01 · A History of Financial Bubbles
From Dutch tulips in 1637 to the crypto winter of 2022, humanity replays the same script every few decades: a new narrative ignites imagination → leverage amplifies gains → everyone joins the party → first cracks appear → panic selling and collapse → regulation arrives late. This article dissects eight great bubbles on a timeline, reviewing each from four angles — background, evolution, peak and collapse, and lessons of human nature — before closing with a checklist of rules common to all eight.
Disclaimer: Everything on this site is for learning and research only and does not constitute investment advice. Markets carry risk; invest with caution.
I. The 1637 Dutch Tulip Mania: The First Mass Speculative Frenzy
1.1 Background
In the 17th century the Netherlands was Europe's wealthiest nation, with advanced shipping and finance; Amsterdam was the global capital of capital. Tulips arrived in Europe from the Ottoman Empire in the late 16th century and became a status symbol among the elite for their rarity and striking colors. In the 1630s, bulbs producing rare colorations (such as the flame-patterned "Semper Augustus") began rising in price uncontrollably.
1.2 How It Evolved
- 1633-1636: Rare bulb prices doubled year after year, becoming a "hard currency" among the wealthy.
- Autumn 1636: Speculation spread from aristocrats to commoners — tailors, bakers, and sailors all began trading bulbs. Deals were done not in nurseries but in taverns via "paper contracts" (similar to today's futures), with deferred delivery — in effect turning into margin trading.
- Winter 1636-37: A single ordinary bulb rose tens of times in months; a rare bulb could cost as much as a canal-side mansion. Tulips even appeared in marriage property inventories.
1.3 Peak and Collapse
In early February 1637, at an auction in Haarlem, a bulb unexpectedly drew no bids — panic swept the country within days. Contract prices lost more than 90% within a week, and many contract holders (especially ordinary people who had bought ahead of their means) went bust overnight. The government briefly attempted to settle contracts at 3.5% of face value, but the effort fizzled out.
1.4 Lessons of Human Nature
- Tulip bulbs had real value (ornamental, horticultural), but they were bid up into an asset far beyond any practical use — the shared starting point of every bubble.
- The last holders in a bubble are almost always the slowest to react and the most heavily leveraged ordinary people: professionals quietly exited before the crash.
- The opening line of "this time is different" was already spoken back in 1637.
Takeaways
- When an asset's price becomes severely detached from the cash flows or utility it actually generates, it has turned from an "investment" into a prop in a game of musical chairs.
- You never know which holder you are, so only ever put money you can afford to lose into any "story-driven" asset.
II. The 1720 South Sea Bubble and Mississippi Bubble: Government Backing and Celebrity Effect
2.1 Background
Around 1720, Britain and France emerged from the War of the Spanish Succession buried in debt. Both countries hit on the same idea: swap national debt for shares in a chartered company, converting sovereign obligations into equity.
- France: John Law's Mississippi Company received the Louisiana colony and the right to issue currency, and set up the Royal Bank, printing money on a massive scale.
- Britain: The South Sea Company was granted a monopoly on trade with South America (trade that never actually materialized) and promised to swap its shares for British government debt.
2.2 How It Evolved
- Both companies' shares soared through 1719-1720: South Sea stock climbed from about £128 to roughly £1,000 by August 1720; Mississippi shares at one point rose from 500 livres to above 10,000 livres.
- Celebrity effect: King George I, members of the royal family, and half the members of Parliament held South Sea stock; aristocratic participation was treated as a "status signal".
- Crowds queued to subscribe to new share issues; newspapers began advertising purely conceptual companies "undertaking a great advantage, but nobody to know what it is" (today's vaporware-token white papers).
- Sir Isaac Newton entered near the top in April 1720 and added to his position in May.
2.3 Peak and Collapse
- Mississippi Bubble: In May 1720, John Law's money printing triggered inflation fears and shares began collapsing; by year-end they had fallen about 95% from the peak. France's financial system was crippled and Law fled into exile.
- South Sea Bubble: In August-September 1720, Parliament passed the Bubble Act, compounded by rumors of insider selling; the price collapsed from around £1,000 back to roughly £135 by year-end.
- Newton lost about £20,000 in this bubble (decades of income for an ordinary person at the time), leaving behind his famous words:
"I can calculate the motion of heavenly bodies, but not the madness of people."
