On Conformity
In their book Social Psychology,1 David Myers and Jean Twenge summarized the landmark research on how our urge to conform affects our perceptions and/or behavior.
1. Sherifian Ambiguous Reality
In 1935, Muzafer Sherif studied group conformity in a laboratory setting.2 As a participant in a Sherif experiment, you sit in a dark room and a point of light appears, disappears, reappears, and disappears fifteen feet in front of you. You are asked to guess how far the light has moved.3 You guess an inch. When the experiment is repeated several times, your guesses continue to average out to about one inch.
The experiments are repeated the next day, this time with two other participants who had also experienced solo experiments the day before. Each participant initially makes his/her best guess from the day before: two inches and eight inches respectively. When it comes to your turn, you wonder if your solo-experiment best guess was too low, so you say “two inches.” The experiment is repeated several times over several days. Over multiple iterations, your guesses and the guesses of the other two participants converge at about two inches.
When the experiments are over, the three of you learn that the light hadn’t moved at all; the perceived “movement” is caused by an optical illusion called the autokinetic effect.4 The real experiment wasn’t to test your ability to detect how much a point of light has moved from fifteen feet away, but to test group suggestibility. Sherif’s conclusion was that group norms can form, even when such norms are false.
Subsequent experiments showed that if participants weren’t told that the light had not moved and were retested alone a year later, they would still tend to continue following the group norm.5 False beliefs can also be transmitted to others. In 1961, Robert Jacobs and Donald Campbell showed that if a confederate deliberately guesses high on the Sherif experiment in the presence of Participant 1 (a real participant), then Participant 1 guesses in front of Participant 2 (who didn’t hear the confederate’s guess), then Participant 2 guesses in front of Participant 3 (who didn’t hear the other participants’ guesses), and so on, the inflation persists through Participant 5.6
The importance of a task can increase conformity. In 1996, Robert Baron, Joseph Vandello, and Bethany Brunsman showed participants a “suspect” and then asked the participants to identify the suspect in a lineup within one second (not easy!). A group of participants was told their answers were very important to the legal community; the other group was told that the experiment was just a trial. The group that was told their answers were important conformed 51% of the time and the other group 35% of the time.7
At Discerene, we are wise to the effect of suggestibility on our view of reality. For example, the intrinsic value of a company is often a Sherifian “ambiguous reality.” If one adds an extra zero to the observed market cap of Company X (e.g., $10 billion instead of $1 billion), the discounted-cash-flow (“DCF”) model that an analyst subsequently builds to value it will likely contain far more optimistic assumptions than the “control” condition. Conversely, if one removes a zero from the observed market cap of the same Company X (e.g., $100 million instead of $1 billion), the DCF model that an analyst then builds to value it will likely contain far more pessimistic assumptions. Our view of reality is often not ours alone.
To counteract this effect, we’ve trained ourselves to maintain our focus on the companies we study (the “pinpoints of light in the darkened room”) and not on the commentary of others (the “guestimates of other participants in the room”). In some cases, our DCF models today yield intrinsic value estimates that are modest fractions of the current market caps of the companies. The natural response is: Surely we made a mistake somewhere! We try to focus on the epistemology of our work: Are our research methods valid? Are the facts upon which we are basing our research sound? Is the scope of our conclusions reasonable? If our work is sound, we try to stand by it, even if the dissonant opinions of other market participants continue to ring in our ears.
2. Aschian Laboratories
What if factual ambiguity is removed? The seminal 1951 experiments conducted by Solomon Asch help answer this question.8 If Sherif is the father of conformity experiments, Asch is the midwife.
