AIB Currency Losses: John Rusnak's Role and the Fraud

John Rusnak’s Role

In July 1993, Allfirst hired Mr. Rusnak. The hiring process was led by the head of treasury funds management for Allfirst, Mr. Ray. Mr. Rusnak had extensive experience in currency trading.

Mr. Rusnak promoted himself as a trader who used options to engage in a form of arbitrage, attempting to take advantage of price discrepancies between currency options and currency forwards. Allfirst had until then engaged in “directional” spot and forward trading, simple bets that particular currencies would rise or fall. Messrs. Cronin and Ray were intrigued by Mr. Rusnak’s style of trading, as he claimed it would diversify the revenue stream arising from simple directional trading.

Mr. Rusnak ws earlier reporting to a trading manager, when in 1999, when the trading manager left, he started directly reporting to mr. Ray. Mr. Ray's knowledge of foreign exchange was limited. However, Allfirst's treasurer, despite his own extensive currency-trading experience, nevertheless relied heavily upon the treasury funds manager to supervise Mr. Rusnak. Mr. Ray however, did not devote significant attention to Mr. Rusnak’s proprietary trading.

The treasury funds manager (Ray) was highly protective of Mr. Rusnak; he often strongly defended Mr. Rusnak in inquiries by the back office and risk assessment personnel.

Mr. Rusnak’s annual bonus was directly related to his net trading profits. In effect, Mr. Rusnak received a bonus equal to 30 percent of any net trading profits he generated in excess of five times his salary.

Mr. Rusnak was regarded by some fellow employees as strong and confident. In the market, Mr. Rusnak was perceived as an active trader and a profitable client for the brokers. Many brokerage firms wanted to cover Mr. Rusnak. The brokers and traders heavily entertained Mr. Rusnak, with meals, hotel stays, golf trips, Super Bowl tickets and other travel. He apparently liked to be wined and dined, and the brokers obliged.

Mr. Rusnak’s Fraud

Trading Strategy: Mr. Rusnak told everyone that he engaged in an arbitrage between foreign exchange options and the spot and forward markets to make consistent profits.

In fact, however, much of Mr. Rusnak’s trading was linear, directional trading. These were simple bets that the market would move in a particular direction. The majority of his real positions were simple currency forwards. He also bought some foreign exchange options with “high deltas” (options that were “deep in the money” and had large premiums). He traded in “exotic” options, although the trading in these products was infrequent.

The Bogus Options: Mr. Rusnak sustained substantial losses at some point in or about 1997, and it was around that time that his fraudulent activities may have begun. Using currency forwards, Mr. Rusnak apparently bet wrongly on the movement of the Japanese yen— he bought a great deal of yen for future delivery, only to see the value of the yen, and thus his forward positions, decline. To hide his losses and the size of his positions, he created fictitious options. These fictitious options also tended to give the appearance that his real positions were hedged.

Through a clever technique, Mr. Rusnak was able to get the bogus options onto Allfirst’s books. Typically, he would simultaneously enter two bogus trades into Allfirst’s trading system.  The two options would involve the same currency, and the same strike price, and they would offset each other from a cash standpoint: the first would involve the receipt of a large premium; and the second would involve the payment of an identical premium; accordingly, there would be no net cash in or out. There was one significant difference in the terms of the offsetting options: the option involving the receipt of a premium would expire on the same day it was purportedly written, but the other option would expire weeks later.

Mr. Rusnak’s bogus options were designed to exploit weaknesses in the control environment around him. Allfirst prepared no reports listing the expiring one-day options. And so no one at Allfirst paid any attention to them. In part, this was due to the fact that the system being used by Allfirst for options did not automatically alert supervisors if such options were not exercised.

At the same time, Mr. Rusnak took advantage of an even bigger hole in the control environment: a failure in the back-office consistently to obtain transaction confirmations. Initially, Mr. Rusnak created bogus broker confirmations to validate his deals, but, with occasional exceptions, he stopped doing that in September 1998. Mr. Rusnak instead apparently managed to persuade an individual in the back office not to seek to confirm the purported pairs of options. There was no need for confirmations, he apparently argued, because there was no net transfer of cash. Perhaps this practice suited the convenience of the back-office staffer; the bogus options were purportedly with the Tokyo or Singapore branches of major international financial institutions, and to have made confirming telephone calls would have required the employee to work in the middle of the night.

The upshot is that Mr. Rusnak’s scheme empowered him to create, at will, assets on Allfirst’s books— false assets— without ever having to pay for them. At the end of a day when he entered the pair of bogus options, the liability represented by the one day bogus option would not appear on Allfirst’s books. What was left— and what did appear on the books— was the purported unexpired deep-in-the-money option for which Allfirst had supposedly paid a large premium. The Allfirst balance sheet would reflect that the bank was holding a valuable asset— one that concealed the losses in Mr. Rusnak’s directional spot and forward trades. Mr. Rusnak would effectively keep the seemingly valuable but nonexistent asset on his book by repeatedly rolling it over into new bogus options as the original ones purportedly came due.

The Prime Brokerage Accounts: In his real trading, Mr. Rusnak  continued to lose money in spot and forward transactions, and as he did so, he wrote more and more of the bogus options to cover up his losses. (There were a few months in late 1999 when he apparently made some money back and reduced his bogus options positions, but that was short-lived.) From 1999 on, the majority of the real, money-losing trading activity was conducted pursuant to “net settlement” agreements that Mr. Rusnak established with various financial institutions, including Bank of America and later Citibank (when his principal trading contact at another institution moved there). The net settlement arrangements with Bank of America and Citibank subsequently evolved into “prime brokerage accounts.”

