exciting and economically advantageous aspects of autonomous vehicle technology is the ability for cars and heavy trucks to “platoon.” Platooning is a driver-assist technology that allows vehicles to travel in tandem, maintaining a close, constant distance. Imagine trucks are racers in a bicycle or foot race. By drafting closely behind one another, the vehicles reduce their energy (fuel) consumption.

I personally find that large-scale platooning should be the ultimate goal of autonomous vehicle technology; the potential time and fuel savings would be enormous if the highways were filled with vehicles drafting behind one another. Imagine a highway system without rubberneckers, the guy on the highway that floors it, and then slams on the breaks during rush hour, or the “Phantom Traffic Jam.” Imagine an organized “train” of cars and trucks instead, following at a close, but technologically safe distance (between 25 and 75-feet) and at a uniform speed.

This future is more likely to begin on a smaller scale, and in the commercial shipping sector, rather than in the consumer vehicle market. The work has already started with some platooning pilot programs involving heavy trucks.

These programs employ short-range communications technology and advanced driver assistance systems in their testing. The technology creates a seamless interface supporting synchronized actions; however, drivers are still needed to steer and monitor the system. When done with heavy commercial trucks — tractor-trailers, 18-wheelers, or semi-trucks (depending on what area of the country you live in) — the trucks are “coupled” through vehicle-to-vehicle (V2V) communication. The V2V technology allows the vehicles to synchronize acceleration, deceleration, and braking to increase efficiency and safety.

The economic incentives for platooning in the freight industry derive from the potential fuel savings, which come from reductions to aerodynamic drag. While both vehicles in a pair of platooning trucks save fuel, the rear vehicle typically saves significantly more. Tests conducted by the National Renewable Energy Laboratory demonstrated average fuel savings up to 6.4 percent for a pair of platooning trucks: a lower amount (up to 5.3 percent) for the lead truck and a higher amount (up to 9.7 percent) for the trailing truck. These numbers varied based on the size of the gap between the two trucks, and the driving speed. The ability to decrease fuel consumption in heavy freight vehicles represents an enormous area to reduce the cost of shipping.

Fuel costs account for roughly one-third of the trucking industries’ cost per mile; a typical heavy-duty freight vehicle incurs between $70,000 and $125,000 in fuel costs each year. Vehicles that reduce their fuel consumption by 6.4 percent would save $4,500 to $8,000 per year. These savings are potentially enormous when extrapolated across the more than 2-million tractor-trailers on the road. The ability to decrease shipping and transportation costs should be a substantial incentive for large shipping companies like Fed Ex, UPS, and Amazon. 

While getting the significant players in the transportation industry is crucial, an estimated 90% of the trucking companies in the U.S. are made up of fleets with six trucks or less, and 97% have fewer than 20. Converting existing truck cabs with the necessary technology could pose a substantial hardship for these small businesses. However, it is projected that owner-operators would recoup their investment in 10-months, and fleet operators would recoup theirs in 18-months. This relatively short period could incentivize even small-scale operators to invest in the technology.

Platooning technology could also help offset the recent spike in the average cost of truck operations. Most of these costs came from increases in driver wages and benefits, likely due to a shortage of long-haul truck drivers. The shortage of drivers is only expected to grow; the combination of long hours, inconsistent schedules, long stretches of solitude, and low pay have increased the turnover rate and disincentivized new drivers from entering the labor market. While the technology is not yet poised to run without drivers, a single truck driver would one day lead a platoon train of autonomous trucks, decreasing the need for drivers in every cab.

My vision of a highway filled with platooning vehicles may not be feasible yet, but with proper investment by businesses, platooning technology could become viable, and cost-effective, within a few years.

Earlier this week, Raphaël wrote about the role for no-fault insurance in an age of automated vehicles. The post raised several important questions about the future of the auto insurance industry as technology advances:

Who do we want to protect? Passengers, for sure. But drivers? There is no driver! Or rather, there are many drivers. To some extent, at least under a layman’s understanding of the term “driver,” all the actors along the supply chain are driving the AV. Or, to be more precise, it is difficult to pinpoint a single driver: the “operator”? The software designer? And that is already assuming that there is a single entity who designed the software or operates what may be a fleet of AVs. And there may be others, as the AV industry continues to evolve

While Raphaël’s questions will be extremely important for the industry going forward, I want to pose a slightly different question here: Should drivers of Level 4 or Level 5 automated vehicles be required to purchase insurance at all?

Today, every US state except New Hampshire and Virginia require vehicle owners to purchase auto insurance. In a world where human drivers are in control of the vehicle, this makes sense. While flawed product designs or other manufacturer errors may contribute to car accidents today, many if not most accidents also involve some element of operator error.

In a world of fully autonomous vehicles however, this will no longer be the case. The human sitting in what is today the driver’s seat will not be in control of the vehicle in any meaningful way. Every movement such a vehicle makes will have been designed and programmed by the manufacturer or some company along its supply chain.

Given the extent of manufacturer control, would it be reasonable to have accident payouts determined by the law of products liability? Or perhaps, as Michigan Law professor Kyle Logue argued in a recent article, liability should be apportioned through a scheme of strict enterprise liability, in which automakers would be “unconditionally responsible for the economic losses resulting from any crashes of their vehicles.”

Even with the heightened manufacturer liability that may accompany highly autonomous vehicles, some room may be left for owner error, and thus for traditional insurance. Owners may be held responsible for routine maintenance of their vehicles, for ensuring software is upgraded in a timely fashion, or for actions taken if the vehicle is ever taken out of autonomous mode. But even if some minimal level of owner responsibility leads lawmakers to reject Professor Logue’s theory of enterprise liability in favor of maintenance of insurance requirements for owner/operators, insurance’s role will surely be different than it is today.

Which brings me back to a version of Raphaël’s questions that I started with. Who is the true operator of an AV, and what are they responsible for? Answering these background questions will go a long way towards determining the future of auto insurance in the era of automated vehicles.

In a recent article published on Reuters Regulatory Intelligence, a DC-area lawyer said the following regarding the potential of implementing no-fault insurance “to” automated vehicles:

“Drivers have an inherent incentive to drive safely, so as not to be injured or killed on the roadways. That inherent incentive is what mitigates the “moral hazard” of a no-fault system. But in a no-fault model for autonomous vehicles, the incentives toward safety would be degraded given that manufacturers do not suffer the physical consequences of unsafe operation, as do drivers.”

Intuitively, this seems right. Yet, I thought: is there more to it? What does a world with an AV no-fault insurance scheme would look like?

This might be puzzling at first: what does one mean with no-fault AV insurance? In a more standard setting, a no-fault insurance system means that one gets the benefits of their own insurance without regard to the actual “fault” (such as negligence) and that civil suits on the basis of one’s fault are banned or severely restricted. No-fault systems are straightforward and predictable, although potentially less “just,” to the extent that a negligent driver may get away with nothing more than a deductible to pay or eventually a higher premium.

There was, and there still is, a good policy ground behind no-fault systems around car accidents: avoiding the social cost of civil litigation, and shifting the financial cost of such litigation towards the insurers, between whom things are more often than not settled out of court. Those we want to protect with no-fault insurance schemes are drivers (and passengers), and that is a majority of the population.

Now let’s consider AVs. Who do we want to protect? Passengers, for sure. But drivers? There is no driver! Or rather, there are many drivers. To some extent, at least under a layman’s understanding of the term “driver,” all the actors along the supply chain are driving the AV. Or, to be more precise, it is difficult to pinpoint a single driver: the “operator”? The software designer? And that is already assuming that there is a single entity who designed the software or operates what may be a fleet of AVs. And there may be others, as the AV industry continues to evolve; we can already see that various paths are taken by industry players, some acting for a form of vertical integration, others relying on a variety of suppliers, in a less streamlined way.

Do these all these industry players deserve extra protection? They are all corporate entities after all, and as the lawyer mentions in the above article, none of them are subject to physical injury in case of an accident. While expensive litigation can drive corporations to the ground, the case for shifting costs to insurers, when it comes to AV drivers, appears less clear. What is clear, though, is that human victims of an accident involving an AV ought to be as protected as if the accident did not involve an AV, and maybe even more.

The final answer will come from lawmakers. Moreover, one should not forget that no-fault insurance is mandatory only in a minority of US states, despite being prevalent in the rest of the world. Yet I believe there might be a case here to adopt a legal scheme which would both guarantee a litigation-free recourse to human accident victims, potentially in the form of an industry-funded guarantee fund, while giving the opportunity to the various players along the supply chain to fight it out, in court if need be, on the basis of the reality of their involvement in the cause of the accident; they are all sophisticated players after all and all share in the benefits of the risk they create. The stories of human victims, though, are what may “kill” the industry, if not enough care is taken to ensure a high level of legal protection.

October 2019 Mobility Grab Bag

Every month brings new developments in mobility, so let’s take a minute to breakdown a few recent developments that touch on issues we’ve previously discussed in the blog:

New AV Deployments

This month saw a test deployment of Level 4 vehicles in London, which even allowed members of the public to be passengers (with a safety driver). Meanwhile, in Arizona, Waymo announced it will be deploying vehicles without safety drivers, though it appears only members of their early-access test group will be riding in them for now. We’ve written a lot about Waymo, from some early problems with pedestrians and other drivers, to the regulations placed on them by Arizona’s government, to their potential ability to navigate human controlled intersections.

Georgia Supreme Court Requires a Warrant for Vehicle Data

This Monday, the Georgia Supreme Court, in the case of Mobley v. State, ruled that recovering data from a vehicle without a warrant “implicates the Fourth Amendment, regardless of any reasonable expectations of privacy.” The court found that an investigator entering the vehicle to download data from the vehicle’s airbag control unit constituted “physical intrusion of a personal motor vehicle,” an action which “generally is a search for purposes of the Fourth Amendment under the traditional common law trespass standard.” Given the amount of data that is collected currently by vehicles and the ever-increasing amount of data that CAVs can and will collect, rulings like this are very important in dictating how and when law enforcement can obtain vehicle data. We’ve previously written about CAVs and the 4th Amendment, as well as other privacy implications of CAVs, both in regards to government access to data and the use of CAV data by private parties.  

Personal Cargo Bots Could Bring Even More Traffic to Your Sidewalk

In May, as part of a series on drones, I wrote about a number of test programs deploying small delivery bots for last-mile deliveries via the sidewalk. A recent Washington Post article highlights another potential contender for sidewalk space – personal cargo bots. Called “gita” the bot can travel at up to 6 mph as it uses it’s onboard cameras to track and follow its’ owner, via the owner’s gait. The bot’s developers see it as helping enhance mobility, as it would allow people to go shopping on foot without being concerned about carrying their goods home. For city-dwellers that may improve grocery trips, if they can shell out the $3,000+ price tag!

Even More Aerial Drones to Bring Goods to Your Door

Last month, as part two the drone series, I looked at aerial delivery drones. In that piece I mentioned that Google-owned Wing would be making drone deliveries in Virginia, and Wing recently announced a partnership with Walgreens that will be part of that test. Yesterday Wired pointed out that UPS has made a similar deal with CVS – though it remains to be seen if the drones will have to deliver the infamously long CVS receipts as well. As Wired pointed out, drugstores, since they carry goods that could lead to an emergency when a home runs out of them (like medication and diapers), speedy air delivery could fill a useful niche. So next time you’re home with a cold, you may be able to order decongestant to be flown to your bedside, or at least to the yard outside your bedroom window.

P.S. – While not related to any past writings, this article  is pretty interesting – Purdue scientists took inspiration from the small hairs on the legs of spiders to invent a new sensor that can ignore minor forces acting on a vehicle while detecting major forces, making it easier for CAVs and drones to focus computing power on important things in their environment without getting distracted.

While AVs have a lot of technological leaps to make before widespread deployment, developers and governments alike also need to also consider the human factors involved, including good old fashioned human fear. Earlier this year, a AAA study showed that almost three out of four (71%) Americans are afraid to ride in an AV. This is a 10% rise in apprehension from earlier studies, a trend that could be connected to the publicity around the 2018 Uber crash in Tempe, Ariz., where a test vehicle struck and killed a pedestrian. This lack of trust in AVs alone should be concerning to developers, but in some places that lack of trust has turned into outright enmity.

Test deployments, like the one undertaken by Waymo in Arizona, have become the targets of anger from drivers and pedestrians, including an incident where man pointed a gun at a passing Waymo test vehicle, in full view of the AV’s safety driver. In that case, the man with the weapon (who was arrested) claimed he hated the vehicles, specifically citing the Uber crash as a reason for his anger. Waymo test vehicles have been also been pelted with rocks, had their tires slashed, and motorists have even tried to run them off the road. The incidents have led to caution on the part of Waymo, who has trained their drivers on how to respond to harassment (including how to spot vehicles that are following them, as witnessed by a group of Arizona Republic reporters last December). Arizona is not the only place where this has happened – in California, during a 3 month period of 2018, 2 of the 6 accidents involving AVs were caused by other drivers intentionally colliding with the AV.

So where is this anger coming from? For some in Arizona, it was from feeling that their community was being used as a laboratory, with them as guinea pigs, by AV developers. Ironically, that line of thought has been cited by a number of people who currently oppose the deployment of test AVs in and around Silicon Valley. It’s rather telling that the employees of many of the companies pushing for AV testing don’t want it to occur in their own towns (some going as far as to threaten to “storm city hall” if testing came to Palo Alto…). Other objections may stem from people seeing AVs as a proxy for all automation, and the potential loss of jobs that entails.

So what can be done to make people trust AVs, or at least accept them enough to not run them off the road? On the jobs front, in June a group of Senators introduced a bill to have the Labor Department track jobs being displaced by automation. Responding to the changes brought on by automation is a center point of Democratic Presidential Candidate Andrew Yang’s campaign, and the issue has been raised by other candidates as well. The potential of automation to take away jobs is a long-standing issue made more visible by AVs on the road, and one that won’t be solved by AV proponents alone. What AV supporters have done and can continue to do is attempt to educate the public on now only potential befits of AV deployment (which PAVE, an industry coalition has done), but also better explain just how AV technology works. At least part of the AV fear stems from not understanding how the tech actually operates, and transparency in that vein could go a long way. Future test projects also need to be sure to get input from communities before they start testing, to ease the feeling of AVs being imposed upon an unwilling neighborhood. A recent debate over AV testing in Pittsburgh, where the city obtained funds for community outreach only after approving testing, leading to push back from community members, is a good example of how a proper pre-testing order-of-operations is vital.

For now, there is clearly a lot of room for public engagement and education. Developers should take advantage of this period where AVs are in the public eye without being widely deployed to build trust and understanding, so that once the vehicles start appearing everywhere they are met with open arms, or at least tolerated, rather than ran off the road. After all, while AVs themselves may not feel road rage, it’s already clear they can be victims of it.

P.S. – If you’re interested in learning more about negative reactions to robots, a good starting point is this NY Times article from January 2018.

I recently wrote about a renewed federal push to regulate automated vehicles. I’ve previously highlighted a range of state regulatory schemes, including California’s relatively strict set of regulations. Meanwhile, the advent of truly automated vehicles, which seemed imminent when Waymo announced its driverless shuttle service in Phoenix, now may be farther away than we expected. Waymo’s shuttle’s still have human safety drivers, and the technology has not advanced as quickly as expected to handle all the vagaries of the road.

But as Congress and the states struggle to get a regulatory handle on this new technology, a recent Tesla update raises an important question. Is the regulatory state agile enough to adapt when the automated vehicle market evolves in unexpected ways?

Last week, Tesla unveiled “Smart Summon,” a feature in some vehicles that allows the user to summon the car to their location. With a range of 200 feet, Smart Summon is primarily designed for use in parking lots. At least one video shows its successful use in a Costco parking lot, avoiding pedestrians and other vehicles to meet its owner at the storefront. However, the feature is not limited to use in parking lots, and videos have emerged of users trying out Smart Summon on public roads, and even running in front of their vehicle to try and make the car chase them. Despite the potential dangers this technology presents, no one has yet been injured or hit by a driverless Tesla being summoned.

Despite the seriousness with which California takes automated vehicle regulation, state authorities have determined that Teslas equipped with Smart Summon are not autonomous, and thus do not need to meet California’s AV standards. Based on regulatory definitions, this is probably the correct. A state DMV spokesperson said the state defines AVs as vehicles able to drive without active physical control or monitoring by a human. Smart Summon requires a user to be attentive to their smartphone. Furthermore, its inability to operate more than 200 feet from a human controller means that it would not satisfy SAE autonomous driving level four or five.

Despite not being a true AV though, it’s clear that Smart Summon presents many of the same dangers as one. It operates in unpredictable parking lots, filled with pedestrians and vehicles and shopping carts all moving around each other in unpredictable ways. It is the sort of environment that can get dicey for a human driver, with our experience and understanding of the subtle signals humans give off to make an otherwise unexpected move a little bit more predictable. And despite a small-print company warning that Smart Summon requires “active driver supervision,” the amount of supervision anyone can truly give a moving vehicle from 200 feet away is certainly questionable.

And yet, these vehicle are not AVs. Instead, they seem to fall within an increasingly muddled gray area of transportation that is something short of fully automated, but requires much less than full and active driver attention. In California’s current regulatory environment, this technology fits neatly into a gap.

A year ago, many people assumed we were rapidly approaching the rise of Level 4 automated vehicles that could operate smoothly on city streets. Regulations developed at the time are written with that sort of technology in mind. Even one year out, legislators were not thinking of how to assure the safety of something like Smart Summon.

So how should something like Smart Summon be regulated? What will autonomous—or semi-autonomous—technology look like a year from now, and how can government agencies prepare for it? Given the unpredictable nature of an early stage technology, regulators will continue struggling to answer these questions.

The European Union recently adopted new rules to help consumers repair household appliances like refrigerators and televisions. The rules require manufacturers to provide spare parts for years after sale – the number of years depending on the device. The “Ecodesign Directive” is intended to help protect the environment by extending the life of consumer appliances. The regulation also applies to servers, requiring firmware updates for 7 years post-production. These regulations are part of a larger battle over consumers’ right to repair their belongings, including vehicles. Vehicles are already part of the right to repair discussion, and the deployment of technically complicated CAVs will ramp up that conversation, as some manufacturers seek to limit the ability of individuals to repair their vehicles.

One current battle over the right to repair is taking place in California. In September of last year, the California Farm Bureau, the agricultural lobbying group that represents farmers, gave up the right to purchase repair parts for farm equipment without going through a dealer. Rather than allowing farmers to buy parts from whomever they’d like, California farmers have to turn to equipment dealers, who previously were unwilling to even allow farmer’s access to repair manuals for vehicles they already owned. A big part of the dispute stems from companies like John Deere placing digital locks on their equipment that prevent “unauthorized” repairs – i.e. repairs done by anyone other than a John Deere employee. The company even made farmers sign license agreements forbidding nearly all repairs or modifications, and shielding John Deere from liability for any losses farmers may suffer from software failures. Some farmers resorted to using Ukrainian sourced firmware to update their vehicle’s software, rather than pay to hire a John Deere technician. The California case is especially ironic, as the state has solid right to repair laws for other consumer goods, requiring companies to offer repairs for electronics for 7 years after production (though companies like Apple have been fighting against the state passing even more open right to repair laws).

In 2018, supporters of the right to repair were boosted by a copyright decision from the Librarian of Congress, which granted an exception to existing copyright law to allow owners and repair professionals to hack into a device to repair it. The exception is limited, however, and doesn’t include things like video game consoles, though its’ language did include “motorized land vehicles.”

So how could battles over the right to repair influence the deployment of CAVs? First off, given the amount of complicated equipment and software that goes into CAVs, regulations like those recently adopted in the EU could help extend the lifespan of a vehicle. Cars last a long time, with the average American vehicle being 11.8 years old. Right to repair laws could require manufactures to supply the parts and software updates needed to keep CAVs on the road. New legislation could protect consumer access to the data within their vehicle, so they don’t have to rely on proprietary manufacturer systems to know what’s going on inside their vehicle. A 2011 study of auto repair shops showed a 24% savings for consumers who used a third-party repair shop over a dealership, so independent access to data and spare parts is vital to keeping consumer maintenance costs down. People are very used to taking their cars to independent repair shops or even fixing them at home, and many consumers are likely to want to keep their ability to do so as CAVs spread into service.

P.S. – Two updates to my drone post from last week:

Update 1 – University of Michigan (Go Blue!) researchers have demonstrated a drone that can be used to place shingles on a roof, using an interesting system of static cameras surrounding the work-site, rather than on-board cameras, though it remains to be seen how many people want a nail gun equipped drone flying over their head…

Update 2 – UPS has been granted approval to fly an unlimited number of delivery drones beyond line-of-sight, though they still can’t fly over urban areas. They have been testing the drones by delivery medical supplies on a North Carolina hospital campus.

A write-up of the afternoon sessions is now available here!

March 15, 2019 – 10:00 AM – 5:30 PM

Room 1225, Jeffries Hall, University of Michigan Law School 

In the case of automated driving, how and to whom should the rules of the road apply? This deep-dive conference brings together experts from government, industry, civil society, and academia to answer these questions through focused and robust discussion.

To ensure that discussions are accessible to all participants, the day will begin with an introduction to the legal and technical aspects of automated driving. It will then continue with a more general discussion of what it means to follow the law. After a lunch keynote by Rep. Debbie Dingell, expert panels will consider how traffic law should apply to automated driving and the legal person (if any) who should be responsible for traffic law violations. The day will conclude with audience discussion and a reception for all attendees.

(Re)Writing the Rules of the Road is presented by the University of Michigan Law School’s Law and Mobility Program, and co-sponsored by the University of South Carolina School of Law.

Schedule of Events

Morning Sessions 

  • 10:00 am – 10:45 am

Connected and Automated Vehicles – A Technical and Legal Primer

Prof. Bryant Walker Smith

Professor Bryant Walker Smith will provide a technical and legal introduction to automated driving and connected driving with an emphasis on the key concepts, terms, and laws that will be foundational to the afternoon sessions. This session is intended for all participants, including those with complementary expertise and those who are new to automated driving. Questions are welcome. 

  • 10:45 am – 11:15 am
Drivers Licenses for Robots? State DMV Approaches to CAV Regulation

Bernard Soriano, Deputy Director for the Califorina DMV and James Fackler, Assistant Administrator for the Customer Services Administration in the Michigan Secretary of State’s Office, discuss their respective state’s approaches to regulating connected and autonomous vehicles.

  • 11:15 am – 12:00 pm
Just What Is the Law? How Does Legal Theory Apply to Automated Vehicles and Other Autonomous Technologies?

Prof. Scott Hershovitz    

Human drivers regularly violate the rules of the road. What does this say about the meaning of law? Professor Scott Hershovitz introduces legal theory and relates it to automated driving and autonomy more generally.                  

Keynote & Lunch

  • 12:00 pm – 12:30 pm
Lunch

Free for all registered attendees!

  • 12:30 pm-1:30 pm

Keynote – Rep. Debbie Dingell

Rep. Dingell shares her insights from both national and local perspectives.  

Afternoon Sessions

(Chatham House Rule)

  • 1:30 pm – 3:00 pm
Crossing the Double Yellow Line: Should Automated Vehicles Always Follow the Rules of the Road as Written?

Should automated vehicles be designed to strictly follow the rules of the road? How should these vehicles reconcile conflicts between those rules? Are there meaningful differences among exceeding the posted speed limit to keep up with the flow of traffic, crossing a double yellow line to give more room to a bicyclist, and driving through a stop sign at the direction of a police officer? If flexibility and discretion are appropriate, how can they be achieved in law?

A panel of experts will each briefly present their views on these questions, followed by open discussion with other speakers and questions from the audience.

Featured Speakers:

Justice David F. Viviano, Michigan Supreme Court

Emily Frascaroli, Counsel, Ford Motor Company

Jessica Uguccioni, Lead Lawyer, Automated Vehicles Review, Law Commission of England and Wales

  • 3:15 pm – 4:45 pm
Who Gets the Ticket? Who or What is the Legal Driver, and How Should Law Be Enforced Against Them?

Who or what should decide whether an automated vehicle should violate a traffic law? And who or what should be responsible for that violation? Are there meaningful differences among laws about driving behavior, laws about vehicle maintenance, and laws and post-crash responsibilities? How should these laws be enforced? What are the respective roles for local, state, and national authorities?

A panel of experts will each briefly present their views on these questions, followed by open discussion with other speakers and questions from the audience.

Featured Speakers:

Thomas J. Buiteweg, Partner, Hudson Cook, LLP

Kelsey Brunette Fiedler, Ideation Analyst in Mobility Domain

Karlyn D. Stanley, Senior Policy Analyst, RAND Corporation

Daniel Hinkle, State Affairs Counsel, American Association for Justice

  • 4:45 pm – 5:30 pm 
 Summary and General Discussion                                     

Participants and attendees close out the day by taking part in wide discussion of all of the day’s panels.

Last week I covered the various companies who are seeking to use aerial drones to deliver goods to your door. Today, in the third part to my series on delivery (you’ll find Part 1 here, and an even earlier post on delivery, from December of 2018, here), I’m going to look at recent proposals to use automated vehicles to deliver consumer goods.

As an introduction, I’m going to include a paragraph from that December 2018 post as an introduction to some of the ways automated vehicles are being used to make deliveries :

The potential for CAVs as delivery vehicles is already being tested by companies like Domino’s and Kroger, among others. Earlier this year Toyota announced delivery partnerships with Amazon and Pizza Hut, and Waymo’s CEO recently highlighted it as an area of opportunity.  This week the New York Times profiled Nuro, the start-up working with Kroger to test robotic delivery cars in Scottsdale, Ariz. Nuro’s vehicles are designed in-house, and look like “toasters-on-wheels,” and are currently followed everywhere they go by human safety drivers in conventionally driven “shadow car.” When the vehicle stops for a delivery, customers enter a PIN code into a small touch pad to open the compartment containing their order. The current charge for same-day delivery using the system is around $6. Ford has also flagged the delivery market as an area they’d like to explore, citing projections that by 2026 the last-mile delivery market for CAVs will hit $130 billion.

Don’t Forget to Tip Your (Robotic) Delivery Driver – Dec. 21, 2018

Since that post, Domino’s has announced a partnership with Nuro as well, with plans to test in Houston at some point this year. Walmart has also jumped in on the action – partnering with another AV developer, Gatik. For now Walmart’s test is limited to a 2-mile route between two of their stores in the company’s hometown of Bentonville, Arkansas. Why the interest? In part because of the potential cost savings – a recent Ford estimate calculates AVs could reduce the cost per mile for deliveries from $2.50 to $1. No doubt the combination of lower costs and ever-greater demand for delivery is a powerful motivator, pushing companies to explore not only AVs, but also drones and delivery bots, as discussed in Parts I and II of this series.

Beyond last-mile deliveries, there is a great deal of interest in automating semi-trucks and other large delivery vehicles. One company, TuSimple, is working with both the U.S. Postal Service (USPS) and UPS to move packages between cities. Interestingly, in UPS’ case, the company only announced the partnership after TuSimple had already been delivering goods for months – which seems to indicate the program is not just a grab for positive PR. The USPS’ test was more limited, running for two-weeks and five round trips. All of the trips included a safety driver and an engineer, and both tests were carried out in the Southwest. Meanwhile, in Sweden, a completely driverless electric truck was deployed in May, a global first. Given a nation-wide shortage of truck drivers (a recent estimate puts the U.S. deficit at roughly 60,000 drivers), automated trucks present a solution that doesn’t overly disrupt a truck-heavy commercial delivery system.

But what would the wide-spread adoption of AVs as part of the delivery ecosystem mean? We can already see that the demand for faster and faster delivery is taking its toll. Recently, the NY Times and Buzzfeed News both published articles detailing the human cost of Amazon’s push for same or next-day delivery. Under-trained drivers pushed to the limit have killed people in seemingly avoidable accidents that don’t often happen with more highly-trained delivery drivers (like those used by the USPS, UPS, and FedEx). Amazon has avoided liability by using a number of third-party companies as contractors, making those companies, and not Amazon, responsible for accidents. AVs would certainly be safer for the public, as they wouldn’t fall prey to the pressures of human drivers, though that does nothing to alleviate the pressures on the human delivery people, who would still be needed to move goods from the vehicle to a door. At the same time, Amazon may continue to escape liability, if the AVs remain owned by third parties. There is also the greater question of the environmental impact of the growing number of delivery vehicles on the road (not to mention the waste created by packing materials and shipping boxes). I’ll leave a greater discussion about those issues to future posts and other forums, but those questions, among so many others (privacy, cybersecurity, and traffic management among them) are important to consider as automated delivery vehicles of all kinds begin to fill our streets and skies.

P.S. – In a follow up to last week’s blog, the USPS has stated to investigate the use of aerial drones, and is now seeking information from drone operators and developers.

Back in January, I wrote about the auto industry’s growing sense that a set of nationwide regulatory standards was needed to govern automated vehicles (AVs). To date, twenty-nine states and Washington, DC have enacted AV-related legislation. A handful more have adopted Executive Orders or developed some other form of AV regulation. As the number of states with varying regulatory regimes continues to rise, the industry and some experts have grown concerned that the need to comply with a patchwork of disparate laws could hinder development of the industry.

Despite these concerns, and bipartisan support, the federal AV START Act died in the Senate at the close of 2018. After passing the House, a group of Senate Democrats became concerned that the bill focused too much on encouraging AV adoption at the expense of meaningful safety regulation. After the bill went down at the end of the year, the industry significantly reduced its lobbying efforts. This led some observers to conclude that the effort to pass AV START would not be renewed any time soon.

Never ones to let a good acronym go to waste, several members of Congress have begun work to revive the American Vision for Safer Transportation Through Advancement of Revolutionary Technologies (AV START) Act. Over the summer, a bipartisan group of lawmakers in both chambers held a series of meetings to discuss a new deal. Their hope is that, with Democrats now in control of the House, the safety concerns that stalled the bill in the upper chamber last winter will be assuaged earlier in the process.

Congress’ efforts, spearheaded by Senator Gary Peters (D-MI) appear to be making at least some headway. Both the House Committee on Energy and Commerce, and the Senate Committee on Commerce, Science and Transportation, sent letters to a variety of stakeholders requesting comments on a potential bill.

The legislature appears to be moving forward deliberately however, and to date no hearings on the subject have been scheduled in either the House or Senate. As Congress once again builds an effort to pass comprehensive AV legislation, this blog will be following and providing updates.

I previously blogged on automated emergency braking (AEB) standardization taking place at the World Forum for Harmonization of Vehicle Regulations (also known as WP.29), a UN working group tasked with managing a few international conventions on the topic, including the 1958 Agreement on wheeled vehicles standards.

It turns out the World Forum recently published the result of a joint effort undertaken by the EU, US, China, and Japan regarding AV safety. Titled Revised Framework document on automated/autonomous vehicles, its purpose is to “provide guidance” regarding “key principles” of AV safety, in addition to setting the agenda for the various subcommittees of the Forum.

One may first wonder what China and the US are doing there, as they are not members to the 1958 Agreement. It turns out that participation in the World Forum is open to everyone (at the UN), regardless of membership in the Agreement. China and the US are thus given the opportunity to influence the adoption of one standard over the other through participation in the Forum and its sub-working groups, without being bound if the outcome is not to their liking in the end. Peachy!

International lawyers know that every word counts, and every word can be assumed to have been negotiated down to the comma, or so it is safe to assume. Using that kind of close textual analysis, what stands out in this otherwise terse UN prose? First, the only sentence couched in mandatory terms. Setting out the drafters’ “safety vision,” it goes as follows: AVs “shall not cause any non-tolerable risk, meaning . . . shall not cause any traffic accidents resulting in injury or death that are reasonably foreseeable and preventable.”

This sets the bar very high in terms of AV behavioral standard, markedly higher than for human drivers. We cause plenty of accidents which would be “reasonably foreseeable and preventable.” A large part of accidents are probably the result of human error, distraction, or recklessness, all things “foreseeable” and “preventable.” Nevertheless, we are allowed to drive and are insurable (except in the most egregious cases…) Whether this is a good standard for AVs can be discussed, but what is certain is that it reflects the general idea that we as humans hold machines to a much higher “standard of behavior” than other humans; we forgive other humans for their mistakes, but machines ought to be perfect – or almost so.

In second position: AVs “should ensure compliance with road traffic regulations.” This is striking by its simplicity, and I suppose that the whole discussion on how the law and its enforcement are actually rather flexible (such as the kind of discussion this very journal hosted last year in Ann Arbor) has not reached Geneva yet. As it can be seen in the report on this conference, one cannot just ask AVs to “comply” with the law; there is much more to it.

In third position: AV’s “should allow interaction with the other road users (e.g. by means of external human machine interface on operational status of the vehicle, etc.)” Hold on! Turns out this was a topic at last year’s Problem-Solving Initiative hosted by University of Michigan Law School, and we concluded that this was actually a bad idea. Why? First, people need to understand whatever “message” is sent by such an interface. Language may come in the way. Then, the word interaction suggests some form of control by the other road user. Think of a hand signal to get the right of way from an AV; living in a college town, it is not difficult to imagine how would such “responsive” AVs could wreak havoc in areas with plenty of “other road users,” on their feet or zipping around on scooters… Our conclusion was that the AV could send simple light signals to indicate its systems have “noticed” a crossing pedestrian for example, without any additional control mechanisms begin given to the pedestrian. Obviously, jaywalking in front on an AV would still result in the AV breaking… and maybe sending angry light signals or honking just like a human driver would do.

Finally: cybersecurity and system updates. Oof! Cybersecurity issues of IoT devices is an evergreen source of memes and mockery, windows to a quirky dystopian future where software updates (or lack thereof) would prevent one from turning the lights on, flushing the toilet, or getting out of the house… or where a botnet of connected wine bottles sends DDoS attacks across the web’s vast expanse. What about a software update while getting on a crowded highway from an entry ramp? In that regard, the language of those sections seems rather meek, simply quoting the need for respecting “established” cybersecurity “best practices” and ensuring system updates “in a safe and secured way…” I don’t know what cybersecurity best practices are, but looking at the constant stream of IT industry leaders caught in various cybersecurity scandals, I have some doubts. If there is one area where actual standards are badly needed, it is in consumer-facing connected objects.

All in all, is this just yet another useless piece of paper produced by an equally useless international organization? If one is looking for raw power, probably. But there is more to it: the interest of such a document is that it reflects the lowest common denominator among countries with diverging interests. The fact that they agree on something, (or maybe nothing) can be a vital piece of information. If I were an OEM or policy maker, it is certainly something I would be monitoring with due care.

Cite as: Raphael Beauregard-Lacroix, (Re)Writing the Rules of The Road: Reflections from the Journal of Law and Mobility’s 2019 Conference, 2019 J. L. & Mob. 97.

On March 15th, 2019, the Journal of Law and Mobility, part of the University of Michigan’s Law and Mobility Program, presented its inaugural conference, entitled “(Re)Writing the Rules of The Road.” The conference was focused on issues surrounding the relationship between automated vehicles (“AVs”) and the law. In the afternoon, two panels of experts from academia, government, industry, and civil society were brought together to discuss how traffic laws should apply to automated driving and the legal person (if any) who should be responsible for traffic law violations. The afternoon’s events occurred under a modified version of the Chatham House Rule, to allow the participants to speak more freely. In the interest of allowing those who did not attend to still benefit from the day’s discussion, the following document was prepared. This document is a summary of the two panels, and an effort has been made to de-identify the speaker while retaining the information conveyed. 

Panel I: Crossing the Double Yellow Line: Should Automated Vehicles Always Follow the Rules of the Road as Written?

The first panel focused on whether automated vehicles should be designed to strictly follow the rules of the road. Questions included – How should these vehicles reconcile conflicts between those rules? Are there meaningful differences between acts such as exceeding the posted speed limit to keep up with the flow of traffic, crossing a double yellow line to give more room to a bicyclist, or driving through a stop sign at the direction of a police officer? If flexibility and discretion are appropriate, how can this be reflected in law? 

Within the panel, there was an overall agreement that we need both flexibility in making the law, and flexibility in the law itself among the participants. It was agreed that rigidity, both on the side of the technology as well as on the side of norms, would not serve AVs well. The debate was focused over just how much flexibility there should be and how this flexibility can be formulated in the law.

One type of flexibility that already exists is legal standards. One participant emphasized that the law is not the monolith it may seem from the outside – following a single rule, like not crossing a double yellow line, is not the end of an individual’s interaction with the law. There are a host of different laws applying to different situations, and many of these laws are formulated as standards – for example, the standard that a person operating a vehicle drives with “due care and attention.” Such an approach to the law may change the reasoning of a judge when it would come to determining liability for an accident involving an AV. 

When we ask if AVs should always follow the law, our intuitive reaction is of course they should. Yet, some reflection may allow one to conclude that such strict programming might not be realistic. After all, human drivers routinely break the law. Moreover, most of the participants explicitly agreed that as humans, we get to choose to break the law, sometimes in a reasonable way, and we get to benefit from the discretion of law enforcement. 

That, however, does not necessarily translate to the world of AVs, where engineers make decisions about code and where enforcement can be automatized to a high degree, both ex ante and ex post. Moreover, such flexibilities in the law needs to be tailored to the specific social need; speeding is a “freedom” we enjoy with our own, personal legacy cars, and this type of law breaking does not fulfill the same social function as a driver being allowed to get on the sidewalk in order to avoid an accident. 

One participant suggested that in order to reduce frustrating interactions with AVs, and to overall foster greater safety, AVs need the flexibility not to follow the letter of the law in some situations. Looking to the specific example of the shuttles running on the University of Michigan’s North Campus – those vehicles are very strict in their compliance with the law. 1 1. Susan Carney, Mcity Driverless Shuttle launches on U-M’s North Campus, The Michigan Engineer (June 4, 2018), https://news.engin.umich.edu/2018/06/mcity-driverless-shuttle-launches-on-u-ms-north-campus/. × They travel slowly, to the extent that their behavior can annoy human drivers. When similar shuttles from the French company Navya were deployed in Las Vegas, 2 2. Paul Comfort, U.S. cities building on Las Vegas’ success with autonomous buses, Axios (Sept. 14, 2018), https://www.axios.com/us-cities-building-on-las-vegas-success-with-autonomous-buses-ce6b3d43-c5a3-4b39-a47b-2abde77eec4c.html. × there was an accident on the very first run. 3 3. Sean O’Kane, Self-driving shuttle crashed in Las Vegas because manual controls were locked away, The Verge (July 11, 2019, 5:32 PM), https://www.theverge.com/2019/7/11/20690793/self-driving-shuttle-crash-las-vegas-manual-controls-locked-away. × A car backed into the shuttle, and when a normal driver would have gotten out of the way, the shuttle did not.

