Tesla's Model 3 Mess Has Become Much Messier

| About: Tesla Motors (TSLA)


You know what was already complicated enough? Building a $35,000 Model 3 by 2020.

A $35,000 Model 3 in 2017? Nope, it just won’t happen. It will cost much more, and you won’t be able to buy one until the middle of 2018.

There are several “tells” that Tesla knows this: a California regulatory clue, failure to invest in Service Centers, failure to invest in Superchargers.

The Chevy Bolt is more evidence about both cost and timing.

Add the SolarCity merger, the Trump election, the Model 3 glass roof and macro worries, and it’s a recipe for disaster.

I began writing about the coming train wreck that is the Model 3, starting in a piece last January, which you can find here.

My principal skepticism was Model 3's announced price. I estimated Tesla (NASDAQ:TSLA), simply to break even, would need to charge at least $45,000 for the base version.

Four months after I wrote that article, Tesla shocked the auto industry by announcing the acceleration of the Model 3: Tesla would deliver between 100,000 to 200,000 Model 3 cars in 2017 and 400,000 in 2018.

A few weeks later (no coincidence), Tesla announced a capital raise of $1.7 billion for Tesla and another $682 million for CEO Elon Musk.

I am 99.98% confident the cheapest Model 3 will cost significantly more than $35,000. And I'm equally confident no production versions of the Model 3 will be available in 2017.


  • Any version of the Model 3 that anyone wants to buy will cost more than $45,000; and
  • Production versions of the Model 3 will not be here until Q2 of 2018, at the very earliest.

Here's a summary of what's changed over the past 11 months to fortify me in my view that, with the Model 3, Tesla is rolling toward disaster.

I. Everybody knows the Model 3 will be late.

With a nod to one of the greatest songs ever, by the late Leonard Cohen (featured in this terrific piece by David Remnick), let's consider what "everybody knows" about when the Model 3 will arrive.

Click to enlarge

(photo by Graeme Mitchell for the New Yorker)

Adam Jonas knows …

Morgan Stanley's Adam Jonas made news last week when he found the SolarCity merger added "zero value" for Tesla shareholders.

Overlooked in that story, though, was Jonas' stunning Model 3 delivery forecast:

We continue to forecast a Model 3 launch at the very end of 2018 (more than 1 year later than company target) with 60k units in 2019 and 130k units in 2020.

Read that again. The launch will occur at the "very end" of 2018.

I think Jonas was actually doing Tesla a favor by setting a price target of $242, even with that dispiriting news.

Indeed, I would not be surprised if Jonas' information about the likely delay came from Tesla itself. By putting the bad news out there now, and yet still maintaining the lofty price target, Tesla and its long-time Wall Street underwriter keep the dream alive.

… and other auto industry experts also know …

Relying on four different Seeking Alpha auto industry experts, I have repeatedly warned that Tesla is unlikely to be able to produce the Model 3 in 2017. You can read the detailed views of the experts here.

And, as always, Steve Funk's seminal article, here, although written six months ago, remains fresh and cogent in its analysis.

Not a single auto industry expert of my acquaintance believes otherwise.

… and Tesla knows, too.

In September, Ed Niedermeyer of the invaluable Daily Kanban wrote an exhaustively researched piece (here) detailing how documents on file at the Bay Area Air Quality Management Board (BAAQMB) reveal Tesla is at least a year away from even beginning work to expand its Fremont paint shop.

Two months have passed since the Niedermeyer article and an inquiry to BAAQMB indicates that, as of last week, Tesla still had not filed any start-up notice for any of its new paint shop equipment. By Tesla's own estimate, such a notice will predate by at least a year the actual operation of the new paint equipment.

The longer Tesla waits, the longer it is constrained by the 12-month, 25-car-per-hour limited described in the Niedermeyer article. That constraint translates to a theoretical maximum annual production of 220,000 units.

Of course, because of maintenance needs, no auto manufacturer can run a paint shop 24 hours per day. Niedermeyer estimates (based on expert input) that the real maximum for Tesla 2017 production is closer to 150,000 cars.

Contrast this with Elon Musk's forecast during the Q2 earnings call:

We'll probably exit next year at unit volume that's 200% or 300% what our current volume is, maybe 400%.

Given Tesla's Q3 production volume of a bit more than 25,000 cars, and its forecast that Q4 will be about the same, 400% of the 2016 exit volume would mean 400,000 cars in 2017.

