Predicting the future this far out, I’m bound to get some things wrong. You may disagree right off the bat. But I guarantee you that if you’re not an expert in auto industry economics, you will learn something new and shocking.
So what’s my prediction?
In the next decade, about 5 million people will be laid off due to self-driving cars. In return, we’ll get an extra 5 work-weeks of productivity, lower overall transportation costs, and save tens of thousands of lives.
Self-driving cars are going to change the world, faster than we think. It will likely have as much impact as cellphones.
Imagine personal transportation as a utility. Cars do a nice job of showcasing status, but in the end they move you (and your stuff) from one place to another. That’s all most people care about. I love cars, but if you told me I could have a robot chauffeur for the rest of my life, for the same price as a car, I’d kiss you.
And unless I'm heading to a red carpet, I don’t really care what the car looks like. If it’s clean and comfortable, we’re good. That gives personal transportation companies (PTCs for short) wide latitude, which can enable great economies of scale. They can build their own cars (and some auto manufacturers will become PTCs), and make existing cars autonomous.
We know how cheap it is to lease a car, and autonomous car service will be even cheaper because of utilization: the percent of time an asset is in use. Most of the time, your car does nothing but take up space. It’s on the road maybe 15% of the time (4 hours/day). If it were autonomous, it could be driving other people around, earning its keep.
That’s what enables current PTCs like Uber and Lyft. Your car has a lot of down time. So do you, if you’re underemployed. Why not capitalize on that?
In essence, PTCs allow us to share cars, which enables utility pricing. I expect they’ll be priced like cell phone service. Until they work the kinks out, there will be various upcharges and pricing bands. Eventually we’ll have flat rates and pre-paid.
Most of us haven’t yet traded our car for a driving service because labor costs make it too expensive. Drivers aren’t paid a lot, but they’re paid a heckuva lot more than AIs. When we eliminate paid drivers, the price will hit a tipping point.
What happens next?
First Million Layoffs
Obviously, this is bad news for cab drivers. On 1/15, Uber employed 160,000 drivers; on 10/15 it was 327,000 drivers. As of 1/16, Uber had over 400,000 drivers and Lyft 315,000, and there are about 200,000 cab drivers. That’s almost one million drivers, and it’s rising (granted, some are part time or hold other jobs). That may not seem like much compared to the 159M person US labor force, but it’s not chump change. You must also keep in mind that our labor participation rate - the percentage of adults who haven’t taken themselves out of the workforce due to retirement, disability, or just plain giving up - is the lowest it’s been in years.
It’s been pointed out that those jobs didn’t exist until recently, and many cab drivers were already laid off because of PTCs. So maybe losing a million jobs isn’t a big deal.
Car Sales Plummet
Car sales will drop significantly, again due to increased utilization (sharing). Determining the total drop is tricky, but you can probably look at the maximum number of cars on the road at one time. That’s peak traffic, plus some reserve. Peak traffic will rise when the cost of rides drops, but I’m sure it will be far less than the 253 million cars we have in US. There are only 159M workers, and not all of them are on the road at the same time. That’s 94M extra cars, almost a 40% drop in need.
But I also predict a big spike in ride sharing. Especially when we redesign cars for privacy; it’d only take basic partitions, like privacy screens in a limo. That will bring down the cost of PTC plans, and further lower the demand for cars.
Then there’s the used market. Any cars that can economically retrofitted for autonomy will be a candidate for purchase. Again, economies of scale will change what is economical.
Cars that can’t be retrofitted (cheaply) will drop in value significantly. First, this may slow the adoption of PTCs for anyone happy to drive themselves. However, I expect mass exports to our neighbors to the south: Mexico, Puerto Rico, and Cuba. I was in PR not long ago and cars are expensive there. You’ve heard the stories about classic cars in Cuba, so they must be dying for a ‘02 Camry. This will likely help their economies a great deal (except manufacturing in Mexico will drop).