As a participant in Asch’s experiment, you are the sixth person in a row of seven participants.9 The experimenter asks each participant which of the following three lines is the same length as the reference line:
%201.png)
The five people before you each call out Line C, and so do you when it’s your turn. A second trial (with new lines) is then conducted, and again all five people before you call out the line that you planned to call out as well, as do you when it’s your turn. However, in the third trial, Participant 1 calls out “Line A.” Then Participant 2 calls out “Line A.” Then Participant 3 calls out “Line A.” So do Participants 4 and 5! When it’s your turn, you stare and stare at the lines. It seems obvious to you that Line C is the line that matches the reference line. Your pulse quickens. You are in the throes of epistemic angst: Which is true — what I observe, or what others are saying?
The experiment is repeated several more times, and each time, all five participants before you identify the same line, which does not agree with your direct observation. You begin to wonder: Are my eyes reliable? How should I continue to answer these questions when it’s my turn?
As it turns out, the rest of the “participants” in Asch’s experiments are confederates and you are the sole subject. The test isn’t really about your perceptual judgment, but about your willingness to conform. Asch ran his experiment multiple times. In control conditions, participants answering alone identified the correct line >99% of the time. In contrast, 75% of participants who sat sixth in a row of confederates who deliberately gave a common incorrect answer conformed at least once. In all, 37% of the responses were conforming.
These experiments are especially striking because participants were not told to try to achieve consensus or that there were points for agreeing with each other. The urge for a participant to conform stems from an automatic heightened arousal from knowing that he or she is standing out.10
In investing, our observation is that epistemic ambiguity isn’t always at issue. We own some incumbent businesses purportedly being disrupted by new entrants (often highly unprofitable) with businesses that are tiny fractions of those of incumbents. As hip as the incumbents are stodgy, such disruptors nonetheless oftentimes have expected unit economics that are far inferior to those of the incumbents. Even so, some such disruptors boast valuations that are comparable to or larger than the businesses they seek to displace. Naturally, we ask: Is there some bigger “game” being played by the disruptors, e.g., some larger market that they are on their way to addressing if they succeed at disrupting existing incumbent business models? The answer is, in some instances, quite unambiguously “no” or “not likely.”
Still, we often find ourselves in metaphorical Aschian laboratories full of people claiming that magic beans that will hopefully grow to three-foot beanstalks are worth more than fully grown ten-footers already generating bountiful, giant edible pods. Whatever pressure we may feel to conform our views to others, we believe that it is important to retain the courage to state, simply, that we believe they are not.
In our industry, courage can sometimes be quite prosaic, but it is courage nonetheless!
1. Myers and Twenge (2018), Social Psychology, McGraw Hill.
2. See Sherif (1935), “A Study of Some Social Factors in Perception,” Archive of Psychology, 187; Sherif (1937), “An Experimental Approach to the Study of Attitudes,” Sociometry, 1.
3. See, e.g.: https://www.youtube.com/watch?v=0DoIxN6B4PQ&t=20s
4. See, e.g.: https://www.britannica.com/science/autokinetic-effect
5. See Rohrer, Baron, Hoffman, and Swander (1954), “The Stability of Autokinetic Judgments,” Journal of Abnormal and Social Psychology, 49.
6. Jacobs and Campbell (1961), “The Perpetuation of an Arbitrary Tradition Through Several Generations of a Laboratory Microculture,” Journal of Abnormal and Social Psychology, 62.
7. Baron, Vandello, and Brunsman(1996), “The Forgotten Variable in Conformity Research: Impact of Task Importance on Social Influence,” Journal of Personality and Social Psychology, 71(5).
8. See, Asch (1951), “Effects of Group Pressure on the Modification and Distortion of Judgments,” in Guetzkow (Ed.) (1951), Groups: Leadership and Men, Carnegie Press; Asch (1952), Social Psychology, Prentice Hall.
9. See also https://www.youtube.com/watch?v=TYIh4MkcfJA
10. Hatcher, Cares, Detrie, Dillenbeck, Goral, Troisi, and Whirry-Achten (2017), “Conformity, Arousal, and the Effect of Arbitrary Information,” Group Processes and Intergroup Relations.