# How Stock Market Crashes Happen

> On 19 October 1987 the Dow Jones Industrial Average fell 22.6% in a single day. How can a market lose so much when no factory or farm has changed? This film explains the machinery: what a share is, how an order book sets a price, why expectations and interest rates move it, how bubbles and herd behaviour build, and how borrowed money turns a fall into a cascade of forced selling. Then it follows the great crashes, told from the official record: 1929 and the Great Depression, Black Monday in 1987, the 1997 Asian financial crisis that began in Thailand, the dot-com bust, the 2008 financial crisis and the 2020 pandemic crash, and the circuit breakers that now pause trading in the United States and Thailand. Send a wave of sell orders through an order book, switch on margin calls, replay five crashes and run a crash day against the brakes. The film is education only: it gives no investment advice and makes no predictions.

Interactive film: https://www.planetpast.com/stories/how-stock-market-crashes-happen/  
Publisher: PlanetPast (https://www.planetpast.com/) · d => T('readPublished', l, { d: dateIn(d, l) }) 2026-10-20 · Society, History

## Prologue

### When everyone sells at once
*Markets and history*

How stock market crashes happen.

How can a market lose a fifth of its value in one day, when no factory, farm or office in the world has changed? On the nineteenth of October 1987, the Dow Jones, America's best-known share index, fell twenty-two point six percent, the largest one-day percentage fall in its history. This film explains the machinery behind it. It is not investment advice and makes no predictions.

- **−22.6%**: Dow Jones Industrial Average in one day
- **−508**: index points

![Nothing physical is destroyed on a day like this: what collapses is the price people will pay. AI-generated still.](https://assets.planetpast.com/a/44/44c5cef225b9ca5e.jpg)

- how prices are set, and how they fall
- six crashes from the official record
- no investment advice, no forecasts

Prologue · 39 years ago

#### Black Monday

Federal Reserve History: on 19 October 1987 the Dow Jones Industrial Average fell 508 points, or 22.6 percent, the largest one-day percentage decline in its history. The Boston Globe’s Today in History gives the same figures.

This film is education only. It is not investment advice and it does not predict what any market will do.

The falling line on screen is a simulated path to the real day’s 22.6% fall, not a minute-by-minute record. There were no market-wide circuit breakers yet; they came after this crash.

How to use: pick a chapter or press ← →; drag to rotate. Several steps have buttons you can press.

Sources: [Federal Reserve History: Stock Market Crash of 1987](https://www.federalreservehistory.org/essays/stock-market-crash-of-1987); [Boston Globe, Today in History, 19 October: Black Monday on Wall Street](https://www.bostonglobe.com/2025/10/19/nation/today-history-october-19-black-monday-wall-street)

## Chapter I · The machine

### A piece of a company
*Chapter I · The machine*

What you own when you own a share.

A share is a small piece of ownership in a company: a claim on a slice of its profits. Companies sell new shares to raise money. After that, investors trade the shares among themselves, and the company gets nothing from those trades. That trading happens on a stock exchange, an organised market for buyers and sellers.

- equity: part-ownership of the company
- a claim on its assets and profits
- usually votes, e.g. electing directors

![Primary market: new shares out, money in. Secondary market: the same shares change hands between investors. Illustration, AI-generated.](https://assets.planetpast.com/a/4c/4c1154235de3eb00.jpg)

> An organized market for the sale and purchase of securities such as shares, stocks, and bonds. (Britannica Money, “stock exchange”)

Chapter I · The machine

#### What a share is

Investor.gov, the US Securities and Exchange Commission’s site for investors, defines a share of stock as an ownership position (equity) in a corporation and a claim on its proportional share of the corporation’s assets and profits; most stocks also carry votes, for example to elect directors. Britannica Money explains that companies raise capital by selling new securities in the primary market, such as an initial public offering, while later trades happen in the secondary market, long after the issuer received its money.

The step’s label puts it in the 1920s, the boom this film turns to next; the mechanics are the same today.

Sources: [Investor.gov (US SEC): glossary, stock](https://www.investor.gov/glossary/glossary_terms/stock); [Britannica Money: primary vs secondary markets](https://www.britannica.com/money/primary-vs-secondary-markets); [Britannica Money: stock exchange](https://britannica.com/money/topic/stock-exchange-finance)

### Where a price comes from
*Chapter I · The machine*

Two queues, and the place they meet.

Every share on an exchange has two queues. Buyers post bids, the most each will pay. Sellers post asks, the least each will accept. This list of waiting orders is the order book. A trade happens where a bid meets an ask: that is the quoted price. A seller in a hurry takes the best bid; when those buyers run out, the price drops to the next bid. Press Sell wave to try.

