Will self-driving trucks increase the demand for truck drivers?

by on February 2, 2018 at 12:09 pm in Economics, Web/Tech | Permalink

For that reason, Woodrow says that he saw their version of self-driving trucks as complementing humans, not replacing them. To make their case, Uber created a model of the industry’s labor market based on Bureau of Labor Statistics data. Then, they created scenarios that looked at a range of self-driving-truck adoption rates and how often those autonomous trucks would be on the road in comparison to human-driven vehicles.

Their numbers for autonomous-truck adoption are intentionally very aggressive, Woodrow says, corresponding to 25, 50, and 70 percent of today’s trucks being self-driven. These do not reflect an Uber prediction that between 500,000 and 1.5 million self-driving trucks will be on the road by 2028, but rather they allow the model to show the dynamics in the labor market that might result from widespread adoption. “Imagine that self-driving trucks are incredibly successful and impactful,” he says. “What would that mean?”

The other set of numbers in the model—the utilization rate of the self-driving trucks—is the component that leads Uber to a different analysis of the effect that these vehicles will have on truckers. Basically, if the self-driving trucks are used far more efficiently, it would drive down the cost of freight, which would stimulate demand, leading to more business. And, if more freight is out on the roads, and humans are required to run it around local areas, then there will be a greater, not lesser, need for truck drivers.

That is from Alexis C. Madrigal at The Atlantic.

1 Anonymous February 2, 2018 at 12:23 pm

In which “smoking bird shit” appears at The Atlantic.

Seems plausible though, that early versions of elf-driving(*) will not be a full human replacement.

* – a humorous typo I learned on the internet

Reply

2 Careless February 3, 2018 at 1:20 am

Anonymous February 1, 2018 at 7:48 pm
Come on Careless, TMC goes full “Fake dossier paid for by Hillary campaign used as excuse to spy on Candidate Trump”

That is not the way FISA court works. You must present to a judge a case that espionage is in play.

So, how’s that working out for you?

Reply

3 Anonymous February 3, 2018 at 1:42 pm

I already answered, back at the place.

Reply

4 Job Bones February 2, 2018 at 12:27 pm

Before I’ll trust any research coming out of Uber, I’d like to see third-party analyses/reviews. They’ve gotten rid of some of the completely untrustworthy people at the top but we know of their past lies, deceptions and gaslighting, and regaining trust is going to mean they need to convince reputationally-rich researchers to endorse their changed ways.

Reply

5 Anonymous February 2, 2018 at 3:23 pm

Third-party analyses are what the entire article is about

Reply

6 Dave Smith February 2, 2018 at 12:28 pm

Yes, self driving trucks will create jobs for other types of truckers, but I doubt we will end up with more of them. This sounds like “tax cuts”/”deficit spending” will pay for itself.

Reply

7 Jeff R February 2, 2018 at 7:11 pm

At the very least, if they’re right about falling freight costs leading to higher demand, that would stimulate demand for forklift drivers.

Reply

8 apoptosis February 3, 2018 at 1:09 am

self-driving forklifts

Reply

9 Jeff R February 3, 2018 at 7:45 am

Tougher to pull off, in my opinion.

Reply

10 SgtDad February 4, 2018 at 3:22 pm

Amazon already has, I am told.

11 Student February 2, 2018 at 12:49 pm

This is an interesting thought but I don’t know.

Let’s see… Redbox drove down the cost of movie rentals, which increased the demand for renting movies and led to more video store workers… not so.

Reply

12 Borjigid February 2, 2018 at 2:22 pm

I think the argument rests on long haul trucking being a different market than short haul trucking, with only the former becoming automated.

So the comparison would be Redbox driving down the cost of movie rentals, increasing the demand of renting movies and leading to more jobs in Hollywood making movies.

Still unlikely, but not prima facie ridiculous.

Reply

13 JWatts February 2, 2018 at 4:19 pm

” increasing the demand of renting movies and leading to more jobs in Hollywood making movies.”

No, that would directly correlate with autonomous vehicles lowering the cost of freight and the lowered cost of freight leading to a lower total cost of goods and thus more production of total goods.

“I think the argument rests on long haul trucking being a different market than short haul trucking, with only the former becoming automated.”

I think this is correct, but I’d doubt the assumption. I can’t imagine there would be much of a window between autonomous long haul trucking and autonomous short haul trucking.

