“If people weren’t wrong so often, we wouldn’t be so rich.”
Charlie Munger
Despite most business school curriculum thinking, it’s now widely accepted that global stock markets are not that perfectly efficient.
If you think about it, how could one conclude that stock prices are efficient after rationalizing the events of the past 20 years?
From the epic bubble that led to the unsustainably high tech stock pricing peak in 2000 all the way to the recent crypto mania, it’s hard to even argue that the aggregate of most investors is pricing public stocks rationally.
There may indeed be an argument that the above inefficiencies above are hard to arbitrage, on the basis that “the market can stay irrational longer than you can stay solvent”.
After all, isn’t this this what allows them to exist in the first place ? And there are indeed plenty of examples showing the existence of over-reaction to the downside. Indeed,many incredible companies have traded at prices well below their intrinsic value over the past 20 years (although certainly at a lesser rate than in the 1980-1990 period)
I need to confess something. This article commit one of the cardinal sins of value investing. I’m about to argue that understanding and predicting mass investor psychology is not only possible, but should be a prerequisite for most good investments nowadays.
The reason is simple. This argument has been brewing in my mind for over a year and formulated after observing many so-called “value investors” buy overpriced tech companies and cyclical businesses at the peak of their earning cycle.
The key question is this: What is that missing ingredient that prevents most investors from identifying great value opportunities, or that leads them to stumble into value traps?
This post argues that this has partly do with the following conventional wisdom : understanding and predicting the aggregate psychology of investors in a particular stock is neither possible nor desirable.
Can Markets Ever Be Wrong?
As Michael Mauboussin makes clear in his wonderful book, More Than You Know, simply creating a market which allows individuals to bet on probabilities typically leads on aggregate to extremely accurate predictions. This is much more than any individual could ever achieve.
In other words, the traditional view that simply “zigging when others zag” leads to profitable investments or that “the crowd is always wrong,” is by markets’ very nature incorrect. The combined predictive power of millions of individuals (and now high-powered computers) has to be more accurate than any individual investor.
This seems to be true, even after accounting for the now widely-known field of behavioral finance first introduced by Daniel Kahnemann. Although individual investors may be affected by powerful cognitive biases, most of these biases are apparently corrected by other investors who are affected by different biases.
Let’s imagine anoversimplified example where an investor – driven by overconfidence bias – buys an overpriced tech stock. On the other side of that trade sits an investor taking profits but prone to the usual tendency that people sell their winners too early and cut their losers too late. In this scenario, both investors have balanced out each other’s ‘irrationality.’
Or we can assume that the investor on the other side of the trade is a hyper-rational institutional investor, who can balance irrationality with billions of assets under management and trying to profit from certain inefficiencies.
The reality of course is obviously much more complex, However, the general idea is that the involvement of a large number of participants in a market tends to lead to far more accurate results than what any individual expert could predict.
In fact, in his other great book, Think Twice, Mauboussin argues that three conditions are required to allow crowds to predict well (i.e., for markets to be ‘efficient’): diversity, aggregation, and incentives.
If we consider any stock market, it generally fulfills all three of these requirements. A diverse investor base with different viewpoints isaggregated into a single market with a common powerful incentive : to maximize their returns.
Of course, despite all this theory which seems to make sense, we nevertheless occasionally witness inefficiencies on an epic scale. It’s undeniable that many companies occasionally do trade well above or below their fair value. And it’s also becoming exponentially harder to identify these inefficiencies.
This is why a new mental model has to be introduced for identifying these inefficiencies more accurately and reducing the error rate of value investments.
The Mass Delusion Hypothesis: The True Source of Inefficiencies Hiding In Plain Sight
Enter the “mass delusion hypothesis” — a term I’m introducing here and that I’m hoping might become a useful tool to make better investing decisions.
Let’s circle back to our earlier example of two separate investor biases canceling each other out. In that example, we ignored the fact that certain biases are more powerful than others. For example, Daniel Kahneman has himself acknowledged that overconfidence ranks as one of the most powerful biases. So there seems to be a certain “pecking order” of cognitive biases.
In addition, humans have a built-in tendency to follow the crowd and this can create powerful self-reinforcing feedback loops.
Why is that? Well, it turns out that this core “cancelation of biases” idea fails to take into account something fundamental : while most investors have very different backgrounds, worldviews, and ideas, we all have a common evolutionary tree. And if we trace this tree back through our long line of ancestors, we will see that our nomadic hunter-gatherer instinct is still there – to survive and reproduce.
This explains the very existence of cognitive biases exist in the first place. They’re shortcuts – originally designed to help us thrive in a far more primitive environment. If we then apply this evolutionary lens on investor psychology, could help us predict which cognitive biases are more powerful than others. Evenven more importantly, we’ll be able to see how they stack on top of each other (we’ll discuss this idea of “stacking” in more depth later in the article).
