Tesla Weekly (8/17): A Squeeze After The Yen Carry Trade Scare
Tesla and other "momo" stocks squeeze higher after last week's yen shock; It's all eerily similar to the August 2007 "stat-arb" sell-off which preceded the 2008 Great Financial Crisis
This is the second issue of Tesla Weekly, in which I’ve included more of my own trading experience this week, along with news that I think is significant for both Tesla and its rivals. This week’s report is divided into the following four parts:
How I traded Tesla and other stocks this week.
How the “yen carry trade” scare last week happened exactly 17 years after the “stat-arb” crash of August 6th, 2007, which preceded the Great Financial Crisis.
This week’s significant news on Tesla
Updated valuation tables including US and Chinese EV rivals of Tesla, as well as how they’re trading this week and YTD (see Figures 9 and 10).
Trading Thoughts This Week: That Squeezy Feeling
Thought I Covered Tesla Too Early: I had a sizeable short position in Tesla via puts which I sold when the stock hit $207 on Friday, August 2nd. I missed out on the 4.2% drop in the stock on Monday, August 5th, after the “yen carry trade” scare. The S&P 500 index has rebounded by 6.8% since that scare while Tesla has only risen by 2.9%.
Trying to Remain Disciplined in Shorting Tesla: As Tesla tried to break above $200 last week, I refrained from re-shorting it due to its low Relative Strength Index (RSI) of around 40 at the time (as a rule of thumb, an RSI of 30 or lower is oversold while 70 or higher is overbought).
Watching Musk Devolve Can Lead to Undisciplined Trading: However, the more unhinged Musk becomes, the harder it is to stay away from shorting Tesla, especially given the deteriorating fundamentals of an auto stock that trades at 117x earnings. So I re-shorted via puts on August 9th only to get my face ripped off this week as Tesla spiked to $216. I added the September 200 puts when Tesla hit $215 yesterday.
Toyota Long Hedge Softens the Blow: My hedge against the Tesla short I initiated on August 9th was Toyota (the October 18 call options at $175 strike), whose rise this week sprayed a sea mist of aloe vera on the remnants of my face that got ripped off by shorting Tesla. Toyota has the strongest fundamentals, globally, right now (13 days of inventory in the US versus an average of 68) and a 60% share of the surging global hybrid car market as EVs become increasingly unaffordable. On top of that, Toyota’s RSI was only 23 (extremely oversold) which was a big factor as the yen carry trade fears subsided.
Lesson Learned—Don’t Become Undisciplined: By simply watching the RSI on Tesla before shorting or covering in size, I did exceedingly well this year, but gave up some of those returns this week by not sticking to my rules. To be honest, it looks like Tesla could rise even higher from here based on the chart’s technicals (RSI is only 51, and other indicators like the MACD imply further upside). This is why I’m only at 50% of a full short position ahead of Q3 results on October 18th (which will be preceded by the October 10th “Robotaxi Day”—an event that will lead to a sell off, in my opinion). Why even be short if it looks like Tesla could rise higher? Because auto demand is quickly weakening (see the below section on US July sales) and Musk is becoming so unhinged that he could “burn the house down” with his recent antics (see section on Twitter’s finances below).
Last Week’s Yen Carry Scare Is Like the August 2007 “Stat-Arb” Crash Before the Great Financial Crisis
I was running a small, $20 million “carve-out” at the hedge fund where I worked as an analyst, and remember the week of August 6, 2007, very well. I bring this up as the “yen carry trade” shock happened on August 5, 2024, which is somewhat eerie.
In August 2007, I was long value and short momentum (growth). Credit jitters that summer rippled throughout the global markets and wreaked havoc on quant funds that month. The quants were the “market wizards” at the time, having gained 60% returns in the five years through August 2007 by being market-neutral.
