Why do markets form bubbles? Why do markets capitulate and collapse in value rapidly? Asked succinctly: why do investors engage in herding behavior? This is the fourth in a series of deep dive articles on behavioral finance and its major biases. Ultimately, I am going to land with A Theory of Behavioral Finance. In the meantime, let’s turn our attention to why investors stampede.

A Helpful Mnemonic Device: LOCHAARM

To my mind the major behavioral biases are:

• Loss aversion
• Overconfidence
• Confirmation
• Herding
• Anchoring
• Availability
• Representativeness
• Mental accounting

A helpful mnemonic device for remembering these biases is LOC HAARM, brain lock that harms investment performance. Hark! What is that sound? The sound of thousands of hooves herding.

Herding Bias: Origins and Manifestations

Surely if you are reading this article you have heard (pun intended) that market prices sometimes form speculative bubbles and have colossal crashes, yes? Investors generally attribute these two phenomena to a seemingly innate behavior in animals of all kinds, even people, to all move in the same direction at the same time unconsciously/without regard to reason. This behavior of course is herding.

Market Bubbles

Most famous in investing to describe herding is journalist Charles MacKay in his 1841 classic, Extraordinary Popular Delusions and The Madness Of Crowds.^[1] Interestingly, the book is not primarily focused on economic bubbles, having just three chapters on the subject. MacKay spends most of time debunking things such as religious fanaticism, fortune-telling, and fashion fads among other things. A single quote from this, his best known work, should serve our purposes here:

“Men, it has been well said, think in herds; it will be seen that they go mad in herds, while they only recover their senses slowly, and one by one.”^[2]

Of the economic bubbles, MacKay recounts a series of historical/hysterical moments in history where investors went crazy for different types of assets. Notoriously, the South Sea Bubble of the 1710s, the Dutch Tulip Bulb Craze of the 1630s, and the Railway mania of the 1840s.

Ironically, MacKay himself is criticized as an ardent booster of bubbles in his role as a journalist.^[3] For example, researchers have found that when writing during the time of the Railway Mania he said, “There is no reason whatever to fear a crash.”[4] Other researchers who have done a deep dive into the Tulip Bulb craze have found that its severity was greatly exaggerated by MacKay. In other words, all of us are susceptible to behavioral bias, even if we are considered classic authorities on the subject.

Market Crashes

It is at this moment though that I need to point your attention to the obvious. Herding is also witnessed when securities prices fall. Though, perhaps they are better called stampedes when there is a rush for the exits.

In my recent article “Fact File: S&P 500 Sigma Events, slight return” I provide a history of the major market moves up and down of this index. In short, while there is a positive skew to daily stock market returns, the sigma events are remarkably symmetrical. The range of possibilities extends out to 10 sigma events, up and down. Ouch!

In investing there are other manifestations of herding behavior that are much more subtle. At the macro level think about the herding effect created by a combination of blind monthly retirement fund contributions and dollar cost averaging, coupled with index fund/ETF as a sponge to absorb all of these monies in defiance of the prices and risks of the component businesses. In other words, this is herding by design. Could it be that a lot of the success of index fund investing is herding by design? Create a list of securities that everyone should buy and then set it up so that they buy regardless of what the performance of that list says.

At the micro level, investment teams can easily lose sight of reality in raging bull markets, and lumbering bear markets, too. Here group think sets in and the same conversations among team members happen over and over again. This is why a hallmark of great investment teams is a constant navigation to truth and not to habit.

Herding Bias: Nuances

Researchers ranging from biologists to mathematicians to economists have all studied herd behavior. For example, in biology it has been noticed that when stampeding each member of a herd reduces its own dangers by orienting itself as closely as possible to the center of the whole herd.^[5] Birds flying in a flock are also known to exhibit this behavior. Namely, there is no one leader; and, essentially, each member of the flock is following the others.

I believe this is an apt description of behavior in the midst of asset bubbles, where absolute notions of value and return are surrendered to relative notions of value. That is, absolute valuation based on valuation theory and as derived by discounted cash flow analysis are surrendered in favor of relative valuation. In fact, I think developing a measure of when relative value takes over from absolute value probably provides an insight as to when a bubble has formed.

During bubbles participants crowd into individual securities, or into markets to orient themselves, “as closely as possible to the center of the whole herd.” Most do not want to be left behind or stampeded over. Here the center of the whole herd is the upward march of the aggregate measures of performance, such as indices.

Mathematicians have tried to model this behavior for many different applications, but obviously an understanding of such phenomenon is useful to investors. In a well cited paper, researchers note two important features of herding behavior:^[6]

• Mechanisms of Transmission. How thoughts and behavior among people are communicated to one another. Given the near omnipresence of interconnected communications via smart devices and social media this has important implications for investors trying to understand herding.
• Patterns of Connections. This refers to the way in which different members of the group interact with one another, and from which emergent behaviors come.

Interestingly, this research was written just after the stock market collapse of March 2009. Many investment researchers believe that one mechanism of transmission that is extremely important to herding behavior is asymmetric information.^[7] Because not all market participants have the same information, when new information emerges there is a tendency of people to overreact to news. This is similar to a scarcity mentality or a gold rush mentality. Limited available gold/information leads to people crowding into similar decisions to avoid being left out of the scramble.

Referring back to my article about the history of S&P 500 sigma events, I would also like to point out that the 9 and 10 sigma events have all occurred in the era of rapid dissemination of information. All of these events have happened in the era of the Internet, e-mail, and instant messaging except for the October 1987 crash. But even then investment pros had access to instantaneous market movements via the Quotron machine.

A backburner project of mine is to research the “Patterns of Connections” within financial markets. That is to map the network and to check out the most important nodes. My educated guess is that when there are crashes that the opinion of the network nodes switches from favorable to unfavorable opinions about the direction of financial markets.

My last point on this subject is that most investors believe that when they hear news about the “Stock Market” such as, “Today investors were concerned that the amount of chicken clucks declined” that “The Market” means every investor. In reality a very low percentage of the market trades each day. Proprietary research I did in 2011 showed that just 0.35% of the market trades, on average. However, in stampeding years it is 0.64%. What this means is that it takes very few marginal idiots to collapse the value of your portfolio.

Conclusion

To me, herding behavior is clearly a rich source of alpha because it leads to distortions in understanding. Sometimes herding leads to volatility, and sometimes to stability. But those distorted perceptions of reality are opportunities for the clear headed active manager to earn alpha.

[1] MacKay, Charles. Extraordinary Popular Delusions and the Madness of Crowds. Public Domain. 1841

[2] Ibid. p. 4

[3] Odlyzko, Andrew. “Charles MacKay’s own extraordinary popular delusions and the Railway Mania.” School of Mathematics, University of Minnesota. 26 February 2012

[4] Ibid.

[5] Hamilton, W.D. “Geometry for the Selfish Herd.” Journal of Theoretical Biology. Vol. 31 (1971): pp. 295-311

[6] Raafat, Ramsey M., Nick Chater, and Chris Frith. “Herding in humans.” Trends in Cognitive Sciences. Vol. 13, No. 10 (2009): pp. 420-428

[7] Brunnermeier, Markus K. Asset Pricing under Asymmetric Information – Bubbles, Crashes, Technical Analysis and Herding. Princeton University Press. 2000