2.4 Lessons of Human Nature
- Government backing doesn't make a bubble safer — it makes it bigger: the South Sea Company's trade monopoly and the Mississippi Company's money-printing rights were both authoritative versions of a "story".
- Celebrities buying is not a reason to buy: the king bought, Newton bought — it still crashed.
- After a bubble bursts, victims collectively demand that "the swindlers be caught" — but the first swindlers had already dumped their holdings onto them at the top.
Takeaways
- Distinguish "fundamentals" from "narrative": the South Sea Company never had meaningful trading revenue; the share price ran on narrative alone.
- Authority (governments, celebrities, institutions) entering prolongs a bubble but never changes the ending — it just makes it harder for you to stay clear-headed.
III. The 1929 Wall Street Crash: Margin Buying and a Decade of Great Depression
3.1 Background
The 1920s were America's "Roaring Twenties": automobiles, radio, and electrification drove industrial prosperity, and the Dow rose from about 100 points in 1924 to 381 points in 1929. Margin requirements were extremely low — buying $1,000 of stock required only $100-200 of your own money (roughly 10%-20% margin) — and borrowing to buy stocks became a national pastime: taxi drivers and shoeshine boys discussed stocks.
3.2 How It Evolved
- Between 1928 and 1929, total margin loans surged from $4 billion to $8.5 billion, far beyond the financing needs of the real economy.
- "Investment trusts" (similar to today's leveraged funds) nested layer upon layer, holding each other's shares and multiplying leverage several-fold.
- In September 1929 economic data began weakening, yet the market kept pushing higher on momentum; from mid-October the Dow started sliding day after day.
3.3 Peak and Collapse
- September 3, 1929: the Dow peaked at 381.17 points (it would take about 25 years to be exceeded again).
- October 24 (Black Thursday): panic selling; volume hit 12.9 million shares (an enormous figure then). A bank syndicate propped up the market and briefly stemmed the fall.
- October 29 (Black Tuesday): the market collapsed entirely; single-day volume reached 16.3 million shares and the Dow fell about 12% (-11.7% on the day).
- The crash continued for three years: by July 1932 the Dow stood near 41 points, down about 89% from the peak.
- Leverage amplified everything: a 20% drop wiped out buyers on 20% margin, and forced liquidations drove further declines — history's first nationwide "forced-liquidation spiral".
- In the ensuing Great Depression (1929-1933), U.S. GDP shrank about 30%, unemployment rose to about 25%, and over 9,000 banks failed.
3.4 Lessons of Human Nature
- Leverage doesn't change direction; it only amplifies swings: the economy was genuinely growing before the 1929 crash, but 10x leverage turned a "reasonable correction" into "collective liquidation".
- When the shoeshine boy gives stock tips and taxi drivers discuss the market — breadth of public participation is the most reliable folk indicator of a bubble top.
- The pain after a crash lasts not one year but a generation: the Great Depression left an entire generation of Americans psychologically scarred toward stocks.
Takeaways
- The Glass-Steagall Act, the Securities Exchange Act of 1934, and the Securities Act of 1933 were all "regulatory legacies" of this crash — every major bubble comes with a round of regulatory catch-up.
- In an environment where margin ratios let you "borrow nine to buy one", you are no longer buying stocks — you're buying time until your luck runs out.
IV. The 1989 Japanese Bubble: Real Estate and Stocks, and the "Lost Decades"
4.1 Background
After the 1985 Plaza Accord sent the yen sharply higher, the Bank of Japan cut rates repeatedly to offset pressure on exports, creating an extremely loose monetary environment. Money flooded into stocks and real estate, amplified by the national belief in the "land myth" (land only goes up), and Japan entered an era of mass speculation.
4.2 How It Evolved
- From 1985 to 1989 the Nikkei rose from about 13,000 to nearly 40,000 points — tripling in four years.
- Land prices in Japan's six largest cities (Tokyo, Osaka, Nagoya, etc.) tripled in four years. In 1989, the assessed value of the Tokyo Imperial Palace grounds was said to exceed that of the entire state of California.
- Corporations parked cheap borrowings in stocks and real estate as "treasury operations"; banks lent against land almost without limit. Financial engineering became standard practice while operating profit stopped mattering.
- At the peak of the bubble in 1990, land in Tokyo alone was estimated to be worth more than all the land in the United States (at prevailing exchange rates).
