The Rogue Trader

At approximately eleven-thirty at night on the twenty-third of February 1995, in a rented apartment in Singapore, a twenty-eight-year-old British trader named Nick Leeson typed two words onto a sheet of paper, fed it into a fax machine, and sent it to his operations supervisor in Hong Kong.

The words were: I'm sorry.

By Monday morning, the bank that Leeson's fax had just ended had been in business for two hundred and thirty-three years. Its name was Barings. It had underwritten the Louisiana Purchase. It had been called, in the diplomatic correspondence of the early nineteenth century, the sixth great power of Europe. On Monday the twenty-seventh of February, the Bank of England placed it in administration. On the sixth of March, the Dutch banking group ING acquired it for one pound.

Account 88888

Nick Leeson had arrived at Baring Futures Singapore in early 1992, aged twenty-four, with no experience running a trading desk. He had come up through the back office — Coutts, then Morgan Stanley, then Barings in London — and his employers had decided that back-office expertise qualified him to set up the new Singapore subsidiary. He was made responsible for both the trading operations and the settlement of those trades.

This combination was prohibited by Barings' own internal risk manuals. The prohibition was written down. It was not enforced in Singapore, because maintaining separate front-office and back-office staff cost approximately seventy thousand pounds a year more than having one person do both jobs.

Eight hundred and twenty-seven million pounds were eventually lost to save seventy thousand a year.

In July of 1992, Leeson opened an account on the SIMEX trading system numbered 88888. He told Barings London it was a clerical-error account — the sort used to temporarily absorb small adjustments from settlement clerks. London agreed to let the daily statements remain in Singapore. They were said to be of no interest.

In late October 1992, Leeson recorded his first real trading loss — approximately twenty thousand pounds — into account 88888 rather than reporting it to his supervisors. This is the decision that mattered. Everything that followed was an attempt to reverse, or to compensate for, or eventually to outrun that first entry.

The arbitrage that wasn't

Leeson was supposed to be doing arbitrage — buying Nikkei 225 futures on whichever exchange was cheaper and selling them simultaneously on whichever was more expensive, profiting from the small spread. Arbitrage is low-risk and low-return by design. Net exposure should be approximately zero, because every buy is matched by a sell.

What Leeson was actually doing was directional speculation — taking large unhedged bets on which way the Japanese stock market would move, and describing those bets to London as arbitrage. The losses went into 88888. The fictitious profits were reported as arbitrage performance. By the end of 1994, Leeson's reported 1994 profit was twenty-eight and a half million pounds. Actual accumulated hidden loss: one hundred and eighty-five million. His bonus for the year, paid in January 1995, was four hundred and fifty thousand pounds.

Throughout 1994 and the first two months of 1995, Leeson requested, and received, approximately eight hundred and thirty-five million US dollars in margin funding from the Barings group treasury in London. The treasurer wired each request without asking why an arbitrage book — which should be cash-neutral — was consuming this much working capital. The Board of Banking Supervision inquiry later noted, with characteristic restraint, that the treasury operation was "reactive, undirected, and lacking any framework by which unusual patterns might be identified."

The forged letter

In December 1994, the external auditors — Coopers and Lybrand Singapore — noticed a fifty-million-pound discrepancy. They asked Leeson to explain it. He produced two letters, purportedly from SLK Securities in New York and from Citibank, confirming a legitimate transfer.

Both letters were fabricated. Leeson had typed them on a home computer, applied correction fluid, photocopied them, and run the photocopies through his own fax machine to simulate received-from-elsewhere artefacts. The auditors accepted them. They did not telephone SLK Securities. They did not telephone Citibank. They signed the 1994 accounts.

Leeson later pleaded guilty to two counts of fraud and forgery in respect of these letters. They were the specific charges for which he served four years of a six-and-a-half year sentence in Singapore's Changi Prison.

Kobe

At five forty-six on the morning of the seventeenth of January 1995, an earthquake measuring 7.3 struck the Japanese city of Kobe. More than six thousand people were killed. The Nikkei 225 dropped seven per cent over the following week.

Leeson had been holding large positions that relied on the Nikkei remaining stable. After Kobe, rather than closing out the losing positions, he doubled down — buying enormous quantities of Nikkei futures on the assumption that the market would rebound. By the twenty-third of February, his positions represented roughly twenty per cent of SIMEX's total open interest in Nikkei futures and fifty per cent in Japanese government bonds. A single person, in a single office, held more Nikkei exposure than every other trader on SIMEX combined.

The Nikkei did not rebound. Leeson sent his fax on the Friday night and flew to Kuala Lumpur the following morning.

The weekend at the Bank of England

On the Sunday the twenty-sixth of February, Eddie George, the Governor of the Bank of England, convened an emergency meeting to attempt a rescue before the Asian markets opened on Monday. Modelled on the 1890 arrangement that had saved Barings from the Argentine debt crisis, George approached a syndicate of merchant banks. He approached the Sultan of Brunei as a potential single buyer. None of the conversations closed.

The problem, in the end, was that nobody could say with confidence, by Sunday midnight, what the total liability was. Leeson's positions were still open. Any buyer would be acquiring a bank whose final loss could not be stated within an order of magnitude. At ten o'clock on the Sunday evening, George informed Peter Baring — the chairman, seventh generation of the Baring family — that a rescue could not be arranged. Barings would be placed in administration before the Asian markets opened.

What the system did not change

The UK Board of Banking Supervision inquiry, published in July 1995, named twenty-three current and former Barings executives who had failed in their supervisory responsibilities. Thirteen received formal criticism. Several were banned by the Bank of England from senior positions for up to three years. None was prosecuted criminally — the Crown Prosecution Service accepted that senior management had not intended to defraud anyone, but had merely failed to prevent someone else from doing so. Within a decade, most of those named were employed elsewhere in London finance.

What the Barings collapse changed was the rule about who can see both sides of a trade. Segregation of duties is now, in every major jurisdiction, enforceable by regulation. External auditors are required to verify bank confirmations independently rather than accepting documents provided by the audited entity.

What it did not change, because it could not be legislated, was the culture that allowed Peter Baring, in late 1994, to believe what Singapore was telling him because the profits appeared to be there. The profits were not there. They had never been there. Every major bank fraud since — Société Générale and Jérôme Kerviel in 2008, UBS and Kweku Adoboli in 2011, Wirecard and Jan Marsalek in 2020 — has involved a version of this same failure. A star performer whose returns nobody in the reporting chain quite understood. A reporting chain that trusted its inputs. A senior executive telling some external party, shortly before the collapse, that performance had been amazing.

In the next one, the rule that gets broken will be a different rule. The fraud will arrive on schedule. The profits will appear to be there. And we will be surprised again.

---

Listen to the full episode on Cooked Books, episode 1: The Rogue Trader. Subscribe wherever you get podcasts, or through the links on crimesfromeurope.com.