That will not happen. Neither will 300,000 (300%). Nor even 200,000 (200%).

Tesla will be fortunate to produce 150,000 cars in 2017, and even more fortunate to sell that many at anything other than another grievous loss.

Click to enlarge

(Montana Skeptic mug in Hamburg, Germany. Photo by Q3 contest winner, Steinlaus Kielfornien)

Tesla's failure to send a start-up notice for the new paint factory facilities is an important "tell" that Tesla knows the Model 3 will not be here in 2017 (and, as detailed hereafter, there are other tells as well).

So, what's really going on here? Why, when it comes to the Model 3, is Tesla talking 2017 and acting like 2018?

Here's what I think: Tesla announced acceleration of Model 3 production early last May to generate market support for the capital raise that came a few weeks later. Even then, though, Tesla management knew the idea of volume production in 2017 was absurd.

And if some members of Tesla management did not know that, or were afraid to say it out loud, they certainly understood it once Peter Hochholdinger came aboard two months ago and sized up the manufacturing operation.

Tesla is now using its Model 3 deposit list to upsell Model S cars. That's a wise course, but it pulls forward Model S demand. So too does Tesla's announcement that free supercharging will end after Q4.

Eventually, the piper must be paid, and Tesla will pay heavily in 2017.

II. The Chevy Bolt shows why the Model 3 won't arrive in 2017.

GM has beaten Tesla to the punch with the first EV priced at under $40,000 and having a range of more than 200 miles.

Yes, it's here. This very week, car buyers in California and Oregon will be driving home in an electric vehicle that:

  • has 238 miles of range;
  • has more interior space than the Tesla Model S; and
  • costs a grand total of (before tax credits and other subsidies and incentives) $37,495.

Let that sink in. The Bolt is here, more than a year before the Model 3 will be here. And lots of knowledgeable auto people think the Bolt is a terrific car.

It's not simply that the Bolt will eat into Model 3 sales (it surely will, though to what extent is TBD).

And it's also not simply that the Bolt is likely to be more reliable than the Tesla's Model S and X have proven to be.

Nor is it that fact that the Bolt has thus far attracted enthusiastic reviews. For instance, Car & Driver:

The range and the price by themselves make the Bolt a car worthy of celebration. Drive it, though, and you'll discover the Bolt also is good enough to be worthy of all the breathless hype that's been showered on the Tesla Model 3.

No, the telling factor is not the fact that the Bolt is first, or more reliable, or enthusiastically greeted in real world reviews. Rather, the telling point is the timeline.

Even if you are inclined to sneer at the Bolt, and are convinced it is no competition for the Model 3, still, the Bolt's timeline is instructive about why the Model 3 will not arrive in 2017.

There have been scores of Bolts on the road since last March. These Bolts came off GM's Orion, Michigan, production line, and GM tested them endlessly to perfect its production process. GM was not willing to release the Bolt for sale until its pre-production testing had racked up hundreds of thousands of miles in all conditions and climates.

(Photo of a pre-production Bolt, courtesy of GM)

From an article last March in Fortune:

GM won't start producing retail-ready versions of its all-electric Chevy Bolt until the end of this year. While pre-production versions of the Chevy Bolt are drivable, they can't be sold to the public. Pre-production is a stage that allows engineers and factory workers to make sure parts fit, the manufacturing process works, and the dimensions on the car are right. For instance, workers might make sure the tool that installs the windshield works correctly, (GM spokeswoman Michelle) Malcho explains.

Eight months of extensive testing elapsed between the first Bolt rolling off the production line and the first Bolt rolling into a paying customer's garage.

Does anyone believe Tesla is better than GM at manufacturing cars and can safely shorten that testing period? Were there no lessons in the Model X debacle?

To have any realistic hope of undertaking adequate pre-production testing and still selling the Model 3 in 2017, Tesla would need to start rolling out production line cars by March.

That won't happen in March. It won't happen in April, or May, or July or October.

Possibly, just possibly, it will happen by December. And, if it does, you may be sure that, in line with the Model X script, Tesla will announce "release" of the Model 3 several months before volume production actually can begin.

III. The Bolt also shows why the Model 3 will cost much more than promised.

The Bolt shows what an immensely experienced automaker can produce and sell for $37,495.