4.3 Peak and Collapse
- December 29, 1989: the Nikkei touched its all-time peak of 38,957 points.
- From 1990 the Bank of Japan raised rates successively and the Ministry of Finance capped real estate lending; the bubble was deliberately punctured. Stocks fell about 40% within a year, and land prices entered a grind lower lasting more than a decade.
- From peak to the 2003 low near 7,600 points, the Nikkei fell more than 80%; it did not reclaim 40,000 points until 2024 — a full 34 years.
- Bad loans piled up; across the decades Japan saw repeated bank failures and mergers (e.g., Hokkaido Takushoku Bank and Yamaichi Securities in 1997); "zombie companies" survived on rollovers, dragging down productivity.
- Household wealth shrank dramatically, consumption stagnated for years, and young people lost hope of advancement — this is the "Lost Decades".
4.4 Lessons of Human Nature
- Sustained central bank easing is the best incubator of bubbles — and their strongest ripening agent: liquidity doesn't stop on its own; it stops when the brakes are applied, and by then it's too late.
- "XX can only go up" is the national-level illusion of every bubble: Japan believed in land, China believed a similar narrative, America believed in house prices, crypto believes in Bitcoin.
- A bubble that is deliberately punctured hurts even more: decades of slow decline do far more damage than one violent year.
Takeaways
- Tops cannot be predicted in time, but "when you buy determines when you retire": those who bought the Nikkei in 1989 waited 34 years to break even — compounding only works if your principal survives.
- When "borrowing to buy assets" becomes society's default truth, assume it is nearing its end.
V. The 2000 Dot-com Bubble: Concept Hype and "Price-to-Dream Ratios"
5.1 Background
The commercialization of the internet began in the 1990s (Netscape doubled on its IPO day in 1995), and with a rate-cutting cycle, venture capital flooded in. ".com" became the password to wealth: simply appending ".com" to a company name unlocked funding, listings, and surging share prices.
5.2 How It Evolved
- From 1995 to 2000 the Nasdaq rose from about 1,000 to 5,048 points — roughly quadrupling in five years.
- Valuations entered the "price-to-dream" era: P/E ratios lost meaning, replaced by "revenue per click", "eyeballs", and "burn rate". Amazon was still losing money in 1999 yet commanded a market cap of tens of billions.
- Typical concept stocks: Pets.com (online pet supplies, unbounded losses), eToys (online toy store, quadrupled on its 1999 IPO day and at one point worth more than Toys "R" Us) — zero profitability, valuation powered entirely by imagination.
- Buybacks, option compensation, and bull-market media narratives reinforced each other; retail and institutional investors chased the market up together.
5.3 Peak and Collapse
- March 10, 2000: the Nasdaq peaked at 5,048.62 points.
- From March, the Microsoft antitrust ruling, rate-hike expectations, and earnings blowups triggered selling; within two and a half years the Nasdaq fell to about 1,114 points by October 2002 — down about 78% from the peak.
- Masses of companies delisted or went bankrupt: Pets.com survived just 268 days after listing and became a textbook case; countless option employees and retail investors who bought the top were wiped out.
- Star funds of the era (such as the USAA-managed funds run by Keith Bannon) lost more than 90%; per the Wall Street Journal, roughly $5 trillion of U.S. market value evaporated between 2000 and 2002.
- Survivors like Amazon and Google grew into giants atop the rubble — a bubble isn't wrong per se; what's wrong is paying too high a price inside it.
5.4 Lessons of Human Nature
- The essence of the "price-to-dream ratio" is anchoring valuation to storytelling ability rather than cash flow — and a story's shelf life is always shorter than a bubble's cycle.
- The new technology was real (the internet genuinely changed the world), but "the technology is real" and "the price is reasonable" are two different things — confusing them was the biggest cognitive error of the bubble era.
- Media coverage and new-issue wealth effects create a self-consistent loop of "the more it rises, the more justified it seems" — until new money stops arriving.
Takeaways
💡 How to Carry Yourself in a Bubble Era
Either get in extremely early or buy extremely cheap — never take the baton in the middle stage "when everybody is talking about it". Value a company by asking "can it still make money ten years from now" instead of "how much can it rise next year"; this is the only vaccine against the price-to-dream ratio.
