Devil Take the Hindmost: Lessons from 400 Years of Speculative Manias
30 USD for Edward Chancellor’s book (Devil Take the Hindmost: A History of Financial Speculation) was a little pricey, so I got ChatGPT to help me understand the key takeaways from his book as well as understand a little more of the past stock market crashes. This is a pretty interesting read, and helps caution us in the highly speculative stock market of today.
I can’t reproduce Edward Chancellor’s exact voice, but I can write an original account with similar broad qualities: financial history told as narrative, attention to characters and institutions, and an emphasis on the recurring psychology and mechanics of speculation.
What makes these episodes interesting is that the object of speculation was usually not nonsense. Tulips really were scarce luxury goods. Railways really did transform civilization. The American economy really was revolutionized in the 1920s. Junk bonds really did open corporate finance to companies shut out of investment-grade markets. Japan really was becoming an industrial powerhouse. The Internet really did change the world.
The mistake was generally subtler:
A true observation about the future became an extravagant assumption about price.
1. Tulipmania (1636–1637): When the Asset Almost Disappeared
Seventeenth-century Holland was one of the richest and most commercially sophisticated societies on earth. Amsterdam’s merchants traded across oceans; its financial markets were unusually advanced; and wealth had created a taste for luxuries whose value lay precisely in their scarcity.
Into this environment came the tulip.
Certain varieties were extraordinarily difficult to reproduce. Especially prized were “broken” tulips, whose petals displayed spectacular streaks and flames of colour. A rare bulb was therefore not quite comparable to an ordinary flower. It resembled a collectible: its value came from beauty, scarcity, fashion, and the possibility that somebody else might prize it still more highly.
That alone did not produce the famous mania. The decisive development was financialization.
Bulbs could only physically change hands during certain seasons. Traders therefore increasingly dealt not in flowers sitting on a table but in contracts for future delivery. By 1636, organised trading was taking place in taverns and informal “colleges.” In many transactions neither side possessed what would ultimately be exchanged: the seller did not yet have the bulb available for delivery, while the buyer did not necessarily possess the cash necessary to pay for it. Modern scholarship describes a market in which little or no margin was required and participants frequently expected merely to settle the difference between their agreed price and the later market price. (Cambridge University Press)
Notice what has happened:
The tulip has almost vanished from the transaction.
A man no longer buys a flower because he desires a flower. He buys a claim on a flower because he expects another man to value the claim more highly tomorrow.
That transformation appears repeatedly in speculative history. An asset begins as something people want to own. Eventually, its principal attraction becomes the expectation that its price will rise.
And once that happens, rising prices perform an extraordinary psychological trick: they appear to confirm the original thesis.
Suppose a bulb rises from 100 guilders to 200. To the skeptic this is evidence that speculation has become excessive. To the participant it seems evidence that he was right. The people who doubted the boom are poorer than the people who embraced it. Success therefore recruits new believers.
The process can become self-reinforcing:
$$\text{Price rises} \longrightarrow \text{Apparent proof} \longrightarrow \text{New buyers} \longrightarrow \text{Further price rises} \longrightarrow \text{Stronger apparent proof}$$
Tulip prices accelerated dramatically during the winter of 1636–37. Then, in early February 1637, buyers simply stopped appearing.
One intriguing modern explanation connects the timing to the biology of the tulip itself. Bulbs planted underground during winter concealed information about how many bulbs were being cultivated. When shoots began appearing in early February, the scale of production became more visible. Whether or not this explanation accounts for the entire collapse, the timing is striking: the speculative market broke in the first week of February, beginning around Haarlem, and trading effectively ceased in many places. (Cambridge University Press)
This is perhaps the purest form of a crash:
There need not be some spectacular piece of bad news. The price merely requires new buyers at still higher prices. Once they disappear, yesterday’s price is irrelevant. A market can therefore collapse even though nothing obvious has happened to the asset.