Reply

14 PaulL February 2, 2018 at 6:01 pm

Long haul – largely on interstates – is a much simpler computational problem than driving around crowded city streets.

There’s also an interesting thought on the quality of the jobs. Short haul trucking is much better for family life than long haul trucking. If it were the case that automated long haul led to more short haul, then that would net improve the quality of life of truckers.

Reply

15 chuck martel February 3, 2018 at 9:12 am

Nobody cares about the quality of life of truckers, except maybe the truckers themselves. Since most of them are paid by the mile, their main concern is how many hours they can drive per day and at what speed.

16 Joshua February 6, 2018 at 7:31 am

While the “problem” of driving in the city might be harder, it’s much easier to deal with issues. You could easily set up a team that dealt with rare cases where a truck might have issues and they would have very little geography to cover. i.e. it’s much easier to fix a flat in the city where all problems happen within 10 miles then it is to send some out to fix a flat 200 miles away in the middle of nowhere.

17 MM February 2, 2018 at 11:31 pm

“Their numbers for autonomous-truck adoption are intentionally very aggressive, Woodrow says, corresponding to 25, 50, and 70 percent of today’s trucks being self-driven”
They needed aggressive adoption rate to change market completely. That may not happen!! Reality would be murkier with job losses for a long period till adaption makes market cheap for demand to go up.

Reply

18 Bill February 2, 2018 at 12:51 pm

Took a course on autonomous vehicle law, and none of the presenters came even close to the prediction that Uber makes. I was surprised at the low numbers until around 2040. What you are more likely to see autonomous vehicles that disengage from autonomy to give control back to the driver. Which means we will have more well rested truck drivers.

Remember: Uber has an interest in hyping the numbers to goose an IPO.

What was interesting about the program was driverless car hacking. The best place to test whether your system is hackable is Russia where they have apparently a great skill set.

Reply

19 JWatts February 2, 2018 at 4:48 pm

“What you are more likely to see autonomous vehicles that disengage from autonomy to give control back to the driver. Which means we will have more well rested truck drivers.”

This seems more likely. At least for the next decade.

A vehicle that starts beeping at the driver and if doesn’t get a response pulls the truck over to the side of the road.

Reply

20 PaulL February 2, 2018 at 6:03 pm

One of the big benefits comes from platooning (was that the term?) Basically one driver and 10 or 20 trucks following him, with no driver in them. They follow real close (good fuel economy) and don’t need a large amount of smarts because unexpected events are handled by the driver in front.

Either the other trucks have drivers, but they’re asleep waiting to get to the destination, or they pull off the freeway at the other end, and someone jumps in and takes over for local driving.

Reply

21 Mark Bahner February 2, 2018 at 6:21 pm

“They follow real close (good fuel economy) and don’t need a large amount of smarts because unexpected events are handled by the driver in front.”

“But Mom, Jimmy’s doing it!”

“If Jimmy jumped off a cliff, would you too?”

Reply

22 PaulL February 3, 2018 at 2:04 am

Amusing, but not quite the same as “my brakes and accelerator were hard-wired to Jimmy’s, so when he went off the cliff I did too.” Not many cliffs on the interstate.

Reply

23 Mark Bahner February 3, 2018 at 12:43 pm

“Amusing,…”

That was the principal intent.

“…but not quite the same as ‘my brakes and accelerator were hard-wired to Jimmy’s, so when he went off the cliff I did too.’ Not many cliffs on the interstate.”

I’ve seen plenty of these:

https://hips.hearstapps.com/amv-prod-cad-assets.s3.amazonaws.com/images/media/51/runaway-ramp-placement-inline-photo-365758-s-original.jpg?crop=1xw:1xh;center,center&resize=800:*

But seriously, my main goal was for amusement. Think of a Simpsons cartoon where the lead truck starts swerving all over the interstate, bumping cars off the road. So the trucks that follow the lead truck do the same thing. Funny!

P.S. Oops…trying to return to seriousness: What most people who don’t understand the likelihood of near-term fully autonomous control often don’t seem to realize is that *one* truck accident involving platooning trucks can result in the “lessons learned” being quickly transmitted to the entire fleet. Contrast this to humans: we know that drunk drivers kill people, but drunk drivers are still killing 15,000+ people year after year. That would never happen with an AI truck fleet, where a single accident will be analyzed thoroughly, and the resulting lessons learned transmitted to the whole fleet. It’s a much, much more rapid boost in safety.

