Tomi Profile picture
Dec 29, 2019 26 tweets 7 min read
1/ THREAD: Weak 1H 2020 will put significant pressure on current FY2020 and FY2021 consensus estimates, which should weigh on the share price in the coming quarters and risk changing the growth narrative.

Other risks are merely the icing on the cake (disc: short). Image
2/ Following 31% q/q deliveries decline in Q1 (as US M3 backlog was exhausted and Europe deliveries had not started en masse), managed to grow deliveries sequentially from Q1’s low base throughout the rest of FY2019.
3/ Q4 will follow Q2 and Q3 in setting new deliveries records, this time at >110k, driven by geographic expansion, release of RHD M3 and expiration of various tax incentives (NL, US FIT).

However, will barely hit the lower end of its delivery guidance of 360-400k in FY2019
4/ In Q4 2018 deliveries report said they ended FY2018 with a 350k deliveries run-rate, and FY2019 will end with ~365k deliveries. Either production didn’t improve at all, or manufactured to demand. (h/t @Badger24)

Declining ASPs suggest demand issue is more likely.
5/ In Q3 2019, auto revenue declined q/q and ~12% y/y. Q4 auto revenue may beat Q3 q/q comp depending on mix, but will be down y/y and mark first LTM revenue decline.

Revenue is already declining despite record deliveries, as mix (and thus ASPs) continue to deteriorate.
6/ To make matters worse, management is guiding to a “tough 1H 2020”. Soft Q1 y/y comp aside, how many q/q declines can the growth story take in a row?

Just meeting the consensus revenue estimates will be a tall order if 1H 2020 starts off far below current record deliveries. Image
7/ Consensus estimates call for $29.4bn of revenue in FY2020E, and $37.5bn in FY2021E.

Assuming stable contributions from Energy and Services segments, and a $55k ASP, that implies >450k and >600k deliveries in FY2020/FY2021, far above current demand and production capacity. Image
8/ will need to rush both MY and CT to market, having realized M3 demand will not sustain the growth without new geographies left to expand to.

Notably cumulative M3 deliveries post-Q4 will barely be at the “420k reservations” number, despite a $1,000 deposit at the time.
9/ MY and CT deliveries are unlikely to start in time to contribute much to 1H 2020 with having guided for MY deliveries starting in late 2020 (and Deutsche expecting early 2020 start) and CT in late 2021. famously doesn’t tend to hit their estimated timelines. ImageImage
10/ The 250k CT “deposits” were $100/each, 90% below M3 where most reservations never translated to sales. With no 300k update from Musk, it’s unlikely CT has gotten there.

Notably never put out a MY reservations number, saying it's not "relevant". Make of it what you will
11/ will need >100k of production from Shanghai in FY2020 to have a shot at their consensus revenue estimates, but that assumes demand for >100k incremental sales, even before looking at profitability, at a time when ASPs seemingly remain the only lever to increase demand.
12/ Consensus estimates call for $5.07 of EPS in FY2020E and $10.89 in FY2021E. $TSLA’s mix is not improving (i.e. ASP will only decrease further), meaning they must rely on further efficiencies and cost cuts to remain margin neutral – and consensus calls for a profitable year.
13/ In addition, need to sell enough vehicles from China to generate $300mm of annual tax revenue to avoid the risk of giving the factory back to the Chinese govt, which helped put it up in record time. That will add pressure to move metal, even at lower ASPs. Image
14/ YTD net income is negative $900mm despite $460mm of regulatory credit sales, with negative cash flow. And this is despite capex being below maintenance levels, even after including the capex hid under finance leases to boost reported FCF (h/t @orthereaboot).
15/ has never had a profitable (or cash flow positive) year, even when they were the only EV option for many consumers, and thus held strong pricing power. That advantage is disappearing with more and more EVs coming to market, putting more pressure on the company.
16/ Each year forward will bring in new and tougher competition from other OEMs, which will spread demand for EVs around a larger pool, mean less demand for regulatory credits, and decrease the impact of any incremental EV tax incentives for .
17/ Let us once more remind everyone this is a company with $13bn of debt and nearly $20bn of purchase obligations. A lot of creditors (not the least the Chinese government…) will want to get paid, which is why $TSLA’s cash generation ability remains important going forward.
18/ If cannot start generating cash to fund its operations and expansion, they will need to again tap the markets for more money. Musk can’t talk about 1mm robotaxis in FY2020 again, and growth is slowing and revenue declining. When will the narrative change?
19/ At some point, the market will realize is just a poorly ran OEM that cannot subsidize the unprofitable EVs with profitable ICE vehicles and cannot generate enough demand at prices that would make its operations sustainable. That is when the wrecking will occur.
20/ When the market comes to the realization that is a regular OEM at moderate to no growth, it should start trading down towards other OEMs – which are at the 0.5x EV/revenue and ~5x EV/EBITDA range.

1x EV/revenue implies $0 for equity; 2x ~60% downside.(h/t @WallStCynic) Image
21/ This has already been highlighted as a possibility in equity research by none other than long-time bull Adam Jonas, who is “…not bullish on Tesla longer term”.

What about the short to mid-term, if 1H 2020 sees two more quarters of q/q revenue decline? Image
22/ To summarize, without record deliveries in 3/4 quarters in FY2020 (seemingly unlikely), has no chance at hitting consensus revenue estimates, may extend their q/q revenue declines to 4 quarters, and risk the end of the growth narrative…
23/ …and that is when they are expected to turn the corner on profitability – something has yet to accomplish in 16 years of operations.

