Advancing Automated and Connected Vehicles: Policy and Planning Strategies for State and Local Transportation Agencies (2017)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

11 C h a p t e r 2 This chapter summarizes the potential impacts of CVs and AVs in five areas in which there are major transportation exter- nalities: safety, congestion, pollution, land development, and mobility. Resource papers covering each area can be found in the interim report for this research, which is available on the website of the Transportation Research Board. This chapter also describes the outcomes that would be most beneficial for society, examines the roles of the primary private-sector actors, and discusses why the actions of the private sector alone may not be enough to realize those outcomes. Potential Impacts of AVs and CVs Safety In 2015, there were 35,092 people killed in motor vehicle crashes in the United States, and an additional 2.44 million were injured (NHTSA 2016d). While fatalities have decreased in the long run—25 percent from 2005 to 2014—the past year has shown a marked increase. The 7.2 percent increase from 32,744 fatalities in 2014 represents the largest increase in over 50 years. This is partly because Americans drove more and partly because they drove more poorly. Drivers, vehicles, and environmental conditions can all cause crashes, but human errors are a critical cause of more than 90 percent of them (NHTSA 2016a). “Critical cause” is defined as the immediate reason for the pre-crash event as col- lected in NHTSA’s National Motor Vehicle Crash Causation Survey, conducted from 2005 to 2007. CV and AV technology have the potential to significantly reduce human error. Safety warnings provided by V2V and V2I technology enable drivers to take actions that could reduce the severity of collisions or avoid them. AVs can avoid many of the common perception, decision, and execution mistakes that humans make, and AVs do not suffer from fatigue or cognitive impairment. However, AVs and CVs may also introduce new safety risks. AVs may potentially be safer than human drivers, but flawed hardware or software could cause accidents, including those caused by the types of errors that humans would not make. AVs and CVs could create cybersecurity risks, which could lead to serious crashes if the vehicles are hacked. Level 3 AVs also introduce risks posed by inattentive drivers being unable to jump in and take safe control of the vehicle when requested. Overall, there is enormous room for improvement over human drivers’ performance, and early research suggests that these technologies have promise, but the safety benefits of AVs and CVs are not guaranteed. Congestion Congestion occurs both regularly (i.e., recurring) and spo- radically (i.e., non-recurring) due to accidents, construction, weather, and so forth. It has enormous societal costs. Travel delays due to traffic congestion caused drivers to waste more than 3 billion gallons of fuel and kept travelers stuck in their cars for nearly 7 billion extra hours—42 hours per rush-hour commuter (Schrank et al. 2015). The total cost to the United States was $960 per commuter, or $160 billion for the nation as a whole. CV safety applications could mitigate non-recurring con- gestion events by reducing delays caused by safety incidents through informing CVs of the delay, thus enabling them to choose a different route. CV mobility applications could pos- itively impact recurring congestion by increasing system effi- ciency and enabling CV-facilitated platoons. These impacts would be maximized if there were widespread adoption of V2V capabilities, widespread V2I infrastructure, and inter- operability among mobility applications. AVs that are safer than human drivers would enable the reduction of crash-related delays. AVs that operate with more precision and control than human drivers could eventually enable infrastructure operators to redesign aspects of their facilities to accommodate more traffic (e.g., narrower lanes and shorter headways), thereby increasing supply. Simultaneously, AVs and CVs are likely to decrease the cost of driving, and are thus expected to induce additional Realizing Societal Benefits of AVs and CVs

