A brief devil's advocate to the SEC Report on Gamestop.
Only 9 months later, or a few seconds in government time, the SEC has released a long-awaited report on the causes of the 2021 Gamestop meme squeeze.
Perhaps disappointingly, they point the finger at the army of retail traders who piled into the squeeze, dismissing theories that short interest (although a factor) and options gamma (although also a factor) were the likely culprits of the massive momentum we observed in late January.
And this is fundamentally correct, regardless of your viewpoint — as I’ve mentioned before, options primarily distort momentum effects; they do not usually by themselves cause the fire. This is fairly intuitive in the idea of gamma regimes — as option dominance (which is measured by NOPE or NOPE-like factors) increases, the importance of short versus long gamma regimes becomes a significant driver of intraday momentum effects.
That said, I’m a dyed-in-the-wool optionista, so I’m going to mainly ignore the parts of the report I found boring or don’t have anything new to add or posit about. I’d fully recommend reading the report yourself as well to get an informed opinion:
https://www.sec.gov/files/staff-report-equity-options-market-struction-conditions-early-2021.pdf
Before I begin some argumentation, I want to make it clear that I in full faith believe the SEC information presented in the report is accurate. It is very likely the SEC worked directly with exchanges and market makers to analyze the options-related data of the event, and therefore I will accept their conclusions on the options market structure of January 2021 Gamestop. These conclusions off the bat are:
1) Options trading volume was observed to increase substantially from January 21st ($58.5 million) to peak at $2.4 billion on January 27th. This represents a meteoric 4100% increase in volume.
2) Based on (I assume good) options data, they conclude that market participants were primarily buying put options on Gamestop and selling call options. I will conclude they most likely have access to high-level exchange information and can see the parties on both sides of option trades, so I will accept this finding.
However, this seems remarkably inconsistent with later findings that call options comprised a high fraction of intraday dollar volume. As noted previously by me, the primary factor in driving the option squeeze was the massive increase in implied volatility, which similarly caused nearly the entire option chain to behave as 50 delta options.
3) Similarly, the report concludes in the heat of the battle, individual customer accounts (aka retail) reached a peak of 91% of non-market maker volume in options. This is very believable given the massive volume contributed by TD Ameritrade and Robinhood alone (nearly half), but is also fairly inconsistent with retail buying Gamestop stock. It is possible but unlikely that put options were also purchased as a way to hedge the stock purchased. However, given this protective put strategy is largely used for tax reasons and by sophisticated investors (it is essentially a synthetic call option, so you are no better off in this case than just buying a call — actually you are significantly worse off due to margining), it seems again very unlikely this is the case.
4) The definition used for gamma squeeze by the SEC is fundamentally incorrect, and only refers to the long side of a gamma squeeze (caused by hedging call options driving up demand to long the underlying).
However, as noted in previous options posts, selling a put option also incurs negative gamma; in this regard, put hedging will require shorting the underlying (assuming naive dynamic hedging, especially if orders cannot be matched).
To rectify some inconsistency in the report with observed behavior (namely the 0-day effect I originally blogged about), we can think about the scenario constructed by the SEC — where market makers are net long calls and short puts.
This is a situation which historically occurs on SPX and related indices with some frequency, due to the massive AUM controlled by systematic option overlay strategies (e.g. the JPM collar). However, the implication here is that there was a substantial imbalance — in this case, it’s implied in the report that long put option flow from non-market makers predominated even compared to short call options (still a bit hard to believe).
In this case, it’s not a huge reach to argue that market makers were left holding substantial inventory of short puts, necessitating hedging by shorting Gamestop shares. By virtue of market impact, this should fairly immediately cause some reduction in stock price, or if during a rally upwards, help brake momentum.
However, it’s also important to remember what happens when an excess of puts exist. As SqueezeMetrics has noted in the past, one of the major drivers of upward momentum from options tends to be the purchase and expiry of put options, given the predictable cycle of hedging (initial shorting) and repurchase of hedges (rebuying).
This is especially salient for the January behavior of Gamestop in two regards:
1) During the rally, implied volatility exploded. As I mentioned in my last post, there is a tenuous but long-term link between implied volatility and realized volatility — we expect increases in realized volatility to spike implied volatility, and for the increase to be reflected in IV long after the realized volatility dissipates. However, what’s more interesting to note here is the volatility of implied volatility, or what we usually call vol of vol. While it was not mentioned in the report itself, any trader scanning the chains during meme squeezes could observe the jaggedness of vol of vol, which reacted to sharp changes in realized volatility. This was especially important in the early phases of the rally (up until January 22), where implied volatility was clearly not pricing correctly the implications of the reddit revolt.
2) As mentioned in the report, the primary options traded were of the nearest expiry, especially for call options.
In the earliest phases of the rally, many on Volatility Twitter dryly noted the retail trader’s first taste of vega, as during the squeeze tiny ($10-15 strike) put options increased in value due to volatility expansion, even as the spot moved further and further away from their strike price. This prolonged elevation in implied volatility kept option prices levitating from the real world for egregious amounts of time; until the last day of expiration (or even last hours) it’s hard to argue they were fairly priced or tied to the real world outcomes.
However, this is the important part. Because those put options still required hedging (shorting in the naive case), the exponential decay of the option in the last days and hours would’ve contributed to upward momentum, as option sellers would have to buy back the hedged delta (given nearly all puts would’ve expired worthless). This was remarkably consistent with the 0-day behavior observed on January 22.
As the day began, the upward momentum was initially marginal, hampered by a week of mean reversionary action that saw Gamestop shares bounce between $30 and $40 (after a gap up on that Monday). However, as the hours quickened and expiry drew closer, we could see substantial escalation of price momentum, which reached a frenzied pace. Again — there is no way for me to determine especially retroactively the contribution of put option hedges being bought back, or even the impact of 0-day call options traded (which, to my recollection, was an obscene amount). However, it is quite in line with the SEC report, given January 22nd was a major options expiration for GME, and the trouble was just beginning.
If truly retail was buying put options, undoubtably the action of expiry would’ve helped propel Gamestop upwards. This would be compounded by the interest of retail momentum traders, who similarly would pile onto the observed upward momentum (in fact, arguably we still see this on mini-rallies in Gamestop and AMC).
In good faith, I have to agree with certain assessments the SEC made. It is not outside the realm of reason to argue that the unique factors and media attention given to the situation led to primarily retail buying, and a squeeze of historic proportions developed. However, I do think that the SEC should have explored the implications of options structure a bit more in depth, and I would welcome any follow up or release of data.
Best,
Lily