Abstract
Purpose
Taking a communication perspective, the paper explores management's rhetoric in profit warnings, whose sole purpose is to disclose unexpected bad news.
Design/methodology/approach
Adopting a close-reading approach to text analysis, the authors analyse three profit warnings of the now-collapsed Carillion, contrasting the rhetoric with contemporaneous investor conference calls to discuss the profit warnings and board minutes recording boardroom discussions of the case company's precarious financial circumstances. The analysis applies an Aristotelian framework, focussing on logos (appealing to logic and reason), ethos (appealing to authority) and pathos (appealing to emotion) to examine how Carillion's board and management used language to persuade shareholders concerning the company's adverse circumstances.
Findings
As non-routine communications, the language in profit warnings displays and mimics characteristics of routine communications by appealing primarily to logos (logic and reason). The rhetorical profiles of investor conference calls and board meeting minutes differ from profit warnings, suggesting a different version of the story behind the scenes. The authors frame the three profit warnings as representing three stages of communication as follows: denial, defiance and desperation and, for our case company, ultimately, culminating in defeat.
Research limitations/implications
The research is limited to the study of profit warnings in one case company.
Originality/value
The paper views profit warnings as a communication artefact and examines the rhetoric in these corporate documents to elucidate their key features. The paper provides novel insights into the role of profit warnings as a corporate communication vehicle/genre delivering bad news.
Keywords
Citation
Edgar, V.C., Brennan, N.M. and Power, S.B. (2022), "The language of profit warnings: a case of denial, defiance, desperation and defeat", Accounting, Auditing & Accountability Journal, Vol. 35 No. 9, pp. 28-56. https://doi.org/10.1108/AAAJ-04-2020-4519
Publisher
:Emerald Publishing Limited
Copyright © 2021, Victoria C. Edgar, Niamh M. Brennan and Sean Bradley Power
License
Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial and non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence may be seen at http://creativecommons.org/licences/by/4.0/legalcode
1. Introduction
Profit warnings (announcements of large negative earnings surprises) occur when companies issue trading updates/earnings press releases forecasting earnings materially below market expectations. Alves et al. (2009, p. 449) characterise profit warnings as “voluntary trading updates that signal a material deterioration in profitability and earnings relative to market expectations.” Under the EU and UK law, [1] listed companies must disclose price-sensitive information immediately. The Financial Conduct Authority (2021, Section 2.1.3) requires price-sensitive disclosures to be “prompt and fair”. Therefore, we consider profit warning documents mandatory, whilst the detailed disclosures are voluntary, absent regulations on these disclosures.
Roberts (2014) characterises investors trading on profit warnings as the equivalent of catching a falling knife. The Investors' Chronicle's Bearbull (2002) provides insights into managers' psychology in disclosing profit warnings: Profit warnings come in threes because managers react to bad news with overconfidence, guilt and shame at failure, such that they are unable to accept the failure in one blow. We use Bearbull's (2002) insights to characterise his three stages as denial (“overconfidence […] that […] things can be pulled round”), defiance (“closer to the truth but still holds something back”) and desperation (“guilt and shame that comes with failure”). In our case, there is a fourth stage, i.e. defeat (the company's collapse). Lok (2010, p. 1324) quotes an equity fund manager interviewee on her reaction to a profit warning. She says as follows:
The reason you had a profit warning is that they had no clue themselves, and that is when you absolutely hit the panic button … First profit-warning regardless of whether the share prices go up, get rid [of the shares], because chances are statistically they will have a second fall.
This interviewee quote highlights two aspects of profit warnings. First, profit warnings suggest management is not in control of the business's financial reporting and is surprised at the unexpected bad news. Second, the quote confirms Bearbull's expectation that profit warnings tend to come in multiples. There is a superstition that “profit warnings come in threes” (coincidently, the situation for our case company, Carillion), but Vincent (2017) suggests fours and fives as well.
Most prior studies research profit warnings by reference to share price reactions using event study methodology. Prior research largely views corporate reports as vehicles to provide useful information to shareholders. Francis et al. (2002, p. 516) characterise earnings press releases as a disclosure mechanism (referring to the document) revealing a “package of information” (referring to the document's content) to investors. Corporate reports can provide useful information to shareholders and persuade corporate report users (Stanton and Stanton, 2002). Davis et al. (2012, p. 845) argue that managers use language throughout an earnings press release “to signal, both directly and more subtly, their expectations about future performance”. Whereas prior accounting research examines the disclosure of bad news and its timing, few studies research the language managers use when companies disclose bad news. We take a different stance to Davis et al. (2012). We believe managers use language in earnings press releases, including profit warnings, not just to signal expectations but to persuade organisational audiences of the profit warnings' disclosure credibility. Examining language use in profit warnings provides a more nuanced analysis of management communication and insights into persuasive trust-building (Joyce, 2020) between managers and their audiences. Such a language may also unconsciously reveal managers' thinking on evolving events.
Rather than question the impact of bad news on investors, our research aim is as follows: What language does Carillion management use to deliver the bad news in profit warning documents? We elaborate our research aim into two research questions. In the context of delivering bad news (profit warnings) before its collapse, we contrast Carillion's rhetoric in the profit warnings with the rhetoric in its contemporaneous conference calls with analysts and board minutes.
How does Carillion management communicate across the stages of the three profit warnings? What is the nature of the use of rhetoric (logos, ethos and pathos) adopted in Carillion's three profit warnings?
Do Carillion's profit warnings tell the whole story? How does the use of rhetoric (logos, ethos and pathos) adopted in Carillion's three profit warnings compare with Carillion's contemporaneous conference calls with analysts and board minutes?
Whilst prior research examines good/bad news in corporate reports, profit warnings constitute a wholly bad-news context whose disclosure is solely motivated by material unexpected bad news. The paper contributes as follows: First, we view profit warnings as rhetorical communication artefacts and examine the rhetoric in these corporate documents. Second, we consider managers strategically use rhetoric to sculpt a business-as-usual reality different from the behind-the-scenes reality. Third, we conjecture that managers' motivation to construct an alternative reality stems from their initial denial of the situation followed by defiance and desperation. We contribute to the literature by building on Sandell and Svensson's (2016) study of the “language of failure”. In their case, failure means not meeting external performance expectations. Like our study, Sandell and Svensson (2017) examine corporate failure events through a rhetorical lens. They argue that management actively constructs those events and uses corporate reports as an institutionalised communicative channel for managing those events through discursive and rhetorical construction.
Regulators do not prescribe the content of profit warnings, conference calls and board meeting minutes. Profit warnings tell investors the story of unexpected bad news. For our case company, this story is first revealed in board minutes, then communicated in writing in the profit warnings and is finally communicated orally in the conference calls. The profit warnings, conference calls and board minutes should tell a consistent story (the reality) since the speaker (Carillion/Carillion's executive directors) in each case is the same. However, leaders can use words to “sculpt reality” (Vignone, 2012, p. 35) and an “audience's interpretation of and reaction to a person, event, or discourse can be shaped by the frame in which that information is viewed” (Benoit, 2001, p. 72, emphasis added).
We examine three “frames” (the profit warnings, board minutes and conference calls), as Davis and Tama-Sweet (2012) conjecture that managers will be strategic in their language choices depending on disclosure outlet. Amernic and Craig (2013, p. 381, emphasis added) argue, “CEO public discourse in whatever form … should be monitored with a view to encourage the construction of a more functional and ethical tone at the top”. Our multi-dimensional perspective provides knowledge about how companies communicate bad news in profit warnings and the strategies used in alternative texts to persuade the audiences and gain readers' acceptance of the writer's preferred definition of reality (Hyland, 1998, p. 232). Our purpose is to compare the language in profit warnings within the broader communication of the company's financial performance. Huse (2007, p. 35, emphasis added) writes, “Accountability is clearly related to understanding, evaluating and balancing various perspectives and interests”. We know little about how companies communicate bad news in profit warnings. Our paper is relevant to a small but growing literature on language use in delivering bad news in corporate reports. Examining board minutes opens the black box of boardroom rhetoric. Examining conference calls contrasts the “carefully crafted” profit warnings to the less-scripted conference calls, particularly the Q&A session with analysts (Danielewicz-Betz, 2016). We believe few papers conduct such comparative analysis to assess the consistency of the story told using language across documentary sources.
Beattie and Davison (2015, p. 655) comment that “narrative theory and analysis remains a relatively neglected area of accounting research”. Previous studies have shown that accounting communications are imbued with rhetoric, including annual reports (White and Hanson, 2000), social and environmental reports (Higgins and Walker, 2012), profit forecasts (Brennan and Gray, 2000), press releases (Brennan and Merkl-Davies, 2014) and takeover defence documents (Brennan et al., 2010). In addition, managers use rhetoric for different purposes, including to legitimise institutional change (Suddaby and Greenwood, 2005); to persuade regarding sustainability development (Higgins and Walker, 2012); to persuade organisations to review their values or improve social and environmental performance (Brennan and Merkl-Davies, 2014); to influence investment judgements (Courtis, 2004) and to persuade concerning an organisation's resilience and long-term survival (Craig and Amernic, 2004). Finally, rhetoric is present in written narratives, pictures (Davison, 2008) and colour (Courtis, 2004). Laine (2009) finds that rhetoric changes depending on the social and institutional context.
Drawing on rhetorical appeals to logos, ethos and pathos (Aristotle, 350BC/2010), we examine the role of profit warnings as a corporate communication vehicle delivering bad news. Rhetoric in this paper refers to the use of language to persuade or influence others. Rhetoric is the “art of ruling the minds of men” (quote attributed to Plato). Our rhetorical analysis draws on Aristotle's art of persuasion, where logos, ethos and pathos are three rhetorical tools used to speak and write effectively in appealing to an audience. Logos concerns appeals to reason using logic and reason; ethos is convincing through authority, credibility and character and pathos is emotional appeal.
