May 7 2021 - Tesla Operating Leverage: Far From Proven

 

  • A key aspect of the Tesla bull case is that the company benefits from strong operating leverage that will make it highly profitable with growth in sales and production.
  • Yet, although sales/production has been growing rapidly, the company is still only minimally profitable and dependent on ephemeral factors like EV credits and Bitcoin sales.
  • Tesla bulls ignore a key factor: with higher production/sales, average production costs fall, but so do average sales prices. Tesla's profitability is a race between these two factors.
  • An analysis of Tesla's recent financial results, cleaned up for some of the more obvious distortions but still a bit "noisy," shows that the case for strong operating leverage is far from proven.
  • The trade-off between falling production costs and falling sales prices will get worse going forward since competition will make the demand curve for Tesla's cars more downward sloping.
Tesla Service Center. Tesla designs and manufactures the Model S electric sedan IV Photo by jetcityimage/iStock Editorial via Getty Images

A key part of the Tesla bull thesis is that, although Tesla is only minimally profitable now, it benefits from "operating leverage" that will make the company enormously profitable as production and sales continue their rapid growth.

However, Tesla has been growing rapidly for a while now and high profits have still not materialized. Tesla bulls say "just wait" - although, strangely enough, they have never shown the same patience with Tesla bear arguments about the arrival of competition or an eventual collapse of the Tesla share price. The bulls also seemingly underestimate the length of the wait and how wrong the predications of imminent profitability have been. Elon Musk's famous "The Secret Tesla Motors Master Plan (just between you and me)" contained the following short description:

23322283-1620321025757594.png

The Master Plan was written in August 2006. It pretty clearly anticipated profitability ("that money") just on the horizon. Tesla went on to lose about .5bn between its IPO in 2010 and its first full year of profitability (but only with the benefit of almost .6bn in EV credit sales) in 2020.

Clearly, like Godot, huge profitability seems to be taking its time showing up. The purpose of this article is to look at why the bulls have been so wrong on this issue.

The Conceptual Error

The Tesla bulls are right to focus on the impact of operating leverage on the average cost of production ("AC"). There are many fixed and "semi-fixed" costs in producing, selling and servicing a car. This is true of the automobile industry in general and Tesla in particular: Tesla's high level of vertical integration has the effect, among other things, of increasing fixed costs. (Incidentally, this is why other manufacturers have generally moved in the opposite direction. They have learned that, in one of the most cyclical industries, having a high degree of vertical integration and fixed costs is not necessarily the disruptively brilliant and innovative idea that many Tesla bulls seem to believe.)

There is only one problem with this line of reasoning. With higher production, two things happen. The first is Tesla works its way down the AC curve by spreading its fixed and semi-fixed costs over a larger production base. But the second is that Tesla also works its way along the demand curve for its cars, which is also downward sloping. To sell more cars, Tesla must be willing to cut prices, either through direct price cuts or through selling cheaper (and likely lower margin) versions of its cars. So, the question of Tesla's profitability with scale is a race between the falling AC versus the falling average sales price ("ASP") needed to sell the additional units.

My contention is that, 17 years after its founding and almost 14 years after Musk's Master Plan, Tesla is barely profitable because it is just holding its own in this race.

The Analysis

We want to focus on the fundamental profitability of Tesla's business and how this profitability changes with scale. Unfortunately, we only have Tesla's published financial results to guide us in this analysis and these results contain a lot of "noise." I have tried to minimize this noise by making some adjustments to (1) remove extraordinary events (such as Tesla's recent gains from the sale of Bitcoin), (2) adjust for timing differences in revenue or costs (such as with Tesla's stock-based compensation ("SBC") or deferred revenue), and (3) avoid issues with Tesla's ability to shift costs between different line items in its income statement (such as with the consistently unprofitable "Services and other" line).

(For the avoidance of any doubt, I am not suggesting that these changes are an appropriate way to look at Tesla's profitability; for example, I would never agree that one should disregard SBC when evaluating a company's profitability. My objective here is to try to understand how Tesla's profitability varies with unit sales. I am focusing on relative movements in profitability (which is what is relevant for an analysis of operating leverage) and not the correct absolute level (which is what is relevant for valuation).)

