Two Potential Market Outcomes

Complementary Infrastructure Requirement

Benefits of Self-Driving Cars

Costs of Self-Driving Cars

Winners

Losers

System Evolution

Driverless (autonomous) vehicles is one of the hottest topics being discussed in the news lately. Some writers have been touting the enormous benefits adoption of driverless cars will bring, emphasizing the utopian scenario associated with the new technology. Others have noted the large industries dislocations their adoption will create, emphasizing the dystopian scenario. This analysis is my attempt to better understand what the market for driverless cars will entail.

Two Potential Market Outcomes

There have been two general market scenarios bandied about in discussions of autonomous vehicles:

• Personal Self-Driving Cars (PSDC): In this scenario people generally own their own vehicles, but instead of people doing the driving, the vehicles drive themselves. This market outcome would yield a vehicle environment that looks relatively similar to the one that exists today, except that cars would have no drivers.

• Shared Self-Driving Cars (SSDC): In this scenario people don’t own their own vehicles. Instead, third-party providers of transportation services own fleets of driverless vehicles, which people hail when they need to go somewhere. In other words, the SSDC scenario conflates autonomous vehicle with peer-to-peer (or sharing) technologies. This market outcome would yield a vehicle environment that is radically different from the one that exists today.

Complementary Infrastructure Requirements

The successful adoption of new technologies always requires the presence of other existing, complementary technologies that enable and support the new technologies. Successful adoption of driverless cars requires the following infrastructure:

• Self-Driving Technologies: These include all the hardware and software technologies in the vehicles that enable them to drive themselves.

• Vehicle-to-Vehicle (V2V) infrastructure: This includes all the hardware and software technologies in the vehicles that enable them to communicate with other vehicles on the road, so as to avoid collisions with each other.

• Vehicle-to-Infrastructure (V2I or V2X) infrastructure: This includes all the hardware and software technologies in the vehicles and on the roadways that enable vehicles to avoid collisions with roadside infrastructure.

• Legal Infrastructure: This includes all the laws and regulations needed to support autonomous vehicle systems, and in particular laws regarding who is liable when autonomous vehicles cause accidents.

Regarding V2V and V2I technologies, the Department of Transportation’s Intelligent Transportations Systems provides five categories of Connected Vehicle Technologies:

• Standards: Connected Vehicle Standards are rules that provide the software programming codes, definitions, and formats needed to create interoperable, consistent, and seamless communications exchange among shared information systems and devices.

• Human Factors Research: Connected Vehicle Human Factors Research is focused on understanding, assessing, planning for, and counteracting the effects of signals or system-generated messages that take the driver’s eyes off the road (visual distraction), the driver’s mind off the driving task (cognitive distraction), and the driver’s hands off the steering wheel (manual distraction).

• Core Systems: Connected Vehicle Core Systems provide the functionality needed to enable trust relationships and data exchanges between and among mobile and fixed transportation users.

• Certification: Connected Vehicle Certification is the process of ensuring that system components, manufactured according to connected vehicle interoperability requirements, to perform as intended. Certification ensures that users can trust that the components will work within the system.

• Test Bed: Connected Vehicle Test Beds are real-world, operational test beds that offer the supporting vehicles, infrastructure, and equipment to serve the needs of public- and private-sector testing and certification activities.

Perhaps the messiest issue surrounding self-driving vehicles is the liability issue. In fact, many media sources have indicated that the primary obstacles to mass adoption of autonomous vehicles are legal, not technological in nature. The Economist cites a RAND study that proposes two possible solutions for addressing the liability issues:

A study in 2009 of the legal risks of increasingly autonomous cars by the RAND Corporation, a research body, suggested two possible solutions: changing the liability laws to require courts to take the benefits of driverless technology into account when punishing carmakers for any failings; and limiting motorists’ ability to sue in state courts when driverless technology mandated by federal laws fails to prevent an accident.

