Bubbles are not an invention of modern finance. Long before candlestick charts and leveraged ETFs, human nature had already run the same play many times over. This week we return to the earliest great bubbles in financial history—the Dutch tulip, the French Mississippi, the English South Sea—to see how they reuse the same script across different nations and different assets, then distill the common anatomy of a bubble through the Kindleberger-Minsky lens. The goal is not to predict the next bubble, but to learn, while inside one, to recognize which act you are standing in.
The Principle
When a thing's price is sustained solely by "the next person will pay more," and is fully detached from the cash flows it can generate, you are no longer buying an asset—you are betting on the greater fool.
Source · Quote
"Men, it has been well said, think in herds; it will be seen that they go mad in herds, while they only recover their senses slowly, and one by one."
— Charles Mackay, Extraordinary Popular Delusions and the Madness of Crowds, 1841
Deeper Reading
The tulip was a rare variety newly introduced to Europe, its dazzling streaks caused by a virus, its supply naturally scarce and hard to value—the absence of a valuation anchor is the very seedbed of a bubble. Trading soon detached from the physical: people bought and sold "futures contracts" on bulbs still in the ground, due to flower the following year (called windhandel, literally "trading in wind"). When the sole basis of price is resale, one broken link in the chain and value evaporates instantly.
The Case
From the winter of 1636 into early 1637, the most prized Semper Augustus bulb reportedly quoted at several thousand guilders—while a skilled craftsman earned about 300 guilders a year. A single bulb was worth a house on the canal. In February 1637, an auction in Haarlem suddenly drew no bids; panic spread down the contract chain, and prices collapsed more than ninety percent within days, leaving a field of contracts no one could honor.
Limits · Checklist
Limit: Mackay's account was exaggerated by later retellings. Historian Anne Goldgar (2007) found in the archives that far fewer were actually ruined than legend holds—most were speculative contracts requiring no delivery; Peter Garber further argues the high prices of some rare bulbs were not wholly irrational at the time. The transferable lesson is not "everyone went broke," but that once price detaches from cash flow, valuation is nothing but story.
- Behind this price, is there estimable cash flow or use-value—or only resale?
- Is my reason for buying essentially "someone will pay more"?
- If the market froze tomorrow and I couldn't sell, would I hold it for the long run?
- How much of the get-rich story I've heard survives a check against the original record?
Essence · Reflection
When the only basis for valuation is "the next person," what you hold is not an asset but a wager on when the crowd sobers up.
Look back at a purchase you most wanted to chase higher: if the rule were no selling for five years after buying, would you still buy? The answer tells you whether it was investing or a relay race.
The Principle
The fiercest fuel of a bubble is not greed but credit expansion. When freshly printed money has nowhere to go, it floods into some class of asset and pushes prices to heights wholly unrelated to real returns.
Source · Quote
"Money is not the value for which goods are exchanged, but the value by which they are exchanged."
— John Law, Money and Trade Considered, 1705
Deeper Reading
The Scotsman John Law was both a pioneer of paper-money theory and the operator of history's first monetary experiment. For a debt-laden France he designed a machine: with one hand the royal bank printed banknotes, with the other the Mississippi Company used the narrative of a North American trade monopoly to soak up those notes into its stock. Money-printing and stock-buying formed a self-reinforcing loop—the more money printed, the higher the stock rose, and yet no real American gold mine ever existed to make good on the promise. This is the oldest specimen of "easy money → asset bubble."
The Case
The Mississippi Company's shares soared from about 500 livres in early 1719 to about 10,000 livres by year end—a twentyfold rise that coined a new word, millionnaire. Yet the Louisiana colony was all but unprofitable. In 1720, when holders began demanding to convert notes into coin, the loop reversed in an instant: bank runs, hyperinflation, a collapsing stock. Law fled France at year's end, and French trust in paper money and joint-stock companies was frozen for nearly a century.