One answer is that we will know it when we see it; or that solutions will emerge out of usage. However, many industry players do not favor such a risk-taking strategy. Indeed, it was argued that smaller players in the AV industry would not be able to keep up if those with deeper pockets decide to go the risky way. 

Another approach to the question is to ask what kind of goals should we be applying to AVs? A strict abidance to legal rules or mitigating harm? Maximizing safety? There are indications of some form of international consensus 4 4. UN resolution paves way for mass use of driverless cars, UN News (Oct. 10, 2018), https://news.un.org/en/story/2018/10/1022812. × (namely in the form of a UN Resolution) 5 5. UN Economic Commission for Europe, Revised draft resolution on the deployment of highly and fully automated vehicles in road traffic (July, 12, 2018), https://www.unece.org/fileadmin/DAM/trans/doc/2018/wp1/ECE-TRANS-WP.1-2018-4-Rev_2e.pdf × that the goal should not be strict abidance to the law, and that other road users may commit errors, which would then put the AV into a situation of deciding between strict legality and safety or harm. 

In Singapore, the government recently published “Technical Reference 68,” 6 6. Joint Media Release, Land Transport Authority, Enterprise Singapore, Standards Development Organization, & Singapore Standards Council, Singapore Develops Provisional National Standards to Guide Development of Fully Autonomous Vehicles (Jan. 31, 2019), https://www.lta.gov.sg/apps/news/page.aspx?c=2&id=8ea02b69-4505-45ff-8dca-7b094a7954f9. × which sets up a hierarchy of rules, such as safety, traffic flow, and with the general principle of minimizing rule breaking. This example shows that principles can act as a sense-check. That being said, the technical question of how to “code” the flexibility of a standard into AV software was not entirely answered. 

Some participants also reminded the audience that human drivers do not have to “declare their intentions” before breaking the law, while AV software developers would have to. Should they be punished for that in advance? Moreover, non-compliance with the law – such as municipal ordinances on parking – is the daily routine for certain business models such as those who rely on delivery. Yet, there is no widespread condemnation of that, and most of us enjoy having consumer goods delivered at home.

More generally, as one participant asked, if a person can reasonably decide to break the law as a driver, does that mean the developer or programmer of AV software can decide to break the law in a similar way and face liability later? Perhaps the answer is to turn the question around – change the law to better reflect the driving environment so AVs don’t have to be programmed to break it. 

Beyond flexibility, participants discussed how having multiple motor vehicle codes – in effect one per US State – makes toeing the line of the law difficult. One participant highlighted that having the software of an AV validated by one state is big enough a hurdle, and that more than a handful of such validations processes would be completely unreasonable for an AV developer. Having a single standard was identified as a positive step, while some conceded that states also serve the useful purpose of “incubating” various legal formulations and strategies, allowing in due time the federal government to “pick” the best one. 

Panel II: Who Gets the Ticket? Who or What is the Legal Driver, and How Should Law Be Enforced Against Them?

The second panel looked at who or what should decide whether an automated vehicle should violate a traffic law, and who or what should be responsible for that violation. Further questions included – Are there meaningful differences among laws about driving behavior, laws about vehicle maintenance, and laws and post-crash responsibilities? How should these laws be enforced? What are the respective roles for local, state, and national authorities?

The participants discussed several initiatives, both public and private, that aimed at defining, or helping define the notion of driver in the context of AVs. The Uniform Law Commission worked on the “ADP”, or “automated driving provider”, which would replace the human driver as the entity responsible in case of an accident. The latest report from the RAND Corporation highlighted that the ownership model of AVs will be different, as whole fleets will be owned and maintained by OEMs (“original equipment manufacturers”) or other types of businesses and that most likely these fleet operators would be the drivers. 7 7. James M. Anderson, et. al., Rethinking Insurance and Liability in the Transformative Age of Autonomous Vehicles (2018), https://www.rand.org/content/dam/rand/pubs/conf_proceedings/CF300/CF383/RAND_CF383.pdf. ×

Insurance was also identified as a matter to take into consideration in the shaping up of the notion of AV driver. As of the date of the conference, AVs are only insured outside of state-sponsored guarantee funds, which aim to cover policy holders in case of bankruptcy of the insurer. Such “non-admitted” insurance means that most insurers will simply refuse to insure AVs. Who gets to be the driver in the end may have repercussions on whether AVs become insurable or not. 

In addition, certain participants stressed the importance of having legally recognizable persons bear the responsibility – the idea that “software” may be held liable was largely rejected by the audience. There should also be only one such person, not several, if one wants to make it manageable from the perspective of the states’ motor vehicle codes. In addition, from a more purposive perspective, one would want the person liable for the “conduct” of the car to be able to effectuate required changes so to minimize the liability, through technical improvements for example. That being said, such persons will only accept to shoulder liability if costs can be reasonably estimated. It was recognized by participants that humans tend to trust other humans more than machines or software, and are more likely to “forgive” humans for their mistakes, or trust persons who, objectively speaking, should not be trusted.

Another way forward identified by participants is product liability law, whereby AVs would be understood as a consumer good like any other. The question then becomes one of apportionment of liability, which may be rather complex, as the experience of the Navya shuttle crash in Las Vegas has shown. 

Conclusion

The key takeaway from the two panels is that AV technology now stands at a crossroads, with key decisions being taken as we discuss by large industry players, national governments and industry bodies. As these decisions will have an impact down the road, all participants and panelists agreed that the “go fast and break things” approach will not lead to optimal outcomes. Specifically, one line of force that comes out from the two panels is the idea that it is humans who stand behind the technology, humans who take the key decisions, and also humans who will accept or reject commercially-deployed AVs, as passengers and road users. As humans, we live our daily lives, which for most of us include using roads under various capacities, in a densely codified environment. However, this code, unlike computer code, is in part unwritten, flexible and subject to contextualization. Moreover, we sometimes forgive each others’ mistakes. We often think of the technical challenges of AVs in terms of sensors, cameras and machine learning. Yet, the greatest technical challenge of all may be to express all the flexibility of our social and legal rules into unforgivably rigid programming language. 

By Wesley D. Hurst and Leslie J. Pujo*

Cite as: Wesley D. Hurst & Leslie Pujo, Vehicle Rental Laws: Road Blocks to Evolving Mobility Models?, 2019 J. L. & Mob. 73.

I.          Introduction

The laws and regulations governing mobility are inconsistent and antiquated and should be modernized to encourage innovation as we prepare for an autonomous car future. The National Highway Traffic Safety Administration (“NHTSA”) has concluded that Autonomous Vehicles, or Highly Automated Vehicles (“HAVs”) may “prove to be the greatest personal transportation revolution since the popularization of the personal automobile nearly a century ago.” 8 8. Federal Automated Vehicles Policy, NHTSA 5 (2016), https://www.transportation.gov/sites/dot.gov/files/docs/AV%20policy%20guidance%20PDF.pdf. × Preparation for a HAV world is underway as the mobility industry evolves and transforms itself at a remarkable pace. New mobility platforms are becoming more convenient, more automated and more data driven—all of which will facilitate the evolution to HAVs. However, that mobility revolution is hindered by an environment of older laws and regulations that are often incompatible with new models and platforms.

Although there are a number of different mobility models, this article will focus on carsharing, peer-to-peer platforms, vehicle subscription programs, and rental car businesses (yes, car rental is a mobility platform). All of these mobility models face a host of inconsistent legal, regulatory and liability issues, which create operational challenges that can stifle innovation. For example, incumbent car rental, a mobility platform that has been in place for over 100 years, is regulated by various state and local laws that address everything from driver’s license inspections to use of telematics systems. Although physical inspection of a customer’s driver’s license at the time of rental is commonplace and expected in a traditional, face-to-face transaction, complying with the driver’s license inspection for a free-floating carsharing or other remote access mobility model becomes more problematic.

Part B of this article will review current federal and state vehicle rental laws and regulations that may apply to incumbent rental car companies and other mobility models around the country, including federal laws preempting rental company vicarious liability and requiring the grounding of vehicles with open safety recalls, as well as state laws regulating GPS tracking, negligent entrustment, and toll service fees. Part C poses a series of hypotheticals to illustrate the challenges that the existing patchwork of laws creates for the mobility industry. 9 9. Note: This article focuses on existing laws applicable to short-term rentals of vehicles, rather than long-term leases (including the federal Consumer Leasing regulations, known as “Regulation M,” which are set forth in 12 C.F.R., Part 213). For a more detailed discussion of long-term vehicle leasing laws, see Thomas B. Hudson and Daniel J. Laudicina, The Consumer Leasing Act and Regulation M, in F&I Legal Desk Book (6th edition 2014). × For instance, whether a mobility operator can utilize GPS or telematics to monitor the location of a vehicle is subject to inconsistent state laws (permitted in Texas, but not California, for example). And vehicle subscription programs are currently prohibited in Indiana, but permitted in most other states. Similarly, peer-to-peer car rental programs currently are prohibited in New York, but permitted in most other states. Finally, Part D of the article will offer some suggested uniform rules for the mobility industry.

First, however, we offer the following working definitions for this article:

  • Carsharing” – a membership-based service that provides car access without ownership. Carsharing is mobility on demand, where members pay only for the time and/or distance they drive. 10 10. About the CSA, Carsharing Ass’n., https://carsharing.org/about/ (last visited May 7, 2019). ×
  • Peer-to-peer Carsharing” or “Rentals” – the sharing of privately-owned vehicles in which companies, typically for a percentage of the rental charge, broker transactions among car owners and renters by providing the organizational resources needed to make the exchange possible (i.e., online platform, customer support, driver and motor vehicle safety certification, auto insurance and technology). 11 11. Car Sharing State Laws and Legislation, Nat’l Conf. of St. Legislatures (Feb. 16, 2017), http://www.ncsl.org/research/transportation/car-sharing-state-laws-and-legislation.aspx. Since most personal auto policies do not cover commercial use of personal vehicles, if the peer-to-peer platform does not provide liability and physical damage coverage, there likely will be no coverage if the vehicle is involved in an accident during the rental period. As noted above, peer-to-peer carsharing platforms currently do not operate in New York, based, in part, on the New York Department of Insurance’s findings that a peer-to-peer platform operator’s insurance practices (including sale of group liability coverage to vehicle owners and renters) constituted unlicensed insurance producing. See RelayRides, Inc. Consent Order (N.Y. Dep’t of Fin. Serv., 2014). Although a detailed discussion of insurance-related issues is beyond the scope of this article, the Relay Rides experience in New York illustrates the need for the insurance industry and insurance laws to evolve to accommodate new mobility models. See Part B.2.d. for a discussion of legislative approaches that several states have taken to address the insurance issues implicated by the peer-to-peer model (including a 2019 New York bill). ×
  • Subscriptions” – a service that, for a recurring fee and for a limited period of time, allows a participating person exclusive use of a motor vehicle owned by an entity that controls or contracts with the subscription service. 12 12. See Ind. Code § 9-32-11-20(e) (2018). The prohibition on vehicle subscription services in Indiana originally expired on May 1, 2019, but was recently extended for another year through May 1, 2020. The Indiana definition also provides that “[Subscription] does not include leases, short term motor vehicle rentals, or services that allow short terms sharing of a motor vehicle.” A bill pending in North Carolina uses similar language to define “vehicle subscription” for purposes of determining highway use tax rates. See H.B. 537 (N.C. 2019). As further discussed in Part C below, it is not clear whether other states would take the same approach and classify a subscription model as distinct from rental or leasing instead of applying existing laws. × Typically, the subscriber is allowed to exchange the vehicle for a different type of vehicle with a certain amount of notice to the operator. This is a developing model with a number of variations, including whether the subscription includes insurance, maintenance, a mileage allowance, or other features and services.
  • Vehicle Rental” – a customer receives use of a vehicle in exchange for a fee or other consideration pursuant to a contract for a period of time less than 30 days. 13 13. See Cal. Civ. Code § 1939.01 (Deering 2019). Although for purposes of this article, we use a traditional 30-day period to define short-term rentals, we note that the time period for rentals varies by state (or even by statute for a particular state) with some defining a short-term rental for periods as long as 6 months or even one year. See, e.g., Md. Code Ann., Transportation § 18-101 (LexisNexis 2019) (defining “rent” as a period of 180 days or less). Compare 35 Ill. Comp. Stat. 155/2 (2019) (defining “rent” as a period of one year or less for purposes of the Illinois Automobile Renting Occupation and Use Tax), with 625 Ill. Comp. Stat. 27/10 (defining “rental company” as one that rents vehicles to the public for 30 days or less for purposes of the Illinois damage waiver law). ×
  • Mobility Operators” – any person or entity that provides access to a vehicle to another person whether by an in-person transaction, an app-based or online platform, or any other means and whether the entity providing the access is the owner, lessee, beneficial owner, or bailee of the vehicle or merely facilitates the transaction.

II.          Existing Laws: Lack of Uniformity and Certainty

As noted above, a patchwork of federal, state (and in some cases city or county) laws regulate short-term car rentals (in addition to generally applicable laws affecting all businesses, such as privacy and data security laws, 14 14. In addition to general privacy and data security concerns applicable to all businesses, the advent of HAVs and connected vehicles may trigger additional privacy and data security issues for mobility operators. For example, issues surrounding the control, access, and use of vehicle-generated data is still unsettled and the subject of much debate. See, e.g. Ayesha Bose, Leilani Gilpin, et al., The Vehicle Act: Safety and Security for Modern Vehicles, 53 Willamette L. Rev. 137 (2017) for additional information on this topic. × the Americans with Disabilities Act (“ADA”), employment law, and zoning laws). Car rental laws have developed over time and typically address:

  1. State and local taxes and surcharges;
  2. Licensing and operational requirements, including airport concessions and permits for picking-up and dropping-off passengers;
  3. Public policy issues, such as liability insurance and safety recalls; and
  4. Consumer protection matters, like rental agreement disclosures, restrictions on the sale of collision damage waivers, prohibitions on denying rentals based on age or credit card ownership, and restrictions on mandatory fees. 15 15. See, e.g., Final Report and Recommendations of the National Association of Attorneys General Task Force on Car Rental Industry Advertising and Practices, 56 Antitrust & Trade Regulation Report No. 1407 (March 1989) at S-3 (“NAAG Report”). The NAAG Report includes “guidelines,” which were intended for use by states in providing guidance to car rental companies on compliance with state unfair and deceptive practice laws, Id. at S-5. ×

As is often the case with regulated industries, state and local vehicle rental laws vary considerably, which can lead to uncertainty and inefficiency. For example, a multi-state operator may need to vary product offerings and pricing, customer disclosures, and agreement forms, depending upon the state in which the rental commences. 16 16. Typically, a state law will apply to a transaction if the renter accepts delivery of the vehicle in that state, regardless of where the rental company’s physical offices are located, where the vehicle is typically parked, or where the vehicle is returned. See, e.g., 24 Va. Code Ann. § 20-100-10 (2019) (“The term [rental in this State] applies regardless of where the rental agreement is written, where the rental terminates, or where the vehicle is surrendered.”). × The uncertainty and inefficiency increases dramatically when considering whether and how existing vehicle rental laws apply to new mobility platforms and services since many of the existing laws do not address or even contemplate modern technology like self-service, keyless access to vehicles, digital agreements, or telematics fleet management.

The following paragraphs provide a brief overview of some of the existing laws.

A.         Federal Law

1. Graves Amendment

The federal Graves Amendment, 17 17. 49 U.S.C.S. § 30106 (LexisNexis 2019). × passed in 2005, preempts any portion of state law that creates vicarious liability for a vehicle rental company based solely on ownership of a vehicle. Specifically:

An owner of a motor vehicle that rents or leases the vehicle to a person . . . shall not be liable . . . by reason of being the owner of the vehicle . . . for harm to persons or property that results or arises out of the use, operation, or possession of the vehicle during the period of the rental or lease, if– (1) the owner . . . is engaged in the trade or business of renting or leasing motor vehicles; and (2) there is no negligence or criminal wrongdoing on the part of the owner . . . 18 18. Before passage of the Graves Amendment, many car leasing and renting companies ceased activities in states with unlimited vicarious liability laws based solely on ownership, such as New York. See Graham v Dunkley, 852 N.Y.S.2d 169 (App. Div. 2008); see also Susan Lorde Martin, Commerce Clause Jurisprudence and the Graves Amendment: Implications for the Vicarious Liability of Car Leasing Companies, 18 U. FLA. J.L. & Pub. Pol’y 153, 162 (2007). ×

Determining whether the Graves Amendment applies to a particular case involves an analysis of both factual and legal issues. The factual issues include a determination of whether:

(A) the claim involves a “motor vehicle”;

(B) the individual or entity is the “owner” of the motor vehicle (which may be a titleholder, lessee, or bailee) or an affiliate of the owner;

(C) the individual or entity is “engaged in the trade or business of renting or leasing motor vehicles”; and

(D) the accident occurred during the rental period. 19 19. Johnke v. Espinal-Quiroz, No. 14-CV-6992, 2016 WL 454333 (N.D. Ill. 2016). ×

    The legal issues include:

(A) whether the owner is being sued in its capacity as owner (as opposed to the employer or other principal of another party); and

(B) whether there are allegations that the owner was negligent or criminal. 20 20. Id. ×

Perhaps not surprisingly, the Graves Amendment has been highly litigated, from early challenges to its constitutionality, 21 21. See, Rosado v. Daimlerchrysler Fin. Servs. Trust, 112 So. 3d 1165 (2013); Garcia v. Vanguard Car Rental USA, Inc., 540 F.3d 1242 (2008); Rodriguez v. Testa, 993 A.2d 955 (Conn. 2009); Vargas v. Enter. Leasing Co., 60 So. 3d 1037 (Fla. 2008). × to later assertions that it does not apply to a particular case because the vehicle’s owner was not “engaged in the business of renting or leasing,” 22 22. See e.g., Minto v. Zipcar New York, Inc., No. 15401/09 (N.Y. Sup. Ct., Queens County Mar. 17, 2010); Moreau v. Josaphat, et al., 975 N.Y.S.2d 851 (N.Y. Sup. Ct. 2013). × or that an accident did not occur during the “rental period.” 23 23. Currie V. Mansoor, 71 N.Y.S.3d 633 (App. Div. 2018); Chase v. Cote, 2017 Conn. Super. LEXIS 3533 (2017); Marble v. Faelle, 89 A.3d 830 (R.I. 2014). ×

Two New York cases are instructive to operators of newer mobility models. In Minto v. Zipcar New York, Inc. 24 24. See Minto v. Zipcar New York, Inc., No. 15401/09. × and Moreau and Duverson v. Josaphat, et al., 25 25. See Moreau, 975 N.Y.S.2d 851. × a New York court examined whether carsharing company Zipcar was “engaged in the trade or business of renting or leasing motor vehicles” for purposes of the Graves Amendment – despite the fact that it touted itself as an alternative to car rental.

In the 2010 Minto case (which the Moreau case closely followed), the court stated that Zipcar’s advertising, which contrasted the company to “‘traditional car rental cars’, d[id] not foreclose the possibility that it is nevertheless also in the rental car business, although not of a traditional sort.” 26 26. See Minto v. Zipcar New York, Inc., No. 15401/09 at 2. × The court then noted that the Graves Amendment did not define “trade or business of renting or leasing motor vehicles.” 27 27. Id. × As a result, it analyzed the “constituent terms” of “renting” and “leasing” to determine whether Zipcar was a rental company for purposes of the Graves Amendment 28 28. Id. See also Moreau, 975 N.Y.S.2d at 855-856. × and concluded that the key features of a “lease” or rental” were the “transfer of the right to possession and use of goods for a term in return for consideration.” 29 29. See Minto v. Zipcar New York, Inc., No. 15401/09 at 2-3. × With these definitions in mind, the court focused on the requirement that Zipcar members pay fees in exchange for the right to use Zipcar vehicles, which it found to be “little different from ‘traditional rental car’ companies, notwithstanding Zipcar’s marketing statements that contrast it with those companies” and held that Zipcar was covered by the Graves Amendment. 30 30. Id. at 3. × As further support of its conclusion, the Minto court noted that the Zipcar marketing “shows that the company competes with traditional car-rental companies and serves a similar consumer need.” 31 31. Minto v. Zipcar New York, Inc., No. 15401/09 at 4. ×

2. Safe Rental Car Act

The Raechel and Jacqueline Houck Safe Rental Car Act of 2015 (“Safe Rental Car Act”) 32 32. Raechel and Jacqueline Houck Safe Rental Car Act of 2015, S. 1173, 114th Cong. (2015) (codified as amended in scattered sections of 49 U.S.C.). × places limits on the rental, sale, or lease of “covered rental vehicles”. 33 33. 49 U.S.C.A. § 30120(i) (2017). × A “covered rental vehicle” is one that: (A) has a gross vehicle weight rating (“GVWR”) of 10,000 pounds or less; (B) is rented without a driver for an initial term of less than 4 months; and (C) is part of a motor vehicle fleet of 35 or more motor vehicles that are used for rental purposes by a rental company. 34 34. 49 U.S.C.A. § 30102(a)(1) (2017). × A “rental company” is any individual or company that “is engaged in the business of renting covered rental vehicles,” and “uses, for rental purposes, a motor vehicle fleet of 35 or more covered rental vehicles, on average, during the calendar year.” 35 35. 49 U.S.C.A. § 30102(a)(11) (2017). ×

Under the Safe Rental Car Act, after receiving notice by electronic or first class mail of a NHTSA-approved safety related recall, a rental car company may not rent, sell, or lease an affected vehicle in its possession at the time of notification, until the defect has been remedied. The rental car company must comply with the restrictions on rental/sale/lease “as soon as practicable,” but no later than 24 hours after the receipt of the official safety recall notice (or within 48 hours if the notice covers more than 5,000 vehicles in its fleet). 36 36. 49 U.S.C.A. § 30120(i)(1) and (3) (2017). The 24-hour/48-hour time requirement applies only to vehicles in the possession of the rental company when the safety recall is received, and does not require rental companies to locate and recover vehicles that are on rent at that time. × If the safety recall notice indicates that a remedy is not immediately available, but specifies interim actions that an owner may take to alter the vehicle and eliminate the safety risk, the rental company may continue to rent (but not sell or lease) the vehicle after taking the specified actions. 37 37. 49 U.S.C.A. § 30120(i)(3)(C) (2017). Once a permanent remedy becomes available, the rental company may not rent affected vehicles until those vehicles have been repaired. ×

Despite the federal recall legislation, several states have introduced bills for similar legislation with California passing a law in 2016 that extends the restrictions on rental, sale, and lease to fleets of any size, as well as to cars loaned by dealers while a customer’s own vehicle are being repaired or serviced. 38 38. Cal. Veh. Code § 11754 (Deering 2019). × Effective January 1, 2019, the California prohibitions on the rental, lease, sale, or loan of vehicles subject to safety recalls also apply to “personal vehicle sharing programs,” which are defined as legal entities qualified to do business in the State of California that are “engaged in the business of facilitating the sharing of private passenger vehicles for noncommercial use by individuals within the state.” 39 39. Cal. Veh. Code § 11752 (West 2019); Cal Ins. Code § 11580.24(b)(2) (West 2011). ×

B.         State Law

Several states, including California, 40 40. Cal. Civ. Code §§ 1939.01 – 1939.37 (West 2017). × Hawaii, 41 41. Haw. Rev. Stat. Ann. §437D (West 2019). × Illinois, 42 42. 625 Ill. Comp. Stat. 27 (West 2019); 625 Ill. Comp. Stat. 5/6-305 (West 2019). × Nevada, 43 43. Nev. Rev. Stat. Ann. §§ 482.295–482.3159 (West 2019). × and New York, 44 44. N.Y. Gen. Bus. Law § 396-z (McKinney 2019). × have comprehensive vehicle rental laws that regulate a variety of issues, including minimum age requirements; sales of damage waivers; limitations on amounts recoverable from renters, fees that a vehicle rental company may charge; recordkeeping practices; general licensing or permit requirements; 45 45. See, e.g., Conn. Gen. Stat. Ann. § 14-15 (West 2018); D.C. Code § 50-1505.03 (2019); Del. Code Ann. Tit. 21 § 6102 (West 2019); Haw. Rev. Stat. Ann. § 251-3 (West 2019); Minn. Stat. Ann. § 168.27 (West 2019); Nev. Rev. Stat. Ann. § 482.363 (West 2019); N.J. Stat. Ann. § 45:21-12 (West 2019); Okla. Stat. tit. 47, § 8-101 (2004); 31 R.I. Gen. Laws Ann. § 31-5-33 (West 2019); W. Va. Code Ann. § 17A-6D-1 (West 2019); Wis. Stat. Ann. § 344.51(1m) (West 2018). × imposition of short-term rental taxes and surcharges; airport concession and permit requirements; limitations on the use of telematics; deposit and credit card restrictions; required display of counter signs; and required disclosures on rental agreements (including specified language, font size/style, and placement on written agreements). California even requires rental companies to warn their customers that operation of a passenger vehicle can expose individuals to certain chemicals that are known to cause cancer and birth defects, and therefore the customers should avoid breathing exhaust and take other precautions. Other states regulate one or more of these issues, with most states varying the specific requirements. For example, approximately 21 states regulate the sale of damage waivers with states taking different approaches on several key issues, including the permissibility of selling partial or deductible waivers, customer disclosures, and the permissible bases for invalidation of a waiver. 46 46. The typical damage waiver statute requires vehicle rental companies to disclose the optional nature of the waiver on the front of the rental agreement form and/or signs at the rental counter. Some statutes also regulate the content of the waiver and its exclusions. See, e.g., Cal. Civ. Code § 1939.09 (Deering 2019). Hawaii, Illinois, Maryland, New York, and Wisconsin require the distribution of brochures summarizing the damages waiver and its terms, and rental companies selling damage waivers in Louisiana and Minnesota must file a copy of the rental agreement before using it. Haw. Rev. Stat. Ann. § 437D-10 (LexisNexis 2019); 625 Ill. Comp. Stat. Ann. 27/20 (LexisNexis 2019); La. Stat. Ann. § 22:1525 (2018); Md. Code Ann. Com. Law § 14-2101 (LexisNexis 2019); Minn. Stat. Ann. § 72A.125 (West 2019); N.Y. Gen. Bus. Law § 396-z(4) (Consol. 2019); and Wis. Stat. Ann. § 344.576 (West 2018). ×

In addition to the issues noted above, most states prohibit rental of a vehicle without first inspecting the renter’s driver’s license to confirm that it is “facially valid” and (1) comparing the signature on the license with the renter’s signature written at the time of rental; and/or (2) comparing the photo with renter. 47 47. See, e.g., Fla. Stat. Ann. § 322.38(1-2) (LexisNexis 2018); 625 Ill. Comp. Stat. Ann. 5/6-305(b) (LexisNexis 2019); Nev. Rev. Stat. Ann. § 483.610 (LexisNexis 2019); Md. Code Ann. Transp. § 18-103(a), (b) (LexisNexis 2019); Wash. Rev. Code Ann. § 46.20.220 (LexisNexis 2019); W. Va. Code Ann. § 17B-4-6 (LexisNexis 2019). × Moreover, case law from various states provide guidance on what may or may not constitute negligent entrustment (which is excluded from the Graves Amendment). Finally, some states have begun to recognize the emergence of new mobility models and have either amended existing laws or passed new legislation to address the new models.

The paragraphs below summarize typical state laws (and how they vary) on several of these issues, including use of telematics systems; tolls and other fees, negligent entrustment, and peer-to-peer car sharing programs.

2. Telematics Systems and Vehicle Technology

Many mobility operators equip their rental vehicle fleet with global positioning systems (GPS) or other telematics systems (collectively “Telematics Systems”) to track vehicles for a variety of purposes, including fleet management; locating and recovering vehicles that are not returned by the due-in date (or that have been reported missing); calculating information related to the use of the vehicle, such as mileage, location, and speed; and providing services to renters, such as roadside assistance, maintenance, and navigation. Connected cars and HAVs will provide even more data that mobility operators can use to manage their fleets and enhance the user’s experience. 48 48. See, e.g., Avis Budget Group Boosts Fleet of Connected Cars with 75,000 In-Vehicle Telematics Units From I.D. Systems, Avis Budget Group (Dec. 17, 2018), https://avisbudgetgroup.com/avis-budget-group-boosts-fleet-of-connected-cars-with-75000-in-vehicle-telematics-units-from-i-d-systems-2/. (last visited May 8, 2019). ×

At the same time, mobility operators that use Telematics Systems to impose fees related to vehicle use (e.g., fees for traveling outside a geographic area or excess speeding), may face customer complaints or even litigation. For example, rental companies have been subject to suit in the past when they used GPS to collect location or speed information about a vehicle while on rent and impose additional fees on customers who violated geographic limitations of the rental agreement or state speed limits. 49 49. See Turner v. American Car Rental 884 A.2d 7 (Ct. App. Ct. 2005); Proposed Judgement, People v. Acceleron Corp., (Cal. Super. Ct. 2004), https://oag.ca.gov/system/files/attachments/press_releases/04-129_settle.pdf. ×

Four states, including California, Connecticut, Montana, and New York, currently have laws that specifically regulate “rental company” use of Telematics Systems. Specifically:

CaliforniaCalifornia generally prohibits rental companies from using, accessing, or obtaining information about a renter’s use of a rental vehicle that was obtained from “electronic surveillance technology” (“a technological method or system used to observe, monitor, or collect information, including telematics, . . . GPS, wireless technology, or location-based technology”), including for the purpose of imposing fines or surcharges.  However, electronic surveillance technology may be used if:

(1) The rented vehicle is missing or has been stolen or abandoned;

(2) the vehicle is 72 hours past the due-in date (and the company notifies the renter and includes required disclosures in the rental agreement);

(3) the vehicle is subject to an AMBER Alert; or

(4)  in response to a specific request from law enforcement pursuant to a subpoena or search warrant. 50 50. See Cal. Civ. Code § 1939.23(a) (West 2019). ×

Rental companies that use electronic surveillance technology for any of the reasons identified above also must maintain certain records of each such use for one year from date of use. 51 51. Id. The records must include any information relevant to the activation of the GPS, including: (1) the rental agreement; (2) the return date; (3) the date and time the electronic surveillance technology was activated; and (4) if relevant, a record any communication with the renter or the police. The record must be made available to the renter upon request, along with any explanatory codes necessary to read the record. × Rental companies may also use telematics at the request of renters, including for roadside service, navigation assistance, or remote locking/unlocking – as long as the rental company does not use, access or obtain information related to the renter’s use of the vehicle beyond that which is necessary to render the requested service. 52 52. See Cal. Civ. Code § 1939.23(b) (West 2019).  In addition, rental companies may obtain, access, or use information from electronic surveillance technology for the sole purpose of determining the date and time of the start and end of the rental, total mileage, and fuel level. × Like most of the other provisions of the California Vehicle Rental law, customers cannot waive these requirements. 53 53. See Cal. Civ. Code § 1939.29 (West 2019). The only provisions of the California vehicle rental law that a customer may waive are those related to business rentals, rentals of 15-passenger vans, and driver’s license inspection exceptions for remote access programs. ×

ConnecticutConnecticut’s non-uniform version of UCC Article 2A, 54 54. Conn. Gen. Stat. § 42-2A-702 (2013). × (which applies to both short-term and long-term consumer and commercial leases) regulates the use of “electronic self-help,” including the use of GPS devices to track and locate leased property to repossess the goods (or render them unusable without removal, such as remotely disabling the ignition of a vehicle). Before resorting to electronic self-help, a lessor must give notice to the lessee, stating:

      • That the lessor intends to resort to electronic self-help as a remedy on or after 15 days following notice to the lessee;
      • The nature of the claimed breach which entitled the lessor to resort to electronic self-help; and
      • The name, title, address and telephone number of a person representing the lessor with whom the lessee may communicate concerning the rental agreement.

In addition, the lessee must separately agree to a term in the lease agreement that authorizes the electronic self-help. A commercial lease requires only that the authorization is included as a separate provision in the lease, which implies that a consumer lease requires the express, affirmative consent of the lessee. 55 55. Conn. Gen. Stat. § 42-2A-702(e)(2)-(3) (2013). Lessees may recover damages, including incidental and consequential damages, for wrongful use of electronic self-help (even if the lease agreement excludes their recovery). Conn. Gen. Stat. § 42a-2A-702(e)(4). In addition, a lessor may not exercise electronic self-help if doing so would result in substantial injury or harm to the public health or safety or “grave harm” to third parties not involved in the dispute – even if the lessor otherwise complies with the statute. Conn. Gen. Stat. § 42a-2A-702(e)(5). ×

Montana Montana requires a “rental vehicle entity” providing a rental vehicle equipped with a GPS or satellite navigation system to disclose in the rental agreement (or written addendum) the presence and purpose of the system. 56 56. See Mont. Code Ann. 61-12-801(1)(a) (2019). For purposes of the Montana law, a “rental vehicle entity” is a business entity that provides the following vehicle to the public under a rental agreement for a fee: light vehicles, motor-driven cycles, quadricycles, or off-highway vehicles. Mont. Code Ann. 61-12-801(2)(b)-(c) (2019). A “rental agreement” is a written agreement for the rental of a rental vehicle for a period of 90 days or less. Mont. Code Ann. 61-12-801(2)(a) (2019). × If the GPS or satellite navigation system is used only to track lost or stolen vehicles, disclosure is not required.

New York – New York prohibits a “rental vehicle company” from using information from “any” global positioning system technology to determine or impose fees, charges, or penalties on an authorized driver’s use of the rental vehicle. 57 57. N.Y. Gen. Bus. Law 396-z(13-a). New York defines a “rental vehicle company” as “any person or organization . . . in the business of providing rental vehicles to the public from locations in [New York]. NY Gen. Bus. Law 396-z(1)(c). × The limitation on use of GPS, however, does not apply to the rental company’s right to recover a vehicle that is lost, misplaced, or stolen.

More recently, vehicle infotainment systems, which may include Telematics Systems like GPS, have come under scrutiny. In a putative class action filed against Avis Budget Group in December 2018, the plaintiff asserted that:

(a) a customer’s personal information may be collected and stored automatically by a vehicle each time the customer pairs his or her personal mobile device to the vehicle infotainment system to access navigation, music streaming, voice dialing/messaging, or other services; and

(b) failure to delete the customer data after each rental violated customers’ right to privacy under the California constitution, as well as the California rental law electronic surveillance technology provisions.

As of the date of this article, the defendant had removed the case to federal court and filed a motion to compel arbitration based on the terms and conditions of the rental agreement. 58 58. See Complaint, Kramer v. Avis Budget Group, Inc., Case No. 37-2018-00067024-CU-BT-CTL (Ca. Super. Ct., San Diego County 12/31/2018). The federal case number is 3:19cv421 (S.D. Cal.). Similar claims have been filed against other companies in California and all were initially removed to federal court, however, one of the cases has been remanded to state court. ×

2. Tolls and Other Fees

Several states, including California, Nevada, and New York, limit the types and even the amounts of fees that rental companies can charge. For example, California prohibits additional driver fees, and Nevada and New York cap those fees. In other states, a fee that appears to be excessive or punitive may be unenforceable. Generally, a fee is more likely to be enforced if it is fully disclosed, and the customer can avoid paying it by either not selecting a particular product or service (such as supplemental liability insurance or an additional driver) or not engaging in a particular behavior (such as returning the car late or with an empty gas tank). 59 59. See, e.g., Blay v. Zipcar, Inc., 716 F. Supp. 2d (D. Mass. 2010); Reed v. Zipcar, Inc., 883 F. Supp. 2d 329 (D. Mass. 2012). Cf. Bayol v. Zipcar, Inc., 78 F.Supp.3d 1252 (N.D. Cal. 2015). ×

Although disgruntled customers may complain about any fee that they believe is excessive or “hidden,” over the past several years, toll program charges have been among the most disputed in the car rental industry. Indeed, several class action claims have been filed against rental companies alleging inadequate disclosure of toll payment terms, failure to disclose use of third parties, unauthorized charges to the customer’s credit card, breach of contract, and similar claims. 60 60. See Doherty and Simonson v. Hertz, No 10-359 (NLH/KMW) 2014 WL 2916494 (D.N.J. Jun. 25, 2014) (approving over $11 million settlement of class action case based on assertions that inadequate disclosure of a rental company’s toll program violated consumer protection laws and breached the rental agreement); see also Mendez v. Avis Budget Group, Inc., No. 11-6537(JLL), 2012 WL 1224708 (D. N.J. Apr. 10, 2012); Readick v. Avis Budget Group, Inc., No. 12 Civ. 3988(PGG), 2013 WL 3388225 (S.D. N.Y. Jul. 3, 2013); Sallee v. Dollar Thrifty Automotive Group, Inc., et al., 2015 WL 1281518 (N.D. Okla. Mar. 20, 2015); Maor v. Dollar Thrifty Automotive Group, Inc., 303 F.Supp.3d 1320 (S.D. Fla. 2017). × State and local attorneys general have also investigated or filed civil claims against rental companies based on similar allegations. 61 61. See infra, note 55. ×

The increase in customer complaints and litigation likely stems from innovations in both toll collection methods and rental car toll payment processing (both of which seem likely to become an integral part of the connected car/HAV ecosystem). For example, an increasing number of toll roads and bridges are all-electronic. At the same time, many rental companies have introduced optional toll service products that permit renters to use electronic toll roads and lanes during the rental, some of which are provided by third parties. Often, a renter who declines to purchase the toll service at the time of rental will be subject to higher fees if he or she incurs toll charges by driving on an all-electronic road or lane during the rental.

