Page 1 of 14
Archives of Business Research – Vol. 9, No.2
Publication Date: February 25, 2021
DOI: 10.14738/abr.92.9768.
Samsa, G. (2021). Bonnie and Clyde Revisited: A Conceptual Model of the Behavior of the Instigators of the GameStop Short Squeeze of
2021. Archives of Business Research, 9(2). 243-256.
Bonnie and Clyde Revisited: A Conceptual Model of the
Behavior of the Instigators of the GameStop Short Squeeze of
2021
Greg Samsa Ph.D
Department of Biostatistics and Bioinformatics, Duke University
ABSTRACT
The efficient market hypothesis assumes that investors act
independently, but social-media-enabled behavior among
amateur investors demonstrably violates this assumption. This
was dramatically illustrated by a recent short squeeze in the
shares of GameStop. By analogy with conceptual models in the
behavioral sciences, we outline a conceptual model attempting
to explain the behavior of the participants, and explore some of
the implications of this conceptual model for investors. This
case study illustrates a systematic inefficiency in option pricing
which only become apparent when extreme market conditions
are encountered. Although the cognitive biases of amateur
investors are of little importance for understanding stock
prices, this is a special case where their cognitive biases do
matter.
Key words; efficient market hypothesis; independence; social media
INTRODUCTION
27 January 2021 was a bad day not only for certain hedge fund managers, but for the
efficient market hypothesis as well. One of the underpinnings of this hypothesis is not only
that stock prices reflect all available information, but that investors act independently [1].
Rather than acting independently, a group of amateur investors (a.k.a. "the Reddit Army",
a.k.a. "the rebels") attempted to execute what was in effect a social-media-enabled populist
rebellion. The most publicized target was GameStop (GME), and their mission was to
execute a massive short squeeze. (A short squeeze occurs when a stock jumps sharply
higher, forcing traders who had bet against it by borrowing the stock, selling it, waiting for
the price to drop, and then planning to buy it back at a profit, to buy it at an inflated price in
order to forestall even greater losses. [2]) "Shorts" (i.e., traders who have bet against GME)
are in a particularly precarious position when they have done so with borrowed money,
because the value of their collateral drops, and in the worst case are required by their
creditors to immediately liquidate their holdings at disadvantageous prices. Various hedge
funds had done precisely that. The revolutionaries had pushed the stock price from a 52-
week low of approximately $3 to more than $450 and the hedge funds in question had lost
billions of dollars. The stated intention of the rebels (e.g., on WallStreetBets (WSB), a
Page 2 of 14
URL: http://dx.doi.org/10.14738/abr.9768. 244
Samsa, G. (2021). Bonnie and Clyde Revisited: A Conceptual Model of the Behavior of the Instigators of the GameStop Short Squeeze of 2021.
Archives of Business Research, 9(2). 243-256.
decentralized Reddit forum [3]) was to bankrupt these hedge funds, arguing that are were
a plaything of rich elites, serve no legitimate economic purpose, and moreover are
dangerous -- for example, having already played a significant role in nearly bankrupting the
global financial system in 2008. These would-be redistributors of wealth styled themselves
along the lines of the American bank robbers and folk heroes Bonny and Clyde, although it
must be acknowledged that this latter story ended in a hail of bullets.
My modest exposure to this revolution consisted of receiving an automated email about
unusual activity in Nokia (NOK), a company in which I had invested. A headline from a
financial site later read "Nokia surges to record gain on record volume, for no apparent
reason" [4]. In approximate numbers, when logging on to my broker it transpired that the
price of NOK had suddenly risen from $5.00 to $7.50, and a 1/29/21 $10 call option could
be sold for $1.50. The reader is invited to reflect on this for a moment. I already owned the
stock, and so writing a covered call option involved no risk of disaster. (The same could not
be said for those speculators swarming to buy GME stock at extraordinarily inflated prices.)
The option purchaser was offering to pay $1.50 per share, a 20% return on the current
stock price, in exchange for me ceding all gains above $10, 33% above the current stock
price, with this option expiring IN THREE DAYS. Thinking that the financial world had
temporarily gone insane, I placed a limit order to sell this option for $1.40, assuming that
the volume of trading might induce a delay and wanting to increase the chance that the
order would be filled -- which it was, at $1.70, a testament to the feeding frenzy which was
playing out in real time. Paying homage to a trader's adage about the appropriate response
to unexpected good fortune, before returning to my day job I briefly opened another
website, took a small part of the profits from the option premium, and bought a sweater.
This is a preliminary effort to apply the principles of conceptual modeling used within the
behavioral sciences to explain the behavior of the rebels, and also to predict features of
their subsequent behavior. Our hypothesis is that such a conceptual model can be
developed. Our assessment criterion is whether or not the model generates testable
predictions of how this behavior will be reflected in stock prices. Ideally, at least one of
these predictions will be unique, in the sense that it could not be directly derived from
existing models of investor behavior.
