Herding, FOMO and Bubbles

Humans are herd animals wearing trousers. When everyone around you starts buying something — a stock, a coin, a tulip — your brain whispers a very old, very reasonable thought: they must know something I don’t. Sometimes they do. That’s the trap. Herding isn’t a sign of stupidity; it’s often a perfectly sensible shortcut that, scaled up to millions of nervous people, can inflate a price into the stratosphere and then drop it like a piano. This lesson is about why crowds form, why you feel the pull, what a bubble actually looks like from the inside — and the uncomfortable truth that “do the opposite of the crowd” is not a strategy, it’s just a different way to be wrong.

Herding and social proof — copying the crowd as a shortcut

Social proof is the tendency to treat other people’s behaviour as evidence about what’s correct. If you walk into a foreign city hungry and see one restaurant packed and the one next door empty, you pick the busy one — without tasting either. You’re outsourcing the judgment to the crowd. That’s not dumb; it’s efficient. Reading every review and inspecting every kitchen would take all night. The crowd is a cheap, fast information shortcut.

Herding is social proof in motion: many people copying many other people until the behaviour becomes self-reinforcing. In markets it looks like buying because the price is going up, and the price is going up because people are buying.

Here’s the crucial, honest point the bible insists on: herding is often rational. If a thousand strangers each glimpsed a little real information and acted on it, the resulting crowd genuinely aggregates that information — this is roughly why prices are usually pretty smart. The problem isn’t that copying is stupid. The problem is that copying stacks: the signal that “everyone is buying” gets counted over and over, even when it traces back to almost no real information at all.

Worked example — when the crowd is real information vs. just an echo

Imagine 100 analysts each independently study a company. 60 conclude it’s a good buy, 40 disagree. If they all vote at once, the 60–40 split is a genuine, information-rich verdict — the crowd is wise.

Now run it differently: they decide one at a time, in public, and each can see what the people before them did. Analyst 1 (a marginal “buy”) buys. Analyst 2, seeing one buy and feeling 50/50, buys to match. Analyst 3 now sees two buys and thinks, “two before me bought; my own weak doubt isn’t enough to override that” — and buys, ignoring their private signal entirely. By analyst 10, everyone is buying because everyone before them bought. The 40% who privately disagreed never get counted. The same 100 people, the same information — but the public, sequential version throws most of the information away. That’s the difference between a wise crowd and a stampede.

When it matters

Whenever you catch yourself doing something because others are — buying a hot asset, avoiding an unloved one, piling into a “consensus” trade — pause and ask: is the crowd’s behaviour adding new information, or just echoing the last person’s? In fast-moving, emotional markets, it’s usually the echo.

Information cascades — when rational people throw their own signal away

This is the engine under the hood, and it has a name: the information cascade, formalised by Bikhchandani, Hirshleifer and Welch (1992). A cascade happens when people, acting in sequence and able to observe each other, rationally ignore their own private information and simply imitate those who went before.

The key word is rationally. Each person is doing something sensible — “lots of people before me chose A; their collective judgment probably outweighs my one weak hunch, so I’ll choose A too.” But once enough people reason that way, everyone stops contributing new information and the herd marches on whatever the first few happened to do. A cascade is fragile precisely because it rests on so little: a tiny piece of fresh news can flip the whole line.

The restaurant-queue analogy

Two empty restaurants, side by side. The first couple to arrive, knowing nothing, pick the one on the left — maybe its door was closer. The next group sees one place with people and one empty, and reasonably joins the busier one. Now there’s a queue on the left and a ghost town on the right. Every newcomer “reads” the queue as quality and joins it. By 9 p.m. the left restaurant has a 40-minute wait and the right one — which might have the better chef — sits empty. Nobody behaved stupidly. Each person made a reasonable inference from the queue. But the queue was built almost entirely on the accident of where the first couple sat. That’s an information cascade: a tower of imitation balanced on a pebble of real information.

Worked example — the cascade in a market

A new biotech IPOs. The first big fund buys (it has a genuine, mildly positive signal). Financial media reports “smart money is buying.” The next funds, many of whom were neutral, buy so as not to be left out and not to look foolish to clients. Retail investors see institutions and headlines piling in and follow. Within weeks the stock has quadrupled — and the only real information in the entire chain was the first fund’s mildly positive signal. Everyone after that was reading the queue. When a single trial result disappoints, the cascade reverses just as fast, because there was never much underneath it.

FOMO — when a rising price becomes a recruiting tool

FOMO — the fear of missing out — is the specifically painful emotion of watching other people get rich while you sit on the sidelines. It turns a rising price into a marketing campaign: every new high is a fresh advertisement, and every friend’s gain is a needle of regret.