This simple train of thought has staggering implications. When a combination of powerful biases around a security or group of securities takes place, most market participants tend to become delusional. In that scenario, there are either no traders or insufficient numbers of traders on the other side with opposing biases, because most participants are strongly affected by these cognitive forces. In other words, the biases can lead to inefficiency.
This tends to happen when certain “high level” or very primitive) biases kick in. This means that other lower-level biases will not be able to balance them out. Think of the fear traders feel during a market sell-off.
There are also insufficient numbers of rational investors who are able to take advantage of this mass irrationality. This may be due to institutional limitations.
For example, a professional investor may be under pressure from clients who may dislike owning a certain stock that is unpopular. Or, perhaps cognitive forces are simply too strong and overwhelming, even for extremely smart and motivated professionals.
In any case, a professional investor willing to bet against a mass delusion is exposed to the risk of suffering huge losses. Buying a cheap stock (or shorting an expensive one) which falls by another 50% over a three-year period after its initial purchase might lead to clients taking their money out of the fund and/or to the manager being fired. Even if it turns out to be a great investment 10 years down the road, mass delusion would still have killed many young contrarian fund managers.
Perhaps most importantly, this cumulation of cognitive biases leads to a breakdown in one of the three prerequisites of high market efficiency: a reduction in diversity. Since all market participants are humans and most will suffer from the same group of powerful, primitive cognitive biases, diversity is thereby reduced. Furthermore, the very fact that the majority of participants are affected by a particular “delusion” makes it even more likely for others to follow suit – a common herd behavior.
Just to make it clear at this stage— I’m not claiming that suddenly, all market participants become brain-dead and unable to think simultaneously. The argument presented here is indeed nuanced. The core idea is that when a majority of market participants is affected by a similar set of biases, it leads to a breakdown in the fabric of market efficiency. Diversity is reduced and therefore inefficiencies are significantly more likely to appear under these circumstances.
After this brief introduction, let’s introduce a formal definition of the mass delusion hypothesis:
Mass delusion hypothesis: The majority of participants in a market occasionally become overwhelmingly subject to a combination of powerful, primitive cognitive forces. This leads to a mispricing in a security or group of securities due to a breakdown in diversity. This mispricing is hard to arbitrage, even for rational institutional investors. Market inefficiencies are unlikely to occur unless a mass delusion is occurring.
How Powerful Stories Can Trigger Mass Delusions
“The four most dangerous words in investing are “this time it’s different.”
Sir John Templeton
Now that we’ve established the definition of mass delusion and the theoretical basis behind a reduction in market efficiency, let’s attempt to determine some factors that trigger this phenomenon.
Let’s start with the most obvious factor :a strong narrative such as a story about the inevitable rise or fall of a particular company or industry.
Our brain has been wired through evolution to have a powerful, built-in response to storytelling. Evolutionary researchers argue that stories are one of the oldest and most primitive ways of communicating used by humans. We see and perceive the world through the lens of narratives — a linear way of understanding the world. In fact, we literally use these stories to make sense of the world around us.
A good basic example would be a certain news headline about stock market movements. A CNBC article from the day before this article was written reads as follows:
“In regular trading the Dow fell more than 350 points, or 1.03%. The S&P 500 and Nasdaq Composite lost 1.4% and 2%, respectively…..The moves came as investors lost hope that the Federal Reserve will be able to engineer a soft landing. Instead, concerns swirled around the state of the economy and whether an economic downturn is approaching.”
The fact is that the Dow fell 350 points. But notice how the journalist can’t help but create a narrative around the market movement. In this case, our brain instantly understands the narrative that investors are worried. It helps the reader make sense of the market move, and even makes the article more memorable.
Yet this is a dangerous way for investors to think. The reality is that the Dow can lose 350 points on any day on the basis of any combination of different factors. It cannot be just a simplistic story about investor worry. In other words, our craving for stories leads us to gobble up naive cause-effect relationships through stories. Our brain is simply not designed to understand the sheer complexity of business and markets.
Now, let’s tie this to our discussion about market inefficiencies. When it comes to analyzing a security or group of securities, the narrative bias is strong.
One has to look no further than the recent rise and collapse of tech stocks, both during and after the Covid-19 lockdowns, when all started with a powerful narrative.
“The world has changed. Everyone is working from home and spending more time at home than ever before. All tech companies whose products are used at home such as Netflix, Zoom and Slack will benefit.”
Intuitively, this linear story makes sense. Since people are stuck at home, so they will be watching Netflix more. In fact, the early phase of market narratives are typically not far from the truth and it all starts out in a reasonable fashion.
