Why every bubble looks the same
Bubbles differ in what they're about and resemble each other in how they unfold. Five phases have repeated for centuries. You can recognize the signs while it's happening, but you can't prove them.
At the start there's always something real: a new technology, a new trade route, lower interest rates. Bubbles don't come out of nowhere. They begin with a change you can understand.
Then comes the boom, in which early buyers make money. In the euphoria, people join who have never been interested in the subject, and more and more is bought with borrowed money.
At some point the early ones start to sell. Prices stop making progress while the story is still being told. And then it tips over, usually without any visible trigger and often set off by forced sales.
Signs that keep coming back: new metrics because the old ones no longer fit. The sentence that this time everything is different. People with no experience at all who are all joining in. And borrowed money as the stake. If you see three of them at once, you don't know when it will end. But you do know where you are.
The economist Charles Kindleberger, building on Hyman Minsky, described five phases: a trigger, the boom, euphoria, first profit taking and panic. At the core is debt. First people borrow money they can repay from their income. Later they borrow in the hope that prices will keep rising. In the end many need new loans just to pay the interest on the old ones.
Charles Kindleberger applied this model to bubbles from many centuries and showed that the course is the same again and again. What's remarkable is that the pattern stays the same across such different things, tulips, railways, houses or internet companies. So the causes lie in how things are bought and financed and in how people behave, less in what's being bought.
What remains difficult is spotting a bubble while it's forming. High prices can be justified by low interest rates or real growth, and bubbles can last a very long time. More reliable than the level of prices is a look at the debt: how much is being bought on credit relative to the total market value? Because the tipping over follows almost mechanically from that debt.
Summary
- Every bubble starts with something real.
- The tipping over follows almost mechanically from the debt, not from mood.
- You can recognize a bubble, but you can't predict when it will end.
Did you get it?
Which five phases does the model describe?
A trigger, the boom, euphoria, first profit taking and panic.
What's the mechanical core of the reversal?
Those who bought on credit have to sell when prices fall, and those sales make prices fall further.
Which metric is more informative than the valuation level?
How much is being bought on credit relative to the total market value.
Sources and further reading
- Hyman Minsky on the financial instability hypothesis, and Charles Kindleberger on the history of financial crises.
- Hyman P. Minsky, The Financial Instability Hypothesis, Working Paper No. 74, Levy Economics Institute, 1992: hedge, speculative and Ponzi finance and the shift toward more debt during long expansions. Retrieved October 9, 2026. View source ↗
- Review of Kindleberger and Aliber, Manias, Panics, and Crashes (NJIT, 2015): stages from displacement through euphoria and profit taking to panic. Retrieved October 9, 2026. View source ↗
Related
- The 1929 crashMarket History
- Bull market, bear market, crash, bubbleStage 1
- The dot-com bubble, 2000Market History