Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable

[MP3Mike]

Yes it does. You need to pay for a loaner for the customer while their car is waiting for those parts.

Not for most of them as they are collision related and not covered by warranty.

 

[Nocturnal]

Playing devils advocate, one possible explanation would be “hardcore cost cutting”. Customers downtime doesn’t cost you immediately.

Yeah that's possible. I'm sure it's not a large number, but if a car is sitting there waiting on a few grand in repairs, that's money you could be pulling in. (aside from costs related to rental cars, customer satisfaction etc.)

 

[sparcs]

All this talk of taking Tesla is foolish. How does one envision Tesla operating going forward under such a massive debt load, that is high now? As EM has hinted, that ship has sailed.

ReflexFunds had some analysis on Tesla's operations in the near term were there to be some adverse issues like demand to deal with: Tesla, TSLA & the Investment World: the 2019 Investors' Roundtable

This is a good question since it’s clearly on investor’s minds so i think it's worth answering in detail.

To start with, I think it’s important to note that cash flow breakeven is much more important than profit and I think Tesla can get Q2-Q4 to cash flow breakeven even with disappointing demand.

• Cash flow is difficult to adjust into a run rate format, but in Q1 free cash flow was very close to zero excluding 1) working capital movements 2) one off cash/deferred revenue received from FCA and FSD special offer and 3) one off cash exceptional restructuring costs. This is the cash flow Tesla would make if it maintained production & deliveries at 51k Model 3 and 12k S&X on a long term basis (working capital is only a significant cash flow item if production and deliveries do not match and if volume is increasing or decreasing significantly quarter on quarter).
• So this is the baseline we are starting with, and clearly several key variables will change run rate free cash flow from this current level of 0 including: 1) change in car volume (likely to increase from Q1 levels), 2) change in cash gross profit per car, 3) change in opex, 4) change in capex (likely to increase from Q1 levels) and 5) will production and deliveries not match (this will cause a one off inventory change in the cash flow, but Tesla will reduce production rather than continue building inventory indefinitely).
• To build GF3 and start installing Model Y equipment, Tesla Is guiding for capex of $2-2.5bn in 2019. This implies c.$0.6-0.7bn per quarter in Q2-Q4 or c.$0.4bn higher than Q1 capex. So adjusting for the increased pace of growth investment but maintaining deliveries flat with Q1, Tesla is on track to reduce cash balance by $0.4bn per quarter in Q2-Q4.
• In addition to this, Tesla intends to continue to end the wave of deliveries at quarter end – they have already made a significant inventory investment towards this already (partly by accident through delayed Q1 deliveries) but this may require another $0.5-1bn inventory investment over the next two quarters. This does have the benefit of making quarter end cash balance much closer to intra-quarter minimum cash balance – which means Tesla’s quarter end liquidity can potentially reduce to as low at c.$0.5-1bn.
• So in total we have Tesla investing $1.7bn to $2.2bn cash balance over Q2-Q4 2019 if all else is kept equal with Q1. This would leave them with $2.2-2.7bn cash balance at year end together with $1bn undrawn bank lines – so $2.2bn to $3.2bn excess liquidity.
• So even with this worst case of no improvement from Q1 will leave Tesla with enough cash to get GF3 into production and likely also Model Y, at which point positive cash flow and profit becomes extremely easy.

But, is it reasonable to assume Tesla deliveries do not improve from 51k Model 3 and 12k S&X over Q2-Q4? This seems extremely unlikely.

• On Model S/X demand side Q1 was impacted by 1) tax credit demand pull forward, 2) discontinuation of the most popular base models for most of the quarter, 3) Customers waiting for a highly rumoured refresh, 4) unfavourable auto market seasonality. In Q2 none of these are still issues (the limitation is mostly ramping production back up to speed), in addition to benefits from lower pricing.
• On Model 3 side – Now Tesla has 1) SR+ availability increasing affordability, 2) SR+ availability meaning current limited cell supply can produce more cars, 3) lower pricing, 4) faded tax credit demand pull forward impact, 5) much better charging speeds, 6) availability in more countries.
• And higher deliveries will also mean higher cash gross profit for each car due to operating leverage benefits on production staff costs. This will however be offset to some degree by reduced ASP now SR+ is available.
• Cash profit per car is also heavily dependent on FSD take rate. Whether or not you believe in Robotaxis, it seems almost guaranteed that Tesla’s new HW3 and recent pace of progress will lead to significantly upgraded FSD functionality this year which is very likely to lead to a higher FSD take rate.
• So all together, these changes vs Q1 should comfortably take Q2-Q4 cash flow to breakeven, even without tesla getting close to their 360-400k delivery guidance. This is why Tesla Is guiding for positive free cash flow in all remaining quarters this year, even despite the delivery wave unwind.

