Wiki › Market history: manias, crashes and manipulation

Tulip Mania, 1637 — the first bubble, and the parts that are not true

Reading for India · about 11 min

The answer

In the winter of 1636 to 1637, prices for rare tulip bulbs in the Dutch Republic rose to absurd levels and then collapsed in about a week in February 1637 . Almost everything else you have heard about it — the ruined nation, the chimney sweeps, the suicides — comes from a book written 204 years later and is not supported by the records.

Why this costs you money

Here is the damage that the popular version of this story does.

You read that tulip mania was obvious madness. You conclude that bubbles are obvious. You then walk through a real one without recognising it, because the real one does not look like a flower selling for the price of a house. It looks like a sensible company, with real customers, at a price that assumes those customers will multiply for 10 years.

The false version teaches you to look for stupidity. The true version teaches you to look for a price that requires a specific future. Those are different searches, and only the second one finds anything.

There is a second cost. Because the tulip story is told as a story about fools, you assume that intelligence protects you. It does not. The people trading bulbs in 1636 were, according to the archival work, wealthy merchants and skilled craftsmen — the informed, numerate class of the richest economy in Europe. They were not confused about what a tulip was. They were making a judgement about what somebody else would pay.

If you think a bubble requires idiots, you will not check whether you are in one.

How it works

Strip the flowers out and 5 things were present. They are present in every episode in this cluster.

1. A real object with a real, hard-to-value quality. Tulips had arrived in Europe from the Ottoman Empire in the 1500s. The most prized ones had streaks of colour on the petals. Nobody then knew what caused the streaks. We now know it was a virus, which is also why those bulbs reproduced poorly. So the rarest bulbs were genuinely rare and genuinely impossible to produce on demand. That is the perfect base for a price with no ceiling: real scarcity plus no way to check what it should cost.

2. A market in promises, not in the thing itself. Tulip bulbs are lifted from the ground in summer and planted in autumn. For most of the year there is no bulb to hand over. So the Dutch traded contracts: a written promise to deliver a named bulb, at a named weight, in the following season. Contemporaries called this trade windhandel, which translates as "trade in wind".

This is the crucial mechanical point. Once you are trading a promise, you can buy without money and sell without goods. The number of contracts is not limited by the number of bulbs.

3. No settlement date in between. A contract signed in December was not settled until the summer. Nothing forced a buyer to produce cash. So a chain of buyers could form, each one confident, because none of them had yet been asked to pay.

4. Prices with no anchor. The highest reliably documented price is around 10,000 guilders for a single Semper Augustus bulb, roughly the cost of a good house on an Amsterdam canal. An auction of orphans' bulbs at Alkmaar in February 1637 raised about 90,000 guilders for around 100 lots. These are real numbers from real documents. They are also a small number of transactions in a small market.

5. The stop. On or about 3 February 1637, at a routine bulb auction in Haarlem, the auctioneer could not find buyers. Word spread within days. Prices for the traded contracts fell to almost nothing.

Note what did not happen. There was no bad news about tulips. No disease, no new supply, no law. The demand stopped, and it stopped because the only remaining reason to buy was that somebody else would pay more. That belief can be destroyed by 1 failed auction, because a failed auction is evidence about the belief itself.

What it tells you, and what it does not

What the episode does tell you. A price supported only by the expectation of resale can go to almost zero without any change in the object. This is called the greater fool structure: you knowingly pay too much, because you expect a greater fool to pay more. The structure is not a moral failure. It is a perfectly coherent trade, right up to the moment the supply of greater fools ends. And it always ends without warning, because the last fool does not announce himself.

What it does not tell you. It does not tell you that manias destroy economies. The Dutch Republic went on being the strongest economy in Europe. Anne Goldgar, who read the notarial and court records, found fewer than a handful of people in financial difficulty from tulips and no bankruptcies she could trace to them. There was no depression. There was a mess.

It also does not tell you how to time anything. Prices had been absurd for months before the stop. Being right about value and being right about when are different skills, and this episode teaches only the first.

The decision rule

If you cannot state what the asset produces, then the only thing supporting the price is the next buyer. Size the position for that.

Not "never buy it". People make money on resale-driven assets all the time. The rule is about honesty of description. If your reason for holding is "more people will want this", you are in the tulip trade, and the tulip trade has one property: it does not decline gently. It stops.