The more important historical correction is that the popular legend exaggerates the economic consequences. Modern scholarship does not support the idea that the entire Dutch economy was ruined. Tulipmania was a spectacular speculative episode, but its macroeconomic effects appear to have been relatively limited. (Cambridge University Press)
That distinction matters: A bubble and a financial crisis are not necessarily the same thing.
2. The South Sea Bubble (1720): When Government Finance Became a Lottery
Eighty years later, speculation moved from flowers into something vastly more consequential: the machinery of the British state itself.
The South Sea Company had been established in 1711. Its arrangement with the government involved assuming government debt in exchange for privileges that included trading rights in Spanish-controlled South America.
The commercial promise sounded magnificent. The practical opportunities were rather less magnificent, particularly since Britain and Spain were frequently hostile and relatively little South Sea trade actually occurred.
But the company’s great opportunity lay elsewhere:
It discovered that public debt could be transformed into corporate equity.
In 1720 the company received permission to take over a substantial portion of Britain’s national debt. Government creditors could exchange their claims for South Sea shares. If those shares rose, the arrangement appeared advantageous to everyone:
- Creditors gained appreciating securities.
- The government reorganised its debt.
- Existing South Sea shareholders became richer.
And rising South Sea shares made the next conversion still easier. The company had stumbled upon one of the most powerful forces in finance:
A rising share price can itself become a financing mechanism.
The higher the shares went, the more valuable the currency with which the company could acquire claims and attract subscribers. Soon the mechanism began feeding upon itself.
The Bank of England records that South Sea shares rose dramatically in 1720, producing a frenzy of investment. Subscription arrangements allowed investors to participate through instalments, effectively increasing their exposure without immediately providing the entire purchase price. The collapse of John Law’s parallel Mississippi speculation in France also sent some capital toward London. (Bank of England)
Then came imitators. New companies sprang up promising profits from mines, insurance ventures, overseas commerce, engineering schemes, and assorted projects. The particular schemes mattered less than the atmosphere surrounding them. Investors had learned that new shares might rise enormously after issuance.
A subtle change again occurred:
- Originally: Buy shares because the enterprise will produce profits.
- Eventually: Buy shares because newly issued shares rise.
At this stage, speculation creates its own evidence. Every newly wealthy speculator becomes an advertisement for speculation:
A gentleman who earns his fortune slowly from land or trade attracts little attention. A neighbour who doubles his money in South Sea stock in a few months produces envy, conversation, and imitation.
Then the direction reversed.
Once South Sea shares began declining, the same financing mechanism that magnified the ascent operated backward. Instalment buyers still owed money. Leveraged participants needed liquidity. Those who had purchased merely because prices were rising discovered that they possessed no independent reason to hold once prices fell.
The boom unwound rapidly. Thousands suffered losses; political investigations uncovered bribery involving prominent public figures; and the episode became a permanent symbol of speculative excess in British political culture. Yet historians caution against treating it as economically equivalent to 1929 or 2008: its broader long-term economic damage appears to have been considerably less severe. (Bank of England)
The South Sea episode nevertheless introduced a phenomenon that would appear repeatedly afterward:
When the price of an asset becomes part of the mechanism supporting the asset, falling prices do more than make investors poorer. They begin dismantling the financial structure that produced the boom.
That is when an ordinary market decline becomes dangerous.
3. Railway Mania (1840s): The Dangerous Bubble Built Around a Technology That Actually Worked
The railway mania presents a much more difficult intellectual problem.
Tulips did not transform civilization. Railways did.
A person standing in Britain in the early 1840s and declaring that railways would change commerce, cities, industry, and travel would have been absolutely correct. This is precisely what made the speculation so persuasive.
Early railway ventures had produced handsome profits. The technology worked. The public could see the trains. Traffic was increasing. Parliament authorized new routes. Britain appeared to be constructing an entirely new economic nervous system.