But this does not mean demand is not an issue at all.

• I think Tesla is most likely going to be able to sell full production this year without any significant actions.
• But it is definitely true that demand is significantly lower than justified by Tesla's product/price superiority due to 1) lacking public awareness of the TCO of EVs, the performance and price of Tesla’s and the huge number of people killed by ICE emissions and 2) heavy PR and FUD attacks against Tesla trying to create a misleading impression of poor quality, fire danger, crash danger, insolvency risk, range anxiety etc.
• Whether Tesla demand would be 1.5x or 5x current production if the whole world was well educated about Tesla vehicles, I don’t know, but to some extent I'm sure Tesla is negatively impacted by this.
• My guess is this is currently impacting Tesla’s pricing power and Tesla would raise prices if people were better educated and demand far outpaced supply.
• But It is also possible that this public ignorance and miseducation will lead to Model 3 demand levels below production capacity. If this scenario plays out Tesla may choose to make some of the more drastic changes you are referring to.

What can Tesla do in the (unlikely given current information in my view) scenario that Model 3 demand does fall below production capacity?

• The first short term move will be to move Model 3 from three to two shifts of production. Most if not all parts of the production line should be able to produce at 4.5k per week on two shifts. This will prevent production outpacing supply and prevent inventory cash outflows.
• Tesla can raise more capital (within 24 hours) if they decide they need more cash to bridge to Model Y production.
• Elon will be forced to finally accept that advertising is a necessary evil for the greater good. If Tesla spends money to better educate people about the benefits of EVs and specs of Teslas Tesla will definitely get incremental demand.
• Tesla could partner with a dealership to open up another new sales channel. FCA in Europe seems an obvious partner given FCA reduces its emissions penalties by $10k for every extra car Tesla sells in Europe next year.
• Tesla could form a JV with an auto company using Tesla’s leading EV powertrain and battery technology. I’m sure many companies would be willing to partner on this given how far ahead Tesla is on technology – it will be Tesla holding out to go it alone at the moment. If Tesla puts its technology into a new brand, it can sell its technology through a new sales channel without comprising on its core Tesla business model.

What about after 2019?

• Tesla’s struggle to get to profitability and positive cash flow is entirely due to operating leverage and lack of scale. Most auto competitors have far lower gross profit per car relative to Tesla, but they need to sell 5 million to 10 million cars per year to leverage their fixed opex costs. Tesla is designing a more efficient business model that is close to breakeven with only c0.25 million cars per year.
• Tesla’s battle gets far easier when they have more factories and more car models. The gross profit from each new factory flows straight through to Tesla’s bottom line and will very rapidly turn Tesla profitable. GF3 will without a doubt increase China demand beyond current levels with lower production costs. Model Y and Semi will without a doubt plug significant new gross profit into Tesla’s largely flat fixed cost base. So in 2020 Tesla’s battle gets far easier, and its very difficult to see a scenario where their current cash balance does not bridge them to Model Y production and positive free cash flow.
• In addition, our best information is that Tesla will get around $0.6bn yoy increase in profit in the EU in 2020 from increased emissions pooling payments.
• Of course, if Tesla does deliver on Robotaxis as Elon expects, all profit and cash flow questions will be answered for good overnight.

 

[gluu]

But there is no greater due diligence than that you'll find on this board about ways to get an idea of production and deliveries. The people on this board have done and continue to do lot of research and find clues into how production is going. I don't for a second think that hedge funds have found a way to track production and deliveries that this board hasn't already tried to do. So the only other way they would "know" is insider information and to me, what you said is fear mongering because there's absolutely nothing to back that up.

AND

Yes.....we do know Q1 was about production constraints. We know they're cell limited on the 3 and we know they had issues with the Raven update. The fact that say it's unclear it was production constraints or demand tells me all I need to know. There's tons of supporting evidence that all says production constrained. If they don't meet Q2 delivery guidance it will be because of production, not demand.

As a buy-side professional I find the TMC estimations far from what alternative data providers are able to do. Alphahat and Second Measure are two examples. One is able to use cell phone tracking data to see when people visit delivery centers, the other is able to scour aggregated credit card transactions. Having this data is extremely valuable. Since I don't trade TSLA professionally I don't have access to that data, but if I had enough money in TSLA I would potentially buy access to that data. There are probably more that I'm not familiar with.

Your chain of reasoning about me 'fear-mongering' is based on faulty assumptions. Please reconsider before you continue to attack a fellow bull.

 

[Tslynk67]

TechCastDaily’s new podcast is a 2 hour long interview with Matt Joyce who was at the autonomy event. Pretty good wide ranging discussion.

Blubrry PowerPress Player

Interestingly, Matt said Elon told a small group of investors that Tesla would hit 10x the current market cap in 3-5 years.

There he goes again!