- bid: the highest price a buyer will pay
- ask: the lowest price a seller will accept
- spread: the gap between them

![Then, orders were shouted on a trading floor; today computers match most of them in microseconds. AI-generated still.](https://assets.planetpast.com/a/50/502ff65b479b7e8a.jpg)

- **100**: starting price
- **1 step**: each time a level of bids is used up

- the best bid has moved down
- the last trade is the new price
- nothing about the company changed

Chapter I · The machine

#### Bids, asks and the order book

Investor.gov: the bid is the highest price a buyer will pay for a given number of shares at a given time; the ask is the lowest price a seller will accept; the difference is the spread. Britannica Money notes that electronic networks now match buy and sell orders directly.

The order book on screen is a simplified model with invented orders, for teaching.

Try it: Sell wave sends market sell orders, which take the best bids first; Buy wave does the opposite; Reset restores the book.

Sources: [Investor.gov (US SEC): glossary, bid price and ask price](https://www.investor.gov/introduction-investing/investing-basics/glossary/bid-price); [Britannica Money: stock exchange](https://britannica.com/money/topic/stock-exchange-finance)

### A bet on the future
*Chapter I · The machine*

Prices move when expectations do.

What moves the queues? A share's value rests on the profits people expect the company to earn, and on what that future money is worth today. When banks pay more interest, money promised years from now looks less attractive than cash in the bank, so share prices tend to fall. To follow many shares at once, people use an index: one number tracking a group of shares, such as the Dow or Thailand's SET.

$$\text{price} \approx \sum_{t} \frac{\text{expected income}_t}{(1+r)^t}$$
- $t$: years from now
- $r$: the rate future money is discounted at; higher rates shrink the sum

Britannica: ideally a share’s price equals the present value of its expected income stream.

- **↑ 0.01%**: five-year interest rates, pushed up by policy
- **↓ ≈0.036%**: stock prices, on average

- **30**: Dow Jones Industrial Average, price-weighted, since 1896
- **500**: S&P 500, weighted by market value
- **1975**: SET Index base year (= 100)

Chapter I · The machine

#### Expectations, interest rates and indexes

Britannica Money: ideally, the price of a share equals the present value of its expected stream of income. A 2022 study summarised by the Federal Reserve Bank of Boston found that a one-basis-point rise in policy-driven five-year interest rates lowers stock prices by about 3.56 basis points on average. S&P Dow Jones Indices: the Dow has 30 components and is price-weighted (launched May 1896); the S&P 500 has 500 and is weighted by float-adjusted market capitalisation. The SET Index covers the common stocks on the Stock Exchange of Thailand’s main board, weighted by market capitalisation, with 30 April 1975 = 100.

The order book on screen is a simplified model with invented orders, for teaching.

Try it: Good news and Bad news shift the whole book up or down, the way a change in expectations does.

Sources: [Britannica Money: stock exchange](https://britannica.com/money/topic/stock-exchange-finance); [Federal Reserve Bank of Boston (2022): monetary policy and stock prices](https://www.bostonfed.org/publications/current-policy-perspectives/2022/monetary-policy-and-stock-prices.aspx); [S&P Dow Jones Indices: the S&P 500 and the Dow](https://www.spglobal.com/spdji/en/index-literacy/the-sp-500-and-the-dow/); [Wikipedia: SET Index (base 1975 = 100, capitalisation-weighted; year-end closes)](https://en.wikipedia.org/wiki/SET_Index)

## Chapter II · Bubbles and borrowed money

### Expectations are contagious
*Chapter II · Bubbles and borrowed money*

A loop that runs both ways.

Expectations are contagious. When prices keep rising, more people buy simply because prices are rising, lifting them far above what profits can justify. That is a bubble. The loop also runs in reverse: falling prices frighten people into selling, which pushes prices lower still. Following the crowd like this is called herd behaviour.

![Britannica: rare bulbs rose to “unwarranted heights”, families mortgaged homes to buy them, and prices collapsed early in 1637. AI-generated still.](https://assets.planetpast.com/a/83/839953320ff47be3.jpg)

> irrational exuberance (a phrase Britannica credits to Federal Reserve chair Alan Greenspan)

- green: buying because the price rose
- red: selling because the price fell
- the pillar: the price they make together

Chapter II · Bubbles and borrowed money

#### Bubbles and herd behaviour

Britannica describes Holland’s tulip mania (about 1633–37), when rare bulbs reached unwarranted heights, families mortgaged homes to buy them for resale, and prices collapsed early in 1637. Its account of the dot-com bubble describes investor overconfidence, often called “irrational exuberance” (a phrase it credits to Federal Reserve chair Alan Greenspan), pushing share prices far above traditionally justified values, with many investors believing traditional measures of value no longer applied.