The problem was that they all piled into the same stocks and got wiped out when signs of a subprime bubble emerged that summer. This is a great snippet from CNN Money on August 9, 2007, describing just how freaked out the market was as these “market wizards” frantically unwound their positions that week:
“The ECB loaned at least $130 billion in overnight funds to banks at a 4 percent rate. The Federal Reserve added $24 billion to temporary U.S. reserves in its regular overnight operations, an amount that some traders said was larger than usual, but not comparable to an infusion of money along the lines of the ECB, Reuters said.” (article here).
At the time, I was long value and short growth due to the S&P 500 having surged by 102% in the five years through July 2007 (which seemed like a lot at the time). As we headed into August 2007, the S&P 500 rose by as much as 1.7% during the first week, but plunged by 7.2% mid-month, only to close roughly flat versus July. I’m not trying to flex here, but things were much easier in those days (i.e. no QE distorting the markets), so my long portfolio dropped by only 0.2% while my short portfolio rose by 5.2%, for a net gain of 5% in August 2007.
To this day, I feel that I simply lucked out by having benefited by massive quant funds like Renaissance, AQR, and others reducing their exposures, leading my portfolio to outperform that month. And it is chilling to think that the “yen carry” crash on August 5th happened literally 17 years after the “stat-arb” crash of August, 6th 2007. In the 19 months after the August 2007 stat-arb crash, the S&P 500 dropped by 55%. Here’s a great article about what happened during the “stat-arb” crisis of August 2007 from the Wall Street Journal (link here). I’m not suggesting that history is now repeating itself, but simply flagging the similarities between the two crashes and how it started by big investors having piled into the same trades.
Tesla Removes Its Lower-Priced Cybertruck Variants From Its Order Site
Tesla’s troubled Cybertruck (which has only sold around 15,000 units since its November launch and seen 4 recalls) has had its more affordable versions at the $61,000 and $80,000 price points removed from its website. The lowest-price Cybertruck available now costs $100,000. The cheaper versions were originally available to order for delivery in 2025.
Given the accelerating trend in hybrid sales since 2H 2023 due to EVs being overpriced, this move by Tesla spells trouble, in my opinion.
It took roughly 6 years to develop the Cybertruck, at an estimated development cost of $3 billion and capex of roughly $1 billion. It has had 4 recalls for quality and safety issues this year.
There’s no way that Tesla would reduce demand via a price hike given all the costs that need to be recouped via mass-producing this model. Orders in Canada—where the Cybertruck was released last week—can be filled immediately despite talk of “over a million orders” globally. Order cancellations are huge, in my opinion, due to the high price and bad quality reputation.
What’s more is that Panasonic confirmed on their earnings call 2 weeks ago that they started shipping large quantities of their 4680 battery cells for the Cybertruck as of July 1st, so Cybertruck production is not “battery-constrained”.
Figure 1 shows a large number of Cybertrucks parked in a Houston parking lot last week. If this were the last few weeks of the quarter, pictures like this would make sense, as Tesla tends to deliver over 60% of global sales in the last month of the quarter. But we’re only midway through the second month of Q3 2024, so this doesn’t look healthy especially given the price changes (see the thread on Reddit here).
Figure 1: Tesla Cybertrucks Piling Up in Houston
Source: Reddit
July Light Vehicle Sales in the US Were Weak
US light vehicle sales in June were hampered by the cyberattack on US car dealerships and dropped by 3% YoY. While many thought this would lead to a strong rebound in July, US light vehicle sales were down 5% month-on-month and 2.8% lower YoY. This is bearish for Tesla, which has roughly 50% of its vehicle sales in the US.
Musk Angers the UK & Then Tells an EU MP To “F*ck [His] Own Face”—Tesla EU Sales Could Suffer
Violence broke out in the UK after 3 young girls were murdered by an unidentified teenager on July 31st. Musk retweeted and amplified claims by far-right British Twitter accounts that the murderer was an “immigrant” and then tweeted that “Civil war is inevitable” in the UK as these riots escalated.
I received loads of DMs from British friends saying that “Tesla is toast in the UK”. My British friends’ opinions are insignificant, but this is pretty serious as the UK is Europe’s second largest car market and Tesla’s sales have been struggling there (July sales were -22% YoY and -13% YTD).
