One Trader Destroyed a 233 Year Old Bank
Barings collapsed in 1995 after losses hidden in an account that was never reconciled. The failure was a control failure, and the specific control that was missing is now standard everywhere.
The Collapse
Barings Bank was founded in 1762 and was among the oldest merchant banks in Britain. In 1995 it failed, and was subsequently sold for a nominal sum, after losses accumulated by a single trader in its Singapore operation exceeded the bank's entire capital.
The losses came from unauthorized positions in Japanese equity index futures that were concealed in an error account, and they grew larger as the trader attempted to trade out of them.
The Control That Was Missing
The central failure is simple enough to state in a sentence. The trader was responsible both for executing trades on the floor and for the back office function that settled and reconciled them.
Segregation of duties exists precisely to prevent this. The person taking risk must not be the person who records and verifies it, because otherwise the record can be made to say whatever is convenient. Every reconciliation designed to catch an error was performed by the person creating it.
An institution that lets one person both take positions and confirm them has not weakened a control. It has removed the control entirely.
How the Losses Grew
The pattern is common in trading disasters. Initial losses, reportedly modest and partly attributable to junior staff errors, were concealed rather than reported.
Concealment creates a trap. The only way to make the hidden loss disappear is to earn it back, which requires taking more risk, and if that fails the hidden position is larger. Each step is individually rational for someone trying to avoid discovery and collectively catastrophic.
Positions eventually became very large bets that the Japanese market would remain stable. An earthquake in Kobe in January 1995 moved the market sharply, and the positions moved decisively against him.
Why Nobody Above Him Noticed
The operation appeared to be enormously profitable, and reported profits from the Singapore desk were a substantial contributor to the division's results. Profitable operations attract less scrutiny than struggling ones, which is exactly backwards from what risk management requires.
The reported strategy was also supposed to be low risk arbitrage between related futures contracts on different exchanges. A genuine arbitrage strategy should not generate outsized profits, and outsized profits from a strategy described as riskless is itself a red flag that management did not interrogate.
Funding requests should also have raised questions. The Singapore operation was requesting very large sums from London to meet margin calls, which is inconsistent with a hedged arbitrage book where positions largely offset.
What It Changed
The failure accelerated the adoption of independent risk management functions reporting outside the trading hierarchy, mandatory segregation of front and back office, position limits monitored independently, and requirements that traders take consecutive leave so someone else must handle their book.
That last measure sounds trivial and is genuinely effective. Concealment usually requires continuous management, and forcing an absence means someone else opens the drawer.
The Bottom Line
Barings failed because one person controlled both trading and its verification, and because unusual profits from a supposedly riskless strategy were treated as success rather than as a question. The controls that followed exist because of exactly this.