The crowd on screen is a schematic, not a simulation of any real market.

Sources: [Britannica Money: tulip mania](https://www.britannica.com/money/Tulip-Mania); [Encyclopaedia Britannica: dot-com bubble](https://www.britannica.com/event/dot-com-bubble)

### Buying on margin
*Chapter II · Bubbles and borrowed money*

How a fall becomes a cascade.

Borrowed money makes the loop stronger. Buying on margin means paying part of the price and borrowing the rest from a broker. In the late 1920s many buyers paid about ten percent and borrowed the rest. Then a ten percent fall wipes out their own money. When prices drop, the broker makes a margin call: add cash, or your shares can be sold without warning. Press With margin calls to see what forced sales do.

- own money: 10
- borrowed from the broker: 90
- price paid: 100

- share now worth: 90
- still owed: 90
- own money left: 0

- the broker can demand cash or securities at once
- it can sell your holdings without contacting you
- it chooses which holdings to sell

- **4 levels**: where margin calls force a sale

Chapter II · Bubbles and borrowed money

#### Margin and forced selling

Federal Reserve History: in the late 1920s investors typically paid about 10 percent of a stock’s price and borrowed the rest, with the stock as collateral; loans to brokers fuelled the euphoria. Investor.gov: if securities bought on margin fall in value, the brokerage can require more cash or securities immediately and can sell your securities without contacting you.

The order book on screen is a simplified model with invented orders, for teaching.

Try it: the same sell wave with and without borrowed money. The orange marks are the prices at which leveraged buyers are sold out.

Sources: [Federal Reserve History: Stock Market Crash of 1929](https://www.federalreservehistory.org/essays/stock-market-crash-of-1929); [Investor.gov (US SEC): glossary, margin call](https://www.investor.gov/introduction-investing/investing-basics/glossary/margin-call)

## Chapter III · 1929

### The great crash
*Chapter III · 1929*

Black Thursday, Black Monday, Black Tuesday.

On Thursday the twenty-fourth of October 1929, a record of nearly thirteen million shares changed hands. Big banks bought stock to stop the panic, briefly. Then on Monday the twenty-eighth, the first Black Monday, the Dow fell nearly thirteen percent, and on Tuesday nearly twelve more. By mid-November it had lost almost half.

![Crowds gathered on Wall Street while the ticker, the machine that printed prices, ran far behind the trading. AI-generated still, not a photograph.](https://assets.planetpast.com/a/4d/4d3b4540a7b2ef7a.jpg)

- **12.9m**: shares traded
- **+4m**: more than the old one-day record

- 3 Sep: record close, 381.17
- 24 Oct: Black Thursday: ~12.9 million shares
- 28 Oct: the first Black Monday: nearly −13%
- 29 Oct: Black Tuesday: nearly −12%
- mid-Nov: almost half the value gone

- **≈ −50%**: Dow, about ten weeks

Chapter III · 1929

#### October 1929

Federal Reserve History: the Dow peaked on 3 September 1929 at 381.17; on Black Monday, 28 October, it declined nearly 13 percent, and the next day nearly 12 percent; by mid-November it had lost almost half its value. Britannica: on Black Thursday, 24 October, nearly 12.9 million shares were traded, almost 4 million more than the previous record, and major banks bought large amounts of stock in a briefly successful effort to stem the panic.

Crash curves are schematic: only the marked points come from the sources; the line between them is smoothed. Each curve is scaled so the level before the fall is 100.

Sources: [Federal Reserve History: Stock Market Crash of 1929](https://www.federalreservehistory.org/essays/stock-market-crash-of-1929); [Encyclopaedia Britannica: Black Thursday (24 October 1929)](https://britannica.com/print/article/68373)

### From a crash to a depression
*Chapter III · 1929*

Twenty-five years to climb back.

The fall did not stop there. By July 1932, the Dow stood eighty-nine percent below its peak. The crash cut spending and firms laid off workers. Then failures elsewhere in the financial system turned a recession, an ordinary downturn, into the longest, deepest depression in American history. The Dow did not regain its 1929 peak until November 1954.

- **381.17**: close, 3 Sep 1929
- **41.22**: close, 8 Jul 1932
- **−89%**: peak to bottom

![Queues for free food became a symbol of the 1930s in American cities. AI-generated still, not a photograph.](https://assets.planetpast.com/a/fc/fcd3a11af00160cb.jpg)

- Sep 1929: peak
- Jul 1932: bottom, lowest close of the century
- Nov 1954: back at the 1929 peak

Chapter III · 1929

#### The crash and the Great Depression

Federal Reserve History: the Dow closed at 41.22 on 8 July 1932, its lowest value of the twentieth century and 89 percent below its peak, and did not return to its pre-crash height until November 1954. The crash reduced spending on big-ticket items, firms slowed production and furloughed workers, and problems elsewhere in the financial system turned “a short, sharp recession into our nation’s longest, deepest depression”.