The typical complaint focuses on alleged lack of or inadequate disclosure of the toll payment-processing program. For example, in recent settlement agreements with the Florida Attorney General, Avis Budget Group, Inc., and Dollar Thrifty Automotive Group, Inc. both agreed to disclose that Florida has cashless tolls, along with details about the rental company’s toll service options, and how the toll service charges can be avoided (such as by paying in cash, programming a GPS to avoid toll roads, contacting local authorities for other payment options, or using a personal transponder that is accepted on the toll road). 62 62. In February 2019, Hertz settled a case with the City Attorney of San Francisco for $3.65 million. The case alleged that the Hertz toll fee program as applied to the Golden Gate Bridge (an all-electronic toll road) failed to adequately disclose the fees or to provide customers the ability to opt-out. See Julia Cheever, Hertz Reaches $3.65 Million Settlement with SF over Golden Gate Bridge Tolls, San Francisco Examiner (Feb. 19, 2019), http://www.sfexaminer.com/hertz-reaches-3-65-m-settlement-sf-golden-gate-bridge-toll-fees/. See also Office of the Att’y Gen. of Fla.v. Dollar Thrifty Automotive Group, Inc., No. 16-2018-CV-005938 (Fla. Cir. Ct Jan. 7, 2019), https://myfloridalegal.com/webfiles.nsf/WF/TDGT-B8NT5W/$file/Final+Signed+DT AG+Settlement+Agreement+1+11+19.pdf.; In re Investigative Subpoena Duces Tecum to Avis Budget Group, Inc. and Payless Car Rental System, Inc., No 2017 CA 000122 (Fla. Cir. Ct. Jul. 7, 2017), http://myfloridalegal.com/webfiles.nsf/WF/JMAR-AP6LZQ/ $file/Settlement+Agreement+Avis.pdf. ×

Finally, state legislatures are taking notice of the tolling issues with several states proposing new legislation to regulate rental company toll programs and fees. As of January 1, 2019, Illinois became the first state to directly regulate toll programs by establishing maximum daily fees for toll programs if the rental company fails to notify the customer of the option to use a transponder or other device before or at the beginning of the rental. 63 63. See 625 Ill. Comp. Stat. 5/6-305. ×

3. Negligent Entrustment.

As noted above, the federal Graves Amendment protects “rental” or “leasing” companies from vicarious liability for their customers’ accidents based solely on ownership of the vehicle; however, the rental or leasing company is still liable for its own negligence or criminal wrongdoing. As a result, one common challenge to a rental or leasing company’s assertion of the Graves Amendment as an affirmative defense is a claim that the rental or leasing company somehow negligently entrusted the vehicle to the customer.

A vehicle owner may be liable for negligent entrustment if: (1) it provides a vehicle to a person it knows, or should know, is incompetent or unfit to drive; (2) the driver is in an accident or otherwise causes injury; and (3) that injury is caused by that person’s incompetence. 64 64. See Osborn v. Hertz Corp., 205 Cal.App.3d 703, 708-709 (1989). × To be found liable for negligent entrustment in the vehicle renting or leasing context, the rental or leasing company generally must have some special knowledge concerning a characteristic or condition peculiar to the renter that renders that person’s use of the vehicle unreasonably dangerous. Plaintiffs’ counsel typically allege that negligent entrustment is at issue where the driver appears to be intoxicated at the time of the rental or has a known substance abuse problem; where a renter is known by the rental company and its agents to be a reckless driver; or  where the rental company has reason to know that the renter may cause injury to others.

On the other hand, courts around the country have found that the following circumstances did not constitute negligent entrustment:

(1) failure to research the renter’s driving record; 65 65. See Flores v. Enterprise Rent-A-Car Co., 116 Cal. Rptr. 3d 71, 78 (2010). ×

(2) failure to recognize the signs of habitual drug use (when renter was not under the influence at the time of rental); 66 66. See Weber v Budget Truck Rental, 254 P.3d 196 (Wash. Ct. App. 2011). ×

(3) renting to an individual whose license had been suspended, but who had not yet received notification of the suspension; 67 67. See Young v. U-Haul, 11 A.3d 247 (D.C. Cir. 2011). ×

(4) failure to administer a driving test or to ensure that the driver is capable of actually operating the vehicle; 68 68. See Reph v. Hubbard, No. 07-7119, 2009 WL 659910 at *3 (E.D. La. 2009). ×

(5) renting to an individual who does not speak English fluently; (6) renting to an individual with an arm splint who did not indicate that the splint would interfere with his ability to drive; 69 69. See Mendonca v. Winckler and Corpat, Inc., No 1-5007-JLV, 2014 WL 1028392 (D.S.D. 2014). ×  and

 (7) renting to a former customer who previously reported an accident in a rental car and also allegedly returned a car with illegal drugs left behind. 70 70. See Maisonette v. Gromiler, No. FSTCV176031477S, 2018 WL 3203887 (Conn. Super. Ct. 2018). ×

4. State Laws Addressing New Mobility Platforms

More recently, some states have begun to recognize the emergence of new mobility models and have amended existing laws or passed new laws to address some of the issues. For example:

  • In 2011, California amended its insurance code to include a “personal vehicle sharing” statute, which regulates insurance aspects of “personal vehicle sharing programs” that facilitate sharing of private passenger vehicles (i.e., vehicles that are insured under personal automobile policies insuring a single individual or individuals residing in the same household) for non-commercial purposes, as long as the annual revenue received by the vehicle’s owners from the personal vehicle sharing does not exceed the annual expenses of owning and operating the vehicle (including the costs associated with personal vehicle sharing). 71 71. See Cal. Ins. Code 11580.24 (West 2018). Oregon and Washington have similar laws. ×
  • In 2012, California amended its driver’s license inspection statute to exempt membership programs permitting remote, keyless access to vehicles from driver’s license inspection requirements. 72 72. Cal. Civ. Code § 1939.37 (Deering 2019). × As of the date of this article, a similar draft bill is pending in Massachusetts. 73 73. H.D. 4139 (Mass. 2019). A similar bill came into effect in Florida on July 1, 2019. See Fla. Stat. Ann. § 322.38 (West 2019). ×
  • In 2015, Florida and Hawaii amended their laws to impose modified car rental surcharges on “carsharing organizations” (i.e., membership programs providing self-service access to vehicles on an hourly or other short-term basis). 74 74. Fla Stat. Ann. § 212.0606 (LexisNexis 2019); Haw. Rev. Stat. Ann. § 251 (LexisNexis 2019). ×
  • Maryland passed the first comprehensive “Peer-to-Peer Car Sharing Program” law in 2018. The Maryland law defines a “peer-to-peer car sharing program” as, “a platform that is in the business of connecting vehicle owners with drivers to enable the sharingof motor vehicles for financial consideration” 75 75. Md. Code Ann., Ins. § 19-520(a)(9) (LexisNexis 2019). Illinois also passed a peer-to-peer car sharing/rental law in 2018, but that law was vetoed by then-Governor Rauner. Michael J. Bologna, Illinois Governor Pumps the Brakes on Car-Sharing Taxes, Bloomberg; Daily Tax Report: State (August 31, 2018), https://www.bna.com/illinois-governor-pumps-n73014482161/ (last visited May 15, 2019). × and extends a number of vehicle rental law requirements, including those related to safety recalls, 76 76. Md. Code Ann., Transp., § 18.5-109 (LexisNexis 2019). ×  collision damage waiver sales, 77 77. Md. Code Ann., Com. Law, § 14-2101 (LexisNexis 2019). ×  limited lines licensing in connection with the sale of car rental insurance, 78 78. Md. Code Ann., Ins., § 10-6A-02 (LexisNexis 2019). × airport concession agreements, 79 79. Md. Code Ann., Transp. § 18.5-106 (LexisNexis 2019). ×  and recordkeeping requirements, to peer-to-peer car sharing programs. 80 80. Md. Code Ann., Ins. § 19-520 (LexisNexis 2019). × It also exempts the Peer-to-Peer Car Sharing Program operator and the shared vehicle’s owner from vicarious liability based solely on vehicle ownership in accordance with the Graves Amendment. 81 81. Md. Code Ann., Ins. § 19-520(e) (LexisNexis 2019). ×

 As of June 2019, the following states have pending, or have passed, peer-to-peer car sharing/car rental (or personal motor vehicle sharing) legislation: Arizona, California, Colorado, Georgia, Hawaii, Indiana, Iowa, Massachusetts, Nebraska, Nevada, New Hampshire, New Jersey, New Mexico, New York, Washington, and West Virginia. 82 82. Arizona H.B. 2559 (Ariz. 2019) and S.B. 1305 (Ariz. 2019); A.B. 1263 (Cal. 2019); S.B. 090 (Colo. 2019); H.B. 378 (Ga. 2019); H.B. 241 HD2 SD 1 (Haw. 2019) and S.B. 662 SD2 (Haw. 2019); Pub. L. No. 253 (Ind. 2019) (to be codified at Ind. Code § 9-25-6-3); H.F. 779 (Ia. 2019); H.D. 4139 (Mass. 2019); L.B. 349 (Neb. 2019); S.B. 478 (Nev. 2019); H.B. 274 (N.H. 2019); A.B. 5092 (N.J. 2019); S.B. 556 (N.M. 2019); S.B. 5995 (N.Y. 2019); H.B. 2071 (Wash. 2019); H.B. 2762 (W. Va. 2019). × The scope of the pending bills ranges from extension of rental tax obligations to peer-to peer rentals to more comprehensive schemes similar to that passed in Maryland in 2018.

III.          The Challenge of Compliance

As demonstrated in the brief survey of existing rental laws above incumbent vehicle rental companies (especially those that operate in several states) must navigate numerous and often-inconsistent federal and state laws in their day-to-day operations. In addition to the challenges created by inconsistencies in the substantive requirements of the laws, not all of the laws use the same definition of “vehicle rental company” (which may vary depending upon the length of the transaction and the type of vehicle rented), so it is possible for an entity or transaction to be considered a “rental” in some, but not all, states or for some, but not all, purposes. 83 83. See Minto v. Zipcar New York, Inc., No. 15401/09 (N.Y. Super. Ct., Queens County Mar. 17, 2010). ×

In recent years, the challenge of compliance with existing laws – most of which did not contemplate anything other than a face-to-face handover of vehicle and keys — has increased as new entrants and incumbent operators attempt to innovate and take advantage of new technology to improve operations and customer experience. For example, use of kiosks, keyless access and GPS fleet management are all innovations that can improve the customer experience, which existing vehicle rental laws fail to facilitate. Enter the newer mobility operators, and things become even more interesting, with a close analysis of the definition of “rental company,” “rental vehicle,” and other key terms becoming even more important. To provide some context, consider a few hypotheticals:

Hypothetical 1 A 26-year old driver with a facially valid, but recently suspended driver’s license, rents a car in Arizona and is involved in an accident injuring a third party. Under Arizona law and indeed the law of all states, the rental car operator meets its statutory obligations by inspecting the driver’s license and confirming that it is facially valid. There is no duty to conduct any further investigation into the status of the driver’s license or the driving record of the prospective renter. Under this simple fact pattern, the rental car company has no liability to the injured third party for the negligence of the renter (beyond any state mandated minimum financial responsibility limit). Should the outcome be the same for a carsharing operation where the user accesses the vehicle through an app without any direct in-person contact with personnel of the operator? What about an owner of small fleet of cars who “rents” his vehicles through a peer-to-peer rental platform? How about a subscription program where an employee delivers a vehicle to a “lessee” or “renter” who has elected to switch the model of car being used?

Hypothetical 2 A California carshare member has had possession of a vehicle for three days and the operator receives notice that the member’s credit card is expired. The member has not responded to inquiries from the operator. If the carsharing transaction is considered to be a rental, as noted above, in California and a few other states, the mobility operator is precluded by statute from utilizing the vehicle’s GPS to locate the vehicle (at least until certain time periods have expired). Should that same limitation apply to the carshare operator? What if the purpose was to make sure that vehicles are properly distributed around a region so that it can serve its members’ anticipated demands? What about the renter of a peer-to-peer vehicle who is late with the car – can either the owner of the car or the peer-to-peer platform assist in locating the car via the vehicle’s GPS system? Can the operator of a subscription program utilize GPS to track the location of vehicles?

Hypothetical 3 A 30-year old renter with a valid license rents a vehicle through a peer-to-peer platform and two days later causes an accident resulting in substantial property damage and injuries. Pursuant to the federal Graves Amendment, if a peer-to-peer rental is characterized as a car rental transaction, the vehicle owner might argue there is no vicarious liability for the actions of the driver (assuming there was no negligence in how the transaction was handled). It is possible the arguments would vary if the owner of the vehicle operated a small fleet of cars, which it placed on a peer-to-peer platform. A few courts have concluded that the Graves Amendment protection extends to carshare operations. 84 84. See id. × Should that protection extend to the individual or small fleet owner that utilizes a peer-to-peer platform? Is there any basis to extend the Graves Amendment protection to the platform operator given that it typically does not own the vehicles?

Currently, the answers to many of the questions raised above are unclear with scant guidance from state legislatures or courts. As a result, a mobility operator generally must look to the definition of “rental company” to determine whether its model is or may be covered by a particular law. And that inquiry may lead an incumbent car rental operator to argue that it should no longer be subject to the outdated vehicle rental laws and regulations either.

IV.          Proposal

There is an ongoing debate in the mobility industry as to the extent that some models need to comply with existing laws and regulations related to the rental car industry. In particular, some peer-to-peer companies resist the application of those rules to their operations and argue that they are merely a technology company providing a platform to connect drivers with cars, and therefore are not subject to taxes, licensing requirements, or consumer protection laws governing incumbent rental companies. 85 85. See Turo, Inc. v. City of Los Angeles, 2019 U.S. Dist. LEXIS 6532 (C.D. Cal. 2019) (dismissing as unripe a peer-to-peer platform provider’s claim that it is immune from liability for state law violations under Section 230 of the Communications Decency Act and denying motions to dismiss claims that the City of Los Angeles misclassified the peer-to-peer platform provider as a rental company). × However, others urge that if all mobility operators are offering essentially the same services (use of a non-owned vehicle), then it seems more accurate to consider all mobility operators in the same business – mobility. As the New York Supreme Court noted in the Zipcar cases discussed in Part B, the services provided by a carsharing company (Zipcar) served a similar consumer need and were “little different from ‘traditional rental car’ companies, notwithstanding marketing statements that contrast it with those companies.” 86 86. See Minto v. Zipcar New York, Inc., No. 15401/09; see also Orly Lobel, “The Law of the Platform,” 101 Minn. L. Rev. 87, 112 (November 2016). ×

Setting aside those differences, there is some value to the mobility industry as a whole in consistent laws and regulations on some issues across the country and, of course, in protecting the safety and privacy of users. What follows are a few recommendations that could form the basis for a set of uniform laws applicable to the mobility industry. 87 87. The authors are unaware of any existing model laws for car rental or the broader mobility industry. Although the National Association of Attorneys General issued the NAAG Report on car rental practices and “guidelines” in 1989, those Guidelines were not intended to serve as model and uniform law, but rather guidance on compliance with state unfair and deceptive trade practice laws. See supra note 8. In addition, the NAAG Guidelines are now 30 years’ old and somewhat outdated in light of the changes in technology and the evolution in the mobility industry discussed in this article. ×

A.         Standardized Terms and Definitions 

Mobility operators, consumers, and regulators would benefit if federal and state laws used more consistent definitions for key terms and phrases. The definitions of the different platforms at the beginning of this article could be a starting point (which we repeat here without citations for ease of reference):

  • “Carsharing” – a membership-based service that provides car access without ownership. Carsharing is mobility on demand, where members pay only for the time and/or distance they drive.
  • “Peer-to-Peer Carsharing or Rentals” – the sharing of privately-owned vehicles in which companies, typically for a percentage of the rental charge, broker transactions among car owners and renters by providing the organizational resources needed to make the exchange possible (i.e., online platform, customer support, driver and motor vehicle safety certification, auto insurance and technology).
  • “Subscriptions” – a service that, for a recurring fee allows a participating person exclusive use of a motor vehicle owned by an entity that controls or contracts with the subscription service. Typically, the subscriber is allowed to exchange the vehicle for a different type of vehicle with a certain amount of notice to the operator. The term of the subscription can vary, but should be subject to a periodic renewal by the subscriber (user).
  • “Vehicle Rental” – a customer receives use of a vehicle in exchange for a fee or other consideration pursuant to a contract for an initial period of time less than 30 days.
  • “Mobility Operators” – any person or entity that provides access to a vehicle to another person whether by an in-person transaction, an app-based or online platform, or any other means and whether the entity providing the access is the owner, lessee, beneficial owner, or bailee of the vehicle or merely facilitates the transaction.

In addition, standard definitions for the terms, “rental” and “rental company” would provide additional clarity for all mobility operators, and to the extent feasible, the more narrow term “rental” and its derivatives should be replaced with “mobility.”

“Rental” should focus on the service provided and be distinguished from long-term leases (which are subject to additional laws and regulations, including federal Regulation M). As a starting point, “rental” could be defined as the right to use and possess a vehicle in exchange for a fee or other consideration for an initial period of less than 90 days. 88 88. Although the definition of “consumer lease” is a transaction for a period exceeding 4 months, we note that other federal laws, such as Graham-Leach-Bliley impose additional requirements on leases of at least 90 days. See 12 C.F.R. § 213.2(e)(1) (2011); 16 C.F.R. § 313.3(k)(2)(iii) (2000). ×

“Rental Company” or “Mobility Company” should be defined as “any corporation, sole proprietorship or other entity or person who is engaged in the business of facilitating vehicle rental transactions.” 89 89. See, e.g., H.B. 2762 (W. Va. 2019). × A de minimis exemption for individuals renting private vehicles through a peer-to-peer or other private vehicle program could apply (e.g., no more than X vehicles available for rent during a 12-month period). 90 90. See id. ×

A more uniform definition for “Rental Vehicle” or “Mobility Vehicle” also could produce more consistency across or even within states since some existing vehicle rental laws currently apply only to “private passenger vehicles,” while others apply more broadly to “motor vehicles.” Before proposing model language, however, we believe that regulators and industry experts need to consider several important (and somewhat thorny) issues.

For example, consider the rental of a pick-up truck to a contractor for use at a construction site. If a law applies only to rentals of “private passenger vehicles,” then the pick-up truck likely would not be subject to the law. On the other hand, if the law applies more broadly to “motor vehicles,” then the pick-up truck rental likely would be covered. The policy argument for covering our hypothetical pick-up truck rental may be weaker for consumer protection statutes, like required disclosures for sales of damage waiver or child safety seat rules. On the other hand, using a broader definition of “rental vehicle,” which would include the hypothetical pick-up truck, may better serve the general public policy goals of the Graves Amendment, the Safe Rental Act, and laws related to liability and insurance.

B.         Use of GPS and Telematics Technology

The use of this technology for locating and monitoring vehicles for a legitimate business, operational, maintenance or safety purpose should be permitted. Those states that have restricted the use of GPS tracking have done so to protect the privacy of renters. Operators in states where there is no statutory limitation often provide a full disclosure to users that vehicle location and other data may be monitored. We believe there are certain mobility models and circumstances where location and other data should be monitored – as long as there is full disclosure. For example, a free-floating carshare operator should be allowed to monitor vehicle location for the purpose of serving anticipated demand. Similarly, an operator of an EV fleet should be allowed to monitor a vehicle’s battery charge and location to ensure an adequate charge level for the next user. Finally, mobility operators should have the right to use GPS or other technology to locate vehicles that have not been returned on time or when the operator otherwise has reason to believe that the vehicle has been abandoned or stolen, or to track mileage driven or fuel used for purposes of charging associated fees (provided there is appropriate notice and full disclosure to the user). On a broader scale, uniform regulation that permits some vehicle monitoring, as long as done in a manner to protect the privacy of a user and with full disclosure, should be adopted across all mobility platforms.

C.         Vehicle Access

Provided there is an initial verification of a driver’s license, a mobility operator that either allows access to vehicles without in-person contact or does not require signing of a rental agreement at the time of rental should be subject to a provision similar to the following:

If a motor vehicle rental company or private vehicle rental program provider facilitates rentals via digital electronic, or other means that allow customers to obtain possession of a vehicle without in person contact with an agent or employee of the provider, or where the renter does not execute a rental contract at the time of rental, the provider shall be deemed to have met all obligations to physically inspect and compare a renter’s driver license pursuant to this article when such provider:

  1. At the time a renter enrolls, or any time thereafter, in a membership program, master agreement, or other means of establishing use of the provider’s services, requires verification that the renter is a licensed driver; or
  2. Prior to the renter taking possession of the rental vehicle, the provider requires documentation that verifies the renter’s identity. 91 91. Id. ×

D.         Graves Amendment    

The Graves Amendment, by its language, applies to the business of “renting or leasing” vehicles. A few state court cases have confirmed that Graves applies to carsharing. That application should be expressly adopted on a national basis and extended to all mobility models that involve a vehicle “owner’s” grant of the right to possess and use a vehicle in exchange for a fee or other consideration (including loaner vehicles).

Similarly, subscription programs which operate somewhere between incumbent car rental and vehicle leasing programs, at their core involve the short-term use of a vehicle in exchange for payment. Provided the subscription program complies with state rental car laws or applicable subscription legislation, the operation should be subject to the Graves Amendment. For that reason, we recommend that state legislatures either refine the Indiana/North Carolina definition of “subscription” to clarify that the model is a rental or lease for purposes of the Graves Amendment or simply state that subscription models are exempt from state vicarious liability laws based on vehicle ownership.

Peer-to-Peer platforms raise some issues when considering the Graves Amendment. On the one hand, an end-user is paying money to use a vehicle that belongs to someone else much like an incumbent rental car operation. On the other hand, a true “peer”-or individual- who occasionally lists his or her personal vehicle for rent when not using it may not really be in the business of renting cars. Much of the recent Peer-to-Peer legislation addresses this and related issues. Our suggestion is that Peer-to-Peer be subject to express state legislation and that such legislation impose sufficient operational, safety and economic obligations on operators, including required insurance coverage. In the absence of Peer-to-Peer legislation, an operator should have to comply with existing state rental car regulations especially if the operator somehow claims it is subject to the Graves Amendment.

E.         Americans with Disabilities Act

    Compliance with and exceptions to the ADA is complex. However, we propose that all mobility operators with fleets above a certain size must provide adaptive driving devices for selected vehicles, as long as the customer provides advance notice (which may vary depending upon the operator’s location and fleet size) and the adaptive driving devices are compatible with vehicle design and do not interfere with the vehicle’s airbag or other safety systems.

F.         Disclosure Requirements

All operators must provide sufficient disclosures to users regarding the following matters: fees, charges, damage waivers, added insurance, and vehicle technology. However, typical requirements in the existing state rental laws, including specified placement and font size for disclosures and in-person acknowledgment of receipt of those disclosures, simply do not contemplate modern technology, including digital agreements and remote access.  We propose the 2018 amendment to the New York vehicle rental law as the model for addressing required disclosures and formatting in electronic and/or master, membership agreements. That amendment provides:

(a) Notwithstanding any other provision of this section, any notice or disclosure of general applicability required to be provided, delivered, posted, or otherwise made available by a rental vehicle company pursuant to this section shall also be deemed timely and effectively made where such notice or disclosure is:

(i)       provided or delivered electronically to the renter at or before the time required provided that such renter has given his or her expressed consent to receive such notice or disclosure in such a manner; or

(ii)      included in a member or master agreement in effect at the time of rental.

(b)  . . . Notices and disclosures made electronically pursuant to this subdivision shall be exempt from any placement or stylistic display requirements, including but not limited to location, font size, typeset, or other specifically stated description; provided such disclosure is made in a clear and conspicuous manner. 92 92. N.Y. Gen. Bus Law § 396-z(16). ×

G.         Other Issues

There are, of course, other issues the industry can consider. For example, some states (New York and Michigan) have laws requiring rental car companies to make vehicles available to younger drivers, subject to certain conditions. Some uniformity on the ability of mobility operators to set minimum age requirements would reduce risk. Additionally, there are inconsistent laws across the country regarding the amount of time a rental car company must wait after a renter fails to return a car before it can notify law enforcement. Appropriate and consistent rules as to when an operator can start to recover a valuable (and mobile) asset would help promote growth in the industry.

The mobility revolution involves a number of different players with disparate and sometimes competing interests. Not all the participants will agree on all the issues, however, we offer the above suggestions to encourage discussion and to advance some level of consistency on a few points.


Wes Hurst is an attorney with a nationwide Mobility and Vehicle Use Practice. He represents rental car companies, carsharing companies, automobile manufacturers and companies pursuing new and emerging business models related to mobility and the use of vehicles. Wes is a frequent speaker and author on mobility related topics. Wes is in the Los Angeles office of Polsinelli and can be reached at whurst@polsinelli.com.

Leslie Pujo is a Partner with Plave Koch PLC in Reston, Virginia. In her Mobility and Vehicle Use Practice, Leslie regularly represents mobility operators of all types, including car rental companies, RV rental companies, automobile manufacturers and dealers, carsharing companies and other emerging models. Leslie is a frequent speaker and author on car rental and other mobility topics and can be reached at lpujo@plavekoch.com.

* The authors wish to thank Naila Parvez for her assistance

By Emily Frascaroli, John Isaac Southerland, Elizabeth Davis, and Woods Parker

Cite as: Emily Frascaroli et al., Let’s Be Reasonable: The Consumer Expectations Test is Simply Not Viable to Determine Design Defect for Complex Autonomous Vehicle Technology, 2019 J. L. & Mob. 53.

Abstract

Although highly automated vehicles (“HAVs”) have potential to reduce deaths and injuries from traffic crashes, product liability litigation for design defects in vehicles incorporating autonomous technology is inevitable. During the early stages of implementation, courts and juries will be forced to grapple with the application of traditional product liability principles to a never before experienced category of highly technical products. Recent decisions limiting the use of the consumer expectations test in cases involving complex products prompted the authors to examine more closely the history behind and the future viability of the consumer expectations test in HAV litigation.

I.          Introduction

In 2016, more than 35,000 individuals died in vehicle crashes in the U.S. and the National Highway Traffic Safety Administration (“NHTSA”) estimated that 94% of these deaths were attributable to human error. 93 93. Automated Vehicles for Safety, NHTSA, https://www.nhtsa.gov/technology-inn ovation/automated-vehicles-safety (last visited May 2, 2019). × In 2017 and 2018, in their own self-driving safety reports, General Motors and Waymo also noted that approximately 1.2 million lives are lost worldwide each year due to car crashes. 94 94. Waymo, Waymo Safety Report: On the Road to Fully Self-Driving 3 (2018), https://storage.googleapis.com/sdc-prod/v1/safety-report/Safety%20Report%20 2018.pdf; General Motors, 2018 Self-Driving Safety Report 3 (2018), https://ww w.gm.com/content/dam/company/docs/us/en/gmcom/gmsafetyreport.pdf. × Each of these entities further agree that highly automated vehicle (“HAV”) 95 95. For purposes of this paper, the terms highly automated vehicle (HAV) or “self-driving” will refer to vehicles defined by SAE Levels 4–5. See SAE International, Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles J3016 (2018), https://saemobilus.sae.org/content/j3016 _201806. The SAE levels of automation are as follows: (0) No automation; the vehicle has zero autonomy, and the driver performs all tasks; (1) Driver Assistance: the vehicle is controlled by the driver, but some driver assistance features are included; (2) Partial Driving Automation: the vehicle has combined automated functions, but the driver must remain engaged with the driving task and monitor the environment constantly; (3) Conditional Driving Automation: the driver is necessary, but is not required to constantly monitor the environment—the driver must be ready to take control of the vehicle at all times; (4) High Driving Automation: the vehicle is capable of performing all driving functions under certain conditions, but the driver has the option of controlling the vehicle; and (5) Full Driving Automation: the vehicle is capable of performing all driving functions under all conditions, with the driver having the option of controlling the vehicle. × technology has the potential to reduce or remove human error from the equation. 96 96. See Waymo, supra note 2; General Motors, supra note 2. × Additional potential benefits include reduced traffic congestion; increased mobility options for currently underserved populations; and, increased comfort and a reduction in lost time during vehicle operation. 97 97. General Motors, supra note 2. × Put simply, the stage is set for HAV technology to revolutionize the mobile world.

During the implementation of HAVs, most sources agree that, due to their highly complex and technical nature, consumer education about the products will be key to successful and effective implementation. For its part, in the 2017 update, Automated Driving Systems 2.0, NHTSA stated that “[E]ducation and training is imperative for increased safety during the deployment of [HAVs],” and encourages the development of “consumer education and training programs to address the anticipated differences in the use and operation of [automated driving systems] from those of the conventional vehicles that the public owns and operates.” 98 98. NHTSA, Automated Driving Systems: A Vision for Safety 2.0 15 (2017), https://www.nhtsa.gov/vehicle-manufacturers/automated-driving-systems#automated-driving-systems-av-20. × General Motors and Waymo echoed these sentiments in their respective self-driving safety reports with Waymo, in October 2017, even helping to launch – Let’s Talk Self-Driving – which it describes as “the world’s first public education campaign about fully self-driving vehicles.” 99 99. Waymo, supra note 2 at 30; General Motors, supra note 2, at 32. × Taking this one-step further, in 2018, Ford provided its Voluntary Safety Self-Assessment Report – A Matter of Trust. 100 100. Ford Motor Company, A Matter of Trust: Ford’s Approach to Developing Self-DrivingVehicles, https://media.ford.com/content/dam/fordmedia/pdf/ Ford_AV_LLC_FINAL_HR_2.pdf. × In it, Ford makes clear “that the central challenge in the development of self-driving vehicles” is not the technology, but, instead, it is consumer trust in the “safety, reliability and experience that the technology will enable.” 101 101. Id.at 3. × Ford reiterates this point stating about consumer education and training that, “[B]ringing self-driving vehicles to market will require a thoughtful and sustained effort to teach customers how they work, why they’re safe and how to use them.” 102 102. Id. at 42 (emphasis added). ×

In light of this, questions remain, particularly with respect to liability, if, and when, an injury or death occurs in an HAV. The question of who is liable when a self-driving vehicle crashes has generated significant debate and conversation. Per NHTSA, “these are among many important questions beyond the technical considerations that policymakers are working to address before automated vehicles are made available.” 103 103. Automated Vehicles for Safety, supra note 1. × NHTSA also posits that questions of liability pertaining to HAVs are something within the purview of each state to manage. 104 104. NHTSA, supra note 6, at 24. × In the wake of some interesting opinions in 2017, this question, and others, prompted the authors to examine the historical development of product defect theories and, in particular, whether the consumer expectations test can reasonably be applied to determine liability in cases involving complex products.

II.         Adoption of Design Defect Tests in the Wake of Section 402A of the Restatement (Second) of Torts.

In 1965, the law of torts and the field of product liability were altered dramatically by the adoption of Section 402A of the Restatement (Second) of Torts. 105 105. See generally George L. Priest, Strict Products Liability: The Original Intent, 10 Cardozo L. Rev. 2301, 2301 (1989). × Section 402A sought to impose strict liability on the manufacturers or sellers of defective products, regardless of negligence, and became perhaps the most cited section of any Restatement of Law in legal jurisprudence. 106 106. See James A. Henderson Jr. & Aaron D. Twerski, Proposed Revision of Section 402A of the Restatement (Second) of Torts, 77 Cornell L. Rev. 1512, 1512 n.1 (1992). ×

A.         The Consumer Expectations Test

Section 402A provides that “[o]ne who sells any product in a defective condition unreasonably dangerous to the user or consumer or to his property is subject to liability for physical harm thereby caused to the ultimate user or consumer, or to his property . . . .” 107 107. Restatement (Second) of Torts § 402A (Am. Law Inst. 1965). × To guide courts in determining whether a product is unreasonably dangerous, the drafters of the Second Restatement included the following comment: “The article sold must be dangerous to an extent beyond that which would be contemplated by the ordinary consumer who purchases it, with the ordinary knowledge common to the community as to its characteristics.” 108 108. Id. at cmt. i. × This comment provided support for the pure consumer expectations test in product defect cases. In turn, this product defect test was embraced by courts in the years following the release of the Second Restatement. 109 109. See, e.g., Aller v. Rodgers Machinery Mfg. Co., Inc., 268 N.W. 2d 830 (Iowa 1978); Phipps v. General Motors Corp., A.2d 955 (Md. 1976); Estate of Pinkham v. Cargill, Inc., 55 A.3d 1 (Me. 2012) (citing Adams v. Buffalo Forge Co., 443 A.2d 932, 940 (Me. 1982)); Simonetta v. Viad Corp., 197 P.3d 127 (Wash. 2008). × Over time, courts across the country recognized that there were significant issues with the Second Restatement’s pure consumer expectations approach to defective design.

For example, in the 1967 case of Heaton v. Ford Motor Co., the Supreme Court of Oregon was faced with application of the consumer expectations test in the context of a design defect claim involving a motor vehicle. 110 110. See Heaton v. Ford Motor Co., 435 P.2d 806 (Or. 1967). × In Heaton, the plaintiff’s vehicle struck a rock in the roadway. After the accident, the rim of the wheel was found to have separated from the rest of the wheel assembly. The court utilized the consumer expectations test to determine design defect, stating:

In the type of case in which there is no evidence, direct or circumstantial, available to prove exactly what sort of manufacturing flaw existed, or exactly how the design was deficient, the plaintiff may nonetheless be able to establish his right to recover, by proving that the product did not perform in keeping with the reasonable expectations of the user. When it is shown that a product failed to meet the reasonable expectations of the user the inference is that there was some sort of defect. 111 111. Id. at 471–72 (emphasis added). ×

However, the court recognized that in Heaton, the jury could not possibly state from their own experience what the expectations of the average consumer would be. 112 112. See id. at 472–73. × After all, high-speed collisions with large rocks are not so common that the average person would know from personal experience how the wheel assembly would perform in such a situation. 113 113. Id. at 473. × As such, “[t]he jury would therefore be unequipped, either by general background or by facts supplied in the record, to decide whether this wheel failed to perform as safely as an ordinary consumer would have expected.” 114 114. Id. × Unfortunately, the Heaton court ultimately refused to acknowledge that the consumer expectations test simply did not apply in this situation, but instead seemed to suggest that expert testimony would be required to establish the consumer expectations. 115 115. See id. at 474. × The paradox is obvious: if an expert is required to tell the consumer what to expect, is that truly the expectation of an ordinary consumer?

Fortunately, courts have begun to recognize that utilizing the consumer expectations test in cases involving alleged design defects in technically complex products is simply not workable. 116 116. See, e.g., Montag v. Honda Motor Co., Inc., 75 F.3d 1414 (10th Cir. 1996) (citing Camacho v. Honda Motor Corp., 741 P.2d 1240, 1246–48 (Colo. 1987)). See also 2 Louis R. Frumer & Melvin I. Friedman, Products Liability§ 11.03 (Matthew Bender, Rev. Ed.). × However, there are courts that have found the consumer expectations test applicable, even where the requisite knowledge is not within the purview of lay jurors. 117 117. See, e.g., Bresnahan v. Chrysler Corp., 38 Cal. Rptr. 2d 446, 451–52 (Cal. Ct. App. 1995). See also 2 Frumer & Friedman, supra note 25. ×

B.         Risk Utility Test

As a result, many courts began to apply the test commonly referred to as risk-utility balancing. Under this test, to establish a prima facie case of design defect, the plaintiff must show that on balance, the utility of the challenged product design outweighs the risk of danger inherent in the design. 118 118. See, e.g., Thibault v. Sears, Roebuck & Co., 395 A.2d 843 (N.H. 1978). × Traditionally, under risk-utility, courts consider a multitude of factors to determine whether a defect exists, including the following factors identified in an influential article by Dean John W. Wade in 1973:

    1. The usefulness and desirability of the product;
    2. the safety aspects of the product;
    3. the availability of safer substitute products;
    4. the possibility of elimination of dangerous characteristics of the product without impairing its usefulness;
    5. the user’s ability to avoid danger by safe use of the product;
    6. the anticipated dangers inherent in the product due to general knowledge or the existence of warnings; and
    7. the possibility of loss-spreading by the manufacturer through price setting or insurance. 119 119. See John W. Wade, On the Nature of Strict Tort Liability for Products, 44 Miss. L.J. 825 (1973). ×

Further, “[t]he utility of the product must be evaluated from the point of view of the public as a whole, because a finding of liability for defective design could result in the removal of an entire product line from the market.” 120 120. See Thibault, 395 A.2d at 807. ×

In 1998, the element of a reasonable alternate design was written into the new Restatement (Third) of Torts. 121 121. See Restatement (Third) of Torts § 2 (Am. Law Inst., 1998). × Under § 2of the Third Restatement, a product is:

 “[D]efective in design when the foreseeable risks of harm posed by the product could have been reduced or avoided by the adoption of a reasonable alternative design by the seller or other distributor, or a predecessor in the commercial chain of distribution, and the omission of the alternative design renders the product not reasonably safe. 122 122. Id. The Third Restatement explicitly rejects consumer expectations as an independent standard for determining design defect. See id. at §2 cmt. g. ×

As noted in 2009, the “reasonable alternative design” standard of the Third Restatement ultimately came to embody the “risk-utility test” that is applied in the majority of United States jurisdictions today. 123 123. See Aaron D. Twerski & James A. Henderson, Jr., Manufacturer Liability for Defective Product Designs: The Triumph of Risk Utility, 74 Brook. L. Rev. 1061, 1065 (2009). Notably, even courts that continue to utilize the consumer expectations test exclusively often acknowledge that evidence of an alternative design is the most appropriate and useful means of showing that a product is unreasonably dangerous. See, e.g., Ford Motor Co. v. Trejo, 402 P.3d 649, 655 (Nev. 2017). ×

Significant developments since 2009, some of which are discussed more fully below, further exemplify the national trend towards applying risk-utility in complex design defect cases and moving away from the consumer expectations test in this context. Indeed, in 2017, the Ninth Circuit recognized that, “when the ultimate issue of design defect calls for a careful assessment of feasibility, practicality, risk, and benefit, the case should not be resolved simply on the basis of ordinary consumer expectations.” 124 124. See Edwards v. Ford Motor Co., 683 Fed. App’x 610, 611 (9th Cir. 2017) (quoting Soule v. General Motors Corp., 882 P.2d 298, 305 (Cal. 1994)) (emphasis supplied in original). ×

C.         Hybrid Test

Other jurisdictions utilize a dual-approach to design defect claims. California, for example, utilizes the consumer expectations test when consumers are capable of developing expectations about the characteristics of a product from everyday use. 125 125. See Soule, 882 P.2d at 310–311. × For more complex products, where the characteristics are outside the knowledge of an everyday consumer, courts apply the risk-benefit test. 126 126. See id. × Thus, the determinative issue in many cases in California and similar jurisdictions is whether a product is too complex or unfamiliar for average consumers to develop expectations, such that utilization of the consumer expectations test is improper. 127 127. See, e.g., Saller v. Crown Cork & Seal Co., Inc., 115 Cal. Rptr. 3d 151, 160–61 (Cal. Ct. App. 2010). × Making this determination in the context of autonomous technology should not be an issue.