Investors drew what seemed an obvious conclusion:
If some railways are enormously valuable, more railways must mean more wealth.
Between 1843 and the autumn of 1845, British railway share prices more than doubled. By autumn 1845, 562 proposed railway schemes had been submitted to Parliament. Railway investment spread beyond an established financial elite, drawing substantial participation from Britain’s expanding middle classes. (Cambridge University Press)
The Trap in the Structure: Part-Paid Shares
Many railway shares were part-paid. Suppose a £100 railway share required only £10 initially:
- You could obtain £100 of economic exposure while committing only £10 today.
- Economically, that resembles leverage.
- A 10% increase in the value of the underlying share represented a 100% return on cash initially committed.
During the boom this arrangement was wonderful. But companies had the right to demand the remaining capital as construction progressed.
Suddenly the apparent £10 speculation became a £100 obligation.
Research on Railway Mania finds that these part-paid shares substantially magnified investors’ exposure and helped fund multiple schemes simultaneously; subsequent capital calls then placed downward pressure on prices as shareholders needed to produce additional cash. (Cambridge University Press)
This is one of the recurring ironies of speculative booms:
The financing technique that makes the boom possible becomes one of the mechanisms that breaks it.
And yet, thousands of miles of railway survived the investors who financed them. That is why Railway Mania is so important for understanding later technological bubbles:
A bubble does not require the underlying technology to be fraudulent. Indeed, the most seductive bubbles often arise around revolutionary technologies because conventional valuation becomes unusually difficult.
How valuable is a nationwide railway network when no nationwide railway network has ever existed? How valuable is electricity before electrification? How valuable is the Internet before e-commerce?
When the future is genuinely unprecedented, almost any valuation can be defended by telling a sufficiently optimistic story.
4. America and the Panic of 1873: Building Ahead of Demand
America repeated the railway story on a continental scale. Following the Civil War, railroads seemed almost synonymous with national progress. They opened western land, connected agricultural regions to eastern markets, created towns, facilitated migration, and supported entire industries.
Again, the central thesis was correct: America really did need railways.
But there is an enormous difference between:
- America needs railways, and
- Every railway currently being financed is economically justified.
Railway investment expanded faster than the demand necessary to support all the proposed lines. European investors held substantial American railroad securities. When financial disturbances in Vienna in 1873 caused European investors to retreat from American assets, railroad financing became increasingly difficult. (Federal Reserve History)
At the centre stood Jay Cooke & Co. Cooke had become famous for financing the Union during the Civil War and subsequently threw his financial power behind the Northern Pacific Railway.
The project required immense sums of capital. Railways consume money long before they produce it:
- Land must be surveyed.
- Track must be graded.
- Bridges constructed.
- Rails purchased.
- Workers paid.
- Locomotives acquired.
A railway under construction therefore lives on continued confidence:
If investors stop providing new capital halfway across the continent, the fact that the finished railway might someday be profitable offers little consolation.
By 1873 construction expenses were outrunning financing. Jay Cooke’s firm could no longer place sufficient Northern Pacific securities. On September 18, it failed.
Confidence vanished almost immediately. Brokerage houses failed. Depositors withdrew money. Railway financing froze. On September 20, the New York Stock Exchange suspended trading—the first such closure in its history. (Library of Congress Guides)
The physical railway tracks had not disappeared. What had disappeared was the willingness to finance the next mile.
This illustrates another essential distinction:
A solvent long-term project can still be destroyed by a short-term funding crisis.
Finance deals not only with whether an investment will eventually make money, but whether its owners can survive long enough to reach that eventually.
5. 1929: When Prosperity Became Extrapolation
The American boom of the 1920s was again built upon realities: mass production was changing industry, cars spread rapidly, radio created an entirely new medium, electrification transformed factories and homes, consumer credit expanded, and corporate productivity improved.
America looked like the future. And gradually investors concluded that a new economic age required a new valuation regime.