The counter-condition: if the thing produces cash — rent, interest, profits, dividends — then there is a floor somewhere, because at some price the cash flow alone justifies holding. Assets with cash flows fall. Assets without them can stop having a price at all.

Try this now

Five minutes, on your own holdings. This is the single most useful thing in the article.

  1. Open your holdings. Pick the position that has risen the most since you bought it, in percentage terms.
  2. Find its current market capitalisation. It is on the stock's page in your broker app, usually under Fundamentals or Overview.
  3. Now write one sentence, by hand, on paper or in your notes app. Fill in the blanks with actual numbers: "For today's price to be reasonable, in 10 years this company must earn about ₹\_\_\_\_ crore a year, which means revenue of roughly ₹\_\_\_\_ crore, which means \_\_\_\_ times its current sales." A rough method: take today's market cap, divide by 20. That is approximately the annual profit a mature business would need to justify it. Then use the company's current profit margin to get back to the revenue.
  4. Read your sentence out loud.
  5. Answer 1 question in writing: do I believe that? Yes or no. Not "possibly". Yes or no.

What you should see. For most holdings, the sentence is unremarkable and you will answer yes. For 1 or 2, the required revenue will be larger than the entire market that company sells into. That is the tulip position in your portfolio, and you found it without anybody telling you the price is too high.

Keep the sheet of paper. In a year, read it again. Writing it down means you cannot later pretend you expected something smaller.

Three real cases

1. Semper Augustus, Dutch Republic, 1630s — the documented extreme The most famous striped tulip. Prices of around 10,000 guilders for a single bulb are recorded, against a skilled craftsman's annual income of a few hundred guilders. What it turned out to be: a bulb infected with a virus that was, unknown to everybody, slowly destroying the plant's ability to reproduce. The rarest and most expensive item in the market was the sickest one.

2. The Alkmaar auction, 5 February 1637 — the last good print An auction of bulbs belonging to the estate of a deceased innkeeper raised a very large sum, in the region of 90,000 guilders, only days after the Haarlem auction had failed. What it turned out to be: the last high price in the record. The sellers were paid. Several of the buyers refused to honour their contracts weeks later.

3. The Haarlem settlement, 1637 — how it was actually resolved Buyers refused to pay; sellers sued; the courts of Holland declined to enforce the contracts. In May 1637 Haarlem proposed that buyers could walk away by paying 3.5% of the contract price. What it turned out to be: not a wave of ruin but a legal shrug. Almost nobody had paid, so almost nobody lost cash. What was lost was trust between merchants, which in that society was the more serious injury.

The question that resolves it

A novice looks at a rising price and asks: how high can this go?

An expert asks: who is the buyer after me, and why would they buy?

If you can name the buyer — a pension fund that must own the index, a company buying back its own shares, a customer base that keeps growing — you have an answer. If the honest answer is "somebody who thinks it will go up", you are holding wind.

What would make this wrong

Start with the part that is genuinely uncertain, because this article rests on it.

The revisionist case could be overstated. Anne Goldgar's archival work, and Peter Garber's earlier economic work, made a strong case that tulip mania was small and nothing like the catastrophe of legend. But absence of recorded bankruptcies is not proof of absence of loss. Court records survive unevenly and informal debts leave no trace. The honest position is that the documented damage is small and the legendary damage is unsupported.

Now the hindsight trap, which matters more.

It is tempting to write that anyone could see this was madness. Consider what a careful, intelligent bulb trader in December 1636 actually knew.

Tulips had been rising in price for decades. Every previous rise had held. The rarest varieties genuinely could not be produced to order, so scarcity was real. The Dutch Republic was in a genuine boom — new trade routes, new wealth, a new merchant class that wanted status objects. The buyers were not strangers; they were people you knew. And the contracts were enforceable, or so everybody assumed, because contracts in Holland had always been enforceable.

The one thing a careful person could genuinely have seen is this: the price had detached from any use. Nobody was buying Semper Augustus bulbs to plant a garden at those prices. Every buyer in the chain intended to sell. That fact was visible in December 1636 to anybody who asked "who is the end user?" — and it was the correct question then, as it is now.

What that careful person could not have seen was the date. They could have known the structure was fragile. They could not have known it would snap in the first week of February. If somebody tells you they can see both, they are describing something nobody has ever reliably done.

In India

India has no 17th-century tulip episode, and inventing one would be dishonest. The nearest genuine Indian parallel comes 228 years later, and it is closer than you might expect.

