Just Say 'No' To Tesla's Misleading Margin Metric

Summary

• There's no Tesla metric more misleading than its gross margin percentage.
• How is it misleading? Let us count the ways.
• What happens when we calculate Tesla’s gross margin in the same manner used, for instance, by Ford? Watch and learn.
• Yes, the Model 3 may eventually achieve a 15% gross margin. But even if it does, Tesla will still be losing $10,000 per delivery.
• Before we get to margin in all its grossness, let’s discuss the latest rumblings about how Q1 is shaping up.

I. The Q1 Rumors And The Plight Of The Parts Suppliers

Since CoverDrive and I published last week's article forecasting a Q1 GAAP loss of $855 million, the Bloomberg Model 3 tracker has again dropped its estimate of the weekly production rate, this time to 599.

Naturally, Tesla Charts is on the case:

(Chart courtesy of the magnificent Tesla Charts, whom you can and should find on Twitter @TeslaCharts. Christmas colors, he explains, because this tracker has become something of a gift to Tesla skeptics.)

During the time I’ve written about Tesla (NASDAQ:NASDAQ:TSLA) here, I don’t recall ever hearing more rumors of serious manufacturing problems, more auguries about a sizable shortfall in deliveries, or more speculation about when the next capital raise is coming or whether there might be impediments to it.

In recent articles, both Anton Wahlman and Bill Maurer questioned whether Tesla might actually deliver fewer cars this quarter than it did last quarter despite the ramp-up of the Model 3.

Separately, jaberwock, in a blog post, took a stab at estimating the Model 3 reservation cancellation rate. He employed a methodology similar to one CoverDrive and I used in determining (accurately, I believe) that the yield rate on Model X reservations was approximately 30%. With careful and appropriate caveats about the limitations of his data set, jaberwock estimates almost half the Model 3 reservations have been cancelled.

Meanwhile, over at Twitter, Minor Threat already can taste the next capital raise, while the formidable luis carruthers (@orthereaboot), who assiduously tracks the stunning number of Tesla lemon lawsuits, believes there are securities law impediments that are stopping it from happening.

As for me, I’m just taking it all in, and not ready to draw any conclusions (beyond those I drew in my crossing-the-Rubicon article several weeks back). However, I can’t help wondering what life is like right now for Tesla’s parts suppliers.

Presumably, Tesla last year gave those suppliers the same over-optimistic production estimates that it gave its investors and customers. And, presumably, the suppliers relied on those estimates in ramping up for volume production.

What production volumes did Tesla tell those suppliers to expect? What volumes have they shipped? What happens to their pricing if anticipated volumes shrink? How much financial duress are Tesla’s production problems causing them?

And what is happening as all those Model 3 parts arrive, and stack up, in Fremont? Tesla recently completed an Automated Storage & Retrieval systems building. I think we can safely assume that building is getting an accelerated shakedown.

II. Tesla’s Mesmerizing Gross Margin Magic

The greatest trick Tesla management ever performed was to enchant the firm’s analysts into focusing on two meaningless metrics: Deliveries and gross margin.

The deliveries metric is useless because the more cars Tesla delivers, the more money it loses. In fact, lately, the more cars it delivers, the more money it loses per car.

Vincent Wolters recently wrote a fine analysis showing that the greatest driver of Tesla’s losses are its costs under the heading of Sales, General & Administrative (commonly, SG&A). For years, Tesla investors have expected the firm’s losses to shrink as its deliveries increase. In actual fact, the opposite has happened and, as the Wolters details, the opposite is likely to continue to happen.

As for gross margin, let us review the ways in which Tesla’s calculations are hugely misleading, and then add another to the list.

A. Tesla’s Calculation Ignores The Cost Of Its Service Center Network

Because it owns its distribution network, Tesla incurs costs other automakers shift to their third-party dealerships.

The cost shifting comes at a price. The other automakers discount their cars by about 10% in selling them to dealerships. As a result, their sales revenues are 10% lower than they would be if they adopted Tesla’s model, which means their gross margins are almost 10% lower.

Let's do the math. Imagine Ford (NYSE:F) has automotive revenue (let’s call it AutoRev) of $3 billion and costs of goods sold (COGS) of $2.5 billion. Our formula is:

Gross Margin = (AutoRev – COGS) / AutoRev

Thus, Ford would have a gross margin of 16.67% ($3.0 billion minus $2.5 billion is $0.5 billion, which divided by $3.0 billion and expressed as a percentage is 16.67%). (In actual fact, Ford’s gross margin last year – 15.34% – was pretty close to that number.)