Figure 2: Tesla UK Sales Are Way Under Peak in January 2023
Source: TMC
Musk virtually poured gasoline on the fire in Europe by tweeting the below meme (Figure 3) to an EU commissioner in charge of “EU censorship”. Many Europeans have already been turned off by Musk and it shows from Tesla’s struggling sales numbers there. The bigger problem is that Tesla has a German plant with 400,000 units of annual capacity to supply Europe but only produced 50% of that in the 1H of 2024.
Figure 3: Musk’s Tweet to an EU Commission Member
Twitter Finances Look Doomed—Musk May Be Forced To Sell Tesla Shares To Keep It Afloat
After telling advertisers who didn’t like Musk’s management of Twitter to “Go fuck yourself” last November, US ad sales on the platform plunged by 53% YoY in Q2 2024 to $114 million, according to the New York Times.
To give you an idea how bad this is, the last quarterly results that Twitter reported in Q2 2022 showed that US revenues were $661 million (see Figure 4 below). While Japan is the second largest market for Twitter (and may not have declined as much), the recent weakness of the yen and the Japanese economy may have also dragged down ad revenues from Twitter’s second largest market.
If the New York Time’s reporting of only $114 million of Q2 US revenues is correct, this implies an 83% drop from Q2 2022 US revenues of $661 million before Musk took over (see Figure 4 below). Using generous assumptions of a 35% drop in Japan revenues since Q2 2022 (the yen has weakened by nearly 9% since then) and flat revenues in the “Rest of the World” (highly unlikely), Q2 2024 total revenues at Twitter add up to $576 million.
Figure 4: Twitter’s Last Geographical Breakdown in Q2 2022
Source: Twitter Q2 2022 10-Q
We know that Twitter has annual interest payments on their debt of $1.3 billion ($325 million per quarter, with the next payment due in late October), so if Q2 revenues were only $576 million, it’s questionable whether Twitter can pay its interest expenses due in late October. There are also over $200 million worth of lawsuits related to ex-employees and unpaid vendors.
Twitter also became a key shareholder of Musk’s artificial intelligence start-up, xAI, last year. Despite burning cash, Twitter was said to have invested $250 million in xAI on top of Musk’s $750 million. The footnotes state that Twitter didn’t use cash, but lent its data center capabilities for a $250 million stake in xAI. This sounds like a clown show that only Musk could get away with, to me.
Why Were Call Option Volumes So High?
This week saw heavy buying of Tesla call options, which had a 39% spike in trading volume week-on-week (WoW) versus put volumes roughly flat at 2% WoW. My favorite Tesla options commentator on Twitter summed it up nicely yesterday (see Figure 5).
Figure 5: Call Buying Spiked Tesla This Week
Thursday saw the October 18 calls at 210 strike trade 34,341 contracts (a 4,187% increase in volume versus the previous day) and Friday’s top-traded option was the August 23 calls at 220 strike (see Figures 6 and 7 below). Tesla’s trading volume on Thursday and Friday were above the (declining) 2-week moving average for the first time since after its bad Q2 results were announced on July 23rd.
Figure 6: Top-Traded Tesla Options on Thursday
Source: Bloomberg
Figure 7: Top-Traded Tesla Options on Friday
Source: Bloomberg
It is said that Musk’s family office can trade Tesla weekly options without having to file a report of each trade (as their expiry is less than one month), although there’s no proof of this.
While this may all sound like a tin-foil hat theory, a look at Tesla’s trading volumes on Thursday and Friday (Figure 8) are telling. Except for Google last Wednesday, no other US tech stock in the “Magnificent 7” saw its volumes trade above the 15-day moving average this month.
Figure 8: Tesla’s Trading Volume Spikes On Last 2 Days of This Week
Source: Bloomberg
Stock Prices & Valuations Have Become Interesting
Toyota and Honda were the best performing stocks this week after having been whacked in last week’s “yen carry trade” scare. Rivian and Lucid were the worst performers. This might be a sign of a “flight to quality”. If you’re running an automotive portfolio, Toyota and Honda are likely the safest longs against shorts like Tesla, Carvana, etc., in my opinion.