Reply

24 SgtDad February 4, 2018 at 3:24 pm

The Armyis already doing this.

Reply

25 Axa February 2, 2018 at 1:11 pm

It happened with air travel.

Technological development in plane autopilot make made flying cheaper (fuel efficiency) which has lead to an increase in the number of travels meaning more pilots, not less.

Reply

26 msgkings February 2, 2018 at 1:20 pm

Excellent comment.

Reply

27 Todd K February 2, 2018 at 1:53 pm

No, many factors have led to cheaper air travel.

Reply

28 Slocum February 2, 2018 at 2:49 pm

With the Airline Deregulation Act being the main factor:

https://en.wikipedia.org/wiki/Airline_Deregulation_Act

Reply

29 Axa February 3, 2018 at 12:23 pm

Some deregulation. The US is still quite close to real competition by not allowing foreign carriers to serve domestic flights.

30 Mark Bahner February 2, 2018 at 6:29 pm

My prediction is that AI is going to lead to computer-driven vertical takeoff and landing (VTOL) passenger aircraft, which is going to result in significant competition with airports. This will greatly reduce present congestion at airports, and make traditional airports increasingly involved in long-haul flights.

It seems to me that pilots are even more vulnerable than truckers to AI…but I’m not a pilot or a trucker.

Reply

31 Mark Paskowitz February 3, 2018 at 5:05 am

What’s the connection between AI and VTOL? I don’t think I’ve ever heard this before?

Reply

32 Mark Bahner February 3, 2018 at 12:53 pm

“What’s the connection between AI and VTOL? I don’t think I’ve ever heard this before?”

My thought is that VTOL takes very good pilots…because the transition between the vertical takeoff and horizontal flight is difficult. AI is checking hundreds of different aspects of the flight every second, so it should be better at flying VTOL aircraft than humans.

Plus, VTOL aircraft make a lot of sense for relatively short-distance flights. For example, in a flight from NYC to LA, the takeoff and landing time is a trivial percentage of the time, and the time spent in the hassle of the airports is also a smaller percentage of the total travel time. But in a flight of say 200-400 miles, the time in the airport and taking off and landing is a significant part of the total flight time. It would be much, much better, for a 200-400 mile flight, to just go to a parking lot somewhere, get on board an airplane (no terminal even necessary), and fly to another parking lot somewhere.

Remember, you read it here first. 😉 Or else other places I’ve written this, since I’ve been making this same point for probably 3+ years. 🙂

Reply

33 Mark Paskowitz February 7, 2018 at 9:17 am

Interesting. My gut still says the engineering of the tilt-rotor (or whatever) would still be the limiting factor, but I’d never even thought about pilot skill being an issue.

34 derek February 2, 2018 at 1:26 pm

Did containers make dock worker jobs disappear? Don’t think so.

Reply

35 derek February 2, 2018 at 1:43 pm

An anecdote

The Port of Vancouver had a long labor dispute and strike I think in the late 80’s or early 90’s. The disagreement was on how to handle container traffic. The union agreement stipulated that members of the longshoremen union would tear down the shipping bundles and repackage them for the destination shipping. It was taken to mean that the containers would be unloaded and reloaded at the Port. Which was absurd, containers were designed to be a source to destination packaging. There was a long strike, months. The people who ran the ports were competing against Seattle and other ports who handled container traffic rationally. The union backed down, and the port has seen substantial growth in traffic since then as one of the western gateways into the continent.

Reply

36 Viking February 2, 2018 at 2:41 pm
37 mkt42 February 4, 2018 at 5:30 pm

Yup, Portland’s been a sh*tshow of bad labor relations for years, to the detriment of the port, the regional users of shipping, and not least the union itself.

I’ve always figured that what happened in places such as Vancouver is that the union finally recognized that the quantity of labor in the market simply had to fall — that’s what containerization and automation do, they replace labor with more efficient technology. But they were able to extract wage-protecting contracts, resulting in wages that are well above free market levels. The employers were able to go along with this because with such massive cost savings they could share some of the profit with the union.

So a higher wage now, for a smaller quantity of lucky union workers with seniority.