This hybrid approach combines elements of both the consumer expectations test and the risk-utility test. One example is the “either-or” concept, which posits that:

[A] product is defective in design either (1) if the product has failed to perform as safely as an ordinary consumer would expect when used in an intended or reasonably foreseeable manner, or (2) if, in light of the relevant factors . . . the benefits of the challenged design do not outweigh the risk of danger inherent in the design. 128 128. Barker v. Lull Eng’g Co., 573 P.2d 443, 446 (Cal. 1978). ×

This approach allows courts more flexibility in applying the appropriate test based upon all of the relevant circumstances. For example, in Barker v. Lull Engineering, the plaintiff sustained injuries while operating a loader at a construction site and alleged that his injuries were caused by a defective design of the product because it was not equipped with a roll bar or seat belts. 129 129. Id. at 447–48. × The California Supreme Court rejected a pure consumer expectations test and a pure risk-utility test, instead articulating the two-prong test allowing a plaintiff to establish a design defect through either test. 130 130. Id. at 455–56. × In so holding, the court noted the benefits of the more flexible approach stating:

[I]t subjects a manufacturer to liability whenever there is something “wrong” with a product’s design – either because the product fails to meet ordinary consumer expectations as to safety or because, on balance, the design is not as safe as it should be – while stopping short of making the manufacturer an insurer for all injuries which may result from the use of its product. This test, moreover, explicitly focuses on the trier of fact’s attention to the adequacy of the product itself, rather than on the manufacturer’s conduct, and places the burden on the manufacturer, rather than the plaintiff, to establish that because of the complexity of, and trade-offs implicit in, the design process, an injury-producing product should nevertheless not be found defective. 131 131. Id. at 456. ×

Other courts have taken a different approach in formulating a hybrid consumer expectations and risk-utility test, incorporating risk-utility factors into the consumer expectation analysis, and vice versa. 132 132. See, e.g.,Potter v. Chicago Pneumatic Tool Co., 694 A.2d 1319, 1333–34 (Conn. 1997). × For example, in Potter v. Chicago Pneumatic Tool Co., the plaintiffs alleged that pneumatic hand tools manufactured by the defendant were defective in design because they exposed the plaintiffs to excessive vibration, resulting in injuries to the plaintiffs. 133 133. Id. at 1325. × Although Connecticut courts had long applied the Second Restatement’s consumer expectations test, the court recognized that “there may be instances involving complex product designs in which an ordinary consumer may not be able to form expectations of safety.” 134 134. Id. at 1333. × In recognizing this issue with the pure consumer expectations test, the Connecticut Supreme Court adopted a “modified consumer expectation test, provid[ing] the jury with the product’s risks and utility and then inquir[ing] whether a reasonable consumer would consider the product unreasonably dangerous.” 135 135. Id. × In determining a consumer’s reasonable expectations, the jury should consider various factors, including but not limited to the “relative cost of the product, the gravity of the potential harm from the claimed defect and the cost and feasibility of eliminating or minimizing the risk.” 136 136. Id. × In sum, under this approach, “the consumer expectation test would establish the product’s risks and utility, and the inquiry would then be whether a reasonable consumer would consider the product design unreasonably dangerous.” 137 137. Id. × The Connecticut Supreme Court’s approach was based, in part, on early drafts of the Restatement (Third) of Torts on Product Liability. 138 138. Id. at 1331. ×

III.       More and More Courts Are Recognizing the Limitations of the Consumer Expectations Test in Complex Design Defect Cases.

In March 2017, the United States Court of Appeals for the Ninth Circuit examined the question of whether the consumer expectations test or risk-utility balancing test should be applied to determine whether a design defect existed in a product liability case involving vehicle roof deformation. 139 139. See Edwards v. Ford Motor Co., 683 Fed. App’x 610 (9th Cir. 2017). × In Edwards v. Ford Motor Co., the plaintiffs claimed that the roof of their vehicle was defectively designed because it deformed inward eight inches into the passenger compartment during a multiple rollover event. The plaintiffs alleged that a properly designed roof should have resulted in less than three inches of deformation in the subject crash.

The Edwards plaintiffs sought to prove design defect by showing that the roof did not perform as the average consumer would have expected. Ford filed a motion contending that the jury should be instructed on the risk-utility test alone. Ford’s motion was granted and plaintiffs appealed. The Ninth Circuit held that the risk-utility test was the proper test to be applied, specifically recognizing the “lack of consumer expectations regarding the extent to which the [vehicle]’s roof would crush in a multiple rollover accident.” 140 140. Id. at 611. × The Ninth Circuit went on to note that “[d]rivers’ everyday experiences do not allow for the formulation of reasonable expectations as to the degree that a vehicle’s roof should crush during a rollover.” 141 141. Id. × The Ninth Circuit did not feel it necessary to state whether or not this product was too complex for the consumer expectations test to govern; instead, it was enough to know that consumers simply would not have expectations related to roof performance in a rollover. 142 142. See id.; See also Branham v. Ford Motor Co., 701 S.E. 2d 5, 13–14 (S.C. 2010) (finding that the consumer expectations test was not appropriate in design defect case after examining the issue in the context of an automotive rollover case involving an allegedly defective seatbelt design). × Thus, the risk-utility test was the appropriate test. 143 143. See id. ×

Another recent case decided by the Court of Appeal of California also limited the applicability of the consumer expectations test. The plaintiff in Trejo v. Johnson & Johnson contracted a rare condition known as SJS/TEN as a reaction to taking over-the-counter ibuprofen produced by Johnson & Johnson. Plaintiff sought to show that the drug was defectively designed through utilization of the consumer expectations test.

The Court of Appeal found the consumer expectations had no place in proving design defect under these facts, noting that “‘[t]he consumer expectations test is reserved for cases in which the everyday experience of the product’s users permits a conclusion that the product’s design violated minimum safety assumptions, and is thus defective regardless of expert opinion about the merits of the design.’” 144 144. Trejo v. Johnson & Johnson, 220 Cal. Rptr. 3d 127, 165 (Cal. Ct. App. 2017) (quoting Soule v. General Motors Corp., 882 P.2d 298, 308 (Cal. 1994). × The plaintiff essentially attempted to use consumer expectations to avoid having to confront the more difficult risk-utility standard or any showing of a reasonable alternative design, but also wished to introduce expert testimony to establish that the ibuprofen did not meet consumer expectations. The court found this fact alone sufficient to demonstrate the consumer expectations test was inappropriate for that case. 145 145. See id. at 168. ×

Succinctly explaining the problem with applying consumer expectations in the case of complex products or products with which consumers are unfamiliar, the court stated: “[I]t could be said that any injury from the intended or foreseeable use of a product is not expected by the ordinary consumer. If this were the end of the inquiry, the consumer expectations test always would apply and every product would be found to have a design defect.” 146 146. See id. at 167 (emphasis added). ×

As to a non-complex product, the Tenth Circuit’s examination of consumer expectations in Kokins v. Teleflex, Inc. is instructive. 147 147. See Kokins v. Teleflex, Inc., 621 F.3d 1290 (10th Cir. 2010). × Kokins involved the determination of what design defect test should be used under Colorado law in the context of a claim involving a metal marine cable, a seemingly simple product. The court initially noted that, under Colorado law, the risk-utility test and consumer expectations test are not mutually exclusive of each other and can sometimes even be applied in the same case. 148 148. See id. at 1297. × However, the Tenth Circuit held that in the context of this particular product, only the risk-utility test was proper, due to the technical and specific information related to metallic corrosion. 149 149. See id. × Quite simply, in cases where technical and scientific issues predominate, use of the consumer expectations test, alone or in conjunction with the risk-utility test, is inappropriate. 150 150. See id. ×

Finally, as recently as November, 2017, the Colorado Supreme Court determined that the “risk-benefit test is the appropriate test to assess whether a product was unreasonably dangerous due to a design defect when . . . the dangerousness of the design is ‘defined primarily by technical, scientific information.’” 151 151. See Walker v. Ford Motor Co., 406 P.3d 845, 850 (Colo. 2017) (quoting Ortho Paharm. Corp. v. Heath, 722 P.2d 410, 414 (Colo. 1986)). × In Walker v. Ford Motor Co., the plaintiff proceeded to trial against Ford for injuries sustained in a rear-end impact. 152 152. See id. at 847­–48. × The plaintiff alleged the seat in his vehicle was defectively designed, alleging theories based in both strict liability and negligence. 153 153. See id. at 848. × At the end of trial, the trial court instructed the jury that it could apply either a consumer expectation test or risk-benefit test, and the jury found in favor of the plaintiff. 154 154. See id. at 848. × The court of appeals reversed the jury verdict. 155 155. See id. at 849. ×

In affirming the Colorado Court of Appeals, the Colorado Supreme Court recognized that it had “stated repeatedly that the risk-benefit test, not the consumer expectation test, is the proper test to use in assessing whether a product like the car seat . . . is unreasonably dangerous due to a design defect.” 156 156. Id. at 850. × The Court further noted:

[P]roducts-liability law has developed in part to “encourage manufacturers to use information gleaned from testing, inspection and data analysis” to help avoid product accidents. Using the risk-benefit test . . . helps further this objective, as it directs the fact-finders to consider the manufacturer’s ability to minimize or eliminate risks and the effect such an alteration would have on the product’s utility, other safety aspects, or affordability. 157 157. Id. at 851 (quoting Camacho v. Honda Motor Co., 741 P.2d 1240, 1247 (Colo. 1987)). ×

While the authors recognize the debate about whether to apply the consumer expectations test or the risk-utility test continues to this day, and that some jurisdictions still apply the consumer expectations test, even in cases of complex products, the above referenced opinions illustrate the issues and concerns with asking jurors to determine the expectations of an ordinary consumer when evaluating a highly technical products in design defect matters.

IV.       The Consumer Expectations Test Is Not the Appropriate Test of Design Defect as Applied to Autonomous Vehicle Technology

The arrival of any new product technology will bring with it litigation, and along with that arguments for the legal standard that will place the lightest burden on plaintiffs in this new arena. Thus, it is likely that as lawsuits begin with autonomous vehicle technology, plaintiffs will argue that the consumer expectations test should apply to their claims for alleged design defects in autonomous vehicles. The argument will likely follow the reasoning employed by courts that refuse to adopt the Third Restatement approach, or that still strictly follow the consumer expectations test, i.e. that risk-utility balancing, especially when a reasonable alternative design is required, places too great of a burden on plaintiffs that do not have the resources to make showings that are so technical in nature. 158 158. See, e.g., Potter v. Chicago Pneumatic Tool Co., 694 A.2d 1319, 1332 (Conn. 1997); Vautour v. Body Masters Sports Indus., 784 A.2d 1178, 1183 (N.H. 2001). ×

A.         Highly Automated Vehicles Are Too Complex for Consumer Expectations to Govern.

The Society of Automotive Engineers lists six (6) levels of automation for HAVs. 159 159. See SAE International, supra note 3. × Currently, all vehicles on roadways are levels one and two, with Audi unveiling the world’s first production Level 3 vehicle in July 2017. 160 160. As reported in an article by IEEE Spectrum, Audi claims to have achieved level 3 through its “AI Traffic Jam Pilot” feature, which can only be activated when driving at less than 37 mph. See Philip E. Ross, The Audi A8: The World’s First Production Car to Achieve Level 3 Autonomy, IEEE Spectrum (July 11, 2017), https://spectrum.ieee.org/ca rs-that-think/transportation/self-driving/the-audi-a8-the-worlds-first-production-car-to-achieve-level-3-autonomy. × Further, even if fully autonomous vehicles were on the road today, the vast majority of consumers will remain unfamiliar with the technology for the foreseeable future. Drivers keep their vehicles on the road for over eleven years on average, 161 161. Reno Charlton, American Drivers Keeping Cars on the Road for Longer: Average Age Now 11.4 Years, Huffington Post (Aug. 9, 2013), https://www.huffpost.com/entry/american-drivers-keeping_b_3718301?guccounter=1. × so vehicles of lower automation levels will continue to be the predominant means of automotive transportation for years to come. 162 162. See Brian A. Browne, Self-Driving Cars: On the Road to a New Regulatory Era, 8. J. L., Tech. & Internet 1, 3 (2017) (Giving examples of the various lower level features many OEMs have planned for the coming years). ×

Further, NHTSA acknowledges the lack of consumer experience with autonomous vehicle technology, as well as how different these vehicles are from conventional vehicles on the roads today. In 2017, in Automated Driving Systems 2.0: A Vision for Safety, NHTSA pronounced that:

Proper education and training is imperative to ensure safe deployment of automated vehicles. Therefore, manufacturers and other entities should develop document, and maintain employee, dealer, distributor, and consumer education and training programs to address the anticipated differences in the use and operation of HAVs [highly automated vehicles] from those of conventional vehicles that the public owns and operates today. Such programs should be designed to provide the target users the necessary level of understanding to use these technologies properly, efficiently, and in the safest manner possible. 163 163. National Highway Traffic Safety Administration, Federal Automated Vehicles Policy: Accelerating the Next Revolution in Roadway Safety 24 (2016), https://www.transportation.gov/sites/dot.gov/files/docs/AV%20policy%20guidance%20PDF.pdf (emphasis added). ×

Essentially, NHTSA is recommending a completely new dimension of consumer education on how to use these products. Likewise, this education will be aimed at and received primarily by consumers who actually purchase and use autonomous vehicle technology and not automotive consumers generally.

On this point, in a 2014 survey conducted by researchers at the University of Michigan’s Transportation Research Institute, Americans were asked, “[h]ow interested would you be in having a completely self-driving vehicle . . . as the vehicle you own or lease?” The most commonly chosen answer, comprising 33.7% of responses, was “not at all interested” with another 22.4% of respondents answering that they would be only “slightly interested.” 164 164. See Brandon Schoettle & Michael Sivak, Public Opinion About Self-Driving Vehicles in China, India, Japan, the U.S. and Australia 16 (Univ. of Mich. Trans. Res. Inst. Report No. 2014-30, 2014), https://deepblue.lib.umich.edu/handle/202 7.42/109433 (emphasis added). × This information suggests that not only are most Americans personally unfamiliar with HAVs, but that a majority of Americans will not become familiar with such vehicles any time soon.

Another striking result of that survey was that, of Americans with Internet access, only 70.9% of respondents had even heard of autonomous or self-driving vehicles. 165 165. See id. at 5. × If these respondents were placed on a jury in a jurisdiction applying the consumer expectations test, roughly three of twelve jurors would be deciding liability based on the ordinary expectations of a consumer for a product about which they had never heard

Moreover, a study by various researchers in the MIT AgeLab suggests that naming conventions for autonomous or “advanced driver assistance systems” can influence the expectations that a consumer may have about these systems. 166 166. Hillary Abraham, et al., What’s in a Name: Vehicle Technology Branding & Consumer Expectations for Automation, AutomotiveUI ‘17 Proceedings of the 9th International Conference on Automotive User Interfaces and Interactive Vehicular Application 226-234 (2017), available at http://st.sigchi.org/publications/ toc/auto-ui-2017.html. × In particular, the authors of this paper observed that:

[D]rivers’ attitudes and beliefs about system capability and performance are known to influence their use of technology. Factors such as a driver’s prior experience with similar technologies, predisposed trusting tendencies, and attitudes formed from exposure to media and societal opinion might all contribute to a driver’s belief that a system can handle a task outside of its [operational design domain].” 167 167. Id. ×

Further, the authors found that “the name of a driver assistance system also has the potential to impact their perceptions of system capability. 168 168. Id. × These same perceptions or misconceptions developed by unfamiliar consumers simply from the name of a particular system are sure to carry over to these consumers ability to judge the systems if called upon in a legal setting.

This is important because, while the consumer expectation test is intended to be an objective test that is applied based on the ordinary consumer’s expectation, the gravamen of the test is that “the everyday experience of the product’s users permits a conclusion that the product’s design violated minimum safety assumptions . . . .” 169 169. See Edwards v. Ford Motor Co., 683 Fed. App’x 610, 611 (9th Cir. 2017) (quoting Soule v. General Motors Corp., 882 P.2d 298, 305, 308 (Cal. 1994)). × At least initially, and most likely for quite a period thereafter, the average juror will simply not possess the everyday experience necessary to properly assess the product in a consumer expectations analysis. Rather, it is much more appropriate and fair to aid a jury by allowing the greater body of evidence encompassed within a risk-utility analysis.

B.         Consumer Expectations of Autonomous Vehicle Technology are Inconsistent and Unrealistic at this Point.

Even when consumer expectations are drawn broadly (i.e., safe versus unsafe), instead of in terms of how a particular aspect of an autonomous vehicle should perform at a technical level, consumer expectations at this point in time have not reached any kind of meaningful consistency. For example, many consumers are highly skeptical of new HAV technology and believe that the technology is inherently unsafe. 170 170. Jeremy Hsu, 75 Percent of U.S. Drivers Fear Self-Driving Cars, But It’s an Easy Fear to Get Over, IEEE Spectrum (Mar. 7, 2016, 15:01 GMT), http://spectrum.ieee.org/ cars-that-think/transportation/self-driving/driverless-cars-inspire-both-fear-and-hope. × On the other hand, some organizations anticipate large reductions in automotive accidents and injuries as a result of this new technology and propound this message to the general public. 171 171. See, e.g., Mothers Against Drunk Driving, MADD Statement on Autonomous Vehicle Technology Legislation, (October 4, 2017), https://www.ma dd.org/press-release/madd-statement-autonomous-vehicle-technology-legislation/. × For its part, NHTSA helped promote the narrative that the promise of self-driving vehicles will lead to a marked increase in automotive safety, noting in their 2017 update that, “in the transportation sector, where 9 out of 10 serious roadway crashes occur due to human behavior, automated vehicle technologies possess the potential to save thousands of lives, as well as reduce congestion, enhance mobility, and improve productivity.” 172 172. NHTSA, supra note 6, at ii. × Some manufacturers are no different: in GM’s 2018 Self-Driving Safety Report, the manufacturer optimistically stated that as a result of self-driving technology, they “envision a future with zero crashes.” 173 173. General Motors, supra note 2 at 3 (emphasis added). ×

Further, HAV manufacturers, eager to explain the admittedly revolutionary technology their vehicles employ, may inadvertently present consumers with the impression that these vehicles truly can do no wrong. Consider the following language from Delivering Safety: Nuro’s Approach:

Our vehicle is engineered to be safer than nearly any other – it is lighter than a passenger vehicle, narrower and more nimble, and operates at lower speeds. This approach gives us more time to react, shortens our stopping distance, and provides an additional safety buffer to the side of the vehicle. Together, these advantages help prevent accidents that standard vehicles cannot avoid, such as someone jumping out from between parked cars or swerving across the road. 174 174. Nuro, Delivering Safety: Nuro’s Approach 8 (2019), https://tonnietal ler.files.wordpress.com/2019/03/d5d69-delivering_safety_nuros_approach.pdf. ×

It is certainly true that HAV technology will revolutionize automotive safety overall. However, these types of statements may lead many consumers to believe that autonomous vehicles should perform to the point of infallibility, which is simply not possible, especially at this early stage of development.

For example, on May 7, 2016, a driver of a Tesla Model S was killed when the driver collided with a tractor-trailer who was crossing an uncontrolled intersection. 175 175. NHTSA Office of Defects Investigation Report, available at https://static.n htsa.gov/odi/inv/2016/INCLA-PE16007-7876.pdf. × The vehicle’s data resulted in three important findings:

  1. That the Tesla was being driven in autopilot mode at the time of the accident;
  2. the automatic emergency braking (AEB) system did not automatically brake or warn to avoid the collision, and;
  3. that the driver did not take any preventive steps, i.e. braking or steering, to avoid the collision. 176 176. NHTSA Office of Defects Investigation Report, supra note 83, at 1. ×

Because of the accident, both the National Transportation Safety Board (“NTSB”) and the National Highway Traffic Safety Administration (NHTSA) through their Office of Defects Investigation (“ODI”) conducted investigations. 177 177. See NHTSA Office of Defects Investigation Report, supra note 83;see also NTSB, NTSB/HAR-17/02, Collision Between a Car with Automated Vehicle Control Systems and a Tractor-Semitrailer Truck (2017), available at https://www.ntsb.gov/investigations/AccidentReports/Reports/HAR1702.pdf. ×

For example, the ODI investigated: (1) the AEB system design and performance; (2) human-machine interfaces related to operating in autopilot mode; (3) additional accident data regarding Tesla’s autopilot and AEB systems; and, (4) the changes if any Tesla has made to such autopilot and AEB systems. 178 178. See NHTSA Office of Defects Investigation Report, supra note 83, at 1. × The result of the investigation was that there were no defects in the design or performance of the autopilot or AEB systems in the vehicles studied – nor was there a situation to which the systems did not perform as designed. 179 179. See id. at 12. ×

Given the situation, is it reasonable to task an “ordinary consumer” with properly determining whether the AEB and autopilot systems are in fact functioning properly or improperly? Compare the reported results of the investigations by NHTSA’s ODI and the NTSB with the statements by Forbes contributor, Brad Templeton, in his article, “Tesla Autopilot Repeats Fatal Crash; Do They Learn From Past Mistakes? 180 180. Brad Templeton, Tesla Autopilot Repeats Fatal Crash; Do They Learn From Past Mistakes?, Forbes (2019), https://www.forbes.com/sites/bradtempleton/2019/05/2 1/tesla-autopilot-repeats-fatal-crash-do-they-learn-from-past-mistakes/#400f773f2f2e. × To wit, Templeton posits, “Even so, most would hope the Tesla Autopilot would have detected the truck crossing in front of it, which appeared not to happen. No braking or evasive actions were taken. The Autopilot was engaged just 10 seconds before the collision.” 181 181. Id. × He further opines:

As such, having already had a fatality from (the old system’s) failure to identify the broad side of a transport trailer, that would have to be very high on the list of the sort of thing they would want their fleet to find and identify for them, so they can confirm it never fails to perceive a crossing truck. Somehow, it still failed. Of all the things you would expect Tesla to identify, these few things which resulted in fatal accidents, like a truck side and a highway crash attenuator, should be at the very top of the list. 182 182. Id. (emphasis in original). ×

Although Templeton is likely more informed than the ordinary consumer, the opinions expressed in his article and the conclusions reached by the NHTSA and NTSB are in clear contradiction of one another. Thus, when considering the expectations of the everyday consumer, it is clear the necessary information is simply not available to conduct investigations such as the one carried out by the ODI or the NTSB, which can take months of analysis and result in sixty-three-page accident reports, and ultimately determine what actually occurred.

Similar to the California Court of Appeal’s reasoning in Trejo that the consumer expectations test could lead to virtually unlimited liability in cases of complex products, the current climate of high expectations regarding HAVs would likely mean that a HAV manufacturer would lose every time when the consumer applications test is applied. Consumers will expect that HAVs should avoid accidents one hundred percent of the time, so any time one of these vehicles is involved in an accident, it has already failed the consumer expectations test. This type of res ipsa loquitur conclusion undermines the concept of design defects in products liability law and would allow plaintiffs to completely sidestep the requirement of a showing that an HAV was in fact defective, effectively making manufacturers of HAVs insurers of those products’ safety. 183 183. See Funkhouser v. Ford Motor Co., 736 S.E.2d 309, 314–15 (Va. 2013) (noting that in failure to warn cases, as well as in products liability cases, removal of the defect requirement could allow plaintiffs to attribute any generalized danger to a manufacturer without any showing of defect in that product). × In essence, plaintiffs would no longer bear the burden of making a showing of product defect.

Further, much of an individual consumer’s expectations about the way a vehicle should perform in an accident scenario are shaped by the behavior of other drivers. 184 184. See Michael Sivak & Brandon Schoettle, Road Safety with Self-Driving Vehicles: General Limitations and Road Sharing with Conventional Vehicles 5 (Univ. of Mich. Trans. Res. Inst. Report No. 2015-2, 2015). × Without the traditional feedback from other drivers to which consumers are accustomed, these expectations are wholly lacking to describe how autonomous vehicle technology will perform in an accident situation. 185 185. See id. × As noted in a report issued by the University of Michigan’s Transportation Research Institute, “[t]he degree of importance of both driver expectations and feedback from other drivers, and the consequent effects on the safety of a traffic system containing both conventional and self-driving vehicles, remain to be ascertained.” 186 186. See id. (emphasis added). ×

V.        Policy Reasons for Not Applying the Consumer Expectations Test to Autonomous Vehicle Technology.

As noted, autonomous vehicle technology has the potential to decrease traffic injuries and deaths. 187 187. See, e.g., Mothers Against Drunk Driving, supra note 79. × By applying the consumer expectations test, in which unknowledgeable consumers are not required to take into account the utility of a product, or the possibility of a feasible alternative design, courts could expose manufacturers to significant uncertainty in product liability litigation. If the standard by which a product will be judged is on the unpredictable expectations of consumers in such a complex and changing technology, rather than by demonstration of the product’s utility, the threshold for deployment by a manufacturer may change:

Thus, even though an autonomous vehicle may be safer overall than a conventional vehicle, it will shift the responsibility for accidents, and hence liability, from drivers to manufacturers. The shift will push the manufacturer away from the socially optimal outcome—to develop the autonomous vehicle. 188 188. See Gary E. Marchant & Rachel A. Lindor, The Coming Collision Between Autonomous Vehicles and the Liability System, 52 Santa Clara L. Rev. 1321, 1334 (2012). ×

To the contrary, under a risk-utility analysis, particularly one that requires proof of a safer, practicable alternative design, automotive manufacturers will be able to show that the societal benefits from the use of HAV technology as opposed to other technologies outweigh the risk of individual malfunctions in individual cases. 189 189. See id. ×

Consider the following example that illustrates the possible effect of unbridled consumer expectations on the introduction of beneficial new technology:

Suppose . . . that a particular type of “autobrake” crash-avoidance technology works to prevent crashes 80 percent of the time. The other 20 percent of the time, however, the technology does not work and the crash occurs as it would have in the absence of the technology. Victims in those crashes may sue the manufacturer and argue that the product was defective because it failed to operate properly in their crashes. Under existing liability doctrine, they have a plausible argument: The product did not work as designed . . . . A manufacturer facing the decision whether to employ such a technology in its vehicles might very well decide not to, purely on the basis of expected liability costs. 190 190. James M. Anderson et al., RAND Corp. Autonomous Vehicle Technology: A Guide for Policymakers 125 (2016). ×

Without any balancing of the utility of these vehicles or the requirement of a reasonable alternative design, it would be possible, even reasonable, for juries applying the consumer expectations test to find defective design every time. This will be especially true in situations such as those involving self-driving vehicle technology, since consumers tend to have unrealistic expectations about the benefits of this new technology as a whole. 191 191. See id. at 125. ×

Further, the consumer expectations test will not allow for consideration of non-safety related societal benefits that HAV technology provides, since the only consideration will be on whether the product performed as expected in that one instance. The average American commuter spends about one week of his or her life in traffic each year—a statistic that HAV manufacturers have set their sights on reducing. 192 192. See General Motors, supra note 2, at 3; see also David Schrank, Bill Eisele, Et Al., The Texas A&M Transportation Institute & INRIX, 2015 Urban Mobility Scorecard 1–2 (2015), https://static.tti.tamu.edu/tti.tamu.edu/documents/mo bility-scorecard-2015.pdf (noting that, as of 2014, the American commuter spends an average of approximately 42 hours per years in traffic). × The potential time saved by commuters on the whole is not a factor that would be considered under the consumer expectations test.

Another benefit of HAV technology outside of the realm of safety is the potential for added mobility for those who cannot currently drive. 193 193. See generally Waymo, supra note 2, at 6. × According to a report from NHTSA, 3 million Americans are blind or suffer from poor vision. 194 194. NHTSA, DOT HS 811 304, Quieter Cars and the Safety of Blind Pedestrians: Phase I 6 (2010), https://www.nhtsa.gov/DOT/NHTSA/NVS/Crash%20Avoidance/Technical%20Publications/2010/811304rev.pdf. × Further, 79 percent of Americans over the age of 65 live in car-dependent communities. The independence these communities could gain with the widespread use of HAV technology would be yet another consideration the jury could not take into account when utilizing the consumer expectations test.

VI.       Conclusion

Courts should reject the consumer expectations test as grounds for determining design defect in cases involving autonomous vehicle technology. This technology is simply too complex and unfamiliar for consumer expectations to have developed enough to have any real meaning or reasonable application. Utilization of risk-utility balancing is a more appropriate means of establishing whether or not a design is defective and will encourage manufacturers to continue to develop and implement this important technology, which stands to have a truly revolutionary impact on automotive safety.


  John Isaac Southerland is a partner at Huie Fernambucq & Stewart LLP in Birmingham, Alabama. Mr. Southerland’s practice areas include automotive product liability, personal injury, heavy equipment product liability, trucking litigation, and towing and recovery liability. He also serves as national coordinating discovery counsel for a major automotive client. Mr. Southerland is a frequent lecturer at various industry conferences and has written and spoken about the emergence of highly automated vehicles and technology on numerous occasions.  He also serves as a Barrister and the Programs Chairperson in the James Edwin Horton Inn of Court at Cumberland School of Law. 

Elizabeth Davis is an associate at Huie, Fernambucq & Stewart LLP in Birmingham, Alabama. Ms. Davis concentrates her law practice in the areas of automotive litigation, product liability and discovery practice and procedure, including serving as national coordinating discovery counsel for a major automotive client. Ms. Davis is an active member of Alabama Defense Lawyers Association, Birmingham Bar Association, and Defense Research Institute.

Woods Parker is an associate at Huie, Fernambucq & Stewart LLP in Birmingham, Alabama. Mr. Parker concentrates his law practice in the areas of automotive litigation, product liability, trucking litigation, consumer lemon law, and discovery practice and procedure, including serving as national coordinating discovery counsel for a major automotive client. Mr. Parker is an active member of Alabama Defense Lawyers Association, Birmingham Bar Association, and Defense Research Institute.

As we move towards a future of fully automated vehicles, the types of crime – and attendant need for criminal enforcement – committed with cars is likely to evolve. As our transit system becomes more automated, the danger of a hack, and the difficulty of discovering the crime through ordinary policing tactics, is likely to increase. Some experts have expressed concerns that automated vehicles would be just as easy to use for delivery of drugs or guns as for more innocuous packages. Others, such as Duke University professor Mary Cummings, say that vehicles are too easy to hack and steer off course.

Going beyond relatively ordinary crimes such as theft, an unclassified
FBI report
obtained by The Guardian revealed the agency’s concern that
autonomous vehicles could be commandeered and utilized as a “potential lethal
weapon” or even self-driving bomb.

The likelihood that automated vehicles will generally obey
the traffic laws complicates the ability of police to find crimes being
committed with these vehicles using traditional methods. As I have written
previously
, traffic stops prompted by minor violations are a point of
contact at which cops often look for evidence of more serious crime. While
there is some hope that a reduction in such stops may reduce racial bias in
policing, it also highlights the need for law enforcement to reduce dependence
on this method of tracking serious crime.

While the potential for criminal activity or even terrorism using automated vehicles is a real possibility, some experts are less concerned. Arthur Rizer, from the conservative think tank R Street Institute, argued that the lives saved by adoption of driverless technology will far outweigh any risk of criminal or terror threat from a hacking. Rizer calls the risk “minute compared to the lives that we will save just from reducing traffic accidents.”

If a significant portion of the roughly 40,000
traffic fatalities
per year can be prevented by the adoption of automated
vehicles, Rizer is likely correct that the benefits will outweigh any risk that
vehicles will be hacked by bad actors. Nevertheless, there is a possibility
that, as CalTech professor Patrick Lin warns, automated vehicles “may enable
new crimes that we can’t even imagine today.” Going forward, it will be
important for law enforcement to develop new techniques of tracking crime
facilitated by automated vehicles.

For the past several months, this blog has primarily focused on new legal questions that will be raised by connected and automated vehicles. This new transportation technology will undoubtedly raise novel concerns around tort liability, traffic stops, and city design. Along with raising novel problems, CAVs will also add new urgency to longstanding legal challenges. In some ways, this is best encapsulated in the field of privacy and data management.

In recent decades, the need to understand where our data goes has increased exponentially. The smartphones that most of us carry around every day are already capable of tracking our location, and recording a lot of our personal information. In addition to this computer/data generation machine in our pockets, the CAV will be a supercomputer on wheels, predicted to generate 4,000 gigabytes of data per day. Human driven vehicles with some automated features, such as Tesla’s with the company’s “Autopilot” functionality, already collect vast amounts of user data. Tesla’s website notes that the company may access a user’s browsing history, navigation history, and radio listening history, for example.

In response to this growing concern, California recently passed a sweeping new digital privacy law, set to take effect in 2020. Nicknamed “GDPR-Lite” after the European Union’s General Data Protection Regulation, California’s law “grants consumers the right to know what information companies are collecting about them, why they are collecting that data and with whom they are sharing it.” It also requires companies to delete data about a customer upon request, and mandates that companies provide the same quality and cost of service to users who opt out of data collection as those who opt in.

In comparison to the GDPR, California’s law is relatively limited in scope. The California Consumer Privacy Act (CCPA) is tailored to apply only to businesses that are relatively large or that are primarily engaged in the business of collecting and selling personal data. Furthermore, CCPA contains few limitations on what a business can do internally with data it collects. Instead, it focuses on the sale of that data to third parties.

In many ways, it remains too early to evaluate the effectiveness of California’s approach. This is in part because the law does not take effect until the beginning of next year. The bill also enables the California Attorney General to issue guidance and regulations fleshing out the requirements of the bill. These as-yet-unknown regulations will play a major role in how CCPA operates in practice.

Regardless of its uncertainties and potential shortcomings though, CCPA is likely to play a significant role in the future of American data privacy law and policy. It is the first significant privacy legislation in the US to respond to the recent tech boom, and it comes out of a state that is the world’s fifth largest economy. CCPA’s implementation will undoubtedly provide important lessons for both other states and the federal government as they consider the future of data privacy.

With roughly a clip a month – most of these being corporate fluff – Waymo’s YouTube channel is not the most exciting nor informative one. At least, those (like me) who keep looking for clues about Waymo’s whereabouts should not expect anything to come out of there.

That was until February 20th, when Waymo low-key published a 15 second clip of their car in action – the main screen showing a rendering of what the car “sees” and the corner thumbnail showing the view from the dash cam. The key point: Waymo’s car apparently crosses a broken-lights, police-controlled intersection without any hurdle. Amazing! Should we conclude that level 5 is at our very doorsteps?

The car and tech press was quick to spot this one, and reports were mostly praise. Yet Brad Templeton, in his piece for Forbes pinpoints at a few things that the clip does not say. First, we have the fact that Waymo operates in a geographically-enclosed area, where the streets, sidewalk and other hard infrastructure (lights, signs, and probably lines) are pre-mapped and already loaded in the algorithm. In other words, Waymo’s car does not discover stuff as it cruises along the streets of Northern California. Moreover, the street lights here do not work and so technically, this is just another four-way stop-signed intersection, with the difference that it is rather busy and there is a traffic police directing traffic in the middle. Finally, the car just goes straight, which is by far the easiest option (no left turn, for example…)

Beyond that, what Waymo alleges and wants us to see, is that car “recognizes” the policeman, or at the very least, recognizes that there is something person-shaped standing in the middle of the intersection and making certain gestures at the car, and that the car’s sensors and Waymo’s algorithms are now at the level of being able to understand hand signals of law enforcement officers.

Now I heard, less than a year ago, the CEO of a major player in the industry assert that such a thing was impossible – in reference to CAVs being able to detect and correctly interpret hand signals cyclists sometime use. It seems that a few months later, we’re there. Or are we? One issue which flew more or less under the radar, is how exactly does the car recognize the LEO here? Would a random passerby playing traffic cop have the same effect? If so, is that what we want?

As a member of the “Connected and Automated Vehicles: Preparing for a Mixed Fleet Future” Problem Solving Initiative class held at the University of Michigan Law School last semester, my team and I have had the opportunity to think about just that – how to make sure that road interactions stay as close as possible as they are today – and conversely how to foreclose awkward interactions or possible abuses that “new ways to communicate” would add. Should a simple hand motion be able to “command” a CAV? While such a question cuts across many domains, our perspective was a mostly legal one and our conclusion was that any new signal that CAV technology enables (from the perspective of pedestrians and other road users) should be non-mandatory and limited to enabling mutual understanding of intentions without affecting the behavior of the CAV. Now what we see in this video is the opposite; seemingly, the traffic police person is not equipped with special beacons that broadcast some form of “law enforcement” signal, and it is implied – although, unconfirmed – that there is no human intervention. We are left awed, maybe, but reassured? Maybe not.

The takeaway may be just this: the issues raised by this video are real ones, and are issues Waymo, and others, will at some point have to address publicly. Secrecy may be good for business, but only so much. Engagement by key industry players is of the highest importance, if we want to foster trust and avoid having the CAV technology crash land in our societies.

Earlier this month, the Journal of Law and Mobility hosted our first annual conference at the University of Michigan Law School. The event provided a great opportunity to convene some of the top minds working at the intersection of law and automated vehicles. What struck me most about the conference, put on by an organization dedicated to Law and mobility, was how few of the big questions related to automated vehicles are actually legal questions at this point in their development.

The afternoon panel on whether AVs should always follow the rules of the road as written was emblematic of this juxtaposition. The panel nominally focused on whether AVs should follow traffic laws. Should an automated vehicle be capable of running a red light, or swerving across a double yellow line while driving down the street? Should it always obey the posted speed limit?