CONCEPTUAL MODELS OF HUMAN BEHAVIOR
The behavioral sciences utilize numerous conceptual models (sometimes called theoretical
models, analytic frameworks, etc.) of human behavior. For example, the Transtheoretical
Model of behavior change, developed by Prochaska and DiClemente [5], posits that
behavioral change typically proceeds sequentially through the various stages of
precontemplation, contemplation, preparation, action, maintenance, and termination. This
model provides an intellectual structure which can be used to explain observed behavior,
and also to motivate the development of interventions to alter that behavior. For example,
the Transtheoretical Model explains why a smoking cessation intervention which has
physicians explain to all their patients how to quit smoking is destined to fail for those
patients are in the precontemplation stage rather than the action stage. One possible
intervention motivated by the Transtheoretical Model would be to first triage patients
based on their stage of change and then, for example, providing those in the
Page 3 of 14
245
Archives of Business Research (ABR) Vol 9, Issue 2, February-2021
precontemplation stage with information about the benefits of quitting, providing those in
the contemplation stage with information about how quitting can be accomplished,
providing those in the action stage with psychological support and problem solving around
barriers to their current attempt to quit, etc.
A characteristic of behavioral science models in general and the Transtheoretical Model in
particular is that they attempt to describe why and how people will act. Indeed, the "how"
is what turns these models into scientifically testable constructs. For example, although no
single study will prove or disprove the Transtheoretical Model, studies can test
interventions (such as the above) which are based on its principles. To the degree that
these interventions are effective (or not), their underlying conceptual model is supported
(or not).
CONCEPTUAL MODELS OF INVESTOR BEHAVIOR
The two main models of investor behavior can be termed "rational" and "behavioral". The
rational model assumes that investors dispassionately assess the financial impacts of their
decisions and act in order to maximize their expected utility. Moreover, it is often also
assumed that markets consist of large numbers of investors, that these investors act
independently, that their actions affect prices in aggregate but not as individuals, etc. --
from which set of assumptions the intellectual infrastructure of quantitative finance,
including the efficient market hypothesis, is built.
The behavioralist model instead asserts that people in general and investors in particular
don't always behave in an entirely rational fashion. A core idea underpinning behavioral
finance is that the way we think, which was honed by evolution, uses various heuristics
("cognitive biases") which are inconsistent with the principles of economics. These
cognitive biases include excessive reliance on recent information, extrapolation of trends
into the future, herd-following behavior, and a tendency to gamble, among others. The
behavioral finance literature demonstrates what we all recognize in ourselves: namely, that
our thought processes aren't entirely rational, in every situation, at all times [6].
As with other models of human behavior, what is ultimately important is that both the
rationalist and behavioralist models can be used to generate testable predictions about
how investors will act, which actions will then provide these models with a greater or
lesser degree of empirical support. A classical test of the rationalist model runs along the
lines of:
"the efficient market hypothesis (derived from the rationalist model)
predicts that investor behavior will not exhibit exploitable
patterns, whereas behavioral finance can be used to generate a
prediction about an exploitable pattern (e.g., excessive trend
following), and that a specific investment strategy derived from
this prediction (e.g., some form of a momentum- based strategy) will
lead to superior returns, after risk adjustment".
Page 4 of 14
URL: http://dx.doi.org/10.14738/abr.9768. 246
Samsa, G. (2021). Bonnie and Clyde Revisited: A Conceptual Model of the Behavior of the Instigators of the GameStop Short Squeeze of 2021.
Archives of Business Research, 9(2). 243-256.
The extensive literature on this topic more or less suggests a stand-off: some behaviorally- based strategies appear to lead to outperformance, but its magnitude is modest and the
debates about what constitutes proper risk adjustment are intense. Moreover, when
behaviorally-based strategies lead to superior returns this tends to be in the short-to- intermediate term (e.g., no more than 3-5 years), whereas in the long run rationality tends
to prevail [7].
A third model considers the incentives of key players in financial markets [6], and is
discussed below.
STRUCTURAL AND INCENTIVE-BASED MODELS OF INVESTOR BEHAVIOR
Accepting the premise that markets are more efficient in the long-term than the short-term,
it is natural to ask why. We believe that much of the answer to "why" lies within the
structure of the market, including the incentives of its participants. A key to understanding
the structural factors in play is to ask "whose actions serve to set stock prices?". To provide
a realistic answer, the assumption that "prices are set by the aggregate and independent
impact of numerous small investors, none of which have a large enough market share to
affect prices as individuals" should be replaced with "for any particular time period, prices
are set by the largest entities who trade regularly". In other words, if an individual investor
trades in volume X and an institutional investor trades in volume 100,000 X, the latter will
effectively set prices whereas the former will take them.
