In the wake of Tesla’s recent announcement of 2015 financial results and a Bloomberg report that the upcoming Model 3 could cost as little as $25,000, it’s worth taking a looking at the prospects for the more affordable electric car with a somewhat more jaundiced eye.
As an auto engineer turned journalist and now technology analyst, I’ve been following Tesla for nearly a decade. I love the fact that Tesla Motors exists and has gotten as far as they have, but I must throw some cold water on the Model 3 party. While I strongly support the efforts and ideals of Tesla, I’ve always been skeptical of the company’s ability to execute. As far back as the spring of 2007 I told my editor at AutoblogGreen, Sebastian Blanco, that I was skeptical that Tesla could deliver the Roadster that summer as originally promised. Sadly, they didn’t surprise me and the company has consistently failed to meet their self-imposed deadlines ever since.
In February 2007, then-CEO Martin Eberhard told me the second Tesla car, codenamed White Star, should be in production by mid-to-late 2009. The first Model S didn’t get to a customer’s garage until June 2012 and the follow-on Model X was similarly late. A modern automobile is an incredibly complex piece of hardware and is subject to vastly more regulation than any other product that comes out of Silicon Valley. Most in the tech industry have hugely underestimated the difficulty of building cars from scratch.
The poet Robert Browning said, “Ah, but a man’s reach should exceed his grasp, Or what’s a heaven for?” Tesla Motors has consistently reached beyond its grasp and that’s a good thing. On occasion it has overachieved such as with the acceleration performance of the Model S, which was better than what Musk and the team originally targeted.
But more often it was fallen short and while there is nothing inherently wrong with that, when you make very vocal public pronouncements about what you will do, it damages your credibility. Not only was the Model S late, the driving range, at least according to the official EPA estimate procedure, was short of the 300-mile promise. At the time the Model S concept was first shown in 2009, the opening line of the press release said “With a 300-mile range and 45-minute QuickCharge, the $49,900 Model S can carry five adults and two children in quiet comfort – and you can charge it from any outlet, without ever stopping for gas.”
While further digging into the details would reveal that a $49,900 Model S and one with a 300-mile (ultimately closer to 250) range were two completely different cars, Musk frequently touted those two numbers as though we would eventually have a car that met both targets. Musk and Tesla never lied about the fact that there would be multiple battery sizes at different price points for the Model S but the phrasing of the information certainly implied something different from reality unless you dug in very closely.
Much the same has been happening in the discussion of the Model 3.
To date Musk has only discussed a $35,000 base price and a 200-plus-mile range. Fortunately at least Musk is now saying that base price is before tax incentives, unlike the $50,000 price which depended on a $7,500 federal tax credit, so that is progress.
During the recent Q4 2015 earnings call, Musk emphasized that the engineers working on Model 3 are taking the lessons learned from the Model S and X and designing the car to be much less expensive and complex to manufacture. However, Tesla has built up a certain premium mystique around its brand and cannot afford to make the Model 3 feel cheap.
There is also the problem of sustainable profitability, something Tesla has not been able to achieve even with cars that have an average selling price that is probably in the neighborhood of $100,000 or more. If Tesla is to survive as a stand-alone business, it must be able to generate more cash than it spends, and sooner rather than later. If it ramps up production volumes with a car that has a starting price half of what the cheapest Model S costs, that becomes an exponentially more difficult problem, especially if the falcon-wing doors on the Model X prove to be as troublesome as most people expect.
I believe that Tesla is unlikely to hit to its late-2017 launch timing for the Model 3 and we likely won’t see deliveries until mid-to-late 2018. While designing a car for easier manufacture is a good thing, increasing volumes by an order of magnitude or more will prove to be far more difficult, with new construction techniques, equipment and employees.
Tesla will also have to significantly up its game on quality with the Model 3. It may sound counter-intuitive but customers buying very expensive cars are actually less sensitive to problems in their cars because they usually have more to choose from. People buying more affordable mainstream products often don’t have another option and rely on a daily driver that has to work right every time.
When it does arrive, I also think the Model 3 will be offered with at least two battery sizes. The larger pack, probably with a capacity of around 60-kWh like the Chevrolet Bolt will offer a range of at least 200 miles. But I remain very skeptical about that variant costing $35,000. Instead, I think that there will be a smaller pack option, probably about 40-kWh with a range of around 140-miles for that low-ball price.
The pressure on Tesla to start making money may lead to another very important change on the Model 3 relative to current products. From the time that Elon Musk announced the introduction of Auto Pilot semi-autonomous capability in fall-2014, the factory began installing all of the necessary sensor hardware on every single car that rolled off the line. What Musk didn’t mention at the time of the announcement was that customers that wanted that new capability would have to pay for a software upgrade, $2,500 at the time of purchasing the car, $3,000 if you decide later that you want to turn it on.
While there is nothing wrong or unusual about charging more for extra functionality, the Tesla approach was peculiar. With other automakers, if you don’t order something when the car is built, they don’t include the relevant pieces. Tesla was building in hundreds of dollars worth of sensors that they might never realize any revenue on. The gross margins on a car costing as much as $140,000 could probably support that and most people spending that much will probably pay for the upgrade.
The math is quite different for a $35,000 car with inherently thinner margins. The take rate on the Auto Pilot option on Model 3 would likely be 10 percent or less based on other similarly priced cars available today with driver assist systems. As a result the base Model 3 will probably have a limited set of sensors only for parking assist and a camera or low-range radar sensor to enable collision warning and automatic emergency braking. The latter is something that Tesla will want to include in order to retain 5-star safety ratings.
Again, nothing out of the ordinary here for most companies. However, Musk has made a big deal about the ability to upgrade customer’s cars with new functions through over-the-air software updates long after they left the factory. While OTA updates will almost certainly be standard on Model 3 as they will be on most new cars in the future, the scope of upgradability will be much more limited and may lead to disappointment for customers.
Finally there is the issue of tax incentives which were at the heart of a recent article claiming the Model 3 could cost as little as $25,000. While that is certainly true based on the $35,000 base price, a federal $7,500 tax credit and a California tax break of up to $2,500, it won’t always be the case. The $7,500 federal incentive is phased out for any manufacturer once they sell 200,000 EVs.
If the Model 3 is late and Tesla manages to sell significantly more Model S and X in the next year as Musk projects, Tesla could hit that 200,000 threshold by mid-2018, eliminating the possibility of that break for customers. Even if the Model 3 is on time, early production will likely be limited to higher-end variants that cost more than $35,000, leaving a net price well above $25,000. Musk understands these numbers which is likely why he has been emphasizing the $35,000 price before incentives.
When it does finally arrive, the Tesla Model 3 and any additional variants Tesla produces will no doubt be a very good car in many ways. But unlike the Roadster and Model S, it will be arriving into a much more crowded field that includes the Bolt, the next-generation Nissan Leaf and Volkswagen e-Golf and many more plug-in vehicles than we have today. Tesla will also have to resolve its retail issues if it’s ever going to sell 500,000 cars a year, but that’s another story.