Under the prime brokerage agreements, spot foreign exchange transactions between Allfirst and its counterparties were settled with the broker and “rolled” into a forward transaction. At the end of each day, all spot foreign exchange trades were swapped into a forward foreign exchange trade between the prime broker and Allfirst.

These forward trades were cash settled in dollars at a fixed date each month. No settlement or cash collateral moved on these accounts on any other date. Prime brokers make money by receiving an agreed upon fee for settlement of foreign exchange spot transactions ($8 to $10 per million settled) and also typically charge full bid-offer pricing on the forward transaction rolls.

These accounts enabled Mr. Rusnak to increase significantly the size and scope of his real trading. Prime brokerage accounts are commonly used by hedge funds and other active traders, and— except for the monthly settlements and lack of collateral requirements (most prime brokerage arrangements call for daily mark-to-market collateral)— the terms of Allfirst’s arrangements with its prime brokers were not particularly unusual. Such accounts are, however, unusual for banks and are not used by AIB’s foreign exchange traders.

Nonetheless, Mr. Rusnak managed to convince his supervisors, including the Allfirst treasurer, that the accounts made sense for Allfirst because they would eliminate the need for extensive back office operations.

The investigation revealed bogus deals in Allfirst’s records of prime brokerage activity. Such deals were input by Mr. Rusnak in the DEVON system (the system used to record prime brokerage account trades) and later reversed prior to the monthly settlement.

Mr. Rusnak’s usage of Allfirst’s balance sheet**:**  Through his use of the prime brokerage accounts, Mr. Rusnak’s trading activity grew. So did his losses and the bogus option positions. And so grew his use of Allfirst’s balance sheet. At some point in 2000, the Allfirst treasurer directed that trading income should reflect a charge for the cost of balance-sheet usage.

In 2001, Mr. Rusnak’s balance sheet usage drew the attention of the finance department, auditors and others (including Mr. Ray, who noted that Mr. Rusnak’s earnings were inadequate to justify his use of the balance sheet).

The deep-in-the-money options: Mr. Rusnak’s solution, beginning in February 2001, was to sell real yearlong, deep-in-the-money options. These options enabled Mr. Rusnak to fund his losses and keep trading. He sold five such options for a total of $300 million. These options thus raised a large amount of cash that was used to help fund the monthly settlement of Mr. Rusnak’s foreign exchange forward transactions. The options also allowed Mr. Rusnak to augment his core directional position.

These real, deep-in-the-money options were essentially synthetic loans made to Allfirst by the counterparties, which included Citibank and Bank of America.

More bogus options: The real options Mr. Rusnak sold were liabilities of Allfirst. And they were recorded as liabilities on Allfirst’s books— initially. But to disguise his losses and his new method of funding them, Mr. Rusnak needed to get them off the books. To do that, he turned again to bogus options— deals that were purportedly transacted with the counterparties to the original deep-in-the-money options, and that gave the impression that the original options had been repurchased. The result, of course, was that Allfirst was saddled with massive, unrecorded liabilities.

Manipulation of the Value at Risk calculation: Mr. Rusnak manipulated the principal measure used by Allfirst and AIB to monitor his trading: Value at Risk (VaR). One way in which Mr. Rusnak manipulated the VaR figures was through the bogus options he created: those options, as noted, appeared to hedge his real positions, and so they reduced the VaR.

But Mr. Rusnak had another technique to manipulate VaR: false figures for so-called “holdover” transactions. Such transactions created the illusion of reducing Mr. Rusnak’s open currency position.

Mr. Rusnak achieved this scheme by directly manipulating the inputs into the calculation of the VaR that were used by an employee in Allfirst’s risk-control group. Thus, while that employee was supposed to independently check the VaR, she relied on a spreadsheet that obtained information from Mr. Rusnak’s personal computer and that included figures for so-called “holdover” transactions—transactions entered into after a certain hour toward the end of each day. But these transactions were not real and, indeed, unlike the bogus options, were not even entered on to the bank’s trading software.

Mr. Rusnak also engaged in a practice of entering false foreign exchange forward transactions in DEVON and reversing them before the next settlement date.

This article is based on the Report to the Boards of Allied Irish Banks, p.l.c., Allfirst Financial Inc., and Allfirst Bank Concerning Currency Trading Losses, which is a part of the FRM syllabus.

Data Science in Finance: 9-Book Bundle

Data Science in Finance Book Bundle

Master R and Python for financial data science with our comprehensive bundle of 9 ebooks.

What's Included:

  • Getting Started with R
  • R Programming for Data Science
  • Data Visualization with R
  • Financial Time Series Analysis with R
  • Quantitative Trading Strategies with R
  • Derivatives with R
  • Credit Risk Modelling With R
  • Python for Data Science
  • Machine Learning in Finance using Python

Each book includes PDFs, explanations, instructions, data files, and R code for all examples.

Get the Bundle for $29 (Regular $57)
JOIN 30,000 DATA PROFESSIONALS

Free Guides - Getting Started with R and Python

Enter your name and email address below and we will email you the guides for R programming and Python.

Data Science in Finance: 9-Book Bundle

Data Science in Finance Book Bundle

Master R and Python for financial data science with our comprehensive bundle of 9 ebooks.

What's Included:

  • Getting Started with R
  • R Programming for Data Science
  • Data Visualization with R
  • Financial Time Series Analysis with R
  • Quantitative Trading Strategies with R
  • Derivatives with R
  • Credit Risk Modelling With R
  • Python for Data Science
  • Machine Learning in Finance using Python

Each book comes with PDFs, detailed explanations, step-by-step instructions, data files, and complete downloadable R code for all examples.