Herding Behavior: How Crowds Overwhelm Private Information

Key Takeaways

• Herding occurs when investors stop trusting their private information and copy the visible decisions of earlier agents instead.
• The Nasdaq fell more than 75% from its 2000 peak after an informational cascade drove prices far above any fundamental anchor.
• Career-risk herding, managers mimicking the benchmark to limit career damage, is rational individually but destructive at the institutional level.
• A herd persists until a triggering event breaks the cascade, and that trigger is almost always visible only in hindsight.

What It Is

In finance, herding describes the clustering of decisions among investors who otherwise have independent information. The foundational theoretical models are Abhijit Banerjee's 1992 Quarterly Journal of Economics paper "A Simple Model of Herd Behavior" and Bikhchandani, Hirshleifer and Welch's 1992 Journal of Political Economy paper "A Theory of Fads, Fashion, Custom, and Cultural Change as Informational Cascades."

Their shared insight: under uncertainty, rational agents who observe earlier agents' actions will often rationally ignore their own private information and follow the pack. The result is an informational cascade, in which early decisions propagate forward and the social belief stops aggregating new information. Cascades can settle on correct or incorrect outcomes with roughly equal ease.

The Intuition

Most investors are not rolling their own valuation models. They are watching what others do and using that as a signal. The reasoning is often sound: other people know things I do not, so their trades contain information.

The problem arises when everyone reasons this way at the same time. Once the first few trades push the consensus, each subsequent participant puts more weight on the visible trades than on their own private signal. Private information stops flowing into prices. The herd can then drift far from any fundamental anchor and stay there for a long time, especially if new entrants keep arriving and keep deferring to the consensus.

This is the story Robert Shiller tells in Irrational Exuberance (2000) about the late 1990s tech bubble. Investors bought tech stocks because other investors were buying tech stocks, fed by media narratives and by the feedback loop between rising prices and apparent confirmation. When the cascade reversed, the Nasdaq fell more than 75 percent from its 2000 peak.

How It Works

Informational cascades arise whenever three conditions hold: agents act in sequence, each agent sees earlier actions, and each agent has private but noisy information. The Bikhchandani-Hirshleifer-Welch model shows that once two or three agents in a row choose the same action, the expected information value of any one subsequent agent's private signal is dominated by the accumulated public signal. Rational updating then favors copying.

Banerjee's 1992 model reaches a similar conclusion with a continuous action space. Bikhchandani and Sharma's 2001 IMF paper extends these ideas to financial markets, distinguishing true informational herding from spurious herding (where everyone reacts to the same public news) and from reputational herding (where fund managers mimic the benchmark or peers to limit career risk).

Career-risk herding is particularly important in institutional markets. A manager who underperforms while deviating from the benchmark loses clients. A manager who underperforms while hugging the benchmark rarely does. That asymmetry quietly rewards closet indexing and crowd-following.

Worked Example

Imagine a new sector, say distributed energy storage, where no participant has strong fundamental information. The first three investors to take a view happen to buy. Each subsequent investor observes three buys and zero sells before they decide.

Even if your own private research suggests the sector is overpriced, Bayesian updating on the public signal of three buys can outweigh your private signal. You buy. The next investor sees four buys. They buy. The cascade is under way, and it can continue through dozens of decisions that include no fresh private information.

If it later turns out the first three investors were mistaken, the cascade produced a wrong consensus at scale. Once news arrives that breaks the cascade (bankruptcy, regulatory action, a credible short seller), the reversal can be violent because there is no second layer of independent buyers to absorb selling.

Common Mistakes

1. Equating herding with stupidity. Herding can be individually rational even when the aggregate outcome is poor. Calling the herd dumb misses the mechanism and tends to make you overconfident about fading it.

2. Assuming you can easily time the reversal. Keynes' observation that markets can stay irrational longer than you can stay solvent applies here. Cascades persist until a triggering event breaks them, and the trigger is usually visible only in hindsight.

3. Mistaking momentum for herding. Momentum is a systematic factor that has shown positive excess returns in many markets over long samples. Herding is a behavioral mechanism that can contribute to momentum but is not identical to it. Trading momentum rules is not the same as riding every herd.

4. Ignoring reputational herding in your own process. Benchmark-hugging, closet indexing, and peer-mimicry are herding behaviors that feel prudent in the moment. They are often the single largest source of underperformance at institutions where managers fear career risk more than market risk.

Frequently Asked Questions

What is herding behavior in investing? Herding is when investors stop acting on their own research and start copying the observable decisions of others. It is individually rational when others have better information, but produces price cascades that drift far from fundamentals when everyone defers to the crowd at the same time.

How does herding behavior affect investment decisions? It creates informational cascades where private signals stop flowing into prices. A herd can push an asset far above any fundamental anchor and sustain it there until a triggering event reverses the cascade, often violently, because there is no independent buyer base to absorb selling.

What is a real-world example of herding behavior? The Nasdaq bubble of 1999–2000 is the textbook case. Retail and institutional investors bought technology stocks largely because other investors were buying, fed by rising prices and media narratives. When the cascade reversed, the Nasdaq fell more than 75 percent from its 2000 peak.

How can investors avoid herding behavior? Write down your own assessment before looking at consensus. Check whether the opinions you are collecting are genuinely independent or traceable to the same source. Be particularly alert to career-risk herding: benchmark-hugging and peer-mimicry feel prudent but are often the largest source of institutional underperformance.

Is herding behavior always irrational? No. When earlier participants genuinely possess better information, deferring to their actions is rational. The problem is that cascades can form on weak foundations and lock in incorrect prices just as easily as correct ones. Calling herders irrational misses the mechanism and tends to make contrarians overconfident about fading the crowd.

Sources

1. Banerjee, A. (1992). "A Simple Model of Herd Behavior." Quarterly Journal of Economics 107(3), 797-817. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4779644
2. Bikhchandani, S., Hirshleifer, D. & Welch, I. (1992). "A Theory of Fads, Fashion, Custom, and Cultural Change as Informational Cascades." Journal of Political Economy 100(5). https://bpb-us-e2.wpmucdn.com/sites.uci.edu/dist/c/362/files/2017/01/Palgrave-information-cascades-Online-version.pdf
3. Bikhchandani, S. & Sharma, S. (2001). "Herd Behavior in Financial Markets." IMF Staff Papers 47(3). https://www.imf.org/external/pubs/ft/staffp/2001/01/pdf/Bikhchan.pdf
4. Shiller, R. (2000). Irrational Exuberance. Princeton University Press. https://press.princeton.edu/books/paperback/9780691173122/irrational-exuberance

Disclaimer

This article is educational content only and is not financial advice. Nothing here is a recommendation to buy, sell, or hold any security. Consult a licensed advisor before making investment decisions.