The knee-jerk reaction of most people would probably be something along the lines of, “of course you shouldn’t program a car that can break the law.” After all, human drivers are supposed to follow the law. So why should an automated vehicle, which is programmed in advance by a human making a sober, conscious choice, be allowed to do any differently?

Once you scratch the surface though, the question becomes much more nuanced. Human drivers break the law in all kinds of minor ways in order to maintain safety, or in response to the circumstances of the moment. A human driver will run a red light if there is no cross-traffic and the car bearing down from behind is showing no signs of slowing down. A human will drive into the wrong lane or onto the shoulder to avoid a downed tree branch, or a child rushing out into the street. A human driver may speed away if they notice a car near them acting erratically. All of these actions, although they violate the law, may be taken in the interest of safety in the right circumstances. Even knowing they violated the law, a human driver who was ticketed in such a circumstance would feel their legal consequence was unjustified.

If automated vehicles should be able to break the law in at least some circumstances, the question shifts – which circumstances? Answering that question is beyond the scope of this post. At the moment, I don’t think anyone has the right answer. Instead, the point of this post is to highlight the type of moment-to-moment decisions every driver makes every day to keep themselves and those around them safe. The rules of the road provide a rough cut, codifying what will be best for most people most of the time. They could not possibly anticipate every situation and create a special legal rule for that situation. If they tried, the traffic laws would quickly grow to fill several libraries.

In my view, the question of whether an AV should be able to break the law is only tangentially a legal question. After arriving at an answer of, “probably sometimes,” the question quickly shifts to when, and in what circumstances, and whether the law needs to adapt to make different maneuvers legal. These questions have legal aspects to them, but they are also moral and ethical questions weighted with a full range of human driving experience.  Answering them will be among the most important and difficult challenges for the AV industry in the coming years.

The “Trolley Problem” has been buzzing around for a while now, so much that it became the subject of large empirical studies which aimed at finding a solution to it that be as close to “our values” as possible, as more casually the subject of an episode of The Good Place.

Could it be, however, that the trolley problem isn’t one? In a recent article, the EU Observer, an investigative not-for-profit outlet based in Brussels, slashed at the European Commission for its “tunnel vision” with regards to CAVs and how it seems to embrace the benefits of this technological and social change without an ounce of doubt or skepticism. While there are certainly things to be worried about when it comes to CAV deployment (see previous posts from this very blog by fellow bloggers here and here) the famed trolley might not be one of those.

The trolley problem seeks to illustrate one of the choices that a self-driving algorithm must – allegedly – make. Faced with a situation where the only alternative to kill is to kill, the trolley problem asks the question of who is to be killed: the young? The old? The pedestrian? The foreigner? Those who put forward the trolley problem usually do so in order to show that as humans, we are forced with morally untenable alternative when coding algorithms, like deciding who is to be saved in an unavoidable crash.

The trolley problem is not a problem, however, because it makes a number of assumptions – too many. The result is a hypothetical scenario which is simple, almost elegant, but mostly blatantly wrong. One such assumption is the rails. Not necessarily the physical ones, like those of actual trolleys, but the ones on which the whole problem is cast. CAVs are not on rails, in any sense of the word, and their algorithms will include the opportunity to go “off-rails” when needed – like get on the shoulder or on the sidewalk. The rules of the road incorporate a certain amount of flexibility already, and such flexibilities will be built in the algorithm.

Moreover, the very purpose of the constant sensor input processed by the driving algorithm is precisely to avoid putting the CAV in such a situation where the only options that remain are collision or collision.

But what if? What if a collision is truly unavoidable? Even then, it is highly misleading to portray CAV algorithm design as a job where one has to incorporate a piece of code specific to every single decision to be made in the course of driving. The CAV will never be faced with an input of the type we all-too-often present the trolley problem: go left and kill this old woman, go right and kill this baby. The driving algorithm will certainly not understand the situation as one where it would kill someone; it may understand that a collision is imminent and that multiple paths are closed. What would it do, then? Break, I guess, and steer to try to avoid a collision, like the rest of us would do.

Maybe what the trolley problem truly reveals is the idea that we are uneasy with automated cars causing accidents – that is, they being machines, we are much more comfortable with the idea that they will be perfect and will be coded so that no accident may ever happen. If, as a first milestone, CAVs are as safe as human drivers, that would certainly be a great scientific achievement. I do recognize however that it might not be enough for the public perception, but that speaks more of our relationship to machines than to any truth behind the murderous trolley. All in all, it is unfortunate that such a problem continues to keep brains busy while there are more tangible problems (such as what to do with all those batteries) which deserve research, media attention and political action.

The common story of automated vehicle safety is that by eliminating human error from the driving equation, cars will act more predictably, fewer crashes will occur, and lives will be saved. That future is still uncertain though. Questions still remain about whether CAVs will truly be safer drivers than humans in practice, and for whom they will be safer. In the remainder of this post, I will address this “for whom” question.

A recent study from Benjamin Wilson, Judy Hoffman, and Jamie Morgenstern at Georgia Tech found that state-of-the-art object detection systems – the type used in autonomous vehicles – demonstrate higher error rates in detection of darker-skinned pedestrians as compared to lighter-skinned pedestrians. Controlling for things like time-of-day or obstructed views, the technology was five percentage points less accurate at detecting people with darker skin-tones.

The Georgia Tech study is far from the first report of algorithmic bias. In 2015, Google found itself at the center of controversy when its algorithm for Google Photos incorrectly classified some black people as gorillas. More than two years later, Google’s temporary fix of removing the label “gorilla” from the program entirely was still in place. The company says they are working on a long-term fix to their facial recognition software. However, the continued presence of the temporary solution several years after the initial firestorm is some indication either of the difficulty of achieving a real solution or the lack of any serious coordinated response across the tech industry.

Algorithmic bias is a serious problem that must be tackled with a serious investment of resources across the industry. In the case of autonomous vehicles, the problem could be literally life and death. The potential for bias in automated systems begs for an answer to serious moral and legal questions. If a car is safer overall, but more likely to run over a black or brown pedestrian than a white one, should that car be allowed on the road? What is the safety baseline against which such a vehicle should be judged? Is the standard, “The AV should be just as likely (hopefully not very likely) to hit any given pedestrian?” Or is it “The AV should hit any given pedestrian less often than a human driven vehicle would?” Given our knowledge of algorithmic bias, should an automaker be opened up to more damages if their vehicle hits a black or brown pedestrian than when it hits a white pedestrian? Do tort law claims, like design defect or negligence, provide adequate incentive for automakers to address algorithmic bias in their systems? Or should the government set up a uniform system of regulation and testing around the detection of algorithmic bias in autonomous vehicles and other advanced, potentially dangerous technologies?

These are questions that I cannot answer today. But as the Georgia Tech study and the Google Photos scandal demonstrate, they are questions that the AV industry, government, and society as a whole will need to address in the coming years.

In the coming decades, advancing technology is likely to strain many tried-and-true legal concepts.  The tort law cause of action for design defects is likely to be among the most impacted. This post will explore the current understanding of design defect claims, and highlight areas where autonomous vehicles and other highly complex technologies will likely lead to a rethinking of the doctrine.

As outlined in the Third Restatement of Torts, design defect claims can be brought against a manufacturer when “the foreseeable risks of harm posed by the product could have been reduced or avoided by the adoption of a reasonable alternative design . . . and the omission of the alternative design renders the product not reasonably safe.” Essentially, plaintiffs who bring a design defect claim after being harmed by a product bear the burden of showing that the product was designed unreasonably for its intended use, and that an alternative design would have been safer for the user and reasonable for the manufacturer to adopt.

Traditionally, courts have adopted one of two tests to determine the reasonableness of a product design under these claims. Under the consumer expectations test, the key question is whether a product performed up to the level at which an ordinary consumer would expect. Because this test is based on the expectations of an ordinary consumer (one who is presumed to not have any special knowledge about the product), a claim can be successful without any expert testimony about the alleged design failure, as in McCabe v. American Honda Motor Corp.

Alternatively, many courts have adopted the risk-utility test for proving a design defect. The risk-utility test is more akin to cost-benefit balancing. The South Carolina Supreme Court in Branham v. Ford Motor Co. noted that the risk-utility test balances “numerous factors . . . including the usefulness and desirability of the product, the cost involved for added safety, the likelihood and potential seriousness of injury, and the obviousness of danger.” As this test requires testimony on cost of the current design and any proposed alternatives, it will require expert testimony and specialized knowledge, unlike the consumer expectations test.

Some have made the case that the consumer expectations test will be inadequate to address claims of design defect in complex technologies such as autonomous vehicles. After all, the argument goes, how could an ordinary consumer possibly have a realistic expectation of how an autonomous vehicle is supposed to perform in a given situation? Any given action by an AV is the result of a series of algorithms that is being constantly updated as the car gathers new information about the world around it. Should a consumer expect the AV to act just as a human would act? Should it be more cautious? Or perhaps even take actions that would seem overly risky for a human driver, because the AV system was certain of what every step in its maneuver would look like going in? How could a human passenger know? If courts are persuaded by these concerns, they will likely need to address them by adopting the more expert-reliant risk-utility test.

On the other hand, some scholars argue that the consumer expectations test is perfectly adequate to handle claims involving advanced technology such as AVs. In a recent article, NYU Law Professor Mark Geistfeld notes that consumers need not understand the intricacies of how a technology works in order to have “well-formed expectations of the product performance.” Under Geistfeld’s approach, a consumer either should have such a well-formed expectation or, in the case where they have yet to develop one, should be warned by the manufacturer or dealer in such a way as to make them aware of the risk they are taking on.

It remains to be seen how design defect claims will be forced to evolve as autonomous vehicles come on the scene. Like many areas of law though, this is a field that will be stressed, and potentially forced to evolve, by the advent of this revolutionary technology.

The European Parliament, the deliberative institution of the European Union which also acts as a legislator in certain circumstances, approved on February 20, 2019 the European Commission’s proposal for a new Regulation on motor vehicle safety. The proposal is now set to move to the next step of the EU legislative process; once enacted, an EU Regulation is directly applicable in the law of the 28 (soon to be 27) member states.

This regulation is noteworthy as it means to pave the way for Level 3 and Level 4 vehicles, by obligating car makers to integrate certain “advanced safety features” in their new cars, such as driver attention warnings, emergency braking and a lane-departure warning system. If many of us are familiar with such features which are already found in many recent cars, one may wonder how this would facilitate the deployment of Level 3 or even Level 4 cars. The intention of the European legislator is not outright obvious, but a more careful reading of the legislative proposal reveals that the aim goes much beyond the safety features themselves: “mandating advanced safety features for vehicles . . .  will help the drivers to gradually get accustomed to the new features and will enhance public trust and acceptance in the transition toward autonomous driving.” Looking further at the proposal reveals that another concern is the changing mobility landscape in general, with “more cyclists and pedestrians [and] an aging society.” Against this backdrop, there is a perceived need for legislation, as road safety metrics have at best stalled, and are even on the decline in certain parts of Europe.

In addition, Advanced Emergency Braking (AEB) systems have been trending at the transnational level, in these early months on 2019. The World Forum for Harmonization of Vehicle Regulations (known as WP.29) has recently put forward a draft resolution on such systems, in view of standardizing them and making them mandatory for the WP.29 members, which includes most Eurasian countries, along with a handful of Asia-Pacific and African countries. While the World Forum is hosted by the United Nations Economic Commission for Europe (UNECE,) a regional commission of the Economic and Social Council (ECOSOC) of the UN, it notably does not include among its members certain UNECE member states such as the United States or Canada, which have so far refused to partake in World Forum. To be sure, the North American absence (along with that of China and India, for example) is not new; they have never partaken in the World Forum’s work since it started its operations in 1958. If the small yellow front corner lights one sees on US cars is not something you will ever see on any car circulating on the roads of a W.29 member state, one may wonder if the level of complexity involved in designing CAV systems will not forcibly push OEMs toward harmonization; it is one thing to live with having to manufacture different types of traffic lights, and it is another one to design and manufacture different CAV systems for different parts of the world.

Yet it is well known that certain North American regulators are not a big fan of such approach. In 2016, the US DoT proudly announced an industry commitment of almost all car makers to implement AEB systems in their cars, with the only requirement that such systems satisfy set safety objectives. If it seems like everyone would agree that limited aims are sometimes the best way to get closer to the ultimate, bigger goal, the regulating style varies. In the end, one must face the fact that by 2020, AEB systems will be harmonized for a substantial part of the global car market, and maybe, will be so in a de facto manner even in North America. And given that the World Forum has received a received a clear mandate from the EU – renewed as recently as May 2018 – to develop a global and comprehensive CAV standard, North American and other Asian governments who have so far declined to join the W.29 might only lose an opportunity to influence the outcome of such CAV standards by sticking to their guns.

Cite as: Kyle D. Logue, The Deterrence Case for Comprehensive Automaker Enterprise Liability, 2019 J. L. & Mob. 1.

I.         Introduction

Automobiles are much safer today than they used to be. Perhaps the best illustration of this fact is the decades’ long decline in the number of auto-related deaths per-mile-driven. 195 195. General Statistics: Fatality Facts, INS. INST. FOR HIGHWAY SAFETY: HIGHWAY LOSS DATA INST., https://www.iihs.org/iihs/topics/t/general-statistics/fatalityfacts/over view-of-fatality-facts (last visited Nov. 13, 2018) (showing motor vehicle crash deaths per 100 million miles driven has declined from 3.35 in 1976 to 1.18 in 2016). × And yet motor vehicles—including cars, trucks, and SUVs— continue to be among the most dangerous products sold anywhere. Automobiles pose a larger risk of accidental death than any other product, except perhaps for opioids. 196 196. In 2016, which is the most recent year for which the Centers for Disease Control has final data as of the time of this writing, there were 58,335 deaths attributable to accidental poisoning, which includes accidental deaths from drug overdose (which, of course, includes accidental opioid overdose). JIAQUAN XU ET AL., CTRS. FOR DISEASE CONTROL & PREVENTION, NATIONAL VITAL STATISTICS REPORTS 34 (vol. 67, no. 5 July 26, 2018) [henceforth, CDC, 2016 Final Death data]. The CDC estimates that roughly 42,000 deaths in 2016 were attributable to opioids, the vast majority of which would presumably be considered accidental deaths. Drug Overdose Death Data, CTRS. FOR DISEASE CONTROL & PREVENTION, https://www.cdc.gov/drugoverdose/data/statedeaths .html (last visited Nov. 13, 2018). Note also that firearms are involved in more deaths per year than motor vehicles, but the vast majority of those deaths are caused intentionally, either suicide (22,938 deaths in 2016) or homicide (14,415). The other leading causes of accidental deaths that year included the following: falls (34,673), firearms (495), and drowning (3,786). CDC, 2016 Final Death data, at 50. × Annual auto-crash deaths in the United States have never fallen below 30,000, reaching a recent peak of roughly 40,000 in 2016. 197 197. For 2016, there were 40,327 motor-vehicle-related accidental deaths. CDC, 2016 Final Death data, at 52. × In addition to these tens of thousands (internationally, millions 198 198. Internationally, the number of annual fatalities attributable to motor vehicle accidents is in the millions. Number of Road Traffic Deaths, WORLD HEALTH ORG., http://www.who.int/gho/road_safety/mortality/number_text/en/ (last visited Nov. 13, 2018) (estimating worldwide auto accident deaths in 2013 to be roughly 1.25 million). × ) of deaths attributable to motor-vehicle crashes, there are many other social costs. Victims of serious auto accidents, for example, often incur extraordinary medical expenses both to provide treatment immediately after the accident and, sometimes, to provide treatment for the rest of their lives. 199 199. LAWRENCE BLINCOE ET AL., NAT’L HIGHWAY TRAFFIC SAFETY ADMIN., THE ECONOMIC AND SOCIETAL IMPACT OF MOTOR VEHICLE CRASHES, 2010 (REVISED) 5 (May 2015) (finding medical costs responsible for $23.4 billion of the total economic cost of motor vehicle crashes in 2010). × Those crash victims whose injuries render them unable to work can experience weeks, months, even years of lost income, which, from their employers’ perspective, is lost productivity. 200 200. Id. (finding $77.4 billion in lost productivity as a result of motor vehicle crashes in 2010). × Auto accidents also cause non-trivial amounts of property damage, mostly to the automobiles themselves though also occasionally to highways, bridges, or other elements of transportation infrastructure. Finally, serious motor vehicle accidents often cause severe noneconomic injuries—that is, severe “pain and suffering”—as a result of accident victims’ painful and debilitating physical injuries. According to some estimates, such noneconomic harms, in the aggregate, amount to more than twice the magnitude of the aggregate economic damages caused by auto accidents. 201 201. DANIEL SMITH, NAT’L HIGHWAY TRAFFIC SAFETY ADMIN., OVERVIEW OF NHTSA PRIORITY PLAN FOR VEHICLE SAFETY AND FUEL ECONOMY, 2015 TO 2017 at 2 (June 2015) (“In addition to the terrible personal toll, these crashes have a huge economic impact on our society with an estimated annual cost of $242 billion, which is an average of $784 for every person in the United States. These crashes also result in $594 billion in societal harm from loss of life and the pain and decreased quality of life due to injuries.”). ×

All of this may be about to change. According to many auto-industry experts, the eventual transition to driverless vehicles will drastically lower the economic and noneconomic costs of auto accidents. 202 202. See, e.g., Adrienne LaFrance, Self-Driving Cars Could Save 300,000 Lives Per Decade in America, THE ATLANTIC (Sept. 29, 2015), https://www.theatlantic.com/ technology/archive/2015/09/self-driving-cars-could-save-300000-lives-per-decade-in-america/407956/ (“Researchers estimate that driverless cars could, by midcentury, reduce traffic fatalities by up to 90 percent.”). × Why might this be so? Because humans are so bad at driving. When it comes to operating motor vehicles, people have bad judgment, slow reflexes, inadequate skills, and short attention spans. They drive too fast. They drive while intoxicated. They drive while sleepy. They drive while distracted. In fact, according to the National Highway Traffic Safety Administration, roughly 94 percent of auto accidents today are attributable to “driver error.” 203 203. NAT’L HIGHWAY TRAFFIC SAFETY ADMIN., CRITICAL REASONS FOR CRASHES INVESTIGATED IN THE NATIONAL MOTOR VEHICLE CRASH CAUSATION SURVEY 2 (Mar. 2018), https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812506. Another 2 percent of accidents are attributable to vehicle component failure, 2 percent to environmental conditions (such as slick roads), and 2 percent to “unknown.” Id. What precisely these statistics mean, however, is not entirely clear. Specifically, it is not obvious how NHTSA’s statistical categories (such as “vehicle component failure”) would relate to analogous legal concepts (such as “defective product”). × Computers can do better. At least that is the hope: that machine-learning computer algorithms, in combination with state-of-the-art sensors and advanced robotics, will be better—much better—drivers than humans are. 204 204. For an extended argument for why driverless cars are better drivers than humans, see, e.g., HOD LIPSON & MELBA KURMAN, DRIVERLESS: INTELLIGENT CARS AND THE ROAD AHEAD (MIT Press 2016). × Whether this will in fact be true is still unproven, but is most likely to be true with respect to so-called fully driverless “Level 5” vehicles, 205 205. See infra note 12. Level 5 vehicles are also sometimes called autonomous vehicles, though that use of the term ignores the distinction between connected vehicles and truly autonomous vehicles. This Article will larger ignore that distinction as well. × which are those autonomous or connected vehicles that are capable of operating on any road and under any conditions that a human driver can handle but with no input from a human passenger other than the choice of destination. 206 206. It is these Level 5 vehicles that hold the real promise for substantial accident-risk reduction. SAE INTERNATIONAL, TAXONOMY AND DEFINITIONS FOR TERMS RELATED TO DRIVING AUTOMATION SYSTEMS FOR ON-ROAD MOTOR VEHICLES J3016 (2018), https://saemobilus.sae.org/content/j3016_201806 (explaining levels 0, no autonomous features, through level 5, where a computer is “operating the vehicle on-road anywhere that a typically skilled human driver can reasonably operate a conventional vehicle”). This is in large part because Level 5 vehicles do not have the “handoff problem,” which occurs at that moment (with levels 1 through 4) when control of the vehicle must be transferred from the algorithm to the human driver. Alex Davies, The Very Human Problem Blocking the Path to Self-Driving Cars, WIRED (Jan. 1, 2017), https://www.wired.com/2017/01/human-problem-blocking-path-self-driving-cars/ (discussing how existence of handoff problem led Google, in 2012, and other companies more recently, to commit to developing level 5 autonomy). × Level 5 vehicles, because they would not suffer from the problems that plague human decision making in the driving context, do hold the promise to be substantially safer than the fully or even partially human-driven alternative. 207 207. One can certainly imagine the possibility of driving algorithms going haywire or sensors failing in ways that cause terrible accidents. But presumably, regulators will not permit Level 5s to be sold until they prove themselves in large numbers of test miles to be substantially safer than human drivers. Some commentators have suggested that regulators not approve Level 5s unless and until they are shown to be twice as safe as human drivers. Mark A. Geistfeld, A Roadmap for Autonomous Vehicles: State Tort Liability, Automobile Insurance, and Federal Safety Regulation, 105 CALIF. L. REV. 1611, 1653 (2017). ×

As promising as a world of highways filled with computer-driven vehicles might be, from an accident-reduction perspective, 208 208. If the advent of autonomous and connected vehicles means more vehicles on the road, it could be bad news for efforts to combat climate change and improve air quality. × such a high-tech world is still only a possibility. And even if it happens, it will not be for a number of years. There continue to be major technological hurdles, as well as potential consumer resistance to actually riding in a driverless vehicle. 209 209. American Drivers Grow More Afraid of Driverless Vehicles, INS. J. (May 22, 2018), https://www.insurancejournal.com/news/national/2018/05/22/490014.htm (noting that 73% of American drivers report being too afraid to ride in a fully automated vehicle). × Therefore, the introduction, spread, and eventual dominance of Level 5s will take some time. 210 210. Many experts think consumers will not even be able to purchase fully autonomous vehicles for another decade. Justin Gerdes, Not So Fast. Fully Autonomous Vehicles are More than a Decade Away, Experts Say, GREEN TECH MEDIA (Feb. 6, 2018), https://www.greentechmedia.com/articles/read/fully-autonomous-vehicles-decade-away-experts (reporting results of informal poll of 300 industry experts). × During that transition, most automobiles will continue to be driven mostly by humans. Indeed, even in the long run, when Level 5 vehicles have been perfected and are available to the general public either through individual purchases and leases or through some ride-sharing arrangement (via Uber or Lyft or some similar web-based platform), we should still expect to see a substantial number of fully or partially human-driven vehicles traveling alongside them. 211 211. See, e.g., Background On: Self-Driving Cars and Insurance, INS. INFO. INST. (July 30, 2018), https://www.iii.org/article/background-on-self-driving-cars-and-insurance (“According to the Insurance Institute for Highway Safety, it is anticipated that there will be 3.5 million self-driving vehicles on U.S. roads by 2025, and 4.5 million by 2030. However, the institute cautioned that these vehicles would not be fully autonomous, but would operate autonomously under certain conditions.”). ×

If I am right about this picture of the automotive future, what should the role of auto tort law be, now and going forward? More specifically, if we conceive of auto tort law—including both automaker product liability and driver negligence liability (and the insurance that covers both types of liability)—as a system of ex post auto-crash deterrence, what would the optimal or efficient auto tort/insurance regime look like? 212 212. For this Article, I assume that the primary role of auto tort law is efficient deterrence. That means, creating incentives that induce all relevant parties—drivers, automakers, even pedestrians—to take efficient or cost-justified steps to minimize the probability and severity of accidents. On this view, the goal is not necessarily zero accidents, because the cost of accident avoidance eventually renders additional investments in accident prevention inefficient and socially undesirable. This is a standard type of normative analysis of accident law. It is, of course, not the only way to evaluate an accident law regime. For example, if the primary function of auto tort law were instead merely compensation for the harms caused by auto accidents, or were to achieve corrective justice (in the sense of reversing wrongfully caused harms), some system other than the one proposed in this Article might make more sense. × Further, how should such an optimally designed auto tort/insurance regime take into account the emergence of Level 5 vehicles?

These questions are the subject of this Article. Specifically, this Article lays out the potential (at this point purely theoretical) deterrence benefits of replacing our current auto tort regime (including auto products liability law, driver-based negligence claims, and auto no-fault regimes) with a single, comprehensive automaker enterprise liability system. 213 213. The term “enterprise liability” has long been used to stand for the idea that “business enterprises ought to be responsible for losses resulting from products they introduce into society.” George L. Priest, The Invention of Enterprise Liability: A Political History of The Intellectual Foundations of Modern Tort Law, 14 J. LEGAL STUD. 461, 463 (1985) (describing intellectual history of enterprise liability idea). See also Gregory C. Keating, The Theory of Enterprise Liability and Common Law Strict Liability, 54 VAND. L. REV. 1285, 1287 (2001) (“[E]nterprise liability expresses the maxim that those who profit from the imposition of risk should bear the costs of the accidents that are a price of their profits.”). The concept of enterprise liability was much discussed in the 1980s and 1990s among tort scholars. See, e.g., Priest, supra; James A. Henderson Jr. The Boundary Problems of Enterprise Liability, 41 MD L. REV. 659 (1982) (discussing line-drawing issues that arise in connection with adopting enterprise liability regimes); and Kenneth S. Abraham & Paul C. Weiler, Enterprise Medical Liability and the Evolution of the American Health Care System, 108 HARV. L. REV. 381 (1994) (applying enterprise liability concepts to medical system). I, together with my colleagues and friends Jon Hanson and Steve Croley, started writing about enterprise liability around this time. See, e.g., Jon D. Hanson & Kyle D. Logue, The First-Party Insurance Externality: An Economic Justification for Enterprise Liability, 76 CORNELL L. REV. 129 (1990); and Steven P. Croley & Jon D. Hanson, Rescuing the Revolution: The Revived Case for Enterprise Liability, 91 MICH. L. REV. 683 (1993). × This new regime would apply not only to Level 5 vehicles, but to all automobiles made and sold to be driven on public roads. 214 214. For a more recent proposal to create a special auto-manufacturer responsibility regime, which has similarities to the one I am describing here, but that—critically—would be limited to accidents involving fully automated vehicles, see Kenneth S. Abraham & Robert L. Rabin, Automated Vehicles and Manufacturer Responsibility for Accidents: A New Legal Regime for a New Era, 105 VA. L. REV. (forthcoming 2019). × Because such a system would make automakers unconditionally responsible for the economic losses resulting from any crashes of their vehicles, it would in effect make automakers into auto insurers as well, although such a change will likely lead to some restructuring in how automobiles are insured and sold. Or so I will argue.

My basic argument is that a comprehensive automaker enterprise liability regime may have previously unexplored, or at least forgotten, deterrence benefits. 215 215. In an article published in 1985, Professor Howard Latin outlined an automaker enterprise liability proposal similar to the one I am describing in this Article. Howard A. Latin, Problem-Solving Behavior and Theories of Tort Liability, 73 CAL. L. REV. 677 (1985). That article, brought to my attention by Professor Stephen Sugarman, also makes a deterrence case for adopting an automaker enterprise liability regime, emphasizing some (though not all) of the same arguments I make here. See also Bryant Walker Smith, Regulation and the Risk of Inaction, in AUTONOMOUS DRIVING 584 (Markus Maurer, et al. eds., 2016), https://link.springer.com/chapter/10.1007/978-3-662-48847-8_27 (exploring (very briefly) the idea of using auto enterprise liability as a means of encouraging automotive safety innovation). × First, it could greatly simplify our existing auto tort regime by replacing all of automaker liability law (including product design defect claims) and driver liability law (as well as existing no-fault regime) with a single enterprise liability regime under which all auto-accident victims could seek recovery. Second, it could encourage automakers to design and manufacturer safer vehicles, whether that means safer human-driven vehicles (with automated features) or Level 5 vehicles. Third, it could incentivize automakers to provide better warnings and instructions with their vehicles, including better ways to deal with the “hand off” problem that occurs when vehicles switch from semi-self-driving mode to human-driven mode. 216 216. Steven Ashley, Level 3 “Hand Off” is Challenging AI Researchers, SAE INT’L (Jan. 17, 2017), https://www.sae.org/news/2017/01/sae-level-3-hand-off-is-challenging-ai-researchers. × Fourth, enterprise liability could result in automobile prices that better reflect the actual costs of driving, leading to more optimal levels of auto sales and miles driven. Fifth, enterprise liability could induce auto companies to coordinate (in a way they are not presently coordinating) with the one industry that has more information than the auto companies have about how the specific driving patterns of individual human drivers affect the risk of auto accidents: namely, the auto insurance industry. Finally, a comprehensive automaker enterprise liability regime would provide an implicit subsidy for the development and deployment of driverless technology, but only to the extent that automakers actually expect such technology to reduce accident costs. All of these points will be developed below.

The argument will proceed as follows. Part II evaluates existing auto tort law—including automaker liability law and driver liability law—from the perspective of optimal deterrence. Part III outlines one plausible version of a comprehensive automaker enterprise liability regime and summarizes the primary deterrence advantages of such a regime. Part IV briefly concludes with a discussion of caveats, concerns, and a list of questions for future research.

II.         Evaluating the Deterrence Implications of Current Auto Tort Law

Automaker Liability Law

To understand the deterrence benefits of an auto enterprise liability regime, it is necessary first to understand the deterrence consequences of the current auto tort regime. To that end, this Part describes the current auto tort system—both automaker liability law and driver liability law—and, drawing on well-known insights from deterrence theory and economic analysis of liability rules, explores what the general deterrence consequences of that regime might be. This is an entirely theoretical discussion. The ultimate question—which auto tort regime comes closes to minimizing the costs of auto accidents—can of course only be answered with empirical research that is beyond the scope of this short paper.

Current automaker liability law, like manufacturer liability law generally, is primarily a negligence-based regime, by which I mean the following: Under current law in most U.S. jurisdictions, individuals who suffer harm caused in an automobile crash can recover from the automaker in tort if they can prove that the harm resulted from negligence (or a lack of reasonable care) on the part of the automaker in designing or constructing the vehicle. 217 217. See, e.g., Larsen v. GM, 391 F.2d 495, 504 (1968) (holding, among other things, that auto manufacturers have a duty to use reasonable care in design and construction of vehicles). × Alternatively, auto accident victims can invoke modern products liability doctrine and argue that a “defect” in the vehicle’s design, manufacturing process, or warnings caused the harm. 218 218. RESTATEMENT (THIRD) TORTS: PROD. LIAB. §§1 & 2 (AM. LAW INST. 1998) (setting forth general rules for liability resulting from product defects). × This latter approach also typically requires some showing of automaker negligence. This is because, in the bulk of U.S. jurisdictions, important aspects of the product defect law are equivalent to negligence law. 219 219. A majority of jurisdictions apply a risk-utility version of the design defect test, which is similar to the common cost-benefit formulations of negligence. RESTATEMENT (THIRD) TORTS: PROD. LIAB., §2, cmt. D (AM. LAW. INST. 1998). Moreover, the adequacy of product warnings is often evaluated according to a negligence-based reasonableness test. RESTATEMENT (THIRD) TORTS: LIABILITY FOR PHYSICAL AND EMOTIONAL HARM, § 3 (AM. LAW. INST. 1998). Also, in the jurisdictions that define design defect according to “reasonable consumer expectations,” there is an obvious reliance on negligence-based principles as well, such as the concept of reasonableness. Current products liability law with respect to warnings is also essentially a negligence-based regime. RESTATEMENT (THIRD) TORTS: PROD. LIAB., §2, cmt. I (AM. LAW. INST. 1998). (“Commercial product sellers must provide reasonable instructions and warnings about risks of injury posed by products.”). × For this reason, auto products-liability, despite sometimes being labeled a form of “strict liability,” 220 220. For an example of lawyers characterizing auto products liability generally, including design defect and warning defect claims, as a form of “strict liability,” see, e.g., DEREK H. SWANSON & LIN WEI, MCGUIREWOODS, UNITED STATES AUTOMOTIVE PRODUCTS LIABILITY LAW 7 (2009), https://www.mcguirewoods.com/news-resources/publications/us-automotive-products-liability.pdf. × is in fact largely a form of negligence liability. 221 221. I am of course not the first person to observe that modern “strict” products liability operates in practice largely as a negligence regime. See, e.g., DAVID G. OWEN, PRODUCTS LIABILITY LAW 38 (3rd ed. 2015) (“These two propositions—that manufacturers must guard against risks only if they are foreseeable, and that manufacturers must guard against those risks only with precautions that are reasonable—are the two major pillars of modern products liability law in America.”). This is not to say that there are no aspects of strict liability in the current auto products liability system. For example, manufacturing defect cases approximate true strict liability. That is, when the product’s design satisfies the risk-utility standard (or is, in a sense, reasonable or non-negligent) but the particular product that caused the harm in the case did so as a result of some sort of malfunction in the vehicle that is not a result of negligent maintenance on the part of the vehicle owner (e.g., the brakes or the steering mechanism simply fails), the automaker is strictly liable. With respect to Level 5 vehicles, presumably a much higher percentage of the accidents would be a result of vehicular malfunction than is the case with human-drive vehicles; thus, there would be a much larger domain of true strict liability if existing product liability doctrine were applied to Level 5 accidents than is currently the case with automaker liability cases. Further, in design defect jurisdictions that apply a consumer expectations test, strict liability also seems likely, assuming courts conclude that consumers reasonably expect Level 5s not to crash. Still, there would be some negligence-based liability with respect to the design of the vehicles and the algorithm that drives them. See, e.g., Bryant Walker Smith, Automated Driving and Product Liability, 2017 MICH. ST. L. REV. 1 (2017) (noting that, applying existing law, automaker liability for Level 5 accidents will likely turn on some version of “unreasonable performance” by the vehicle, which sometimes will approximate strict liability and sometimes negligence); and Mark A. Geistfeld, A Roadmap for Autonomous Vehicles: State Tort Liability, Automobile Insurance, and Federal Safety Regulation, 105 CAL. L. REV. (forthcoming 2018) (observing that applying existing products liability law to Level 5s will sometimes result in strict liability and sometimes negligence-based liability). ×

A negligence-based automaker liability regime can in theory have certain deterrence advantages, if one makes particular assumptions. Those assumptions, however, are keys to the analysis—and do not always apply. For starters, a negligence-based automaker liability regime can create efficient incentives with respect to automaker care levels. Automaker “care levels” are the precautions taken by automakers—in the design, production, and warnings with respect to their vehicles—that reduce the probability or severity of auto accidents. 222 222. See generally STEVEN SHAVELL, ECONOMIC ANALYSIS OF ACCIDENT LAW 73–85 (1987) (comparing theoretical deterrence benefits of negligence liability rule and strict liability rule, in terms of care levels and activity levels); WILLIAM M. LANDES & RICHARD A. POSNER, THE ECONOMIC STRUCTURE OF TORT LAW 54–84 (1987) (same); and A. MITCHELL POLINSKY, AN INTRODUCTION TO LAW AND ECONOMICS 113–123 (3rd ed. 2003) (discussing choice of optimal products liability rules in particular). × Efficient automaker care levels occur when the automaker has made all available investments in care—in crash-risk reduction—that reduce expected auto-accident costs by more than the marginal costs of the additional care. 223 223. Put differently, when an automaker is investing efficiently in care, there are no additional investments in accident reduction that could be made that would reduce expected accident costs by more than the costs of accident avoidance. × Thus, an efficient negligence-based tort liability rule would hold an automaker liable for the harms resulting from a given auto accident only if that automaker failed to take efficient care. For example, if there was an alternative automotive design or alternative warning that the automaker could have used that would have reduced expected accident costs by more than the marginal costs of that design or warning change, failing to deploy that alternative design or warning in their vehicles would constitute negligence on the part of the automaker, and would therefore be potential grounds for tort liability. 224 224. The plaintiff must also demonstrate causation. ×

This sort of efficient negligence-based liability rule would induce automakers to take efficient care if we assume the following to be true: (a) that automakers are aware of the law and respond rationally to it, and (b) that courts applying a negligence-based automaker liability rules perform a thorough and accurate cost-benefit analysis (for example, judges and juries do not tend to make systematic errors in their determinations regarding what constitutes automaker negligence or what counts as a design defect). Under those assumptions, the negligence-based regime would incentivize efficient automaker care levels. Why? Because automakers would under those assumptions realize that they can avoid negligence-based liability entirely if they merely make all cost-justified investments in auto safety (e.g., all cost-justified design and warning changes). Knowing this, they would have a strong legal and financial incentives to do just that. 225 225. This insight is simply an application of a standard conclusion regarding the effects on injurer care levels of a perfectly applied negligence rule. See generally supra sources cited in note 28. ×

In addition, a negligence-based automaker liability regime can also create incentives for efficient driver care-levels—incentives for drivers to drive reasonably carefully—even in the absence of a defense of contributory negligence or comparative fault. 226 226. See generally supra sources cited in note 28. × This is because a negligence-based regime, by its nature, leaves accident costs on victims and their insurers when the automaker is not negligent. That fact will induce drivers to drive carefully, so as to minimize their own risk of uncompensated accident losses. Again, however, this conclusion holds only if certain key assumptions are also true. Specifically, we must assume the following: a) that drivers, like automakers, are knowledgeable about tort law and respond rationally to the potential of tort rules to apply to their future conduct; and b) that drivers actually bear these costs and do not externalize them to someone else.