An automaker, we might note, that is willing to sacrifice profits on its EV line to reduce its liability for the green indulgences it must pay by way of ZEV, GHG and CAFE credits. Because every Bolt sold at break-even, or indeed at a loss, is more profit GM can make from the larger cars its consumers want.

There's no way Tesla will be able to sell a decently equipped base version Model 3 for $35,000. We already suspected that, of course, but the Bolt is powerful proof on four wheels.

Before we leave the Bolt, let's dispel one further myth: that GM won't be able to make many of them. GM's Orion plant produces both the Bolt and the Sonic. It can manufacture about 90,000 cars per year, and can shift production between the two vehicles depending upon demand.

IV. The Model 3's roof is so cool that it's hot.

At one time, I believed the disastrous design decisions of the Model X - falcon wing doors and non-folding second row seats - would teach Tesla a lesson it would never forget.

The Model 3's sweeping glass roof, however, has me scratching my head.

The roof creates three potential problems.

First, such a large glass roof creates body stiffness and other design challenges. Tesla glass roofs already have an unfortunate tendency to leak.

Second, a sweeping glass roof means higher interior temperatures in sunny climates, which means more need for air conditioning, which (particularly in EVs) means a significantly reduced range.

Click to enlarge

Third, and most importantly, the sweeping roof means the Model 3 will have an awkwardly narrow trunk opening and consequently limited cargo capabilities. Once again, Tesla has opted for "cool" instead of practical.

Sure, the sweeping glass roof certainly contributes to the Model 3's sharp looks. (Just as the falcon wing doors and unusual second row seats contribute to the Model X's sharp looks.)

As always, though, there are trade-offs between sexy design and practical utility. How car buyers weigh that trade-off with the Model 3 is TBD, but it's certainly possible that the glass roof will prove to be a net negative.

(Note: Yes, I'm aware the second-row seating in the latest Model X five-seat version folds down to increase cargo space. However, the second row appears to be a bench, and does not fold into the floor, but instead sticks up at an angle. As a fix to the lack of folding second-row seats, Tesla's five-seat solution appears to be a sub-optimal kludge.)

V. Tesla's new "Autopilot" adds to the Model 3 cost.

What's the real back story on Tesla's spat with Mobileye (NYSE:MBLY)? I only know what I read (and the most insightful and informed piece I've read was by Mark Hibben here).

Regardless of who did what to whom, Tesla's switch to Nvidia's (NASDAQ:NVDA) platform increases the cost the Advanced Driver Assistance systems (ADAS) equipment, which will now include four cameras, 12 ultrasonic sensors, and a more powerful computer.

The new equipment will be incorporated not only into all new Model S and X cars, but also into all Model 3 cars. And that, of course, will bump up the Model 3 cost.

By how much? I've seen educated guesses of between $1,500 to $3,000.

Sure, Tesla will recoup some of the added cost from those who agree to pay the $5,000 required to activate the new hardware to the level required simply to enjoy the driver-assist features available with the EyeQ3 system on existing Teslas.

The point, though, is that adding the more expensive hardware to the Model 3 increases the car's cost, and hence its price. And it is an iron rule of economics that when price rises, demand falls.

As the price of the Model 3 continues to rise, the size of the addressable market falls off. The universe of potential buyers for a $50,000 car is considerably smaller than the universe of potential buyers for one costing $10,000 less.

At any average sales price of $50,000 for a small sedan with limited range, can Tesla hope for annual sales of 400,000? Or even 300,000? Or 200,000? Or, indeed, 100,000?

Can it harbor this hope as more EV competitors arrive in 2018?


VI. Tesla's failure to spend on Service Centers is a "tell."

If one credits Elon Musk's announcement that Tesla will produce at least 100,000 Model 3 cars next year and 400,000 in 2018, then it's clear Tesla badly needs more Service Centers and Superchargers.

Tesla's most recent letter to its shareholders stated:

During the quarter, we opened 17 new stores and service centers to increase our customer support network to 250 locations globally.

Tesla does not break out how many of the 17 are stores and how many are service centers, and there is a world of difference between a new store and a service center.

Behold, for instance, one of Tesla's new stores, on Greenwich Avenue in Greenwich, CT:

(Cygnus3000 photo)

This store, located in a small retail strip center, is nothing more than a "gallery" at which one may view, but not buy, a Tesla.

