The views expressed here are those of the author and do not necessarily reflect the
views of Morningstar.
In earliest times, men worked on the task of moving people and things from one place to another. A very few years later, a second problem arose: keeping one bunch of moving things from colliding with something else. Today, a great deal of attention is being paid to the introduction of driverless cars, and one of the benefits of driverless cars will be the reduction of accidents. Every year, more than 30,000 people are killed in the United States alone, and 10 times that number injured. The emotional and monetary cost of all this colliding is an obvious target for improvement. Engineers are working to improve cars’ navigation capabilities with radar, lidar, electronic maps, and considerable computer power, so that each vehicle can make its way unscathed to its destination.
That isn’t the only way to solve the problem. Railroads and airlines have gone to centralized control systems where a combination of humans and computers control all the trains and planes in a large region. So, we have two competing system designs, one controlling each vehicle autonomously, the other controlling the traffic in a top-down fashion. Let’s think about how some of these systems evolved.
About 1830, ingenious engineers in the United Kingdom and then the United States started building railroads. This was a big deal. For the first time, people and things could move around at high speed and low cost. Before long, the railroads had more workers than any other sector except farming. However, the life of railroaders in the early days was dangerous. For example, coupling train cars together was very risky because a worker had to go between the cars to fasten the couplings by hand. This was a tricky job for anyone, and even harder to do when drunk, which many workers in those days were most of the time. Eli Janney solved the problem by inventing the automatic coupler in 1873.
Starting and stopping trains also was very difficult, because there were manually operated brakes on each car. Brakemen had to perch on top of the cars and turn a wheel connected to a long shaft that set the brakes. They also had to jump to the next car. It was easy to fall off the top of a moving car, and if it happened in a Nebraska blizzard, it might be months before anyone found your body.
Then, George Westinghouse1 patented the Westinghouse air brake in 1872, and it debuted on the Pennsylvania Railroad that same year. The brakes were controlled by compressed air in a pipe, which in turn was controlled from the locomotive cab and extended to each car in the train.
The final step in making trains safer was to keep two trains on the same track from running into each other head-on. The first rudimentary signal systems were devised in the 1830s. When the telegraph was invented in the 1840s, railroaders immediately recognized its value. Railroad officials could put a telegraph operator in each station; when the train stopped, the engineer would get the all-clear to proceed from the station. This was not only profitable for the telegraph company, but the railroad rights of way were an ideal place to put up the telegraph lines. The locomotive engineers got “train orders” on slips of paper, and this system, with improvements, lasted until the 1960s. Generations of little boys loved to beg the engineer for his used train orders and put them in a collection.
Today, most railroads use a centralized traffic control, or CTC, where a dispatcher’s office controls all the trains in a large region. Each train moves only under CTC control.
Accidents Still Happen
Still, despite the advent of CTC, railroad accidents still occur today. That’s mostly because some railroads don’t use CTC, with tragic consequences. On May 12, 2015, a northbound Amtrak passenger train derailed and wrecked on a curved section of track in Philadelphia. Eight people were killed and 200 injured. The train was traveling at 102 mph in a 50 mph zone. The engineer made the error of entering the curve at twice the speed limit, with catastrophic results.
The accident would have been prevented if the train was equipped with positive train control, or PTC, which was used elsewhere in the same busy Northeast corridor route, but PTC had not been installed in this part of the track. An older system was in place on the southbound track of the curve but not on the northbound track, where the wreck occurred. Amtrak came up with bureaucratic excuses for not installing either of these systems. The proximate cause was pilot error, but Amtrak management was clearly derelict in not installing a modern CTC system on the busiest track in the United States.
An even worse disaster occurred in the Canadian town of Lac-Mégantic in Quebec. On July 6, 2013, a 74-car freight train was parked six miles away from the town. The tank cars were filled with Bakken Formation crude oil. The Bakken crude is light oil, which is the most valuable variety but more flammable than most crude oil. The engineer parked the train on the main line and went to spend the night in a hotel in town. He did not set the brakes correctly. The train rolled and reached Lac-Mégantic at a speed of 65 mph. It hit a curve and derailed. The tank cars broke open. A blanket of oil flowed into the downtown area of the city and exploded in a gigantic fireball. Forty-two people were killed. At least 30 buildings were destroyed.