Following this line of reasoning, the largest market participants (e.g., pension funds and the
like) are like elephants in a bathtub -- their size precludes them from short-term trading
(e.g., because of the slippage in prices which trading large numbers of shares would
induce), which essentially forces them to try to slowly accumulate shares of businesses
with sound long-term prospects at fair prices and then hold them over time. In other
words, their size (and fiduciary responsibility to invest conservatively) require them to act
"rationally", in the economic sense of the term. However, because of the constraints on
their short-term trading, their actions only serve to set "rational" prices in the long-term.
Now considering the short-term (e.g., hours, days, weeks, months), the largest entities who
trade regularly are hedge funds, growth-based mutual funds, momentum-based mutual
funds, institutional day traders, micro-level arbitrageurs and similar entities whose actions,
to a greater or lesser degree, involve speculation. The answer to "why do they speculate?"
is primarily that their payment structures provide them with incentives to do so, to act in
ways which are contrary to long-term value, and essentially to act as if they are driven by
some of the biases catalogued by behavioral finance. Among these actors, the managers of
hedge funds are influenced by a particularly perverse set of incentives. For example, their
compensation depends on short-term performance, which can be enhanced by using
borrowed money to fund financially risky adventures, which in turn lead to Croesus-like
bonuses when successful. When things go poorly they neither return their previous
bonuses nor go personally bankrupt, and if things blow up in a big way the taxpayers can
be counted on to clean up the mess. By being insulated from the losses associated with
doing so, they are incentivized to gamble. (The question of why these payment structures
exist is outside the present scope.)
Page 5 of 14
247
Archives of Business Research (ABR) Vol 9, Issue 2, February-2021
Following this line of reasoning, using an incentive-based model, the classical test of the
efficient market hypothesis could be more realistically modified to be:
"the efficient market hypothesis predicts that investor behavior will
not exhibit exploitable patterns, whereas the incentives of hedge
fund managers and other speculators who effectively set prices in
the short term can be used to generate a prediction about an
exploitable pattern, and that a specific investment strategy derived
from this prediction will lead to superior returns, after risk
adjustment".
Unit of Analysis
According to the tenets of behavioral finance, the rationalist model of investor behavior is
"normative", in that it describes how individual investors should behave, whereas the
behavioral model is "positive", in that it describes how individual investors actually do [8].
Certainly, the rationalist model is normative in the sense that it begins with a set of
assumptions which are hoped to be approximately true but are primarily selected because
of notions of how an idealized "perfect market" should operate.
A curious aspect of the debate between the rationalists and the behavioralists can be
illustrated by returning to the Transtheoretical model one final time, and noticing that its
unit of analysis is the individual -- in other words, it attempts to describe how individuals
(and not groups) behave. The original assumptions of the rationalists (i.e., that individuals
are rational utility maximizers) are stated at the level of the individual. The same is true
for behavioral finance. However, the classical test of the rationalist model is not executed
at the level of the individual investor -- instead, it asks whether or not stock prices follow a
predictable pattern.
An answer to "why" the test is performed in this fashion is that the rationalists came first,
and their primary interest was not in the behavior of individuals, but instead was in the
behavior of markets -- especially, how closely the stock market approximates an idealized
perfect market. Behavioral finance, among others, could be used to generate theoretically
supported predictions about exploitable patterns in stock prices, and thus strengthen how
the efficient market hypothesis is tested. Indeed, the rationalists were perfectly happy to
concede the point about cognitive biases, and simply revised their assumptions about
markets to be (loosely speaking) that even though individual market participants might not
be rational, groups are (i.e., as reflected in stock prices).
In contradistinction to the rationalists, how individual investors behave is critical to
behavioral finance. Indeed, an ideal test of hypotheses derived from behavioral finance
would be performed at the level of the individual investor. Thus, for example, the question
of whether stock prices exhibit exploitable patterns derived from cognitive biases -- which
effectively pertains to groups -- should be relevant to behavioral finance only if an
additional assumption is made: namely, that individual biases are also exhibited by groups
of investors. On the surface, making this extension seems innocuous and intuitive -- for
Page 6 of 14
URL: http://dx.doi.org/10.14738/abr.9768. 248
Samsa, G. (2021). Bonnie and Clyde Revisited: A Conceptual Model of the Behavior of the Instigators of the GameStop Short Squeeze of 2021.
Archives of Business Research, 9(2). 243-256.
example, if individual investors engage in herd-following behavior shouldn't groups of
investors do the same?
The incentive model provides a different perspective on these questions. By recognizing
that stock prices are set by the actions of a few individuals (e.g., pension fund managers
and the like in the long-term; hedge fund managers and the like in the short term), the
cognitive biases (or lack thereof) of most individual investors can be ignored. Moreover,
hypotheses about the behavior of stock prices effectively becomes hypotheses about those
few individuals whose actions do set prices. In other words, the unit of analysis is
effectively changed from the group to the individual. The second statement of the classical
test of the efficient market hypothesis reflects this.
