In a previous article I disclosed my short position in Tesla Motors (TSLA) and outlined the basic reasons for which I deemed the stock over-priced. So far the short position has moved against me, but I still hold it in my short portfolio.
Today I'd like to look at another aspect of the company which I hadn't included in my original discussion. In particular, I'd like to examine the relationship between the company's vehicle gross margins and overall costs - and its potential impact on the company's longer term financial performance.
Remember that - when making their valuation cases - many bulls rely heavily on the idea that TSLA will be able to maintain 25% gross margins on car sales (even after the EV subsidies are phased out). Though I personally doubt that the company will be able to maintain those margins on the Model S; let alone on the future, lower priced, "gen 3" vehicle, I'm willing to grant the premise for argument's sake. Instead, what I'd like to suggest is that Tesla's gross margins should be evaluated differently than those of other auto manufacturers. Why? Because Elon Musk has famously derided the idea of having service constitute a profit center. Here's a short quote from a CNN article emphasizing the point:
"I've told the Tesla service division that their job is never to make a profit," Musk said. Most auto dealerships make a large portion of their profits from the service department which, Musk pointed out, creates a conflict of interest when it comes to product quality.
"I hate the idea of making money because our product broke," said Musk. "That's just wrong."
Moreover, repair and maintenance services aren't the only liabilities which the company must support on the strength of its new car sales; there's also the build out, maintenance and supply of free electricity to the company's supercharger (SC) network. (For those unaware of it, use of the supercharger network is free to all 85 kWh Model S owners and as $2K option for those buying the 60 kWh model. In either case, the 25% gross margin which buyers pay at the outset, also goes to support the at-cost service and the "free" super-charging network.) As a result, comparing TSLA's gross margins with other auto manufacturers is misleading. Or to put it another way, since Tesla front loads the costs of service and of the SC network in the initial price of the car, the margins have to be enormously high in order to carry the future liabilities.
To get a sense of how impactful this might be, I've estimated new Model S sales as a percentage of the installed base, as well projecting the average fleet age over time. I detail these projections below.
Estimated Model S Deliveries
I start by estimating monthly Model S deliveries. The table below shows my monthly delivery projections along with the resulting quarterly delivery figures for reference. (For past quarters, the monthly estimates are based on known quarterly delivery numbers).
In my projections I have Tesla exiting 2014 at a run rate of over 40,000 Model S deliveries per year, with a total of 37,650 vehicles being delivered during the year.
Deliveries as Percentage of Fleet
Next I look at the monthly deliveries as a function of the total installed base. So, for example, the number in Oct 2012 is 400 / (400 + 263) = 60%. In other words new deliveries that month account for 60% of all the Model S's ever delivered. Given that all of Tesla's profits come from the new deliveries, and that the company incurs substantial profitless costs from servicing the installed base of delivered vehicles, this number is important. The graph below shows the deliveries as a percentage of the installed base through 2014 (based on the projections I presented above).
Through the first half of 2013, the company was delivering >12% of new vehicles for every Model S it had already sold. But by 2014, the number of Model S's delivered monthly will drop below 10% of the installed base. Thus profits available due to the (assumed) high margins become a smaller and smaller percentage of the overall costs associated with warranty repairs, servicing, and the SC network.
Average Fleet Age
We can also get a sense of the importance of service costs by estimating the average fleet age. I do this by assuming that cars delivered within a given month are new (i.e. have an age of zero), and that previous months' deliveries age monthly. So, for example, in Nov 2012 I calculate the average fleet age in months as follows: (263*2 + 400*1) / (263 + 400 + 800) = 0.63 months. The graph below depicts the average Model S fleet age through 2014 (again based on my projected deliveries tabulated above).
Because production is ramping up over the whole period considered, the fleet ages rather slowly, about 0.4 months / month. As a result, even by the end of Dec 2013, the average Model S will be less than 6 months old. Nonetheless, as there are more Model S's on the road and their average age increases, there are more demands on the warranty, service and SC networks. (Indeed as we'll see below, even now in October with the average fleet age < 5 months old, there are signs that the company is already incurring high "legacy" costs.)
Impacts on Service and Owner's Perceptions
An aging fleet means that the honeymoon phase for Tesla aficionados is slowly elapsing - and as a result we're getting more and more realistic reports of not only of the car's benefits, but also its flaws. Moreover, since any flaws are fixed either under warranty (which is a direct cost to Tesla), or as profitless service (which negatively impacts net profit margins), it's important to try to get one's arms around them in order to project the company's future net margins.