To put all of this together, according to standard deterrence theory, an efficiently and accurately applied negligence-based automaker liability rule can produce efficient incentives for both automakers and drivers to take care to avoid auto accidents. 227 227. See generally supra sources cited in note 28. This conclusion also assumes that automakers are well informed about and respond rationally to tort liability rules. ×

But there are obvious problems with this rosy picture. First, consider the effects on automaker care levels if we relax the assumption that courts accurately apply negligence-based standards. If judges and juries are not very good at doing the complex and information-intensive analysis necessary to determine what particular automotive designs, warnings, or instructions are cost-justified or reasonable (or not defective), the outcomes of courts’ negligence determinations become highly uncertain. This can in turn produce incentives for automakers both to over-invest and to under-invest in auto safety. 228 228. See generally Mark Grady, A New Positive Economic Theory of Negligence, 92 Yale L.J. 799 (1983); and John Calfee & Richard Craswell, Some Effects of Uncertainty on Compliance with Legal Standards, 70 Va. L. Rev. 965, 982 (1984). ×

The incentive to over-invest in auto safety can arise when manufacturers expect courts to set the standard of reasonable care (or a non-defective design) inefficiently high—that is, when manufacturers expect that courts may find a design defect notwithstanding the fact that the automaker’s design decisions were consistent with an accurate, objective, comprehensive risk-utility test. If that is the expectation, then automakers would have an incentive to satisfy the inefficiently high court- or jury-imposed design standard (or warning standard) in order to avoid liability. The incentive to under-invest in safety can arise if courts rely too much on custom within the industry as their source for what constitutes reasonable care, or a non-defective design or warning. This is because industry custom can (famously) lag behind what is truly efficient levels of safety. 229 229. The T.J. Hooper, 60 F.2d 737 (2d Cir. 1937) (“a whole calling may have unduly lagged in the adoption of new and available devices”). In product liability design defect cases, of course, courts do not generally permit compliance with industry custom to be totally exculpatory; however, it can be considered relevant to the risk-utility negligence-based balancing test. See, e.g., Carter v. Massey-Ferguson, Inc., 716 F.2d 344 (5th Cir. 1983). However, if a defendant in a negligence-based product liability regime has adopted a design that is the safest in use at the time of manufacturing, it may be difficult for the plaintiff to prevail. See RESTATEMENT (THIRD) TORTS: PROD. LIAB., §2, cmt. D (AM. LAW. INST. 1998). × It is not a surprise, then, that commentators have argued that custom-based standards of care, like those that currently apply to automaker liability, can inhibit innovation. 230 230. See, e.g., Gideon Parchomovsky & Alex Stein, Tort and Innovation, 107 Mich. L. Rev. 285 (2008). ×

A second problem with a negligence-based auto products liability regime has to do with driver care levels. For a negligence-based regime to efficiently incentivize drivers to drive carefully (by imposing on drivers the risk of accidents that are not cost-justifiably preventable by the manufacturer), recall that we assumed that drivers are well informed of both accident risks and how those risks are allocated according to the specific rules of auto tort law. Those assumptions are obviously unrealistic. Drivers simply are not aware of the tort law rules that apply to them or the product liability rules that apply to automakers. Moreover, even when drivers do know about accident risks and legal rules, there are reasons to believe (discussed below) 231 231. See infra, discussion at notes 44–48 and 51. × either that drivers will not respond rationally to that information or that they will externalize those risks to insurance companies. If I am right about that—about drivers’ lack of information about driving risk and auto tort law, and about their cognitive biases and cost-externalization—then the ability of a negligence-based auto products liability regime to optimize driver care levels is substantially undermined. Legally imposing costs on drivers would not, or at least may not, have the desired deterrence effect on driver care levels. 232 232. By contrast, the assumption that automobile manufacturers—with their teams of expert engineers, lawyers, and accountants—are fully informed of the torts liability regime in which they operate and how those rules are likely to affect them. This claim—that auto manufacturers are likely to be better informed (both about the risks of auto accidents and about the relevant liability rules) and more classically rational in their decision-making than drivers—is not new. See, e.g., Latin, supra note 21, at 692–93 (arguing that, because drivers are much less likely to know about and respond rationally to having auto-accident losses imposed on them than auto manufacturers are, a regime of auto-manufacturer enterprise liability could produce an overall improvement in social welfare, through a reduction in overall auto accidents). ×

The final deterrence problem with a negligence-based auto products liability regime would exist even if judges and juries were good (accurate and unbiased) at applying risk-utility or cost-benefit standards. In fact, this problem results because automakers would expect accurate application of the negligence-based rules. The problem involves the effect of a negligence-based automaker liability rule on the number of vehicles sold, or, in the language of deterrence, the effect on automaker “activity levels.” 233 233. See supra sources cited in note 28. × Even an efficiently safe car (one with no defects whatsoever) that is driven carefully by its human or algorithmic driver poses some residual or irreducible risk of crashing. This residual risk will have a tendency to be ignored or externalized by automakers under a negligence-based product liability regime because automakers can virtually insulate themselves against liability by merely complying with the liability standard. 234 234. See supra sources cited in note 28. Of course, this automaker activity level effect is mitigated to some extent when automakers except to be held liable by courts despite having taken reasonable care. These effects are unlikely, however, to be perfectly offsetting. × The result of this externality is that the scale of operation in the auto industry—the number of cars sold—may be higher than the social-welfare maximizing level, even ignoring the effect of automobile emissions on the environment, because the price of vehicles does not include this cost of unpreventable auto accidents.

To summarize, given how our current negligence-based automaker liability regime is applied in practice, there are reasons to be concerned that automaker and driver care levels may be too low and activity levels too high. What’s more, this concern would apply not only to human-driven vehicles, but to Level 5 vehicles as well. That is, there is nothing about the nature of Level 5 vehicles that would suggest these problems are less likely to be present than would be the case for human-driven vehicles. 235 235. With Level 5 vehicles, if there were a problem with “driver care levels,” it would be a problem automaker care levels. That is, with Level 5s, driver care levels are, by definition, included as a part of manufacturer care level. × This activity-level inefficiency associated with current automaker liability law has been totally ignored by those who have argued in favor of applying existing product liability standards, or revised but still negligence-based versions of existing product liability standards, to Level 5 vehicles.

Driver Liability Law

In a majority of states in the U.S., if someone is injured or suffers property damage as a result of a driver’s negligent operation of an automobile, rather than as a result of automaker negligence, the victim may recover from the negligent driver under standard common-law principles of tort. 236 236. For a summary of emergence of fault-based and no-fault auto liability/insurance systems, see JAMES M. ANDERSON, PAUL HEATON, SEPHEN J. CARROLL, THE U.S. EXPERIENCE WITH NO-FAULT AUTOMOBILE INSURANCE: A RETROSPECTIVE 19–61 (2010), https://www.rand.org/content/dam/rand/pubs/monographs/2010/RAND_MG8 60.pdf. × The victim must demonstrate that the harm to her was a result of the driver’s failure to do something that a reasonable driver would have done under the circumstances, or the drivers’ doing something that a reasonable driver under the circumstances would not have done. 237 237. RESTATEMENT (SECOND) OF TORTS § 284 (defining negligent conduct in terms of what reasonably prudent person would do or not do); see also RESTATEMENT (THIRD) OF TORTS: LIABILITY FOR PHYSICAL AND EMOTIONAL HARM, § 3 cmt. h (Am. Law. Inst. 1998) (“Many cases say that negligence consists of “the failure to do something which a reasonably careful person would do, [or] the doing of something which a reasonably careful person would not do.”“) (citations omitted). × Accident victims who can recover include pedestrians, cyclists, passengers, or other drivers—anyone who is harmed as a result of driver negligence.

Because driver liability is also a negligence-based regime, it has similar potential to provide efficient deterrence as does a negligence-based automaker liability regime. Specifically, negligence-based driver liability law can have beneficial deterrence effects on driver care levels, if we make the following assumptions:

  • Drivers are well informed about accident risks (and how their behavioral changes affect those accident risks),
  • Drivers are well-informed about the rules of tort law,
  • Drivers internalize those risks (do not externalize them to insurers, for example), and
  • Drivers process the information about those risks rationally (without any systematic cognitive biases), and we assume again that
  • Courts are good at applying cost-benefit-type negligence-based liability rules.

If all of those assumptions are true, then, for the same reason that automakers would be incentivized by a negligence-based automaker liability regime, drivers too would be incentivized to drive with efficient care—in terms of driving speed, safe braking and passing practices, smart-phone usage (or non-usage), and the like. This is so because, by taking efficient care in driving, drivers would avoid liability for the accidents that nevertheless occur. Again, under a negligence-based regime, driving with efficient care can be seen as a type of insurance for drivers, a fact that—if all of the above-listed assumptions are true—would incentivize safe driving.

The reasons that this vision of negligence-based driver liability law do not describe reality should be clear at this point. The assumptions listed above on which the analysis depends almost certainly do not hold in the real world. While drivers may be generally aware of the broad outlines of the driver liability regime in their state (whether it is fault-based or no-fault), they likely do not understand what the precise implications of that fact are on their chances of being found liable in court for unsafe driving. What’s more, the average driver, while generally and vaguely cognizant of the risks of driving, is almost certainly uneducated about the precise levels of risk associated with various aspects of driving—for example, precisely how much the chance of a crash is increased by texting while driving or changing lanes abruptly with no signal. In fact, there is a good chance that most drivers underestimate those risks.

Why would drivers tend to underestimate such risks? First, there is the long list of well-documented cognitive biases that affect how individuals process information generally. 238 238. The sources here are many. A decent place to start would be the classic essay by Amos Tversky & Daniel Kahneman, Judgment Under Uncertainty: Heuristics and Biases, in JUDGMENT UNDER UNCERTAINTY. HEURISTICS AND BIASES 3 (Daniel Kahneman, Paul Slovic, & Amos Tversky eds., 1982). More recent summaries of the literature include DANIEL KAHNEMAN, THINKING, FAST AND SLOW (2011); and RICHARD H. THALER, MISBEHAVING: THE MAKING OF BEHAVIORIAL ECONOMICS (2015). A classic article summarizing the application of behavioral insights to law and economics is Christine Jolls, Cass R. Sunstein & Richard H. Thaler, A Behavioral Approach to Law and Economics, 50 STAN. L. REV. 1471 (1998). × One famous example is the tendency of individuals to ignore the risk of very low probability events and underestimate the likelihood of some high probability events. 239 239. For a summary of the relative behavioral literature as it relates to products liability, see Jon D. Hanson & Douglas A. Kysar, Taking Behavioralism Seriously: the Problem of Market Manipulation, 74 N.Y.U. L. REV. 630, 643–87 (1999). For early applications of behavioral insights to products liability law, see Latin, supra note 21; and Howard A. Latin, “Good” Warnings, Bad Products, and Cognitive Limitations, 41 UCLA L. REV. 1193, 1194–95 (1994). For a discussion of the findings on very low probability and high probability events, see Hanson & Kysar, supra note 45, at 716–20. Note that there is also research showing that consumers sometimes overestimate the risks of merely “low probability” events—those that fall between very low probability and high probability. Id. (discussing research summarized in Enterprise Responsibility for Personal Injury, 1 A.L.I. 230 (1991)). × Auto-crash risks may similarly be ignored or underestimated. 240 240. Whether auto risks are more likely to be very low, merely low, or high probability events is not entirely clear. However, the most important insight of the Hanson & Kysar article is that, because product manufacturers—including automakers–have considerable influence over how consumers perceive the risk of their products (and because product manufacturers—including automakers–have a strong market incentive to ensure that consumers underestimate the risks of their products), there is every likelihood that consumers on balance underestimate the risks of auto accidents. Id. Moreover, Hanson & Kysar, in a follow up article, provide considerable anecdotal evidence of actual market manipulation of consumer risk perceptions by manufacturers. Jon D. Hanson & Douglas A. Kysar, Taking Behavioralism Seriously: Some Evidence of Market Manipulation, 112 HARV. L. REV. 1420, 1466 (1999). × Also, drivers are especially prone to overestimating their own driving ability and thus their own ability to avoid crashes. 241 241. This finding has proven robust over many years. See, e.g., Ola Svenson, Are We All Less Risky and More Skillful Than Our Fellow Drivers?, 47 ACTA PSYCHOLOGICA 143 (1981); Timo Lajunen & Heikki Summala, Driving experience, personality, and skill and safety-motive dimensions in drivers’ self-assessments 19 PERSONALITY AND INDIVIDUAL DIFFERENCES. 307 (1995); A.F. Williams, Views of US drivers about driving safety 34 J. SAFETY RES. 491 (2003). See also Jolls et al., supra, at 1537–38 (discussing problem of “overoptimisim” among drivers); Hanson & Kysar, Taking Behavioralism Seriously: A Response to Market Manipulation, 6 ROGER WILLIAMS L. REV. 259, 354–55 (1999) (same). × Moreover, drivers not only underestimate their own likelihood of a crash relative to the average driver (which they do), they also overestimate their own likelihood of a crash relative to the actual probability. 242 242. Christine Jolls, Behavioral Economics Analysis of Redistributive Legal Rules, 51 VAND. L. REV. 1653, 1660 (citing Richard J. Arnould & Henry Grabowski, Auto Safety Regulation: An Analysis of Market Failure, 12 BELL J. ECON. 27, 34–35 (1981) and Colin F. Camerer & Howard Kunreuther, Decision Processes for Low Probability Events: Policy Implications, 8 J. POLY. ANALYSIS & MGMT. 565, 566 (1989)). × For all of these reasons, a negligence-based driver liability regime, which relies on assumptions of informed and rational drivers to produce optimal driver care levels, may not produce the deterrence benefits that are predicted by deterrence theory. 243 243. The evidence on whether negligence-based driver liability law reduces auto accidents, or the harms resulting from auto accidents, is probably best characterized as inconclusive. See generally Nora Freeman Engstrom, An Alternative Explanation of No-Fault’s Demise, 61 DEPAUL L. REV. 303, 332–333 (2012) (“[R]oughly half of the studies published thus far claim that no-fault coverage increases fatal accidents, while the other half find no effect, and the notion that no-fault reduces fatalities has been seemingly put to rest.”) (footnotes omitted). The hope that a shift away from a negligence-based driver-liability regime would not substantially reduce auto accidents, along with the overall desire to lower auto insurance rates, was one of the original justifications for the movement towards auto no-fault regimes in the 1970s and 1980s. Id. × In addition, it is commonly argued that drivers have many powerful incentives to drive carefully even in the absence of a negligence-based regime that left on them the uninsured costs of auto accidents, incentives such as the desire to avoid a traffic fines or, more importantly, a crash that could be painful or even fatal to them or their loved ones. 244 244. Latin, supra note 21, at 690–91; Engstrom, supra note 49, at 330. ×

How does this pessimistic picture of driver liability law as a system of incentivizing good driving change if we introduce auto insurance? The answer to that question turns out to be complicated. On one hand, automobile insurance has the potential to correct some of these deterrence-related problems. 245 245. See generally Omri Ben-Shahar & Kyle D. Logue, Outsourcing Regulation: How Insurance Reduces Moral Hazard, 111 MICH. L. REV. 197 (2012) (discussing ways that insurance companies help insureds reduce risk). × Here’s why. Auto insurers are, unlike most drivers, extremely well informed about the intricacies of accident law. They employ teams of lawyers whose job is to understand how driver liability laws in each state affect the liability risks of their customers. Indeed, their profitability and their survival as going concerns depend on this expert understanding of the auto liability laws of all sorts. In addition, auto insurers have unparalleled access to enormous amounts of detailed information regarding the crash-risk characteristics of millions of drivers and automobiles. This is the result of decades of experience providing auto insurance coverage to hundreds of millions of drivers and vehicles, which in turn means pricing millions of auto insurance policies and adjusting millions of auto-crash claims over the years. No other institution or organization would have the same amount of driver-specific, automobile-specific data, as would the auto insurance industry.

In addition, recent innovations in “telematics” (which combines telecommunications, data science, and automotive technology) have increased auto insurers’ ability to gather and analyze risk-relevant driver and vehicle data. 246 246. Background On: Pay-As-You-Drive Auto Insurance (Telematics), INS. INFO. INST., https://www.iii.org/article/background-on-pay-as-you-drive-auto-insurance-tele matics (last visited Nov. 21, 2018). × With this new and emerging technology, not only do insurers have access to information regarding how drivers’ past auto-claims and traffic-ticket histories affect their riskiness as drivers; they also have the ability to gather information on the effects of a range of specific driving behaviors on auto-crash risks. 247 247. Id. See also Ben-Shahar & Logue, supra note 51. × For example, a number of insurers currently gather information about drivers’ braking, acceleration, speeding, turning, and cornering behaviors and then send that information back to the insurers for analysis. 248 248. Yuanjing Yao, Evolution of Insurance: A Telematics-Based Personal Auto Insurance Study, U. CONN. HONORS SCHOLAR THESES, 590, 598 (2018), https://opencommons.uconn.edu/srhonors_theses/590/. × Once this driver-specific data is combined with data gather by insurers and others (including NHTSA) about what factors cause auto accidents generally, it becomes possible for auto insurers to link specific driving behaviors of particular drivers with premium discounts. 249 249. Id. ×

All of this information is to varying degrees already being taken into account by many auto insurance companies in the pricing of their insurance policies. For example, policy discounts are offered to drivers with good safety records 250 250. Most auto insurers give discounts for being accident free for a given period of time. Car Insurance Discounts, VALUEPENGUIN, https://www.valuepenguin.com/car-insurance-discounts (last visited Nov. 21, 2018). See also Yao, supra note 54 (discussing use of behavioral driving discounts among insurers). × as well as for vehicles with particular safety features. 251 251. One survey of the leading car insurers, found the following additional vehicle-safety-related discounts: passive restraint (25% to 30%), new car (10%), daytime running lights (around 3%). Id. Some insurers are starting to offer discounts for semi-autonomous features such as adaptive cruise control, collision avoidance systems, and lane departure warnings. Cherise Threewitt, What Car Insurance Discounts Can I Get?, U.S. NEWS & WORLD REP. (June 29, 2018), https://cars.usnews.com/cars-trucks/car-insurance/car-insurance-discounts. × In addition, insurers are now offering discounts if drivers will improve their driving ability—for example, if they will take defensive driving classes. 252 252. Car Insurance Discounts, supra note 56 (reporting insurers giving discounts of 10% to 15% for completion of defensive driving courses). × Because of telematics revolution, auto insurers are even able to adjust premiums on the basis of the specific driving behavior of individual drivers. For example, some insurers give discounts for a range of driver-care-level factors such as wearing seatbelts, driving at moderate speeds, limiting late night trips, and avoiding aggressive braking. 253 253. Id. See also Barbara Marquand, Comparing Drivewise, Snapshot and Other Usage-Based Insurance Plans, NERDWALLET (Feb. 8, 2016), https://www. nerdwallet.com/blog/insurance/comparing-drivewise-snapshot-usage-based-insurance/. × Also, the advances in telematics have made “pay as you go” auto insurance, under which premiums are a function of the number of miles driven, more accurate—and thus more prevalent—than ever before. 254 254. Usage-Based Insurance and Telematics, NAT’L ASS’N OF INS. COMMISSIONERS (July 26, 2018), https://www.naic.org/cipr_topics/topic_usage_based_insurance.htm. × Driving-behavior-sensitive auto insurance premiums—which could take into account both good and bad driving choices (i.e., driver care levels) and, critically, the number of miles driven (i.e., driver activity levels)—hold the promise of incentivizing risk-reducing driving behavior in a way that even the most sophisticated government regulator could not hope to do. 255 255. See generally Ben-Shahar & Logue, supra note 51 (discussing potential risk reducing benefits of high-tech auto-insurance pricing); Hanson & Logue, supra note 19, at 192–93 (suggesting reasons why auto insurance is better at risk-segregating than other types of first-party insurance). What little empirical research has been done on the subject tends to confirm that incentive-based insurance pricing tends to alter driving behavior in a risk-reducing direction. See Mark Stevenson et al., The effects of feedback and incentive-based insurance on driving behaviours: study approach and protocols, 24 INJ. PREVENTION, 89, 93 n. 27–30 (2018), https://injuryprevention.bmj.com/content/24/1/89; see also Telematics Helps Reduce collisions and Claims, AUTOMOTIVE FLEET (Nov. 10, 2017), https://www.automotive-fleet.com/157806/telematics-helps-reduce-collisions-and-claims. ×

But here is the problem: Under current law and given existing market conditions, auto insurers do not have strong incentives to make full use of their comparative advantage at gathering risk-relevant information and pricing their insurance on the basis of that information, or at least there is reason to be concerned about their incentives to do so. The reason for concern is that the amount of coverage currently being provided by auto insurers presently represents only a fraction (in many cases a small fraction) of the total risks of auto crashes. This is true of first-party auto insurance coverage, which tends to cover only a fraction of the accident risks that any driver faces. 256 256. This assertion requires some explanation. There are no good studies on this particular question. So my claim is derived from circumstantial evidence of a sort. First, note that, whereas all states require some amount of liability insurance coverage for anyone who drives on public roadways, only a small minority of states require drivers to purchase auto insurance that provides any sort of first-party medical or disability benefits. See, e.g., Background on: Compulsory Auto/Uninsured Motorists, INS. INFO. INST. (April 16, 2018), https://www.iii.org/article/background-on-compulsory-auto-uninsured-motorists. What’s more, even when there is mandated medical or disability coverage, the amount of required coverage is almost always far less than would potentially be recoverable under an auto tort claim, whether it be an automaker liability claim or a driver liability claim, and far less than the potentially enormous total costs (in the millions) of any given auto accident. Id. (for example, noting mandatory amounts of bodily injury liability coverage ranging from $15,000 to $50,000 per person). Karl Eisenhower, Personal Injury Protection: How PIP Insurance Works in Your State, WALLET HUB (Jan. 9, 2015), https://wallethub.com/edu/pip-insurance/9248/ (noting PIP mandates ranging from $2000 per person in Utah to $50,000 in New York.) The only state that requires unlimited PIP coverage is Michigan. Id. Further, even when drivers do purchase first-party medical or disability coverage through their auto-insurance policy, that coverage is often secondary to the victims’ other forms of first-party health or disability insurance. For example, in some states requiring PIP coverage in auto policies, the insured can elect to make auto PIP coverage secondary to other first-party health and disability insurance. This is sometimes called the “coordination” option. MICH. DEPART. INS. & FIN. SERV., YOUR GUIDE TO AUTOMOBILE INSURANCE: FOR MICHIGAN CONSUMERS 10 (Sept. 2017), https://www.michigan.gov/documents/ difs/Auto_Insurance_Guide_448003_7.pdf. Because making auto health insurance secondary lowers the insured’s auto insurance premiums, and has little or no effect on her first-party health and disability insurance premiums, most insureds choose the coordination option, which means most insureds choose to make their non-auto first-party insurers primary. In sum, most auto health and disability risks end up being borne by non-auto first-party insurers—such as health insurers and disability insurers. × It is also true of auto liability coverage, owing in part to the fact that the mandatory minimum amounts in most states are far less than the maximum harm threatened by an auto accident that results in even one serious injury or death. 257 257. State mandated minimums for liability coverage for personal injuries to a single person range from a low of $10,000 (Florida) to a high of $50,000 (Alaska and Maine), and by for the most common minimum is $25,000. Car Insurance Laws by State, FINDLAW, https://injury.findlaw.com/car-accidents/car-insurance-laws-by-state.html (last visited Nov. 19, 2018) (gathering links to state laws). A single accident resulting in serious bodily injury or death could easily produce economic losses alone in excess of $1 million. × As a result, many of the costs of auto accidents are currently being externalized to non-auto first-party health and disability insurers who—unlike auto insurers in the telematics age—do not tailor premiums at all based on their insureds’ driving decisions. 258 258. Most first-party health and disability insurers make no effort to price their coverage in a way that reflects the riskiness of insureds’ driving choices—such as how they drive, how much they drive, or even what type of vehicle they drive. See generally Hanson & Logue, supra note 19 (using fact that most non-auto first-party insurers do not price-differentiate on basis of consumer product use to argue for enterprise liability for product accident risks). There is a perfectly sensible reason for this fact: the risks of auto-related health or disability claims are only a small fraction of the overall health and disability risks covered by any given first-party health or disability insurer. It is not worth the insurers’ while to tailor their insurance premiums on the basis of any particular behavioral choice of their insureds, other than perhaps the choice to smoke or not. The result of all this: that portion of auto crash risks that are ultimately born by non-auto first-party insurers get externalized (or largely ignored) by drivers, with obvious deterrence consequences. × Moreover, to the extent auto insurers do attempt to charge individualized, behaviorally- and risk-adjusted auto insurance rates (which, as I noted above, they are increasingly trying to do), this incentive is undermined by the fact that auto insurers cover only a fraction of the risks of auto accidents. 259 259. Because auto insurers do not bear all of the risks of auto accidents, the premium discounts they are willing to offer to induce safer driving habits may not be adequate. The point can be illustrated with a simple example. Suppose there was some investment in driver care that cost $50 but would reduce expected accident costs by $80. Say it would reduce a chance of a $200,000 loss from .001 to .0006. If the auto insurer bore the full $200,000 risk, it would have an incentive to offer a premium discount to cover the cost of driver care, with an additional discount perhaps up to a total just short of the $80 saved by the additional investment in driver care. But what if the auto-insurer bore only, say, $40,000 of the $200,000 potential loss? Then the largest discount it could offer without losing money would be $16, which would be the amount of the savings in going from a .001 to a .0006 risk of, now, $40,000 in covered losses. But that discount would not be enough to induce the consumer to make the investment in care, assuming the other $160,000 in expected accident cost is externalized either because of drivers’ under-estimation of risk or because of non-adjusting non-auto first-party insurance coverage. ×

It should also be noted, however, that there are important ways in which the allocation of auto-accident risks to non-auto first-party insurers has cost-reducing advantages. This may seem incongruous with the argument in the previous paragraph, but it is not: While auto insurers are in a good position, through premium discounts, to help optimize driver care and activity levels, auto insurers are not necessarily in a good position to minimize some other costs associated with providing insurance benefits. For example, primary health care coverage provided through auto insurance companies is almost certainly much more expensive than primary health care provided through regular non-auto first-party health insurers. This would be because, although auto insurers, in a sense, specialize increasingly in reducing driver ex ante moral hazard, it is non-auto health insurance companies who specialize in reducing ex post medical moral hazard—that is, excessive or wasteful use of the healthcare system. 260 260. The claim that first-party health insurers specialize in trying to hold down health care costs may seem controversial, at least for fee-for-service policies. My claim is only that health insurers—especially ones that use managed care tools—are probably better at holding down ex post health care costs than are auto insurers. This is one of the reasons that auto no-fault regimes which make auto PIP coverage primary over non-auto first-party health coverage are so expensive, and why auto-no-fault regimes have not led to the cost savings that were expected. See Engstrom supra note 49. Professor Engstrom notes that “[m]edical insurers . . . reduce costs via discounts and fee schedules, and the limit patient treatment using any number of mechanism, including deductibles, co-payments, utilization controls, and medical protocols . . . , [a]uto insurers . . . tend to pay almost any bills that a victim incurs . . . dramatically increasing . . . the cost of care. Id. at 341 (citations omitted). × My point here is only that the current division of auto-accident costs, allocating so little to auto insurers, may be non-optimal, given auto insurers potential ability to incentivize better (and less) driving. 261 261. In Part III below, I suggest that, by adopting an automaker enterprise liability regime, automakers will have an incentive to strike the efficient balance between amount of auto-crash costs allocated to auto insurers and amount allocated to non-auto insurers. ×

To summarize, because of drivers’ lack of accident-risk information and understanding of auto tort law and their susceptibility to cognitive biases, and because of the presence of cost-externalizing insurance coverage, there is reason to be doubtful that the current negligence-based auto tort laws—automaker liability laws as well as driver liability laws—work to optimize driver care and activity levels. As discussed in the next Part, the adoption of an auto enterprise liability regime could in theory create incentives for automakers, together with auto insurers, to provide better driver-side incentives, as well as better automaker safety incentives.

III.         The Automaker Enterprise Liability Alternative 262 262. The arguments in this section draw on prior work done by me and a number of other scholars on the deterrence benefits of enterprise liability in various contexts. See, e.g., Hanson & Logue, supra note 19; Steve P. Croley & Jon D. Hanson, Rescuing the Revolution: The Revived Case for Enterprise Liability, 91 MICH. L. REV. 683 (1993); Kyle D. Logue & Jon D. Hanson, The Costs of Cigarettes: The Economic Case for Ex Post Incentive-Based Regulation, 107 YALE L.J. 1163 (1998); and Jon D. Hanson & Douglas A. Kysar, Taking Behavioralism Seriously: Some Evidence of Market Manipulation, 112 YALE L.J. 1420, 1553 (1999). In addition, as mentioned in an earlier footnote, the argument here has some overlap with a proposal made by Howard Latin. See generally Latin, supra note 21 (making a deterrence case of automaker enterprise liability). ×

The Basic Proposal

As an alternative to our current negligence-based auto tort regime, consider the possibility of a comprehensive automaker enterprise liability regime. Under such a regime, anyone who suffers a physical injury or property damage in an automobile accident—whether driver, passenger, or pedestrian—would be legally entitled to recover, from the manufacturer of the vehicle involved, compensation for the losses sustained as result of the accident. 263 263. This could be done through the existing court system or through specialized courts or agencies set up to handle auto-crash disputes. × Thus, to recover under this enterprise liability regime, accident victims would not be required to show negligence on the part of manufacturer or anyone else. Nor would accident victims have to prove that the automobiles, or any of the warnings or instructions accompanying the automobiles, are in anyway defective or unreasonably dangerous. Rather, crash victims would need only to prove that the harms for which they seek compensation “arose out of the use of” a vehicle that was designed and built by the manufacturer from whom compensation is sought. Each automaker, therefore, would be financially responsible for the losses resulting from any crash arising out of the use of that automaker’s vehicles. 264 264. The “arising out of the use of” analysis would replace a causation determination. This phrase is used now in standard auto insurance policies. Thus, an automaker enterprise liability regime would be a particular type of cause-based no-fault compensation regime, modeled after similar programs that have been adopted outside of the auto context, such as workers’ compensation laws at the state level or the vaccine compensation program at the federal level. See generally Jon D. Hanson, Kyle D. Logue & Michael S. Zamore, Smokers’ Compensation: Toward a Blueprint for Federal Regulation of Cigarette Manufacturers, 22 S. ILL. L.J. 519 (1998) (discussing deterrence benefits of cause-based no-fault compensation regimes). In the Workers’ Compensation context, claims against employers are limited to injuries or illness that “arise out of the course of employment.” Professors Abraham and Rabin have proposed a similar regime—that would also use the “arising out of” standard—but that would apply exclusively to accidents involving Level 5 automated vehicles. Rabin & Abraham, supra note 20. My idea is to make such a regime comprehensive, to apply to all motor vehicles, subject to transition rules discussed below. Some scholars have expressed concern about the use of enterprise liability regimes that rely on boundary-maintaining doctrines such as the “arising out of” concept used in workers’ compensation regimes, among other places. See, e.g., Henderson, supra note 19. This is of course a reasonable concern, although the evidence suggests that programs such as workers’ compensation have found relatively effective ways to police the borders of their programs. DON DEWEES, DAVID DUFF & MICHAEL TREBILCOCK, EXPLORING THE DOMAIN OF ACCIDENT LAW: TAKING THE FACTS SERIOUSLY 393-394 (1996) (reporting administrative costs for workers’ compensation regimes that are low relative to those of the tort system). ×

That is the most basic picture of the proposal. Now consider a few possible details of such a program. One important initial question is who exactly would fall within the class of “automakers” to whom the enterprise liability regime would apply. The most obvious class of defendants/payers would be the original equipment manufacturers (OEMs) of the vehicles involved in the crash. They are the ones who generally make the key automotive design choices, have control over the manufacturing processes, and decide on the terms of any warning or instruction manual; and they are also the ones with the greatest expertise on such questions. Auto manufacturers also determine the pricing of their vehicles and the number of them to produce, subject of course to the constraints of supply and demand. Given that manufacturer care levels and activity levels are key auto-accident deterrence variables, making OEMs responsible for the auto-crash costs associated with their vehicles has obvious deterrence benefits, discussed further below.

Liability under an enterprise liability regime, however, would not necessarily be limited to auto manufacturers. Liability could also be extended, on a joint and several basis (or on a several basis), to a range of other enterprises that fall within the design, production, sale, and distribution chain of any given vehicle. 265 265. Thus the concept of “automaker” in an automaker enterprise liability regime could be similar to the concept of a “seller” in existing products liability law. × In most cases, it is likely that the crash victim would bring the claim against the manufacturer, and then the manufacturer would either implead the other parties in the chain of production into that suit or would sue them separately in a contribution action. Precisely how the responsibility for the costs of any accident would be allocated among the various parties on the automaker side of the ledger is beyond the scope of this Article. That allocation of responsibility, however, would presumably be determined mostly by contracts among the counter-parties, which contracts should be enforced so long as the cost of auto accidents is not allocated to parties who are insolvent or judgment proof, which if permitted would undermine the deterrence benefits of the regime. 266 266. See Steven Shavell, The Judgment Proof Problem, 6 INT’L REV. OF L. & ECON. 45 (1986) (explaining how presence of insolvent defendants undermines incentive effects of liability law). ×

The types and amount of compensation recoverable under an automaker enterprise liability regime would probably be limited to economic losses—medical expenses, lost income, and property damage. There is of course a deterrence argument for including noneconomic or pain-and-suffering damages as well, since failing to include noneconomic damages could produce a serious externality. 267 267. Noneconomic damages are generally not covered by first-party insurance policies, which means, insofar as drivers (and consumers generally), because of the cognitive biases already discussed, ignore or underestimate the risks of auto accidents, they will externalize noneconomic damages as well. Including noneconomic damages, therefore, has the potential to improve care levels and activity levels. See Hanson & Logue, supra note 19, at 186–89 (describing the “unambiguous deterrence benefits of nonpecuniary-loss damages”). × However, some have argued that individuals do not desire to purchase insurance against non-economic losses (as evidenced by the dearth of pain-and-suffering insurance observed in the marketplace), and therefore should not be forced to purchase such coverage through a mandatory compensation regime. 268 268. Examples of this sort of argument can be found in George L. Priest, The Current Insurance Crisis and Modern Tort Law, 96 YALE L.J. 1521, 1546–47; and Alan Schwartz, Proposals for Products Liability Reform: A Theoretical Synthesis, 97 YALE L.J. 353, 362–67 (1988). For a powerful set of counter arguments, providing arguments why consumers might—and evidence that they in fact do—demand insurance for noneconomic losses, see Steven P. Croley & Jon D. Hanson, The Nonpecuniary Costs of Accidents: Pain-and-Suffering Damages in Tort Law, 108 HARV. L. REV. 1785 (1995). × In any event, limiting compensation to economic losses, and thus not providing compensation for noneconomic harms, is a common and reasonable political compromise that is often made when no-fault cause-based compensation regimes are adopted. 269 269. See generally Hanson, Logue & Zamore, supra note 66, at 556–62 (reviewing arguments for limiting damages in no-fault cause based compensation regimes to economic damages). ×

It is worth emphasizing again that the compensation regime I am imagining is a comprehensive automaker enterprise liability regime. In other words, it would apply to all automobiles (sold after the effective date of the enacting legislation), whether driven by humans, computer algorithms, or any combination of the two. Thus, unlike some other proposals for manufacturer-funded vehicle compensation regimes, my proposal would not apply exclusively to Level 5 vehicles. 270 270. For an interesting proposal to create a special enterprise liability-type regime, similar to the one I am describing, but that would be limited to accidents involving automated vehicles, see Abraham & Rabin, supra note 20. × Which is not to say that the regime would not have special rules for autonomous and connected vehicles. For example, whereas Level 5s may be in fewer accidents, or fewer accidents involving serious physical injuries or deaths (that’s the hope anyway), Level 5 vehicle accidents may involve much higher auto-repair costs than accidents involving human-driven vehicles, because of the expense of repairing high-tech sensors as well as computer hardware and software. 271 271. Because of the higher repair costs, some in the auto insurance industry have proposed mandatory minimum auto repair coverage for self-driving vehicles. TRAVELERS INSTITUTE, INSURING AUTONOMY: HOW AUTO INSURANCE CAN ADAPT TO CHANGING RISKS 11 (2018), https://www.travelers.com/iw-documents/travelers-institute/Final-Digital-2018-0710-AV-White-Paper-No-SAE.pdf. Under enterprise liability, of course, there would indeed be mandatory minimum auto repair coverage, as well as mandatory minimum personal injury coverage, but the coverage mandate would be imposed on automakers instead of auto purchasers. ×

If an automaker enterprise liability regime were adopted, there would be no need for either the existing automaker liability laws (i.e., products liability as applied to automobiles), driver liability laws, or state auto no-fault laws. All of those auto tort regimes would be replaced by a single comprehensive automaker enterprise liability regime. 272 272. Tort liability for non-economic and potentially punitive damages could be retained for especially egregious behavior, such as recklessness or intentionally harmful actions, on the part of defendants. × Further, if a motor-vehicle crash were to involve two or more vehicles made by different auto manufacturers, the enterprise liability regime would handle the accident as follows: First, the victims would simply file claims for their covered economic losses, naming the automakers of all of the vehicles involved in the accident. After a factual determination was made of whether in fact all of the named vehicles contributed in some way to the accident, the victims’ crash costs would be split between or among the automakers (or the auto insurers covering the losses for each automaker). The split among the automakers could either be equal (each vehicle manufacturer bearing its pro rata share of the crash costs) or according to any other reasonable allocation formula that the industry agrees upon. 273 273. The deterrence benefit for automaker and driver care and activity levels would largely remain, without the need for individualized fault determinations in multi-vehicle crashes. The crash data gathered from all the payouts under the program would provide information as to which types of vehicles and which drivers tended to be in accidents, how much damage resulted from those accidents, and under what circumstances. This information would be combined with the data gathered by automakers and auto insurers regarding individual vehicle and driver behavior in contexts not involving accidents. There would be little additional deterrence benefit to investing in the costly judicial determination of which driver, if any, was at fault or which vehicle, if any, was defective. ×

One result of the adoption of a comprehensive automaker enterprise liability regime would be an increase in the apparent (and the experienced or internalized) price of most newly purchased automobiles, relative to vehicles purchased before the effective date of the enacting legislation. 274 274. This assumes that the new law would have a grandfather provision exempting vehicles built and sold before new law’s effective date. As discussed in the conclusion, such complete grandfathering is not the only conceivable approach to handling the transition to the new regime. × This would happen because the cost of auto accidents that had been hidden in non-auto first-party insurance coverage prior to the enterprise liability regime would, with the adoption of the new system, be brought into the open through increases in automobile and auto-insurance prices. Because such a shift would be a significant change in the automotive marketplace, it would probably be prudent (and politically necessary) to institute a delayed effective date and/or an extended phase-in period over which the law would take effect. 275 275. If there is indeed deterrence value to shifting these costs from non-auto first-party insurers to automakers, as argued in the next section, then the overall price of vehicles (including the costs covered by various forms of first-party non-auto-specific insurance) should eventually go down, especially if the pace of the transition to driverless technology is hastened. Indeed, a significant result of adopting a comprehensive automaker enterprise liability regime is that Level 5 vehicles, if they provide as big an advance in safety as many are expecting, would be substantially less expensive overall than conventional human-driven vehicles. Cf. Bryant Walker Smith, Automated Driving and Product Liability, 2017 MICH. ST. L. REV. 1 (2017) (discussing the potentially massive reduction in overall vehicle prices resulting from the shift to automated vehicles under existing product liability law). A bigger concern with the price increase is the effect on low-income drivers. I address this concern briefly in Part IV below, though the cost of mobility generally is a topic worthy of special consideration. ×

The Theoretical Deterrence Benefits

Under a comprehensive automaker enterprise liability regime, because automakers would be responsible for all of the economic costs of auto accidents associated with their vehicles, they would be forced to internalize those costs. As a result, there would be beneficial deterrence consequences for automaker, and potentially driver, care and activity levels. This section explores those consequences.