Crash curves are schematic: only the marked points come from the sources; the line between them is smoothed. Each curve is scaled so the level before the fall is 100.

Sources: [Federal Reserve History: Stock Market Crash of 1929](https://www.federalreservehistory.org/essays/stock-market-crash-of-1929)

## Chapter IV · Black Monday

### Selling by formula
*Chapter IV · Black Monday*

A crash at the speed of computers.

Fifty-eight years later, a crash came faster. After a fall on Friday the sixteenth of October 1987, sell orders swamped the exchanges on Monday. The official inquiry, the Brady report, gave a big role to portfolio insurance: computer models telling large funds to sell automatically as prices fell. They sold index futures, contracts on the index's future level, and each sale pushed prices down, triggering more. Other traders carried the selling into the shares.

![Stocks then settled in three days, futures and options in one, so cash fell out of step between markets (Federal Reserve History). AI-generated still.](https://assets.planetpast.com/a/07/073f8d1db32af503.jpg)

- **$4bn**: futures sold by portfolio insurers
- **>40%**: of futures volume, excluding floor traders dealing for themselves (“locals”)
- **~$2bn**: stock sold by three insurers, of ~$21bn on the NYSE

> An initial price decline starts a vicious circle by causing portfolio insurers to sell, causing further price declines… (Economist Robert Shiller (1988), quoted by Federal Reserve History)

Chapter IV · Black Monday

#### Portfolio insurance and program trading

The Brady report (Presidential Task Force on Market Mechanisms, 1988) found that the decline “ignited mechanical, price-insensitive selling by a number of institutions”, that portfolio insurance “played a prominent role”, and that index arbitrage was the “transmission mechanism” that carried the pressure from futures to stocks. It also cites worse-than-expected trade figures and proposed tax legislation as triggers. Federal Reserve History adds that the Dow fell 4.6 percent on Friday 16 October, and that stocks settled in three days while futures and options settled in one.

Index arbitrage traders profit from gaps between the price of index futures and the prices of the shares in the index, so heavy selling of futures pulls them into selling shares. Federal Reserve History quotes Robert Shiller describing the 1987 loop as a “vicious circle”.

These are the leading explanations given by the official inquiry; economists still debate how much each mattered.

The order book on screen is a simplified model with invented orders, for teaching.

Sources: [Report of the Presidential Task Force on Market Mechanisms (Brady Report), January 1988](https://pages.nes.ru/aobizhaeva/BradyReport_1988_on_market_crash.pdf); [Federal Reserve History: Stock Market Crash of 1987](https://www.federalreservehistory.org/essays/stock-market-crash-of-1987)

### The day after
*Chapter IV · Black Monday*

A promise of cash, and a new kind of brake.

That Monday, markets around the world fell too. On Tuesday the Federal Reserve said it stood ready to supply cash to the financial system, and urged banks to keep lending. The Dow won back over half its loss in two days and passed its pre-crash high within two years. Regulators then added circuit breakers, rules that pause trading after big falls.

- **−60%**: New Zealand, the most severe fall
- **first**: Asian markets fell before New York opened

> …affirmed today its readiness to serve as a source of liquidity to support the economic and financial system. (Federal Reserve Chairman Alan Greenspan)

- **+288**: Dow points in two sessions: 57% of the loss
- **< 2 yrs**: to pass the pre-crash highs

- circuit breakers: market-wide halts after big falls
- paying for and delivering shares, futures and options on one timetable

Chapter IV · Black Monday

#### The Federal Reserve’s response

Federal Reserve History: on 20 October 1987 Chairman Alan Greenspan said the Federal Reserve “affirmed today its readiness to serve as a source of liquidity to support the economic and financial system”, and the Fed encouraged banks to keep lending on normal terms. Asian markets fell before the US opened and New Zealand’s fell 60 percent. In two sessions the Dow regained 288 points, 57 percent of the Black Monday loss, and US stocks surpassed their pre-crash highs in less than two years. Regulators then created rules that allow exchanges to halt trading after large declines and standardised clearing and settlement; the Brady report had recommended “circuit breaker mechanisms (such as price limits and coordinated trading halts)”.

Crash curves are schematic: only the marked points come from the sources; the line between them is smoothed. Each curve is scaled so the level before the fall is 100.

