Anatomy of a Bubble: Common Patterns

The pattern catalogue is real: the same stages, ingredients, and rationalisations appear from Amsterdam's taverns to Reddit's feeds, and financial history's most-cited scholars built a durable model of them. The guardrail is equally real, and it is the documented record's own: pattern recognition in hindsight and bubble identification in real time are different abilities, the second one has defeated nearly every documented contemporary who claimed it — and a respected wing of academic finance disputes that "bubble" is even a well-defined real-time concept. Both halves are the education. Per the hub's standing rule, stated here at maximum strength: this article teaches historical literacy — vocabulary for understanding arguments and reading history — and is not a crash-prediction toolkit, a market-timing method, or a commentary on any present market.

The framework: Minsky's stages, Kindleberger's history

The canonical model comes from Hyman Minsky's financial-instability theory as organised by economic historian Charles Kindleberger in Manias, Panics, and Crashes (1978) — the book that catalogued three centuries of episodes and found one recurring arc, in five stages. Displacement: a genuine novelty changes what seems possible — a technology (railways, internet), a policy regime (cheap credit), a financial innovation (securitisation, joint-stock debt conversion), a reopened world. Every documented episode in this pillar begins with something real. Boom and credit expansion: rising prices attract money, and — Minsky's signature insight — the financing degrades as it grows: cautious borrowing gives way to lending that depends on prices continuing to rise, whether as 1929's 10%-margin call-money machine, 2008's teaser-rate refinancing assumption, or 1720's instalment plans and loans against shares. Euphoria: the stage with the tells — valuation metrics get replaced when the old ones stop justifying prices (eyeballs for earnings, rarity for yield, attention for revenue); "this time is different" reasoning becomes respectable; new participants arrive in waves (the shoeshine-boy folklore of 1929, the day-trading culture of 1999, the lockdown account surge of 2020); and prices are sustained by expected resale rather than by cash flows — the definitional core every version of the model shares. Distress: insiders and early money begin selling; credit tightens; the marginal buyer thins — visible mostly in retrospect. Revulsion: the self-reinforcing machinery runs in reverse — forced selling, credit withdrawal, the feedback anatomy this pillar has documented in both directions — followed, in the full historical arc, by investigation and architecture.

The five recurring ingredients — a cross-episode table

Beneath the stages, five ingredients recur across the documented cases; the table maps them.

Two honest annotations, per the record. The ingredients are diagnostic of the past, not predictive of the future: every documented bubble had them, but so have booms that never collapsed — the ingredients' presence is far more common than the outcome, which is precisely the identification problem the next section takes seriously. And the table's tidiness is partly survivorship: episodes get into the canon because they crashed, so the catalogue documents what bubbles looked like, not what proportion of similar-looking moments became bubbles — the base-rate blindness the behavioural article warns about, operating on financial history itself.

The guardrail: why recognition is not prediction

Three documented facts discipline everything above. First, the timing record: the pillar's episodes are strewn with correct diagnoses that were catastrophically early — "irrational exuberance" preceded the dot-com peak by more than three years and a near-quadrupling; documented short sellers of famous bubbles were carried out before being vindicated; and being early, in leveraged markets, is functionally identical to being wrong. Recognising euphoria's features and dating its end are different problems, and the record shows the second defeating even the era-defining experts of each episode. Second, the definitional dispute, reported two-sided per house rule: a substantial academic tradition — the efficient-market school, with Eugene Fama its most prominent voice — disputes that "bubble" is a well-defined real-time concept at all, arguing the label is applied reliably only after crashes and that predicted bubbles fail to materialise often enough to make the term unfalsifiable ex ante; against this, the behavioural and historical tradition (Shiller's Irrational Exuberance, the Kindleberger canon) holds that identifiable excess exists and documents it — a genuine, unresolved scholarly argument in which this portal, as always, referees nothing and reports both corners. Third, the asymmetry of the word itself: "bubble" is a conclusion wearing the costume of an observation — deployed in real time, it is a forecast that prices will fall, which is exactly the category of claim this portal never makes and this pillar's hub rules out. What the framework is for, then, stated plainly: reading history with structure (the stages organise every episode in this pillar); understanding arguments (when a commentator calls something a bubble, the reader now knows which ingredients they're claiming and which school they're standing in); and recognising mechanisms — credit dependence, metric substitution, resale-based pricing — as things that exist and matter, without pretending their presence carries a date. The pillar's closing article completes the synthesis: what all of this teaches about risk itself.