VI. The 2007-08 Subprime Crisis: MBS/CDO and the Chain of Leverage
6.1 Background
The Fed kept rates low for years after 2000; U.S. home prices kept climbing and "house prices never fall" became national faith. Chasing scale and profit, banks issued subprime mortgages to borrowers with poor credit and invented adjustable-rate loans (low teaser rates for two years, then jumping sharply), luring vast numbers of undocumented-income buyers into homes.
6.2 How It Evolved
- Banks packaged subprime loans into MBS (mortgage-backed securities), then assembled CDOs (collateralized debt obligations) and other structured products; rating agencies stamped them AAA, slicing, diluting, and hiding risk layer by layer.
- Wall Street held these products at extreme leverage (some institutions above 30x), while insurers like AIG underwrote CDS (credit default swaps). Risk passed around the globe like a relay baton: European and Asian banks were all buying.
- From 2004 to 2006 U.S. home prices rose over 10% a year and a "flipping" craze swept the country; chasing fees, banks lowered standards to "no down payment, no income proof" (NINJA loans: No Income, No Job, no Assets).
6.3 Peak and Collapse
- Mid-2006: U.S. home prices peaked and turned down; from 2007 subprime defaults soared, and ARM resets ignited a wave of foreclosures.
- August 2007: BNP Paribas froze three funds and global credit markets tightened abruptly — the "credit freeze" began.
- March 2008: Bear Stearns was emergency-sold to JPMorgan at $2 per share (later raised to $10); on September 15, Lehman Brothers declared bankruptcy (see 02 - Famous Crashes and Black Swans); within a week AIG was taken over by the U.S. government and global credit markets froze.
- The full transmission chain replayed: home prices fall → subprime defaults → MBS/CDO writedowns → holding institutions blow up → interbank lending freezes → real-economy financing seizes → recession.
- Global losses from the crisis ran into trillions of dollars; U.S. unemployment rose from 4.7% in 2007 to 10% by October 2009; China launched a four-trillion-yuan stimulus after 2008 to offset shrinking exports.
6.4 Lessons of Human Nature
- The longer the chain of risk transmission, the more people believe "the risk isn't mine" — every link thinks it merely "earns the middle spread", while every link in the system is running naked.
- An AAA rating is not fact; it is a paid-for label. "Everyone else bought it" has never been a reason to buy.
- Structuring risk into smaller slices (CDOs of CDOs) does not diversify it — it hides it; however finely sliced, everything goes to zero together when the underlying assets default.
Takeaways
- For financial products you don't understand (synthetic CDOs, securitized derivatives, complex structures), assume they are toxic.
- True diversification happens at the level of asset class, country, and currency — not the fake kind of "buying ten funds that all hold property debt".
VII. The 2021-22 Crypto Bull and Collapse: DeFi Leverage, Luna, and FTX
7.1 Background
In 2020-2021, post-pandemic global money printing fueled an epic crypto bull market. DeFi offered "stable yields" of 20%-1000% APR under the banner of yield farming; centralized exchanges offered 100x perpetual contracts; new narratives — "Web3", "on-chain world", "institutional adoption" — piled on endlessly.
7.2 How It Evolved
- From the pandemic low of March 2020 (Bitcoin ~$3,800) to November 2021, Bitcoin topped out around $69,000, and Ethereum rose from about $90 to roughly $4,800.
- Stablecoins became leverage fuel: Terra's UST promised 20% APY (via Anchor Protocol); users staked UST for yield, borrowed against it to buy LUNA, and nested the loop again and again.
- Exchanges offered high-leverage contracts plus copy-trading and lending services; across 2021-22 multiple funds (Three Arrows Capital among them) held crypto and DeFi positions at 3-6x leverage.
- Celebrity shilling, NFT avatars, and lending-platform blowups (Celsius froze withdrawals in 2022) kept reinforcing the "crypto is the future" narrative.
7.3 Peak and Collapse
- November 10, 2021: Bitcoin touched roughly $69,000, its all-time high, and the bear market began.
- May 2022, the Luna/UST death spiral: UST depegged (fell below $1) → arbitrageurs dumped → the ecosystem sold LUNA to buy back UST → LUNA collapsed (from about $80 to a handful of decimal places, effectively zero) → UST permanently lost its 1:1 peg. Korean investors alone lost billions overnight; reports of investor suicides shook public opinion.
- June-July 2022: hedge funds such as Three Arrows Capital blew up, and crypto lenders (Celsius, Voyager) froze withdrawals.