The stock market reflected the optimism spectacularly. The Dow Jones Industrial Average rose roughly sixfold, from 63 in August 1921 to 381 in September 1929. (Federal Reserve History)
The Mechanism of Leverage
The most dangerous development was not simply that people liked stocks; it was that they could buy much more stock than they could afford. Margin accounts permitted investors in some cases to put down roughly 10% of the purchase price and borrow the rest, with shares serving as collateral. (Federal Reserve History)
Imagine a $100 stock purchased with:
- $10 your money
- $90 borrowed
If the stock rises to $120, your equity rises from $10 to $30. The stock gained 20%, but you gained 200% on your original capital. That is the intoxicating effect of leverage.
Reverse the arithmetic: If the stock falls from $100 to $90, your $10 equity has completely disappeared. A decline that is uncomfortable for an unleveraged investor is fatal for a leveraged one.
This produces forced selling. The investor does not sell because he has calmly reconsidered the company’s long-term prospects. He sells because the lender demands its money. And forced selling drives prices lower, which forces someone else to sell:
$$\text{Falling prices} \longrightarrow \text{Shrinking collateral} \longrightarrow \text{Margin calls} \longrightarrow \text{Forced sales} \longrightarrow \text{Lower prices}$$
By autumn 1929, the market had become unstable. Selling overwhelmed attempts by prominent bankers to restore confidence. On October 28 (Black Monday), the Dow fell nearly 13%. On October 29 (Black Tuesday), it fell almost another 12%.
By mid-November, nearly half the market’s peak value had disappeared. The decline eventually continued until July 1932, when the Dow stood roughly 89% below its 1929 peak. It did not regain that 1929 level until 1954. (Federal Reserve History)
Crash vs. Depression
There is an important historical subtlety here. It is tempting to draw a straight line:
$$\text{Stock Crash} \Longrightarrow \text{Great Depression}$$
The actual history was more complicated. The American economy had already begun contracting before the crash. The market collapse weakened spending and confidence, but the Depression became catastrophic through subsequent banking panics and financial crises in 1930–33, policy mistakes, monetary contraction, and international stresses associated with the gold standard. (Federal Reserve History)
This is crucial because it explains why some stock bubbles burst without producing depressions:
The decisive question is: What is connected to the bubble?
If people lose money in stocks, that is painful. If those stocks collateralize enormous amounts of borrowing, banks fail, credit disappears, and businesses cannot finance themselves, the consequences become vastly larger.
6. The Junk-Bond Age: Speculation Moves to the Capital Structure
The speculative excitement of the 1980s looked very different. Instead of tulip bulbs or railway shares, the object at the centre of the drama was corporate debt.
High-yield bonds, commonly called junk bonds, were bonds issued by companies with below-investment-grade credit ratings. The underlying financial insight was legitimate: a risky company might pay a sufficiently high interest rate that a diversified investor could still earn attractive returns. A bond should not automatically be dismissed merely because a rating agency called it speculative.
The market grew rapidly: from near insignificance around 1980, the American high-yield market reached nearly $200 billion by the end of the decade. (FRASER) Michael Milken and Drexel Burnham Lambert became its dominant figures.
Fuelling the Leveraged Buyout (LBO)
The importance of junk bonds extended beyond investing: they provided fuel for the leveraged-buyout boom.
Suppose a company worth $10 billion generates substantial cash flow. An acquirer might contribute only part of the purchase price himself and borrow the rest. The acquired company’s own future cash flows would then service the debt used to acquire it. If everything worked, a relatively small equity investment could control an enormous enterprise.
Again:
- Leverage magnifies returns.
- Leverage magnifies errors.
During good economic conditions, interest payments appeared manageable. Rising asset values made transactions look successful, encouraging more aggressive financing. Eventually, companies were burdened with capital structures requiring nearly everything to go right.