GM remains the best performing auto stock in 2024: GM’s stock price has likely peaked, in my opinion. They’re a play on the US auto market, which is rapidly deteriorating (demand has dropped by 3% YoY in both June and July, while incentives have spiked). Chinese EV maker NIO is the worst-performing auto stock due to greater than expected cash burn and massive price wars in China. I wouldn’t be a dip-buyer of any Chinese auto stocks except for BYD and Li Auto at this point, unless they’re growing their exports or production in Europe and the APAC region (Geely falls into this category).
Tesla rose by 4.1% this week versus the S&P 500’s rise of 3.9% mostly on call buying. There was no positive news to cause Tesla to rise more than the market this week. The only positive catalyst I can see is that July sales (outside of North America, where monthly sales aren’t reported) appear to have risen by 20% YoY. This looks good and some prominent Tesla bulls tweeted about Q3 at Tesla likely seeing its first YoY rise in a long time. But here are the cold, hard facts: Tesla is offering aggressive financing deals (0% in China and 1.99% in the US for 5 years, both roughly 500 bps above the normal lending rates) and last year’s Q3 saw a shutdown of the Model 3 line in China, as Tesla retooled it for the refresh. This impacted China and EU sales the most and formed a low hurdle for Q3 this year.
Toyota seems the most attractive both from a fundamental standpoint and a valuation perspective. Inventory as of June end in the US was only 13 days (1/3 of its target rate in the US and way below the ideal rate of 60 days). Toyota also is the biggest winner of the huge surge in hybrid car sales—which is killing EV sales in most countries—and has over a 60% share of the global hybrid market.
Honda is very attractive from a shareholder return perspective: The Japanese government has put a flame under listed companies that trade below 1x book value (shareholders equity per share). Honda trades at a price to book ratio (PBR) of 0.7x. While Honda’s car business has a strong brand and is recovering well, it is still weak by global comparisons. However, Honda’s motorcycle business is worth more than two times its current market cap (based on rival valuations), which means the auto business is valued at zero. They have more net cash/equity than Tesla does to survive another “black swan” event like covid and are increasing dividends and buybacks to raise their share price. Tesla’s CEO seems to be doing the opposite.
Beware of “cheap” German auto stocks: If you look at Figure 10, the average 2025 EV/EBITDA (ex-Tesla), is 2.4x while the average PER is 6.7x. The Big-3 German carmakers trade on average at a 1.1x EV/EBITDA and a 3.8x PER. This is a value trap in my opinion. They have roughly 40% earnings exposure to China (the worst car market right now), and another 30% exposure to the EU market (the second worst car market, after China). The huge difference in Toyota’s 2025 EV/EBITDA of 4.2x versus Mercedes-Benz Group’s (MBG) 1.6x multiple is of keen interest, in my opinion (see Figure 10). Toyota is slated to generate a 13% pretax margin in 2025 while MBG is expected to make an 11% margin. Toyota’s EV/EBITDA valuation is 163% higher than MBG’s, yet its 2025 pretax margin is only 18% higher than MBG’s. Consensus may be too high for Toyota, but my sense is that the German carmakers face tougher times from here on, given the lower-than-expected demand for EVs, which is causing Toyota’s hybrid sales to surge. Hence the seemingly low valuations for German carmakers, which will likely negatively surprise in the coming quarters.
Figure 9: Top Performers This Week
Source: Bloomberg
Figure 10: 2025 Valuations & Returns Among Major Tesla Rivals
Source: MH estimates & Bloomberg
Great "cover the waterfront" summary, Brad, thank you.
Personally, I just avoid TSLA altogether. That said, I think the market has not yet absorbed the reality that Musk is likely to lose in Tornetta, and that Plaintiff will be awarded a record size legal fee. (But that's just one of many realities the market is ignoring.)
great stuff as always brad. couple thoughts here in case anyone is interested:
* TSLA up 4+% vs SPY up 3+% is underperforming given tsla's beta of 2.3x.
