What every stock market crash has in common: five repeated patterns

Which five patterns repeat in every stock market crash?

Five patterns recur across the crashes of 1929, 1987, 1990, 2000, 2015, and 2020. First, prices outrun earnings. Second, borrowed money turns a price fall into forced selling. Third, a trigger shifts the cost of money or the flow of news. Fourth, selling feeds on itself through rules and panic. Fifth, the policy response and the state of the banks decide how long the damage lasts.

A timeline shows the patterns better than a list of causes. The triggers differ widely: a trade deficit report, a rate rise, a magazine warning, an antitrust ruling, a currency cut, a virus. The conditions underneath look similar each time, while the size of the fall differs: the Dow lost 22.6% in 1 day in 1987, the Nikkei 225 lost more than 35% during 1990, and the Nasdaq lost 78% by October 2002. This page gives one section to each pattern, with figures from the pages of this series, so you can compare crashes side by side.

Do rising prices and optimism come before every stock market crash?

A long climb with rising optimism comes first. Before the 1987 crash the S&P 500 tripled in 5 years and traded at 23 times earnings against a postwar average of 14.5. Before the 2000 crash the Nasdaq reached a price-to-earnings ratio of over 200. Before the 2015 crash Shanghai prices rose more than 150% in 12 months, and Nicholas Lardy wrote that the market was priced way beyond perfection.

Japan in 1989 follows the same shape. The Nikkei 225 closed at 38,915.87 on 29 December 1989, and Tokyo commercial land rose about 42% in 1985 and about 122% in 1986. Stretched prices set the size of the fall that follows. They give no date, since markets can stay expensive for years, which is why the second pattern matters for timing.

How does borrowed money show up in every stock market crash?

Borrowed money magnifies a fall. In 1929, margin buying and leverage fueled investor enthusiasm, the Dow fell 12.8% on Black Monday and 23% across Black Monday and Black Tuesday, and it lost 89% by July 1932. In 1987 the Brady Commission recorded margin calls about 10 times their average size, and 11 firms received margin calls above their adjusted net capital.

In Japan, loans rested on land and share prices, so each fall cut the collateral behind the loans. In China in 2015, retail investors who borrowed to buy equities supplied an unusually large part of the run-up, and 3 weeks of falling prices produced forced sales. The mechanism stays constant: a loan backed by a price needs the price to hold.

What triggers a stock market crash once prices are high?

A trigger changes the price of money or the flow of information. The Bank of Japan raised its discount rate 5 times between May 1989 and August 1990, from 2.5% to 6.00%. In 1987 a trade deficit report on 14 October weakened the dollar and pushed interest rates up. In 2000 the Federal Reserve raised rates the day after a Barron’s article on burn rates, and the yield curve inverted.

Triggers also come from outside the financial system. In 2020 the spread of covid-19 and an oil price war arrived in the same fortnight, and the Dow reached a bear market 21 days after its record. In 2015 the yuan was cut 1.86% on 11 August. Surveys keep triggers in proportion: in Robert Shiller’s 1987 survey of 889 investors, only 3 institutional investors cited the merger tax bill, and the most common answer was a gut feeling of an impending crash.

Why does selling feed on itself in a stock market crash?

Selling feeds on itself through rules and crowds. In 1987 portfolio insurance sold index futures as prices fell, and the Federal Reserve notes that the strategy accelerated the crash because early losses triggered further rounds of selling. In 2015 around 1,300 firms, representing 45% of the market, suspended trading from 8 July while the index fell 30% in 3 weeks.

Halts and curbs shape the pace. Trading curbs activated 4 times within a week in March 2020, and prices kept falling after each reopening. In January 2016 Chinese authorities suspended their circuit breaker, citing concern that it may have intensified investors’ concerns. A rule that pauses trading manages the pace of a fall, and the cause of the fall still needs a policy answer.

What decides how long a stock market crash lasts?

Policy and the state of the banks decide the length. In 1987 the Federal Reserve affirmed its readiness to serve as a source of liquidity on 20 October, and the S&P 500 regained its high on 26 July 1989, about 21 months after the crash. In 2020 the Fed cut its target to 0% to 0.25%, the US passed a $2.2 trillion fiscal package, and US indices returned to January levels in November 2020, within 9 months of the low.

Where banks carried bad debt, the damage lasted longer. In Japan, bailouts, cheap central bank credit, and delayed loss recognition turned many banks into zombie banks, and the Nikkei 225 needed 34 years to close above its 1989 record, on 22 February 2024. The Nasdaq needed about 15 years after the 2000 peak to set a new record close. Recovery time therefore ranges from months to decades, and the range depends on credit and policy. The EQOS Asia method page shows how a score reads credit strength and policy room together.

How can you use the five patterns before the next stock market crash?

Turn the patterns into 5 questions. Are prices far above earnings? Who borrowed against those prices? What could change the price of money or the news flow? Which rules would sell at the same time? How strong are the banks and the policy room? Write the answers each quarter, and note which patterns turn red. As of October 2026 this routine ties your view to evidence instead of dates.

A score combines structural dimensions into 1 measurement, and the EQOS Asia simulator compares a soft landing, a correction, and a crash for your own case. The series on the stock market crash holds the other guides. This page is general information, not investment advice.

Frequently asked questions

About the author

Stéphane Jambu

President, EQOS Forecast Asia

Stéphane Jambu leads EQOS Forecast Asia and works as a consultant in e-commerce, search visibility, and AI. LinkedIn profile