12 demand for driving and increase VMT (Anderson et al. 2014). SAE Level 4/5 AVs could also increase travel demand by enabling individuals who were previously unable to drive to do so (Smith 2012). Additionally, with fully automated (SAE Level 4/5) vehicles, the opportunity costs of a motorist’s time could be completely removed through the motorist not even being present in the vehicle. This could enable many dif- ferent services and opportunities for motorists, which would also likely increase demand for vehicular travel. As with any disruptive technology, the net effects of its complex forces cannot be predicted. Pollution Automobiles emit local air pollutants (e.g., particulate mat- ter, hydrocarbons, nitrogen oxides, and carbon monoxide) and global air pollutants (greenhouse gases) when they com- bust fuels, primarily fossil fuels. This pollution poses an enor- mous environmental and public health cost, both locally and globally in the case of greenhouse gases. The total emission in the road transportation segment can be described across four primary categories (Transportation Research Board 2000). CVs and AVs could have an effect in each category. The following list highlights some but not all the complex potential effects. 1. Travel-related factors. The effect of AVs and CVs on travel demand (whether an increase or a decrease due to the fac- tors as described in the congestion discussion) would have corresponding effects on pollution, to the extent that vehi- cles continue to burn fossil fuels. 2. Vehicle-related factors. To the extent that CVs and AVs are safer, and AVs are shared among multiple users, there could be reductions in vehicle production rates and reduced parking needs. The use of right-sized vehicles (i.e., vehicles that have capacity equivalent to the number of passen- gers or goods) could also reduce emissions. On the other hand, passengers may prefer larger AVs to allow them to take better advantage of the opportunity to do things other than driving, resulting in lower fuel economy and greater emissions. AV and CV capabilities could also lead to an increase in electric, hydrogen, or other renewable/ low-emission vehicles because vehicles would be lighter and could potentially drive themselves to refueling areas. 3. Driver behavior. AVs could encourage or enable eco- driving, increasing fuel economy of vehicles. Shared AVs (SAVs) may encourage higher vehicle occupancies, but AVs may also reduce vehicle occupancies with the ability to have zero-occupancy vehicles. 4. Highway-related factors. Increased effective capacity enabled through CV and AV capabilities may result in less need for new capacity projects and lower construction- related emissions. Land Development Automobile use has influenced the form and extent of land development in the United States, leading in large part to sprawl (i.e., low-density, inefficient land use patterns; Burchell et al. 2002). The land allocated to automobile infrastructure poses a cost to society: it could otherwise be used for farms, open space, homes, businesses, and other facilities, with asso- ciated environmental, economic, and public health effects (Delucchi and Murphy 2008). Factors that have influenced land development patterns in the United States can be divided into two categories—market forces and public policy decisions. In terms of market forces, CVs/AVs could increase safety and convenience of vehicle travel, lowering transportation costs and thus increasing people’s willingness to travel farther and adding to sprawl. On the other hand, if the technology is incorporated into transit vehicles and shared vehicles, the effect would be the opposite—decreasing vehicle ownership and use in favor of transit and shared mobility. The result could be growth in higher-density areas. As another effect, Level 5 AVs could reduce the need for parking adjacent to destinations, which is currently mandated through parking minimums for new developments. If this were the case, then parking requirements may be altered or eliminated, and parking in urban areas could be reused for other land uses that more directly benefit society. Mobility Access to transportation is essential for a high quality of life for nearly all Americans yet is often a significant challenge for aging adults, youth under age 16, and individuals with dis- abilities. Many live in car-dependent areas but do not drive, and transit alternatives may be geographically inconvenient or inaccessible. AVs represent an opportunity to reduce this negative exter- nality. By leveraging the existing infrastructure that favors motor vehicles, fully automated vehicles (Level 4/5) may offer significant—potentially transformative—improvements in mobility for aging adults, youths, and individuals with dis- abilities. This is particularly true for those who live in areas with limited alternative modes. The benefits of less-than-full auto- mation and CVs are unclear and probably fewer because of the demands on the human driver behind the wheel, but this limitation may depend on the specific disability. Positive Societal Outcomes There are several potential ways in which the private sector could create desirable outcomes for society. The first is safety. If safe AVs and CVs were developed and used wisely and responsibly, the current public health crisis in the U.S. trans-

