February 2006
Robi Dattatreya, managing director of Total Solutions Europe and Guido Kalmijn, consultant for ABN AMRO Asset Management provide an analysis

Presently most banks are looking at Basle II requirements and are in various stages of meeting the associated regulations. Basle II also covers the operational risk side of the business. This gives operational risk the attention it deserves. If Barings had adhered to elementary operational risk principles; they might still be an independent bank.

In this article we examine to what extent the Barings organisation adhered to the elementary controls in the case of the fraud committed by Nick Leeson, which brought Barings to the brink of collapse in 1995. After giving an account of the organisation and situation involved, we shall then examine whether the fraud could have been discovered if six elementary controls had been implemented. We shall then draw conclusions on what went wrong. 

Introduction
Nick Leeson joined Barings in London in 1989. At the time he worked in the settlement department. In April 1992 Barings London transferred him to Singapore to establish the settlement process in the bank's local office. In 1992 it was decided to have Leeson take part in Barings Futures Singapore's (BFS) trading activities. In July of that year he began as floor manager at Singapore International Money Exchange (SIMEX). The idea was that Leeson would carry out orders from other Barings offices and clients only. BFS ultimately had four clients, three of which were other Barings offices. Leeson was supposed to limit himself to these activities and to report on his transactions and results. In 1993 Leeson's portfolio was extended from executing orders to trading on his own account. This mainly concerned arbitrating price differences between identical options and futures contracts on SIMEX on the one hand and the Tokyo and Osaka stock markets on the other. Leeson was supposed to trade within intraday trading limits and not to hold any un-hedged positions until the following day.

Structure of the organisation
At BFS Nick Leeson was in charge of both the front and back offices. The price tag of a separate back office manager was considered too high. As a result 'independent' accounting or reports on Leeson's trading activities were non-existent. Between 1993 and 1995 Barings had a matrix structure in which functional and regional lines intersected. Supervision at BFS was unclear and insufficiently regulated. Various parties shared the task of supervising BFS: Barings Securities London was assigned the task of supervising trading activities, Barings Securities Japan was expected to supervise the risks attached to intraday trading and Barings Singapore was given a part in supervising the operations department. Since Barings' managers relied on each other to supervise Leeson, he was effectively free to do as he pleased. Barings London actually assumed that Leeson's trading activities constituted little risk and that they in fact brought in significant earnings.

In 1984 Barings internal accounting department came to the conclusion that its organisation was relatively 'uncontrolled'. Barings' internal accounting department noted the lack of a proper control structure at BFS, commenting at the same time that it involved trading for other Barings offices, meaning they were in fact controlled based on the reconciliation of advances and transactions. In response to the internal accounts department's findings, Barings took the view that it was saving money by not appointing a separate head of settlements.

The fraud itself
In July 1982 Leeson found himself having to deal with a mistake made by one of his junior administrators who had confirmed a purchase as a sale to an opposite party.
 
Leeson corrected the original transaction by executing it in the proper manner.

The order originally intended was entered as a normal transaction. At the same time he had to reverse the sale to arrive at the position envisaged by the original order.

However, Leeson waited too long before making the correction and the market had in the meantime moved against him. He decided to book a fictitious order with retrospective effect in order to reverse the original sale. The actual loss-making transaction executed to reverse the original mistake was entered in account 88888.

This first error resulted in a loss of $35 000, while the annual salary of the junior administrator in question was just $7 000. Leeson and his department were understaffed and overworked. Leeson wanted to take on more experienced administrators to be able to withstand the workload but Barings denied him permission to do so. This resulted in further errors.

A week after the first irregular correction of a trading error there was a second incident. Leeson and one of his floor traders had a difference of opinion about a particular order, i.e. whether it was a purchase or sale order. Once again Leeson used account 88888 to conceal the error and to keep it out of official reports.

In a subsequent incident, at a time when the Osaka stock exchange was having computer problems and part of its trade was diverted to Singapore, BFS staff was hard pressed to keep up with the trade volume. During this time BFS was faced with an error that ultimately cost Barings $1.7 million.

At a later stage Leeson executed transactions for other offices as well, adjusting the execution price to that of the confirmation, enabling the other office to book a profit.

The corresponding loss was then entered into account 88888. Over time the losses became so great that they could no longer be covered by fraudulent or non-existent transactions in the general ledger. Leeson decided to adopt the tactic of selling short, his intention being to cover the losses in account 88888 with the premiums on such sales. The options were obviously entered into account 88888 as well. These sold option positions resulted in additional losses for Barings. It was not until 24 February 1995 that the accounting of these fraudulent transactions finally came to light.

Elementary back-office controls
The comprehensive list of controls comprises of 25 key control items for operations. In this article we limit ourselves to the following elementary controls:

Elementary controls
Segregation of duties between front office and back office
Transaction confirmations
Reconciliation of cash accounts
Reconciliation of settlement accounts
Co-coordinating position with the front office
Co-coordinating P&L with the front office

Below we examine to what extent these controls were actually carried out or where they failed to have effect.

Segregation of duties
It is both abundantly clear and practically incomprehensible that responsibility for the front office and back office merged at a point just above the executing level. Where tasks are separated in the proper manner, responsibility for both offices should meet at the highest possible levels. If this is not possible, for instance because of the sheer size of the organisation, the question should be whether adequate support is available for a front office. BFS back-office staff reported directly to the trader.

