Markets crash. They always have. Since 1929, US stocks have fallen more than 20% from a peak — the common definition of a bear market — over a dozen times, and more than 40% on several occasions. If you invest for decades, you will live through a few. That's not pessimism; it's arithmetic.
This chapter walks through five famous collapses: 1929, 1987, the dot-com bust, 2008, and 2020. The goal isn't to frighten you, and it's definitely not to teach you to predict the next one — nobody has managed that reliably, and this chapter won't pretend to. The goal is calibration. Investors who know what crashes look like, how long recoveries have taken, and what panicking cost people in the past are much harder to shake out at the bottom.
History is the closest thing to a vaccine that investing offers. It doesn't prevent the disease. It softens the symptoms.
The crash of 1929 remains the benchmark for catastrophe. After a decade of speculation — much of it with borrowed money — the Dow Jones Industrial Average peaked at 381 in September 1929. By July 1932 it sat at 41, a fall of 89%. A dollar invested at the top was worth about eleven cents at the bottom.
The recovery is the sobering part. The Dow didn't close above its 1929 peak again until November 1954 — twenty-five years later. The full story is somewhat kinder: prices of everything fell during the Depression, and dividends kept paying, so an investor who reinvested dividends recovered real purchasing power in roughly half that time. Still slow. Still brutal.
The deepest lesson isn't about stocks; it's about leverage. Investors then could buy shares with as little as 10% down, and when prices dipped, forced selling fed on itself. Nearly every modern rule limiting borrowed money in markets descends from that spiral.
October 19, 1987 — Black Monday — remains the worst single day in US stock market history. The Dow fell 22.6% between the opening and closing bells. Nothing in that morning's news explained it; a mix of automated portfolio-insurance selling, crowd panic, and overwhelmed trading systems turned a bad day into a historic one.
Here's what most people forget: the market finished 1987 slightly up for the year. The crash mostly erased a big earlier rally, and within about two years prices had regained their pre-crash peak. An investor who slept through 1987 entirely saw little damage on their year-end statement. An investor who sold in terror on that Monday locked in the loss and then had to decide, every day afterward, when it felt safe to return. It rarely feels safe.
1987's lesson is narrow and useful: a crash isn't always a message about the economy. Sometimes the market's machinery just breaks. No recession followed.
The late 1990s convinced millions that the internet had repealed the old rules of valuation. Companies with no profits — sometimes no revenue — commanded billions. The tech-heavy NASDAQ index roughly quintupled in five years and peaked at 5,048 in March 2000.
Then gravity returned, in slow motion. There was no single crash day; the NASDAQ just fell, rallied, and fell again for two and a half years, finally bottoming 78% below its peak in October 2002. The broader S&P 500 lost about 49%. The NASDAQ didn't see 5,000 again until 2015 — fifteen years later.
Two details deserve memory. First, the internet really did change the world — the visionaries were right about the technology and still lost fortunes, because price matters even when the story is true. Second, boring diversification worked: while tech cratered, many unloved value stocks and bonds actually rose. Owning both halves turned a catastrophe into an inconvenience.
2008 was different: the crash came from inside the financial system. Years of reckless mortgage lending had been packaged into complex securities that big institutions held while barely understanding them. When US house prices fell, losses cascaded, and in September 2008 Lehman Brothers — a 158-year-old investment bank — went bankrupt. Credit froze worldwide.
The S&P 500, which had peaked in October 2007, ultimately fell 57%, bottoming in March 2009. Unlike 1987, this crash reflected real economic devastation: millions of jobs and homes were lost, in the deepest downturn since the Depression.
Recovery took about five and a half years — the S&P 500 regained its 2007 peak in early 2013, sooner counting reinvested dividends. The investors who fared worst weren't those who held through the fall; they were those who sold near the bottom and stayed out during 2009's violent rebound, when the index rose more than 60% off its March low by year-end. Missing the recovery, not enduring the decline, did the lasting damage.
The COVID crash of 2020 was the fastest bear market ever: the S&P 500 fell 34% in about five weeks in February and March as the world economy deliberately shut down. Then, propelled by massive government support and rate cuts, it staged the fastest recovery ever — back at record highs by August, roughly five months after the bottom.
Line the five up and patterns emerge. Every crash felt like the end of the world in real time. Every one produced confident voices declaring stocks dead. Every one, so far, eventually gave way to new highs — but recovery time has ranged from five months to twenty-five years. That range is the honest headline. The comforting slogan that markets always come back has been true in the US so far, but the wait has sometimes outlasted a career, and some other countries' markets have waited even longer. Nothing guarantees the next recovery resembles the last.
So hold cash you'll need soon outside stocks, size positions so you could survive a 50% drawdown, and distrust anyone selling crash predictions. The forecasters who called one crash almost never called the next.
Previous: Macro Indicators That Move Markets · Next: Machine Learning in Investing: Promise vs Reality · Financial glossary
← Back to all investing concepts