But Portland hasn’t been able to move into the 21st century. Some of this may be the fault of the longshoremen, some of it may be the fault of management, and IIRC in Portland there’s also a fight over union turf with some other union the electricians or somebody, claiming that certain work tasks belong to them.

Reply

38 anon February 2, 2018 at 3:34 pm

I thought automation at the ports has massively reduced jobs?

Reply

39 derek February 2, 2018 at 9:14 pm

But it has massively increased shipping. Employment? I’m not sure.

Reply

40 Matt2 February 2, 2018 at 9:16 pm

Total longshore jobs are way down in US.

Reply

41 SgtDad February 4, 2018 at 3:28 pm

Largely caused by the union. Much traffic is diverted to BC and Prince Rupert, which have better rail connections to the Midwest. Prince Rupert is also closer to the Asian ports. The upgraded Panama Canal allows shippers to skip the west coast & head straight to Houston.

Reply

42 Dick the Butcher February 2, 2018 at 1:40 pm

I’m wondering what effects self-driving trucks will have on the demand for self-driving ambulances.

Reply

43 PDV February 2, 2018 at 1:53 pm

This seems disingenuous. The people who drive long-haul trucks and the people who drive local trucks are not the same.

Reply

44 Joël February 2, 2018 at 2:19 pm
45 BDub February 2, 2018 at 2:37 pm

Long-haul truckers down, short-haul truckers up. I could see it.

Reply

46 derek February 2, 2018 at 2:54 pm

Some long haul truckers. There are defined routes that are very regular on consistent roadways. Once you get away from that, the scenario gets more complicated.

Reply

47 BikeRound February 2, 2018 at 3:21 pm

The answer to the question in the title of this posting is no. While self-driving trucks may increase the demand for labor needed to produce goods that are transported by self-driving trucks, the demand for truck drivers will be precisely zero in very short order, including local drivers. The verb “driving” will lose its current meaning as it relates to machinery and revert back to its original meaning. And all of this will happen within the next few years since the technology is ready.

Reply

48 msgkings February 2, 2018 at 3:23 pm

“And all of this will happen within the next few years” LOL no

Reply

49 BikeRound February 2, 2018 at 3:56 pm

You want to bet? The technology is there and ready to go. Especially when it comes to fully rational actors like trucking companies, the new technology will be adopted very quickly.

Reply

50 Taylor February 2, 2018 at 4:25 pm

The Uber folks don’t seem to think so. Their article said adoption by 2028 was very optimistic.

Reply

51 msgkings February 2, 2018 at 8:54 pm

Yes I want to bet.

Reply

52 Anonymous February 2, 2018 at 3:26 pm

Many truck drivers load and unload moving trucks full of furniture. It will be a while longer before we have robots to do that, I think.

Reply

53 Art Deco February 2, 2018 at 3:33 pm

CUCK!

Reply

54 BikeRound February 2, 2018 at 3:59 pm

But then the job function is being a mover, not being a truck driver. The truck will still drive itself.

Reply

55 Anonymous February 2, 2018 at 4:34 pm

Sure, but many jobs with the “truck driver” label involve doing a lot more than driving a truck.

Reply

56 derek February 2, 2018 at 9:16 pm

Nope. You should go around with these folks some time. Self driving vehicles are a long long ways away from being able to handle the routes that these trucks use.

Reply

57 Elite February 2, 2018 at 3:35 pm

The real story in Silicon Valley is that investors and entrepreneurs in the autonomous vehicle space are extremely concerned about loss of jobs and what federal, state, and local governments will do about it. So of course Uber published this.

Reply

58 Al February 2, 2018 at 11:11 pm

Ding ding ding!

Winner of thread.

Reply

59 rayward February 2, 2018 at 3:44 pm

I’m always impressed when economists provide predictions of the future even as the accuracy of their predictions in the past are, shall we say, a little less than reassuring. I suppose the benefit of providing so many predictions of the future is that it keeps one’s attention off the errors of the past.

Reply

60 Bumby Hemingway February 2, 2018 at 4:02 pm

I liked this argument better when it was “if we cut taxes a lot then business will become so good that we will actually GAIN tax revenue”

Reply

61 Anonymous February 2, 2018 at 4:35 pm

Obviously the devil is in the details. Without a doubt that argument is true given certain circumstances

Reply

62 Benny Lava February 2, 2018 at 4:10 pm

I’d be ashamed to post something like this. Is this what economics has become? How many industries in the past have introduced automation to a specific job function and not reduced employment? See:

http://www.themoneyillusion.com/?p=32209

Reply

63 PaulL February 2, 2018 at 6:05 pm

I always like the analogy that the invention of spreadsheets didn’t result in fewer accountants. Instead, it made it possible to do far more innovative accounting, and the accounting profession exploded.

