It would be hard for most to understand how many links and stories I receive from the public these days. Too often they’re accompanied by the terse suggestion, which sometimes feels more like a threat, that we’re holding back this massively important story from the public. The pedigrees of these earth-shattering so-called stories never seem to tie them to any legitimate news source. A great example would be two years ago, when we were warned that a military exercise called Jade Helm was really a secret plan for the president to take over the Great State of Texas. And yet when you remind some sender that in the past they also believed that particular story, they fire back that the only reason it didn’t happen is because it became public knowledge. Really?
So when I received an email and link to a story with the headline, “Peak Car Ownership is Near – Beginning of the Fall of Car Ownership,” I first thought it could simply be another anti-automotive screed by a modern-day Luddite. After all, it had been forwarded as a story apparently posted on Facebook, under nobody’s byline. It started off by relating that Columbia University had published a study showing that Uber could replace every cab in New York with just 9,000 self-driving cars, none more than 36 seconds away, that would take you to your destination for a mere 50 cents per mile. In fact, for just $4.3 billion, we could replace all taxis and, apparently, mass transit in America, with these same self-driving vehicles.
From there the story went to a Price Waterhouse Coopers study, which claims our self-driving future will be so efficient that the total number of vehicles in America will drop from 245 million to just 2.4 million, an industry collapse of 99 percent; and with that, the auto insurance, car finance, parking and automotive aftermarket industries will also collapse. They left out one critical part, though: If there’s only 2.4 million cars left, there won’t be any highway funds to speak of from gasoline and diesel taxes; so how does America pay for fixing and constructing roads and freeways? We have right at 4.1 million miles of roads now, which means each self-driving car would have to raise enough revenue each year to take care of 1.7 miles of road.
As it is, we’re having an extremely hard time fixing our highways with 245 million vehicles’ fuel taxes contributing to the cause. No wonder Price Waterhouse really wanted La La Land to win the Oscar for Best Picture.
Of course, both claims are so outrageous that who could ever believe this tripe? Only it turns out, it’s not false news at all. After some digging and research on this story, I found the documents being written about. The first part is from The Earth Institute at Columbia University, a report titled, “Transforming Personal Mobility,” and it credits Dr. Lawrence D. Burns as the report’s lead director. Burns’ credentials are unquestionably among the finest in the industry; for nine years he was head of R&D and Advanced Planning at General Motors; he’s on the National Academies of Engineering and sits on numerous advisory boards for corporations, think tanks and universities. In fact, the only slightly off thing in his official resumé is that his PhD is from Berkeley. But that merely makes you think he was a lot of fun in college.
But, even though it’s not fake, this report is not valid as its origins might lead one to believe. It was revised and republished over four years ago, and it makes some suppositions that are so simplistic that one questions how anyone of Dr. Burns’ background and knowledge could ever make them. Case in point: how much a self-driving vehicle costs to make, own and operate.
To calculate their costs the study uses a car costing $25,000, but assumes that the technology to make that vehicle self-driving costs just $2,500. Nice dream: The lidar unit alone that numerous companies working on this technology today use costs around $85,000. Not to mention the auxiliary lidar units, cameras, computers and so on.
Recently Oliver Cameron, who is in charge of autonomous vehicles at Udacity, a break-off team from Google’s program, put the cost of the equipment for a self-driving car today at a quarter of a million, sans car. We’re talking decades and decades before that per-vehicle cost drops to $10,000, much less to $2,500.
What do I mean? In the past, auto manufacturers seem to have doubled the cost of new technology to come up with the mark-up for those options on their cars. As I recall, as late as 1986 when Delco offered an aftermarket CD player for GM cars, the cost was over $1,000. So, to be a $2,500 option, self-driving equipment would have to cost the manufacturer around $1,250 net; and by the time that happens, the car will cost way more than $25,000.
But that’s not the only corner cut in this imaginative thesis. The author assumes the car will last for 250,000 miles, that insurance will cost $3,000 a year — assuming you can get it because, remember, Price Waterhouse claims self-driving cars will collapse the auto insurance market — registration and taxes will cost $600 per year, and gasoline will cost 15 cents a mile. Further, they are basing some of those assumptions on a 2012 report by AAA titled, Your Driving Costs.
So, by page 11 of the report, either I’m completely lost or I’m questioning how Columbia University and someone of Burns’ legendary stature put out this report. Really? You’re basing the average cost of ownership of a future self-driving car on a AAA report of automotive averages from 2012?