13 portation system could be reduced, if not mitigated. This out- come may take many decades, given rates of vehicle turnover, but the long-term impact could be significant. The second relates to mobility. If safe and usable Level 5 AVs were devel- oped, mobility could be increased for millions of Americans who currently have limited mobility. A third relates to pollu- tion and congestion. If the potential increase in VMT created by AVs and CVs were mitigated, or if VMT were decoupled from fossil fuels, there could be enormous environmental and public health benefits. Similarly, if SAVs were widely available and widely used, congestion, pollution, and land use benefits could occur. The fourth relates to liability issues. Uncertainty over the magnitude of the liability risks may deter and delay introduction of these technologies, which have the potential to greatly benefit society. A review across externalities shows common actions in which the private sector (i.e., consumers or producers of AVs and CVs) should engage to enable these positive societal outcomes. For CVs, producers need to implement effective safety, mobility, and environmental applications (through evaluation and testing), and consumers need to use them widely and appropriately. The analysis assumes that V2V safety applications will be implemented through federal man- date, so this implementation is not one of the enabling actions. Such actions will benefit crash, congestion, and pol- lution externalities. However, there is little evidence from the literature that CV applications will affect land development or mobility externalities. For AVs, producers need to develop and sell AVs that are safe and efficient. They also need to act upon communications with road operators to ensure that the infrastructure (e.g., lane striping) is in place to support safe and efficient operation or to enable the changes and maintenance necessary to ensure operation. To maximize social welfare, consumers need to pur- chase safe AVs and use them appropriately but not increase their appetite for travel (more trips, more VMT, and more sprawl) or vehicle size (larger vehicles). SAVs in particular could offer many of the benefits of AVs while not increasing travel. If, in addition, vehicles are coordinated with transit to solve last- mile connectivity, they could increase the use of transit. These enabling actions are listed in Table 2 for CVs and in Table 3 for AVs. However, not all enabling actions lead to positive outcomes. Some conflicts exist. In particular, the increased mobility for aging adults and individuals with dis- abilities may increase VMT, leading to more congestion and pollution. Aligning Public- and Private-Sector Interests Producers and consumers of AVs and CVs are the primary private-sector actors whose market decisions will determine whether and how AVs and CVs benefit society. Produc- ers include automobile manufacturers (e.g., Ford, Toyota); technology firms (e.g., Google, Apple); and Tier 1 suppliers (e.g., Delphi, Bosch).1 Consumers include private individuals Actions of Producers and Consumers Externalities Crashes Congestion Land Develop- ment Air and Noise Pollution Mobility Producers develop and sell interoperableV2V or V2I mobility applications X X Producers develop and sell interoperableV2V or V2I environmentapplications X Consumers purchase vehicles with V2V/V2I capabilities X X X Consumers purchase and use aftermarketV2V safety applications X X X Consumers are attentive to V2V and V2I safety warnings in vehicles X X X Consumers use V2V/V2Imobility applications X X Consumers are attentive to V2V and V2I mobility messages in vehicles X X Consumers use V2V/V2I environment applications X Consumers are attentive to V2V and V2I environmentalmessages in vehicles X Note: X indicates a reduction of a negative externality. Table 2. Private-sector actions enabling positive outcomes of CVs. 1 USDOT and university research institutions are also producers of CV tech- nology but not private-sector actors.

