Loading signal…
Four centuries of 'this time is different.'
Every crash begins the same way. A new thing that can't fail — tulips, railroads, radio, dot-coms, mortgages, crypto. Credit expands. A class of insiders sell early. A class of outsiders buy late. Then something — a bank, a borrower, a country — doesn't pay. The crash is the moment the mathematics asserts itself against the mood.
Dutch tulip bulbs — specifically, the ones infected with a mosaic virus that produced striped 'broken' blooms — became the speculative asset of the 1630s. At the peak, a single Semper Augustus bulb traded for more than a luxurious Amsterdam canal house. Then, one February morning in 1637, a Haarlem auction failed. Prices collapsed 99% inside three months.
The South Sea Company was granted a monopoly on trade with Spanish America — a business that turned out to be almost entirely theoretical. To mask the lack of actual revenue, the company proposed converting British government debt into its own shares. Insiders pumped the price with orchestrated rumors, sold into the retail mania they had manufactured, and walked away with fortunes before the inevitable collapse.
The Dow peaked at 381 in September 1929. It bottomed at 41 in July 1932 — a decline of 89% over 34 months. Recovery to the 1929 peak took until 1954. Margin lending wiped out the middle class. Unemployment reached 25%. One in four US banks failed. The New Deal, modern securities law, the FDIC, and most of the 20th century's economic institutions exist because of this crash.
On October 19, 1987, the Dow fell 22.6% in a single session. It remains the largest one-day percentage decline in the history of the US stock market. Unlike 1929, there was no obvious economic cause. The consensus explanation is a feedback loop between portfolio insurance and program trading — algorithms selling because prices fell, then prices falling because algorithms sold.
On the last trading day of 1989, the Nikkei 225 closed at 38,957. Japanese corporate real estate was valued so highly that the grounds of the Imperial Palace were said to be worth more than all the real estate in California. Over the following twenty years, the index lost 82% of its value. Recovery to the 1989 peak did not happen until 2024 — thirty-four years later.
Between 1995 and March 2000, the Nasdaq rose more than fivefold. Any business with a .com in its name could raise money on a spreadsheet. The thesis was real — the internet would reshape the economy — but the valuations were a different thing entirely. Pets.com, Webvan, Kozmo, eToys. When the cash ran out, it ran out everywhere at once.
The US subprime mortgage market cracked in 2007. It took the investment banking industry with it in 2008. Lehman Brothers filed bankruptcy on September 15; the following two weeks, the global financial system came within hours of seizing. AIG, Fannie Mae, Freddie Mac, Wachovia, and Washington Mutual were rescued in succession. $33T in global wealth was destroyed. Recovery took until 2013.
COVID produced the fastest bear market in history — the S&P 500 fell 34% in 33 days — and the fastest recovery, powered by a $5T Fed and Treasury response. Two years later, the same liquidity bubble that drove the recovery unwound in crypto. Luna/UST lost $60B in three days. FTX — once valued at $32B — turned out to be a fraud. The Nasdaq fell 33%.
Each bar is drawn on a log scale — the relative intensity of that era against the others. On a linear scale, the earliest eras would disappear into a single pixel next to the most recent ones.