Do UK audit committees really engage with auditors on audit planning and performance?

Abstract

In the wake of the financial crisis, regulators intend to increase the responsibilities of audit committees (ACs), yet little is known about how ACs discharge their existing responsibilities and interact with auditors and management. This study investigates the involvement of the AC, the AC chair (ACC), the audit partner (AP) and the chief financial officer (CFO) in relation to a range of audit-related matters in UK-listed companies in the 2007 regulatory environment, which remains fundamentally unchanged. The level of AC and ACC engagement in seven AC responsibilities set by the Combined Code is high (over 80%). However, only 50% of 16 audit planning, performance and finalisation matters are routinely discussed. The ACC acts without the full AC in 11% of discussions, while 25% involve only the CFO and AP without either the ACC or the AC. The extent of discussion and/or ACC involvement is influenced by background characteristics (company size, auditor size and ACC experience and qualifications). This evidence of less than full AC engagement with audit-related issues suggests that regulators may risk creating an AC expectations gap if AC duties under the extant model are significantly increased without structural change.

Keywords:

• audit committee
• audit interaction
• discussion
• engagement
• involvement
• negotiation

Acknowledgements

This paper is adapted from the P.D. Leake Lecture delivered by the authors at the Institute of Chartered Accountants in England and Wales on 21 June 2011. We would like to thank The Institute of Chartered Accountants in England and Wales’ charitable trusts for financial support on this project. We thank Graham Roberts, Robert Hodgkinson, Richard Macve, Stuart McLeay and other audience members at the 2011 P.D. Leake Lecture for helpful conversations and suggestions. Thanks also go to: John Coombe, Ken Lever, Ian Percy and Gerald Russell who have acted as general advisors to the project (including pilot testing the questionnaire); Steve Maslin who also helped pilot test the questionnaire; and the 100 Group of Finance Directors, the ICAEW's Audit and Assurance Faculty and many audit firms who gave their support. Particular thanks go to the 498 individuals who completed the questionnaire on which this paper is based.

Notes

International Standard on Auditing (ISA) (UK&I) 260.

Revisions have subsequently been made to the Code (FRC 2006, 2008, 2010) which has now been renamed the UK Corporate Governance Code.

The UK Listing Authority requires listed companies to provide a ‘comply or explain’ statement in their annual report which explains how the Combined Code has been applied by the company (FRC 2006). The code provisions extant in 2007 were in five parts, dealing with: directors; remuneration; accountability and audit; relations with shareholders; and institutional shareholders (FRC 2006).

The title was shortened to Communication with Those Charged with Governance.

Strictly, the auditors monitor (i) the financial control systems put in place by management and which produce the annual report and (ii) the reliability of external reporting by management.

AIM companies are excluded because they are not yet required to adopt IFRS; investment trusts are excluded because their accounting, auditing and governance is very different.

For a variety of reasons, several companies in the initial sample were reselected (e.g. company had delisted, merged or demerged, or moved domicile since their last annual report; company reported under US GAAP). A replacement was sought from the same industry group and with the closest market capitalisation).

The initial company sample resulted in 58 companies (33 involving the top 250) where the audit committee chair had been selected more than once (in three cases, four times). The 27 cases involving non-top 250 companies were reselected, but this often produced new duplications.

A few weeks prior to mailing the questionnaire, the integrity of the mailing list details was checked using RNS data (which requires announcement of changes in board and directorate membership). It was found that several individuals had changed job and the database was updated accordingly.

This approach has two important advantages compared with most prior studies. First, the responding auditors are all members of the target group (and not audit partners with no current experience of listed company audit or audit managers without partner-level experience of interactions). Second, the partners are drawn not only from the Big 4 audit firms but also the next tier of firms with listed company clients. The latter feature permits the investigation of the impact of auditor type.

ACC respondents were all audit committee chairs, with the exception of two who were deputy chairs; included AP respondents were all listed company audit engagement partners (four responses were eliminated as they did not fall within the criteria set for the following reasons: client reported under US GAAP only, client not yet on IFRS (AIM company), AP audited investment trusts only; and client was a public sector organisation).

The breakdown into all 16 response groups for three of the responsibilities (oversight of auditor independence, non-audit services provider and non-audit services fees) is reported in Beattie et al. (2009).

In September 2011, according to the corporate governance database Boardex (http://www.boardex.com/), 78 of the FTSE all-share index companies reported in their annual report that they had a risk committee of some form (38 combined risk with audit while 40 either had a stand-alone risk committee or combined risk with another function).

Four per cent of respondents reported zero AC members with recent and relevant financial experience, contrary to the Combined Code (FRC 2006), 47% had one such member and the remaining 49% had more than one such member.

Fifty-nine per cent of respondents reported no AC members who were former auditors; 34% reported one such member and 7% reported two such members.

Fifteen per cent of respondents reported no AC members holding a recognised accounting qualification, 56% reported one such member and 29% reported two or more such members.