We base our analysis on three profit warnings issued by UK facilities management and construction multinational, Carillion plc, before its collapse in January 2018. We select Carillion as our case because the UK parliamentary enquiry (House of Commons, 2018a, p. 86) found Carillion management's behaviour to be self-serving, “The individuals who failed in their responsibilities, in running Carillion and in challenging, advising or regulating it, were often acting entirely in line with their personal incentives”. In addition, the availability of contemporaneous analysts' conference calls and private board minutes (which became publicly available during the UK parliamentary enquiry which followed Carillion's collapse) enhanced the potential insights from this case.
The main rhetorical strategy we observe is logos, which dominates in all three profit warnings but drops in Carillion's third profit warning. A key challenge for management disclosing a profit forecast is credibility, which language use in narrative disclosures influences. Under pressure from a credibility perspective, honesty within the ethos rhetorical strategy doubles between the first two profit warnings and the final profit warning. Beason (1991, p. 326) believes ethos “can be the most potent means of persuasion”. We suggest the increase in ethos in the third profit warning is explained by the failure of previous logos appeals to persuade shareholders of the adverse situation, consequently the urgent need to enhance credibility. By comparing the rhetoric used in the conference calls and board minutes, we suggest that profit warnings do not tell the whole story. Instead, we show profit warnings are vehicles of constructed discourse, a medium through which company leaders can persuade regarding unexpected bad news.
Section 2 and Section 3 discuss prior empirical research and our rhetorical theory. We summarise our methodology in Section 4. Section 5 presents our findings, and the paper concludes in Section 6.
2. Prior empirical research
This section discusses prior research on language use in narrative reporting and delivering bad news.
2.1 Language use in narrative reporting
Corporate narrative documents can be routine (for example, annual reports, CSR reports and earnings press releases) or non-routine (for example, profit warning press releases and takeover defence documents). Merkl-Davies and Brennan (2017, p. 434) observe that “despite its growing importance […] we know little about the nature of accounting communication with external audiences, particularly the ways in which technical accounting is mediated through language”. The importance of examining “language (particularly rhetoric)” is that it “enables us to understand how organisations use corporate narrative documents to communicate with external parties in a clear and transparent manner, to shape messages to suit their own agenda, or, worse still, to mislead audiences” (Merkl-Davies and Brennan, 2017, p. 434).
Leaders have an important role in corporate narratives. Commenting on the work of Amernic and Craig (2006), Beelitz and Merkl-Davies (2012, p. 102) write, “The words of CEOs have ‘clout’ […] in that they not only shape the perceptions of stakeholders and society, but also create ideology, thus impacting on the way the world is run […] This is particularly important in times of crisis, when CEOs are expected to assume rhetorical leadership.” Liff and Wahlström (2018) suggest failed crisis communication by top management as a plausible explanation for the collapse of Northern Rock. Beelitz and Merkl-Davies (2012) conclude that Chief Executive Officers (CEOs) use discourse strategically. Amernic and Craig (2013, p. 381) argue that CEO public discourse should be monitored “based on the presumption that the public utterances of corporate leaders potentially can reveal important insights”.
Davis et al. (2012) conjecture that managers' language in routine earnings press releases may provide an imprecise forecast of earnings or other performance metrics. They suggest (p. 849) that “language possesses several characteristics that, in combination with its inherent subtlety, are likely to influence investors' ability to evaluate the credibility” of earnings press releases. They add (p. 850) that “language provides more subtle signals regarding managers' future-earnings expectations”. Bochkay et al. (2020) provide evidence that investors are influenced not just by what managers say but by how they say it – using the case of extreme language in conference calls. Guo et al. (2020) observe ample evidence that managers often make strategic choices in the language they use in communication. In studying managers' language style, the prior literature focusses on linguistic features such as optimistic/pessimistic tone (e.g. Davis et al., 2012), readability, pronoun use (e.g. Chen and Loftus, 2019), etc. Language matters. In reporting management forecasts, a key managerial challenge is their credibility. Several studies examine the importance of language in perceptions of forecast credibility (e.g. Guan et al., 2020; Hutton et al., 2003). For example, in an experimental study, Chen and Chang (2017) find unsophisticated investors perceive managers more trustworthy if they use vivid language and plausible explanations in earnings warnings. If language matters in routine press releases, we expect managers to place the same, if not more, importance on language in non-routine profit warning press releases. Yet, we know little about language use in profit warning press releases.
Like our study, Davis and Tama-Sweet (2012) compare language use (positive/negative tone) across disclosure outlets – earnings press releases vs Management Discussion and Analysis documents (MD&As). They find more optimistic and less pessimistic language in earnings press releases compared with MD&As. Bochkay et al. (2020) contrast the formal and boilerplate language in regulatory filings with conference calls involving spoken, rather than written, language. They observe that conference calls tend to be less formal and scripted than typically seen in regulatory filings.
2.2 Delivering bad news
Language use varies depending on whether the news is positive or negative. Corporate narrative documents generally contain good news and may also include bad news. In contrast, profit warning press releases are solely motivated by the need for companies to deliver bad news. Previous literature suggests that routine bad news in annual reports is subject to the Pollyanna effect (excessive optimism) and managers use complex language to obfuscate bad news (Rutherford, 2005, 2013). In such routine communications, the opportunity for the Pollyanna effect exists because the presence of good news would not necessarily seem disproportionate or out of place. However, investors expect bad news in profit warnings, so excessive optimism would seem out of place. This bad-news context limits the opportunity for the Pollyanna effect. Therefore, understanding language use when delivering bad news in profit warnings is an important area of study. In experimental research, Chen and Loftus (2019) study the effect of language on perceptions of managers' credibility, finding higher credibility perceptions for self-inclusive singular pronouns vs collective plural pronouns when performance news in earnings conference calls is negative. Guo et al. (2020) examine plain, straightforward language vs complex and vague language in negative earnings surprises for its effect in signalling weakness on the competitive activity of rival firms.
Profit warnings are viewed somewhat differently in the UK (see Alves et al., 2009 definition in the opening paragraph to the paper) and the USA. In their seminal study, Kasznik and Lev (1995, p. 113) consider profit warnings as communications to investors in advance of fourth-quarter negative earnings surprises. Profit warnings negatively impact a company's reputation as well as its value. Given their impact, how managers write profit warnings is consequential. Prior research has examined the writing characteristics of profit warnings, finding explanations in the forecast (Baginski et al., 2004), soft-talk disclosures and verifiable forward-looking statements (Hutton et al., 2003) and disaggregation of the forecast into its components (Elliott et al., 2011) to influence investors. In an experimental study, Chen and Chang (2017) coin the term “earnings warnings language” in terms of the vividness of the language and the plausibility of the explanations provided.
In summary, the puzzles motivating our study are as follows: First, that language use in corporate narrative documents is important, yet we know little about it (Merkl-Davies and Brennan, 2017). Second, we know little about language use in non-routine corporate narrative documents. Third, most studies examine communication one-dimensionally within a particular context. We examine profit warnings by contrasting them with contemporaneous conference calls and board minutes. Finally, previous research examines delivering bad news in narrative documents where investors expect good news. Profit warnings are bad news, and studying good news in such documents seems out of place. So we need a better understanding of the language within profit warnings.
3. Theory
This section discusses our rhetorical approach
Beattie and Davison (2015, p. 655) state that writers use various devices in storytelling, including “structure, plot, viewpoint, character and rhetorical techniques”. We examine rhetorical language, which is “sometimes seen as synonymous with discourse […] is, however, distinguished by a focus on persuasion” (Higgins and Walker, 2012, p. 197). Rhetoric is particularly pronounced during crises or change when organisations require the support of key stakeholders, the media and the general public (Merkl-Davies and Brennan, 2017). Scholars have studied the role of rhetoric in corporate failure using voluntary disclosures in annual reports. Craig and Amernic (2004) examine how studying leaders' rhetorical discourse can help understand Enron's failure. Bujaki and McConomy (2012) highlight the use of metaphor as rhetoric in the voluntary disclosures of the failed company, Nortel. The role of more immediate forms of accounting communication, such as press releases and profit forecasts, has been examined in the context of corporate controversy (Brennan and Merkl-Davies, 2014) and corporate takeovers (Brennan and Gray, 2000) but not corporate failure. Few, if any, studies view profit warnings as a communication genre. Hursti (2011) links the strength of argument in the profit forecasts with their accuracy for a large sample of European forecasts, finding that strong arguments are associated with more accurate forecasts. In discussing Northern Rock's profit warning, Liff and Wahlström (2018, p. 248) highlight management's efforts to persuade shareholders that Northern Rock's business model was solid and its future prosperous. Since rhetoric has the “capacity to influence thought and action”, can be “used for good or base ends” and “has moral implications” (Herrick, 1992, p. 133), Bujaki and McConomy (2012) call for a better understanding of the use of rhetorical devices other than metaphor in corporate disclosures.
For Aristotle (350BC/2010), knowledge is contingent and open to rhetorical construction and interpretation. Classical rhetoric focusses on how we use words. It emphasises the intentional and deliberate use of persuasive language to influence meaning and shape action. Persuasion of ourselves and others plays a critical role in how we familiarise and evaluate our world. Classical rhetoric suggests that there are three primary modes or strategic devices for persuasion (Aristotle, 350BC/2010): logos (persuasion through reasoning), ethos (persuasion through personality or authority) and pathos (persuasion through the arousal of emotion) (Beattie, 2014). “Organisations use rhetoric, retrospectively, to respond to existing rhetorical situations or proactively to shape or frame future rhetorical situations” (Cheney et al., 2004, p. 87, quoted in Brennan and Merkl-Davies, 2014, p. 607). So when crafting a profit warning, a company can appeal to either logos, ethos or pathos, or a combination of these rhetorical strategies, to explain the bad news to shareholders and the media in anticipation of critique (Sandell and Svensson, 2017). The choice of rhetorical strategy is important to try to minimise the anticipated negative reactions to a profit warning.