 

[Remus]

I'm super bullish on Tesla long term, please don't accuse me of FUD. Being wise is different than fear mongering. I work in the financial industry and alternative data sources are a huge source of edge. If one had sufficient capital under management it would make sense to do really rigorous P&D estimation, and these big funds fall in that area. It may have been coincidence that Fidelity sold in March, but there's some non-zero probability that they had insight into P&D before that number was given by Tesla.

We can read the tea leaves in June based on volumes/price action, but make no mistake, this is a really high leverage moment for Tesla. In Q1 it was unclear if Tesla simply had production constraints or was demand limited. The latter would be validated if Tesla fails to increase deliveries in a quarter with a tax credit cliff. I'm personally conflicted about when TSLA will bounce back...it may be the Q2 P&D report or Q3.

There is no denial that somebody has alternative source of information. For instance Google has pretty accurate data on the traffic to unemployment benefit application website, via which they can put pretty safe bet on SP500.

However there is no evidence these institutions, without big data sources such as Google, would put boots on the ground collecting data w.r.t. to all their significant holdings. The cost seems too high. Being a financial illiterate, I just checked 13F of Q3 last year, don't see many institutions gets in early, before the surprise profit. there is a high probability that I misread the 13F though.

 

[StarFoxisDown!]

As a buy-side professional I find the TMC estimations far from what alternative data providers are able to do. Alphahat and Second Measure are two examples. One is able to use cell phone tracking data to see when people visit delivery centers, the other is able to scour aggregated credit card transactions. Having this data is extremely valuable. Since I don't trade TSLA professionally I don't have access to that data, but if I had enough money in TSLA I would potentially buy access to that data. There are probably more that I'm not familiar with.

Your chain of reasoning about me 'fear-mongering' is based on faulty assumptions. Please reconsider before you continue to attack a fellow bull.

Yeah...….I'm not going to reconsider anything because I didn't attack anyone. As you just mentioned yourself, tools such as Alphahat are not exclusive to Fidelity and T Rowe Price. The comment insinuated very clearly that the funds that sold such as Fidelity have access to information that we can't get ahold of. That's false.

 

[Fred42]

There he goes again!

Of course this reflects his vision and optimism on Robotaxis. He believes this path to success.

 

[gluu]

There is no denial that somebody has alternative source of information. For instance Google has pretty accurate data on the traffic to unemployment benefit application website, via which they can put pretty safe bet on SP500.

However there is no evidence these institutions, without big data sources such as Google, would put boots on the ground collecting data w.r.t. to all their significant holdings. The cost seems too high. Being a financial illiterate, I just checked 13F of Q3 last year, don't see many institutions gets in early, before the surprise profit. there is a high probability that I misread the 13F though.

These kinds of things are done in secret. Not so secretly, Loup Ventures recently attempted to count Model 3's coming out of the factory back in Q2 2018 I believe. I'm not sure how much they have under management but its not that expensive to have boots on the ground to count things.

 

[gluu]

Yeah...….I'm not going to reconsider anything because I didn't attack anyone. As you just mentioned yourself, tools such as Alphahat are not exclusive to Fidelity and T Rowe Price. The comment insinuated very clearly that the funds that sold such as Fidelity have access to information that we can't get ahold of. That's false.

Is there a shared subscription to AlphaHat and Second Measure here on TMC that I'm unaware of? I'm saying individual investors typically do not buy these data sources because they don't have enough capital at risk to make it worth it.

 

[AlMc]

The current continued negative narrative (some filed by EM/Tesla itself) has, IMO, hurt some sales. I was at a weekend get together in Massachusetts with family and friends. Some are, agin IMO, fairly astute investors and to a man/women they had serious concerns mostly based on the continued negative droning about Institutional selling and having some of them and their extended family and friends opting not to consider buying a Tesla.

The brand image can recover but we need a MUCH better internal and external comms team. EM should have ALL his emails to staff vetted by a competent comms person.

Just my opinion. Opinion of one.

Will be buying more at $180 IF we get there.

 

[NicoV]

The car will become and entertainment machine that happens to also drive you around. Similar to smart phone being a full computer that also happens to make calls. Long and strong

It also shows how strongly Telsa thinks that FSD will become a reality, as it’s pretty useless without fsd: without fsd the passenger could use it, but then the driver would lose all screen info.

 

[CaliBear1]

I don't think the earlier poster is correct about Tesla stock dipping below $200/share as being a catalyst for more selling..I believe that the stock will be okay if it closes above $200/share...an intraday reversal is a healthy sign...if it dips below a strategic price pint and rebounds...

Right about now I need a "strategic price pint." A Guinness would be perfect.

 

[Remus]

These kinds of things are done in secret. Not so secretly, Loup Ventures recently attempted to count Model 3's coming out of the factory back in Q2 2018 I believe. I'm not sure how much they have under management but its not that expensive to have boots on the ground to count things.