There are one million people manufacturing cars and trucks and three million in sales and maintenance. That won’t go to zero, but it will get slashed considerably. Perhaps another million layoffs.
Finance Industry Slammed
Did you know the outstanding debt for car loans and leases is over $1 trillion dollars? I didn’t! I knew it was a big part of the industry, sure. GM Financial, the lending arm of GM, had $6.5 billion in revenue last year. That’s on roughly $65B in loans and leases, and it’s for just one car maker. Think about every bank and credit union out there.
Also in the news is how car loans are being compared to subprime mortgages, saddling borrowers with negative equity, and the word “predatory” is being cast about again.
The silver lining (I hope) is that instead of a low income family having to finance the purchase of a $7K car that ends up costing them $20K, just so they can get to work, it’s more affordable to have a contract with a PTC (plus public transportation).
Auto Insurance Industry Evaporates
KPMG reported that the personal auto insurance market will shrink by 60% by 2040. Sounds optimistic to me, I predict more and sooner. Insurance losses (payouts) are currently $125 billion per year, and premiums total $200 billion per year. KPMG predicts that accident payouts will drop 80% with both safer and autonomous cars, and that will translate to lower premiums.
But wait! Most major companies self-insure their vehicles. We can expect PTCs will do the same, and unlike today, they’ll collectively own most of the cars on the road. That business is gone.
I don’t have employment numbers for the auto insurance industry, in part because so many insurance giants cover all sorts of policies. But we know it’s going to take a big hit.
Trucking Industry Transformed
I wasn’t even halfway done with this article when I had to edit this section, because autonomous trucks arrived while I was writing it. You might assume truck drivers are safe from layoffs because you need them for loading/unloading the cargo and other safety reasons. Not to mention how much harder it is to drive the darn things. Heck, I don’t even want to deal with a UHaul!
As Uber and Otto have shown, the highway is where it’s at. Our biggest fear with big rigs is a tired truck driver. But AIs don’t get tired. Trucks won’t replace rail, which is four times more efficient, but I’m sure the industry will grow.
We’ll only need local drivers for final delivery. In reality, the biggest stumbling block to autonomous trucks will be the truckers’ unions. When waterless urinals came out, new buildings didn’t need water lines for them. Unions negotiated to install useless lines to keep the lucrative work. Maybe something similar will happen here. (For the record, I’m assuming the AIs won’t unionize.)
I don’t know how many people are employed at truck stops, which exist more for the truckers than the trucks. Those businesses will probably turn into automated fuel stations. But there are about 3.5 million truck drivers, of which 1.8 million drive tractor-trailers.
Delivery Service Primed
Remember all that buzz from Amazon about drone delivery? A squadron of drones going back and forth between warehouses and residences and offices? Some found the idea amusing. Most figured it would be unsightly, annoying, and potentially hazardous to have the sky blackened with package-laden drones.
I predict there will be drone delivery, but they won’t be flying from warehouses. They’ll be flying in and out of autonomous delivery vans. If your package isn’t too heavy or has to be signed for, a drone will carry it to your door. Why wouldn’t it? It should be noted that:
- UPS employs 240,000 drivers
- The Postal Service employs 315,000 mail carriers
The bigger question is what happens to physical retail when delivery is even faster and cheaper? Yes, overall it shrinks. But what is left is a lot more interesting and pleasant (“experiential”) than what we have today.
Rental Cars
Q: What’s the difference between getting a ride from an autonomous taxi, or renting a self-driving car from Hertz by the hour or mile?
A: I don’t know either.
What I do know is that the current car rental fleets combined are about double the number of taxis (again, including Uber and Lyft).
To cut to the chase, I am very bearish on rental car companies. They have so much work to do to catch up, and it’s like nothing they’ve ever done. I see them being liquidated rather than transformed.
This will further damage auto makers, beyond the drop in fleet sales. I learned that rental companies buy a good percentage of their inventory under a repurchase program (“program cars” in the industry). It’s an insurance policy that protects them from losing money when they sell (e.g., the model becomes undesirable).