Personally, I don’t find either of these two solutions particularly satisfying. They beg the question as to why the auto industry would get such a pass on liability, while, say, the pharmaceutical industry doesn’t.

On the liability issue, Chunka Mui and Paul B. Carroll, in The New Killer Apps: How Large Companies Can Out-Innovate Start-Ups, note that while airplanes can fly themselves, liability issues require every flight to have two pilots. However, they go on to cite the Rand Study, which “suggested that government might intervene and mandate self-driving cars if they prove to be half as safe as Google claims.”

Only if the automobile manufacturers are held liable for accidents will they have the proper incentives to minimize (or, more accurately, optimize) damages associated with accidents. If this liability scenario prevails, then Chunka Mui, in “Google Car + Uber = Killer App,” notes that

… with the transition to driverless cars, much of car insurance will probably take the form of product liability insurance bought by manufacturers, as opposed to personal liability insurance bought by drivers.

Benefits of Self-Driving Cars

The benefits associated with mass adoption of self-driving cars are widely touted in the media. These include:

• Greater Safety: By bar, the largest benefit of autonomous vehicles is the greater safety they will bring. Human error is responsible for most accidents. Automation will eliminate human error, thereby greatly reducing the number of driving-related accidents, injuries, and deaths.

• Time Savings: When freed up from the responsibility of driving, people will be able to spend their travel time on more productive activities.

• Greater Traffic Efficiency: The precision driving enabled by automated vehicles will enable them to avoid most of the congestion currently encountered on roadways. This will save driver time, gasoline usage, and pollutant emissions.

• Reduced Environmental Impact: Autonomous vehicles are better able to optimize driving patterns so as to minimize fuel usage and emissions.

• Greater Access: Mass adoption of driverless vehicles will provide people with greater access to transportation services, including people who are too young or old to drive or those who are handicapped.

• Lower Costs of Transportation: The costs to users of transportation services will decrease significantly under either the private (PSDC) or shared (SSDC) vehicle scenarios. In particular, the greater safety of autonomous vehicles systems will relieve users from the costs of car insurance and vehicle safety equipment, and their greater efficiency will reduce fuel costs. In a shared vehicle economy, users will be further relieved from the costs of vehicle ownership, maintenance, and parking,

Costs of Self-Driving Cars

Other than the liability issue, few writers have discussed any serious disadvantages to the use of driverless vehicles (aside from industry losers, which is discussed below). Randal O’Toole, in “Driverless Cars Yes, V2I No!” is the exception; he posits four problems associated with autonomous vehicle systems:

First, V2V and V2I communications pose serious security risks for travelers and cities. With V2V communications, an automobile that suffers a fender-bender would communicate to all nearby vehicles that they ought to take a different route to avoid congestion.

That sounds good, but what happens when someone hacks the system and puts out radio signals in a hundred or a thousand critical urban intersections that effectively shut down traffic in an entire city?...

Second, V2I communications will allow the nanny state to monitor and control when and where you travel. For example, PC Magazine observes that V2I is “so accurate a revenue-hungry town could write tickets for doing 57 in a 55 zone.”

Worse, suppose your state decides to cut per capita driving in half, which isn’t far fetched considering that in 2008 the Washington legislature passed a law mandating such a reduction by 2050. With V2I communications, the government could decide you have driven enough and simply shut off your car

Third, what happens when all cars are dependent on V2I systems that the government can’t afford to maintain? The federal government is notorious for funding capital projects and then providing inadequate money to maintain them, and state and local governments are little better.

Finally, V2V and V2I communications will be an unnecessary added expense to auto ownership. The Department of Transportation says it won’t even have a draft of rules mandating V2V before 2017, and such rules won’t possibly go into effect before 2018. Yet partially autonomous cars that improve safety by providing steering assistance and collision avoidance are already on the market.

Winners

There will definitely be winners and losers when autonomous vehicles become widely adopted. Winners include

• Transportation Users: People will experience lower time and money costs associated with using transportation services.