Logos uses “facts and figures to back up a claim” (Brennan and Merkl-Davies, 2014, p. 608) and justifies a course of action “for reasons of efficiency or reasonableness” (Green, 2001, p. 44). Ethos uses “writers or speakers who are trusted and respected” (Beason, 1991, p. 326) whose arguments audiences are more likely to accept. Ethos is persuasive because of the character of the communicator as a credible source of ideas and views. Ethos is arguably the most potent form of persuasion “often being more effective than either logical or emotional appeals” (Beason, 1991, p. 326). Pathos influences audiences “by evoking an emotional response”. Pathos involves “the use of figurative language, particularly metaphor” (Charteris-Black, 2004 quoted in Brennan and Merkl-Davies, 2014, p. 608). Pathos persuades “the emotional part of the mind” (Green, 2001, p. 44). Several authors have adopted Aristotelian classical rhetoric to study accounting communication. In their study of social and environmental reports of three New Zealand companies, Higgins and Walker (2012) draw on Aristotle to analyse the discursive struggles concerning the appropriate role of business in society. Brennan and Merkl-Davies (2014) conduct a rhetorical analysis through an Aristotelian lens of an exchange of press releases in an argument between Greenpeace and six organisations in the fashion trade. Aerts and Yan (2017) study meta-discourse markers in CEO letters to shareholders, enabling writers to connect with their audience by expressing their stance and engagement with readers. They argue that meta-discourse markers are highly instrumental in materialising rational (“logos”), credible (“ethos”) and affective (“pathos”) appeals core to the classical Aristotelian concept of rhetorical communication. These appeals represent three distinct but often intersecting and inseparable dimensions of constructing persuasive messages. Hossain et al. (2019) explore 24 Fortune 500 companies' rhetoric in sustainability reporting to persuade stakeholders and to legitimise their strategies. They operationalise persuasion as logos, ethos and pathos.
Rhetoric focusses primarily on “the persuasive aspects of messages” (Merkl-Davies and Brennan 2017, p. 445). Rhetorical analysis is appropriate for our study of non-routine profit warnings delivering bad news to shareholders since profit warnings trigger negative stock price reactions (e.g. Kasznik and Lev, 1995). Likely, the consequences of disclosing a profit warning include triggering sell-outs, downgrading company stock and speculative attacks on shares (Liff and Wahlström, 2018, p. 248). A rhetorical perspective allows scholars to answer various questions, including which rhetorical strategies organisations use most frequently, which are most effective and how organisations persuade their shareholders and stakeholders to support them in times of crisis and change (Merkl-Davies and Brennan, 2017, p. 445). By adopting a rhetorical perspective, we can examine whether rhetoric before a company's collapse differs from the rhetoric used in alternative contexts. In addition, the comparison in rhetoric across documentary sources allows an understanding of how rhetorical strategies differ when telling the same story (we discuss earlier the necessity, importance and interest of this). However, it does not tell us whether investors view the profit warnings as useful or credible or the ethical implications, including whether readers can trust profit warnings. Using a rhetorical perspective, the consequences of the communication between organisations and their audiences, including share price reactions, cannot be answered. Alternative theoretical perspectives would help answer these questions and are areas for future research.
4. Research methodology
Our paper identifies Carillion's rhetorical appeals in its profit warnings and contrasts the rhetorical appeals used in conference calls with analysts and board of directors' minutes. Our research employs Aristotle's strategies of rhetoric, logos, ethos and pathos. In the 20th century, the new-rhetoric movement further developed this classic rhetoric. We identify the relevance of this new-rhetoric movement for our paper. We highlight the importance of context in two ways: First, we consider Carillion's context, facing unexpected bad news and before its collapse. Second, we highlight context by examining rhetoric across three documentary sources with different contexts, including whether these documentary sources tell the same story and use recurrent or different patterns when persuading audiences. This context consists of Merkl-Davies and Brennan's (2017) three interrelated elements. The speaker in each text is Carillion/Carillion's executive directors. The audience for the profit warnings is primarily Carillion investors. But, as press releases, profit warnings can have multiple audiences due to their public nature (Brennan and Merkl-Davies, 2014), particularly if they “are brought to the attention of the media they can have a strong public relations effect” (Brennan et al., 2010 p. 265). The conference calls' primary audience is financial analysts (Danielewicz-Betz, 2016). The board minutes' audience is Carillion board members. In this section, we present our case, summarise the data analysed and describe our analytical approach.
4.1 The case
We base our analysis on three profit warnings issued by UK facilities management and construction multinational, Carillion plc, in the seven months before it collapsed in January 2018. We summarise the key events in our case in Figure 1.
Figure 2 shows the share price at the time of the profit warnings, which we interpret as a proxy of how successful the profit warnings were in retaining investor trust.
4.2 The data
We analyse Carillion's three profit warnings, transcripts of two investor conference calls following Profit Warning 1 and Profit Warning 2 and board minutes that became public following the UK House of Commons (2018a) parliamentary enquiry into Carillion. We download the profit warnings from the London Stock Exchange's Regulatory News Service, which are in the form of press releases. We examine the data commencing at the headline in the profit warnings and ending before the housekeeping details at the end of each press release. We obtain conference call transcripts from Bloomberg. We only analyse the transcript text where Carillion management is speaking, excluding any text relating to the operator or analysts and others attending the call (i.e. excluding the analyst questions but including Carillion management responses). We download the board minutes from the UK House of Commons (2018b) parliamentary enquiry website. We exclude the date of the meeting and list of attendees at the start of the board minutes. Board Minutes 5 includes some information on the Grenfell Tower fire disaster on 14 June 2017 (326 words), which we do not include in our data. Board Minutes 5 is the only document that includes text not directly related to the profit warnings. We exclude the names of speakers in conference call transcripts and board minutes. Table 1 summarises the data.
4.3 Data analysis
We adopt Amernic and Craig's (2013) close-reading methodology. The manual content analysis provides an in-depth understanding of the language and context of language use. Documentary content analysis, including corporate communications, “seeks to quantify content in terms of predetermined categories” (Bryman, 2012, p. 289). This type of quantitative content analysis is form-orientated and non-contextualised. It relies on a “systematic and objective application of neutral rules” (Bryman, 2012, p. 289). Our research examines profit warnings in the context of Carillion's collapse. Therefore, we adopt a meaning-orientated approach to content analysis that considers context. This approach “views the meaning of accounting communication to be dependent on contextual factors, such as the social and institutional setting, and to emerge in a process of close reading and interpretation in the form of qualitative text analysis” (Merkl-Davies and Brennan, 2017, p. 448).
Amernic and Craig (2006, p. 6) describe close reading as constituting “an intensive, almost forensic, scrutiny of the explicit and implicit composition of the text”. In line with Amernic and Craig (2013, 2017), we first review written material on Carillion's collapse, including newspaper articles and the UK House of Commons (2018b) parliamentary enquiry's oral and written evidence. The enquiry's May 2018 final report (House of Commons, 2018a) facilitates our understanding of the context of Carillion's collapse. We begin our analysis by close reading the narratives at least three times, intensively scrutinising each genre (profit warnings, analyst conference calls and board meeting minutes) and comparing them for internal consistency. Amernic and Craig (2017, p. 68) acknowledge the method's subjectivity when they say, “an interpretative process can produce many useful insights, it can also lead to contestable conclusions because of the complexity of a social phenomenon and a ‘plurality of plausible explanations’”. Therefore, like Amernic and Craig (2017, p. 68), “we do not claim that the explanations we offer are necessarily unique or better than some others”, but our “resulting knowledge is an interpretation”.
We adopt a manual content analysis approach. Content analysis can also be computerised. We chose not to adopt a computerised approach because of subjectivity in analysing the “keywords” relating to logos, ethos and pathos [2]. For example, often readers recognise pathos based on context and surrounding words, e.g. we code “quick fix” as pathos but neither “quick” nor “fix” as individual keywords are “pathos”. Our coding of keywords is also specific to a business context. For example, “balance” and “sheet” are used in the context of “balance sheet”, which is the rhetoric of business (and therefore, logos). In a non-business context, it is unlikely that a computer would code “balance” or “sheet” as logos.
In developing a protocol for close-reading methodology, Craig et al. (2001, p. 7) identify close-reading as a useful means of “fathoming” rhetoric and that this “perspective helps make persuasive language … explicit”. Close-reading methodology emphasises the importance of context, differing perspectives and the techniques of argumentation. Researchers can employ this methodology to examine various aspects of communication, including “hypertextuality, rhetoricity and perception-fashioning” (Craig et al., 2001, p. 7). When researchers use close reading to analyse rhetoric, Bryman (2012) refers to this as “rhetorical analysis”. Our research focusses on all three strategies of rhetoric (logos, ethos and pathos) similar, for example, to Higgins and Walker (2012) and Brennan and Merkl-Davies (2014). Other research focusses only on one rhetorical strategy or a sub-set, for example, metaphor, which is an aspect of pathos, (Bujaki and McConomy, 2012), or virtue, which is an aspect of ethos (Shanahan and Seele, 2015).