When the entire used car market plummets, they’ll invoke this clause to sell at a decent price and buy autonomous cars. They may also start requesting this clause more often. Given this symbiotic mess, perhaps automakers will create an “autonomous retrofit” clause that replaces the repurchase program. Maybe that will save them. Maybe.
Personal Net Worth Drops
Over 90% of Americans own a car, and for many that is their most valuable asset. What happens when that drops in value because nobody wants it? Our collective net worth goes down.
Unless you have a loan or lease - then bank owns it. This doubles back to that $1T in automotive debt. I think we’re used to being underwater on car loans, but to this extent? Is the finance industry prepared for this? Not with the damage I described above.
An interesting twist: car companies loan you money for cars, even though they are not worth the loan amount, because they enable your employment. And that empowers you to pay the bank a lot more than the car is worth.
What happens when you’re unemployed, your car isn’t worth close to what you’re paying for it, and PTCs enable you to get around without owning a car? We have another loan crisis.
Government
A lot of government infrastructure exists to support drivers. Obviously, the DMV is a big one. I imagine PTCs will be able to integrate their systems with the DMV’s to expedite car registration. But what about licenses? We’ll need government IDs, but just as fewer people are learning to drive a stick-shift, over time fewer people will learn to drive at all. So no tests.
Then there’s enforcement. There are 41 million speeding tickets per year, with $6.4 billion in fines. And that’s just speeding; we’ve left out all the other violations you can get like running a red light, tailgating, failing to yield, etc. Not to mention the most expensive, damaging ticket of all: DUI/DWI.
A reviewer pointed out that such fines are a significant source of revenue to local governments. We are talking billions. Without it, taxes will go up. But if the cause is greater public safety, then the effect is fewer police and parking enforcement personnel on the payroll. It’s anyone’s guess how that will balance out. The government is insanely bad at shrinking itself, but this is a case where policing pays for police.
Saving Lives (and Money)
At last, we see the light at the end of the tunnel! Autonomous vehicles will save lives. How many? From the CDC:
“...more than 32,000 people are killed and 2 million are injured each year from motor vehicle crashes. In 2013, the US crash death rate was more than twice the average of other high-income countries. In the US, front seat belt use was lower than in most other comparison countries. One in 3 crash deaths in the US involved drunk driving, and almost 1 in 3 involved speeding.”
That’s 90 deaths a day. The medical and work loss costs related to deaths was $44 billion. That’s just from the deaths! On a personal note, I have two friends who lost their lives on motorcycles, and neither was at fault; it was the drivers who killed them.
In addition:
“Americans spend more than 1 million days in the hospital each year from crash injuries. Crash injuries in 2012 totaled $18 billion in lifetime medical costs. ...
Lifetime work lost because of 2012 crash injuries cost an estimated $33 billion.”
In another part of the site, they claimed the total was $80 billion. So the total medical and work-loss costs from deaths and injuries is $95B to $124B per year.
Then there’s property damage from roughly 5 million car crashes per year. The average insurance claim for property damage is little over $3K. That’s $15 billion per year in damages. Well, reported damages.
All told, that is truly wonderful news: we will save tens of thousands of lives and over $100 billion in related costs per year. I don’t know how much that will offset the hit to the economy from everything above.
Productivity
The average American spends ~50 minutes a day commuting, or 200 hours/year. That gives you 5 work weeks of extra productivity. What could you do with 5 extra weeks per year?
What if your work day started as soon as you got in the car, because you didn't have to drive it? (It can't be less productive than your open office environment.)
Where are the new jobs?
I posited that we may reclaim many manufacturing jobs, and people told me that’s impossible. I know we’ve forgotten how to manufacture many things, supply chains are optimized for Asia, etc. But if not manufacturing, what are millions of laid off blue collar workers going to do? Nobody seems to have an answer for that. If you do, I’d love to hear it.