PwC tight-lipped over AIB audit

Auditor PricewaterhouseCoopers (PwC) will not comment on any aspect of its audit of AIB Group accounts, citing client confidentiality…

Auditor PricewaterhouseCoopers (PwC) will not comment on any aspect of its audit of AIB Group accounts, citing client confidentiality. It is not known if the auditor sent "management letters" to the head of treasury or management at Allfirst or to group management at any time over the course of the 1997 to 2001 audits.

Fraud over this period at the Allfirst foreign exchange division cost the group some $691 million (€781.9 million).

Management letters are used by auditors to communicate concerns about controls or other issues discovered during the course of an audit to the audit client. The Allfirst operation was audited by the local PwC office in the US, which reported into the main group audit in Dublin.

AIB chairman Mr Lochlann Quinn appeared to absolve PwC of any accountability over the Allfirst foreign exchange fraud. At the press conference to release the Ludwig report, Mr Quinn stated that the fundamental responsibility rested with Allfirst management and systems and indicated AIB was not considering any action against its auditors. But should the auditors have been able to alert the group to control weaknesses or abuses at Allfirst, to the potential for fraud or to the actual fraud in its early years?

READ MORE

The fraud at Allfirst started in 1997, according to AIB, with losses intensifying over the period to discovery in January/February 2002. AIB Group 1997 accounts were audited by Coopers & Lybrand. From 1998 the accounts have been audited by PwC - the merged Pricewaterhouse and Coopers & Lybrand firms.

The Ludwig investigation of the fraud pointed out a number of weaknesses at the Allfirst treasury division over the period. It showed that, while there were controls in place, they were not operated and it set out a number of "missed opportunities to detect Mr Rusnak's fraud".

These ranged from failure to adequately supervise Mr Rusnak's trading activities to inadequate responses when issues were raised about his trading, including problems on his prime brokerage accounts and back-office difficulties confirming his trades. They included instances of when the group was alerted to the huge size of Mr Rusnak's positions and trading or should have been alerted (Allfirst's 1999 and 2000 10-K filings in the US and associated file notes and the preparation of the year 2000 financial reports).

This information was available at Allfirst and within the group. In his report, Mr Ludwig stated "because of time pressures and the determination not to interfere with the ongoing audit work of PricewaterhouseCoopers, we did not interview representatives of PricewaterhouseCoopers".

Whether PwC became aware or should have become aware of problems at Allfirst would largely have depended on the extent to which the external auditor relied on the work of the internal audit department in carrying out its audit. Company directors are responsible for preparing annual accounts while the auditor's role is to express an opinion on whether the accounts give a true and fair view and are properly prepared.

Auditors often explain that they are "watchdogs" rather than "bloodhounds". By this they mean that their specific responsibility is to express an opinion on the accounts and that the audit process is not designed to uncover fraud but could uncover fraud as part of the process.

In the audit of a large group with foreign subsidiaries such as AIB, the auditors are likely to base their opinion on an assessment of the strength of the controls in place and their reliability rather than on a detailed testing of transactions. The external auditor often relies heavily on the work of the internal auditor in deciding whether these controls can be relied on - but this would only follow an assessment of the work of the internal audit. The external auditor is likely to replicate some of the work of the internal audit department to test its reliability.

Discovering a fraud would largely depend on the location and extent of the fraud and how well the test sample is drawn. The greater the extent of the fraud, the more it is that a fraudulent transaction or failed control would turn up in the auditor's sample.