Reply

64 Beefcake the Mighty February 2, 2018 at 11:28 pm

It’s Scott Sumner, what do you expect?

Reply

65 Taylor February 2, 2018 at 4:23 pm

There was a lot in this article that didn’t really ring true. Right off the bat, it started with “The outlook for trucking jobs has been grim as of late”, which is the opposite of what’s been going on in the trucking industry for the past 6 months at least. I know they were talking about long term projections, but that’s still a stretch. Trucking capacity is very tight, mostly due to the lack of eligible drivers. Active trucking utilization has been hovering at or above 100% for several months now, exasperated by the recent electronic logging device mandate. Most of the issues they claimed were facing truckers have mostly been solved by a tight labor market (for truckers and for the economy more broadly). That “workers are employed by a nasty labor model that holds down wages” is no longer really true. The Atlantic’s evidence for that claim is one of their own articles from May 2016, when the US was still recovering from the 2015 freight recession. Truck drivers are seeing much more money in their pockets now. The article very briefly mentioned the trucker shortage – it is really the largest story in trucking today.
The article also talks as if big companies are keeping truck drivers poor. Really, trucking is one of the most fragmented industries there is; there are 350,000 or so owner operators in the US, so most of the hauling is done by mom and pop truckers. Collective bargaining is therefore mostly a non-starter for truck drivers.The idea of Sherman anti-trust preventing truck drivers from collectively bargaining is laughable. Finally, it is not obvious that drayage is the worst job in the business. Obviously hauling short distances requires less time away from home, which is a major benefit. We can see evidence of this in the fact that long haul truckers tend to get paid better than short haul truckers.
Overall, I thought the article was less informative than the Uber publication on which it was based. I went to the FTR (Freight, Trucking, and Rail) conference in Indianapolis last September, and many attendees there thought that self-driving trucks would lead to more trucking employment, mostly by winning shippers over from rail and maritime. The Uber Freight rep there made many of the same points as the Medium post – it was cool to see them fleshed out in such detail.

Reply

66 JWatts February 2, 2018 at 5:18 pm

+1, informative

Reply

67 Taylor February 2, 2018 at 7:08 pm

Thanks. Wish my formatting had come through, as nobody likes to read a big block. I guess I need to hit enter twice to actually break it up.

Reply

68 TMC February 2, 2018 at 7:23 pm

Hear that Ray?

Reply

69 Al February 2, 2018 at 11:14 pm

Excellent.

Reply

70 daguix February 2, 2018 at 4:35 pm

Truckers will be paid less for the last mile logistics and engineers will be paid more for managing the autonomous fleet. Average will be over.

Reply

71 Dan Lavatan February 2, 2018 at 4:51 pm

There will be less than 70 self-driving trucks total rather than 70 percent on the road by 2028 so to won’t matter.

We already know there is no real market for self-driving passenger or cargo vehicles. If you look at aviation for instance, we know the technology exists to mutiplex labor, because the military does it in limited circumstances and the aircraft already have the necessary equipment. But there is no demand or you would see pilots flying four aircraft at once remotely. They can’t even get rid of attendants even though they are useless and cost lives by putting too many people on the plane. There is no effective constituency for reducing labor costs.

There are less than two instances in my life when I might have used a self-driving car, even though I make long road trips. And we know this is true in general, because there is almost no demand for chauffeured cars or coaches that people could sleep in, outside of rock bands that have a bunch of staff anyway. There is a lot of demand for cars that are fun. If there is an accident and a self-driving car had to go around something they would just get stuck for days, holding everything else up.

Heck, they can’t even make self-driving trains, inspite of the fact they are on rails, it would take zero effort and the software would know where the curves are. But passenger trains are useless anyway. Maybe 50 years after self-driving trains are universal someone should wake me up and think about trucks.

Reply

72 JWatts February 2, 2018 at 5:20 pm

“But there is no demand or you would see pilots flying four aircraft at once remotely. ”

Umm, the military flies a massive amount of drones at this point. Where a pilot (based on the ground) flies multiple drones at once.

