Herding Effect in the Stock Market — Examples, Mechanics, Protection
Tulip bulbs 1637, railway shares 1870, telecom stocks 2000, GameStop 2021. The story changes, the mechanism stays: investors buy because others are buying. Herding is evolutionarily smart (in the savannah the herd was the safe place), but in markets exactly wrong — the crowd piles in at the top and runs out at the bottom. Majority behaviour is therefore almost always the worst strategy. Here are the five biggest historical herds, the mechanics, and five filters that keep you out of the swarm.
What herding is and why it works (for the herd)
Bikhchandani/Hirshleifer/Welch described it in 1992 in “A Theory of Fads, Fashion, Custom and Cultural Change” — the concept of the information cascade: each individual rationally decides that observing others provides valuable information. If 100 people say “BUY!”, odds are at least one knows something. Individually rational behaviour produces collectively irrational outcomes.
When following prices means prices rising, the swarm becomes a self-fulfilling prophecy — until liquidity runs out. Classic reflexivity loop (Soros). The top arrives when the marginal benefit of additional buyers is smaller than the realisation tendency of the early buyers.
Five historical herds
- Tulip mania 1636-1637: A single tulip bulb (Semper Augustus) reached the value of an Amsterdam townhouse. Within weeks prices collapsed ~99%. First documented asset bubble in history. Lesson: the tulip wasn’t the problem — the collective belief that the next buyer would always pay more was.
- Railway bubble 1845-1849 (UK): 263 new railway companies founded, many without realistic routes. Share prices rose 100-300%, then 80% of companies went bankrupt or merged. Effect: UK stock market needed 30 years to recover pre-bubble levels.
- Dotcom 1999-2000: Pets.com IPO at $11, bankrupt within 9 months. Webvan: $1bn valuation, bankrupt in 18 months. NASDAQ −78% from March 2000 to October 2002. Buying tech-heavy mid-1999 meant being underwater into 2007.
- Crypto 2017-2018: Bitcoin from $1,000 to $19,000 in 12 months, then back to $3,200 in 12 months. Ethereum from $8 to $1,400 and back to $80. ICO mania: 80% of boom-year ICOs are worthless today.
- GameStop / meme stocks 2021: WallStreetBets, Reddit collective, GME from $4 to $480 in four weeks. Six months later back at $38. AMC, Bed Bath & Beyond, Hertz: identical pattern. Lesson: organised herds can move markets short-term, never long-term fundamentals.
How is herding measured?
| Indicator | What it shows | Warning level |
|---|---|---|
| Lakonishok-Shleifer-Vishny LSV (1992) | How strongly funds simultaneously buy/sell the same stock | > 0.15 = herd behaviour |
| Christie/Huang CSAD (1995) | Cross-sectional dispersion of returns — low = herding | Dispersion > 30% below average |
| Margin debt growth | NYSE Margin Debt to market cap | > 2.5% = late-bubble level |
| Google Trends search volume | Searches for “buy stock”, “crypto” | 3-month high in rating = hype phase |
| IPO queue | IPOs in last 4 weeks | > 20 = top phase likely |
If 3+ indicators are at warning levels, the market is most likely in a herd phase. Anti-herd behaviour is then not “brave” — it is statistically smart.
Five filters against the herd
- Consensus test: Before each buy ask: “What is the consensus on this position?” If 8 of 10 influencers, brokers and forum users are positive — you’re in the herd. Real outperformance comes from positions where 6-7 of 10 are negative and you have a reasoned contrarian thesis.
- Search-volume filter: Check Google Trends — if the asset is hitting search highs, you are late in the swarm. Instead of buying: watchlist, wait 6 months, re-check.
- Sector-IPO filter: If your sector saw more than 5 IPOs in the last 60 days (cannabis 2018, EV 2021, AI 2024), top phase is likely. Hold existing positions, no new ones.
- Anti-FinTwit day: One full day per week without Twitter / YouTube / Reddit. That day is when you make investment decisions — eliminates the social cascade.
- Counter-cyclical pre-commit: Write down what you do before a crisis — not during. “If S&P drops 20%, raise equity allocation by 5 points.” You cannot invent these rules under stress.
How much do anti-herd strategies earn?
Over 30 years, wealth doubles every ~12 years instead of every ~16 years — anti-herd behaviour is the most expensive mental skill retail investors can train.
Pros & cons: total contrarian
- Statistically better risk-reward profile
- Lower drawdowns in crashes because you didn’t enter overbought
- Counter-cyclical buys are cheap because they happen during fear
- Some trends are real — Apple 2010-2024, NVIDIA 2023-2025
- Pure contrarian without fundamentals = falling-knife trap
- Permanent anti-position = severe underperformance in bull markets
The asymmetry: not always against the herd, but at the top against the herd, at the bottom with the non-existent anti-herd. Rule of thumb: defensive when 3+ warning indicators trigger. With 0 warning indicators, keep running the standard savings plan.
Common questions
What is the difference between herding and momentum?
Momentum investing is rule-based trend-following with defined exits. Herding is unstructured following without an exit plan. Both start similarly (stock rises → buy more), but momentum has built-in stops, herding has “to the moon”. So momentum is statistically profitable, herding is not.
Are index funds a herd?
Debated. Index investing is rule-based (market weights), not story-driven. But when 50%+ of US stocks are index-held (2024 ~60%), a collective reduction in index exposure can amplify market reactions despite the rules. No crash trigger so far, but on regulators’ radar.
How do I spot a hype top?
Classic signals: 1) Mainstream media on retail-investor wins, 2) Influencers in premium magazines / TV, 3) Margin debt at multi-year high, 4) Reflexive price moves without news, 5) IPO wave in the sector. When 3+ align — go defensive.
Does passive investing help against herding?
Indirectly yes. Pushing the same monthly amount into MSCI World ignores all swarm signals by design. Weakness: savings-plan investors still panic-sell in crises. Solution: do not manually pause the savings plan when markets are red.
What is “The Big Short” success in herding terms?
Burry, Eisman, Cornwall were strict anti-herd in 2007 — holding against consensus that US housing was safe. Reward: 100-1000% returns. Cost: years of painful pre-positioning, high transaction costs (credit default swaps), and the risk that the crash thesis simply doesn’t materialise.
How do I practice anti-herd behaviour?
Small steps: 1) Maintain a watchlist of stocks no one wants (52-week low, sector unloved). 2) Allocate 5% of portfolio there. 3) Three-year horizon. The learning effects from this setup are higher than from mainstream allocation.
Related Hubs: Investor Glossary | Legendary Quotes
Try TradingView Free for 30 Days
Plus get a $15 discount on your first subscription through this link.