Now, imagine Ford owned its dealerships. In that case, of course, Ford would not discount its cars to the dealership network, and its automotive revenue would be $3.3 billion. Its gross margin would rise to 24.24%.

Well, you ask, does it not all work out the same in the end? Yes, it does, but not for gross margin purposes. Tesla pays the extra 10% in operating its own distributor and service network, but it allocates that cost to SG&A rather than COGS. So, the cost implicit in Ford’s gross margin calculation is excluded from Tesla’s gross margin calculation

B. Tesla Excludes Its R&D From Its Calculation

While most automakers include Research & Development (R&D) in their COGS, Tesla does not. Is there any justification for this exclusion?

No, I don’t think so, because R&D is in the nature of an operating cost. The automotive market is hugely competitive, with ever-changing technologies, regulations, and mandates, and consequently R&D spending is something every automaker must ceaselessly undertake. An auto company not spending R&D on the next new thing, which will roll out three or four years down the road, is an auto company that will die. R&D is, quite inescapably, a cost of doing business.

Think of the refresh of the interiors of the Model S and X, which likely has been under development for the past several years, and will finally see the light of day in a few months. This is simply to sustain demand, to keep the company alive. One must adapt or die, and the adaptation is an unending imperative.

Obviously, the bet Tesla investors are making is not on what the company has done or is doing, because what it's doing is losing lots of money on its existing cars. Rather, investors are betting on what they believe Tesla will do: The Model Y, the semi, the Roadster 2, the pickup truck, the Solar Roof Tiles.

Developing every one of those future products, though, will require the expenditure of R&D dollars, as will refreshing the existing products.

C. Tesla Allocates Some Supercharger Costs To SG&A

Tesla allocates about half its Supercharger costs to SG&A on the theory that those costs are in the nature of a marketing expense. This practice is, at best, questionable. Many Tesla cars were sold with free supercharging. Imagine if GM were to promise free gasoline for the life of each new car it sold, but then charge half the cost of the gasoline to marketing instead of COGS.

Moreover, if Tesla were to shut down its sales operation tomorrow, it still would need to maintain the Supercharger network for so long as there are Tesla cars on the road.

(hmmm. Sent my way by Matthew L. Alden. Thanks, Matt!)

III. Tesla’s Lease Accounting Inflates Its Gross Margin

The distortions in the gross margin calculation detailed above have been discussed before. The focus today is how Tesla’s lease accounting has grossly inflated its gross margin. And not by just a little.

This lease accounting inflation has happened in two different ways: The expensing of warranty costs and the front loading of gross margin. Let’s explore.

A. Expensing Of Warranty Costs

For each car it sells, Tesla creates a warranty reserve. Because they are Tesla’s estimate of future costs for that car, the amounts so reserved are not counted in “Automotive revenues.” Thus, the creation of the reserve reduces the gross margin on the sales transaction.

If, down the road, Tesla actual warranty costs are lower than the amounts it reserved, then Tesla can reduce the reserve and pick up the difference in “Automotive revenues.” Tesla on at least one occasion did reduce its warranty reserve. That reduction, though, was short lived. The overall trend has been to add to the reserve as warranty costs stack up.

So, the warranty reserve exists for cars Tesla sells. For cars that it leases, however, Tesla creates no such reserve. Instead, it expenses the warranty costs as they are incurred.

In 2015, as bonvivanttraveler recently reminded me, warranty expenses on leased cars totaled $9.5 million. In 2016, $19 million. Last year, $35.5 million.

That $35.5 million amounts to 1.6% of the automotive gross margin Tesla reported last year. Not earthshaking, but enough to account for 0.3% of Tesla’s 22.9% gross margin.

B. Front-Loading Of Gross Margin

Tesla historically has front loaded the entire amount of gross margin into the first year of lease payments. But the lease residual value is usually unrealistically high (caused in large part by Tesla adding the federal income tax credit amount to the residual value). As a consequence, the lessee’s incentive is to return the car rather than purchase it at the inflated residual value.

After the lessee returns the car, Tesla writes down its value or sells it at a loss. But at that point, it is “Services & other” that takes the hit, not “Automotive revenues.”