We follow a two-stage process. First, we analyse the profit warnings for the language used, quantification, specificity, forecast horizon, causal explanations and management undertakings. Second, we analyse the profit warnings, conference calls and board minutes for rhetoric. To reduce the subjectivity of close-reading analysis, we develop an analytical framework (Table A1) to identify the rhetoric deployed. We use ten rhetorical sub-categories for each of the three rhetorical strategies, citing sources from the prior literature to justify our ten sub-categories. Example 1 summarises and illustrates our ten rhetorical sub-categories. Our close reading entails examining texts to discern meaning and help make the texts' persuasive language explicit (Craig et al., 2001). We read each sentence/phrase (our unit of analysis) and paragraph and mark them as persuasion by logos, ethos or pathos. We repeat this process several times. From our close reading, some sentences persuade using more than one rhetorical strategy; other sentences do not contain any rhetorical strategies.
5. Findings
This section describes the findings, first concerning the characteristics of and rhetoric in the three profit warnings and second, the rhetorical analysis of the conference calls and board minutes. This section concludes by discussing the language of profit warnings as a case of denial, defiance, desperation and defeat. We apply these labels to Carillion's profit warnings (and collapse) to characterise the stages of management's use of rhetoric as Carillion's collapse unfolds.
5.1 Description of the profit warnings
Profit warnings are expressed in coy language. For example, a clear phrase such as “profit warning” is rarely, if ever, used. Table 2 summarises the wording and other characteristics of the three profit warnings. The focus of the profit warnings shifts from “management's previous expectations” (i.e. previous management forecasts) in Profit Warning 1 to “current market expectations” (i.e. current analyst forecasts) in Profit Warning 2 and Profit Warning 3. Profit Warning 3 discloses that “current market expectations” are likely to be “materially lower” than “management's previous expectations”. Carillion replaces the word “lower” in Profit Warning 2 with the phrase “materially lower” (a strong phrase) in Profit Warning 3, emphasising the extreme deterioration in performance. Only sophisticated readers familiar with this corporate report genre would be able to “read between the lines” and interpret what these phrases and words mean. The specificity of Carillion's three profit warnings/forecasts decreases over time. Profit Warning 1 is a quantified range forecast of revenue, together with an unquantified profit forecast (“overall performance”). Profit Warning 2 and Profit Warning 3 are unquantified/qualitative profit forecasts. Further, the forecast horizon period (the period from the date of the profit warning to the year-end date of 31 December 2017) becomes shorter and shorter. Shorter forecast horizons should lead to greater certainty around actual results and greater forecast accuracy. Increasingly, imprecise language in Profit Warning 2 and Profit Warning 3 replaces the confidence of the quantified range forecast in Profit Warning 1. Specificity decreases despite the forecast horizon reducing. We would expect the opposite. As the forecast horizon decreases, Carillion should be able to more accurately specify the impact on profits as it comes closer to its year-end.
Causal explanations and attributions (Aerts, 2005) (i.e. explanations for what caused the profit warning and attribution to the factors causing the performance to deteriorate) almost always accompany profit warnings. Causal attributions attribute the cause of company performance to either internal factors or external factors. Managers engage in attributional behaviour to explain the data in a self-serving manner. As shown in Table 3, the causal explanations are neutral, neither attributing cause for the deteriorating performance internally nor externally, although most explanations hint at external causes. Notably, the number of causal explanations increases from one in Profit Warning 1, two in Profit Warning 2 and four causal explanations in Profit Warning 3. One causal explanation is repeated in each of the three profit warnings – phasing of public private partnerships (PPP) equity disposals. The lack of attribution in Profit Warning 1 points towards denial, whilst the increase in attributions in Profit Warning 2 and Profit Warning 3 points towards defiance (blaming others) and desperation (trying to justify).
Statements of board and management remedial actions almost always accompany profit warnings to address the problems implied by the poor performance. Table 4 summarises Carillion's proposed management actions. The number of actions decreases from 14 in Profit Warning 1, to 8 in Profit Warning 2, to only 4 in Profit Warning 3. The most common action is cost-cutting followed by management change. The language specificity decreases over the three profit warnings/forecasts, with vague phrases appearing in Profit Warning 3, such as “in discussions with”, “expects to” and “kept under continuous review”. Whilst the number of causal explanations doubles in Profit Warning 3, board/management actions slump from 14 in Profit Warning 1 to 4 in Profit Warning 3.
5.1.1 Description of the rhetoric in the profit warnings (RQ1)
This section analyses the rhetoric in the profit warnings. Table 5 shows that ethos and pathos are minor compared with logos in the profit warnings, especially Profit Warning 1 and Profit Warning 2. All three profit warnings appeal strongly to logos (71, 71 and 57%, respectively). Profit Warning 1's logos appeals are to both logic (32%), referencing facts and figures, and reason (39%), which are claims or attributions. The remaining appeals are to ethos, primarily honesty (11%), the board being honest about something and inclination to succeed (9%), referencing future achievements, such as new work. Profit Warning 2's logos appeals are primarily to logic (54%) compared with reason (17%). Appeals to ethos/honesty and ethos/inclination to succeed remain similar but appeals to pathos (emotion) increase to 10% compared with Profit Warning 1. Profit Warning 3 reduces and reverses its logos appeals compared with Profit Warning 2. Appeals are primarily to reason (48%) compared with logic (9%). Appeals to ethos/honesty double to 22% and appeals to ethos/inclination to succeed disappear (0%). Appeals to ethos/authority increase (9%) but appeals to pathos decrease (4%).
5.1.2 Description of the rhetoric in conference calls and board meeting minutes (RQ2)
This section analyses the rhetoric in the conference call transcripts and the board minutes. We then compare the rhetoric in our three documentary sources. We compare Profit Warning 1 primarily to Conference Call 1 (held immediately after the release of Profit Warning 1) and Board Minutes 1–6 (held in the months before Profit Warning 1). We compare Profit Warning 2 primarily to Conference Call 2 (held immediately after the release of Profit Warning 2) and Board Minutes 7 (held one month before Profit Warning 2). No conference call or board minutes were released relating to Profit Warning 3. We acknowledge that the production of the three documentary sources is quite different, which limits comparability. However, we believe such a comparison adds insights to our analysis as discussed in Section 1.
Table 6 shows that compared with Profit Warning 1, Conference Call 1's appeals to logos are lower (49%) but appeals to pathos are higher (20%). Appeals to ethos are similar (31%). Conference Call 2's appeals to logos are lower (40%) than Profit Warning 2 and Conference Call 1. Appeals to ethos (30%) and pathos are higher (30%) than Profit Warning 2.
Example 2 illustrates pathos – emotional appeals – in the conference calls. These appeals intensify in Conference Call 2 following Profit Warning 2 with some evidence of repetition.
Carillion management uses phrases such as “clearly”, “frankly” and “obviously”, emphasising alignment with, and thus expressing empathy with, observations in the profit warnings.
Table 7 shows that, compared with Profit Warning 1, Board Minutes 1–6 appeals to logos are on average lower (42%, ranging between 22-60%). Appeals to ethos are on average higher (53%, ranging between 36-78%). Whilst the appeals to ethos vary across the board minutes, on average, the highest appeals are to ethos/authority (16%), ethos/honesty (11%), ethos/deference (9%) and ethos/values (9%). All board minutes contain appeals to ethos/honesty, except for Board Minutes 6, which is an outlier compared with the other board minutes. The directors held this board meeting on 9 July 2017 at 5.30 p.m., the day before Carillion published the first profit warning. Board Minutes 6 is very short (578 words) compared with the other board minutes leading to Profit Warning 1 (average 2,276 words).
Drawing together the findings in Tables 5–7, we compare differences in rhetoric between the profit warnings, conference calls and board minutes. The use of logos dominates in profit warnings, in conference calls and overall. The board minutes appeal to ethos/honesty to a greater extent overall (17% in the board minutes vs 12% in the profit warnings). Most ethos/honesty only appears in Profit Warning 3, whereas the board minutes consistently appeal strongly to ethos/honesty (except Board Minutes 6). The dominant rhetorical sub-categories in conference calls are logos/reason followed by pathos/emotion. Pathos features most in conference calls, possibly because conference calls comprise the spoken word, whereas profit warnings and board minutes are the written word and are easier to control.
5.2 A case of denial, defiance, desperation
This section discusses our findings in Section 5.1. We characterise the profit warnings' rhetoric as moving from denial, to defiance and to desperation and ultimately, ending in defeat.
5.2.1 Denial
We label Profit Warning 1 as a denial of the situation. In Profit Warning 1, Carillion's management provides and discusses the facts and figures, despite the bad news. We see this in the use of logic (facts and figures) and reason (reasons, arguments, justifications and criticisms). We see further evidence in the characteristics of the language used. Carillion's Profit Warning 1 is quantitative, “revenue now expected to be between £4.8bn and £5.0bn and overall performance expected to be below management's previous expectations”. Carillion provides the following causal reason, “phasing of Public Private Partnerships (PPP) equity disposals, which are now expected to be in H2” (Profit Warning 1). Carillion is specific in its language and proposes 14 management remedial actions to deal with lower expected profits.
We believe Board Minutes 1–6 (leading up to Profit Warning 1) and Conference Call 1 (immediately following Profit Warning 1) reinforce our view that management does not appear to appreciate the seriousness of the situation. The board minutes refer to the profit warning as an “announcement” or “trading announcement”. In addition, a profit warning is a serious issue, and we would expect the final board minutes before the profit warning to reflect the seriousness of the situation. The board did not take the decision to issue the profit warning. Rather, Board Minutes 5 reveal Carillion's “brokers” requested the board to make the announcement. Carillion's legal advisors “Slaughter and May confirmed that that was their view [to make an announcement]” and “in the light of the decision of Morgan Stanley and Stifel [to withdraw as underwriters/sponsors of the proposed equity issue]”. The board never discussed the severity of the situation in the board minutes. Instead, Board Minutes 5 record some board querulousness that there was “no obvious reason” for the withdrawal of the sponsors and that “the rationale given by Mr Moorhouse [of Morgan Stanley] […] was not credible given that there had been no obvious change to the market position, the position of the business or its prospects since he and Mr Arch [of Stifel] had represented to the Board that an equity issue was viable and that the two banks would expect to underwrite it”.