What I Have Adjusted

Operating Profit: We want to focus on the operating profit line (a line that Tesla calls "Income from Operations") in order to exclude the impact of taxes and financing decisions, such as the choice of financing the business with debt or equity. Importantly, operating profit picks up both the revenue (ASP) and the cost (AC) side of expanding production and sales. It also avoids problems with Tesla's accounting choices which can move costs between different line items.

Exclude Regulatory Credit Sales: These sales should be excluded for multiple reasons. First, their timing and magnitude have nothing to do with the concurrent production and sale of cars, the former because Tesla recognizes these sales when they sell the credits and not when they earn the credits by selling cars, and the latter because the price of the credits depends on things that have nothing to do with Tesla's performance. They should also be excluded because they are a vanishing source of income for Tesla, something that the company acknowledges ("…in the long-term, regulatory credit sales will not be a material part of the business….")

Exclude Stock-Based Compensation ("SBC"): Many argue that SBC should be excluded because it is not a "cash" cost. This is total nonsense. The dilution associated with SBC is unquestionably a real cost, as can be seen by considering a hypothetical: What if Tesla paid all of its suppliers and employees only with shares. Would anyone claim that the company has no costs of sales? Of course not.

The reason to exclude SBC isn't because it's unreal. The reason is that the timing of this expense has little to do with current period production or sales and therefore cannot help us understand the operating leverage of the company.

Automotive Leasing: The issue here is, once again, timing. A lease produces income and cost over its entire life, which extends well beyond the production of the car. Therefore, I will eliminate both leasing income and costs, as well as leased cars from units delivered. (Logically, I should do the same with Tesla's solar leasing business. Unfortunately, although Tesla breaks out solar lease income in its financials, it does not do the same for costs, making this adjustment impossible.)

Deferred Revenue: This is also a timing issue. For example, when Tesla sells a car with the so-called "Full Self Driving" software, it does not immediately recognize all of the software sales price as revenue. Instead, the cash is offset with "Deferred Revenue" which is subsequently recognized as "features" of this software are delivered. Since the timing of these features has nothing to do with current production and sales, I have backed this out of the numbers. Although FSD is the largest component of deferred revenue, there are also other elements, such as sales with an obligation to provide free Supercharging. I will make this adjustment by adding back the quarterly number from the cash flow statement, which is an agglomeration of all of these deferred revenue effects.

Adjustments for Residual Value Guarantees ("RVG"): Tesla has sold cars, predominately to leasing companies, with a repurchase guarantee at a fixed price. Tesla has to periodically estimate the potential liability for this guarantee. If the estimated liability has increased, then Tesla treats this as a reduction in its automotive sales, partially offset with a reduction in the cost of these sales. Tesla discloses the quarterly effect of these changes on its earnings and I have added this back since this has nothing to do with current production or sales.

Restructuring and Other: These items are excluded because they are totally unrelated to current production and sales. The recent 1mn profit from the sale of Bitcoins, for example, says nothing about Tesla's operating leverage.

What I Haven't Tried to Adjust

Exclusion of Solar and Storage Business: In an ideal world, this analysis would focus exclusively on the automotive business. Unfortunately, Tesla does not provide nearly enough information to make this possible; for example, there is no allocation of SGA or R&D expense to the different segments of the business. In addition, the Tesla bulls insist that Tesla is more than just a car company, so it is appropriate to lump everything together.

Operating Decisions: I strongly believe that Tesla is deliberately underinvesting in R&D and customer service in order to eke out its minimal profits. For example, here are Tesla's R&D expenditures over the last few years:

2017

2018

2019

2020

1Q 2021

,378mn

,460mn

,343mn

,491mn

6mn

This is remarkably flat for a company that is allegedly developing a revolutionary battery technology, autonomous driving, a semi and a pickup truck, a solar roof, improved and industrial-scale battery storage, a major refresh of the Models S and X, a rumored compact Model 2, and the world's most advanced sports car, among other things. This thin R&D probably accounts for the fact that Tesla only has an extremely thin product offering in an increasingly crowded EV market.