The cognitive biases demonstrated by hedge fund managers and similar large short-term
speculators include gambling, herd-following, recency bias, and miscalibration of risk,
among others. We argue that large speculators are sophisticated and well-informed
individuals, are familiar with the literature on behavioral finance, could avoid these biases
if they wish, and instead intentionally choose to manifest them because they are
incentivized to do so. Indeed, if the choice in framing a hypothesis is between "because of
their evolution-derived cognitive biases hedge fund managers cannot prevent themselves
from gambling" and "hedge fund managers choose to gamble because they are incentivized
to do so", we believe that the latter version is positive and the former normative. Upton
Sinclair made this point quite succinctly: "it is difficult to get a man to understand
something, when his salary depends upon his not understanding it" [9].
Regarding the unit of analysis, the rebels provide a special case. In the case of GME, by
acting in concert and in large numbers, their impact was similar to that of any other large
speculator. In other words, for our purpose they can be treated as a single individual, one
whose actions can set prices, and thus whose beliefs, cognitive biases and incentives
matter.
The Task
Operationally, we propose to restate the classical test of the efficient market hypothesis,
replacing "hedge fund managers" with "rebels":
"the efficient market hypothesis predicts that investor behavior will
not exhibit exploitable patterns, whereas the beliefs and other
characteristics of the rebels can be used to generate a prediction
about an exploitable pattern, and that a specific investment strategy
derived from this prediction will lead to superior returns, after risk
adjustment".
In order to do so, we will first attempt to outline a conceptual model of the behavior of the
populist rebels. Analogous to the conceptual model of the behavior of hedge fund
managers, our intention is to identify inconsistencies between their behavior and economic
reality. In turn, these inconsistencies should give rise to investment strategies which
attempt to obtain superior risk-adjusted returns, at least until the hail of bullets begins (or
Page 7 of 14
249
Archives of Business Research (ABR) Vol 9, Issue 2, February-2021
short selling is better regulated or eliminated) and the opportunity is lost. These strategies
are intended to be more systematic than the experience of just happening to hold a stock
which becomes the focus of a Reddit mob, at which point the obvious "strategy" is to simply
take the money and run (i.e., by either profitably selling the stock or profitably selling a
covered call option, as illustrated above).
Conceptual Model: Structural Inefficiencies
The proposed conceptual model contains two elements: (1) a description of the structural
market inefficiencies that the rebels exploit (or cause); and (2) a description of the typical
behavior of the rebels (e.g., which stocks they tend to purchase).
Most fundamentally, the market inefficiencies in question derive from the observation that
financial markets respond poorly to extreme conditions. One element of poor response to
extreme conditions pertains to operational aspects -- for example, in the usual run of
events the execution of trades is effectively instantaneous with accurate information about
bid and asked prices, whereas when trading volumes are extremely heavy execution is
delayed and the stated bid and asked prices can be inaccurate. In the worst case, trading is
halted, either by intention or by necessity because systems crash.
Another element of poor response pertains to the nature of short sales. In the usual run of
events, even though the potential losses associated with a short sale are theoretically
infinite, a short sale is not fundamentally different from purchasing a put option -- it is a bet
against the stock which might or might not turn out well. The extreme condition occurs
when the volume of shares which can be used to replace the shares which were borrowed
dries up, at which point the localized dynamics of supply and demand can cause the stock
price to rise significantly and a short squeeze takes place. Indeed, short squeezes (usually
of smaller magnitudes) do regularly occur, with institutional investors being on both sides
of the trade, and no objections are raised (except by taxpayers, when the short squeeze
threatens to collapse the global financial system and they are required to foot the bill,
although these objections are quickly forgotten). Apparently, the financial establishment
believes that the above short squeezes are allowable, whereas it is something
fundamentally different when a group of amateur investors band together to stick it to the
man.
Yet another element of poor response pertains to economic models of stock behavior,
which typically include assumptions about smoothness and continuity. As an example, the
unreasonably high price associated with the short-term call options on NOK was likely
generated by a mathematical model of implied volatility, which in turn was short-circuited
by the massive short-term movement in the price of the underlying stock. Such
mathematical models generally work quite well, but not necessarily when encountering
stock price behavior which is discontinuous and more extreme than their assumptions are
intended to accommodate.
More specifically, because option pricing models are based on classical financial theory,
they embed its premises, among which are that stock prices reflect all available
information and that individual investors act independently. Thus, when a stock such as
Page 8 of 14
URL: http://dx.doi.org/10.14738/abr.9768. 250
Samsa, G. (2021). Bonnie and Clyde Revisited: A Conceptual Model of the Behavior of the Instigators of the GameStop Short Squeeze of 2021.
Archives of Business Research, 9(2). 243-256.
NOK has a dramatic spike in volatility, these models effectively assume that this volatility
must be a response to a real and significant new signal about corporate prospects -- for
example, the possibility of a merger or other major event. Assuming that such a signal is
real, then major movements in stock prices are anticipated, and large option premia are
justified. Here, however, no such signal existed, and the jump in the stock price was an
artifact of investor behavior which was coordinated rather than independent, and so the
options in question were profoundly over-priced.