Reply

73 BikeRound February 2, 2018 at 10:39 pm

The demand for cars is going to dramatically change in the next 24 months and demand for manual cars is going to collapse in the next five years. Just twelve months from now, think back to this conversation and it will be very clear as to who was right. Already in 2018–this year–when Waymo launches its commercial pod service in a Phoenix suburb, the perception of manual cars will be completely different from what it is now. The most high-end Mercedes will look obsolete and will no longer project an aura of sophistication upon its owner. While manual cars will still be salable in 2018, people are going to start switching to more mundane models because luxury cars will no longer be status symbols; you might as well get anything that gets the job done, and the lowliest Kia works just as well as any Mercedes. There are a lot of dimwits who are writing a lot of articles about how self-driving cars are still a long ways off, and they all tend to repeat the same brainless nonsense. I have a folder that serves as a library of these articles so that I can make fun of them in 2019. Self-driving technology is ready and it will transform transportation in the next few years.

Reply

74 BikeRound February 2, 2018 at 11:10 pm

Now, in 2019, of course Mercedes may come out with a model that is also fully self-driving. But then what is the difference between the Waymo pod and the luxury car? That the luxury pod has cushier seats? What is going to happen is that these different services will start to differentiate themselves in ways that are completely unrelated to transportation. For example, Mercedes Pod may strike an exclusive deal with HBO that the latest episode of Game of Thrones that will air on HBO on Thursday will be available in the MBP on Wednesday night if you book you commute trip home by Sunday evening. Are you starting to understand how this is going to be a completely different business from the current business of selling useless macho machines with ever more horsepower? Do you seriously think that anybody is going sit there in their cubicle thinking that they would rather be “driving” home, whatever that may mean?

Reply

75 chuck martel February 3, 2018 at 9:36 am

For the average American, who constantly grouses about his or her commute, their time behind the wheel at the controls of the second most expensive thing they’ll ever buy is the best portion of an otherwise mundane, boring and meaningless day. They’re in command of their own little world, they pick the music they’ll hear, the route they’ll take. There’s no boss looking over their shoulder, no kids or spouse whining, they’re free to chat on the phone. They’re engaged, on a personal level, with the most exciting thing that can occur that day, the inevitable conflict with their road psycho counterparts. There’s no other normal sensory stimulus as important and consequential as rush hour traffic. It can be literally life or death but practically never is. Who would want to give this up in exchange for a slow-speed crawl in a crappy autonomous vehicle littered with gum wrappers and discarded Starbucks cups? How much different would the autonomous vehicle journey be from a bus trip with a single passenger?

At present, an automobile trip, while strictly regulated in many ways, remains the almost very definition of freedom in the US.

Reply

76 Mark Bahner February 3, 2018 at 12:19 am

“We already know there is no real market for self-driving passenger or cargo vehicles.”

I should somehow put this comment on Long Bets. I predict that twenty years from now, this comment will be viewed like a comment in 1975 that “there is no real market for personal computers.”

Reply

77 Mark Bahner February 3, 2018 at 12:30 am

“There are less than two instances in my life when I might have used a self-driving car, even though I make long road trips.”

Every time I drive I would have used a self-driving car, *if* the self-driving car was some combination of: 1) less expensive per mile, 2) quicker, and 3) as safe or significantly safer.

In twenty years, when the majority of cars are self-driving cars providing transportation as a service, they will be: 1) half the cost or less per passenger-mile (for people like me driving less than 6000 miles per year), 2) at least 50 percent faster for most trips, and 3) result in 90%+ fewer accidents and deaths than at present.

Regarding long distance trips, I predict that in 20 years, it will be possible to drive the 2800 miles from LA to NY in less than 28 hours in a self-driving car.

Reply

78 msgkings February 3, 2018 at 2:34 am

That last bit, I wanna bet on that one too LOL. Man you and BikeRound need to get a room.

Reply

79 Komori February 3, 2018 at 12:38 pm

Just a couple of weeks ago I was driving along a highway frontage road, speed limit 55. Guy in a pickup exited the highway, dive-bombed the right-most lane, the one I was in, just in front of me, and proceeded to come to a complete stop. Then he shifted into park (I saw the reverse lights come on briefly). On a 55 mph road three lanes wide. When I honked at the moron, I saw him fumble his cell phone and drop it into the passenger seat, at which point he looked around and finally realized where he was parked. He proceeded to flip me off before moving, of course.