Board Minutes 5 record as follows: “In conclusion, the Chairman noted that work continued toward a positive and upbeat announcement for Monday, focusing on the strength of the business as a compelling and attractive proposition […] .” The UK House of Commons (2018a, p. 31) expressed astonishment at the attitude of the board: “The Monday announcement comprised an £845 million write-down. It is difficult to believe the Chairman of the company was not aware of the seriousness of its position, but equally difficult to comprehend his assessment if he was”. Acknowledgement of problems in board minutes suggests the board is aware of the situation but the reference in the minutes to a “positive and upbeat announcement” indicates the board is oblivious of its seriousness. On average, Board Minutes 1–6 are lower in appeals to logos and higher in appeals to ethos than Profit Warning 1. Ethos is persuasion through authority, credibility and character. We see evidence of appeals to several ethos sub-categories. The appeals to ethos/authority may reflect Carillion management's reluctance to disclose a profit warning, believing disclosure is unnecessary and only doing so under duress/pressure from advisors (reflecting Carillion management's denial of the situation).
Although Conference Call 1 has a higher percentage of appeals to pathos, logos still dominates. Logos dominance may further indicate denial – the board does not experience a need to persuade with emotion, as it is in denial of the situation. Appeals to pathos still maintain the situation is not so serious. For example, Conference Call 1 references “a perfect storm” to explain the bad news, not any particular shortcomings and “already from some of the customer calls we've received this morning, we're getting fantastic support from customers saying we're, in fact, happy to go on the record and speak and talk about the high quality of Carillion's work”.
Conclusive evidence of denial is Carillion's payment of a £55 million dividend, paid exactly one month before Profit Warning 1, prompting the House of Commons (2018a, p. 3) to observe, Carillion's “spectacular fall was a story of recklessness, hubris and greed”.
5.2.2 Defiance
We label Profit Warning 2 as defiance of the situation. Defiance can describe resistance, opposition or confrontation. Carillion's share price was 192.30p on 7 July 2017 (the Friday before Profit Warning 1), but by Monday 10 July 2017 (the day of Profit Warning 1), the share price had dropped to 117.10p (39% drop) and by 12 July 2017, it was at 57.2p (70% drop). Carillion's shareholders responded negatively to the first profit warning and did not reflect the board's assessment of the situation. In our examination of Profit Warning 2, Carillion's leaders severely reduce any attempt to reason with its investors. Instead, they resort almost entirely to logic (facts/figures). For example, the following statements present facts but no reasons: “No change to previously announced provision of £845m for construction contracts” and “Further £200m provision for support services contracts, but minimal impact on cash”. Profit Warning 2's middle section is a table of key figures comparing H1 2017 with H1 2016. Further, the profit warning is no longer quantified. There appears to be reduced engagement with investors, reflecting resistance or defiance to its shareholders. Profit Warning 2 is less specific and qualitative.
Carillion's leaders' defiance is also evident in Board Minutes 7 (before Profit Warning 2). There is some realisation within the board that the situation is more serious than it had previously believed. Ethos/honesty is highest in Board Minutes 7. For example, “problems had arisen because they [major projects] were signed-up too quickly in order to get cash in”. However, Carillion's management seems to blame others for the problems, which supports the idea of defiance. For example, Board Minutes 7 refer to “wilful blindness” (on the part of longer-serving staff), but that actions were now “being pursued with vigour” and that “some of the pressure would undoubtedly have come from government, for example in a need to complete before government pulled funding” [3]. Combined with stronger appeals to pathos than previously, perhaps this reflects the board's fear that the brokers and investors were right. We suggest board members attempted to reason why these problems arose but blamed others for the problems, suggesting the board is closer to the truth but still holds something back.
Conference Call 2 reveals further evidence of defiance. Pathos is higher in Conference Call 2 than both Profit Warning 2 and Conference Call 1. Defiance is an emotional response. We suggest this increase in pathos arises now that there is some realisation that the situation is more difficult than initially thought. Carillion feels on the back foot, particularly from analysts' questioning, to which Carillion management was unable to anticipate and craft carefully worded responses. Carillion's limited attempts to engage with investors continue from Profit Warning 2 into Conference Call 2. For example, “Let's just work on getting the £75 million first, then we can have another conversation”, and “let's be very clear, we're not giving specific guidance. … All I'm giving you is the direction of travel. It's going to go up a bit, and then it's going to come down a bit”.
5.2.3 Desperation
We label Profit Warning 3 as desperation of the situation. Profit Warning 3 is almost the opposite of Profit Warning 2 concerning logos/logic and logos/reason. An increase in appeals to reason within the logos rhetorical strategy using claims, attributions and opposing images became necessary in Profit Warning 3 since the facts and figures (logic) told a worse story. The board provides four attributions for the materially lower expectations (up from one in Profit Warning 1). The board also reveals the highest level of honesty in Profit Warning 3. For example, “These self-help measures […] will not be sufficient to enable the Group to achieve its target net debt to EBITDA ratio” […] “Based on its latest forecasts, reflecting the items mentioned above, the Board now expects a covenant breach” and “significant challenges remain”. This acceptance of reality suggests the board is no longer in denial or defiant but rather is desperate. Shortly after Profit Warning 3, Carillion's board announced it was bringing forward the commencement date for the new CEO by three months, from April 2018 to January 2018, perhaps in an act of desperation to save the company. The absence of a conference call following Profit Warning 3 may indicate quiet desperation and surrender to the circumstances.
6. Discussion and conclusions
Individuals use language as a resource to make novel circumstances understood, to articulate a role for themselves, to construct a preferred reality and to meet the voices or perceived expectations of relevant others. Bujaki and McConomy's (2012) research suggests metaphors (pathos) in annual reports change during periods of growth and decline. In contrast, our research suggests that rhetoric (logos, ethos and pathos) in profit warnings favours logos throughout. The dominance of logos creates a business-as-usual impression, mimicking the characteristics of routine communication. This logos dominance suggests Carillion attempts to justify the profit warnings (large provisions and lower than expected profitability) by convincing its audience (shareholders, customers, public and media) using facts and figures (appeals to logic) and justifying its performance using reason, such as management actions to address the performance (appeals to reason) based on the rhetoric of effectiveness and reasonableness. One would expect Carillion to use ethos to enhance the credibility of the profit warning forecast. But initially, it does not do so, possibly because management is in denial/defiant about the situation. It is only at the desperation stage, with Carillion's share price continuing to fall, that Carillion starts to use ethos to enhance credibility by persuading audiences that Carillion management can be trusted and respected. But it is too little too late. The main way in which management tries to enhance credibility is by increasing appeals to honesty. We base our conjecture relating to trust on prior literature that suggests language use is related to persuasive trust building (Joyce, 2020) and specifically that ethos is related to “trust” (Beason, 1991). Some research suggests that ritual and visibility of executives bring meaning for participants to in-person meetings, such as board meetings and Q&A sessions of annual-result presentations (e.g. Roberts et al., 2006). We do not interview Carillion board members or the analysts, so we cannot conclude whether written profit warnings or Carillion managers' behaviour or a combination of both build trust. The parliamentary enquiry into Carillion's collapse concluded that “Effective stewardship by investors depends in large part on the availability of trustworthy financial reporting and on honest engagement with board members in response to the raising of concerns. The Carillion board failed on both these counts” and that Carillion's leaders “failed to publish the trustworthy information necessary for investors who relied on public statements to assess the strength of the company. Investors who sought to discuss their concerns about management failings with the board were met with unconvincing and incompetent responses” (House of Commons, 2018a, pp. 49–50). We similarly conclude based on our analysis of rhetoric.
We contribute theoretically by examining rhetoric in the context of a company's collapse. We examine Carillion's leaders' words in three documentary sources. We extend previous research (White and Hanson, 2000; Higgins and Walker, 2012; Brennan and Gray, 2000; Brennan and Merkl-Davies, 2014; Brennan et al., 2010) that, as immediate forms of communication to deliver bad news, profit warnings are written rhetorical vehicles. Rhetoric is also present in spoken forms of communication. We also extend previous research concerning the use of rhetoric as we show how rhetoric can be used to construct a “business-as-usual” reality. However, this reality differs from the realities constructed in the conference calls and the private board minutes. Our findings extend Davis and Tama-Sweet's (2012) research regarding differences in optimistic/pessimistic language across disclosure outlets to a consideration of rhetoric across such outlets. We analyse the behind-the-scenes choreography highlighting Carillion's leaders not telling the whole story.
We do not suggest that not telling the whole story led to Carillion's collapse. Indeed, telling/not telling the whole story does not preclude/guarantee failure. Most companies issuing profit warnings do not collapse. Rather, our findings reveal that whilst profit warnings might follow the letter of the law, they do not follow the spirit for which profit warnings are intended. Carillion's profit warnings inform investors that results are materially below expectations. However, the language used is reserved, moves from more helpful quantitative to less helpful qualitative forecasts and decreases in specificity. Furthermore, Carillion's managers used rhetoric to construct an alternative reality that the news was not bad but it was “business as usual”. The profit warnings only represent the publicly revealed tip-of-the-iceberg bad-news story. Carillion's leaders used language and rhetoric to construct a story to appear not as bad as the behind-the-scenes reality. Whether investors can trust the language in profit warnings is an area for future research. Empirically, we contribute by arguing that Carillion's leaders' motivation to construct a “business-as-usual” reality is linked to denial, defiance and desperation.