The store offers no service or repair facilities for any Tesla owner needing service. For that, the Greenwich owner must go to Mt. Kisco, NY (24 miles away) or Milford, CT (36 miles away).

There is just no getting around the numbers: Tesla is falling behind in building new Service Centers.

At the end of 2013, there were 513 Tesla cars per Service Center. By the beginning of 2014, the number had risen to 696. At year-end 2015, the number had climbed to 903.

Through the first three quarters of this year, Tesla added 54,033 cars but only 10 Service Centers. So, the ratio has now risen to an all-time high of 1,252 cars per Service Center.

Click to enlarge

Given these numbers, it's no surprise there is increasing discontent among Tesla owners about wait times for service appointments.

The Service Center shortage will become even more acute if Tesla meets its 25,000 delivery guidance for Q4 2016 and its 180,000 (low end) guidance for 2017. (Never mind the 500,000 deliveries promised for 2018.)

In the shareholder letter and during earnings calls, Musk and Jason Wheeler boasted about the deferral of capital spending as capex "efficiency."

I think the deferral of Service Center spending is a "tell" - Tesla management knows the Model 3 will not arrive in 2017, and hence it has not engaged in the needed spending to build out the Service Center infrastructure.

Remember: besides the Service Centers, Tesla will need a much more fully developed distribution network if it hopes to sell the Model 3 in the quantities promised. There's no evidence Tesla has even started to develop that network.

VII. Tesla's failure to spend on Superchargers is another "tell."

What's true of the Service Centers is also true of the Superchargers: the addition of Supercharger locations is not keeping pace with the number of Tesla cars on the road.

There were 661 Supercharger locations at the end of Q2. Another 54 locations were added during Q3, for an increase of 8.2%.

However, during Q3, the number of Tesla cars on the road grew from 136,652 to 161,473, making for an increase of 18.2%.

In other words, during Q3, the number of Tesla cars grew at more than double the rate of the number of Supercharger locations.

(Teslas lined up at Mountain View, CA Supercharger station on October 10, 2016.)

Sure, Supercharger overcrowding is a problem only at some locations, and even at those locations only at some times. But particularly in California, where Tesla density is greatest, the congestion at some locations is significant and growing.

And, given that state's high land prices, strict zoning regulations and abundant regulatory hurdles, solving the problem requires substantial capital investment. Investment which, this year at least, Tesla has chosen not to make at the pace required to keep up with its car production.

This problem will become much more acute when volume production of the Model 3 begins. (And, one can imagine the culture clash at the Supercharger stations when the Model 3 owner, who paid $30,000 after state and federal incentives, noses in on the Model X owner who paid four times as much, or more.)

Again, the fact that Tesla is dawdling in building out the Supercharger network is another "tell" that it knows the Model 3 won't be here in 2017.

VIII. The SolarCity merger will hinder the Model 3.

Let me be brief: SolarCity is an anchor around Tesla's neck.

Starting with the current quarter, Tesla already faces at least five deeply consecutive negative quarters. The SolarCity merger will deepen the operating losses, and the only question is by how much.

The unknowns include:

  • The terms of the deal Panasonic (OTCPK:PCRFY) and Tesla ultimately strike for the Riverbend factory in Buffalo;
  • What effect, if any, the continuing corruption surrounding the Riverbend project has on Tesla;
  • How quickly Tesla can slash SolarCity overhead; and
  • How much longer the market continues to put any stock in the absurd story of the solar roof (which EnerTuition has capably eviscerated here and Rogier van Vlissingen here).

Tesla management has spent much of this year with its focus diverted to the SolarCity merger (and, as well, has spent tens of millions of dollars in legal fees and investment banking fees).

The latest example of "squirrel!" are the solar roof tiles. Before the solar roof tiles it was autonomous driving, Tesla energy, battery swapping, wall snakes, etc.

Look for more squirrels in 2017 to continue distracting investor attention from the Model 3.

A Note about SolarCity's "Debt"

In a recent article discussing Musk's promise that Tesla will assume all SolarCity's "debt," I wrote that the debt totaled $6 billion. Seeking Alpha member Chuck Jones noted in a comment that SolarCity's balance sheet as of the end of Q3 included only about $3.4 billion of convertible notes, solar bonds, credit line debt.