First, automakers would have a strong legal and financial incentive to develop and implement cost-justified auto-safety innovations, whatever those might be. That is, if an automaker determined that there was some new brake design (such as a new computer-assisted automatic braking system) or some new guided cruise control mechanism that would reduce overall accident costs relative to its costs of development and implementation, then enterprise liability would reward them implementing those innovations, and punish them for not doing so. What’s more, there would be no inefficient incentive to stick with existing industry customs or consumer expectations if such customs or expectations were lagging behind proven safety innovations. Likewise, there would be no incentive to over-invest in safety features that are likely to impress a court or jury in a negligence-based lawsuit (such as a design defect lawsuit) but that, in actuality, provide less additional accident-risk reduction than they cost to produce. 276 276. Obviously, automakers already have some incentives to develop such safety technology, in part because of consumer tastes for safer vehicles and perhaps because of the threat of potential liability under existing tort law. See, e.g., Press Release, NHTSA, NHTSA-IIHS Announcement on AEB, (Dec. 21, 2017), https://www.nhtsa.gov/press-releases/nhtsa-iihs-announcement-aeb (“Twenty automakers pledged to voluntarily equip virtually all new passenger vehicles by September 1, 2022, with a low-speed AEB system that includes forward collision warning technology proven to help prevent and mitigate front-to-rear crashes.”). Consistent with this fact, it is common knowledge that safety innovation has been happening for decades without the presence of an automaker enterprise liability regime. My argument is that, according to a nuanced application of standard deterrence theory, safety-enhancing automotive innovations would be more likely to be adopted and would be adopted more quickly under an automaker enterprise liability regime. See, e.g., Latin, supra note 21, at 691 (making similar argument with respect to airbags, suggesting that adoption of automaker enterprise liability would have hastened the industry’s move to include airbags). ×

Second, enterprise liability would force the price of automobiles to reflect the full-expected costs of auto accidents. That cost-internalization, in turn, could result in a scale of automotive manufacturing and sales that would be closer to the social optimum than is currently the case, as drivers would—in deciding whether to purchase a vehicle—be more likely to take into account something closer to the full social costs of that decision. In other words, auto enterprise liability could push us in the direction of optimal manufacturer activity levels: the optimal number of vehicles being sold. If that were to happen, it would be a clear improvement—in terms of overall efficiency—over the existing negligence-based automaker liability regime.

It is worth pausing here to emphasize the potential effects of a comprehensive automaker enterprise liability regime on the development of and transition to Level 5 vehicles. Because it to would be a comprehensive regime, it would apply to both driverless and human-driven vehicles. Assuming automakers expect Level 5s to bring a dramatic reduction in expected accident costs relative to human-driven vehicles, then Level 5s, when they eventually are available for sale to consumers, would have a substantially lower enterprise liability “tax” relative to human-driven vehicles made and produced after the new regime is adopted, since the human-drive vehicles would have much higher expected accident costs. 277 277. For now, I am assuming that “old” vehicles, produced before the enactment of the new regime, would be totally grandfathered and thus exempt from the new law. I return to this assumption in the conclusion. × As a result, there would be a natural enterprise liability subsidy in favor of the production of Level 5 vehicles; and this subsidy, in effect, would be funded by a relatively high enterprise liability tax on human-driven vehicles, again, assuming such vehicles are not nearly as safe as Level 5s. Thus, the adoption of a comprehensive automaker liability regime would, under present assumptions, strongly incentivize and reward auto manufacturers to proceed, as quickly as is feasible, with the development and distribution of Level 5 vehicles. By contrast, if a special liability regime were adopted just for Level 5s, that increased their potential accident liability relative to human-driven vehicles, there would be a disincentive to move to Level 5s in the absence of a separate subsidy regime, perhaps funded by federal income taxes. 278 278. Abraham and Rabin make such a proposal, including the need for a subsidy for Level 5 vehicles. Abraham & Rabin, supra note 20, at 45. ×

There are efficiency reasons to prefer a Level 5 vehicle subsidy that is funded through an enterprise liability tax on auto sales, with the amount of the cross-subsidy depending on the relative risk of vehicles (and drivers), over a subsidy funded by federal income taxes. The main advantage has to do with information. Under the direct subsidy, the regulator—or whatever government body would be asked to determine the amount and structure of the subsidy—would have to determine which particular safety technologies to subsidize and which not to subsidize and how much the subsidy should be. This would require an enormous amount of information and expertise that is not within the government’s comparative advantage relative to the auto industry. By contrast, under the subsidy structure inherent in a comprehensive enterprise liability regime, it is the auto industry who would calculate the appropriate amount of the subsidy ex ante, based on their educated guesses about (a) the amount of costs to be imposed on them under the regime for accidents involving human-driven vehicles, (b) the amount of costs that would be imposed on them if they make the investment necessary to develop and implement Level 5 vehicles, and (c) the R&D, design, manufacturing, marketing, training, and other costs that would be necessary to get Level 5s fully up and running. 279 279. It is at least possible that the expected accident costs associated with Level 5 vehicles will not be, overall, a lot less than that of human-driven vehicles, once the cost of repairing the vehicles is taken into account. That is, while Level 5s are expected to reduce frequency collisions and the number of auto-related deaths and serious bodily injuries, they may result in increased repair costs, due to the expense of repairing or replacing the damaged technology in a Level 5 vehicle. In that sense, we might be trading one sort of cost for another, which of course can be socially desirable. This fact, however, might counsel in favor of including at least some noneconomic (or pain-and-suffering) damages in the enterprise liability regime, to make sure that such costs get included in the enterprise liability subsidy for Level 5 vehicles. ×

If an enterprise liability regime is likely to have deterrence benefits on the automaker side, what about its deterrence effects on driver behavior? How an auto-enterprise liability regime would affect the driving behavior of human drivers is of course an especially important question, given that, with non-Level 5 vehicles, human drivers make most of the important operational decisions. In fact, enterprise liability could actually help with driver care and activity levels in a number of ways. First, enterprise liability would create strong legal and financial incentives for automakers to develop and adopt the most cost-effective ways of warning drivers about crash risks and of instructing drivers about how best to avoid certain types of accidents. 280 280. It is a standard conclusion of deterrence theory that enterprise liability would provide strong incentives for manufacturers to develop effective warnings. Croley & Hanson, supra note 68, at 786–792. × This effect flows from the fact that enterprise liability makes automakers’ responsible for all the economic costs of their vehicles’ accidents: If an automaker could actually reduce the frequency or severity of accidents in its vehicles by altering the wording, design, or placement of warnings or instructions, it would have an incentive to do so. On the other hand, if some new or revised warning would be more likely to confuse or annoy drivers than to educate them, the automaker would be incentivized under enterprise liability not to add that sort of unhelpful warning—even if it would have gotten the automaker “off the hook” under a more traditional negligence-based warning-defect standard. Automakers would do whatever works best to reduce accident costs, which would redound to their benefit as reduced auto-accident claim payouts over time.

In addition, enterprise liability could incentivize automakers to restructure the ways that automobiles are insured and sold in order to improve driver care and activity levels. First, consider how an enterprise liability regime might affect how auto insurance is provided. Note that under an enterprise liability regime automakers would have an incentive to shift contractually much of the expected costs of auto accidents to auto insurers. This somewhat counterintuitive result flows from the fact that auto insurers’ have a comparative advantage with respect to monitoring and regulating driver care- and activity-levels. If automakers could get auto insurers to take on somewhat more of the risk of auto accidents, the insurers would have a strong incentive to help drivers reduce expected accident costs. That is, because of competition for customers in the insurance industry, auto insurers would be incentivized to use the tools at their disposal—including individualized, driving-behaviorally-sensitive, risk-adjusted insurance premiums—in ways that would tend to encourage better driving habits and perhaps less driving, especially by high risk drivers. 281 281. Ben-Shahar & Logue, supra note 51, at 220–223. ×

What does this mean for how auto insurance would be sold? Auto insurance under an enterprise liability regime might be sold in the same way that it is today. An individual auto purchaser, in other words, might pay the automaker one price for the vehicle itself and then purchase a separate auto insurance policy at the same time from a separate auto insurance company. However, given that automakers would be ultimately responsible legally (through the doctrine of subrogation) for the auto-accident losses paid by the auto insurers, there would be strong incentives for contractual coordination between automakers and auto insurers. Individual auto manufacturers might even be induced to partner with particular auto insurers in an effort to offer the best, most competitively priced, combined product of vehicle and vehicle-insurance coverage. 282 282. Why are automakers and auto insurers not incentivized now, without the adoption of enterprise liability, to coordinate in the way described in the text? This is a fair question, one that I have put to representatives of both industries and to which I have yet to get a good answer. I suspect that under current legal rules and market conditions, those incentives are dampened. Automakers can largely avoid liability by complying with the largely negligence-based product liability rules, and auto insurers make profits from insuring the residual accident risk. Neither industry—car makers or car insurers—are being forced to bear the full losses of auto accidents. In addition, because much of the risk of auto accidents are externalized by drivers to their non-auto first-party insurers, as discussed above, there is little demand-side incentive for either industry to coordinate with the other. ×

Another way that enterprise liability could improve driver care and activity levels is through its effect on how automobiles are sold. For example, the introduction of an enterprise liability regime might push the automotive industry in the direction of lease transactions rather than outright sales. This is because leasing would make it easier for automakers to enforce the terms of the auto insurance policies, which again might be sold by an insurer who was contractually partnered with the automaker. Under a lease arrangement, for example, if a driver became uninsurable (because of bad driving behavior and/or increased claim payouts), or if the driver simply stopped paying her premiums, there might be a provision in the lease empowering the automaker to reclaim the vehicle.

In addition to favoring leasehold arrangements, the introduction of enterprise liability might create market pressure on auto manufacturers to sell vehicles to commercial purchasers rather than individual consumers. That is, automakers under enterprise liability might be incentivized to sell to commercial entities—fleet operators—who would agree contractually to indemnify the manufacturer for any enterprise liability payments made to victims harmed by vehicles in their fleets. These commercial purchasers, in turn, would either lease the vehicles to individual drivers or perhaps make them available through ride-share arrangements. Automakers in turn could be incentivized to choose commercial purchasers who are financially responsible and would be incentivized to purchase efficient auto insurance contracts to cover the enterprise liability payouts. Such a trend toward commercial fleets would be consistent with already existing market trends towards ride-sharing companies, which trends are expected to accelerate with the advent of Level 5 vehicles. 283 283. Andrew G. Simpson, If They Try It, They’ll Like It: How Ridesharing, Autonomous Cars Will Win Over the Public, INSURANCE JOURNAL (Jan. 23, 2018), https://www.insurancejournal.com/news/national/2018/01/23/478073.htm. ×

I am not suggesting that comprehensive automaker enterprise liability would necessarily result in auto lease arrangements replacing individual sales, or ride sharing replacing driving. Rather, the point is that, once automakers are made legally responsible for the cost of auto accidents (or for most of those costs), they will have an incentive (and the ability) to structure automobile distribution markets in ways that are closer the social optimal.

IV.         Caveats, Concerns, and Conclusions

The description I have given here of an automaker enterprise liability regime is necessarily only a rough outline of an idea, a jumping off point for further discussion. The actual design of such a program would require empirical research into a range of topics, including whether shifting to enterprise liability would actually, and not just theoretically, produce substantial deterrence benefits. Among the other questions that would need to be answered include the following:

Under any real-world version of an automaker enterprise liability regime, there is the question of how long the automakers’ responsibility for insuring their vehicles would remain in effect. Would it be for the useful life of the vehicle or for some set period of time, say, ten years? If for some set period of time, who then would be responsible for covering the accidents arising out of the use of the vehicle? Also, what would the precise relationship be between an automaker enterprise liability regime and state mandatory insurance/financial responsibility laws? Presumably, rescission of coverage by the insurer due to excessive accident experience or failure to pay premiums would result in a suspension of driving privileges, but how would that be enforced? All good questions.

Similarly, if an auto enterprise liability regime were adopted, would it in fact have a grandfather provision perhaps exempting all vehicles manufactured and sold before a given date, as suggested above? Or would older vehicles made before the new law goes into effect be transitioned into the new regime over time? If older vehicles were fully exempted from (or grandfathered out of) the new regime, how would we deal with the resulting, potentially large, price differential between new vehicles (which would be priced with full accident costs internalized into the purchase price) and used vehicles (which would not be)? What role could increased mandatory minimum levels of auto insurance play in assisting with that transition? 284 284. An alternative to a comprehensive automaker enterprise liability regime would be simply to increase auto insurance mandates to provide coverage closer to what would be provided under an enterprise liability regime. Such a regime could be made comprehensive, in the sense that it would apply both to human-driven and computer-driven vehicles, with perhaps higher repair cost minimum coverage for Level 5 vehicles. A comprehensive mandatory auto insurance regime could also replace all of existing automaker liability and driver liability law in much the same way as I propose for auto enterprise liability and would also likely result in contractual coordination between the automakers and the auto insurers to provide the best combination of auto safety and auto-crash risk coverage. A full discussion of this auto-insurance-mandate alternative to automaker enterprise liability is of course beyond the scope of this Article. ×

There is also a whole range of question regarding how an automaker enterprise liability regime would deal with the threat of auto crashes (or stolen or destroyed data) resulting from criminal hacking of a connected system. Existing and growing markets in cyber insurance coverage might be able to handle the risks of posed to data stored in the vehicles, but the market may have more difficulty covering cyber risks to life, limb, and property. 285 285. Most cyber polices include exclusions for physical damages to persons or property. John Buchanan, Dustin Cho, and Patrick Rawsthorne, When Things Get Hacked: Coverage for Cyber Physical Risks, 2018 A.B.A. LITIG. SEC., INS. COVERAGE LITIG. COMM. HOT TOPICS FOR ICLC’S 40TH—THE COVERAGE BATTLES OF 2028 TUCSON, ARIZONA—MARCH 3, 2018, 2. × Solutions range from expanding the role of the federal government as a reinsurer of last resort to limiting liability for cyber-related physical risks to the amount of mandatory liability insurance coverage. All of these details, and many others, would need to be addressed before any comprehensive automaker enterprise liability regime could seriously be considered.

The final concern raised by the idea of an automaker enterprise liability regime involves the cost. The concern is not that the “experienced” price of autos would rise, although that would certainly be true in the short run. As already noted, such a price increase would be the source of much of the deterrence benefit of an enterprise liability regime, the mechanism through which deterrence would work, incentives for accident-avoidance optimized. Rather, the concern has to do with the problem of affordability. For some households, owning an automobile is already unaffordable, which is a source of hardship and an obstacle to social mobility. For those households, a program that raised the price of autos, even in an effort to make them safer, may not be a welcome change without some form of compensating subsidy. My own view is that some type of taxpayer funded transportation subsidies for the low-income drivers may indeed be desirable (from a social justice perspective), whether or not an automaker enterprise liability regime were adopted. But that topic too must await another day.


   Douglas A. Kahn Collegiate Professor of Law, University of Michigan Law School. Professor Logue received helpful suggestions on the ideas in this Article from numerous participants at a conference on Traffic Accident Liability and the Future of Autonomous Vehicles held at Wake Forest University School of Law and from his Michigan Law colleagues at a Fawley Lunch Workshop in Ann Arbor. Thanks also to Bryant Walker Smith for his comments and notes.

 

Recently, I wrote about the prospects for federal legislation addressing connected and autonomous vehicles. While the subject will be taken up in the new Congress, the failed push for a bill at the end of 2018 is an indication of the steep hill any CAV legislation will have to overcome. Despite the lack of federal legislation, the Department of Transportation (DOT) has been active in this space. In October 2018, the Department issued Preparing for the Future of Transportation: Automated Vehicles 3.0, DOT’s most comprehensive guide to date outlining their plan for the roll-out of CAVs. The document indicates that the department expects to prioritize working with industry to create a set of voluntary safety standards over the development of mandatory regulations.

Given the Trump administration’s broad emphasis on deregulation as a driver of economic growth, this emphasis on voluntary standards is unsurprising. A handful of consumer groups focused on auto safety have raised the alarm over this strategy, arguing that mandatory regulations are the only way to both ensure safety and make the general public confident in automated driving technology.

The remainder of this post will discuss the effectiveness of voluntary safety standards relative to mandatory regulation for the CAV industry, and consider the prospects of each going forward. While little information is available about the response to either option in the CAV field, I will seek to draw lessons from experience with regulation of the traditional automobile industry.

The National Highway Transportation Safety Administration (NHTSA) has undergone a dramatic strategic shift over its half-century existence. In its early days, NHTSA was primarily devoted to promulgating technology-forcing regulations that sought to drive innovation across the industry. Jerry Mashaw and David Harfst have documented the agency’s shift away from adopting regulations in favor of an aggressive recall policy for defective products in the 1980s. As they write, the agency then returned to a regulatory policy in the 21st century. However, rather than attempt to force technology, they chose to mandate technologies that were already in use across most of the auto industry. While these new standards still took the form of mandatory regulation, they operated as virtually voluntary standards because they mandated technologies the industry had largely already adopted on its own. Mashaw and Harfst found that this shift was essentially a trade-off of slower adoption of new safety technology, and potentially lost lives, in favor of greater legitimacy in the eyes of the courts and industry. Particularly given the rise of pre-enforcement judicial review of regulations, this shift may be seen as a defensive mechanism to allow more regulations to survive court challenges.

Even as NHTSA has pulled back from technology forcing regulations, there has been no sustained public push for more aggressive auto safety regulation. This may be because the number of traffic fatalities has been fallen slightly in recent decades. This shift is likely due more to a reduction in drunk driving than improved technology. With studies showing that the public is particularly wary of CAV adoption, it remains to be seen whether NHTSA will seek to return to its technology-forcing origins. While the auto industry has traditionally preferred voluntary adoption of new technologies, it may be the case that government mandates would help ease public concern about CAV safety, speed the adoption of this new technology, and ultimately save lives.

To date, twenty-nine states have enacted legislation related to connected and autonomous vehicles (CAVs). Eleven governors have issued executive orders designed to set guidelines for and promote the adoption of CAVs. In response to this patchwork of state laws, some experts have argued that the federal government should step in and create a uniform set of safety regulations.

Partially responding to such concerns, the National Highway Traffic Safety Administration (NHTSA) issued A Vision for Safety 2.0 in September, 2018. The guidance document contains voluntary guidance for the automotive industry, suggesting best practices for the testing and deployment of CAVs. It also contains a set of safety-related practices for states to consider implementing in legislation.

The NHTSA document is likely to have some effect on the development of safety practices for the testing and deployment of automated vehicles. While not mandatory, the guidance does give the industry some indication of what the federal government is thinking. Some companies may take this document as a sign of what they will be required to do if and when the Congress passes CAV legislation, and begin to prepare for compliance now. Furthermore, this nudge from the federal government could influence state action, as legislators with limited expertise on the subject look to NHTSA for guidance in drafting their CAV bills.

Without new legislation however, the force of NHTSA’s guidance will be blunted. No manufacturer is required to follow the agency’s views, and state legislatures are free to continue passing conflicting laws. Such conflicts among states could make it difficult to design a vehicle that is able to meet all state standards and travel freely throughout the country. To date, this has not been an acute problem because CAVs, where they are deployed, operate only within a tightly limited range. As use of these vehicles expands however, uniform standards will begin to appear more necessary.

A late push for CAV legislation in the last Congress petered out in the December lame duck session. After unanimously passing the House in 2017, the bill stalled when Senate Democrats balked at what they saw as its lack of sufficient safety protections. With Congress’ schedule blocked by the government shutdown, CAV legislation has been put on the back burner so far in 2019. At some point though, Congress is likely to take up a new bill. The Senators who were key drivers of the CAV bill in the past Congress, Gary Peters (D-MI) and Jon Thune (R-SD) remain in the Senate. Both Senators retain their influential positions on the Committee on Commerce, Science, and Transportation. The key change from the previous Congress will be the dynamic in the newly Democratic-controlled House. While a bill passed unanimously last term, it remains to be seen whether the new House will be held back by the same consumer safety concerns that led the Senate to reject the bill last term.

As autonomous vehicle technology continues to march forward, and calls for a uniform nationwide regulatory system are expected to grow. We will be following major developments.

Welcome to 2019! Over the past several months, this page has focused a lot on deployment of connected and autonomous vehicles (CAVs) in US cities. 2018 was indeed a big year for CAVs in the United States. The vehicles were deployed commercially in Arizona, California began to allow testing of the technology without a safety driver, and policymakers and urban planners across the nation thought seriously about how to integrate CAVs into their existing transportation grid.

Running through much of this work is the fear that, if left unchecked, wide-scale deployment of CAVs will kick off an accelerated version of the problems associated with the initial popularization of the automobile – suburban sprawl, increased congestion, deeper economic inequality, and more. Most American cities have proposed addressing these issues – to the extent they have considered them at all – through modest incentive programs. To kick off the new year, I want to briefly examine a city that has taken a much more aggressive tack on curtailing the problems associated with sprawl and traffic.

Even before the widespread adoption of CAVs, Singapore is moving beyond modest incentives to combat congested roads. The city of nearly 6 million people charges commuters nearly $15,000 per year to own a vehicle and use it during rush hour. In 2017, Singapore took the extreme step of announcing a freeze in the growth rate of private car ownership. While such measures seem exorbitant from an American perspective, they have contributed to reduced congestion. Singapore in 2015 was less congested than the year before, and suffered less congestion than cities such as New York, London, or Beijing. Only around 11% of Singapore’s population owns a car, in comparison to 46% of New York City residents and nearly 90% of Angelenos.

The city is also taking steps to prepare for a future dominated by CAVs. Singapore recently removed a requirement that cars have human drivers, and has mandated that all new development meets standards that accommodate CAVs while discouraging car ownership. These new real estate requirements include narrow streets, road markings designed to be easily recognizable by CAVs, and fewer parking spaces.

Such aggressive maneuvers are out of sync with policy across the United States. Many US cities have created carpool lanes to encourage ride-sharing, and Oregon has experimented with a per-mile charge to reduce congestion and plug infrastructure funding gaps that have traditionally been filled with a gas tax. However, such programs have typically been modest. Perhaps most strikingly, in comparison to Singapore’s large yearly fees, the average annual tax levied on vehicle ownership in the US comes in at a little over $200.

In many parts of the US, abundant cheap land and low vehicle taxes set the stage for suburban sprawl and maddening levels of congestion brought on by the first automobile revolution. The same factors are aligned to accelerate these problems in the upcoming CAV revolution. None of this is to say that the Singaporean approach is right for the US. It is certainly possible that, as CAVs are deployed nationwide, their benefits will outweigh any social cost brought on by sprawl and congestion. When setting their own policy though, our cities should examine a full range of options, including places like Singapore that are modeling a more aggressive regulatory posture. Regardless of the approach we choose to take, there are valuable lessons to learn from countries that approach these challenges from a different governance tradition.

This fall we’ve spent a fair amount of time talking about how connected and automated vehicles (CAVs) will change the structure of our cities, from the curb, to public transit, and beyond. In my last post before the holidays, I want to take a look at how CAVs could change the way goods are transported and delivered within cities. While they probably won’t reach Santa-levels of delivery efficiency, CAVs may help make last-mile deliveries more efficient (and could help fill the current shortage of truck drivers in the US, but that’s a subject for another day).

CAVs are already being tested as delivery vehicles by companies like Domino’s and Kroger, while earlier this year Toyota announced delivery partnerships with Amazon and Pizza Hut, and Waymo’s CEO recently highlighted it as an area of opportunity.  This week the New York Times profiled Nuro, the start-up working with Kroger to test robotic delivery cars in Scottsdale, Ariz. Nuro’s vehicles are designed in-house, and look like “toasters-on-wheels.” Currently they followed everywhere they go by human safety drivers in conventionally driven “shadow car,” since the vehicles are still in testing. When the vehicle stops for a delivery, customers enter a PIN code into a small touch pad to open a compartment containing their order. The current charge for same-day delivery using the system is around $6. Ford has also flagged the delivery market as an area they’d like to explore, citing projections that, by 2026, the last-mile delivery market for CAVs will hit $130 billion.

But the roads are not the only path automated vehicles may soon tread in their mission to bring you your takeout order. A number of companies, including Postmates, are working on delivery robots that will cruse down the sidewalk and roll right up to your door. Last year I even personally witnessed Postmates’ bot rolling along the streets of Washington. As exciting as it would be to have R2-D2’s cousin deposit an order of egg rolls on your doorstep, the deployment of delivery bots raises an interesting question of how much space we’re willing to give up to automated devices. The sidewalk is a human dominated space, and, especially in cities, is already busy with foot traffic. Will people be willing to cede some of this space to a robot? Yet another question that city regulators and individual citizens will be forced to answer as automation makes greater inroads to our daily lives.

P.S. – Last week a delivery robot caught fire in Berkeley, leading some locals to build a memorial in its honor.

I’ve written in recent weeks about the impact of autonomous vehicles on city design. Choices made by both city planners and CAV operators in the coming decades will play key roles in determining whether our new transportation paradigm is one of compact, walkable cityscapes that accommodate traffic of all sorts, or one that spurs increased suburban and exurban sprawl and is truly designed only with car transport in mind. One particularly important aspect of this question is to what extent CAVs will integrate with current mass transit rather than attempt to replace it.

Some companies, such as Ann Arbor based May Mobility,are purposely seeking out opportunities to integrate with local transit. The company recently contracted with Columbus, OH to begin operating their CAVs on a short loop through downtown along the Scioto River. A small-scale project like this has the potential to improve traffic flow in the central city without incentivizing people to move ever farther away from the urban core. A deal was also announced between May and the state of Rhode Island to run autonomous shuttles that will connect public transit lines in the nearby cities of Providence and Olneyville.

Its certainly possible that May’s long-term ambitions are bigger. They may hope to use their autonomous technology to compete with companies like Ford, Waymo and Uber to provide people with a primary mode of transportation. For now though, services like this should be viewed as a model for cities seeking to promote vibrant urban centers.

Many cities across the country, even those without longstanding strength in public transit, have already committed serious resources to revitalizing and maintaining their urban cores. Kansas City is planning a roughly $300 million extension of an existing light rail line. Phoenix, which first opened its light rail line ten years ago, passed a ballot initiative in 2015 to raise new funds fora 66-mile expansion of the system. And public transit is not supported only by public money. In South Florida, a privately owned, high speed commuter train recently opened to carry passengers between Miami, Fort Lauderdale, and West Palm Beach.

Cities investing so heavily in large-scale public transit certainly have a demonstrated interest in the economic development that comes with urban revitalization. Furthermore, they see transportation as a key factor in spurring renewed growth. If these localities are not careful though, they may see their careful plans laid to waste by the onset of CAVs. In the post-WWII era, the dominance of the automobile contributed to the emptying out of city centers and the paving over of vast swaths of land. Looking ahead, it’snot hard to see the rise of this new technology thwarting the plans of the most well-intentioned cities.

Those that hope to back up their commitment to public transit and sustainable living will need to think carefully about how transportation technologies should be accommodated. For now, the May Mobility model may be attractive for its intentional compatibility with other forms of transit. Looking ahead, as CAVs become more advanced, such companies will likely move to take over more of the transportation market. Cities need to be aware of that possibility and consider how to design their infrastructure and transportation policies to integrate CAVs into existing plans, lest they betaken over by them.

California has become the second state in the nation to permit connected and automated vehicles (CAVs) to operate on public roads without a safety driver. With the recent announcement that Waymo has obtained approval to test driverless CAVs in a handful of Northern California communities, the state joins Arizona on the leading edge of the driverless vehicle revolution. Similar to the Arizona experiment, which I wrote about recently, California has positioned itself to play a key role in shaping the speed and direction of growth in the CAV industry.

California’s regulatory apparatus, while not without its critics, will provide an interesting contrast to the relatively lax system enacted by Arizona. The remainder of this post will explore the key differences between these two approaches to governance of CAVs.

Arizona requires merely that CAV operators submit written confirmation to the State that each vehicle complies with all relevant federal law, that it is capable of reaching a “minimal risk condition” when necessary, and that it be capable of complying with traffic and safety laws. California, by contrast, has a handful of more specific requirements. In addition to the need to comply with federal law, California requires that driverless CAV operators:

  • Notify local authorities in communities where testing will take place
  • Submit a law enforcement interaction plan
  • Certify that the vehicles meet the autonomous vehicle Level 4 or 5 definition of the Society for Automotive Engineers
  • Maintain a communication link between the vehicle and a remote operator
  • Inform the DMV of the intended operational design domain
  • Submit an annual disengagement report to the DMV
  • Submit collision reports to the DMV within 10 days of a crash

In addition to these requirements for driverless vehicle testing, California has a further set of requirements before driverless CAVs can be certified for public use. This supplemental set of requirements generally revolves around data recording and security against cyber-attacks.

Critics have argued that even California’s approach to CAVs is not safety conscious enough. Consumer Watchdog, a California public interest group, has raised an alarm that the state is merely taking Waymo’s word that it has met requirements “without any real verification.” The organization has also suggested that California’s regulations are not substantively strict enough, arguing that they are turning people “into human guinea pigs for testing [Waymo’s] robot cars.” Proponents though, argue that the safety concerns are overblown in light of the potential for vast improvements relative to error-prone human drivers.

While the debate over how much regulation is proper persists, it is notable how quickly California seems to be following in the footsteps of Arizona in the rollout of CAVs. One key argument in favor of Arizona’s light touch regulation is that it has positioned the state to take the lead in development of this new technology. California’s oversight, while not enough for some, is undoubtedly more rigorous than that of its neighbor to the east. Arizona’s approach does appear to have given the state a short head start in the CAV race. California’s progress, though, indicates that modest increases in state oversight may not present a substantial barrier to the adoption of this new technology.

In the U.S., Thanksgiving represents the busiest travel period of the year, with AAA predicting that this year 54 million people will travel 50 miles or more before sitting down for turkey and stuffing. So how will CAVs and other mobility innovations change how we travel, not just at Thanksgiving, but yearlong? Lets take a look at a few recent stories that could point the way:

Waymo’s Self-Driving Service Hits the Bigtime

Back in August, Dan mentioned some issues Waymo’s automated vans had run into in Phoenix. Those issues don’t seem to have slowed the Alphabet (Google) owned company, as they have announced (as noted by Kevin) the launch of commercial service in December. The company is planning a slow roll out, and some cars will still have backup drivers, but by the Christmas travel season, some people in Arizona will be able to hail a driverless taxi to shuttle them to the airport.

Multimodality – Instead of a Taxi to the Airport, How About an E-Scooter and a Bus?

Uber has recently started to personalize suggestions on how to complete a trip. Depending on the distance to be traveled, the app will suggest you use a JUMP bike instead. Travelers in select cities can use Citymapper to plan trips across rideshares and public transit. In Chicago, for example, the app coordinates city buses, Divy bike shares, the ‘L’ system, and commuter rail. In London, Citymapper users can even hail a rideshare via the app’s own fleet. Meanwhile, bike and scooter startup Lime is expanding their services to include cars on their platform, and plans to deploy up to 500 cars in Seattle by the end of the year.

These companies are far from the only parties trying to synchronize how we use various mobility services. While the promise of a single app for all our mobility needs is yet to be fulfilled, the momentum is clearly there. Such an app would further enhance the congestion (and environmental) benefits that are projected to come with wider adoption of CAVs. While CAVs can better coordinate the cars that are on the road, multimodal programs can take even more cars off the road by pointing users to more efficient public transit or bikes/scooters.

Leaving Car Ownership Behind (Eventually…)

While some drivers may use self-driving cars and multimodality services to supplement their personal vehicles, there is an increasing push to replace vehicle ownership altogether. Lyft has launched a “ditch your car” challenge in a number of cities, encouraging users to try to live without their vehicles for a month. They’ve also launched a subscription service, offering 30 rides (up to $15 each) a month for $299.

Not interested in completely ditching your car? GM’s Maven platform lets you rent out your own vehicle, and is expanding in 2019 to include non-GM vehicles. Or you can opt for a more old-fashioned carpool, facilitated by Waze, which is slowly expanding a service to connect potential carpool members. So by next Thanksgiving, you may be able to snag a Waze carpool while leaving your personal vehicle behind to earn a little extra cash on Maven.

The point of this round up is not to provide a commercial for these platforms, but to highlight the ongoing disruption of the way people move through the world, a disruption that will only continue as CAVs reach greater deployment.

CAVs and the Traffic Stop

The traffic stop has long been a primary point of interaction between police and the community. As consent Department of Justice (DOJ) investigations into local police departments in Ferguson, Baltimore, and Chicago made clear in recent years, they are also a moment that is open to large-scale abuse. The rise of connected and autonomous vehicles (CAVs) will fundamentally alter, and perhaps dramatically reduce the occurrence of, this common police tactic. In order to avoid replicating the problematic aspects of traffic stops, communities need to grapple with the ways in which their current system has failed, and how policing should look in the future.

Local police departments in at least some parts of the country have been found to use routine traffic stops as a fundraising tool for the city. Due to either implicit or explicit bias, such policies frequently have an outsized impact on minority members of the community. DOJs investigation of the Ferguson Police Department unearthed a city government primarily concerned with the use of traffic stops to “fill the revenue pipeline.” Particularly in light of decreased sales tax revenue, city officials saw the need to increase traffic citations as “not an insignificant issue.” This attitude filtered down from the City Council and Financial Director to line officers, who were regularly reminded of the need to increase “traffic productivity.” In Ferguson, demand that the police department be a revenue generation machine contributed to racial bias in the city’s criminal justice system. African American drivers were the subjects of 85% of the traffic stops, despite constituting only 67% of the population. Among those stopped, 11% of black drivers were searched, compared to only 5% of white drivers. While the Ferguson report throws the twin problems of racialized policing and use of the police for revenue generation into stark relief, the city is far from alone. The investigations in Baltimore and Chicago found similar abuses. A review of academic literature by researchers at Princeton found that “Blacks and Hispanics are more likely to be stopped by the police, convicted of a crime, and . . . issued a lengthy prison sentence” than similarly situated whites.

These findings highlight the centrality of the traffic stop to modern policing. Traffic stops not only lead directly to citations – for speeding, missing stop signs, and the like – but also to searches of individuals and vehicles that may lead to more serious crimes for things like possession of drugs or weapons. The importance of traffic stops has been spurred on by a Supreme Court that has given its blessing to pretextual stops, in which an officer can stop a car as long as there is a valid reason, regardless of their actual reason. Widespread use of CAVs, however, could seriously cut down on pretextual stops. If a CAV is programmed to travel no faster than the speed limit, to always signal turns, and to never run a red light after all, the number of available pretexts is significantly reduced. While many commentators have been hesitant to think that this shift will lead to large-scale shifts in police tactics or a significant reduction in abuses, they have at least highlighted that possibility.

While CAVs and other new technology may lead to a shift in police tactics, they alone will not eliminate, and may not even reduce, biased policing. Unless addressed through changes to underlying structures of taxation or spending, the financial imperative to turn the police force into a revenue generator will continue to drive over-policing of minor violations. Without addressing implicit bias, this over-policing will continue to disproportionately target minority communities. The CAV era may channel these pressures in new directions. But cities that wish to address the ongoing challenge of racially biased policing must initiate structural changes, rather than merely hope that technology will save them.

City design has long been shaped by modes of transportation. The transition is easy to spot as you move westward across America. Relatively compact eastern cities initially grew up in the 18th and 19th centuries, when people traveled by foot or by horse. Scattered across the plains, and particularly throughout the vast expanses of Texas and the Southwest, are cities filled with wide thoroughfares and sprawling suburbs, designed to match the rise of car culture. A large-scale shift to autonomous vehicle transportation will once again mold our cities in new ways. I wrote recently about this coming shift, focusing in particular on the reuse of space currently dominated by parking. This post will build on that theme by exploring the ways in which big data generated by new transportation technologies will guide city planners and business strategists in creating new urban environments.

Many cities already take advantage of more traditional forms of transportation data to improve urban planning. For example, analysis of population density and traffic patterns facilitated Moscow’s 50% increase in public transit capacity, which enabled the city to reduce driving lanes in favor of more space for pedestrians and cyclists. Looking to the future, New York University’s Center for Urban Science and Progress seeks to help cities harness the power of big data to “become more productive and livable.” Today, more data exists regarding our transportation habits than ever before. Ride-hailing services such as Uber and Lyft, along with the popularity of “check-in” apps such as Foursquare, have exponentially increased the amount of data collected as we go through our daily routines. The advent of CAVs, along with smaller scale technologies such as bike-share and scooter-share programs, will only accelerate this trend.

Currently, most of this data is collected and held by private companies. This valuable information is already being aggregated and used by companies such as Sasaki, a design firm that uses data from Yelp, Google, and others to help businesses and developers understand how their planned projects can best fit in with a community’s existing living patterns. The information is able to help businesses understand, on a block-by-block basis, where their target market lives, shops, and travels. As companies such as Uber and Waymo roll out fleets of autonomous vehicles in the coming years that collect data on more and more people, such information will increasingly drive business planning.