If have a weak 1H 2020, there will be pressure to revise down estimates. If that happens, the share price should reflect it.
24/ As @WallStCynic says, isn’t going bankrupt overnight. However, this year’s WeWork story is an illustrative example of just how quickly 80% of market value can evaporate.

Weak 1H 2020 may finally change the narrative. Trade carefully; all feedback welcome.
25/ Final note: This is not meant to be an exhaustive coverage of the story, which will one day make for one hell of a book.

There are countless other on-going risks such as federal investigations and questions about accounting, and has those covered.
26/ This news from China (h/t @TESLAcharts) reporting has no equipment to stamp, weld or paint vehicles at the Shanghai factory bodes well for my thesis.

Getting to consensus revenue may be impossible not only from demand, but also from supply perspective.

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Tomi

Tomi Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @tomi

Dec 27, 2021
1/ You might hear about the metaverse daily, but the concept remains confusing.

My framework for the metaverse: “the collective sum of all activity that takes place online” makes its profound impact on the world easy to understand with limited prior knowledge.
2/ Metaverse is sometimes thought of as a rich, seamlessly integrated alternate 3D universe, but that’s unlikely to ever materialize.

However, the Internet today is already facilitating countless meaningful human interactions and experiences that augment/amplify everyday lives.
3/ Life’s various constraints and the increasing convenience of the Internet are continuously driving more and more interactions online.

Online already dominates in quantity, so quality is increasingly going to determine the equilibrium of online vs. in-person interactions.
Read 25 tweets
Jul 10, 2021
1/ To help shine some light on the magnitude of the Counter-Strike ecosystem's issues, TO losses in 2019-2020 per public filings:
- ESL ~$80mm (incl. all games)
- BLAST ~$25mm

Combine team losses, and the ecosystem (excl. Valve) is easily losing upwards of $50mm/year.
2/ Worth noting that 2020 was actually the better year financially for ESL -- they were able to keep much of their sponsorships/revenue with significantly reduced costs, driven by layoffs and lack of offline events.

2019 was *worse* than 2020. Scary for return to "normal."
3/ Why are investors (note: not VC funds -- primarily family offices, strategics, etc.) funding this?

Assuming continuous growth (my pinned tweet's article explains some of the high-level rationale for industry as a whole), eventually this turns into a lucrative media business.
Read 20 tweets
Jul 9, 2021
1/ Thread: Some of the #IEM games have seemingly lasted forever, so I went through two extreme examples to understand the breakdown of time.

Summary: Counter-Strike needs rule changes to make it more viewer friendly for a growing part of the audience. @CSGO
2/ Astralis vs FaZe (nuke only)

Actual game: 1h 14m 38s / 71%
Freezetime: 15m 40s / 15%
Total game: 1h 30m 18s / 86%
Tech: 4m 30s / 4%
Timeouts: 4m / 4%
Breaks: 6m 26s / 6%

Total: 1h 45m 14s

Series ended 2-0, but lasted over 3h 7m (due to OT). #IEM
3/ Liquid vs mouz (inferno only) #IEM

Actual game: 1h 35m 58s / 70%
Freezetime: 17m 40s / 13%
Total game: 1h 53m 38s / 82%
Tech: 12m / 9%
Timeouts: 4m / 3%
Breaks: 8m 17s / 6%

Total: 2h 17m 55s

Series went the distance (incl. OT), and took over 4h 50m (!) to finish. #IEM
Read 12 tweets
May 14, 2020
1/ THREAD: I believe the increasing length of matches is becoming a problem in Counter-Strike.

Maps used to run at most 40-45 minutes with BO3 series often finishing in ~2h30. Now +4 hours long series are common, as we’ve recently seen with matches running late into the night.
2/ Various fundamental rule changes have led to a gradual increase in match length in the past years. There include increases in:

- freezetime
- roundtime
- bomb timer
- timeouts
- half-time break

While each alone is relatively minor, cumulatively the impact is meaningful.
3/ In addition, the modern economy in CS:GO appears to make matches more drawn out – teams don’t tend to get blown out because there are fewer save rounds, leading to an increase in rounds played per map on average, compounding the impact of each individual change outlined above.
Read 21 tweets
Apr 26, 2020
1/ THREAD: With sports on pause globally, there's talk about esports (competitive gaming) being a beneficiary, as many tournaments and competitions can be held online and thus take place, albeit in a different setting, regardless of restrictions imposed on society by COVID-19.
2/ Over the past few years, we’ve seen an increasing number of headline-worthy comparisons between traditional sports and esports, many of them comparing apples to oranges, at best, if we're being generous.
3/ Examples include comparing total unique viewers who tuned in even for a second across a week(s) long tournament on a free online live stream, to pay-TV AMA for a 3 hour game, or even worse – those who attended the Super Bowl in person.
Read 34 tweets
May 25, 2019
1/ Finally read the Kotaku esports bubble article.

I'm obviously biased, with a vested interest in the industry (and teams in particular) doing well, but I think the text missed the mark in many ways.

Thread below, focusing on the team (arguably most criticized) POV.
2/ Much like with other fast-growing, nascent industries, there are of course plenty of bad actors (see e.g. @DenialEsports) who use the apples to kiwis comparisons (differing viewership metrics, full events vs. single games, etc.), but it doesn't mean it's standard practice.
3/ From my experience, the smaller teams (w/ most to gain) tend to play more fast and loose, whereas the bigger teams (w/ most to lose) tend to be more conservative with data, assumptions, etc.

Bad actors can exist anywhere, but faking numbers isn't a viable long-term strategy.
Read 22 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!

:(