* on this week's move: jump in volume w/ a burst of call buying smells like a) short covering (thurs right after retail sales beat) and some aggressive call option activity as brad noted. that said, cannot ignore that correlations and volatility crashed & market makers went right back into volatility dispersion trades in options market (selling vol on SPY, buying it on TSLA and other basket stocks---this trade requires low correlations). --doesn't mean musk family wasn't in on the call buying, but this week has massive OI on DTEs (1.2MM+) due to 3rd friday phenomenon and that, along with falling correlations, might be reason enough for the call buying bonanza (in contrast, next week's 0dte expiry has current OI of 240k--and i doubt it'll beat 0.6M as it rises through the week). ---IMO the oct high call volume activity was again the 3rd fri phenomenon, guessing in support of dispersion trade as 3rd fri is when monthly index options expire; the total OI on SPY went up from 19MM to 20+MM on fri alone (im likely not capturing all exchange option OI, just reporting on SPY total agg OI).
* that said, i can get behind the notion that musk is in on call buying, despite it being a lousy week for him to try (high OI). ---he's acting like a cornered rat, and i'm not going to be the slightest bit surprised if he gets hit by a regulator (CA or SEC/DOJ) against FSD / autopilot in the immediate future. (there's only a handful on earth who know the timing: a couple DOJ / state attorneys, a bunch of tesla attorneys, and the tesla ceo). further handicapping why:
a) moving all businesses out of CA stinks of pre-positioning against any legal action against tsla, so he can cry foul on politicization of justice dept
b) endorsement of trump. ---he endorsed trump 7/13 IIRC and went all in on him before biden dropped out (7/21 or so), which is an incredibly reckless move for a CEO (normally send checks quietly). trump is transactional and would go w/ a quid pro quo w/ elon, so elon went public knowing that cozying up to trump before any regulatory action would further advance his 'politicization of justice' argument to the masses in light of any regulatory action against tsla, and that trump would call off the dogs if he wins (which it sure looked like when elon went all in on trump). (side note: his angling for a cabinet job in the X interview, in light of the collapse in the car business, sure seems like elon is thinking of the exits---his ego wont allow him to stay and watch tsla collapse on his watch)
c) lashing out at EU and b) above---tsla as a brand is dead to huge segments of the car buying population in CA, DE and UK and other EU countries b/c of musk's politics (all the analysts covering this stock have been eerily quiet on this point). elon has turned his back on the well-educated blue state libs who brought him to the dance, and while he may sell a few more cars in red states as a result, tsla's are in outright decline in CA, DE etc and that volume isn't getting made up by sales in the dakotas.
d) needs cash for twitter interest pmt. --just like w/ apr and early july, run it up before or immediately around bad news as liquidity & timing permits. and if regulators are coming, best come at it from a higher perch or let elon sell into strength before the 2x whammy
all that said, near term everything is on the table: have to agree w/ brad that it can go higher (ugh) as market might continue higher and drag tsla up with it. (those god damn CTAs and momentum funds love liquidity and only name beating TSLA on liquidity is NVDA---and it was up ~ 25% in 2 weeks IIRC)
or we can see a reversal lower in the market (as volatility typically clusters) and take tsla down---which could be a long ways.
or we can see the market go up and tsla down like we did in Q1, vs Q4 last year when market pulled tsla up, key diff b/w the two scenarios is the outlook for tsla:
12/31/23 at its peak $260: NTM earnings ~ 3.80 or 68 NTM fwd pe. while cal 25 eps ~ 5.35 yielded fwd p/e of 49 for cal 25
now: $216 w/ NTM ~ 2.85 or 77 NTM fwd pe while cal 25 eps 3.16 yields fwd pe 68 for cal 25 (and these earnings are due to get revised down hard for reasons brad articulates well).---so its already way more expensive than it was at the end of an epic run in Q4 23. can it get even more expensive ? sure, but only by hot money IMO (liquidity attraction)
tough freaking stock to short. i sincerely believe were it not for option liquidity this thing would already be < 80. i sure hope i make it there, these rallies are a mother******.