Sources: [Federal Reserve History: Stock Market Crash of 1987](https://www.federalreservehistory.org/essays/stock-market-crash-of-1987); [Report of the Presidential Task Force on Market Mechanisms (Brady Report), January 1988](https://pages.nes.ru/aobizhaeva/BradyReport_1988_on_market_crash.pdf)

## Chapter V · Baht and dot-coms

### The crisis that began in Thailand
*Chapter V · Baht and dot-coms*

Borrowed dollars and a currency peg: the Tom Yum Kung crisis.

In 1997 borrowed money struck again, in Thailand. Firms and banks had borrowed heavily in foreign currency, much of it short term, trusting the baht's fixed link to the dollar. On the second of July 1997, after months of traders betting against the baht had drained the central bank's dollar reserves, Thailand devalued it, and foreign debts suddenly cost far more. Remembered in Thailand as the Tom Yum Kung crisis, it spread across Asia, and the SET Index ended 1997 fifty-five percent down.

![Federal Reserve History names overheated real-estate markets and rapid credit growth among the weak points. AI-generated still.](https://assets.planetpast.com/a/16/160cd5fd5163ed47.jpg)

![Long-standing currency pegs had masked the exchange-rate risk of foreign borrowing (Federal Reserve History). AI-generated still.](https://assets.planetpast.com/a/b0/b0d39a28c452203b.jpg)

- **−55.17%**: SET Index in 1997: 831.57 → 372.69 (SCMP)
- **$118bn**: loans mobilised for Thailand, Indonesia and South Korea

- gold figures: foreign lenders pulling back
- the pillar: share prices in Bangkok

Chapter V · Baht and dot-coms

#### The 1997 Asian financial crisis

Federal Reserve History: on 2 July 1997, after months of speculative pressure had substantially depleted its foreign-exchange reserves, Thailand devalued its currency relative to the US dollar. Rapid credit growth, heavy short-term foreign borrowing and long-standing pegs that masked exchange-rate risk left firms exposed; when the pegs broke, the local-currency cost of foreign debt surged. Malaysia, the Philippines and Indonesia let their currencies weaken substantially, South Korea came close to default, foreign creditors pulled back, and about $118 billion in loans was mobilised for Thailand, Indonesia and South Korea. The South China Morning Post reported the SET Index down 55.17% over 1997, from 831.57 to 372.69. In Thailand the crisis is widely called the “Tom Yam Kung” (Tom Yum Kung) crisis, and the Bangkok Post dates the baht’s float to 2 July 1997.

The 1997 crisis and the IMF programmes that followed are told in more detail in this site’s film on the IMF and the World Bank.

Sources: [Federal Reserve History: Asian Financial Crisis (1997–98)](https://www.federalreservehistory.org/essays/asian-financial-crisis); [South China Morning Post (Dec 1997): SET Index down 55.17% in 1997, 831.57 to 372.69](https://www.scmp.com/article/224927/rupiah-set-index-worst-97); [Bangkok Post: the lowdown on the 1997 financial crisis (“Tom Yam Kung” crisis; baht floated 2 July 1997)](https://www.bangkokpost.com/business/general/2740424/the-lowdown-on-the-1997-financial-crisis)

### The dot-com bust
*Chapter V · Baht and dot-coms*

A bubble in the future of the internet.

Then the expectations loop built a bubble around the internet. Investors poured money into young online companies, many with no profits. The technology-heavy Nasdaq index peaked in March 2000, fell about seventy-seven percent by October 2002, and needed fifteen years to close above that peak again.

![Britannica: the sell-off began in early 2000, after the Federal Reserve announced a modest interest-rate rise. AI-generated still.](https://assets.planetpast.com/a/87/8738dab28de2ff72.jpg)

- 10 Mar 2000: record close, 5,048.62
- Oct 2002: 1,139, about −77%
- 23 Apr 2015: new record, 5,056.06

- **15 yrs**: to a new record close, 23 April 2015
- **>75%**: the fall at the bottom, as NPR put it

Chapter V · Baht and dot-coms

#### The dot-com bubble

Britannica: the Nasdaq Composite reached 5,048 in March 2000 and fell to 1,139 by October 2002 (about 77 percent lower), erasing nearly all of the bubble’s gains; investors had funded many debt-ridden companies with no realistic path to profit, and the sell-off began after the Federal Reserve announced a modest rate increase. NPR, via WAMC: on 23 April 2015 the Nasdaq closed at 5,056.06, above its previous record close of 5,048.62 on 10 March 2000.

Crash curves are schematic: only the marked points come from the sources; the line between them is smoothed. Each curve is scaled so the level before the fall is 100.