My reckoning of the total SolarCity debt was certainly imprecise. I apologize for creating confusion, and thank Mr. Jones for illuminating the issue.

What remains unclear is what Musk meant by "debt." Does debt also include accounts payable and other current obligations? Contractual obligations to the State of New York? Build-to-suit lease liabilities?

In other words, does "debt" include all liabilities? Total liabilities on the Q3 balance sheet (the last one we shall ever see for SolarCity as a stand-alone entity) totaled $6.68 billion.

Just how broad the definition of SolarCity "debt" that Musk would have Tesla "assume" will be determined in the future when, in going to market for another capital raise, Tesla details the uses to which the funds will be put.

IX. The Trump election complicates the Model 3's future.

I'll certainly admit I didn't see that one coming.

Neither did Tesla CEO Musk. Here's what he said just four days before the election (I've corrected published reports of his remarks after listening to the audio):

Well, I feel a bit stronger that, probably, he's not the right guy. He just doesn't seem - he doesn't seem to have the sort of character that reflects well on the United States. It's just, I don't know. Just, no.

Anton Wahlman has speculated about the likely effect on Tesla of the Trump victory, and Seeking Alpha members have had plenty to say about it in the comments.

  • What will Trump's policies be?
  • Will CAFE standards be relaxed, or even repealed?
  • Will the 30% solar investment tax credit be reduced, or even eliminated?
  • Will the "blank check" cost of the year-long phase-down period of the federal electric vehicle tax credit be reformed?

When it comes to most policy issues, I find Trump to be unpredictable, probably because in many areas he lacks clearly formed policy views.

However, given his announced intention to nominate Myron Ebell as Secretary of Energy, and his generally negative views about global warming theories, it is easier to imagine more downside risks than upside benefits for Tesla.

Regardless of Trump's ultimate policy on some renewable energy issues, I would simply remind readers that one of Donald Trump's character traits is a thin-skinned sensitivity to criticism.

I think it's a good bet that Trump either made note of the Musk remarks quoted above, or that he will be reminded about them. (And, listening to the interview, one sees how easy it would have been for Musk to simply have made some anodyne statement. Alas, he couldn't resist weighing in for Hillary Clinton and disparaging Donald Trump as having "character" that does not "reflect well" on his country. "Just, no.")

Again, this one is unpredictable. For all I know, President-elect Trump could tomorrow announce he is nominating Elon Musk as Secretary of Transportation. But, at present, the negatives to Tesla of a Trump Administration are more apparent than the positives.

X. Other macro factors also complicate the Model 3's future.

We are again reminded of Yogi Berra's injunction: Predictions are difficult, especially about the future.

At the risk of being wrong about the dangers we see, and failing to detect those hidden from view, let's summarize some storm clouds on the horizon:

  • New car auto sales likely reached their peak in 2016. For the next few years, price pressure will be downwards.
  • The hypertrophic leasing activity of the past several years will soon result in a flood of used cars, creating yet more gravitational force on car prices.
  • The credit card binge that gave rise to much of the auto spending (and other consumer spending) may well have an ugly ending (for a smart take from that natty nabob, see the article and charts here).

I don't presume to say what 2017 will bring, except to note that after a seven-year bull market with unprecedented interest rate suppression and all-time highs in price/earnings ratios, some corrections to equity market exuberance would not come as a surprise.

Reality eventually catches up with us all. That's true even in the progressive paradise of California, which, as detailed here, here and here, will soon have to reckon with massive unfunded pension costs when doling out its lavish subsidies to "alternative" energy.

A note about Fear, Uncertainty and Doubt

I am ever so weary of reading that my articles should be dismissed as "FUD" (Fear, Uncertainty and Doubt).

To me, the accusation of "FUD" is intellectually bankrupt. Anyone investing in Tesla, long or short, who is not filled with a sense of fear, uncertainty and doubt is simply not paying attention to the many dangers of this most dangerous stock.

A note about my contributors

Special thanks this time to Bonaire and CoverDrive.

And, obviously, to Ed Niedermeyer of dailykanban.com who, I understand, has taken a respite from Twitter in order to focus on finishing a book.

Which, I would venture to guess, will make for fascinating reading.

Disclosure: I am/we are short TSLA VIA OPTIONS.

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.

Editor's Note: This article discusses one or more securities that do not trade on a major U.S. exchange. Please be aware of the risks associated with these stocks.