• Makers of Autonomous Vehicles

• Providers of Autonomous Vehicle Operating Systems, such as Google

• Operators of Fleets of Autonomous Vehicles, such as Uber

• Providers of Passenger Entertainment

The extent to which fleet operators will benefit from the new technology will depend on whether the personal (PSDC) or shared (SSDC) vehicle system is adopted, as previously discussed.

Several authors (e.g., Morgan Stanley, “Autonomous Cars: Self-Driving the New Auto Industry Paradigm”) have noted that the market for autonomous vehicles will resemble the smartphone market, in which hardware is relatively commoditized, and the bulk of value is created and extracted by systems and content providers.

Losers

Mass adoption of autonomous vehicles will create losses for various industries, where the extent of the losses will depend upon the eventual market scenario that prevails, personal (PSDC) or shared (SSDC) vehicle systems.

Big Losers with Either PSDC or SSDC

• Automobile Insurers: With fewer accidents, automobile insurers will provide a much smaller role in the new industry. As indicated above, automobile insurance for users may disappear entirely and be replaced by product liability insurance for car manufacturers.

• Healthcare Service Providers: With fewer accidents, there will be much less need for healthcare services to treat people involved in accidents.

• Healthcare Insurance Providers: With fewer accidents, there will be much less need for healthcare insurance to pay for healthcare services to treat people involved in accidents.

• Government: With cars on the road that are programmed not to break laws, significant amounts of government revenues – those associated with parking and speeding tickets– will disappear. 

Big Losers with SSDC

• Automotive Financers: With significantly fewer cars on the road and a potentially different structure of ownership (fleet owners, rather than private owners), automobile financers will lose a significant portion of their business.

• Parking Industry: With significantly fewer cars on the road combined with greater utilization of existing cars, much less parking space will be required.

• Drivers: Autonomous vehicles will eliminate the need for most car, truck, and bus drivers. Note that in the future UPS and FedEx drivers won’t be needed to drive their delivery trucks. However, they will still be needed to load packages onto their trucks at the warehouses and remove the items from their trucks and deliver them into the hands of the recipients at homes or offices.

• Automotive Aftermarket: With significantly, fewer cars on the road, combined with a change in the needs of users – from driving and performance accessories to entertainment accessories – much of the auto aftermarket as it exists today will disappear.

• Rental Car Companies and Public Transportation Systems? Zack KJanter, in “How Uber’s Autonomous Cars Will Destroy 10 Million Jobs and Reshape the Economy by 2025” claims that Uber’s driverless car network will wipe out rental car companies and public transportation systems. While this is certainly a possibility, it’s not necessarily a foregone conclusion. I would venture to say that while there is some overlap, the people who use Uber are not the same people who use public transportation or rental cars. In other words, I claim that the three different groups cater to different market segments. Rental car companies cater mostly to people traveling medium to medium-long distances. These people trend to take taxis for short distance travel and airplanes for long distance travel. Uber, on the other hand, is currently strongest in short distance markets. Public transit, in contrast, offers greater efficiency at transporting groups of people along a route, which enables lower prices to transit users. With self-driving cars (Uber) and self-driving public transportation vehicles (buses), the cost differential associated with joint ridership does not disappear, which means the two sets of transportation (Uber and buses) could very well co-exist together in the market, even if/when vehicles become autonomous. So, as I said, with driverless vehicles, Uber could conceivably grow and take over the rental car and public transportation markets, but it’s not necessarily a foregone conclusion. They could all continue to coexist together in the market as they do currently.

System Evolution

Which Scenario, PSDC or SSDC, Is Most Likely to Prevail?

In the spring of 2014 Uber announced that it would eventually transition its fleets to self-driving cars. It seems like after that time, all media reports on autonomous cars have focused exclusively on the shared vehicle scenario (SSDC). Is the SSDC scenario now the foregone conclusion, or is the PSDC scenario still possible?