Just as this wealth of data is impacting business decisions, making it available to the public sector would mark a significant upgrade in the capabilities of urban planners. To be sure, granting the government easy access to such fine-grained information about our daily lives comes with its own set of challenges, which my colleague Ian Williams has explored in a previous post. From the perspective of planning utility however, the benefits are clear. By better understanding exactly what times and locations present the worst traffic challenges, cities can target infrastructure improvements, tollways, or carpool benefits to alleviate the problem. A more detailed understanding of which routes people take to and from home, work, shopping, and entertainment districts can allow for more efficient zoning and the development of more walkable neighborhoods. This type of improvement has the potential to improve the livability of city centers so as to guard against the danger that CAVs will facilitate a new round of exurban flight.

As with previous shifts in transportation, the widespread move to CAVs expected in the coming years will be a key driver of the future shape of our cities. Urban planners and business strategists will play a featured role in determining whether this technology ushers in a new round of sprawl, or facilitates the growth and attractiveness of metropolitan centers. The intelligent and conscientious use of data generated by CAVs and other emerging technologies can help fuel smart development to ensure that our downtown spaces, and the communities they support, continue to thrive.

 

Tesla’s enthusiastic marketing of its Autopilot feature may be landing the company in legal hot water. Last week, a Florida man sued the car manufacturer after his Model S crashed into a stalled vehicle at high speed. The driver, who allegedly suffered spinal and brain injuries, claims that Tesla’s “purposefully manipulative sales pitch” had duped him and other Tesla owners into the mistaken belief that their vehicles can travel on the highway almost without supervision. The outcome of the case may carry key lessons not only for Tesla, but for all automakers as they develop more autonomous features.

This isn’t the first time Tesla has faced legal challenges related to the Autopilot feature. In May, the company paid $5 million to settle a class action suit claiming its Autopilot 2.0 upgrade was unusable and dangerous. This case, while currently only involving one plaintiff, could have even broader ramifications. The plaintiff’s products liability suit claims that the company has systematically duped consumers through a “pervasive national marketing campaign.” If successful, this suit could open the door to recovery for others who crash while using Autopilot.

While Tesla has typically been more grandiose in their advertising techniques than more traditional automakers, their legal challenges do serve to highlight the struggles that auto manufacturers will face in the coming years. This year alone, Ford has packaged its driver assist features into a system called Co-Pilot 360 and GM has called its Super Cruise system “the world’s first true hands-free driver assistance feature for the freeway.” In the near future, other car manufacturers are expected to join these companies in developing ever more autonomous features.

As the auto industry collectively drives toward the creation of truly autonomous vehicles, there will be an understandable temptation to hype up every new technological feature. Arguably, many of these features will increase auto safety when used properly. Certainly, road testing such features is a key step on the path towards fully driverless cars. The challenges facing Tesla should serve as a warning though. Companies need to be cautious in describing their driver-assist technologies, and ensuring that customers understand the limits of such new features. Doing so will have the dual benefit of reminding drivers that they should still be in control of the vehicle, and shielding themselves from the type of liability Tesla faces today.

 

The rapidly approaching deployment of commercially available CAVs has led city planners to begin grappling with the ways in which this new technology is expected to shape our built environment.  A 2017 report from MIT’s Urban Economics Lab and Center for Real Estate, financed by Capital One, explores potential real estate changes driven by CAVs. The report describes two theories of what the effect will be. First, CAVs could reinforce demand for central city living by relieving congestion and need for parking, making cities more livable. Alternatively, they could lead to a new wave of suburbanization by increasing the distances people are willing to travel.

As much as CAVs will shape the future of cities though, design choices made by city planners today will also impact the ways in which CAVs are utilized. Cities that are designed primarily for drivers, with limited walkability and few public transit options, are likely to experience a rehash of all the problems with 20th century suburban sprawl: congestion, increasing infrastructure needs on the urban fringe, and a reduced tax base within city limits, to name a few. There are, however, affirmative steps that cities can take to disincentivize sprawl in favor of growth in the urban core. Two of these policy options, which I will discuss below, are smart pricing of vehicle travel and increased walkability of city centers.

Many cities have already taken steps to make solo trips in cars less attractive. Whether these policies take the form of increasing options for light rail and other public transportation, designating carpool lanes, or varying parking costs depending on the time of day, many of them may not be significantly altered by the arrival of CAVs. One change that could be facilitated by CAVs is the possibility for more fine-grained trip pricing. A city that is committed to reducing congestion could vary ride pricing for people who carpool, or for trips made outside of the heaviest use periods. Those hoping to incentivize public transit could provide reduced fares for “last mile” trips to and from light rail or bus stations.

The prevalence of CAVs will also provide cities an opportunity to rethink the design of their urban landscapes. Most American cities are dominated by parking, with 30% of the space in many downtown areas being taken up by parking spaces. This is unsurprising in light of the fact that the typical car is parked around 95% of the time. The rise of CAVs will provide cities with an opportunity to adapt much of this space to more productive use through business development, building downtown housing, and expanding green space. A key challenge here for cities will be in managing the transition. A study by the Regional Planning Association for New York, New Jersey and Connecticut found that land use planning is unlikely to be “permanently altered” by CAVs until 2040 and beyond. In the intervening years, cities can begin to take steps to plan for adaptive reuse of space. This includes such design choices as building parking garages with features that allow them to be easily converted into housing and considering zoning changes that will facilitate a more livable, walkable urban core.

CAVs have the potential to contribute to the continued revitalization of city centers through the creation of more resident-friendly downtowns, or to kickstart an accelerated urban sprawl. Smart, data-driven trip pricing and infrastructure designed to smooth the transitioning needs of cities can help guide the use of CAVs in ways that facilitate compact growth and walkable communities.

Two recent news stories build interestingly on my recent blog post about CAVs and privacy. The first, from Forbes, detailing law enforcement use of “reverse location” orders, where by investigators can obtain from Google information on all Google users in a given location at a given time. This would allow, for example, police to obtain data on every Google account user within a mile of a gas station when it was robbed. Similar orders have been used to obtain data from Facebook and Snapchat.

Look forward a few years and it’s not hard to imagine similar orders being sent to the operators of CAVs, to obtain the data of untold numbers of users at the time of a crime. The problem here is that such orders can cast far too wide a net and allow law enforcement access to the data of people completely uninvolved with the case being investigated. In one of the cases highlighted by Forbes, the area from which investigators requested data included not only the store that was robbed, but also nearby homes. The same situation could occur with CAVs, pulling in data from passengers completely unrelated to a crime scene who happen to have been driving nearby.

The other story comes from The Verge, which covers data mining done by GM in Los Angeles and Chicago in 2017.  From the article:

GM captured minuted details such as station selection, volume level, and ZIP codes of vehicle owners, and then used the car’s built-in Wi-Fi signal to upload the data to its servers. The goal was to determine the relationship between what drivers listen to and what they buy and then turn around and sell the data to advertisers and radio operators. And it got really specific: GM tracked a driver listening to country music who stopped at a Tim Horton’s restaurant. (No data on that donut order, though.)

That’s an awful lot of information on a person’s daily habits. While many people have become accustomed (or perhaps numb) to the collection of their data online, one wonders how many have given thought to the data collected by their vehicle. The article also points out scale of the data collected by connected cars and what it could be worth on the market:

According to research firm McKinsey, connected cars create up to 600GB of data per day — the equivalent of more than 100 hours of HD video every 60 minutes — and self-driving cars are expected to generate more than 150 times that amount. The value of this data is expected to reach more than $1.5 trillion by the year 2030, McKinsey says.

Obviously, creators and operators of CAVs are going to want to tap into the market for data. But given the push for privacy legislation I highlighted in my last post, they may soon have to contend with limits on just what they can collect.

~ P.S. I can’t resist adding a brief note on some research from my undergraduate alma mater, the University of Illinois. It seems some researchers there are taking inspiration from the eyes of mantis shrimp to improve the capability of CAV cameras.

 

By the end of this year, Alphabet subsidiary Waymo plans to launch one of the nation’s first commercial driverless taxi services in Phoenix, Arizona. As preparations move forward, there has been increasing attention focused on Arizona’s regulatory scheme regarding connected and automated vehicles (CAVs), and the ongoing debate over whether and how their deployment should be more tightly controlled.

In 2015, Arizona Governor Doug Ducey issued an executive order directing state agencies to “undertake any steps necessary to support the testing and operation of self-driving vehicles” on public roads in the state. The order helped facilitate the Phoenix metro area’s development as a key testing ground for CAV technology and laid the groundwork for Waymo’s pioneering move to roll out its driverless service commercially in the state. It has also been the target of criticism for not focusing enough on auto safety, particularly in the aftermath of a deadly crash involving an Uber-operated CAV in March.

As the technology advances and the date of Waymo’s commercial rollout approaches, Governor Ducey has issued a new executive order laying out a few more requirements that CAVs must comply with in order to operate on Arizona’s streets. While the new order is still designed to facilitate the proliferation of CAVs, it includes new requirements that CAV owners affirm that the vehicles meet all relevant federal standards, and that they are capable of reaching a “minimal risk condition” if the autonomous system fails.

Along with these basic safety precautions, the order also directs the Arizona Departments of Public Safety and Transportation to issue a protocol for law enforcement interaction with CAVs. This protocol is a public document intended both to guide officers in interactions with CAVs and to facilitate owners in designing their cars to handle those interactions. The protocol, issued by the state Department of Transportation in May, requires CAV operators to file an interaction protocol with the Department explaining how the vehicle will operate during emergencies and in interactions with law enforcement. As CAVs proliferate, a uniform standard for police interactions across the industry may become necessary for purposes of administrative efficiency. If and when that occurs, the initial standard set by Waymo in Arizona is likely to bear an outsized influence on the nationwide industry.

Critics have called the new executive order’s modest increase in safety requirements too little for such an unknown and potentially dangerous technology. Even among critics however, there is no agreement as to how exactly CAVs should be regulated. Many have argued for, at minimum, more transparency from the CAV companies regarding their own safety and testing procedures. On the other hand, advocates of Arizona’s relaxed regulatory strategy suggest that public unease with CAVs, along with the national news coverage of each accident, will be enough to push companies to adopt their own stringent testing and safety procedures.

This more hands-off regulatory approach will get its first close-up over the next few months in Arizona. The results are likely to shape the speed and direction of growth in the industry for years to come.

 

For many people, syncing their phone to their car is a convenience – allowing them to make hands-free calls or connect to media on their phone through the car’s infotainment system. But doing so can leave a lot of data on the car’s hardware, even after a user believes they have deleted such data. That was the case in a recent ATF investigation into narcotics and firearms trafficking, where federal law enforcement agents were issued a warrant to search a car’s computer for passwords, voice profiles, contacts, call logs, and GPS locations, all of which they believed had been left on the car’s on-board memory. While it’s uncertain just what was recovered, an executed search warrant found by Forbes claims the information extraction was successful.

While this case doesn’t necessarily raise the same issues of government access to data found in the Supreme Court’s recent Carpenter decision, it does illustrate the growing amount of personal data available to outside actors via the computer systems within our vehicles. And while the 4th Amendment can (usually) shield individuals from overreach by government, personal data represents a potential target for malicious actors, as shown by the recent data breach at Facebook which exposed the data of 30 million users. As cars become yet another part of the greater “internet of things,” (IoT) automakers have to confront issues of data protection and privacy. Security researchers have already began to prod vehicle systems for weaknesses – one group was able to breach the computer of a Mazda in 10 seconds.

There has of late been a great deal of talk, and some action, in Washington, Brussels, and Sacramento, towards mandating greater privacy and security standards. Earlier this month, the Senate Commerce Committee held a hearing on Data Privacy in the wake of the European Union’s General Data Protection Regulation, which took effect in May, and California’s Consumer Privacy Act, which was passed in June. Last month, California also passed a bill that sets cybersecurity standards for IoT devices – and there are similar bills that have been introduced in the House and Senate. While it remains to be seen if either of those bills gain traction, it is clear that there is an interest in more significant privacy legislation at the state and federal level, an interest that has to be considered by automakers and other CAV developers as CAVs move closer and closer to wide-scale deployment.

Last week’s release of the Intergovernmental Panel on Climate Change (IPCC) special report highlights the “rapid and far-reaching” societal transformations required in order to limit warming to 1.5, or even 2 degrees Celsius. A new study by researchers at the University of Michigan, published in the journal Environmental Science & Technology, highlights the role of connected and automated vehicles (CAVs) in ushering in a low-emissions future. This research sheds new light on a largely understudied aspect of the coming CAV revolution. In my first post for the Journal of Law and Mobility, I will summarize that study and provide key takeaways for policymakers.

The Michigan study identifies several factors that will cause CAVs to emit more greenhouse gases than comparable human-driven vehicles. The weight of sensors and the computer system necessary to operate an CAV, the power consumed by the computer system, in particular the mapping function used to create high definition charts of the car’s surroundings, and the increased drag from cameras and sensors mounted on the outside of the vehicle, all operate to increase emissions. Depending on the weight of the equipment and power usage of the computer, these factors were found to increase emissions by between 2.8% and 20% relative to a comparable human driven car.

These factors are expected to be partially, if not entirely, offset by the car’s ability to create more favorable traffic patterns and identify more efficient routes. This extra efficiency is expected to more than offset any increase in weight, power usage, and drag under some scenarios, and turn a relatively large emissions increase into a more modest one under others. All things considered, the study finds that the emissions impact of CAVs will range from a 9% reduction to a 5% increase.

Many of the obstacles to reducing CAV emissions are engineering challenges: reducing the weight and power consumption of computer systems, and improving the aerodynamics of external sensors. Policymakers role in solving these challenges are likely to primarily take the form of support for university research and/or tax credits for private sector research on improving the efficiency of CAVs.

Policymakers do however have a significant role to play in improving the network effects of automated vehicles, such as reduced congestion. After the proper levels of safety, security, and reliability are obtained, a high volume of CAVs on the road increases the efficiencies that can be gained through cars communicating with each other to ease the flow of traffic. Laws that ensure high standards for data privacy and CAV safety can give consumers the confidence needed to use CAVs at a higher rate. Regulatory schemes that ease the entrance of CAV fleets into a city’s vehicular landscape can promote early adoption.

Particularly as the technology advances, CAVs have a role to play in reducing harmful greenhouse gas emissions. Widespread adoption of the technology can maximize these benefits, paving the way for large fleets of CAVs that create strong network efficiencies. As the technology advances to a point of being safe for public use, policymakers should account for these potential benefits as they consider the advent of CAVs in their cities.

As per my last post, our law school problem solving class is looking at problems created by the interaction between connected and automated vehicles and other roadway users. This article from The Information offers some interesting insights on the difficulties Waymo is facing as it deploys its robo-taxi service in Phoenix.  Basically, the problem comes down to . . . people.  A blurb from the article:

The biggest issue for Waymo’s vans and other companies’ prototypes is human drivers or pedestrians who fail to observe traffic laws. They do so by speeding, by not coming to complete stops, by turning illegally, texting while driving, or with an endless array of other moving violations that have become an accepted part of driving. Waymo’s prototypes sometimes respond to these maneuvers by stopping abruptly in ways that human drivers don’t anticipate. As a result, human drivers from time to time have rear-ended the Waymo vans.

This fall, the University of Michigan Law School is offering its third Problem Solving Initiative (“PSI”) course concerning connected and automated vehicles. The first class, offered in the Winter 2017 semester, involved a team of fifteen graduate students from law, business, engineering, and public policy who accepted the challenge of coming up with commercial use cases for data generated by connected vehicles using dedicated short-range communication (“DSRC”) technology.

In the Fall of 2017, we offered our second PSI Course in CAV—this one to 23 graduate students. That course focused on the problem of Level 3 autonomy, as defined by the Society of Automotive Engineers (“SAE”). Level 3 autonomy, or conditional automation, is defined as a vehicle driving itself in a defined operational design domain (“ODD”), with a human driver always on standby to take over the vehicle upon short notice when the vehicle exits the ODD. As with the first course, our student teams spent the semester collecting information from industry, governmental, and academic experts and proposing a series of innovative solutions to various obstacles to the deployment of Level 3 systems.

This semester, our PSI course is entitled Connected and Automated Vehicles: Preparing for a Mixed Fleet Future. I will be co-teaching the course with Anuj Pradhan and Bryant Walker Smith. Our focus will be on the multiple potential problems created by unavoidable future interactions between automated vehicles and other road users, such as non-automated, human-driven vehicles, pedestrians, and bicyclists.

Although cars can be programmed to follow rules of the road, at its core, driving and roadway use are social activities. Roadway users rely heavily on social cues, expectations, and understandings to navigate shared transportation infrastructure. For example, although traffic circles are in principle governed by a simple rule of priority to vehicles already in the circle, their actual navigation tends to governed by a complex set of social interactions involving perceptions of the intentions, speed, and aggressivity of other vehicles. Similarly, while most states require bicyclists to obey stop signs and traffic lights, most cyclists do not; prudent drivers should not expect them to.

Can cars be programmed to behave “socially?” Should they be, or is the advent of robotic driving an opportunity to shift norms and expectations toward a greater degree of adherence to roadway rules? Will programming vehicles to be strictly rule compliant make CAVs “roadway wimps,” always giving in to more aggressive roadway users? Would that kill the acceptance of CAVs from a business perspective? Is reform legislation required to permit CAVs to mimic human drivers?

More generally, is the advent of CAVs an opportunity to reshape the way that all roadway users access roadways? For example, could the introduction of automated vehicles be an opportunity to reduce urban speeds? Or to prohibit larger private vehicles from some streets (since people may no longer be dependent only on their individually owned car)? These questions are simply illustrative of the sorts of problems our class may choose to tackle. Working in interdisciplinary groups, our graduate students will attempt to identify and solve the key legal, regulatory, technological, business, and social problems created by the interaction between CAVs and other roadway users.

As always, our class will rely heavily on on the expertise of folks from government, industry, and academia. We welcome any suggestions for topics we should consider or experts who might provide important insights as our students begin their discovery process next week.

Transportation as we know it is changing dramatically.  New technology, new business models and new ways of thinking about how we move are being announced almost daily.  With all this change, come inevitable questions about legality, responsibility, and morality.  Lawyers and policy makers play a leading role in answering these challenging questions.  The newly launched Journal of Law and Mobility, will serve an important role as the leading source for scholarship, commentary, analysis, and information, and enable a meaningful dialogue on a range of mobility topics.

In order to facilitate this needed dialogue, it is important at the outset that we ground ourselves in the terminology used to describe “mobility.”  There are a lot of terms being used by different people in the industry, government and media that can be confusing or ambiguous to those not familiar with the technology.  Terms such as “semi-autonomous,” “highly automated” or “connected and automated vehicles” can describe a wide range of vehicles, from “self-driving cars” that actually have self-driving capability, to cars that are connected and communicating with each other, but have lower levels of automation that provide assistance to drivers.

It is very important that we are clear and concise when having a discussion about mobility, because while there are common issues in each area, there are many unique aspects of each technology that merit different discussion.  Fortunately, we have a framework that helps us have clearer discussion about automated technology, the SAE levels of driving automation.  This document describes 6 levels of automation, from Level 0 – no automation, to Level 5 – full automation, and the responsibilities associated with each level of automation in terms of monitoring and executing the Dynamic Driving Task (DDT).  The SAE taxonomy has become so widespread, that even governmental entities such as the National Highway Traffic Administration (NHTSA) and the California Department of Motor Vehicles (CA DMV) are utilizing these levels of automation in their policy statements and rulemaking.

The CA DMV went even further, and specifically regulates the use of certain terminology.  In their Driverless Testing Regulations issued in February, 2016, they specifically require that “no manufacturer or its agents shall represent in any advertising for the sale or lease of a vehicle that a vehicle is autonomous” unless it meets the definition of SAE Levels 3-5.

Lawyers know the importance of words for legal purposes, but terminology is also important for consumers, particularly for building the trust that will be required for successful deployment of self-driving vehicles.  There is already some data suggesting that consumers are confused, for example a finding from an MIT AgeLab survey question that asked respondents if self-driving vehicles are available for purchase today, with nearly 23% saying “yes” – despite the fact that no Level 3 or higher vehicle is actually for sale yet.

NHTSA’s 2017 policy statement addresses this concern, it includes “Consumer Education and Training” as one of the twelve safety design elements of the Voluntary Safety Self-Assessments it suggests that manufacturers complete, citing a need for explicit information on system capabilities to minimize potential risks from user system abuse or misunderstanding.  Legislation that passed the House last year, the SELF DRIVE Act, would take this a step further by mandating that the Department of Transportation (DOT) do research to determine the most effective method and terminology for informing consumers about vehicle automation capabilities and limitations, including the possible use of the SAE levels.

SAE is not the only organization to tackle this problem, there are similar definitions developed in Europe by the German Association of the Automotive Industry (VDA) and the Germany Federal Highway Research Institute (BASt).  Whether we utilize one of these definitional frameworks or not, what is most important is that we are specific about what we are discussing, to enable clear and effective dialogue as we endeavor to solve the important issues ahead.

Cite as: Daniel A. Crane, The Future of Law and Mobility, 2018 J. L. & Mob. 1.

Introduction

With the launch of the new Journal of Law and Mobility, the University of Michigan is recognizing the transformative impact of new transportation and mobility technologies, from cars, to trucks, to pedestrians, to drones. The coming transition towards intelligent, automated, and connected mobility systems will transform not only the way people and goods move about, but also the way human safety, privacy, and security are protected, cities are organized, machines and people are connected, and the public and private spheres are defined.

Law will be at the center of these transformations, as it always is. There has already been a good deal of thinking about the ways that law must adapt to make connected and automated mobility feasible in areas like tort liability, insurance, federal preemption, and data privacy. 286 286. See, e.g., Daniel A. Crane, Kyle D. Logue & Bryce Pilz, A Survey of Legal Issues Arising from the Deployment of Autonomous and Connected Vehicles, 23 Mich. Tel. & Tech. L. Rev. 191 (2017). × But it is also not too early to begin pondering the many implications for law and regulation arising from the technology’s spillover effects as it begins to permeate society. For better or worse, connected and automated mobility will disrupt legal practices and concepts in a variety of ways additional to the obvious “regulation of the car.” Policing practices and Fourth Amendment law, now so heavily centered on routine automobile stops, will of necessity require reconsideration. Notions of ownership of physical property (i.e., an automobile) and data (i.e., accident records) will be challenged by the automated sharing economy. And the economic and regulatory structure of the transportation network will have to be reconsidered as mobility transitions from a largely individualistic model of drivers in their own cars pursuing their own ends within the confines of general rules of the road to a model in which shared and interconnected vehicles make collective decisions to optimize the system’s performance. In these and many other ways, the coming mobility revolution will challenge existing legal concepts and practices with implications far beyond the “cool new gadget of driverless cars.”

Despite the great importance of the coming mobility revolution, the case for a field of study in “law and mobility” is not obvious. In this inaugural essay for the Journal of Law and Mobility, I shall endeavor briefly to make that case.

I. Driverless Cars and the Law of the Horse

A technological phenomenon can be tremendously important to society without necessarily meriting its own field of legal study because of what Judge Frank Easterbrook has described as “the law of the horse” problem. 287 287. Frank H.Easterbrook,Cyberspace and the Law of the Horse, 1996 U. Chi. Legal F. 207, 207-16. × Writing against the burgeoning field of “Internet law” in the early 1990s, Easterbrook argued against organizing legal analysis around particular technologies:

The best way to learn the law applicable to specialized endeavors is to study general rules. Lots of cases deal with sales of horses; others deal with people kicked by horses; still more deal with the licensing and racing of horses, or with the care veterinarians give to horses, or with prizes at horse shows. Any effort to collect these strands into a course on “The Law of the Horse” is doomed to be shallow and to miss unifying principles. 288 288. Id. ×

Prominent advocates of “Internet law” as a field rebutted Easterbrook’s concern, arguing that focusing on cyberlaw as a field could be productive to understanding aspects of this important human endeavor in ways that merely studying general principles might miss. 289 289. Lawrence Lessig, The Law of the Horse: What Cyberlaw Might Teach, 113 Harv. L. Rev. 501 (1999). × Despite Easterbrook’s protestation, a distinct field of cyberlaw has grown up in recent decades.

“The law of the horse” debate seems particularly apt to the question of law and mobility since the automobile is the lineal successor of the horse as society’s key transportation technology. Without attempting to offer a general solution to the “law of the horse” question, it is worth drawing a distinction between two different kinds of disruptive technologies—those in which the technological change produces social changes indirectly and without significant possibilities for legal intervention, and those in which law is central to the formation of the technology itself.

An example of the first species of technological change is air conditioning. The rise of air conditioning in the mid-twentieth century had tremendous effects on society, including dramatic increases in business productivity, changes in living patterns as people shifted indoors, and the extension of retail store hours and hence the growing commercialization of American culture. 290 290. Stan Cox, Losing Our Cool: Uncomfortable Truths About Our Air-Conditioned World (and Finding New Ways to Get Through the Summer) (2012). × The South’s share of U.S. population was in steady decline until the 1960s when, in lockstep with the growth of air conditioning and people’s willingness to settle in hot places, the trend abruptly reversed and the South’s share grew dramatically. 291 291. Paul Krugman, Air Conditioning and the Rise of the South, New York Times March 28, 2015. × The political consequences were enormous—from Richard Nixon through George W. Bush, every elected President hailed from warm climates.

One could say, without exaggeration, that the Willis Carrier’s frigid contraption exerted a greater effect on American business, culture, and politics than almost any other invention in the twentieth century. And, yet, it would seem silly to launch a field of study in “law and air conditioning.” Air conditioning’s social, economic, and political effects were largely indirect—the result of human decisions in response to the new circumstances created by the new technology rather than an immediate consequence of the technology itself. Even if regulators had foreseen the dramatic demographic effects of air conditioning’s spread, there is little they could have done (short of killing or limiting the technology) to mediate the process of change by regulating the technology.

Contrast the Internet. Like air conditioning, the Internet has had tremendous implications for culture, business, and politics, but unlike air conditioning, many of these effects were artifacts of design decisions regarding the legal architecture of cyberspace. From questions of taxation of online commercial transactions, 292 292. See, e.g., John E. Sununu, The Taxation of Internet Commerce, 39 Harv. J. Leg. 325 (2002). × to circumvention of digital rights management technologies, 293 293. See, e.g., David Nimmer, A Rif on Fair Use in the Digital Millenium Copyright Act, 148 U. Pa. L. Rev. 673 (2000). × to personal jurisdiction over geographically remote online interlocutors, 294 294. Note, No Bad Puns: A Different Approach to the Problem of Personal Jurisdiction and the Internet, 116 Harv. L. Rev. 1821 (2003). × and in countless other ways, a complex of legal and regulatory decisions created the modern Internet. From the beginning, law was hovering over the face of cyberspace. Al Gore may not have created the Internet, but lawyers had as much to do with it as did engineers.

The Internet’s legal architecture was not established at a single point in time, by a single set of actors, or with a single set of ideological commitments or policy considerations. Copyright structures were born of the contestation among one set of stakeholders, which was distinct from the sets of stakeholders contesting over tax policy, net neutrality, or revenge porn. And yet, the decisions made in separate regulatory spheres often interact in underappreciated ways to lend the Internet its social and economic character. Tax policy made Amazon dominant in retail, copyright policy made Google dominant in search, and data protection law (or its absence) made Facebook dominant in social media—with the result that all three have become antitrust problems.

Whether or not law students should be encouraged to study “Internet law” in a discrete course, it seems evident with the benefit of thirty years of hindsight that the role of law in mediating cyberspace cannot be adequately comprehended without a systemic inquiry. Mobility, I would argue, will be much the same. While the individual components of the coming shift toward connectivity and automation—i.e., insurance, tort liability, indemnification, intellectual property, federal preemption, municipal traffic law, etc.—will have analogues in known circumstances and hence will benefit from consideration as general questions of insurance, torts, and so forth, the interaction of the many moving parts will produce a novel, complex ecosystem. Given the potential of that ecosystem to transform human life in many significant ways, it is well worth investing some effort in studying “law and mobility” as a comprehensive field.

II. An Illustration from Three Connected Topics

It would be foolish to attempt a description of mobility’s future legal architecture at this early stage in the mobility revolution. However, in an effort to provide some further motivation for the field of “law and mobility,” let me offer an illustration from three areas in which legal practices and doctrines may be affected in complex ways by the shift toward connected and automated vehicles. Although these three topics entail consideration of separate fields of law, the technological and legal decisions made with respect to them could well have system-wide implications, which shows the value of keeping the entire system in perspective as discrete problems are addressed.

A. Policing and Public Security

For better or for worse, the advent of automated vehicles will redefine the way that policing and law enforcement are conducted. Routine traffic stops are fraught, but potentially strategically significant, moments for police-citizen interactions. Half of all citizen-police interactions, 295 295. Samuel Walker, Science and Politics in Police Research: Reflections on their Tangled Relationship, 593 Annals Am. Acad. Pol. & Soc. Sci. 137, 142 (2004); ATTHEW R. DUROSE ET. AL., U.S. DEP’T OF JUSTICE, OFFICE OF JUSTICE PROGRAMS, BUREAU OF JUSTICE STATISTICS, CONTACTS BETWEEN POLICE AND THE PUBLIC, 2005, 1 (2007). × more than forty percent of all drug arrests, 296 296. David A. Sklansky,Traffic Stops, Minority Motorists, and the Future of the Fourth Amendment, 1997SUP. CT. REV. 271, 299. × and over 30% of police shootings 297 297. Adams v. Williams, 407 U.S. 143, 148 n.3 (1972). × occur in the context of traffic stops. Much of the social tension over racial profiling and enforcement inequality has arisen in the context of police practices with respect to minority motorists. 298 298. Ronnie A. Dunn, Racial Profiling: A Persistent Civil Rights Challenge Even in the Twenty-First Century, 66 Case W. Res. L. Rev. 957, 979 (2016) (reporting statistics on disproportionate effects on racial minorities of routine traffic stops). × The traffic stop is central to modern policing, including both its successes and pathologies.

Will there continue to be routine police stops in a world of automated vehicles? Surely traffic stops will not disappear altogether, since driverless cars may still have broken taillights or lapsed registrations. 299 299. See John Frank Weaver, Robot, Do You Know Why I Stopped You?. × But with the advent of cars programmed to follow the rules of the road, the number of occasions for the police to stop cars will decline significantly. As a general matter, the police need probable cause to stop a vehicle on a roadway. 300 300. Whren v. U.S., 517 U.S. 806 (1996). × A world of predominantly automated vehicles will mean many fewer traffic violations and hence many fewer police stops and many fewer police-citizen interactions and arrests for evidence of crime discovered during those stops.

On the positive side, that could mean a significant reduction in some of the abuses and racial tensions around policing. But it could also deprive the police of a crime detection dragnet, with the consequence either that the crime rate will increase due to the lower detection rate or that the police will deploy new crime detection strategies that could create new problems of their own.

Addressing these potentially sweeping changes to the practices of policing brought about by automated vehicle technologies requires considering both the structure of the relevant technology and the law itself. On the technological side, connected and automated vehicles could be designed for easy monitoring and controlling by the police. That could entail a decline in privacy for vehicle occupants, but also potentially reduce the need for physical stops by the police (cars that can be remotely monitored can be remotely ticketed) and hence some of the police-citizen roadside friction that has dominated recent troubles.

On the legal side, the advent of connected and automated vehicles will require rethinking the structure of Fourth Amendment law as required to automobiles. At present, individual rights as against searches and seizures often rely on distinctions between drivers and passengers, or owners and occupants. For example, a passenger in a car may challenge the legality of the police stop of a car, 301 301. Brendlin v. California, 551 U.S. 249 (2007). × but have diminished expectations of privacy in the search of the vehicle’s interior if they are not the vehicle’s owners or bailees. 302 302. U.S. v. Jones, 565 U.S. 400 (2012). × In a mobility fleet without drivers and (as discussed momentarily) perhaps without many individual owners, these conceptions of the relationship of people to cars will require reconsideration.

B. Ownership, Sharing, and the Public/Private Divide

In American culture, the individually owned automobile has historically been far more than a transportation device—it has been an icon of freedom, mobility, and personal identity. As Ted McAllister has written concerning the growth of automobile culture in the early twentieth century:

The automobile squared perfectly with a distinctive American ideal of freedom—freedom of mobility. Always a restless nation, with complex migratory patterns throughout the 17th, 18th, and 19thcenturies, the car came just as a certain kind of mobility had reached an end with the closing of the frontier. But the restlessness had not ended, and the car allowed control of space like no other form of transportation. 303 303. Ted v. McAllister, Cars, Individualism, and the Paradox of Freedom in a Mass Society. ×

Individual car ownership has long been central to conceptions of property and economic status. The average American adult currently spends about ten percent of his or her income on an automobile, 304 304. Máté Petrány, This Is How Much Americans Spend on their Cars. × making it by far his or her most expensive item of personal property. The social costs of individual automobile ownership are far higher. 305 305. Edward Humes, The Absurd Primacy of the Automobile in American Life; Robert Moor, What Happens to the American Myth When You Take the Driver Out of It?. ×

The automobile’s run as an icon of social status through ownership may be ending. Futurists expect that the availability of on-demand automated vehicle service will complete the transition from mobility as personal property to mobility as a service, as more and more households stop buying cars and rely instead on ride sharing services. 306 306. Smart Cities and the Vehicle Ownership Shift. × Ride sharing companies like Uber and Lyft have long been on this case, and now automobile manufacturers are scrambling to market their vehicles as shared services. 307 307. Ryan Felton, GM Aims to Get Ahead of Everyone with Autonomous Ride-Sharing Service in Multiple Cities by 2019. × With the decline of individual ownership, what will happen to conceptions of property in the physical space of the automobile, in the contractual right to use a particular car or fleet of automobiles, and in the data generated about occupants and vehicles?

The coming transition from individual ownership to shared service will also raise important questions about the line between the public and private domains. At present, the “public sphere” is defined by mass transit whereas the individually owned automobile constitutes the “private sphere.” The public sphere operates according to ancient common carrier rules of universal access and non-discrimination, whereas a car is not quite “a man’s castle on wheels” for constitutional purposes, 308 308. See Illinois v. Lidster, 540 U.S. 419, 424 (2004) (“The Fourth Amendment does not treat a motorist’scaras hiscastle.”). × but still a non-public space dominated by individual rights as against the state rather than public obligations. 309 309. E.g., Byrne v. Rutledge, 623 F.3d 46 (2d Cir. 2010) (holding the motor vehicle license plates were nonpublic fora and that state’s ban on vanity plates referencing religious topic violated First Amendment). × As more and more vehicles are held and used in shared fleets rather than individual hands, the traditional line between publicly minded “mass transit” and individually minded vehicle ownership will come under pressure, with significant consequences for both efficiency and equality.

C. Platform Mobility, Competition, and Regulation

The coming transition toward ride sharing fleets rather than individual vehicle ownership described in the previous section will have additional important implications for the economic structure of mobility—which of course will raise important regulatory questions as well. At present, the private transportation system is highly atomistic. In the United States alone, there are 264 million individually owned motor vehicles in operation. 310 310. U.S. Dep’t of Energy, Transportation Energy Data Book, Chapter 8, Household Vehicles and Characteristics, Table 8.1, Population and Vehicle Profile, https://cta.ornl.gov/data/chapter8.shtml (last visited May 29, 2018). × For the reasons previously identified, expect many of these vehicles to shift toward corporate-owned fleets in coming years. The question then will be how many such fleets will operate—whether we will see robust fleet-to-fleet competition or instead the convergence toward a few dominant providers as we are seeing in other important areas of the “platform economy.”

There is every reason to believe that, before too long, mobility will tend in the direction of other monopoly or oligopoly platforms because it will share their economic structure. The key economic facts behind the rise of dominant platforms like Amazon, Twitter, Google, Facebook, Microsoft, and Apple are the presence of scale economies and network effects—system attributes that make the system more desirable for others users as new users join. 311 311. See generally DavidS.Evans& Richard Schmalensee, A Guide to the Antitrust Economics of Networks, Antitrust, Spring 1996, at 36; Michael L. Katz & Carl Shapiro, Systems Competition andNetworkEffects, 8 J. Econ. Persp. 93 (1994). × In the case of the mobility revolution, a number of features are suggestive of future scale economies and network effects. The more cars in a fleet, the more likely it is that one will be available when summoned by a user. The more cars connected to other cars in a fleet, the higher the quality of the information (on such topics as road and weather conditions and vehicle performance) available within the fleet and the steeper the machine learning curve.

As is true with other platforms, the mere presence of scale economic and network effects does not have to lead inexorably to market concentration or monopoly. Law and regulation may intervene to mitigate these effects, for example by requiring information sharing or interconnection among rival platforms. But such mandatory information sharing or interconnection obligations are not always advisable, as they can diminish a platform’s incentives to invest in its own infrastructure or otherwise impair incentives to compete.

Circling back to the “law of the horse” point raised at the outset, these issues are not, of course, unique to law and mobility. But this brief examination of these three topics—policing, ownership, and competition—shows the value of considering law and mobility as a distinct topic. Technological, legal, and regulatory decisions we make with respect to one particular set of problems will have implications for distinct problems perhaps not under consideration at that moment. For example, law and technology will operate conjunctively to define the bounds of privacy expectations in connected and automated vehicles, with implications for search and seizure law, property and data privacy norms, and sharing obligations to promote competition. Pulling a “privacy lever” in one context—say to safeguard against excessive police searches—could have spillover effects in another context, for example by bolstering a dominant mobility platform’s arguments against mandatory data sharing. Although the interactions between the different technological decisions and related legal norms are surely impossible to predict or manage with exactitude, consideration of law and mobility as a system will permit a holistic view of this complex, evolving ecosystem.

Conclusion

Law and regulation will be at the center of the coming mobility revolution. Many of the patterns we will observe at the intersection of law and the new technologies will be familiar—at least if we spend the time to study past technological revolutions—and general principles will be sufficient to answer many of the rising questions. At the same time, there is a benefit to considering the field of law and mobility comprehensively with an eye to understanding the often subtle interactions between discrete technological and legal decisions. The Journal of Law and Mobility aims to play an important role in this fast-moving space.


Frederick Paul Furth, Sr. Professor of Law, University of Michigan. I am grateful for helpful comments from Ellen Partridge and Bryant Walker Smith. All errors are my own.