Transaction confirmations
The documents reveal little about the irregularities in the transaction confirmations. However, since Leeson was also in charge of the back office he was able to manipulate confirmations issued officially by BFS. One particular case has been discovered in which a control by means of a transaction confirmation which, had it been implemented properly, could have revealed the ongoing fraud.

In January 1995, during the annual audit, the accountants tried to explain a discrepancy amounting to $70 million between the ledge and the statement issued by SIMEX. Leeson gave a number of explanations and within several days six different versions of the cause of the discrepancy were circulating at Barings' head office in London. Common to several of these explanations was the allegation of an unauthorized transaction. After a few days Nick Leeson finally produced a transaction confirmation that was supposed to account for the discrepancy. He had falsified it and faxed it from his home. The top line of the fax clearly showed the names Nick and Lisa. No consequences were attached to the fact that he had abused his trading powers and he was paid his normal bonus.

Reconciliation
The losses accumulated by Leeson were futures contracts traded on SIMEX. When trading in futures one is supposed to maintain a margin account to hold a balance that serves as collateral for the obligations arising from any particular futures position. If the value of one's futures falls, the account must be topped up: this is known as the margin call. For this margin call it makes no difference whether or not one reports to one's own organisation; if the value declines, more money has to be put into the account to provide sufficient collateral.

To all appearances, Leeson's trading activities were quite plain: on the instructions of four clients three of which were Barings offices - he traded in futures on SIMEX.

Leeson needed enormous sums of money to finance his trading activities and margin obligations. For these he used money from Barings London and Barings Tokyo. In addition, at the request of BFS, Barings Securities London paid advances for margin obligations on the stock exchange. A remarkable fact is that even though Leeson did not report his position correctly to London, London received reports of the margin obligations that arose from the position in the 88888 account. This information was only used once the fraud had been discovered. As early as 1993 London had an unaccountable reconciliation discrepancy of between $26-35 million. This discrepancy widened to $176 million in 1994 and $565 million in the first two months of 1995. This problem had already been discussed in London in 1993. His unrelenting need for money during the years in question and the simultaneous inability to reconcile it was no reason for London to stop making payments to BFS.

Leeson also booked a number of advances he required for his position as loans to clients. The department responsible for lines of credit should have set up lines of credit and organised the proper documentation in order to formalize such loans. The fact was, however, that the department concerned took no further action.

Positions
At the end of 1994 and beginning of 1995 various signals emerged from the market indicating that all was not well at BFS. By the closing weeks of 1994 Leeson had taken up positions resulting in obligations on the part of Barings that constituted over 75% of its group capital. Even though this was communicated to the Bank of England in the 'large exposure report', neither it nor Barings senior management took adequate measures.

Signals from the market invested in intensity at the beginning of February. Not only had SIMEX requested a guarantee for BFS's margin obligations, even the Bank of International Settlements in Basle made enquires with Barings around that time. Barings' response, however, came too late. On 11 and 25 January 1995 BFS also received letters from SIMEX which cast doubt on the information BFS had provided, referred to a possible breach of stock exchange regulations and requested additional security for BFS's payment of margin obligations. Barings' management in Singapore asked Leeson to draw up a draft reply but neglected to investigate the problems listed by SIMEX.

In the end Barings London discovered on 17 February 1995 that there was a discrepancy of $130 million between the margin deposits listed in the books and SIMEX's own statement. In addition to this, Leeson still needed money to meet his margin obligations. Ultimately, between 17 and 23 February, the day Leeson fled Singapore, another $280 million had been transferred.

The fraud
The consequence of Leeson's fraudulent conduct was that between 1992 and 1994 BFS and Barings were able to publish good results, all thanks to the 88888 account. Losses in this account rose from $5.5 million in September 1992 to $230 million in 1994, and finally $1.4 billion when Barings finally crashed. According to the inspectors of Barings Futures, the losses in the 88888 account up to the end of January 1995 were such that Barings could still have survived the fraud.

Summary of what went wrong
The major conclusions to be drawn from the case in question are summarized below. Each point listed could in itself have prevented the fraud from occurring if translated into action.

The lack of separation of tasks between trading activities and administration gave Leeson a free hand to execute unauthorized transactions, dress up figures for existing transactions, report handsome profits and at the same time expose Barings to enormous market risks.

Advances were granted for the margin calls but not reconciled in arrears with actual transactions and positions. BFS and Barings' other offices all failed in this respect.

Leeson received little supervision (or correction) from above in performing his activities. Barings' confusing matrix structure led to various senior managers concluding that fellow managers were supervising Leeson.

Senior management had too little understanding of Leeson's activities to be able to appreciate that profits reported from arbitration activities were unrealistic. It is not realistic to believe that trading activities bearing a low risk are able to yield high profits over a sustained period of time.

Barings in London kept financing Leeson's activities up to 75% of the group's capital without taking measures to reduce positions or cover risks.

Barings did not respond adequately to signals from the market that something was amiss. In particular, those from SIMEX and BIS in January 1995 were serious enough to prompt immediate action. Despite this, Barings kept financing Leeson's activities. Barings' collapse could perhaps have been averted if Leeson had been stopped in January 1995. A major part of the losses occurred in February 1995.

The Bank of England concludes as follows:
"We have concluded that the system of checks and balances, necessary for the correct management and control of a financial institution, failed in the case of Barings vis--vis BFS in a most serious manner on a number of levels and in more than one location. It is clear to us that lessons can be learned from the collapse".

Total Solutions Europe
Total Solutions Europe, established in 1998, is a consultancy firm specializing in the financial service industry. For this segment Total Solutions Europe provides consulting, recruitment and training services.

© Banker Middle East 2006