Conceptual Model: Investor Behavior
The typical behavior of the rebels can be classified as culturally driven, and manifested by
both "sociopolitical trades" and "meme-based trades". The attempted short squeeze on
GME is an example of a sociopolitical trade -- apart from hoping to make a profit, the
avowed purpose was also to apply the principles of cancel culture to hedge funds. Indeed,
the behavior of hedge fund managers highlights the distinction between an observation
and a value judgment: a "value judgment" describes this entire system as corrupt, whereas
an "observation" simply provides a description. When considering the hedge funds which
were short of large amounts of GME stock, by deciding not to take profits but instead to
intensify the short squeeze, some of the rebels in effect made a value judgment (and also a
financial decision which was at best dangerous and at worst foolhardy).
The meme-based trades involve speculation in the bright shiny objects which typically give
rise to bubbles. Current examples include cannabis, cryptocurrency, and the like. Many of
these memes are consistent with a particular world view -- for example, cryptocurrency is
framed as a potential alternative to traditional currencies untethered from the influence of
central bankers, and the rationale for cannabis should be self-evident.
Another distinguishing characteristic of the rebels is their lack of sensitivity to
considerations of price and value. If Reddit posts are to be believed, to the rebels this
almost appears to be a point of honor. "To the moon or to zero", or a scatological
equivalent of the same, is their rallying cry.
Implications of the Conceptual Model
Both of the above types of trades can give rise to bubbles: a short squeeze is a temporary
bubble, whereas meme-based behavior leads to bubbles which are longer lasting. Indeed,
the rebels' meme-based bubbles should in many respects be similar to the typical stock
market bubble [10] such as the dot-com frenzy (which was also meme-based), with the
exception that the buying patterns would be better coordinated, likely inducing even more
volatility than usual. Because of their lack of sensitivity to price and value, their actions
should also induce price momentum in the short-to-medium term -- for example, by
causing meme-based bubble prices to exceed corporate valuations by even more than is
usually the case for bubbles.
A final implication pertains to the internal dynamics of the meme-based bubbles that the
rebels create (or intensify). In a typical bubble, once their prices become sufficiently
inflated the stocks in question become "priced to perfection", and any negative news about
that company can cause its stock price to fall precipitously. In large part this is caused by
Page 9 of 14
251
Archives of Business Research (ABR) Vol 9, Issue 2, February-2021
the recognition, on the part of institutional investors, that they are engaged in a speculative
exercise, the stock in question is over-priced, and it is wise to already have one foot out the
door. As a bubble inflates to increasingly unsustainable levels, less speculative
stockholders are replaced with increasingly speculative ones, with this final group of
speculators attempting to obtain massive returns in the final, parabolic phase of the bubble
by becoming the second-to-last person to buy the stock before the inevitable crash. (The
magnitude of the potential returns is the incentive for this behavior.) It goes without
saying that amateur investors should not attempt to do so, as this one aspect of speculation
for which they are at a significant disadvantage.
Within this dynamic, to the degree that the rebels are attracted to an idea rather than an
individual company, the impact of confirmation bias will tend to discount company-specific
negative news and cause the group in question -- for example, cannabis stocks -- to rise
together. In other words, the actions of the rebels might cause individual companies to
drop out of a meme-based bubble more slowly than is usually the case.
In any bubble there is always a last buyer before the crash. In the case of GME, some of the
rebels had all the usual characteristics of the last buyer-- they lacked sensitivity to price
and value; they lacked experience about investment; they were attracted by the possibility
of quick profits; they observed a herd of people doing something and followed suit; and
they believed the meme. To this toxic brew they added a new reason for becoming the last
buyer: namely, a sociopolitical cause. Is it any wonder how this story ends for them?
DISCUSSION
While a cursory perusal of their websites suggests humorously sociopathic tendencies, the
rebels are also engaged in a principled (if quixotic) social-media-enabled critique of
modern financial capitalism. We argue that much of their behavior, at least to the degree
that their stock-related posts can be believed, can be partially explained by a simple and
culturally-driven conceptual model. This model does not apply generally: instead, its scope
is limited to those relatively few stocks for which the rebels trade in sufficient numbers to
affect prices. The conceptual model not only provides general predictions about investor
behavior but suggests at least one unique one -- namely, that the dynamics of rebel-induced
bubbles will be subtly different from the dynamics of a more typical bubble -- and thus we
conclude that the answer to whether or not our assessment criterion has been satisfied is
"yes". We also argue that the rebel's behavior has also uncovered systematic inefficiencies
in option pricing which only become apparent when extreme conditions are encountered.
Finally, we argue that although the cognitive biases of amateur investors are usually of
little importance for understanding stock prices, the rebels are an exception.