Limits · Checklist
Limit: not every round of easy money must breed a bubble—low rates can merely support valuations mildly; what matters is whether credit floods into assets lacking a cash-flow anchor and is accompanied by self-reinforcing leverage. The misuse is treating "central-bank liquidity" as a permanent license to be long. The real signal is when price gains lean more and more on liquidity itself rather than underlying earnings.
- Is this rally earnings growing, or merely liquidity swelling?
- Is the rise self-reinforcing (up → more leverage → further up)?
- If credit tightens and rates rise, does this valuation still stand?
- Am I buying on verifiable returns, or on a grand story not yet delivered?
Essence · Reflection
When prices rest more on "there's more money" than on "it earns more," you are standing on liquidity—and liquidity can ebb at any time.
Examine your best-performing holdings: how much of the rise came from the business getting better, and how much from there being more money in the market? That ratio is your margin of safety.
The Principle
Bubbles do not screen for IQ. However brilliant the mind, once captured by envy at watching others get rich fast, reason yields to the fear of missing out.
Source · Quote
"I can calculate the motion of heavenly bodies, but not the madness of people."
— Attributed to Isaac Newton (after his South Sea loss; authenticity disputed, widely repeated)
Deeper Reading
The South Sea Company's core was not trade but a piece of financial alchemy: swapping ever-rising shares of its own stock for British government debt. The higher the share price, the better the debt-for-equity conversion, so the company was motivated to inflate the price further with exaggerated South American prospects—one of history's earliest "stock-price-as-fuel" self-referential frauds. The frenzy spawned countless imitators, one of whose prospectuses reportedly read: "carrying on an undertaking of great advantage, but nobody to know what it is."
The Case
South Sea stock rose from about £128 in January 1720 to about £1,000 in August, then crashed back to about £124 by year end. Newton took an early profit, but at high prices could not resist temptation and bought back in, ultimately losing an estimated £20,000—a large slice of his life savings. This disaster spared not even the greatest rational mind humanity has produced.
Limits · Checklist
Limit: "clever people lose too" does not mean "expertise is useless"—Newton's error was not in intellect but in having no preset discipline to resist FOMO; conservatives who held to valuation that same year were unharmed. The defense is never being smarter, but rules written in advance and not revised in the moment.
- How much of my urge to buy comes from the anxiety that "others are making money"?
- Do I have a discipline line set beforehand that I won't raise just because prices rose?
- Do I understand how this asset makes money, or only that "it's going up"?
- If everyone around me is buying, is that a signal to be more cautious or more excited?
Essence · Reflection
A bubble pierces not intelligence but discipline—Newton lost because he could compute the heavens yet set no gate on his own greed.
Write down the one opportunity you most fear missing right now, and the exit rule you've set for it. If the rule is blank, envy is already making the decision for you.
The Principle
Three centuries, three nations, three assets—yet bubbles reuse the same skeleton: displacement → credit boom → euphoria → insider selling → panic. Recognizing the stage is more workable than predicting the top.
Source · Quote
"There is nothing so disturbing to one's well-being and judgment as to see a friend get rich."
— Charles Kindleberger, Manias, Panics, and Crashes, 1978
Deeper Reading
Building on Minsky's financial-instability hypothesis, Kindleberger split a bubble into five acts: ①Displacement (a real change—new lands, new technology—resets expectations) → ②Boom (credit expansion fuels it) → ③Euphoria (price detaches from fundamentals, "this time is different" becomes the slogan, and swindles cluster) → ④Insider selling (the well-informed quietly exit) → ⑤Panic (confidence reverses, selling self-reinforces). It is the same thing as Howard Marks's "pendulum" and the positive-feedback phase transition of complex systems.
①Displace
②Boom
③Euphoria
④Selling
⑤Panic
The Case
Fit the three bubbles into the frame and the skeleton is strikingly identical: displacement was a rare new variety, new-world trade, and debt restructuring; boom was fueled by futures contracts, royal banknotes, and debt-for-equity; euphoria produced distorted signals like "a bulb worth a house" and "a prospectus for no one knows what"; panic wiped out ninety percent within months. 1929, dot-com 2000, and subprime 2008 map act-by-act just as well—the asset changes, the structure does not.