Between 1861 and 1865, the American Civil War cut off the supply of American cotton to British mills. Indian cotton, shipped from Bombay, filled the gap. Money poured into the city. That money went into shares of banks, land reclamation companies and shipping companies. The most famous was the Back Bay Reclamation Company, which proposed to reclaim land from the sea on the western edge of Bombay. Its shares traded at large multiples of their issue price.

The structure matched the tulip trade in 1 crucial respect. Shares were bought on time bargains — forward contracts settled at a future date, with little or no money paid up front.

The American Civil War ended in April 1865. Cotton prices fell. The Bombay share market collapsed in the second half of 1865. The Bank of Bombay was severely damaged, and Premchand Roychand, the most prominent broker of the boom, lost most of his fortune in Back Bay and similar ventures. In 1875, brokers who survived formed the Native Share and Stock Brokers Association, which is today the BSE. India's oldest exchange was founded by the survivors of a crash.

The modern Indian version of windhandel is legal and widely used. A weekly index option can be bought for a few hundred rupees and represents a contract of several lakh rupees. Most are never settled by delivery. They are trade in wind, with an exchange guaranteeing the other side. SEBI has repeatedly published data on this segment, and the consistent finding is that a large majority of individual traders lose money.

In the United States

The United States did not exist in 1637, but the tulip structure has a precise modern American home: assets with no cash flow, traded on promises, with a settlement date far away or absent.

Two American episodes carry the same skeleton.

Florida land, 1925 to 1926. Plots were sold on binders — a small deposit that reserved the right to buy. Buyers resold the binder before any payment was due. The number of binders exceeded the number of buildable plots by a wide margin. A hurricane in September 1926 ended it, but the structure had already run out of new buyers.

Non-fungible tokens, 2021 to 2022. Digital certificates of ownership, with no cash flow and no legal claim on anything. Prices for some collections rose by a factor of 100 and then fell by more than 90% within roughly 18 months. The resemblance is not the technology. It is that every buyer was a future seller.

The regulatory difference matters. In the United States, an asset with no cash flow that is sold as an investment can fall under securities law, and the SEC has used that authority in several digital-asset cases. There was no such protection for a tulip, and none in 1637.

Where they differ, and what that tells you

The difference is not psychology. It is what happens when the buyer refuses to pay.

In 1637 the courts of Holland refused to enforce tulip contracts. The buyer simply walked away. That is why the recorded losses are small: the promises were never collected.

In modern India and the United States, a forward promise on an exchange is guaranteed by a clearing corporation, and it is collected. Your losses are deducted from your margin the same evening. There is no walking away.

What that tells you is uncomfortable. Modern markets are far better regulated than 1637, and that improvement makes the individual outcome worse, not better. In 1637 the mania was mostly notional and settlement was a negotiation. Today, when a leveraged position moves against you, the money leaves your account automatically, before you have decided anything.

So the lesson does not transfer as "it was not that bad". It transfers as: the greater-fool structure is the same, and the collection machinery is much stronger now. That combination is more dangerous, not less.

Carry this

  • A price supported only by the next buyer does not decline. It stops.
  • Ask "who is the end user?" If every buyer intends to sell, you know the structure.
  • Most of the famous tulip story is from Charles Mackay's 1841 book and is not in the records.

Knowledge check

Q. Two assets have both risen 300% in 18 months.

  • Asset A is a company whose profits have risen 250% over the same period. Its price now assumes profits keep growing at 40% a year for 10 years.
  • Asset B produces no cash at all. Its price has risen because the number of people who want to own it has risen.

Which one has the tulip structure?

Explanation. The tulip structure is not "the price went up a lot". It is "the only support under the price is the next buyer".

Asset A is probably expensive and it may fall a long way. But it has a floor somewhere, because at a low enough price the profits alone justify owning it. It falls, then it finds a level.

Asset B has no such level. If people stop wanting it, there is no price at which the asset itself starts paying you to hold it. That is why greater-fool assets do not glide down. They gap.

The first option is tempting because 40% for 10 years really is a heroic assumption, and this article's own exercise is designed to expose exactly that. But an unrealistic growth assumption makes an asset overpriced. It does not make it a tulip. Hold those 2 criticisms apart: one implies a painful fall, the other implies no bid at all.

The South Sea Bubble, 1720 — when the smartest man alive lost his money →