Now please understand that, by all accounts, the cars are still seen as performing well, but the veneer of "perfection" is slowly fading as the experience base grows. More importantly, with respect to service and warranty costs, the idea that EVs - by virtue of not having a combustion engine and associated drive train - are virtually "maintenance free" is slowly evaporating.
Instead there are reasons to think that service and warranty costs could substantially eat into the high gross margins of new vehicle sales. To help readers get a sense of this, I've rounded up a few of the issues currently being prominently mentioned on the forums and in the media. Of course the recent TSLA battery fire is getting the most attention, but currently I don't include it on the list, as it's still a one-off event with no statistical meaning. (That's not to say that it hasn't been of terrible PR value - indeed my sense is that the fire has already impacted new car demand as well as prompted a small but noticeable increase in the number of Tesla Model S's listed for sale on eBay.) Setting that issue aside, here are some of the more prevalent worries with the car's reliability and service/warranty costs:
Loud humming noise when car travels above 60 or 70 mph
Many owners are reporting a loud humming noise coming from the motors when the car is traveling at freeway speeds. In some cases the noise is present very early in the car's life, in others it develops over 4 to 6 months. The solution so far has been a warranty repair by Tesla involving a complete replacement of the drive units. See for example this forum thread for details. Note too that the drive noise has impacted the dynamic of Tesla owners acting as "unpaid salesmen" in some cases:
In my case, they are refusing to do anything even though it is so loud that I no longer give test rides out of embarrassment (having told everyone how quiet the car is).
Tire Wear and Sunroof (Pano roof) Leaks
In performing their long term testing, Edmunds experienced the type of accelerated tire wear that many owners had previously complained about. Similarly, Edmunds encountered problems with the panoramic sunroof that has plagued many owners. In commenting on their experience with the car, Edmunds remarked:
Including the tire wear issue, it seems like a lot of stuff for a six-figure car with less than 10,000 miles on the odo.
There have also been a series of problems with the car's 12V batteries. Here's a typical recent report from an owner who had to have his car towed as a result of the 12V battery dying. One can search the forums to find many similar stories.
Long Service Times
In what I think might be a consequence of the service needs becoming a much greater factor vis-à-vis new monthly deliveries, many owners are commenting on lengthy service times. Sometimes it's a function of problems that are new and not completely diagnosed (such as the hum mentioned above), sometimes it seems to be due to lack of service capacity.
There are a number of reports detailing inordinately long service times (for relatively new cars). For example this owner has had two months of total service on his new car, while this owner notes his car has been in for service for 1.5 months of its 6 month life. This owner is even considering invoking the California Lemon law.
(Note too that while the cars are being repaired, Tesla loans the owner a model S, which is great for the customer, but adds substantial costs to the service portion of the company.)
Finally there's a looming realization that the car's energy consumption isn't limited to the power needed to propel the car, there are significant "vampire" losses which come from the car being in a ready state (and software bugs), as well as from the battery's thermal management. As winter approaches, and with more cars having been sold in cold northern climes, expect to hear more and more about this. (As an example, consider this Norwegian owner who loses 20-30 km/day due the car having to keep its batteries warm, and it's only October!)
On top of warranty costs and the break-even service department, Tesla also has the costs of building out, maintaining and upgrading its nascent Supercharger network. To get a sense of these costs, it's worth reading a recent trip report from a typical user of the existing network, and to then dwell on the idea that after buying a $100K car - travel times for longer trips double. Assuming Tesla wants to improve that outcome, there will have to be substantial money put into the network in even the most built out areas (California from the Bay area south) and then much, much more for the rest of the country and Europe. Factor in the additional costs of maintenance in areas with more severe weather (Northern US, Norway, etc.) and you're talking about real money. All of which has to be supported by the gross margins on new car sales.
In summary, the high gross margins that many bulls expect the company to achieve on an on-going basis, should not be compared with the gross margins of other automakers. For other automakers, servicing their cars is a profit center, and there are no "legacy" costs to support over a car's lifetime. Tesla, on the other hand, is likely to see reduced net margins as it intends to service its cars at cost, it has substantial warranty issues to deal with, and it has a Supercharger network to expand, maintain and supply with free electricity. As a result, as the company's installed base of cars expands and ages, we can expect net margins to decrease relative to gross margins. Given that the risk so far doesn't appear to be reflected in any analyst or financial commentator's valuations, the decrease may weigh heavily on a stock that's currently priced for perfection.