Another way to see this is to compare Tesla's R&D expenditures/sales to the ratio of its "Big-Six" competitors (Toyota, VW, Daimler, BMW, Honda and GM). Here are the results:

R&D Expenditures/Sales

 

2017

2018

2019

2020

Big-6

4.6%

4.9%

4.8%

5.0%

Tesla

11.7%

6.8%

5.5%

4.7%

Source: Motörhead (@BradMunchen)

It's striking that supposedly the most technologically sophisticated manufacturer - and not just of automobiles - spends a relatively lower amount on R&D than its major competitors, despite being a much smaller company.

With respect to the underinvestment in customer service, judging from the pleading "I love my Tesla but…" tweets constantly sent to Elon Musk, the following table must be a pretty accurate summary of the deterioration. An earlier article from me also supports this conclusion.

23322283-1620321025748942.png

Finally, Tesla itself has acknowledged the inadequacies of its customer service multiple times. Here is a quote from a recent shareholder Q&A:

23322283-16203210263471088.png

Being able to get an appointment in less than 10 days in only 70% of the service centers for a premium car in that car's most developed market doesn't exactly qualify for bragging rights.

Although I believe that Tesla's current levels of expenditures on R&D and customer service are unsustainably low, trying to adjust for this would be enormously contentious and invariably somewhat arbitrary. Therefore, I will leave it out. However, this means that my results will definitely overstate Tesla's operating leverage.

Warranty Reserves: Like with deferred revenue, this is another area subject to timing differences. It is also an area where many Tesla bears feel that Tesla is using "aggressive" accounting to flatter current results. This is hugely complicated and contentious area that I will not dive into.

Costs for Model Refreshes: In Q1 2021, Tesla claimed a 0mn direct P&L impact from the changeover of Models S and X. Some would claim that this should be excluded. However, these types of costs are routine for automobile manufacturers, which must continuously refresh their cars to maintain sales. No adjustment was made.

Other Time Periods: The analysis runs from the first quarter of 2019 until the first quarter of 2021. This is somewhat arbitrary but the period has been chosen because 1Q 2019 is a period of (1) significant Model 3 production, (2) the commencement of the "phase out" of the federal tax credit for buyers, and (3) after the "cherry picking" of high margin orders out of the Model 3 backlog that occurred in Q3 and Q4 of 2018.

Foreign Exchange Effects: I have made no attempts to adjust for FX effects on Tesla's sales, costs, assets and liabilities. This would be hugely difficult and prone to substantial error.

Accounting Anomalies: Tesla bears point to a number of accounting anomalies (eg., accounts receivable, mysterious sudden drops in production costs, weird pricing of used car sales possibly affecting RVG calculations, and others). Although I believe that there are a large number of valid questions, it is way beyond the scope of this article to attempt to adjust for any of these things. I will take Tesla's financial statements at face value.

Results

The information in the following table comes from Tesla's shareholder decks, Forms 10-Q and Forms 10-K, which can be found here.

Tesla's Adjusted Operating Profit

(All figures in $mn, except car sales and per-share amounts)

 

Q1 '19

Q2 '19

Q3 '19

Q4 '19

Q1 '20

Q2 '20

Q3 '20

Q4 '20

Q1 '21

Operating Profit

(522)

(167)

261

359

283

327

809

575

594

EV Credit Sales

(216)

(111)

(134)

(133)

(354)

(428)

(397)

(401)

(518)

SBC

208

210

199

281

211

347

543

633

614

Auto Leasing Income

(215)

(208)

(221)

(225)

(239)

(268)

(265)

(280)

(297)

Auto Leasing Costs

117

106

117

119

122

148

145

148

160

Deferred Revenue

318

158

149

176

53

(73)

138

203

162

RVG Adjustment

92

15

(10)

0

0

23

7

0

0

Restructur'g (incl BTC)

44

117

0

0

0

0

0

0

(101)

Adjusted Op Profit (equal to the sum of the above numbers)

(174)

120

361

577

76

76

980

878

614

                   

Units Sold (thousands)(excludes leased)

61.6

89.2

88.1

103.2

82.4

86.2

129.6

167.0

171.3

                   

Wtd Avg Diluted Shares (approx)

865

885

897

901

994

1,029

1,064

1,099

1,133

Adj Op Prof/Diluted Share

(0.20)

0.14

0.40

0.64

0.08

0.07

0.92

0.80

0.54

I have graphed and regressed these numbers. However, since it is pretty clear that there has been a "step change" in the numbers once the Shanghai factory has begun large-scale production in 3Q 2020, I have treated the two periods separately. Here is the graph and regression for the pre-Shanghai period.