A shared vehicle system would require most people to give up on owning their own vehicles and turn, instead, to renting shared vehicles for all of their travel needs. A decade or two ago, this might have seemed a highly unlikely situation. In the past, authorities have tried many times to get people to give up their cars and instead turn to public transportation for their travel needs. However, all such attempts have been futile; people simply would not give up their cars. The most likely explanation is that transportation authorities failed to consider that private car ownership in the US has long been a symbol of freedom, pride, personality, and status.

However, things seem to be changing. The Millenials appear to have a very different attitude towards driving and car ownership than their predecessors. As Jordan Weissmann, in “Why Don't Young Americans Buy Cars?” states,

…today's teens and twenty-somethings don't seem all that interested in buying a set of wheels. They're not even particularly keen on driving.

Toyota USA President Jim Lentz offered up a fairly doleful summary of the industry's challenge. "We have to face the growing reality that today young people don't seem to be as interested in cars as previous generations," Lentz said. "Many young people care more about buying the latest smart phone or gaming console than getting their driver's license."

Millenials’ lack of car ownership has been blamed on (i) their high levels of unemployment and thus inability to afford to own their own cars; (ii) their propensity for city-living, where public transportation is easily accessible; and (iii) their propensity to prefer connecting with others through virtual, rather than physical, presences.

If the younger generations continue to eschew car ownership even after markets rebound and their finances become more conducive to owning their own cars, then the SSDC scenario seems likely, at least for most short-distance travel.

However, the logistics of medium- and long-term travel, especially for users located in suburban and rural areas, combined with the quirks of people, will still make private ownership of vehicles desirable for a significant portion of the population. In particular,

• People hate waiting for things. One of the biggest advantages of private car ownership is the ability of people to use their cars on the spur of the moment. Even with efficiently designed fleet of shared cars, people in suburban and rural locations would have to either book their rides in advance and/or wait several minutes for the shared vehicles to arrive at their pick-up locations.

• People like having their own space where they can leave their own stuff and know it will be there the next time they want to use it. Many people would be loathe to have to bring their stuff with them each time they get into a vehicle and take all their stuff with them each time they leave.

• Cars serve as more than just a means of traveling from one location to another. As mentioned above, cars provide people with a sense of freedom, pride, personality, and status. And especially if people end up traveling longer distances with autonomous cars because they are freed up from the responsibility of driving, they will especially want to outfit their transportation pods to be as personal and comfortable as possible. 

Operating Systems

The AI nature of operating systems in autonomous cars means that each vehicle has the ability to learn from the experiences every other vehicle in the same system. This creates a natural monopoly scenario. That is, the AI nature creates a scenario in which it is most efficient to have a single operating system control all the cars on the road, gaining experiences from each vehicle and transmitting each of those experiences to every other vehicle.

If, indeed, there ends up being a single operating system that is used on all vehicles, then who will own this operating system? Will it be privately held and controlled by one company, say, Google? Will it be operated and controlled by the government? Will it become open source?

A much more realistic scenario is that autonomous vehicle operating systems will roll out much like the electronic medical records (EMR) systems have been introduced: There will be a plethora of different systems offered by different suppliers. Currently, the different EMR systems are having great difficulties in providing seamless communications across different information seekers. If the same thing happens in the market for autonomous vehicles, there will most likely be a great many accidents during the initial stages of adoption as the different operating systems work out all the bugs affecting inter-system communications. Will the government stand by as the free market moves toward consolidation, or will it step in and pick a winner, justifying the necessity of doing so by the number of lives it will save?

Will We Experience Jevon’s Paradox?

Many of the proponents of self-driving cars in the shared vehicle scenario claim that the greater safety and utilization of vehicles will lead to drastic reductions both in the number of vehicles on the road, as well as the in costs of using transportation services. However, wouldn’t such a scenario be likely to lead to a Jevon’s paradox? That is, wouldn’t the new driving environment (lower costs, less congestion, greater convenience) cause demand for transportation services to skyrocket, thereby eventually leading to even larger amounts of congestion, pollution, and demand for vehicle services than we currently have?