An Academic Critique of the SEC’s GameStop Report

On October 18, 2021, SEC staff released a long-awaited report on equity and options trading in connection with the meteoritic rise of GameStop’s share price in January 2021.  The staff report addressed several issues surrounding the GameStop episode, but one of the most widely reported was the conclusion that neither a short squeeze nor a gamma squeeze caused the increase in GameStop’s share price.  Rather, staff concluded, “it was the positive sentiment, not the buying-to-cover, that sustained the weeks-long price appreciation of GameStop stock.”[1]

As members of the Ad Hoc Academic Committee on Equity and Options Market Structure Conditions in Early 2021, we issued our own report earlier this year, and it identifies additional data suggesting that GameStop’s shares may have been subject to a short squeeze and gamma squeeze.  First, by extending the sample and considering securities lending transactions, we find evidence that a nontrivial fraction of the trading volume in GameStop’s stock consisted of purchases by short sellers covering their positions.  Second, we show that SEC staff’s analysis of a gamma squeeze can be extended to include delta-hedging by put-and-call options market makers to better approximate the contribution of options positions to the increase in GameStop’s share price.

Short Squeeze Analysis

Footnote 78 of the staff report states:

We identify traders with large short positions by first calculating traders’ average inventory positions as of January 15, 2021, and isolating the Firm Designated IDs (“FDIDs”) with an average negative position, excluding market makers and high frequency traders (i.e., identified as traders that offset their trades within a day). We then isolate the FDIDs with negative inventories below (i.e., more negative than) the median as our sample of heavily shorted traders. We then identify the buy trades initiated by these FDIDs over the next two weeks (January 19 – February 5). Note that since the CAT sample only begins on December 24, 2020, we are not able to include FDIDs’ inventory positions accumulated prior to this date.[2]

First, staff‘s use of the Firm Designated ID (FDID) is a problematic way to identify unique customers.  CAT NMS LLC, the entity responsible for developing the consolidated audit trail (“CAT”) system, makes clear that the FDID designation represents trading accounts; however, it also states: “A customer can have more than one trading account, and a trading account can have more than one customer.”[3]  The guide is clear that the FDID is broker-specific and need only remain constant across vendors used by that broker.[4]  Staff’s use of FDIDs is particularly odd because the CAT specification specifically includes a customer designator, the CCID, the collection of which is provided by Regulation NMS Rule 613.

Second, by limiting the sample of “heavily shorted traders” to those below the median, staff potentially omits a large volume of cover purchases by market participants who had a net short position in GameStop’s shares in a volume above the median.

Finally, staff’s inability to include data prior to December 24 when calculating inventory positions is likely to materially distort its conclusions.  Rather than utilizing a “stock” measure of each market participant’s short position as of January 15, staff is apparently inferring that short position by calculating the net trading “flows” from December 24 to January 14.  Staff’s justification for this approach is the availability of data from the CAT, which begins on December 24.  While that makes sense from an operational standpoint, December 24 is a relatively late date to begin a calculation of inventory positions based on trading flows.  A consequence of the staff’s sampling procedure is that, per this methodology, there were zero short positions on December 23, 2020.  But as we describe in our report, there were anecdotal reports that short sellers had opened large short positions earlier in December, and publicly available data on securities lending positions are consistent with those reports.[5]

Gamma Squeeze Analysis

SEC staff examined the possibility that GameStop’s shares were subject to a gamma squeeze.  On this point, staff wrote that the “increase in options trading volume was mostly driven by an increase in the buying of put, rather than call, options. Further, data show that market-makers were buying, rather than writing, call options. These observations by themselves are not consistent with a gamma squeeze.”[6]

As a preliminary matter, there are internal inconsistencies in the staff report on this question.  More fundamentally, there are three ways to induce market makers to hedge options exposure by purchasing shares of the underlying security. First, one may purchase put options from a market marker prior to an increase in the underlying share price.  The market maker will initially hedge its exposure by shorting shares of the underlying security.  As the share price increases, the delta of the put options increases (in a signed sense), ordinarily leading the market maker to purchase shares.  Second, one may sell put options to a market maker (i.e., open a short position in put options).  The sale of put options to a market maker will ordinarily induce that market maker to hedge its exposure to those put options by purchasing the underlying security.  And hedging induced by market participants buying call options are a third way that options transactions can apply upward pressure on the share price.

What matters is not the type of option at issue – put or call – but rather the direction of the hedging induced by the (a) initial transaction and (b) subsequent share price movement.  This is a direct consequence of put-call parity, which holds that the value of a put option is mathematically equivalent to the value of a call option, after adding the strike price and subtracting the future price of the security. A consequence of put-call parity is that the hedging activity induced by an options position does not depend on the type of option employed (put or call), as staff incorrectly implies, but rather by the direction of the initial transaction and any subsequent share price change.

We find that, even at relatively high levels of assumed netting of hedging order flow, it appears that a large volume of trading volume in GameStop’s shares consisted of options hedging. This evidence calls into question the basis for the conclusion that “staff did not find evidence of a gamma squeeze in GME during January 2021.”

To be sure, without access to the sort of nonpublic, deanonymized data available to the SEC and its staff, we are unable to pin down the magnitude of a short squeeze or gamma squeeze in GameStop’s shares, much less identify the actors involved in or responsible for such trading.  We hope that our analysis can helpfully guide the commission and its staff in identifying additional data that might shed light on the extent to which the increase in GameStop’s share price may have been the product of a short squeeze and gamma squeeze.  We are ready and willing to provide any assistance to the commission and staff in that effort.

ENDNOTES

[1] Sec. & Exch. Comm’n, Staff Report on Equity and Options Market Structure Conditions in Early 2021, at *26, https://www.sec.gov/files/staff-report-equity-options-market-struction-conditions-early-2021.pdf .

[2] Id. at 28 n.78.

[3] https://catnmsplan.com/sites/default/files/2020-01/FDID-Guidance-April-2019.pdf

[4] Id. at 6.

[5] See supra Section I.

[6] Id. at *29.

This post comes to us from Joshua Mitts at Columbia Law School, Robert H. Battalio at the University of Notre Dame, Jonathan Brogaard at the University of Utah’s David Eccles School of Business, Matthew D. Cain at the University of California’s Berkeley Center for Law and Business, Lawrence R. Glosten at Columbia Business School, and Brent Kochuba at SpotGamma. It is based on their publication, “A Report by the Ad Hoc Academic Committee on Equity and Options Market Structure Conditions in Early 2021,” available here.