14 and private-sector fleet owner/operators who would buy and use AVs and CVs. Their choices about what technologies to develop, when to deploy them, and how to use them could lead to (or fail to lead to) beneficial mechanisms and outcomes. However beneficial, the desired outcomes may not actually be realized because many of them accrue to society rather than to either the producers or consumers of AV and CV technol- ogy. Consumers may be unwilling to pay for expensive tech- nology if most of the benefits go to others, and consequently, producers may be less willing to develop them. Thus, there is less incentive for producers and consumers to take actions that would achieve beneficial outcomes. Note: The term AVs in this table refers to Levels 3–5. When a specific level of automation is the subject of the action, it is labeled accordingly (e.g., Level 5). X indicates a reduction of a negative externality. X(-) indicates an increase of a negative externality. Actions of Producers and Consumers Externalities Crashes Congestion Land Develop- ment Air and Noise Pollution Mobility ProducerActions Producers develop and sell safe AVs X X X Producers of AVs act upon communicationswith road operators about infrastructure/maintenance necessary to ensure safe operations and system efficiency (across different use cases/operating conditions) X X X Producers develop and sell connected AVs that harmonize traffic flow X X Private, shared-vehicle services purchase and operate SAVs X X Private, shared-vehicle services prioritize ride-sharing and linkages with line-haul mass transit X X X Developers build fewer parking facilities or build parking facilities that can be adapted to other purposes X X Producers develop and sell AVs that are lower polluting X Producers develop and sell AVs with eco-driving operating objectives X Producers develop and sell Level 4/5 AVs that are usable by aging adults and individualswith disabilities X(-) X(-) X ConsumerActions Consumers purchase safe AVs X X X Consumers follow safe AV maintenance and operating procedures X X X Consumers purchase connected AVs that harmonize traffic flow X X Consumers of AVs minimize VMT growth, though the technology decreases travel cost and enables mobility among some who cannot otherwise drive X X Consumers of AVs do not drive farther for housing, even though the technology decreases travel cost X X X Consumers use SAVs rather than privately owned AVs to minimize VMT growth X X X Consumers use Level 5 vehicles to avoid parking in urban centers X(-) X X(-) Consumers purchase AVs that are lower polluting X Consumers purchase AVs with eco-driving operating objectives X Aging adults and individualswith disabilities (consumers) purchase Level 4/5 AVs X(-) X(-) X Aging adults, youth, and individualswith disabilities (consumers) use Level 4/5 SAVs X(-) X(-) X Table 3. Private-sector actions enabling positive outcomes of AVs.

15 Congestion offers a useful illustration of this phenomenon. Each driver that takes to a busy road not only experiences congestion but also adds to the congestion of his or her fellow travelers. While drivers bear their own congestion costs (their own time spent in traffic, added fuel use, etc.), they do not bear the costs they impose on other drivers. This means that drivers have only some incentive to reduce congestion, even though it would be enormously beneficial to society. Much of the cost of congestion (and the benefit of reducing it) is external to each driver’s decision making. This leads to a sub- optimal level of congestion for society, and society is worse off than it would be if the drivers bore the full cost of adding to congestion or received the full benefits of reducing it. This is an example of an externality. An externality is an effect that one party imposes on another party without compen- sating for the effect if it is negative or charging for it if it is positive (Buchanan and Stubblebine 1962). The free market allocates resources inefficiently and produces suboptimal outcomes in the presence of externalities (i.e., when the costs faced by individual actors do not include the costs of their actions upon fellow citizens).2 Externalities have important implications for realizing the benefits of AVs and CVs. Suppose, for example, that AVs were more efficient than traditional vehicles—reducing the sharp acceleration and braking that contributes to congestion and perhaps decreasing the required safe following distance between vehicles (due to faster and/or coordinated reaction time among AVs through connectivity). Users of such AVs would increase roadway efficiency and reduce congestion not only for them- selves, but for all other road users (all other things being equal). This would have positive effects on public health, the economy, and the environment. However, if the costs of congestion (or the benefits of reducing it) remain external, the market for effi- cient AVs would be weaker. There would be less of an incentive to produce and consume efficient AVs, even though this would be better for society. This effect is not limited to congestion. All of the five areas of AV and CV impact (safety, congestion, pollution, land use, and mobility) involve existing driving externalities that could result in suboptimal social welfare because the producers or consumers of the AVs and CVs might not consider the full social costs and benefits when making choices. Externalities are one reason governments interfere in mar- kets. Social welfare can be increased if externalities are internal- ized so that the costs faced by individual actors represent the social costs of their actions (including the externalities). This can be done with subsidies, user fees, mandates, and privileges to equalize the public and private benefits. As one example, car- pooling reduces congestion, but the costs and inconvenience of doing so accrue to the driver, while the benefits of carpool- ing (reduced congestion) typically accrue to other travelers. There is little incentive for anyone to carpool. High-occupancy vehicle (HOV) lanes attempt to change that. They internalize the positive externalities of reducing congestion by enabling carpoolers to themselves bypass congestion and get to their destinations faster. Social welfare can also be improved if the negative externalities are reduced, even if the externalities are not necessarily internalized in market decisions. 2 Even if individual actors do not pay for the costs they impose on others, those costs still exist and are real, but are borne by society at large.