Do Used Tesla Model S Buyers Believe Elon Musk About Teslas Appreciating? #CleanTechnica Analysis

April 28, 20195 years ago Michael Barnard 0 Comments

Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!

Recently, Elon Musk was interviewed by an MIT artificial intelligence researcher, Lex Fridman, and made a claim he’s made before.

“Essentially, buying a car today, is an investment in the future … the most profound thing is that, if you buy a Tesla today, I believe you are buying an appreciating asset, not a depreciating asset.”

At an investor event a couple of weeks later, Musk went further and indicated that Tesla owners could make a $30,000 a year gross profit from their cars. That was qualified with a million mile and 11 year vehicle lifetime, but it’s not like Musk showed his math on this one.

We decided to look at this in a few different ways to assess this claim. What’s the background for it? Which Teslas did it apply to? And has the used car market picked up on the signal?

Background

This is not news, Musk is just talking more about it at present. Projecting the creation of an Uber competitor with owned Teslas driving themselves has been on Musk’s radar for quite a while. Given that, last quarter, Tesla also paid off $920 million in debt, started shipping to other continents (leading to a delivery delay that impacted revenues), and was spending a lot of money building its next Gigafactory in Shanghai for the Chinese and Asian markets (leading to a Q1 loss and projected Q2 losses), it’s quite likely that a positive note was needed.

A relevant October 2016 announcement was that all Teslas were manufactured from that point forward with all of the hardware, including sensors, neural nets, and actuators to be fully self driving once the software component was finalized, and that upgrading to full autonomy was going to be an over-the-air update. They’ve tweaked their processor since, but Musk has assured owners who purchase full self-driving capabilities that the processors will be replaced for free. And new vehicles come with the new processors.

~6 months before it is in all new production cars. No change to sensors. This is simple replacement of the Autopilot computer. Will be done free of charge for those who ordered full self-driving.

— Elon Musk (@elonmusk) October 16, 2018

“Ordered” vs “order new” might mean some additional upgrade charge for those who didn’t pre-commit, but we’ll see how that plays out. Regardless, that’s not greatly pertinent to the further analysis.

Assessment of the claim

When full autonomy is available, customers will be able to opt in to having their cars be used for ride sharing when they are not driving them. As the average car is driven 5% of the year, that’s a lot of time that they could be making money instead.

Let’s assume 50% of Uber prices because they don’t have to pay the driver. That’s about $1 per mile. Let’s assume a limited 10 fares a day, an average of of 3 miles each, and operation five days a week. That turns into $30 per day, $150 per week, ~$600 per month, and $7,200 per year.

30 miles takes about 10 kWh of electricity at an average of $0.12 in the US, so that’s $3.60 per day in additional expenses at worst (most EVs will charge in low time-of-use billing periods in the middle of the night or on free chargers as available). That’s a cost of $18/week, ~$72/month, and ~$864 per year.

You are almost doubling average daily US driving with 30 miles extra a day, so there will be additional wear and tear, cleaning, and tire costs. Call it another $1,000 in expenses a year.

$7,200 − $1,100 − $864 = $5,336 gross profit annually for something you don’t have to think about. Double the rides to 60 miles a day, it’s $10,000 per year. Triple them to 90 miles a day, $15,000. Etc.

Average New York taxis drive 180 miles in each 12 hour shift, or 360 miles in a day. Much of the time they are traveling to fares, not with paying customers, so let’s assume 60% of the miles are under fare, leaving 216 miles per day of paid travel. Take away the 5% of personal usage and that’s close enough to 200 miles per day under fare. That could net over $30,000 per year, assuming Tesla wasn’t taking a significant cut off of the top. Even that would be manageable with a reasonable adjustment upward of rates.

Tesla Model 3s are coming with all the autonomous hardware in the $35,000 version. Let’s call it $45,000 with autonomy turned on. Used Tesla Model S 60 and 70 vehicles built after October 2016 are in the low $50,000 range.

Most cars depreciate perhaps 30% when you have them delivered. Say your Tesla has lost $15,000 in the first year, and probably another 10% per year after that. Let’s work up some numbers to see if you break even on depreciation.

Chip in a few dollars a month to help support independent cleantech coverage that helps to accelerate the cleantech revolution!

Yes, we could get ahead of depreciation by a long shot pretty quickly with moderate carsharing. And, of course, these are depreciation rates that seem reasonable, but those would be impacted by the greater utility.

If I could buy a second-hand, one-year-old Tesla Model 3 for $30,000 and have it make me $30,000 per year, that’s actually an amazing deal. It’s arguable that people would pay a premium for that. Even a used Model S 60 or 70 at that price would be very interesting.

Does the used car market show any signals on this?

One of the places we would expect to see some evidence that smart people are seeing this opportunity is in the used car markets.