By 1989, strains emerged. Highly leveraged companies including Campeau encountered serious financial difficulty. Investors became more sensitive to whether corporate cash flows could genuinely support debt-service obligations. Bond defaults increased, spreads widened, and issuance contracted sharply. (FRASER)
Then Drexel itself failed in February 1990. This mattered because Drexel had not merely sold junk bonds; it had provided liquidity, inventory, and market-making for the entire ecosystem. The New York Fed described investors worrying not merely about Drexel’s holdings, but about how the high-yield market would function after the loss of its largest market maker. (New York Fed)
The lesson was not that high-yield debt itself was economically useless—the market eventually returned and remains an important part of modern finance.
The deeper lesson concerned the danger of confusing:
a useful financial innovation with every use of that financial innovation being sensible.
The same distinction had existed with railways a century earlier.
7. Japan (1985–1990): When Collateral Created Its Own Prosperity
Perhaps the most extraordinary modern example occurred in Japan.
By the mid-1980s, Japan’s postwar economic achievement was undeniable. Japanese manufacturers dominated industries that Western companies had once controlled. The country appeared disciplined, technologically sophisticated, and extraordinarily productive.
From this reality emerged a larger proposition: perhaps Japanese assets deserved permanently higher valuations.
Money and credit expanded. Banks lent heavily. Real-estate prices rose. But rising real-estate prices did something particularly important in a bank-centred financial system: they increased the value of collateral.
Imagine a company owns land worth ¥1 billion:
- A bank might comfortably lend ¥500 million against it.
- Then land prices double to ¥2 billion.
- The company’s collateral appears stronger, so the bank lends more.
- The company uses that credit to purchase more property or financial assets.
- Those purchases drive asset prices higher, increasing collateral values once more.
$$\text{Land rises} \longrightarrow \text{Collateral rises} \longrightarrow \text{Lending rises} \longrightarrow \text{Purchasing power rises} \longrightarrow \text{Land rises}$$
This is one of the most dangerous feedback loops in finance because the bubble manufactures evidence of financial strength.
Research from the Bank of Japan describes the late-1980s bubble as combining three features: rapidly rising asset prices, overheating economic activity, and expanding money and credit. Other research finds that increased bank lending to real estate played an important role in the initial asset-price rise. (Bank of Japan IMES)
Japanese equities became extraordinary:
- The Nikkei 225 ended 1985 around 13,113.
- Four years later, on the final trading day of 1989, it closed around 38,915. Shares had nearly tripled. (IMF eLibrary)
Then monetary conditions tightened. Equity prices fell. Property prices followed. Now the collateral machine operated backward:
$$\text{Land falls} \longrightarrow \text{Collateral falls} \longrightarrow \text{Bank balance sheets weaken} \longrightarrow \text{Lending contracts} \longrightarrow \text{Borrowers sell assets} \longrightarrow \text{Land falls further}$$
The aftermath differed radically from Tulipmania. Japanese banks had lent against inflated collateral. When asset prices fell, the loans remained, but the collateral supporting them did not.
Businesses and financial institutions spent years repairing their balance sheets. Bad loans accumulated. Investment weakened. The economy slowed dramatically, and later in the 1990s consumer prices entered a prolonged period of deflation. (IMF eLibrary)
The destruction was not principally caused by somebody buying an expensive stock. It resulted from the toxic interaction between:
asset prices + collateral + credit + bank balance sheets.
Whenever those four are deeply intertwined, a falling market becomes a profound economic event.
8. The Internet Bubble: Being Right About the Revolution and Wrong About the Investment
Then came perhaps the most beautiful example of the paradox.
The Internet enthusiasts were right. Staggeringly right.
The Internet would transform retail, advertising, communications, entertainment, finance, media, and almost every information-based industry.
But investors quietly substituted another proposition:
“If the Internet changes everything, Internet companies are worth almost anything.”
Those statements are not equivalent.
