An auditor’s report, and in particular a long-form auditor’s report, can, when well written, provide unique and valuable insights into the audit process and the areas of risk in a business that the audit focuses on. Despite this, their potential usefulness is tainted by criticism that the reports are overly complex or boilerplate, hampered by technical jargon, or too vague.

Words matter

The auditing profession is built on numbers, but auditors must also communicate effectively, including in their auditor’s report. Auditors know that these reports can enlighten readers, if properly used, and help inform long-term investment decisions. However, in writing their reports, auditors are subject to conflicting pressures. An auditor’s report needs to be technically accurate as well as understandable, brief as well as informative, and clear and robust regarding auditor concerns, but not emotive or misleading. And they have evolved to meet the demands of an increasingly litigious and risk-averse environment. These pressures can drive the language used almost as much as auditing standards.

With this in mind, ICAEW’s Audit and Assurance Faculty commissioned research into the linguistic features of an auditor’s report, with a view to understanding how they help or hinder the stakeholders who read them. Dr Veronika Koller, Professor of Discourse Studies at Lancaster University, read and analysed eight long-form auditor’s reports and provided findings regarding the linguistic features of the reports, and recommendations regarding what auditors might consider doing differently. Dr Koller had no significant prior knowledge of auditing. This study focused primarily on the key audit matters in long-form auditors’ reports but also covered a material uncertainty disclosure in one report. 

Dr Koller’s study identified several linguistic features that were common across the sample of auditors’ reports. Dr Koller did not find anything that will be news to auditors and, importantly, many findings are unavoidable because of the legal and regulatory requirements. Nevertheless, we set out below an overview of each of these linguistic features. We also offer for the consideration of both auditors and regulators, some simple examples of how auditors might use more helpful language, without creating a significant change in substance. Taking a step back to think about the overall impression given by the language used in an auditor’s report and making some simple but effective changes in overall tone can help them become more informative and useable. 

Beating about the bush? 

The first term Dr Koller’s study refers to is ‘pervasive hedging’. Hedging in a linguistic sense means using qualifying words or statements that increase subjectivity. Common examples are words related to likelihood (‘might have a material effect’; ‘might or might not emerge’) or that impose conditions (‘if it was concluded that a breach had occurred…’), often combined with other qualifiers (‘... there are a number of potential penalties that could be imposed’). 

Modifiers are a type of hedging that qualify an object or process by making it less than absolute (‘Our analysis indicated that in substance…the entry fees represent…’), or relate to cognition (‘We believe…; ‘We consider…’; ‘In our view…’). 

The use of multiple qualifiers and modifiers in close proximity can give the impression of a high level of subjectivity or uncertainty (‘alternative performance measures can provide investors with appropriate additional information…however when improperly used and presented, these kinds of measures might mislead investors’). 

Many qualifiers accurately represent the uncertainties being described, as uncertainty is an inherent feature of doing business. No-one would thank auditors for being definitive where genuine uncertainty exists, as it often does. However, Dr Koller’s study suggests that the pervasive use of qualifying language can result in unnecessarily complex sentences. Auditors might therefore consider whether all of the qualifiers are truly necessary. For example, consider whether the following would, in context, have a significantly different meaning. 

What we didn’t do

The second term is ‘unmodalised’ statements. These are the opposite of hedging – they are unqualified statements. Many such statements in an auditor’s report provide information about the audit, often in a negative form (‘…This statement is not a guarantee as to the company’s ability to continue as a going concern’). Dr Koller’s study notes that these statements may appear to be an attempt by auditors to distance themselves from their own conclusions. 

Again, many of the examples in an auditor’s report, both positive and negative, are statements derived from, or required by, auditing standards and simply describe the audit process. However, these statements can sometimes be convoluted, thus reducing their clarity and informativeness. It is for regulators to simplify some of this required wording.

Trust us

The third term is ‘legitimation’. ‘Legitimation’ is described as an attempt by auditors to counteract a lack of informativeness in their report by using ‘legitimising’ language in the audit (‘In our professional judgement…’), (‘Our global audit team has deep industry experience…’) (‘We performed an extensive reassessment of…’).

The use of ‘legitimation’, often involving hyperbole, can inadvertently undermine the value of work performed in the eyes of a sceptical reader – exactly the opposite of what is intended. When applied selectively it may also give the impression that less effort was exercised in other areas, which may not be true. 

What could regulators and users of auditors' reports do?

The FRC currently provides details about auditors’ responsibilities on its website, and auditors’ reports include a link to this information. The FRC’s consultation on auditor reporting proposes to add more details to this statement, including on reporting by exception, as part of their plans to declutter the audit report. This includes removing the requirement to describe the extent to which the audit was considered capable of detecting irregularities, including fraud, limiting this requirement to PIEs and other listed entities. However, regulators could consider further ways to reduce mandatory language or consider placing it in another location.

Investors and other stakeholders could increase their engagement with audit committees, auditors, professional bodies and regulators, on how auditors' reports could be made more informative.