In early January, CleanTechnica published two pieces on the used car market and value for Tesla Model 3s. The first found that Tesla Model 3s were barely present in the used car market — via an analysis of Autotrader, we compared the Model 3 to its Audi and BMW competitors, and also compared it to equivalent-volume, cheaper cars.

We’ve just updated that assessment, three months later, to look at only 2019 model year vehicles from other manufacturers and to see if people weren’t selling their Model 3s in part due to potential loss of their tax rebate in a pre-2019 sale. With a more apples-to-apples comparison, Tesla Model 3s are still vastly underrepresented in the used car market. That data point could suggest that owners are thinking of the utility value of the car, especially as many buyers of the Model 3 are likely stretching to acquire the more desirable car. The several hundred thousand pre-orders for the Model 3 attested to its desirability at the price points that Tesla was projecting it at, as does its status as the 6th best selling car in the USA over the past 3 quarters combined.

Then CleanTechnica pointed out the head-scratching Kelley Blue Book take on Model 3 resale value. Kelley Blue Book titled it the best of any electric car for expected resale value, but their own published numbers made it clear the Model 3 would hold its value better than any other car, period, yet wasn’t titled that way. The second-best electric car was the Model X in that assessment. Per the thesis, this could indicate that the folks at KBB saw the value of autonomy, but when you read KBB’s actual comments, there’s no indication of it. It doesn’t seem as if the resale analysts are valuing this in the same way Musk is asserting it should be or the earlier numerical work suggests it should be, at least not yet.

Another data point that’s worth assessing is a recently release iSeeCars.com analysis of resales of the Tesla Model S versus other cars. They found that the Model S was the fastest selling used car in its class by a substantial margin, almost 10% quicker than the much cheaper Audi A7 and A8 and with close to half the time-on-market of the Maserati Quattroporte. BMW’s similarly valued 7 Series takes almost two weeks longer to move, and the cheaper 6 Series is in the same boat. It’s clear from this analysis that Tesla’s used cars are strongly valued in the market as well, but does this send a signal yet about how buyers are valuing them based on autonomous taxi service?

We reached out to iSeeCars for clarification on several points to see if further poking at the data would uncover anything interesting.

The iSeeCars team made a point in their report that the Model S was the most heavily discounted of the cars, but it was unclear how they had arrived at that assertion from the report text itself. What they’d done was to look at the average across all used cars of that model to look for outliers. The Tesla Model S had the greatest variance in used price per that.

But this seemed potentially aligned, as the Tesla Model S has come in an enormous range of new price points for different configurations, from the sub-$60,000 original Model S 60 to the up-to-$135,000 P100D. That’s a much greater MSRP range than its competitors, which have much more narrowly segmented product lines and hence multiple offerings covering the same range. Could that have been the cause of the variance that iSeeCars was seeing? They looked more deeply and provided this interesting insight:

“P100d has the lowest percentage of all trims to be 10% or more below market value. The 60D had the highest percentage of listings that are 10% or more below.”

This seems to imply that the 60D had the largest number of cars that were less desirable. This might have been an age thing for the prospective owners, but the desirability of the P100D seems to bely this. On the used market, buyers want as much Tesla as they can get.

Finally, there was the question of whether the underlying data showed anything about vehicle value with full self-driving features. If it did, there would be a visible change in statistics for cars between 2016 and 2017. Once again, we asked that specific question of iSeeCars.

“Our analysis shows a breakdown of the model year, or manufactured year in Tesla’s case. 2016 and 2017 are nearly identical for days on market: 2016 is 35.8 days, while 2017 is 35.9 days. 2018 is 29.0 days.”

Once again, the used car market isn’t showing that purchasers are clearly valuing the potential for full self driving, but simply that they want the newest, best Teslas.

Where does this leave us?

Musk’s assertion seems right. It would be possible to have a Tesla with full self driving and a Tesla robotaxi autonomous ridesharing service make a good deal of money annually for Tesla owners. However, there’s no clear signal from the underlying used car market that anyone cares about that yet.

Will everyone opt in for rideshare? Of course not. But will everyone get a family vehicle that could deliver every family member everywhere they need to go without Mom or Dad driving all the time? Yes, and that’s worth a lot too. In fact, it’s quite possible that many two-car families with kids will be able to have a single autonomous car and get the same utility. As US governmental estimates peg the annual cost of a car at $9,576 per year, maybe Musk should be touting one-car families instead of Tesla Taxi!

Teslas are highly desirable used vehicles, selling faster than competitors and non-competitors, and holding resale value longer in many cases, but that doesn’t seem to be aligned to making money off of the future Tesla Taxi capability. Of course, this is like Henry Ford in the early 20th century saying, “If I had asked people what they wanted, they would have said faster horses.”

Most people haven’t internalized the thought of their Teslas turning into money spinners instead of costs yet. When that starts sinking in, the data may change. But by that point, other cars probably will have full self-driving capabilities as well, and companies such as Uber will be leveraging them at competitive price points, leading to the same valuation on the used car market. Maybe.