Earlier, we identified three stages in Bearbull's characterisation of disclosure of profit warnings in threes. In Section 1, we adapt Bearbull's characterisation by relating our findings to the four stages in our case. Figure 3 illustrates our findings. We extend Sandell and Svensson's (2016) language of failure by connecting our analysis of rhetoric in the three profit warnings with our three stages of communication: denial, defiance and desperation, which in our case company culminated in a fourth stage, i.e. defeat.
In conference calls, Carillion tries to convince the audience (analysts representing institutional investors) by appealing to logos (logic and reason) and pathos (emotion), which is high in both conference calls. The conference calls use the spoken word. The conference calls include a presentation by management (primarily the chairman, CEO and chief financial officer (CFO)) and a question-and-answer session between management and analysts attending the conference call. Therefore, some parts of the conference calls more closely represent a conversation. The high presence of pathos suggests that Carillion management wanted to evoke an emotional response from its audience. This pathos might indicate that whilst Carillion's management was aware of the company's problems, it was in denial about their seriousness.
Board discussions in the board minutes appeal to logos (logic and reason) and ethos (honesty). The Carillion board seemed oblivious (whether consciously or unconsciously) to the company's problems and looked forward to an upbeat announcement in Profit Warning 1. The board minutes are notably more honest about Carillion's problems. Whereas the profit warnings attribute the problems primarily to specific contracts and markets, the board minutes reveal issues of sloppy accounting and unhappiness that a proposed equity issue was not underwritten. However, despite these problems, the board justifies these issues by appealing primarily to reason (logos), for example, that the problems not only were understandable or based on estimates but also appealing to authority (ethos) that the auditor, KPMG, [4] was satisfied with the results or that the accounting methods complied with accounting standards. The board minutes show that the profit warnings did not reveal the extent of Carillion's problems. This disparity could be deliberately self-serving, giving the appearance of rationality, when the board knew the situation was otherwise; or unconscious bias in that the board believed the problems were not that disastrous.
This research is one of the first to analyse profit warnings with depth as a corporate communication genre. We base our findings on a single case company and three profit warnings and are, therefore, limited. The circumstances of the production of our three documentary sources are quite different and make comparison challenging. The next step is to analyse the content of a larger sample of profit warnings to find systematic patterns in these types of corporate reports (Brennan et al., 2021). In contrast to our rhetorical approach, other researchers have examined managers' use of impression management when disclosing bad news (e.g. Westphal and Park, 2020). Studying profit warnings through an impression management lens is likely to add to our understanding of these under-researched documents.
Merkl-Davies and Brennan (2017) conjecture that companies remain silent on activities with negative impact, which Shrives and Brennan (2017, p. 32) term “the rhetoric of silence”. The temptation for silence is likely to be acute in the case of profit warnings in which context Elder (2020) observes “their [companies'] silence now is just saving up problems for later”. Silence not only includes non-disclosure, but also delayed timing in disclosing a profit warning. The Financial Conduct Authority (Pooley, 2018) announced an investigation into the timing of Carillion's profit warning [5]. The study of silence in profit warnings, a difficult construct to research, merits attention. Another difficult but worthy area of study would be the preparation and production of trading updates/forecasts/profit warnings. Gibbins et al.’s (1990, 1992) research merits extension to corporate genres other than annual reports.
Profit warnings are a relatively immediate corporate report for reducing the information gap between managers and investors. However, they are largely unregulated. Our study reveals the importance of subjecting these narratives to closer examination and discussion. Managers are strategic and can influence the narrative publicly presented to investors about bad news. Managers must consider the moral and accountability implications of expressing their position to investors, particularly issues that are difficult to identify, for example, the accounting issues discussed in the board minutes.
We opened our paper with Bearbull's (2002) caution/conventional wisdom that profit warnings come in threes and with Roberts' (2014) caution that investors trading on profit warnings is like catching a falling knife. Our paper tells the more human story of why that happens, identifying the different stages of human psychology and emotions revealed in management's use of rhetoric as Carillion's collapse unfolds. Thus, to Bearbull's (2002) and Roberts' (2014) admonitions, we add our four Ds reflected in the language/rhetoric in the profit warnings, conference calls and board meeting minutes describing the pattern of management's reactions in the eye of the storm: denial, defiance, desperation and defeat.
Figures
Data chronology and data analysed
| Corporate document | Date | |||
|---|---|---|---|---|
| Panel A: Chronology of data | ||||
| Board minutes 1 | 09.05.2017 | |||
| Board minutes 2 | 15.05.2017 | |||
| Board minutes 3 | 23.05.2017 | |||
| Board minutes 4 | 08.06.2017 | |||
| Board minutes 5 | 05.07.2017 | |||
| Board minutes 6 | 09.07.2017 | |||
| Profit warning 1 | 10.07.2017 | |||
| Conference call 1 | 10.07.2017 | |||
| Board minutes 7 | 22.08.2017 | |||
| Profit warning 2 | 29.09.2017 | |||
| Conference call 2 | 29.09.2017 | |||
| Profit warning 3 | 17.11.2017 | |||
| Panel B: Data analysed | No. sentences/ phrases | No. Words | Total words | |
| Profit warning 1 | 10.07.2017 | 32 | 714 | |
| Profit warning 2 | 29.09.2017 | 46 | 549 | |
| Profit warning 3 | 17.11.2017 | 18 | 552 | 1,815 |
| Conference call 1 | 10.07.2017 | 369 | 8,029 | |
| Conference call 2 | 29.09.2017 | 584 | 10,344 | 18,373 |
| Board minutes 1 | 09.05.2017 | 125 | 2,772 | |
| Board minutes 2 | 15.05.2017 | 57 | 1,283 | |
| Board minutes 3 | 23.05.2017 | 104 | 2,389 | |
| Board minutes 4 | 08.06.2017 | See note | ||
| Board minutes 5 | 05.07.2017 | 134 | 2,658 | |
| Board minutes 6 | 09.07.2017 | 28 | 578 | |
| Board minutes 7 | 22.08.2017 | 145 | 3,192 | 12,872 |
| Total | 1,642 | 33,060 | 33,060 | |
Note(s): We do not analyse the 08.06.2017 board minutes due to significant redaction
Illustrations of rhetorical strategies/sub-categories
| Rhetorical strategy | Rhetorical sub-category | Illustrative example/sentence (key words guiding analysis underlined) | |
|---|---|---|---|
| Logos | 1 | Logic | H1 revenue expected to be similar to that in 2016 at approximately £2.5bn. (Profit warning, 10 July 2017) |
| 2 | Reason | Only undertaking future construction work on a highly selective basis and via lower-risk procurement routes. (Profit warning, 10 July 2017) | |
| Ethos | 3 | Values | Zafar and I also intend to back this up with transparency. (Conference call, 10 July 2017) |
| 4 | Authority | Mrs Horner noted she wanted Slaughter and May's view. (Board minutes, 9 May 2017) | |
| 5 | Similarities | KPMG concurred with the output of management's review (board minutes, 23 May 2017) | |
| 6 | Deference | The chairman noted the position and invited questions from the board for Morgan Stanley and Stifel. (Board minutes, 5 July 2017) | |
| 7 | Expertise | Carillion has a strong brand with a long history. (Conference call, 29 September 2017) | |
| 8 | Honesty | In other words the entries were in the balance sheet but in the wrong place. (Board minutes, 9 May 2017) | |
| 9 | Inclination to succeed | […] and our outlook is well underpinned by our order book and ability to win good work (Conference call, 10 July 2017) | |
| Pathos | 10 | Emotion | Without doubt, Carillion is a complex business and there is no such thing as a quick fix. (Conference call, 29 September 2017) |
Source(s): (See analytical framework in Table A1)
Language and other characteristics of the profit warnings
| Profit warning 1 10.07.2017 | Profit warning 2 29.09.2017 | Profit warning 3 17.11.2017 | |
|---|---|---|---|
| Wording of the profit warning | Revenue now expected to be between £4.8bn and £5.0bn and overall performance expected to be below management's previous expectations | lower than current market expectations | Materially lower than current market expectations |
| Quantification | Yes | No | No |
| Specificity | Range forecast | Not applicable | Not applicable |
| Revenue £4.8bn - £5.0bn | Not applicable | Not applicable | |
| Forecast horizon | 174 days | 93 days | 44 days |
Causal explanations/attributions in the profit warnings
| Forecast | No | Causal explanation |
|---|---|---|
| Profit warning 1 | 1 | […] Phasing of public private partnerships (PPP) equity disposals, which are now expected to be in H2 |
