Uddrag fra Goldman, Nomura og Zerohedge:
Volatility finally did its job on Friday.
After a quarter of suppressed index vol due to the relentless grind higher in chip/memory stocks on the back of endless “spot up, vol up” market dynamics (traders buying calls to force a dealer gamma squeeze, pushing prices higher, leading to more call buying, squeezing dealers even more, until all prices had completely disconnected from fundamentals), the week ended with an explosive real reminder that positioning and complacency still matter. The VIX soared above 20 – the first 20-handle of the quarter after printing 15.40 yesterday (12th percentile of the last 6 months).
The intraday move was remarkable in its steady, methodical unwind of risk, coupled with rising fear in both the Vix…
… as well as the collapse of this bubble’s “ground zero“, South Korea, which spawned the memory bubble and which saw the KORU 3x levered Korea ETF lose more than half its value in 2 days, a move reminiscent of the 2018 collapse of the XIV inverse volatility ETN, which ended up with the termination of the product.
Incidentally, we pointed out that going into Friday, the pace of foreign account sales of Kospi stocks was unprecedented: literally every single day since May 6 had seen aggressive selling of Korea’s bubble stocks by foreigners, typically offloading to local momentum-chasing retail investors…
… resulting in this shocking chart of new highs and new lows: almost as if the original Hindenburg Omen formulation is for amateurs.
Still, it was only a matter of time before Korea’s levered bubble burst.
… which is bad news for those same Retail Investors who as we profiled two weeks ago, had “Maxed Out On Margin Debt, Choosing To “Risk Complete Collapse” Than Miss Stock Rally.” Well, it appears they got “complete collapse.”
According to the weekly recap note from Goldman’s Brian Garrett published after Friday’s close, what stands out most is the market was already paying for protection ahead of this move. Going into today, SPX 3m ATM implied vol closed at 15.3v vs 21d realized at just 10.15v – a +5.2v VRP, one of the wider gaps of the quarter. In hindsight, the Goldman trader notes, “the market was right to pay up.”
You could also see hedging demand quietly building under the surface all week:
- 1m SPX IV: 12.59v → 12.63v (+0.03v)
- 3m SPX IV: 14.87v → 15.33v (+0.47v)
- 6m SPX IV: 16.61v → 16.97v (+0.36v)
- 1y SPX IV: 18.43v → 18.73v (+0.29v)
Almost as if several big accounts decided (and successded) to rugpull retail investors just as bullish concentration surpassed dot com levels. Not surprisingly the most active option contract on Friday across single stocks, indexes and ETFs was the VIX 25 Call expiring in two weeks.
As for the longer-dated part of the curve, that was already lifting into payrolls even while spot kept grinding higher.
Skew was also quietly steepening into today:
- 3m 25d puts: 15.34v → 15.83v (+0.49v)
- 3m ATM: 14.87v → 15.33v (+0.47v)
- 3m 25d calls: 13.70v → 13.97v (+0.26v)
The 25d put-call wing differential widened to +1.86v from +1.63v last week. Put buyers were already getting more active before macro finally broke today.
Single-stock and sector vol told a similar story. Semis and tech ETF vol had already started getting bid aggressively all week before today’s tape:
- TSM +3.1v WoW to 49.7v
- XLK +1.5v to 31.7v
- QQQ +1.1v to 23.2v
- AAPL +1.5v to 25.5v
- XLE +1.5v to 27.4v on Iran/oil headlines
- NVDA actually lagged (-0.1v WoW) despite broader semi stress
In some ways, the unwind happened precisely how Nomura’s Charlie McElligott modeled it would as a reminder, this is what he said on May 28:
… we can start by seeing if Wednesday’s modest Tech sector pullback (and first meaningful drawdown in the Semis-Energy 50/50 Barbell in quite some time too, while we’re at it), which from the “Vol Signals” perspective COULD then begin to see 1) ATM Vols leaking lower, or perhaps see 2) Call Skew flattening (where we start to lose the OTM Call demand after rallying so far, so far, so incessantly) and-or 3) Skew steepening (with Puts firming relative to Calls softening)
IF any of these begin to pivot, it will tell us a lot about rotation potential…how “stuffed” the recent buyers of Tech now are, and whether there are incremental buyers remaining around to perpetuate Upside demand…
Because as is my common refrain, “Spot Up, Vol Up” often-times resolve in “messy” fashion, collapsing under the significant weight of its own $Delta, i.e. seeing a modest profit-taking turn into something uglier and forced / mechanical selling the lower we go….
This is especially the case then with the certainty that a period of Tech / AI thematic weakness in those Stocks / ETFs / Indices means the “pro-cyclical” Leveraged ETF rebal flows will agnostically turn the other way and need to sell more the lower Spot goes, particularly “Loaded” with regard de-risking flow potential—AUM we track across LevETF remains at ATHs ~$187B, and where 87% of said LevETF AUM underlying is in “AI Tech Leadership” ($67.6B in Tech, $39.0B in Semis, $26.2B in Mag&+ -linked products alone, BEFORE index -linked, where those same largest Hyperscaler / Semis / Memory names are > 41% of SPX and > 52% of NDX weights)
So now what.
According to Garrett “overall it felt less like outright panic and more like the market finally remembering that macro still matter.” Perhaps, although to thousands of reddit “traders” most of whom had gone max levered into whatever the top 0DTE flavor of the day was, Friday was the biggest shakeout event since covid and as for the Korea open on Monday, it won’t be a pretty sight.
Finally, for those who followed our advice in Thursday’s post, “Deep Dive Inside The Mechanics Of The SpaceX Offering: How To Trade The World’s Biggest IPO“, and added the recommended SMH long tails and Index protection trades…
Sell SMH Sep26 Put Ratio (Buying the Wing): Semis could become a source of funds to buy SpaceX. Given the extent of the rally (3x levered long semis >600% off March lows) and the short gamma, SMH wings have more potential to pay despite the high starting level of vol. SMH put skew is at YTD lows following the melt higher. Indicatively, SMH Sep26 590/500 Put Ratio – 1×3 (buying the wing), offered 10.5, -13d, ref 639.29
Index Protection: The selling flows in the equity market expected over the next month create risk of high volatility. VIX & VVIX are near YTD lows. Indicatively, VIX Jul26 40 Call, offered 0.56, 13d, spot VIX ref. 16.01. Whilst passive and retail flows we anticipate to be focused on NDX and Semis, SPX vol is more attractive to hedge a correlated move lower. QQQSPY vol spread is at YTD highs. Indicatively, SPY 10Jul26 725 Put, offered 5.1, -23d, ref 754.24.
… Friday was a very good day with more than 500% gains in less than 24 hours.
