FTX has dominated my twitter feed recently: the utter collapse of a $32 billion company backed by the venerable VC Sequoia Capital – an unthinkable history. The FTX blowup, did not occur in isolation. The NASDAQ is down 29% year-to-date, with many tech stocks down over 80%. Such boom-busts are not anomalies. The 1630s Tulip-mania, 1930s Great Depression, the dot-com bust, and US housing bubble, are some of the biggest meltdowns in financial history.
This is not supposed to happen. Aren’t markets efficient? Did the present value of the US economy really change so much in the last six months? I won’t concern myself with the well-discussed limitations of the efficient market hypothesis; suffice to say that “investing is more art than a science.” But what about the wisdom of crowds? There is evidence of this in Francis Galton’s Vox Populi experiment. Galton asked 787 attendees at a livestock exhibition to guess his Ox’s weight. Each wrote their guess on a ticket, and the average was 1,197lbs, surprisingly close to the real 1,198lbs. Crowds have been shown to accurately predict elections, sports games, and even the location of a missing submarine.
If crowds are so wise, why do they fail to predict the market? How can they get something so important so horribly wrong? Conventional wisdom explains booms and busts through the primal emotions of greed and fear. Whilst greed and fear are undeniably at work, this explanation is not fully satisfactory to me. All serious investors have studied market cycles and are familiar with Warren Buffett’s famous aphorism “be fearful when others are greedy, and greedy when others are fearful.” How do professional investors (even pre-eminent ones) get sucked into speculative euphoria?
A key, often ignored, criterion in Galton’s experiment was that responses were private. This does not hold in financial markets where we have changing share prices, sell-side analyst predictions, and countless ‘experts’ – who are decidedly un-expert – with weak opinions strongly yelled (on CNBC Squawk Box). This inundation of information facilitates groupthink and conformity, changing perceptions, opinions, and behaviors of people in ways that are consistent with group norms. There are two types of conformity.
Informational influence is when we use others’ behavior or attitudes as information or evidence about reality. Evidence for this effect is found in Sherif’s Autokinetic Effect Study. In this study, participants were asked to judge how far a bright dot on a dark screen had moved. Subjects first made judgments individually, and then in groups. Over time, estimates converged because participants saw information in others’ guesses. Informational influence underpins financial bubbles. There is, as John Kenneth Galbraith calls it, this “feeling that with so much money involved, the mental resources behind them cannot be less.” This belief that others know something we don’t leads to a lemming-like behavior from investors, bidding up asset values, confirming personal and group wisdom – until the inevitable crash.
The second type is normative influence, when people conform to others’ views or expectations in pursuit of social approval. This is evidenced in Asch’s Line Experiment, wherein subjects viewed a card with a line on it, followed by another with three labelled lines. Students had to say which line matched the length of that on the first card. Other students were in fact actors paid to give incorrect answers, meaning “a group unanimously contradicted the evidence of [subjects’] sense[s].” The pursuit of social approval and fear of being ostracized meant 75% of participants conformed at least once. Professional investors are susceptible to normative influence because they are continually evaluated on their performance by peers and client, whose approval they seek. This means they may buy whatever is in vogue, even if they do not intrinsically believe that is the right decision.
Clearly, social influence undermines the “wisdom of crowds” effect. This is supported by Lorenz et al’s study, wherein subjects were asked to reconsider their responses to factual questions after having received average or full information of others’ responses. Although groups were initially “wise,” knowledge about others’ estimates made them unwise. Social influence diminished the diversity of the crowd and “range reduction” pushed the true answer to peripheral regions of the range of estimates. Ironically, even as the crowd became less accurate, the convergence of their estimates made individuals more confident.
Having just experienced the bang that (likely) marks the end of a speculative episode, one may pause to reflect on investors’ agency. Booms are born out of a kernel of truth relating to innovation – but they escalate into mass insanity. As asset prices go up, people perceive information in these movements: there must be a reason. This drives prices up further. Eventually, irrational herd behavior sets in, and non-believers feel compelled to participate or else look bad in front of clients and peers. In this way, the informational and normative influence that underpins speculation literally “buys up... the intelligence of those involved.” CONFORMITY KILLS.