I think people panicking about the safety of autonomous cars really don’t pay attention to what the drivers around them are doing all day long. The AI doesn’t even have to be as good as the average human driver to vastly improve road safety.

Reply

80 msgkings February 3, 2018 at 1:27 pm

I agree completely re robot driving being safer than human. I was shooting down the idea that in less than 20 years you will be able to have a robot drive you from NY to LA at over 100 MPH avg speed the whole way.

81 Mark Bahner February 3, 2018 at 9:26 pm

“I was shooting down the idea that in less than 20 years you will be able to have a robot drive you from NY to LA at over 100 MPH avg speed the whole way.”

I’d say the odds are above 50/50. A human has already driven 2803 miles from NYC to LA in 28 hours, 50 minutes and 30 seconds:

http://www.cnn.com/2013/10/31/us/new-york-los-angeles-cannonball-speed-record/index.html

I would say the odds that the *technology* could do it are much higher than 90%. In 20 years, a thousand dollar computer will likely be able to perform more than 1000 times the calculations per second of a human brain. Plus, autonomous cars will have vision that will include infrared, LiDAR, radar, sonar, conventional cameras, and mapping systems. Plus, the guy who averaged close to 100 mph got only 40 minutes of sleep, which was a problem. It won’t be a problem for a computer.
.

82 msgkings February 4, 2018 at 6:03 pm

That’s what makes markets, differing valuations and opinions. I stick by my claim.

83 Mark Bahner February 4, 2018 at 10:45 pm

“That’s what makes markets, differing valuations and opinions. I stick by my claim.”

Is your claim based on the legality of it…that it won’t be legal? Or you think the technology won’t be capable, even 20 years from now?

84 Dan Hanson February 2, 2018 at 6:11 pm

I wonder how many of the people making predictions about the future of truck drivers have ever ridden with one to see what they do?

One of the big failings of high-level analyses of future trends is that in general they either ignore or seriously underestimate the complexity of the job at a detailed level. Lots of jobs look simple or rote from a think tank or government office, but turn out to be quite complex when you dive into the details.

For example, truck drivers don’t just drive trucks. They also secure loads, including determining what to load first and last and how to tie it all down securely. They act as agents for the trunking company. They verify that what they are picking up is what is on the manifest. They are the early warning system for vehicle maintenance. They deal with the government and others at weighing stations. When sleeping in the cab, they act as security for the load. If the vehicle breaks down, they set up road flares and contact authorities. If the vehicle doesn’t handle correctly, the driver has to stop and analyze what’s wrong – blown tire, shifting load, whatever.

In addition, many truckers are sole proprietors who own their own trucks. This means they also do all the bookwork, preventative maintenance, taxes, etc. These people have local knowledge that is not easily transferable. They know the quirks of the routes, they have relationships with customers, they learn how best to navigate through certain areas, they understand how to optimize by splitting loads or arranging for return loads at their destination, etc. They also learn which customers pay promptly, which ones provide their loads in a way that’s easy to get on the truck, which ones generally have their paperwork in order, etc. Loading docks are not all equal. Some are very ad-hoc and require serious judgement to be able to manoever large trucks around them. Never underestimate the importance of local knowledge.

I’ve been working in automation for 20 years. When you see how hard it is to simply digitize a paper process inside a single plant (often a multi-year project), you start to roll your eyes at ivory tower claims of entire industries being totally transformed by automation in a few years. One thing I’ve learned is a fundamentally Hayekian insight: When it comes to large scale activities, nothing about change is easy, and top-down change generally fails. Just figuring out the requirements for computerizing a job is a laborious process full of potential errors. Many automation projects fail because the people at the high levels who plan them simply do not understand the needs of the people who have to live with the results.

Take factory automation. This is the simplest environment to automate, because factories are local, closed environments that can be modified to make things simpler. A lot of the activities that go on in a factory are extremely well defined and repetitive. Factory robots are readily available that can be trained to do just about anything physically a person can do. And yet, many factories have not automated simply because there are little details about how they work that are hard to define and automate, or because they aren’t organized enough in terms of information flow, paperwork, processes, etc. It can take a team of engineers many man years to just figure out exactly what a factory needs to do to make itself ready to be automated. Often that requires changes to the physical plant, digitization of manual processes, Statistical analysis of variance in output to determine where the process is not being defined correctly, etc.