Sources: [Encyclopaedia Britannica: dot-com bubble](https://www.britannica.com/event/dot-com-bubble); [WAMC / NPR (23 Apr 2015): 15 years after the dot-com bust, Nasdaq closes at a new record](https://www.wamc.org/2015-04-23/15-years-after-the-dot-com-bust-nasdaq-closes-at-new-record)

## Chapter VI · Banks and a virus

### From houses to banks
*Chapter VI · Banks and a virus*

When the lenders themselves fall.

In 2008 borrowing met housing. American home prices had more than doubled from 1998 to 2006, fed in part, says the Federal Reserve's history, by home loans bundled into securities and sold worldwide. When home prices fell, firms holding those securities lost heavily, and on the fifteenth of September 2008 the investment bank Lehman Brothers went bankrupt. From its 2007 peak to March 2009 the S&P 500 fell almost fifty-seven percent.

- **×2+**: home prices, 1998–2006
- **61% → 97%**: mortgage debt as a share of GDP

![The next day the Federal Reserve stepped in to support the insurer AIG (Federal Reserve History). AI-generated still.](https://assets.planetpast.com/a/d9/d92234adcc508826.jpg)

- 9 Oct 2007: record close, 1,565.15
- 9 Mar 2009: 676.53, −56.8%
- 28 Mar 2013: above the 2007 record

Chapter VI · Banks and a virus

#### The 2008 financial crisis

Federal Reserve History: US home prices more than doubled between 1998 and 2006 and household mortgage debt rose from 61% to 97% of GDP; observers point both to low interest rates after 2001 and to the growth of mortgage-backed securities. Home prices peaked in early 2007. Lehman Brothers filed for bankruptcy on 15 September 2008 (HISTORY.com); the next day the Fed supported AIG. The recession lasted from December 2007 to June 2009. The S&P 500 closed at 1,565.15 on 9 October 2007, at 676.53 on 9 March 2009 (56.8% lower), and passed its 2007 record on 28 March 2013.

Crash curves are schematic: only the marked points come from the sources; the line between them is smoothed. Each curve is scaled so the level before the fall is 100.

Sources: [Federal Reserve History: The Great Recession and its aftermath](https://www.federalreservehistory.org/essays/great-recession-and-its-aftermath); [HISTORY.com: Great Recession timeline (Lehman Brothers files for bankruptcy, 15 Sep 2008)](https://www.history.com/articles/great-recession-timeline); [Wikipedia: closing milestones of the S&P 500 (2007 peak, 2009 low, 2013 recovery, 2020)](https://en.wikipedia.org/wiki/Closing_milestones_of_the_S%26P_500)

### The fastest fall
*Chapter VI · Banks and a virus*

A pandemic, four halts, and a rebound.

In 2020 the shock was a virus. As the pandemic shut down daily life, the S&P 500 fell thirty-four percent in about five weeks. Prices fell so fast that the circuit breakers, first added after 1987, halted trading four times that March. Then, helped by record-low interest rates and government stimulus, it regained all its losses by August.

![Offices emptied as lockdowns spread; markets kept trading electronically. AI-generated still.](https://assets.planetpast.com/a/5b/5bff9914a57c1c77.jpg)

- 9 Mar: S&P 500 down 7%: 15-minute halt
- 12 Mar: second halt
- 16 Mar: third halt
- 18 Mar: fourth halt

- **−34%**: 19 Feb to 23 Mar
- **3,389.78**: record close, 18 Aug

Chapter VI · Banks and a virus

#### The 2020 crash and rebound

NPR (via Maine Public): the S&P 500 plunged 34% from its 19 February 2020 peak, so fast that automatic circuit breakers halted trading several times; it bottomed on 23 March and closed at a record 3,389.78 on 18 August, as investors bet on recovery supported by record-low interest rates and other stimulus from the Federal Reserve and Congress. Optiver: Level 1 market-wide circuit breakers were triggered on 9, 12, 16 and 18 March 2020.

The intraday chart on screen is a simulated crash day, not a record of any real session.

Sources: [Maine Public / NPR (18 Aug 2020): S&P 500 closes at record high six months after coronavirus plunge](https://www.mainepublic.org/2020-08-18/the-big-comeback-s-p-500-closes-at-record-high-6-months-after-coronavirus-plunge); [Wikipedia: closing milestones of the S&P 500 (2007 peak, 2009 low, 2013 recovery, 2020)](https://en.wikipedia.org/wiki/Closing_milestones_of_the_S%26P_500); [Optiver: five things to know about market-wide circuit breakers (March 2020 triggers; 27 Oct 1997)](https://www.optiver.com/insights/explainers/5-things-you-should-know-about-market-wide-circuit-breakers/)

## Chapter VII · The brakes

### Circuit breakers
*Chapter VII · The brakes*

Pausing the loop.

