Week 36 Day 1: Herding: The Instinct to Follow the Crowd

In 1999, your neighbor, your barber, and your taxi driver were all buying tech stocks. You bought too. In 2008, everyone was panicking and selling. You sold too. Both times, understanding that you were herding would have saved you a fortune: buy when nobody wants stocks, sell (or hold) when everyone is euphoric.

Herding in market history: (1) The Tulip Bubble (1637). Dutch tulip prices rose 2,000% in months because everyone was buying. When the first sellers appeared, the crowd reversed and prices crashed 99%. (2) The South Sea Bubble (1720). Isaac Newton invested early and made a profit. Then he watched others make even more. He re-invested at the peak and lost 20,000 pounds (approximately $4 million today). 'I can calculate the motion of heavenly bodies, but not the madness of people.' (3) The Dot-Com Bubble (1999-2000). The NASDAQ rose 400% in 5 years. Everyone was a stock market genius. It crashed 78%. (4) The Housing Bubble (2006-2008). 'Real estate never goes down.' Everyone was flipping houses. It crashed 30-50% in major markets. (5) The Meme Stock Mania (2021). GameStop rose 1,500% because of Reddit herding. Most late buyers lost 70%+. (6) The Crypto Bubble (2021). Bitcoin to $69,000, driven by social media FOMO. It crashed to $16,000. The pattern: herds form during rising prices, accelerate into euphoria, reverse at the peak, and stampede during the crash. How to not herd: (a) have a pre-set investment plan, (b) automate contributions so they happen regardless of market sentiment, (c) ignore financial media during volatile periods, (d) never buy something because 'everyone is buying it.'

Herding in financial markets was modeled by Banerjee (1992) and Bikhchandani, Hirshleifer, and Welch (1992) as an 'information cascade': when an investor observes others buying, they infer that others possess positive private information. If the inferred information outweighs the investor's own negative private information, the rational response is to ignore private information and follow the crowd. Once a cascade starts, all subsequent investors follow regardless of their own information -- making the cascade fragile (built on thin informational foundations) and vulnerable to sudden reversal. Scharfstein and Stein (1990) showed that professional fund managers herd because of career concerns: a manager who deviates from the consensus and fails is fired ('stupid and wrong'), while a manager who follows the consensus and fails is retained ('wrong but smart enough to agree with everyone'). This career incentive creates institutional herding that amplifies price movements. Empirically, Wermers (1999) documented significant herding among mutual fund managers, particularly in small stocks and during periods of market stress. Nofsinger and Sias (1999) found that institutional herding is positively correlated with contemporaneous returns (pushing prices up during buying herds and down during selling herds) but negatively correlated with subsequent returns (herded stocks mean-revert), confirming the pattern of bubble and crash. The anti-herding defense for individual investors: (1) dollar cost averaging eliminates the decision of WHEN to invest (immune to crowd timing), (2) index investing eliminates the decision of WHAT to invest in (immune to crowd stock selection), (3) automation eliminates the emotional trigger that initiates herding behavior.