A lot of pundits have a sense that automation is accelerating in replacing jobs. In fact, I predict it will slow down, because we have been picking the low hanging fruit first. That has given us an unrealistic idea of how hard it is to fully automate a job.

Reply

85 clamence February 2, 2018 at 7:44 pm

Very good post. Self driving trucks aren’t as simple as Juicero or Facebook even if some in Silly-Con Valley think so.

Reply

86 Taylor February 2, 2018 at 7:47 pm

+1

Reply

87 PaulL February 2, 2018 at 8:02 pm

I think the reality is that most automation won’t replace people, it will be automation that enhances productivity of people. And this should be welcomed, as it’s what allows pay to increase. Truck drivers won’t be replaced, they’ll be augmented, perhaps allowing them to sleep whilst on the motorway, or perhaps to pilot multiple trucks in parallel.

Reply

88 Dan Hanson February 3, 2018 at 10:48 pm

I comletely agree. But of course, this has been going on for a long time. GPS took away the need to do extensive route planning on paper. CB Radios allow truckers to share road and traffic conditions in real time. Internet connections allow for automated dispatching and route updating. Modern trucks have all sorts of digital aids that make being a trucker more productive. This trend will continue, essentially forever.

But that’s not what people are worrying about. The modern Luddites are convinced that somehow the nature of automation has changed, and now it will start permanently destroying jobs overall – to the point where they believe we need to implement a guaranteed income to provide for the coming hordes of automation-displaced workers. There is absolutely no evidence for that. We have been automating labor for centuries, and unemployment is at a record low.

Reply

89 BikeRound February 2, 2018 at 10:28 pm

I think we are talking about very different things here. For sure, somebody will still manage the business of moving goods around. But the actual act of driving a truck on roads will be handled by an automated system.

Reply

90 Dan Hanson February 3, 2018 at 10:50 pm

Yeah, probably. And that may allow truckers to drive more hours, if the law catches up. That makes trucking more efficient. But lots of things have made trucking more efficient, and it didn’t result in trucker employment armageddon.

Reply

91 Jeremy Lott February 19, 2018 at 2:33 pm

Dan Hanson: Valuable information. Would you be willing to write me at Jeremy{foil robots}At[and spammers]MichaelHyatt[Because I] Dot [hate them] Com to discuss an idea?

Reply

92 Transnational Pants Machine February 2, 2018 at 10:41 pm

So this is what MR has come to.

OK sorry, I couldn’t resist. But this post is…. sad. Truly.

Not only will self-driving trucks NOT be a thing in the lifetime of anyone reading this dreck, it will also NOT cause more truck drivers to be hired. Much like robots will not create a need for more assembly line workers.

What the hell is wrong with Tyler specifically, and econ people in general? Is this entire field nothing more that Readers Digest clickbait? Christ on a unicycle. Sad.

Reply

93 Al February 3, 2018 at 12:02 am

Assuming that the US wins the race to autonomous trucks them, yes, there will be an increase in truck driver employment as the unions and, more importantly, the government steps in and places absurd regulation upon the technology.

However, even if the US wins, China will fast follow. China will not give one damn about the truck drivers and will deploy the technology efficiently. The gains in efficiency will be denied, of course, by the truck drivers and their friends on the left. However, within a couple of years the truth will become evident and the regulations will fall in the US.

Reply

94 Al February 3, 2018 at 12:25 am

As a meta point, I look forward to China stomping the west’s face in the coming decades. Somehow a tumor took hold in the West in the 60s and grew, parasitically, throughout our real and political economy. Strangling everything other than IT, which somehow befuddled the carcinoma, it was able to stop almost all progress.

Finally, a new contender to the title of world leader has appeared. And, having lived through a near death experience with the cancer, it will give no pause to any of cancer’s feelings/desires.

China is the shining star of humanity. It is a great time to be alive.

Reply

95 msgkings February 3, 2018 at 1:28 pm

“China is the shining star of humanity. It is a great time to be alive.”

This is why I know Thiago/Charbes is just a troll. If he were sincere he’d be all over this guy.

Reply

96 Ariesusduabelas February 8, 2018 at 6:55 am

corporate need to cut costs.

Reply

Leave a Comment

Previous post:

Next post: