Stock market crashes in history: 1987, the dot-com bust, 2008 and Covid, and the risk lessons

A stock market crash is a sudden, steep fall in share prices across a whole market, usually defined loosely as a drop of 20% or more over days or weeks, or a double-digit fall in a single day. The four best-documented modern examples are Black Monday in October 1987, the dot-com bust of 2000 to 2002, the global financial crisis of 2007 to 2009, and the Covid crash of February and March 2020. Each had different causes and a different shape, but all of them punished leverage and showed that stop losses and diversification work less well when you need them most.

This guide sets out the documented facts of each episode, then draws out the lessons for anyone trading indices or shares with leverage. It is a history, not a forecast: none of it says when the next crash will happen or what it will look like. For background on the indices themselves, see what is a stock index.

Crashes, corrections and bear markets

The vocabulary is loose but useful:

• A correction is commonly described as a fall of 10% or more from a recent high.
• A bear market is commonly described as a fall of 20% or more from a high, usually lasting months.
• A crash describes the speed: a very large fall in a very short time.

A crash can mark the start of a long bear market, as in 2008, or be followed by a quick recovery, as in 2020. Nobody knows which at the time.

Black Monday: 19 October 1987

What happened

On Monday 19 October 1987 the Dow Jones Industrial Average fell 508 points, or 22.6%, in a single session. It remains the largest one-day percentage fall in the Dow's history. The S&P 500 fell about 20.5% the same day. The fall spread around the world, with large declines in markets from London to Hong Kong to Tokyo; the Nikkei 225 fell 14.9% the next day.

The crash followed a strong rally earlier in the year and a week of losses before the Monday. Explanations that are commonly cited include high valuations after that rally, rising interest rates and bond yields, and the role of portfolio insurance, a strategy that sold index futures automatically as prices fell, which added selling pressure as the market dropped. Order handling systems were overwhelmed, and many orders could not be executed at anything close to their intended prices.

What came next

The day after the crash, the Federal Reserve, under its new chairman Alan Greenspan, issued a short statement affirming its readiness to serve as a source of liquidity to the financial system. Markets stabilised over the following weeks. The US economy did not enter a recession as a result, and the Dow regained its pre-crash high in about two years.

The crash led directly to reforms. US exchanges introduced market-wide circuit breakers, which halt trading when an index falls by set percentages, so that a panic has a pause built into it. Today's US rules halt trading for 15 minutes when the S&P 500 falls 7% or 13% from the previous close during most of the session, and close the market for the day at 20%. The trading halts and circuit breakers guide covers how they work.

The lesson

The speed. A fall of more than a fifth in one day means a stop placed a few percent below the market could not be filled anywhere near its level. On a leveraged position, the loss is determined by where the market actually traded, not by where the stop was.

The dot-com bust: 2000 to 2002

What happened

Through the late 1990s, shares in internet and technology companies rose to valuations that, in many cases, assumed decades of rapid growth. The Nasdaq Composite, home to most of those companies, closed at a record 5,048.62 on 10 March 2000.

There was no single crash day. Instead, the index fell in waves over two and a half years, with sharp rallies along the way that convinced many investors the worst was over. By 9 October 2002 it had closed at 1,114.11, a fall of about 78% from the peak. Many individual companies did far worse, and a large number disappeared entirely. The broader S&P 500 fell about half from its 2000 peak to its 2002 low. The period also included the 11 September 2001 attacks, after which US stock markets stayed closed for several days, and a series of accounting scandals at large companies.

What came next

The Nasdaq Composite did not close above its March 2000 peak until 23 April 2015, more than 15 years later.

The lesson

A crash does not have to be fast. The dot-com bust was a slow grind with convincing rebounds, which is arguably harder to manage than a single bad day, because it invites traders to add to losing positions on each rally. It also shows concentration risk: an index dominated by one theme can fall much further than the broad market. That is relevant today to any index heavily weighted in a few large technology companies; the Nasdaq-100 trading guide covers that concentration.

The global financial crisis: 2007 to 2009

What happened

The S&P 500 closed at a record 1,565.15 on 9 October 2007. Losses on US mortgage lending, especially subprime loans packaged into securities held across the financial system, spread through 2007 and 2008. Several large financial institutions failed or were rescued. On 15 September 2008 Lehman Brothers filed for bankruptcy, and in the weeks that followed credit markets seized up and share prices fell sharply around the world. On 29 September 2008, after the US House of Representatives initially rejected a bank rescue plan, the Dow fell 777.68 points, at the time its largest points fall ever.

The S&P 500 closed at 676.53 on 9 March 2009, a fall of about 57% from its 2007 peak. Central banks cut interest rates sharply and began large-scale asset purchases, known as quantitative easing, and governments recapitalised banks.

What came next

The S&P 500 closed above its 2007 high on 28 March 2013, about five and a half years after the peak.

The lesson

Correlation. In the crisis, almost everything linked to economic growth fell together: shares in every country, commodities, and higher-risk currencies. Traders who thought they were diversified because they held long positions in several indices found that they effectively held one large position. The currency correlation guide explains why correlations tend to rise in a crisis, and the same applies to equity indices.

The Covid crash: February to March 2020

What happened

The S&P 500 closed at a record on 19 February 2020. As the Covid-19 pandemic spread and governments imposed lockdowns, it fell about 34% to its closing low on 23 March 2020, only 23 trading days later. It was one of the fastest falls of that size on record. The US market-wide circuit breakers, introduced after 1987, were triggered four times in March 2020, on 9, 12, 16 and 18 March. The VIX volatility index closed above 80 in mid-March, a level last seen during the 2008 crisis. The VIX guide explains what that measure captures.

Liquidity was strained well beyond the stock market. Even US Treasury bonds, normally the deepest market in the world, experienced disorderly trading for a period, and the Federal Reserve cut rates to near zero and announced very large asset purchases and lending programmes.

What came next

The S&P 500 regained its February high on 18 August 2020, about six months after the peak, making it one of the fastest recoveries from a bear market on record.

The lesson

Both the speed of the fall and the speed of the recovery caught traders out. Many who were stopped out or margin-called near the lows were not in the market for the rebound. Being forced out at the worst moment, because of leverage rather than choice, is the most expensive way to experience a crash.

Other crashes worth knowing

• Japan, 1989 onwards. The Nikkei 225 closed at 38,915.87 on 29 December 1989 at the height of an asset bubble, then fell for years. It did not close above that level again until 22 February 2024, more than 34 years later. It is the clearest example that a major index can stay below a previous peak for decades.
• Japan, 5 August 2024. The Nikkei fell 12.4% in one session, its largest points fall on record, as a sharp rise in the yen triggered rapid selling. The Nikkei 225 trading guide covers that episode.
• The 1929 crash and the Great Depression. The most famous crash of all, and the one after which the Dow took about 25 years to regain its 1929 peak. Market structure, regulation and policy were so different then that direct comparisons are limited.

The maths of losses and recoveries

A loss requires a bigger percentage gain to recover, because the gain is calculated on a smaller base. The required gain is 1 ÷ (1 − loss) − 1.

• A 22.6% fall (Dow, 19 October 1987) needs a 29% gain to recover.
• A 34% fall (S&P 500, 2020) needs a 52% gain.
• A 57% fall (S&P 500, 2007 to 2009) needs a 131% gain.
• A 78% fall (Nasdaq Composite, 2000 to 2002) needs a 355% gain.

The same arithmetic applies to a trading account. A trader who loses half the account needs to double what is left to get back to the start. The drawdown and recovery guide works through this in detail, and the drawdown calculator shows the gain needed and the number of trades it might take.

Worked example: a crash day on a leveraged index position

All numbers are illustrative.

Two traders each have a 10,000 euro account and are long an index CFD when the index gaps 20% lower overnight, a move of the same order as Black Monday.

Trader A: sized by leverage

Trader A uses most of the available leverage. At 20:1 on a major index, 10,000 euros of equity allows up to 200,000 euros of exposure; Trader A holds 100,000 euros, 10 times the account.

• Loss from a 20% gap: 100,000 × 20% = 20,000 euros.
• That is twice the account balance.

The stop and the 50% margin close-out rule cannot help: the market opens below both, and the position is closed at the first available price. In the EU and UK, a retail client's loss is limited to the account balance by negative balance protection, so Trader A loses the full 10,000 euros. Without that protection, the trader would owe the remaining 10,000.

Trader B: sized by risk

Trader B risks 1% of the account per trade with a stop 2% below entry.

• Risk budget: 10,000 × 1% = 100 euros.
• Position size: 100 ÷ 2% = 5,000 euros of exposure.
• Loss from a 20% gap: 5,000 × 20% = 1,000 euros.

That is ten times the planned loss and a painful 10% of the account, but Trader B is still trading the next day, with 9,000 euros and the ability to take part in any recovery. The difference between the two traders is not their view of the market. It is position size. The position sizing guide and the instruments position size calculator show how to size from risk rather than from available margin.

Lessons for traders

• Stops are not a ceiling on losses. In a gap, a stop becomes a market order and is filled at the next available price. The gap and weekend risk guide explains how stops behave in gaps and how to size positions for them.
• Spreads widen and liquidity thins. In fast markets, providers widen spreads and execution slows. The cost of getting out rises exactly when you most want to leave.
• Correlations rise. Several long index positions in a crisis behave like one big position. Count your total exposure, not the number of trades.
• Margin requirements can rise. Exchanges and providers may raise margin when volatility jumps, which can trigger margin calls on positions that were comfortable the day before. The margin call and stop-out guide explains how that process works.
• Recoveries can be fast or extremely slow. 2020 recovered in six months; the Nasdaq after 2000 took 15 years; the Nikkei after 1989 took 34. Planning on a quick rebound is not a risk plan.
• Being forced out is the real damage. Most of the permanent harm to traders in crashes comes from leverage that leaves no room to survive the move.

Common mistakes

• Assuming the last crash is the template for the next. 1987 was one day, 2000 to 2002 was a slow grind, 2008 was a credit crisis, and 2020 was a pandemic. Each looked different.
• Treating a crash as a buying signal. Some falls recovered quickly and some took decades. Hindsight makes the lows look obvious.
• Averaging down with leverage. Adding to a losing leveraged position during a slide increases exposure exactly when risk is rising.
• Relying on diversification across stock indices. In a broad crash, most stock indices fall together.
• Using the maximum leverage available. ESMA's 20:1 cap on major indices is a ceiling, not a recommendation. A 5% fall at 20:1 is a 100% loss of the margin committed.

Frequently asked questions

What was the biggest stock market crash in history?

By one-day percentage fall in the Dow Jones Industrial Average, it was Black Monday, 19 October 1987, when the index fell 22.6%. By total decline over a longer period, the 1929 to 1932 crash was far larger. Among modern episodes, the Nasdaq Composite's fall of about 78% from 2000 to 2002 stands out.

How long does it take the stock market to recover from a crash?

It has varied enormously. The S&P 500 regained its pre-Covid high in about six months in 2020, and its 2007 high after about five and a half years. The Nasdaq Composite took 15 years to recover its 2000 peak, and the Nikkei 225 took 34 years to regain its 1989 high. There is no reliable average to plan around.

What are circuit breakers?

Rules that pause trading when an index falls by set amounts. In the US, market-wide circuit breakers halt trading for 15 minutes when the S&P 500 falls 7% or 13% from the previous close during most of the trading day, and close the market for the rest of the day at 20%. They were introduced after the 1987 crash and were triggered four times in March 2020.

Can a stop loss protect me in a crash?

Only partly. A stop loss closes the position when the price reaches its level, but in a gap or a very fast market it is filled at the next available price, which can be far worse. Position size, chosen so that even a large gap would be survivable, is the more dependable protection.

Can I lose more than my deposit in a crash?

For retail CFD clients in the EU and UK, negative balance protection limits losses to the money in the account. Elsewhere, and for products such as exchange-traded futures, losses can exceed the deposit and you may owe the difference. In either case, a leveraged position can lose the whole account in a single session. Most retail CFD accounts lose money.

Do all stock markets crash at the same time?

Not always, but in the largest crises they usually fall together. In 1987, 2008 and 2020, markets around the world fell sharply within days of each other. Local crashes also happen, such as Japan after 1989, so the timing and depth can differ by country.