But then, the so-called « narrative fallacy » seeps in. Feedback loops then amplify the phenomenon :
“XYZ tech company has grown its earnings per share by 30% over the last year, but this is expected to continue for a long time. The addressable market is huge. This is a revolution after all.”
Never mind the base rate that most companies end up growing at or near GDP growth rates and never mind regression to the mean, where it’s likely that growth and perhaps even earnings will get back to where they were before. A powerful story has entered the mind of most market participants and overwhelmed our rational senses.
How Cognitive Biases Reinforce Each Other to Reduce Diversity and Create Inefficiencies
It’s at this point that participants begin buying the security in droves. Even at unreasonable price-to-earnings ratios, everything can be justified by the narrative.
And then, other biases such as « crowd bias » or « herding bias » kick in. As more and more people repeat the the narrative, the harder it becomes to resist. Classic overconfidence leading to a bubble.
Then there’s the « availability heuristic. » The more the narrative is disseminated and repeated, the more investors overvalue its likelihood of happening. In parallel, its little sister, the « salience bias » then kicks in, predisposing us to focus on items that are more emotionally striking. And that’s then the previously discussed « base rate fallacy » rears its ugly head.
Charlie Munger’s “Lollapalooza effect” has begun – a “confluence of multiple tendencies that reinforce each other and lead to extreme consequences.” The consequences of following the herd into a mass delusion are obvious – extremely low or negative returns for very long periods of time. It’s really a high price to pay for a bit of excitement.
Obviously over the past 20 years, the majority of such mass delusions have leaned toward the upside. Driven by record-low interest rates, narratives keep popping up that capture investors’ imagination, preventing them from “thinking clearly”.
It appears that periods of low interest rates mostly drive mass delusions to the upside (overpricing of companies), whereas periods of high interest rates tend to lead to mass delusions to the downside (underpricing). Think of the early ‘80s – one of the prime periods when value investors such as Buffett generated extremely high returns. I would argue that this comes in part from taking advantage of mass delusions and excessive pessimism.
Empirical Evidence for the Existence of Mass Delusions
One cannot claim to develop a framework without empirical evidence of its existence. In closing, I will argue that the well-documented “momentum factor” is concrete evidence for the phenomenon of self-reinforcing mass delusions.
Let’s start with the basics. The momentum factor is simply the tendency of winning stocks to continue performing well in the near term. However, it’s also well known that momentum eventually leads to reversal : “what can’t go on forever will eventually end.” Some academics have argued that momentum arises as a result of initial under-reaction to a trend, and then eventual over-reaction.[1]
From a psychological perspective, the momentum factor is, in my view, simply a great illustration of the narrative fallacy triggering a large number of cognitive biases which eventually create a mass delusion. Here’s the most standard case:
- A narrative enters the mainstream about a company or group of companies.
- That narrative has a relatively sound logical basis, so participants begin accepting and trading on the basis of that narrative. This leads to an initial price rise.
- A few cognitive biases then begin working together to cloud the judgment of participants. Prices then begin to rise above ‘rational’ levels.
- As more and more participants trade under the influence of delusion, the powerful herding tendency (social proof) further reduces their ability to think clearly. Prices then enter bubble territory.
- The feedback loop continues to reinforce itself until the narrative begins to break due to a shock to the system (e.g., interest rate changes, big event, etc.)
- Finally, once some market participants beging to realize their delusion, prices quickly return to normal as market diversity is restored.
In other words, whichever narrative has captured the imagination of investors over the past 6-to-12 months is likely to continue doing so until it abruptly ends.
How to Profit From Mass Delusions
In closing, the underlying question on this information is this : how can investors actually exploit mass delusions?
There’s obviously never any foolproof way of generating excess returns. However, I do believe that understanding mass delusions can give an edge to rational investors.
The main use case this model seems to allow is a deeper understanding of why particular securities or industries are underpriced. It’s unfortunately common for value investors to ignore this important question when selecting investments with the quesstion: “why is the market giving me this great opportunity”?
Unless there’s a clear operating framework based on investor psychology, that important question will often prove hard to answer. I would argue that, if it cannot be answered correctly from the perspective of investors’ biases, then the stock might well not be undervalued at all.
It might also prove wise to operate from the inverse perspective: once an investor identifies a mass delusion (leading to pessimism), it would be worthwhile to look at the company or companies affected by that particular delusion as they might offer excess returns once the mass delusion goes away.
Overall, adding the mass delusion framework to an investor’s toolkit will likely help avoid key mistakes that would lower returns (e.g., investing in companies that are neither actually undervalued nor even overvalued) and provides another “lens” through which to find and evaluate potential investments.
SOURCES
[1] https://joi.pm-research.com/content/29/3/38