The notion that, during a pandemic-induced lockdown which has served to exacerbate
various social tensions, a group of amateur investors could outsmart the "financial elites"
seemed to capture the popular imagination and become a meme of its own. Indeed, the
rebels were even indirectly featured in a Super Bowl commercial by Reddit which began
with "Wow, this actually worked", and included "One thing we learned from our
communities last week was that underdogs can accomplish just about anything when they
Page 10 of 14
URL: http://dx.doi.org/10.14738/abr.9768. 252
Samsa, G. (2021). Bonnie and Clyde Revisited: A Conceptual Model of the Behavior of the Instigators of the GameStop Short Squeeze of 2021.
Archives of Business Research, 9(2). 243-256.
come together around a common idea. [11]" Whether it actually did work in the end is
discussed in the postscript.
Recognizing that our interest is in investor behavior rather than social commentary, and
accepting for the sake of argument the above description of the rebel's behavior, a natural
question becomes whether investment strategies (or, perhaps more accurately: speculation
strategies) can be developed which can intentionally exploit the economically unsound
elements of the rebel's behavior, and do so with sufficient safety to produce superior
returns, even after adjusting for the increased risk. One such strategy was illustrated in the
introduction: namely, "in the midst of a short squeeze, simply take the money and run". In
other words, if an investor happens to be holding a stock which the rebels cause to go into a
short-squeeze-induced temporary bubble, they should take short-term profits -- whether
directly, by selling the stock, or indirectly, by selling expensive covered call options. Such a
strategy is self-evidently profitable, although not intentional, since it relies on having
previously purchased the stock for unrelated reasons.
Once a short squeeze of the magnitude of GME advances to a certain point, no sufficiently
safe method of taking advantage is apparent. Certainly, buying GME at $400 and hoping it
will rise to $1,000 crosses the line between speculation and gambling and, indeed,
represents gambling with very poor odds. This is not even to mention that the rebels are
customers, whereas the capitalists in question are owners -- for example, they own the
banks, they own the brokerage houses, they control various points of leverage allowing
them to halt trading and otherwise manipulate the system, they have information about
supply and demand through order flow, they are in a position to affect legal and regulatory
decisions, they can monitor and infiltrate websites, etc., not even to mention that GME was
never worth anything near $400 per share. Competing with professional speculators in the
arena of speculation has an obvious downside. In the end Bonnie and Clyde were
outgunned, too, not even to mention getting some of their gang members killed along the
way in addition to themselves.
Presumably, identifying reasonably-valued stocks to purchase at the beginning of a short- squeeze would be profitable on a risk-adjusted basis. NOK might plausibly fall into this
category. For example, NOK closed near $5 on 1/26/21 and, despite rising to nearly $10 on
1/27/21, had dropped back to $5 by the morning of 1/29/21. Without digressing into an
assessment of its corporate performance, if an investor believed that NOK was reasonably
priced on 1/16/21 they would likely believe the same on 1/29/21. Indeed, NOK might be
an even better value, because an at-the-money 2/5/21 call option could be sold for an
approximate 10% premium.
In theory, trying to guess the next candidate for a short squeeze would be a profitable
strategy. Likely candidates could be found in lists of most heavily shorted stocks and from
perusal of relevant blogs. Most extensively-shorted stocks are shorted for good reasons,
and only a small subset of such stocks would provide adequate fundamental value. Also,
even though the market might not be perfectly efficient participants do learn very quickly,
institutions are likely already performing due diligence to search for such candidates, and
prices of the stocks in question are likely to rapidly adjust in response. Moreover, hedge
Page 11 of 14
253
Archives of Business Research (ABR) Vol 9, Issue 2, February-2021
funds might reduce their level of exposure to shorting, or perhaps execute shorting in a
way which makes proletarian short squeezes more difficult to accomplish, or perhaps
lobby to change the rules in a way that hamstrings the rebels. Of course, the behavior of
the rebels can evolve as well.
Even if their antagonists can successfully prevent future short squeezes, our conceptual
model suggests that the rebels will continue to engage in meme-based trades. In this case,
the same general principles surrounding bubble-based behavior would likely apply. In
particular, once a bubble has inflated to a certain point no sufficiently safe method of
directly taking advantage of price momentum is apparent. At the early stages, though,
writing covered calls likely provides superior risk-adjusted returns, the rationale for doing
so being two-fold: (a) option premia are large, reflecting the extra volatility in prices which
the coordinated action of the rebels induces; and (b) because of the confirmation bias
induced by their meme-based thinking, the rebels can (for speculative purposes) be
counted on to buy any dips in price. Presumably, someone choosing to speculate on
writing covered calls on meme-based stocks should choose very short-dated calls, because
(a) the premium per unit time period is highest; and (b) there is much to be said for exiting
speculative trades as soon as possible.
As an illustration of how a meme-based approach might work in practice, Aurora Cannabis
(ACB) has one of the largest market capitalizations within the cannabis sector, and closed
at $18.92 on 10 February 2021. The actual value of ACB is pure guesswork, even to an
order of magnitude. The stock price exceeded $140 in 2018 during a previous bubble.