Maybe you hadn’t thought about this, but if there is a future world of self-driving cars then it would bear little if any resemblance to our automotive world of today. After all, when you don’t own a car, just rent one when you need it, do you care if it has 50 or 500 horsepower? Because it’s never going to speed, even if you’re really late to the most important meeting of your life. It’s not going to go one mile over the speed limit, won’t run stop lights and won’t brake constantly from aggressive driving. That means it would be incredibly fuel efficient under any circumstances. And, knowing all of that, who will care about the engine size or horsepower? So why, if the entire exercise of creating a world of self-driving cars is for the efficiency, would you use a AAA report from 2012 on fuel efficiency and gasoline costs as a baseline?
Of course, then there’s the part about the small electric city car, designed just to haul two people around urban areas; and here they tag a cost of just $6,500 plus the $2,500 for technology. And at this point I’m done. This cost estimate seems to me nothing short of ridiculous. OK, that’s an overstatement, but that incredibly low cost is something unlikely to happen for 40 or 50 years. Much in the same way that a microwave oven can be bought for under $100 today — but, when Amana introduced the first Radarange for home use in 1967, it cost $495. That’s around $3,610 in today’s money.
And that spins us back around to the idea put forth in the report that just 9,000 Uber cabs could handle the job of every cab in New York, pick you up in just 36 seconds and cost just 50 cents per mile. That should be fun to watch. Because around 1.5 million people live in Manhattan, but the average workday population swells to around 4 million as people stream into the city to work. Well, maybe under the most ideal conditions those 9,000 self-driving cars can replace the 13,000 New York City taxis, but that number is never going to handle the millions rushing into the city in the morning, going out for lunch (and back) at noon and returning home at 5. (For that price per mile, in a world of self driving cars loads of mass transit users will flip to a personal ride.) It seems, when dealing in unperfected future technologies, one truism that holds up over time is that you can’t use the numbers from the past to forecast the future. I can assure you that the costs and efficiency of the Wright Brothers’ biplane bore no relation to those of the first Douglas DC-3, which in turn gave no usable information to those who designed the Boeing 707 jetliner.
As for the Price Waterhouse study covered in its Autofact newsletter, the total overall number of vehicles falling from 245 million to just 2.4 million suspiciously resembles the figure that most cars today are parked 90 percent of the time. Self-driving cars will be constantly in motion, so you could get by with 90 percent fewer vehicles. Then again, some people drive for a living, from truckers to pharmaceutical reps and so on. Farmers use their trucks 14 hours a day during planting and harvest — and the list goes on and on. Again it ignores that 124.7 million people go to work at 8 in the morning, lunch at noon and home at 5. 2.4 million self-driving cars, even if stunning in their efficiency, are going to have a major problem with that daily workload.
Additionally, that study points out that the auto industry employs almost 10 million people, from manufacturing to those who drive professionally, parts, service and dealerships. The “good” news is that almost all of those ten million jobs will disappear within 10 – 15 years. Yet the report says zip about the economic impact of 10 million reasonably well-paid Americans’ losing their jobs and no longer being consumers. Then again, remember all of the manufacturing jobs we sent to Mexico? Forget Trump, apparently self driving cars will wipe them out, too.
Overall it first looked like fake news. Then I found the reports and the article was factual. But then I found the reports the article was based on, and those reports seemed overly simplistic, no matter how much advanced algebra they used in their calculations.
Oh, and without owning the equipment, Uber today in some markets, certainly during surge pricing, can earn more than Columbia’s study suggests New York City rides will cost in the future, yet Uber loses money. Their losses would go higher once their corporation has to pay all of the expenses for fuel, cars, maintenance, depreciation, clean-up and so on. That’s a key number to overlook.
It also leaves out the human element: Americans like owning cars and trucks and driving ourselves around, even if our vehicles do sit parked 90 percent of the time.
© 2017 Ed Wallace
Ed Wallace is a recipient of the Gerald R. Loeb Award for business journalism, conferred by the Anderson School of Business at UCLA. He reviews new cars every Friday morning at 7:20 on Fox Four’s Good Day and hosts the top-rated talk show, Wheels, 8:00 to 1:00 Saturdays on 570 KLIF AM. E-mail: firstname.lastname@example.org