No matter how secret it is, what is the benefit if they don't trade it? If they didn't get in before the surprised q3 result last year, how much confident you have that they knew?

Don't tell me they can hide their trade too.

 

[kakamora]

How and why can't Tesla match the cuts ?

Tesla's operational costs will be low compared to ICE - and nobody else has shown an ability to make profitable cheaper EVs. Don't assume others will have cheaper cars, even ICE - they apparently will rely on more expensive sensors.

ps : With 1 M cars and $1/mile charge, TN makes $81B per year. If they cut to 0.50/mile, they make $40B. Currently Uber has highly discounted rate of $2/mile avg.

Vot? 1MM cars will have to drive 81,000 Miles a year per car at $1/Mile to make 81BN for the network - of which TN's cut will be 20 to 30%.

81000Miles/year/car seems high no?

 

[Matias]

Yeah...….I'm not going to reconsider anything because I didn't attack anyone. As you just mentioned yourself, tools such as Alphahat are not exclusive to Fidelity and T Rowe Price. The comment insinuated very clearly that the funds that sold such as Fidelity have access to information that we can't get ahold of. That's false.

How we get ahold of Alphahat and Second Measure data?

 

[UnknownSoldier]

I took losses to get out the day of the earnings report but honestly I look like a genius right now to bail when I did.

There aren't any support levels left according to TA so this is uncharted territory and there is no bottom. So let's see how far down it can go I guess.

 

[bhtooefr]

Re: costs of porting game engines to the MCU: I suspect those are actually rather low.

The MCU is already based on commodity PC hardware, the game engines already run on Linux on that commodity PC hardware, so it's just a matter of integrating them into Tesla's GUI instead of a normal Linux/X11 GUI.

I want to see management taking customers seriously for once. Cars designed to suit a larger demographic of customers better for actual use. Not just a Model 3 designed to compete with the sedan version of the BMW 3-series, etc. Truth is, those are show-off cars. People who NEED cars to facilitate life itself, drive hatchbacks, estates, etc. A fixed parcel shelf, letterbox boot, small frunk, it just doesn't add up to a car the market will demand in limitless numbers.
Small capitulation on Tesla's part: tow hooks are being allowed, shruggedly, going forward. Perhaps there is hope.

There's multiple markets.

In the US and some developing markets, hatchbacks and wagons aren't popular due to image concerns (hatchbacks being the cheap option for those who can't afford a sedan, wagons in the US having picked up a stigma of being uncool family cars), and we do not buy them, unless we're Jalopnik readers demanding the mythical brown manual turbodiesel RWD wagon (and then we still don't buy them new). Let's put it this way: the Golf is being discontinued in the US market, because it doesn't sell enough, with the exception of the GTI and Golf R (which have enough enthusiast buyers to justify their existence).

Until the past 5-10 years, sedans were the normal "practical" cars in the US. There were options if you needed more than a sedan could handle - up through the mid 1980s, wagons were it, then minivans started replacing wagons, and in the mid 1990s, SUVs started replacing minivans. And, there were pickups if you needed more than that.

Now, though, crossovers (car-based SUVs) are replacing sedans. They're simply hatchbacks and wagons that, by virtue of being lifted a few inches and then stretched vertically a few more inches, and given butch styling to look like a rugged off-road vehicle, Americans are actually willing to buy.

So, Tesla's version of the car that you want is the Model Y.

(No, I don't like this. However, I suspect that a Model Y would have higher efficiency than the car you actually want - the drag coefficient of a cut-off Model 3 hatchback would likely cause more drag than the Model Y's additional height causes.)

 

[NicoV]

This is Tesla stupidity at its finest. The CEO is personally reviewing every one in 10 invoices with the CFO taking a look at all of them, you brand gets a beating because it can't communicate, logistics are a shamble and all of your products are not rolling out due to production issues, yet this is A-OK? Spending salary to port game engines to a platform just because you can? No. I can't see it being anything else as Elon's personal little toy idea, everything else be damned.

It’s not a toy idea if you (I mean Tesla) think fsd is arriving soon. I don’t think everybody in an fsd car will kill the time by making pornhub videos, so playing games is a good alternative. An watching movies, I’m pretty sure they will make a Netflix client too. The Model 3 screen is already in the right orientation.

 

[Nocturnal]

Vot? 1MM cars will have to drive 81,000 Miles a year per car at $1/Mile to make 81BN for the network - of which TN's cut will be 20 to 30%.

81000Miles/year/car seems high no?

Not that high. Full time Uber drivers hit over 1k a week. Plenty of regular owners put 20-30k a year on their cars.
A robot car could theoretically double a full time driver's usage. If taxi prices come down by half due to robo taxis, then demand will increase as well.