During the previous 52 weeks, it ranged from $4 to $21, approximately. None of the dozen- plus analysts who cover the company rate ACB stock as a buy. It would be reasonable to
anticipate that ACB will be a laggard within the cannabis sector: in other words, that it will
move similarly but belatedly. During the previous day, a headline from a financial site read
"Tilray (TLRY) shares soar 39% as Reddit message board sets sights on cannabis stocks".
Indeed, during the week ACB also rose from $12.84 to $18.92, which is notable but still
underperformed TLRY on a relative basis. On 10 February 2021 ACB rose by $3.31, on an
above-average volume of over 83 million shares. Quite likely, trading volume of this
magnitude suggests not action by the rebels, who appeared to be primarily focused on
TLRY, but instead illustrates the phenomenon mentioned above: namely, of institutions
performing due diligence to predict meme stocks and bidding up their prices beforehand.
Continuing this illustration, Table 1 below illustrates the impact of short-term volatility on
short-term option premia. The table reports bid prices (i.e., a worse-case scenario for the
option seller, although a competent broker can often obtain a better price) for 12 February
2021 options -- that is, options which come due in two days. A similar phenomenon is
observed as the NOK example in the introduction -- not as extreme, but still noteworthy.
Although trading strategies which are based on this ensemble of option prices should be
obvious, our interest is more academic, and asks the question of whether such option
prices are so high as to represent an inefficiency in the option market. In the purely
mathematical sense, the answer is probably "no" -- the prices in question are based on
expected volatility during the next two days and the recent trading history suggests that
Page 12 of 14
URL: http://dx.doi.org/10.14738/abr.9768. 254
Samsa, G. (2021). Bonnie and Clyde Revisited: A Conceptual Model of the Behavior of the Instigators of the GameStop Short Squeeze of 2021.
Archives of Business Research, 9(2). 243-256.
this volatility could be extraordinarily high. (Indeed, the closing price of ACB on 12
February 2021 was $12.48, sufficiently far from $18.92 for the option-price-setting
algorithm to say "see, I told you so".) That the 12 February 2021 call premia are being
driven by very short-term volatility can be illustrated by noting that the 19 February 2021
$19 option premium is $2.74, in contrast to a 12 February $19 option premium of $2.25.
In a more practical sense, though, we propose that the answer is "yes", and offer two
arguments. The first argument takes the perspective of a trader, who anticipates that the
meme-based behavior of the rebels will provide support ACB's price in the case of
temporary dips. Such a trader might sell a call option at 17, notice that the effective
purchase price is little more than the closing price on 09 February, and anticipate being
called away with a 6% return. This would be a particularly appropriate choice if the trader
believes that the price of ACB has temporarily risen too far and too quickly. Or, such a
trader might sell a call option at 19, whose premium offers a 13% return on the effective
purchase price. Or, noting that the premia degrade rather slowly as the strike prices
increase, the trader might sell an out-of-the-money call which still generates a significant
premium and holds out the possibility of greater gains. In other words, because the trader
anticipates that the price will recover in case of dips, they are happy when the stock is
called away and they are equally happy when it isn't. The call options in question should
be covered calls, as writing a naked call on a stock with this level of volatility is an
extremely risky endeavor. Although the assessment of a trading strategy shouldn't be
based upon the results of a single trade, in this case, regardless of which covered call option
was selected to sell, as of 12 February 2021 the investor would have an unrealized capital
loss, albeit one which has been reduced by obtaining a discounted purchase price. In other
words, the premise that the investor would actually be equally happy when the stock isn't
called away is put to a test.
Table 1: 12 February 2021 call options on ACB, as of 10 February 2021
Strike price Option
premium (bid)
Effective
purchase price
Intrinsic
value
Time value Maximum
profit
15 4.20 14.72 3.92 0.28 0.28
16 3.55 15.37 2.92 0.63 0.63
17 3.00 15.92 1.92 1.08 1.08
18 2.59 16.23 0.92 1.67 1.67
19 2.25 16.67 0 2.25 2.33
20 1.90 17.02 0 1.90 2.98
21 1.65 17.27 0 1.65 3.73
22 1.39 17.55 0 1.39 4.47
The second argument mirrors the NOK example in the introduction. In other words, if the
recent jumps in price are the result of anticipatory positioning by institutional investors,
they in effect are "trading signals" rather than new information about the underlying value
of ACB. To the degree that option premia also code information about the "true value" of
the stock in the underlying company then, for example, offering a 13% return in two days if
the price of ACB remains unchanged represents an inefficiency in option pricing. (A
believer in the efficient market hypothesis would reject the distinction between economic
Page 13 of 14
255
Archives of Business Research (ABR) Vol 9, Issue 2, February-2021
signals and trading signals, and instead assert that the volatility in prices simply reflects a
substantial and suddenly increased level of disagreement about the economic value of the
underlying company, and also that this level of disagreement is likely to continue, at least in
the near future.)
