£827M
of hidden losses when the bank fell
88888
the error account where losses lived for years
£1
what ING paid for all of Barings

Barings was not a bank that took stupid risks. Founded in 1762, it had financed the Louisiana Purchase and survived Napoleon, two world wars and the Great Depression. In 1995 it was destroyed in three days — not by a market crash, but by the oldest trading mistake in the book, compounded quietly for four years by one man whom nobody checked.

The star nobody audited

Nick Leeson arrived in Singapore in 1992, at 25, to run Barings’ operation on SIMEX, the futures exchange. His official job was low-risk: arbitraging tiny price differences in Nikkei 225 futures between Singapore and Osaka. On paper the desk printed steady, safe profits, and London loved him for it.

The fatal detail was structural, not personal: Leeson ran both the trading desk and the back office. He placed the trades, and he settled and reported the same trades. The person generating the risk was the person auditing the risk. Every disaster that followed fit through that one gap.

Account 88888

Soon after arriving, Leeson opened an internal “error account” — number 88888 — nominally for booking small clerical mistakes. It became the place where losing trades went to hide. A junior’s £20,000 error went in first. Rather than report it, Leeson traded to win it back. The recovery trades lost too, and went into the same account.

Here is the mechanism worth studying, because it is the same mechanism that empties retail accounts every week: a loss was treated as something to be won back, not something to be taken. Each losing position was doubled, on the logic that the market “had to” come back. By the end of 1992, account 88888 hid £2 million. By the end of 1994 — £208 million. All the while, the official reports showed a profitable desk, and London wired margin money whenever Singapore asked.

Averaging into a losing position does not reduce a mistake. It converts a mistake into a destiny.

Selling volatility with no way out

To generate cash for the mounting margin calls, Leeson began selling straddles on the Nikkei — collecting premium for simultaneously selling calls and puts. A short straddle earns a little money as long as the market stays calm, and loses without limit the moment it does not. It is a bet that nothing will happen, financed by everything you have.

On 17 January 1995 the Kobe earthquake hit Japan. The Nikkei broke. The short puts exploded into losses, and the futures position Leeson had built to “support” the market — tens of thousands of long contracts, bought as the index fell — grew in the exact direction of the damage. In the final weeks he was no longer trading a view. He was buying because he was already long, the definition of a trader who has replaced analysis with hope.

Three days in February

On 23 February 1995 Leeson left a note on his desk — “I’m sorry” — and fled with his wife. The losses in account 88888 came to £827 million as at 27 February 1995 — and £927 million once every position had been closed out — roughly twice the bank’s available capital. On 26 February, Barings was declared insolvent. The Dutch bank ING bought two centuries of history for one pound and assumed the liabilities. Leeson was arrested in Frankfurt and served four years in a Singapore prison.

What this actually teaches

The comfortable reading is “one rogue trader”. The useful reading is that every safeguard that failed at Barings has a personal-account equivalent:

  • Leeson audited himself. So does every trader without a written plan and a journal. If the only person checking your trades is the person who wants them to win, your risk control is a mood. A journal is the retail version of an independent back office.
  • The first small loss was never taken. £20,000 became £827 million through nothing more exotic than refusing to be wrong. A stop-loss is not a technical tool; it is the pre-agreed price of being wrong, decided while you are still sane.
  • Size grew as losses grew. The correct direction is the opposite one. Position size is a function of your stop distance and your account — a calculation, not a feeling. (Do it with the position size calculator; it takes thirty seconds.)
  • He sold unlimited risk to fund limited problems. Collecting option premium without a defined exit is renting out the roof of your house during a lightning storm. If a position has no price at which it is closed, it is not a position — it is a hostage situation.

Barings had survived 233 years of markets. It did not survive four years of one unwatched trader refusing to take a small loss. The account size is different from yours; the mechanism is not.

Not financial advice. Everything on this page is educational — history, simulations, and reasoning, not recommendations. It is not a signal service and not investment advice. Trading futures and options carries a substantial risk of loss. Never risk money you cannot afford to lose.