| Profit warning 2 | 1 | The phasing of PPP equity disposals |
| 2 | The trading of contracts with H1 provisions at zero margin | |
| Profit warning 3 | 1 | A combination of delays to certain PPP disposals |
| 2 | A slippage in the commencement date of a significant project in the Middle East | |
| 3 | Lower than expected margin improvements across a small number of UK support services contracts | |
| 4 | Partially offset by cost savings initiatives realised in the fourth quarter |
Proposed management actions in response to poor performance
| Forecast | No | Board/Management action |
|---|---|---|
| Profit warning 1 10.07.2017 | 1 | Cost reduction underway |
| 2 | Disposal of 50 per cent of the economic interest in the Group's business in Oman, Carillion Alawi, for an immediate cash consideration of £12.8m | |
| 3 | Board to undertake an enhanced review of all of the Group's material contracts, with the support of KPMG and its contracts specialists | |
| 4 | Disposals to exit non-core markets and geographies to raise up to a further £125m in the next 12 months | |
| 5 | Further annual cost savings to be quantified as part of the strategic and operational review | |
| 6 | Maximising the recovery of receivables | |
| 7 | 2017 dividends suspended resulting in a cash saving of approximately £80m | |
| 8 | The board announces today that it is undertaking a comprehensive review of the business and the capital structure, with all options to optimise value for the benefit of shareholders under consideration | |
| 9 | Significant actions already taken to reposition the business | |
| 10 | Exit from construction PPP projects | |
| 11 | Exit from construction markets in Qatar, the Kingdom of Saudi Arabia and Egypt | |
| 12 | Only undertaking future construction work on a highly selective basis and via lower-risk procurement routes | |
| 13 | Richard Howson has stepped down as Group Chief Executive and from the board with immediate effect […] Richard will stay with the Group for up to one year to support the transition | |
| 14 | Keith Cochrane, previously our Senior Independent Non-Executive Director, will take over as interim Group Chief Executive, whilst a search is underway for a new Group Chief Executive | |
| Profit warning 2 29.09.2017 | 1 | Business refocussed on core strengths and markets – support services, infrastructure and building |
| 2 | New leadership team and operating model – delayered structure, greater accountability and transparency | |
| 3 | Initial cost reduction target of £75m by mid-2019 | |
| 4 | Actions underway to improve cash flow and strengthen balance sheet | |
| 5 | Expected proceeds from non-core business disposals increased to £300m from £125m | |
| 6 | Discussions ongoing regarding sales of Carillion's business in Canada and the UK healthcare business | |
| 7 | Pension deficit reduction of £80m, potential to reduce further by £120m | |
| 8 | Agreed further £140m committed facility with a number of banks | |
| Profit warning 3 17.11.2017 | 1 | Since July, the Group has been focussed on reducing costs, collecting cash, executing its disposals programme and implementing its new operating model |
| 2 | The Board is therefore in discussions with stakeholders regarding a broad range of options to further reduce net debt and repair and strengthen the Group's balance sheet | |
| 3 | The Board expects to commence steps to implement the chosen option during the first quarter of 2018 and a further announcement will be made in due course | |
| 4 | The Board has kept under continuous review the risk that receipts from contract claims and/or disposals forecast to be received during November and December 2017 might slip beyond 31 December 2017 |
Rhetoric in profit warnings (RQ1)
| Rhetorical strategy | Rhetorical sub-category | Profit warning 1 10.07.2017 | Profit warning 2 29.09.2017 | Profit warning 3 17.11.2017 | Total | ||||
|---|---|---|---|---|---|---|---|---|---|
| Sentences analysed | No | (%) | No | (%) | No | (%) | No | (%) | |
| Logos | 1. Logic | 14 | (32) | 26 | (54) | 2 | (9) | 42 | (37) |
| 2. Reason | 17 | (39) | 8 | (17) | 11 | (48) | 36 | (31) | |
| 31 | (71) | 34 | (71) | 13 | (57) | 78 | (68) | ||
| Ethos | 3. Values | 2 | (5) | 2 | (4) | 1 | (4) | 5 | (4) |
| 4. Authority | 1 | (2) | 0 | (0) | 2 | (9) | 3 | (3) | |
| 5. Similarities | 0 | (0) | 0 | (0) | 0 | (0) | 0 | (0) | |
| 6. Deference | 0 | (0) | 0 | (0) | 1 | (4) | 1 | (1) | |
| 7. Expertise | 1 | (2) | 0 | (0) | 0 | (0) | 1 | (1) | |
| 8. Honesty | 5 | (11) | 4 | (8) | 5 | (22) | 14 | (12) | |
| 9. Inclination to succeed | 4 | (9) | 3 | (6) | 0 | (0) | 7 | (6) | |
| 13 | (29) | 9 | (19) | 9 | (39) | 31 | (27) | ||
| Pathos | 10. Emotion | 0 | (0) | 5 | (10) | 1 | (4) | 6 | (5) |
| 44 | (100) | 48 | (100) | 23 | (100) | 115 | (100) | ||
Note(s): The number of profit warning phrases/sentences in this table are greater than those in Table 1, as some phrases/sentences contain more than one rhetorical strategy
Rhetoric in conference calls (RQ2)
| Rhetorical strategy | Rhetorical sub-category | Conference call 1 10.07.2017 | Conference call 2 29.09.2017 | Total | |||
|---|---|---|---|---|---|---|---|
| Sentences analysed | No | (%) | No | (%) | No | (%) | |
| Logos | 1. Logic | 98 | (19) | 102 | (17) | 200 | (18) |
| 2. Reason | 156 | (30) | 142 | (23) | 298 | (27) | |
| 254 | (49) | 244 | (40) | 498 | (45) | ||
| Ethos | 3. Values | 13 | (3) | 23 | (4) | 36 | (3) |
| 4. Authority | 7 | (1) | 20 | (3) | 27 | (2) | |
| 5. Similarities | 12 | (2) | 7 | (1) | 19 | (2) | |
| 6. Deference | 13 | (3) | 10 | (2) | 23 | (2) | |
| 7. Expertise | 32 | (6) | 26 | (4) | 58 | (5) | |
| 8. Honesty | 45 | (9) | 61 | (10) | 106 | (9) | |
| 9. Inclination to succeed | 35 | (7) | 35 | (6) | 70 | (6) | |
| 157 | (31) | 182 | (30) | 339 | (29) | ||
| Pathos | 10. Emotion | 106 | (20) | 189 | (30) | 295 | (26) |
| 517 | (100) | 615 | (100) | 1,132 | (100) | ||
Illustrations of pathos – appeals to emotion
| Conference call 1 | Conference call 2 |
|---|---|
|
|
Rhetoric in board minutes (RQ2)
| Rhetorical strategy | Rhetorical sub-category | Board minutes 1 09.05.2017 | Board minutes 2 15.05.2017 | Board minutes 3 23.05.2017 | Board minutes 5 05.07.2017 | Board minutes 6 09.07.2017 | Board minutes 7 22.08.2017 | Total | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Sentences analysed | No | (%) | No | (%) | No | (%) | No | (%) | No | (%) | No | (%) | No | (%) | |
| Logos | Logic | 1 | (1) | 6 | (11) | 12 | (11) | 7 | (8) | 1 | (11) | 6 | (4) | 33 | (7) |
| Reason | 39 | (34) | 25 | (49) | 40 | (38) | 29 | (35) | 1 | (11) | 46 | (31) | 180 | (35) | |
| 40 | (35) | 31 | (60) | 52 | (49) | 36 | (43) | 2 | (22) | 52 | (35) | 213 | (42) | ||
| Ethos | Values | 6 | (5) | 3 | (6) | 7 | (7) | 5 | (6) | 2 | (22) | 21 | (14) | 44 | (9) |
| Authority | 14 | (12) | 4 | (8) | 17 | (16) | 8 | (10) | 3 | (34) | 2 | (1) | 48 | (9) | |
| Similarities | 12 | (10) | 4 | (8) | 2 | (2) | 0 | (0) | 0 | (0) | 5 | (3) | 23 | (4) | |
| Deference | 13 | (11) | 4 | (8) | 2 | (2) | 3 | (4) | 2 | (22) | 0 | (0) | 24 | (5) | |
| Expertise | 1 | (1) | 0 | (0) | 0 | (0) | 1 | (1) | 0 | (0) | 7 | (5) | 9 | (2) | |
| Honesty | 21 | (18) | 3 | (6) | 17 | (16) | 14 | (17) | 0 | (0) | 35 | (23) | 90 | (17) | |
| Inclination to succeed | 2 | (2) | 0 | (0) | 0 | (0) | 10 | (12) | 0 | (0) | 9 | (6) | 21 | (4) | |
| 69 | (59) | 18 | (36) | 45 | (43) | 41 | (50) | 7 | (78) | 79 | (53) | 259 | (50) | ||
| Pathos | Emotion | 7 | (6) | 2 | (4) | 9 | (8) | 6 | (7) | 0 | (0) | 19 | (13) | 43 | (8) |
| 116 | (100) | 51 | (100) | 106 | (100) | 83 | (100) | 9 | (100) | 150 | (100) | 515 | (100) | ||
Analytical framework and sources from the prior literature for rhetorical sub-categories
| Logos | |||
|---|---|---|---|
| 1 | Appeal to logic |
| Brennan and Merkl-Davies (2014, p. 608) |
| 2 | Appeal to reason |
| Higgins and Walker (2012, p. 198); Brennan and Merkl-Davies (2014, p. 608); Hossain et al. (2019, Table 4) |
| |||
| |||
| |||
| |||
| |||
| |||
| Ethos | |||
| 3 | Appeal to values, morals, standards, ethics of the speaker |
| Brennan and Merkl-Davies (2014, p. 608) |
| 4a | Appeal to the authority of the speaker |
| Brennan and Merkl-Davies (2014, p. 608) |
| 4b | Appeal to the authority of another social actor (expert, independent authority, person of high social or moral standing) |
| Brennan and Merkl-Davies (2014, p. 608) |
| 4c | Appeal to the authority of the law |
| Brennan and Merkl-Davies (2014, p. 608, 621) |
| 5 | Appeal to the similarities between speaker and audience |
| Higgins and Walker (2012 p. 197) |
| 6 | Appeal to deference (speaker's respect for rights and feelings of the audience) |
| Higgins and Walker (2012, p.198) |
| 7 | Appeal to the expertise of the speaker |
| Higgins and Walker (2012, p. 198) |
| 8 | Appeal to the honesty of the speaker |
| Higgins and Walker (2012, p. 198) |
| 9 | Appeal to the inclination to succeed |
| Higgins and Walker (2012, pp. 197–198) |
| Pathos | |||
| 10 | Appeal to emotion |
| Brennan and Merkl-Davies (2014, p. 608); Higgins and Walker (2012, p. 198) |
| |||
| |||
Notes
Under Article 17 of the European Union's (2014) Market Abuse Regulation and the UK Listing Rules (Financial Conduct Authority, 2018), listed companies must disclose price-sensitive inside information. These disclosures are colloquially referred to as “trading updates”, which include “profit warnings” (Annex 2, Financial Conduct Authority, 2021).