Notice how this connects to the loss aversion you met in Prospect Theory. Missing a gain everyone else is grabbing doesn’t feel like a neutral “no change” — your brain reframes it as a loss relative to the crowd’s new, higher reference point. And losses, as you know, hurt about twice as much as equal gains feel good. So watching your neighbour double their money while you didn’t isn’t mild envy; it’s processed as real pain, and pain demands action.

Worked example — the regret math that pulls you in

A coin you eyed at $10 is now $40. Your System 1 doesn’t compute “expected future return from $40.” It computes the counterfactual: if I’d put in $5,000, I’d have $20,000 right now. That phantom $15,000 “loss” feels vivid and specific, while the abstract risk of buying at $40 feels fuzzy and distant. The asymmetry — concrete regret vs. fuzzy risk — is exactly what FOMO exploits to get you to buy precisely when the price is highest and the risk is greatest. The recruiting poster works best at the top.

When it matters

FOMO is at its most dangerous when three things coincide: a fast-rising price, a vivid story of people getting rich (social media is rocket fuel here), and a vague sense that this is your last chance. That cocktail — price + envy + urgency — is the signature of the late, manic stage of a bubble, which is exactly where we go next.

The anatomy of a bubble

A bubble is when an asset’s price detaches from any reasonable estimate of its underlying value, driven up by the self-reinforcing loop of herding, cascades, and FOMO — until the buyers run out and it collapses. Bubbles aren’t a fringe theory; they’re recurring historical fact. The phases below repeat with eerie consistency across four centuries, because the psychology driving them doesn’t change.

Step through the phases. Watch the price (solid line) pull away from fundamental value (the dashed line), and read the crowd-psychology note at each stage.

The classic phases, and the psychology that powers each:

The historical roll-call

The same movie, different costumes:

• Tulip mania (1630s, Netherlands). Rare tulip bulbs traded for the price of a house, then collapsed. The original cautionary tale of a price unmoored from any use-value.
• The South Sea Bubble (1720, Britain). Shares in a trading company with vague prospects rocketed on hype and government entanglement, then crashed — wiping out fortunes. Even Isaac Newton lost a packet, reportedly sighing that he could calculate the motions of the heavens but not the madness of men.
• The 1929 crash (USA). A frenzied, leveraged stock boom in the Roaring Twenties — ordinary people buying “on margin” (with borrowed money) — gave way to the crash and the Great Depression.
• The dot-com bubble (~2000). Internet companies with no profits (sometimes no revenue) commanded enormous valuations on the story that “the internet changes everything.” The internet did change everything; the valuations were still nonsense, and the Nasdaq fell roughly 78% from its peak.
• The 2008 housing bubble (USA + global). House prices were treated as if they could only ever rise; mortgages were handed out freely and repackaged into securities. When prices fell, the whole edifice — and much of the global financial system — came down with it.
• Crypto cycles (2017, 2021, …). Repeated, vivid boom-and-bust cycles in cryptocurrencies and tokens, with textbook mania, FOMO, and capitulation phases compressed into months rather than years.

Worked example — spotting the detachment

In early 2000, a profitless dot-com might trade at a market value of $10 billion while generating $50 million in annual revenue (not profit — revenue) and burning cash. To justify $10 billion, you’d need the company to eventually earn, say, $1 billion in profit a year forever — a 20× leap, from a company losing money. The number only “made sense” if you assumed the rise itself would continue. That circularity — the price is justified because the price keeps rising — is the tell. When the justification for a price is the price’s own momentum, you’re looking at the mania phase.

The contrarian trap — going against the crowd isn’t automatically right

Here’s where most “smart” beginners faceplant. Having learned that crowds can stampede and bubbles are real, they draw the wrong lesson: so I should just do the opposite of the crowd. This is the contrarian trap, and it is every bit as lazy as blindly following the herd — it just feels cleverer.

Three hard truths puncture it:

1. The crowd is usually right. Markets aggregate information well most of the time. Betting against the consensus as a default rule means betting against the truth as a default rule. Reflexive contrarians lose money continuously in exchange for being occasionally, gloriously right.

2. Bubbles are obvious only in hindsight. When you’re inside one, the “this is clearly a bubble” call and the “this is a genuine new paradigm” call look identical in real time — both are confident, both have smart people on each side. The internet really did change everything; some 2000-era valuations were insane and some early-stage giants were actually undervalued. Hindsight makes the line look crisp; it never was.

3. “The market can stay irrational longer than you can stay solvent.” This famous line (attributed to Keynes) is the killer. Even if you correctly identify a bubble, betting against it — shorting — can bankrupt you before you’re proven right, because the mania can run far higher and far longer than seems possible. Being early is, financially, indistinguishable from being wrong.

Recap

A mixed recap pulling from the whole lesson:

Key Takeaways