Barings, 1995

One trader, one hidden account, and a 233-year-old bank sold for £1. Not a pricing failure — a controls failure that any organisation chart would have caught.

What happened

  • 1992 — Barings sends a young trader to Singapore to run futures operations, arbitraging price differences in Nikkei 225 futures between the Osaka and Singapore exchanges. Low-risk work by design.
  • 1992 onwards — an error account, numbered 88888, is used to hide losses rather than report them. Crucially, the same person runs both the trading desk and the back office that settles and reconciles it.
  • 1993–94 — the hidden position grows. Reported profits are excellent; the trader is treated as a star and his reported results account for a large share of the bank's stated earnings.
  • 17 January 1995 — the Kobe earthquake. Japanese equities fall hard.
  • January–February 1995 — instead of cutting, the position is doubled: a large long Nikkei futures position plus short options, betting on a recovery that does not come.
  • 23 February 1995 — the trader leaves a note and flees. Losses total roughly £827m, more than the bank's entire capital.
  • 26 February 1995 — Barings, founded 1762, is declared insolvent. ING buys it for £1 and assumes the liabilities.

The mechanism

The two lines are the whole case. The real position grew for three years while the reported one barely moved — and the gap only became visible when the cash to fund it could no longer be raised.
DiscoveredActual positionReported to managementTimePosition size and reported P&L
  • The trades were not exotic. Long Nikkei futures and short straddles: the first is a directional bet, the second is short volatility. Both are covered on this site's futures and strategy builder pages. Nothing here required a Nobel prize to understand.
  • The short straddle is the accelerant. It pays a premium if the index sits still and loses without limit if it moves. After Kobe it moved a long way, in the wrong direction, and the losses compounded against the futures position rather than offsetting it.
  • Doubling down is not irrationality — it is the incentive. Once losses exceed what can be confessed, the only path that avoids certain career destruction is a bet large enough to recover them. Every rogue-trading case follows this shape.
  • The cash was the tell. Sustaining the position required enormous variation margin, funded by requests to London that were met without anyone asking why a low-risk arbitrage business consumed hundreds of millions in cash. Margin does not lie; someone has to read it.

What it teaches

  • Segregation of duties is not bureaucracy. One person controlling both the trade and its confirmation can make any position invisible. This single control failure is the case in one line, and it is now the first thing any operational-risk review looks for.
  • Profits that nobody can explain are a risk finding, not a success. A genuinely low-risk arbitrage generating outsized returns is arithmetically suspicious — the edge is small by definition. Unexplained profit and unexplained loss are the same signal.
  • Follow the cash, not the P&L. Reported results are an assertion; margin payments are a fact confirmed by a third party. Any reconciliation of the two would have ended this in 1993.
  • Concentration risk applies to people. A single individual generating a large share of group profit is a control question long before it is a compensation question.
  • Barings is not a historical curiosity. The same pattern — a hidden position, a compliant back office, escalating size — recurred at Sumitomo, Allied Irish, Société Générale and UBS. The instruments changed; the control gap did not. Archegos is the modern version, with the concealment done by the market structure rather than by one trader.