23322283-16203210263606741.png

This graph shows strong operating leverage, to the extent of .5mn per 1,000 cars, or incremental profits of ,493 per car. This number is implausibly high. It is also almost entirely driven by the sharp increases of the "miracle quarters" of Q3 and Q4 of 2019, when Tesla's operating profits shot up on small increases in production, in a way that experienced automotive analysts found highly questionable.

The graph for the results after Shanghai tell a very different story:

23322283-1620321025754853.png

This graph suggests that the company is now generating negative operating leverage to the tune of about ,394 per vehicle. This is also implausible, but it could be the effect of sales in China, which is the only source of growth for Tesla, benefitting in Q3 (and, to a lesser extent, in Q4) 2020 from the same type of "cherry picking" that pumped up Tesla's results when the Model 3 was introduced in the United States in Q3 and Q4 2018.

I have graphed the total adjusted operating profits. However, a more relevant metric for a shareholder would be to do this on a per-share basis, since the average diluted shares have exploded by more than 30% over this period. This would make all the numbers look considerably worse.

Conclusions

A key part of the Tesla bull case is the belief that the company will become enormously profitable once it achieves scale. Yet even a devoted Tesla bull must be wondering why this day never seems to arrive and why Tesla's minimal profitability is still dependent on ephemera like EV credit sales and gains from selling cryptocurrencies.

The purpose of this article has been to try to analyze this question using Tesla's financial information. This is a very "noisy" exercise, although I have attempted to adjust for some of the more obvious distortions. However, even with the noise, it is clear that, although Tesla is reducing AC by spreading its fixed costs over a higher volume, it is also having to cut prices to sell the increased production. Below are the figures for Tesla's ASP over the period under study. (To do this calculation, I have had to adjust Tesla's reported automotive sales revenue for things like EV credits, RVG adjustments (in this case, the gross amount), and deferred revenue (in this case, the amount only relating to automotive sales, which Tesla only discloses annually; I have had to allocate this to the quarters and the Q1 2021 number is assumed to repeat the Q4 2020 amount).)

23322283-1620321026346059.png

A reduction in ASP of almost 10% per annum, combined with share dilution of almost 13% per annum, is a mighty big headwind for Tesla shareholders to overcome.

Even after my adjustments, I think there is too much "noise" in the numbers to definitively conclude that Tesla has no operating leverage. However, the blithe assumption of the bulls that Tesla has huge operating leverage is equally unproven, an assumption that the failure of profitability to arrive already makes highly questionable.

It is also true that this analysis covers a period of relatively light competition for Tesla, which translates into a milder downward slope for the demand for its cars due to the lack of alternatives. This is no longer true. Barring a major breakthrough on something like battery technology or autonomous driving, the trade-off between volume and price can only get worse for Tesla going forward.

0
0
0
0
0
0
0 0 1963
Submit comment
    No comments yet

Nov 12 2021 - Tesla: Trillion Of Speculation

Apr 27 2020 - How Strong Is China's Leverage Over Tesla?

Aug 17 2020 - Tesla Investors Are Flying Blind In China

July 17 2020 - Tesla's Positive Feedback On Share Price And Possible B On S&P Inclusion

Oct 16 2020 - Major New Developments Racking Up For Tesla In Asia

Oct 16 2020 - Tesla: The Story Unfolds

My Fearless Forecasts For Tesla's 2020

Aug 13 2020 - Strong Drivers For Growth For Tesla Auto Business On 3 Continents

Submit media
Enter your nickname

Show

Show

Enter your email address and we will send you an email explaining how to change your password or activate your account.

Back to login form

Close