In summary, economic history might ultimately decide that January 27, 2021 was no more
than an interesting footnote, or perhaps GME will come to represent the start of a
successful revolution. The fact that the rebels are outgunned doesn't necessary guarantee
their failure and, in any event, the other side hardly inspires sympathy. Rather than simply
observe the battle with fascination, we recommend that investors profit therefrom. As they
do so, it can be useful to recognize that, when faced with coordinated action on the part of
individual investors options can become over-priced, and thus it will be advantageous to be
a seller of such options rather than a buyer.
Postscript
By 5 February 2021 the price of GME had dropped from a high of $483 to $64, with the
anticipation that it would drop still further. The parabolic phase of the short squeeze
ended quickly, as is often the case. Anyone who had bought before 22 January 2021 could
still exit at a profit, though.
Perhaps because the GME stock squeeze became a cultural phenomenon, in part
representing a revolution against the financial elites, the post-mortem commentary
strongly depended on one's point of view. A particular source of disagreement was the
decision to temporarily stop trading in GME (more accurately: to temporarily allow selling
but not buying). The rebels argued that this was an illustration of how the old guard
manipulates the system to protect their interests (and, who knows, might not have been
following their own rules about social distancing as they did so). Regulators focused on the
desirability of protecting inexperienced investors from unsound speculation. In some
circles, the participants in WSB were blamed for egging on the uninformed, and words such
as "conspiracy" were even bandied about. The business model of hedge funds underwent
additional scrutiny.
One indisputable fact is that someone bought GME at $483 per share and, as typically the
case for bubbles, that "last buyer" was likely a poorly-informed amateur investor. Indeed,
if that last buyer was intending to make a sociopolitical statement it was an unsuccessful
one: amateur investors cannot bring a putatively corrupt financial system to its knees by
losing money. In their defense, during the previous six months the participants of WSB had
initially reached a rough consensus that, were the short squeeze to appear to be working,
the most prudent strategy would be for them to sell enough shares to ensure a profit, and
thus be playing with the house's money. This intention was to some extent forgotten in the
heat of the moment -- perhaps out of greed, and perhaps out of the desire to maximize
harm to the affected hedge funds. In any event, one of the post mortem postings expressed
things rather well (paraphrased):
Page 14 of 14
URL: http://dx.doi.org/10.14738/abr.9768. 256
Samsa, G. (2021). Bonnie and Clyde Revisited: A Conceptual Model of the Behavior of the Instigators of the GameStop Short Squeeze of 2021.
Archives of Business Research, 9(2). 243-256.
"When you're buying in at the top and memeing about your diamond
hands after joining the sub less than a week ago you aren't
investing, you aren't even gambling, you're just losing."
To conclude with some points of optimism: (1) so long as they avoided buying GME during
the final parabolic phase of the short squeeze and remembered to sell along the way, the
rebels made a profit; (2) the rebels successfully made their point; and (3) although social
media contains significant amounts of misinformation, it also facilitates the
democratization of knowledge. We haven't heard the last of the rebels and, indeed,
"rebels2.0" is likely to be a significant upgrade over "rebels1.0". The financial
establishment should take notice.
References
1. Langevoort, D.C. (1992). Theories, assumptions and securities regulation: Market efficiency
revisited. University of Pennsylvania Law Review, 140(3), 851-920.
2. https://www.investopedia.com/terms/s/shortsqueeze.asp. Accessed 30 January, 2021.
3. reddit.com/r/WallStreetBets. Accessed 30 January, 2021.
4. https://www.marketwatch.com/story/nokias-stock-soars-toward-a-record-gain-on-record-volume- for-no-apparent-reason-11611774702?siteid=bigcharts&dist=bigcharts. Accessed 29 January, 2021.
5. Prochaska, J., & DiClemente, C. (1983). Stages and processes of self-change of smoking: toward an
integrative model of change. Journal of Consulting and Clinical Psychology, 51(3), 390–395.
6. Samsa, G. (2020). Now that behavioral finance has demonstrated that investor behavior is
(somewhat) predictable does that mean that arbitrage must make the biases in question disappear?
Archives of Business Research. 7(8), 146-160.
7. Samsa, G. (2020). Why does momentum persist? Archives of Business Research. 8(11), 119-125.
8. https://www.econlib.org/library/Enc/bios/Hume.html. Accessed 30 January, 2021.
9. https://quoteinvestigator.com/2017/11/30/salary. Accessed 07 February 2021.
10. Aliber, R.Z., Kindelberger, C.P., Kindleberger, C.P. (2011). Manias, panics, and crashes: A history of
financial crises. New York: Palgrave Macmillan.
11. https://www.sportingnews.com/us/nfl/news/reddit-super-bowl-ad- wallstreetbets/otv1bj28z66i1m5kx8r59dhxk. Accessed 9 February 2021.
12. https://www.marketwatch.com/story/tilray-shares-soar-29-as-reddit-message-board-sets-sights- on-cannabis-sector-11612977363?siteid=bigcharts&dist=bigcharts. Accessed 21 February 2021.