Today in the United States, if the S&P 500 falls seven percent in a day, trading pauses fifteen minutes; at thirteen percent, again; at twenty, it stops for the day. So a fall like Black Monday's would now be stopped for the day at twenty percent. Press Run a crash day to watch. Thailand's exchange uses eight, fifteen and twenty percent. Its breaker tripped on the fourth of March 2026, and trading resumed in the afternoon session. A pause cannot change what anything is worth, but it gives people time to think.

- Level 1: 15-minute halt: 7%
- Level 2: 15-minute halt: 13%
- Level 3: closed for the day: 20%

- real thresholds, simulated prices
- red band: time the market stood still
- first the US rules, then Thailand’s
- the 1987 fall would stop at Level 3

- Level 1: 30-minute halt: 8%
- Level 2: 30-minute halt: 15%
- Level 3: 1-hour halt: 20%

![Single US stocks have their own pauses too: limit up–limit down bands stop one share for five minutes. Illustration, AI-generated.](https://assets.planetpast.com/a/60/605c0ab5a2cbd8be.jpg)

Chapter VII · The brakes

#### Circuit breakers in the US and Thailand

Investor.gov: market-wide circuit breakers are triggered by a single-day fall in the S&P 500 of 7% (Level 1), 13% (Level 2) or 20% (Level 3) from the prior day’s close. Levels 1 and 2 halt trading for 15 minutes if they occur before 3:25 p.m.; Level 3 halts trading for the rest of the day. The Stock Exchange of Thailand’s April 2026 consultative paper gives its current levels as falls of 8% (30-minute halt), 15% (30 minutes) and 20% (one hour) from the previous close. The SET’s own rule page adds that if less time remains in the session than the halt, trading stays halted until that session ends and resumes in the next one. The April 2026 paper proposes only to drop the requirement that members suspend program trading during a halt; the levels are unchanged. On 4 March 2026 the SET Index fell 117.52 points, or 8.01%, at 12:18; Thansettakij reports trading halted from 12:18 to 12:30 and resumed in the afternoon session, and linked the volatility to the situation in the Middle East. Thai PBS counts six uses of the SET breaker in its history, including in 2008 and 2020.

Try it: Run a crash day plays a simulated day with steadily falling prices; switch between US and Thai rules. Simulated prices, real thresholds.

Sources: [Investor.gov (US SEC): stock market circuit breakers](https://www.investor.gov/introduction-investing/investing-basics/glossary/stock-market-circuit-breakers); [Stock Exchange of Thailand (Apr 2026): consultative paper on circuit breaker rules (current levels 8/15/20%)](https://media.set.or.th/set/Documents/2026/Apr/No19-ConsultativePaper-Rules-on-Circuit-Breaker-en.pdf); [Stock Exchange of Thailand: trading procedure, circuit breaker](https://www.set.or.th/en/market/information/trading-procedure/circuit-breaker); [Thansettakij (4 Mar 2026): SET halts trading 12:18–12:30 after an 8.01% fall](https://www.thansettakij.com/finance/stockmarket/652954); [Thai PBS (4 Mar 2026): SET uses circuit breaker after index falls 8.01%](https://www.thaipbs.or.th/news/content/502919)

## Epilogue

### When everyone sells at once
*Epilogue*

Different triggers, the same machine.

Put the crashes side by side and the triggers all differ, but the machinery repeats: expectations that spread, borrowed money, and sellers forced to sell. Recovery has taken from six months to twenty-five years, and no one can know in advance which kind a crash will be. Press a year to replay each fall and compare how deep it went and how long the climb back took.

- Dow 1929–32: −89%
- Nasdaq 2000–02: ≈ −77%
- S&P 500 2007–09: −56.8%
- S&P 500 2020: −34%
- Dow, 19 Oct 1987 alone: −22.6%

- 1929: 25 yrs
- 2000: 15 yrs
- 2007: 5.5 yrs
- 1987: < 2 yrs
- 2020: 6 months

![All stills and illustrations in this film are AI-generated. Education only: no investment advice and no forecasts.](https://assets.planetpast.com/a/91/91f68fc2ce7a9e70.jpg)

Epilogue · Sources

#### How stock market crashes happen

The figures in this film come from the Federal Reserve’s own histories, the US SEC’s investor site, the 1988 Brady report, Britannica, the Stock Exchange of Thailand and news reports, each named where it is used. Explanations of why a crash happened are given as the sources give them; economists still debate the weight of each cause.

Education only: this film is not investment advice and makes no predictions. Order books and crash days on screen are simplified models; crash curves are schematic between sourced points. All stills are AI-generated.