Venture capital poured into technology companies. IPOs produced spectacular first-day gains. Companies discovered that attaching “.com” to a business plan could attract attention and financing. Conventional measures such as profits seemed old-fashioned when market share, website traffic, “eyeballs”, and network effects promised domination of enormous future markets.
Once again, the difficulty of forecasting a genuine technological revolution loosened the constraints of valuation:
- How large would online commerce eventually become? Nobody knew.
- If nobody knew, why couldn’t it be enormous?
- And if it would be enormous, perhaps extraordinary valuations were justified.
The Nasdaq rose 170% between September 1998 and March 2000. (IMF)
This produced familiar circularity:
- Technology shares rose because investors believed the Internet would create enormous wealth.
- Rising prices created wealth for employees, founders, and investors.
- That wealth funded more start-ups.
- Start-ups bought advertising, servers, software, and office space from other technology companies.
- Those companies reported rapid revenue growth.
- That revenue growth seemed to validate tech valuations.
Capital-market enthusiasm began appearing directly in the operating results used to justify the capital-market enthusiasm.
Then expectations reached levels that companies could not satisfy. In March 2000, the Nasdaq peaked. Within a year it had fallen roughly 60%. Within two years it had fallen around 75% from its peak. One IMF study estimates approximately $5 trillion of market wealth disappeared from peak to trough. (IMF)
Numerous companies vanished. Yet the Internet did not:
- Amazon survived.
- Digital commerce grew.
- Broadband spread.
- Online advertising became enormous.
- Cloud computing eventually appeared.
The technological revolution continued largely as its advocates had predicted. The securities simply had been priced as though too much of that future belonged to the companies then trading at extravagant prices.
A revolutionary technology can be a terrible investment at the wrong price.
Indeed, the stronger the technology’s promise, the easier it becomes to rationalise paying almost anything for exposure to it.
9. What All These Crashes Have in Common
Across nearly four centuries, the objects change almost beyond recognition: flowers become government-finance companies, which become railways, stocks, corporate bonds, Japanese property, and Internet companies.
The financial machinery evolves. Human behaviour changes surprisingly little.
The recurring sequence looks roughly like this:
- Something genuinely changes: A new technology, financial innovation, political arrangement, or economic regime creates real opportunity.
- Early investors make real money: Their success provides convincing evidence and attracts imitators.
- Credit or financial innovation expands participation: Futures contracts, instalment subscriptions, part-paid shares, margin loans, junk bonds, or bank collateral allow investors to command more assets with less cash.
- Price appreciation becomes evidence of the thesis: Instead of asking whether rising prices are justified, investors interpret rising prices as proof that they were right.
- Valuation gives way to extrapolation: “This is valuable” becomes “this will continue becoming more valuable.”
- The system requires a continuous supply of new money: Once that money stops arriving, leverage and forced selling transform a normal decline into a crash.
Bubble Size vs. Systemic Disaster
There is one further distinction worth keeping in mind:
The size of a bubble does not determine the size of the economic disaster. Its financing does.
- Tulipmania collapsed without destroying the Dutch banking system.
- Technology stocks lost trillions after 2000 without producing another Great Depression.
- But when inflated assets sit underneath large quantities of debt, when banks depend upon them as collateral, or when short-term funding supports long-term speculative positions, falling prices begin to damage the mechanism that supplies credit to the rest of the economy.
That is when speculation ceases to be merely a story about foolish investors. It becomes a story about financial systems.
The Dangerous Sentence
Perhaps the most unsettling lesson from this history is that identifying the next mania is unlikely to be as simple as finding something obviously ridiculous. The great speculative episodes normally begin with an idea sufficiently compelling to persuade intelligent people:
- The railway really will transform Britain.
- America really has entered an age of mass prosperity.
- Japanese industry really is extraordinary.
- The Internet really will change the world.
The dangerous sentence comes immediately afterward:
“Therefore, the price does not matter.”