The railroad was the Montreal, Maine and Atlantic Railway, which was owned by a small company. The company skimped on maintenance of the track and the locomotives; the lead locomotive on the doomed train was actually on fire just before the accident. The number of maintenance shortfalls and engineer errors were numerous, but a modern CTC could have prevented the accident, perhaps by warning the engineer not to park on the track. I say only “could have,” because the level of incompetence of the railroad company might well have outwitted a computerized safety system.
Aviation has followed the same general path of railroads with air traffic control systems giving orders to each plane in the region. Although the current system could use an update, most years no commercial airliner has a fatal accident.
Centralized Network for Cars
How will the driverless car network evolve? There are strong technical arguments for having a single giant computer controlling every car, certainly on freeways. This would allow for more efficient highway travel at a higher volume. Cars could be packed more densely and move at higher consistent speeds, because individual drivers would not be at the wheel trying to keep a safe distance from each other or swerving in and out of traffic, causing accidents. For local driving, there are good arguments for each vehicle having its own guidance system. It will be interesting to see how it works out.
The advantages of a computerized road network over individual smart cars are explained by a new must-read book by Yuval Noah Harari2:
“Indeed, if we forbid humans [from driving] vehicles altogether and give computer algorithms a monopoly over traffic, we can then connect all vehicles to a single network, thereby rendering car accidents far less likely. In August 2015, one of Google’s experimental self-driving cars had an accident. As it approached a crossing and detected pedestrians wishing to cross, it applied its brakes. A moment later, it was hit from behind by a sedan whose careless human driver was perhaps contemplating the mysteries of the universe instead of watching the road. This could not have happened if both vehicles had been guided by interlinked computers. The controlling algorithm would have known the position and intentions of every vehicle on the road and would not have allowed two of its marionettes to collide. Such a system would save lots of time, money, and human lives.”
What It Means for Investors
Driverless cars, with or without central control, would cause lots of changes. These cars are likely to be electric, as electric vehicles are easier for computers to control. The design of cars would be freed up. Imagine losing a complex gasoline engine and its exhaust cleaning system, not to mention the steering wheel and driver’s seat. The reason to buy a Cadillac instead of a Chevy would be only the quality of the TV and video game console, or, for worker bees, extending a home office into their car. Your car could be like a detachable room plugged into a port in your house. It detaches from your home, drives to your work address, and plugs into a similar port in your office building. Handy on the occasions you need to meet someone face to face.
The existing network of gas stations would be obsolete. There would be a lot of battery charging to do, but not along the roadside. Oil and gas stocks would evolve, with less consumption of petroleum but an increase in natural gas use as the electric utility fuel. The electric utility industry would profit from an increase in demand with a shift toward night battery recharging. The insurance industry would be hurt when individuals stop being liable for accidents. You don’t need liability insurance when you ride in a train or an airplane, and you won’t when your car is under centralized computer control.
Labor markets would be restructured. Suburban parents would no longer bear the brunt of driving the kids to ballet school and the soccer field, freeing more time in the complex work/life juggle. Seniors would regain the power to travel easily, for work and social events. At the same time, millions of professional drivers of trucks, buses, Ubers, and delivery vans would be redundant. There would be a lot of restructuring to do, with consequences for investors. Some of the restructuring would require changing the shape of cities. Revolution in the transport network demands novel urban architecture.
It will be interesting to see whether we choose to control driverless cars autonomously or through a central command. Either way, they will permanently alter society.
1 George Westinghouse (1846–1914) was a major inventor. He created electric signaling to replace the train order paper system. He also bought Nikola Tesla’s patents for alternating current. He had a marvelous battle with Thomas Edison, who favored direct current. Westinghouse won that war, and today your house has AC. 2 Harari, Y.N. 2017. Homo Deus: A Brief History of Tomorrow, Harper Collins, p. 315.
This article originally appeared in the February/March 2018 issue of Morningstar magazine. To learn more about Morningstar magazine, please visit our corporate website.
Ralph Wanger does not own shares in any of the securities mentioned above. Find out about Morningstar's editorial policies.