Limits · Checklist
Limit: the framework explains perfectly in hindsight yet cannot time precisely—"euphoria" may last years, and shorting too early often bankrupts you first (Keynes: the market can stay irrational longer than you can stay solvent). Its proper use is not predicting the top but calibrating position and emotion: on reaching Act ③, cut leverage, hold cash, refuse "this time is different"—rather than betting the farm on the turn.
- Which of the five acts does the current market match? On what evidence?
- The "this time is different" case—am I testing it, or echoing it?
- Are insiders (management, early investors) adding or trimming right now?
- Do my position and leverage match the act I judge us to be in, leaving room for "longer than expected"?
Essence · Reflection
The assets in a bubble are refreshed every era; the skeleton has not changed in three hundred years—recognizing your act beats vainly predicting the day of collapse.
Pick a hot field you're watching and check it against the five acts. Which act is it in? And more crucially: do your cash and leverage levels earn the right to that judgment?
Further Reflection
If the five-stage structure of bubbles has been so stable for three hundred years, why do rational investors still fall into them again and again—is it a lack of knowledge, or the structure itself?
Mainly the structure. A bubble is a social game: in Acts ② and ③, following the crowd and chasing higher is often the individually rational optimum—not participating means watching peers get rich and clients walk away. Keynes's "beauty contest" nails it: clever people are not judging value but judging "how others will judge." So even if everyone has read the history of bubbles, incentives keep pushing the system toward the top. Knowledge lets you recognize the stage but cannot rewrite the incentives that draw you in—the real moat is discipline set in advance and externalized.
Kindleberger says a bubble begins with a real "displacement." So how do you tell early on a "genuine world-changing shift" from a "bubble wrapped in the narrative of one"?
The difficulty is precisely that both share the same true core—railroads, the internet, AI are all real shifts, and the bubble parasitizes on them. The dividing line is not "is the technology real" but "how much is already priced in": does the growth implied by today's valuation exceed any reasonable path? Do gains lean more and more on new entrants and leverage than on cash flow? Has financing become so easy that even undifferentiated imitators are chased (the South Sea "no one knows what" prospectus is the classic signal)? A real shift does not make today's price real—after most transformational bubbles burst, the technology still changes the world; the early shareholders are simply washed out first.
As a long-term investor, once you recognize you're in Act ③, how should you weigh "cut leverage, hold cash" against "missing a rally that may run for years"?
The key is to shift the decision from "timing" to "reshaping the odds." You cannot foresee the top, so you shouldn't go all-in or all-out in black and white, but manage exposure asymmetrically: keep a core position so you're not absent, while cutting leverage, raising cash, and refusing to add at highs—preserving some upside while lowering the permanent loss "if it crashes." Marks puts it plainly: "You can be more aggressive or more defensive, but it should be commensurate with where you are in the cycle." The cost is forgoing a slice of the final rally; the reward is that when Act ⑤ arrives, you have cash to be the one greedy while others panic—the greatest returns often come not at the peak but from the rubble.
In today's world of AI and algorithmic trading, where information and capital move in milliseconds, will the five stages be compressed, smoothed out, or amplified?
More likely compressed and amplified at once, not smoothed. The human core (greed, envy, fear of missing out) is not rewritten, only accelerated: social media pushes a narrative worldwide in days, leveraged tools let retail pile in instantly, and the arc from boom to euphoria is sharply compressed—2021's meme stocks and crypto's boom-bust cycles ran the full five acts in weeks, not years. Algorithmic herding then amplifies the Act ⑤ stampede (the 2010 flash crash was a rehearsal). The implication: your window to recognize the stage and retreat calmly is shorter, so the discipline must be pre-set as rules rather than judged in the moment.