- November 2022, FTX collapses: tens of billions of dollars in related-party transactions and misappropriated funds linked Alameda Research and FTX (see 02 - Famous Crashes and Black Swans); under a bank run, FTX went from industry leader to bankruptcy within days, with much of customer funds unaccounted for.
- Across 2022 total crypto market capitalization fell from about $3 trillion to under $1 trillion — a loss of more than 70%.
7.4 Lessons of Human Nature
- A "stablecoin"'s stability rests on other people's confidence: any "risk-free 20% annual yield" means someone, somewhere you can't see, is carrying outsized risk — the high interest you collect is someone else's principal being lost.
- An algorithmic stablecoin's "arbitrage peg" works in normal markets and turns on itself in a panic: the mechanism's correctness assumes the very thing a panic removes.
- A centralized exchange's "segregation of customer assets" is a promise, not a fact: if you can't withdraw it, it isn't yours. FTX's misappropriation was the model working as designed, not an accident.
Takeaways
- For any "stable yield" product whose returns far exceed the risk-free rate, ask three questions first: where does the money come from? who bears the risk? can I withdraw anytime?
- In crypto, whoever holds the private keys holds the assets; use a hardware wallet for large sums, and never deposit more than a "zero wouldn't matter" fraction of your wealth.
VIII. A Checklist of Rules Common to All Eight Bubbles
Put the eight bubbles side by side and the script repeats frame by frame:
| Stage | Tulips 1637 | South Sea/Mississippi 1720 | 1929 | Japan 1989 | Dot-com 2000 | Subprime 2008 | Crypto 2021-22 |
|---|---|---|---|---|---|---|---|
| New narrative | Rare flower = status | National trade privilege | New industrial age | Land myth | Internet revolution | House prices never fall | Blockchain revolution |
| Leverage vehicle | Paper contracts | Debt-for-equity swaps | 10% margin | Bank property loans | Venture capital | MBS/CDO | Contracts + stablecoins |
| Mass participation | Tavern crowds | Subscription queues | Shoeshine-boy tips | Everyone trades stocks and land | Retail + funds | Everyone buys homes | Influencers + retail |
| First crack | Failed auction | Bubble Act | Data weakens | Rate hikes | Earnings blowups | Subprime defaults | UST depeg |
| Crash | -90% in a week | -95% in months | -89% in three years | -80% over thirty-four years | -78% in two years | Two years of global recession | -70% in a year |
| Regulatory legacy | Tulip trading curbed | Bubble Act | Securities/banking laws | Big Bang reforms | Sarbanes-Oxley | Dodd-Frank | Crypto legislation worldwide |
💀 Leverage Is the Only Amplifier of Bubbles
Without leverage, a bubble is just a slow drawdown; with leverage, it becomes a liquidation spiral. Every major bubble came paired with a matching leverage instrument — margin, credit, derivatives, stablecoins.
The common-rules checklist:
- The new narrative must first "sound plausible and grand": tulips meant status, the internet meant revolution, crypto meant freedom — the grander the narrative, the more believers, the bigger the bubble.
- Leverage is the bubble's amplifier: without it, a bubble is a slow drawdown; with it, a liquidation spiral. Every major bubble had its matching leverage instrument (margin, credit, derivatives, stablecoins).
- Mass participation is the top signal: when "people who never invest start investing, and investors start leveraging up," the top is usually only months away.
- Bad news always arrives looking insignificant: a failed auction, one earnings report, a single depeg — the market starts falling while "everything looks normal."
- Crashes are far faster than rallies: three years up, three weeks down is the norm; the wilder the rise, the steeper the fall, and it never leaves you enough time to exit.
- Regulation always arrives after the crash: the Bubble Act, the Securities Act of 1933, Dodd-Frank — history shows rules are mended fences, not foresight.
- After every crash, survivors say "this time really is different": new technology, new mechanisms, new regulation give the next bubble fresh skin — but human nature hasn't changed since 1637.
⚠️ Risk Warning
The historical cases here are for teaching and research only and do not constitute investment advice. Bubbles and crashes are "known in outcome, unpredictable in process": no one can call the top, and attempting a "perfect exit" is itself high-risk behavior. Historical patterns exist to identify risk and size positions — not to catch bottoms or short tops. Leveraged trading can wipe out your capital and even leave you in debt (negative balance); participate only with money you can afford to lose.