Sources: [Federal Reserve History: Stock Market Crash of 1987](https://www.federalreservehistory.org/essays/stock-market-crash-of-1987); [Boston Globe, Today in History, 19 October: Black Monday on Wall Street](https://www.bostonglobe.com/2025/10/19/nation/today-history-october-19-black-monday-wall-street); [Investor.gov (US SEC): glossary, stock](https://www.investor.gov/glossary/glossary_terms/stock); [Britannica Money: primary vs secondary markets](https://www.britannica.com/money/primary-vs-secondary-markets); [Britannica Money: stock exchange](https://britannica.com/money/topic/stock-exchange-finance); [Investor.gov (US SEC): glossary, bid price and ask price](https://www.investor.gov/introduction-investing/investing-basics/glossary/bid-price); [Federal Reserve Bank of Boston (2022): monetary policy and stock prices](https://www.bostonfed.org/publications/current-policy-perspectives/2022/monetary-policy-and-stock-prices.aspx); [S&P Dow Jones Indices: the S&P 500 and the Dow](https://www.spglobal.com/spdji/en/index-literacy/the-sp-500-and-the-dow/); [Wikipedia: SET Index (base 1975 = 100, capitalisation-weighted; year-end closes)](https://en.wikipedia.org/wiki/SET_Index); [Encyclopaedia Britannica: dot-com bubble](https://www.britannica.com/event/dot-com-bubble); [Investor.gov (US SEC): glossary, margin call](https://www.investor.gov/introduction-investing/investing-basics/glossary/margin-call); [Federal Reserve History: Stock Market Crash of 1929](https://www.federalreservehistory.org/essays/stock-market-crash-of-1929); [Encyclopaedia Britannica: Black Thursday (24 October 1929)](https://britannica.com/print/article/68373); [Report of the Presidential Task Force on Market Mechanisms (Brady Report), January 1988](https://pages.nes.ru/aobizhaeva/BradyReport_1988_on_market_crash.pdf); [Federal Reserve History: Asian Financial Crisis (1997–98)](https://www.federalreservehistory.org/essays/asian-financial-crisis); [WAMC / NPR (23 Apr 2015): 15 years after the dot-com bust, Nasdaq closes at a new record](https://www.wamc.org/2015-04-23/15-years-after-the-dot-com-bust-nasdaq-closes-at-new-record); [Federal Reserve History: The Great Recession and its aftermath](https://www.federalreservehistory.org/essays/great-recession-and-its-aftermath); [HISTORY.com: Great Recession timeline (Lehman Brothers files for bankruptcy, 15 Sep 2008)](https://www.history.com/articles/great-recession-timeline); [Wikipedia: closing milestones of the S&P 500 (2007 peak, 2009 low, 2013 recovery, 2020)](https://en.wikipedia.org/wiki/Closing_milestones_of_the_S%26P_500); [Maine Public / NPR (18 Aug 2020): S&P 500 closes at record high six months after coronavirus plunge](https://www.mainepublic.org/2020-08-18/the-big-comeback-s-p-500-closes-at-record-high-6-months-after-coronavirus-plunge); [Optiver: five things to know about market-wide circuit breakers (March 2020 triggers; 27 Oct 1997)](https://www.optiver.com/insights/explainers/5-things-you-should-know-about-market-wide-circuit-breakers/); [Investor.gov (US SEC): stock market circuit breakers](https://www.investor.gov/introduction-investing/investing-basics/glossary/stock-market-circuit-breakers); [Stock Exchange of Thailand (Apr 2026): consultative paper on circuit breaker rules (current levels 8/15/20%)](https://media.set.or.th/set/Documents/2026/Apr/No19-ConsultativePaper-Rules-on-Circuit-Breaker-en.pdf); [South China Morning Post (Dec 1997): SET Index down 55.17% in 1997, 831.57 to 372.69](https://www.scmp.com/article/224927/rupiah-set-index-worst-97); [Bangkok Post: the lowdown on the 1997 financial crisis (“Tom Yam Kung” crisis; baht floated 2 July 1997)](https://www.bangkokpost.com/business/general/2740424/the-lowdown-on-the-1997-financial-crisis); [Britannica Money: tulip mania](https://www.britannica.com/money/Tulip-Mania); [Stock Exchange of Thailand: trading procedure, circuit breaker](https://www.set.or.th/en/market/information/trading-procedure/circuit-breaker); [Thansettakij (4 Mar 2026): SET halts trading 12:18–12:30 after an 8.01% fall](https://www.thansettakij.com/finance/stockmarket/652954); [Thai PBS (4 Mar 2026): SET uses circuit breaker after index falls 8.01%](https://www.thaipbs.or.th/news/content/502919)