Obviously, it costs the same to build the cars that are leased as it does to build the cars that are sold. The higher gross margin on the leases is simply a function of Tesla pulling forward the gross margin. And, as noted earlier, at the end of the lease, when the piper must be paid, it’s not the automotive gross margin that takes the hit. Rather, the damage is buried in “Services & other.”

How much difference has the lease accounting made to Tesla’s gross margins? A fair amount, as Seeking Alpha commenter xonkd recently detailed in an illuminating interchange with doggydogworld and Bill Cunningham that starts here.

Xonkd detailed how much higher gross margin is on the leased car transactions than the car sales, and how the higher leasing percentages pull up the overall percentages. With thanks (again) to Tesla Charts, here’s what it looks like graphically:

The red line data inflates the blue line data to give us the only percentage Tesla reports, which is the grey line data. (Auto sales revenues are much larger than auto leasing revenues, that's why the inflation effect is relatively small.)

OK, the inflation is significant, but not exactly dramatic.

Now, let’s adjust for some of the other factors I discussed earlier to show what Tesla’s gross margins would look like if calculated in the manner employed by other automakers.

At the outset, let’s use for our AutoRev figure only the revenues and COGS of sales (rather than leases). Then, let’s adjust for the dealership factor by decreasing the AutoRev number by 10%.

Finally, we’ll include 90% of Tesla’s R&D in the cost portion of the gross margin calculation. (Why only 90%? Because some of Tesla’s R&D is for Tesla Energy products rather than its cars and trucks.)

With these adjustments, made to mirror the method used by Ford, General Motors (NYSE:GM), Volkswagen (OTCPK:OTCPK:VLKAY), Toyota (NYSE:TM), and others in calculating gross margin, let’s take a look at Tesla’s 2017 gross margins (and, once again, thank you, Tesla Charts, whose Twitter posts are the first ones I look at each day):The grey line is the gross margin percentage Tesla reports. The blue line is the gross margin percentage adjusted to remove the distortions.

Now, that’s quite a difference.

The conclusion is clear: Any Tesla analyst or business journalist who places any importance on Tesla’s calculation of gross margin is being willfully misled.

C. What About Regulatory Credits?

All the revenue numbers in my Tesla Charts graphics include Tesla’s revenues from ZEV, GHG, and CAFE credits and payments. Those regulatory credits totaled $360 million in 2017.

My long-time friend, xonkd, hesitated at including the regulatory credits in AutoRev. I, however, do not. However distasteful may be the policies that led to the creation of those credits, they are for Tesla reliable sources of revenue.

That said, there is a strong case to be made for smoothing out the regulatory credits over the course of the year so that the gross margin numbers are not distorted by Tesla’s practice of hoarding the credits for several quarters and then selling them all at once. However, while such smoothing would change the quarterly calculations, it would leave the year-end total calculations untouched.

IV. Can We Please Stop Talking About The Gross Margin Percentage?

Watch what happens as the Model 3 ramps. Tesla’s revenues will rise to new heights. Its gross margin dollars will achieve new records.

And its cash burn and losses will surpass prior marks as well.

Why? Because, as mentioned earlier, and as Vincent Wolters so brilliantly illustrated, SG&A is what is driving Tesla’s losses.

So, what’s the lesson here?

The lesson, gentle readers, is that it is completely foolish to focus on what Tesla reports as its gross margin percentage. That percentage, be it up or be it down, tells us nothing of any value whatsoever.

Rather, the focus should be, as Nathan Weiss of Unit Economics LLC preaches to his analysts, on gross profit dollars per delivery.

Weiss believes Tesla may be able to achieve gross profit dollars per delivery of close to $20,000 once the Model 3 is fully ramped. That will amount to a lofty-looking 15% gross margin.

Alas, at that level, the combined SG&A and R&D will be closer to $30,000 per delivery. In other words, a net loss of $10,000 per car.

That’s what happens even if and when Tesla achieves Model 3 gross margins of 15%. And that’s why reading any importance into Tesla’s reported gross margin percentages is a fool’s game.

So I send out this message to all those writers at CNBC, and Forbes, and The Motley Fool, and BusinessInsider, and Bloomberg, and The Wall Street Journal, and etc.:

Please, stop being taken in by Tesla’s reporting of its gross margin percentage.

You’re smarter than that, aren’t you?

Aren’t you?

Analyst’s Disclosure: I am/we are short TSLA. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

I am short TSLA via long-dated options