As a robustness check of the manual analysis and to triangulate our approach, we use a Python script to automate the textual analysis (see Anand et al. (2020) for a review of the use of this software in accounting research). We conduct a computer-aided analysis at the word level. In contrast, the manual analysis is at the phrase/sentence level. First, we generate a combined word frequency table for all documents. Then, we generate bags of words (Loughran and McDonald, 2016). The author who manually coded the documents assigned the most frequent words into three categories (bags of words): logos, ethos and pathos. We use a Python Natural Language Processing (NLP) package to process and analyse our data. We test for differences in the percentage frequency of logos, ethos and pathos keywords in the bag of words to total words in each documentary source – profit warnings, conference call transcripts and board meeting minutes. Assuming a consistent story across disclosure vehicles/genres, our hypothesis is that there is no difference in the use of rhetoric in our three documentary sources. Our dependent variable is the percentage frequency of words in each document from the bag of words to total words in each document. The independent variable is the document type/group (profit warnings, conference calls and board minutes). Given the small sample size, we adopt a Kruskal–Wallis H non-parametric test to test for differences in medians between the three documentary sources. We find statistically significant differences in the intensity of logos keywords between the three documentary sources. At the 10% level, there is a statistically significant difference in the intensity of pathos keywords between the three groups. There is no statistically significant difference in the intensity of ethos keywords between the three groups. Full details of this analysis are available from the authors on request.
Baroness Morgan of Huyton made this comment. She joined the Carillion board on 30 June 2017, 11 days before the first profit warning. This board meeting was only her second.
In September 2021, the Financial Reporting Council (FRC) (2021) announced that it would conduct a disciplinary tribunal over allegations that KPMG provided false or misleading information to the FRC regarding its audits of Carillion.
In September 2020, the Financial Conduct Authority (FCA) (2020) gave Carillion and certain executive directors a warning notice each. In November 2020, the FCA publicly censured without financial penalty unnamed executive directors for recklessly making “[…] misleadingly positive statements about Carillion's financial performance generally […] which did not reflect significant deteriorations in the expected financial performance […].” The wording suggests the statements referred to were the profit warnings.
References
Aerts, W. (2005), “Picking up the pieces: impression management in the retrospective attributional framing of accounting outcomes”, Accounting, Organizations and Society, Vol. 30 No. 6, pp. 493-517.
Aerts, W. and Yan, B. (2017), “Rhetorical impression management in the letter to shareholders and institutional setting: a metadiscourse perspective”, Accounting, Auditing & Accountability Journal, Vol. 30 No. 2, pp. 404-432.
Alves, P., Pope, P.F. and Young, S. (2009), “Cross‐border information transfers: evidence from profit warnings issued by European firms”, Accounting and Business Research, Vol. 39 No. 5, pp. 449-472.
Amernic, J. and Craig, R. (2006), CEO-speak: the Language of Corporate Leadership, McGill-Queen’s University Press, Kingston.
Amernic, J. and Craig, R. (2013), “Leadership discourse, culture, and corporate ethics: CEO-speak at news corporation”, Journal of Business Ethics, Vol. 118 No. 2, pp. 379-394.
Amernic, J. and Craig, R. (2017), “CEO speeches and safety culture: British Petroleum before the Deepwater Horizon disaster”, Critical Perspectives on Accounting, Vol. 47, pp. 61-80.
Anand, V., Bochkay, K., Chychyla, R. and Leone, A.J. (2020), “Using Python for text analysis in accounting research”, Foundations and Trends in Accounting, Vol. 14 No. 3, pp. 128-359.
Aristotle (350BC/2010), Rhetoric, (Trans by W. Rhys Roberts), The Pennsylvania State University, University Park, Pennsylvaniya, PA.
Baginski, S.P., Hassell, J.M. and Kimbrough, M.D. (2004), “Why do managers explain their earnings forecasts?”, Journal of Accounting Research, Vol. 42 No. 1, pp. 1-29.
Bearbull (2002), Warn, Then Warn Again, Investors Chronicle, 13 December 2002, available at: https://www.investorschronicle.co.uk/2012/03/07/warn-then-warn-again-2hllw55gYpOoOdYvLlL9tN/article.html (accessed 18 April 2020).
Beason, L. (1991), “Strategies for establishing an effective persona: an analysis of appeals to ethos in business speeches”, The Journal of Business Communication, Vol. 28 No. 4, pp. 326-346.
Beattie, V. (2014), “Accounting narratives and the narrative turn in accounting research: issues, theory, methodology, methods and a research framework”, The British Accounting Review, Vol. 46 No. 2, pp. 111-134.
Beattie, V. and Davison, J. (2015), “Accounting narratives: Storytelling, philosophising and quantification”, Accounting and Business Research, Vol. 45 Nos 6-7, pp. 655-660.
Beelitz, A. and Merkl-Davies, D.M. (2012), “Using discourse to restore organisational legitimacy: ‘CEO-speak’ after an incident in a German nuclear power plant”, Journal of Business Ethics, Vol. 108 No. 1, pp. 101-120.
Benoit, W.L. (2001), “Framing through temporal metaphor: the ‘bridges’ of Bob Dole and Bill Clinton in their 1996 acceptance addresses”, Communication Studies, Vol. 52 No. 1, pp. 70-84.
Bochkay, K., Hales, J. and Chava, S. (2020), “Hyperbole or reality? Investor response to extreme language in earnings conference calls”, The Accounting Review, Vol. 95 No. 2, pp. 31-60.
Brennan, N. and Gray, S.J. (2000), Rhetoric and Argument in Financial Reporting: Disclosures in Profit Forecasts and Takeover Documents, Association of Chartered Certified Accountants Occasional Research Paper 31, Certified Accountants Educational Trust, London.
Brennan, N.M. and Merkl-Davies, D.M. (2014), “Rhetoric and argument in social and environmental reporting: the dirty laundry case”, Accounting, Auditing and Accountability Journal, Vol. 27 No. 4, pp. 602-633.
Brennan, N.M., Daly, C.A. and Harrington, C.S. (2010), “Rhetoric, argument and impression management in hostile takeover defence documents”, British Accounting Review, Vol. 42 No. 4, pp. 253-268.
Brennan, N.M., Edgar, V.C. and Power, S.B. (2021), “COVID-19 profit warnings: delivering bad news in a time of crisis”, The British Accounting Review, in press, doi: 10.1016/j.bar.2021.101054.
Bryman, A. (2012), Social Research Methods, 4th ed., Oxford University Press, Oxford.
Bujaki, M. and McConomy, B. (2012), “Metaphor in Nortel's letters to shareholders 1997–2006”, Accounting, Auditing and Accountability Journal, Vol. 25 No. 7, pp. 1113-1139.
Charteris-Black, J. (2004), Corpus Approach to Critical Metaphor Analysis, Palgrave-Macmillan, Basingstoke.
Chen, M. and Chang, A.C.C. (2017), “Effects of language vividness and explanation plausibility on nonprofessional investors' reactions to earnings warnings”, Journal of Accounting and Finance, Vol. 17 No. 9, pp. 46-59.
Chen, Z. and Loftus, S. (2019), “Multi-method evidence on investors' reactions to managers' self-inclusive language”, Accounting, Organizations and Society, Vol. 79, p. 101071.
Cheney, G., Christensen, L.T., Conrad, C. and Lair, D.J. (2004), “Corporate rhetoric as organisational discourse”, in Grant, D., Hardy, C., Oswick, C. and Putnam, L. (Eds), The Sage Handbook of Organizational Discourse, Sage Publications, London, pp. 79-104.
Courtis, J.K. (2004), “Colour as visual rhetoric in financial reporting”, Accounting Forum, Vol. 28 No. 3, pp. 265-281.
Craig, R.J. and Amernic, J.H. (2004), “Enron discourse: the rhetoric of a resilient capitalism”, Critical Perspectives on Accounting, Vol. 15 Nos 6/7, pp. 813-851.
Craig, R., Garrott, L. and Amernic, J. (2001), “A ‘close reading’ protocol to identify perception-fashioning rhetoric in web site financial reporting: the case of Microsoft”, Accounting and The Public Interest, Vol. 1, pp. 1-16.
Danielewicz-Betz, A. (2016), “External corporate communication: quarterly earnings conference calls”, in Danielewicz-Betz, A. (Ed.), Communicating in Digital Age Corporations, Macmillan Publishers, London, pp. 277-307.
Davis, A.K. and Tama‐Sweet, I. (2012), “Managers' use of language across alternative disclosure outlets: earnings press releases versus MD&A”, Contemporary Accounting Research, Vol. 29 No. 3, pp. 804-837.
Davis, A.K., Piger, J.M. and Sedor, L.M. (2012), “Beyond the numbers: measuring the information content of earnings press release language”, Contemporary Accounting Research, Vol. 29 No. 3, pp. 845-868.
Davison, J. (2008), “Rhetoric, repetition, reporting and the ‘dot.com’ era: words, pictures, intangibles”, Accounting, Auditing and Accountability Journal, Vol. 21 No. 6, pp. 791-826.
Elder, B. (2020), Companies' Silence on Coronavirus Could Be Storing up Problems, Financial Times, 29 February 2020, available at: https://www.ft.com/content/23debec6-5a0f-11ea-a528-dd0f971febbc (accessed 18 April 2020).
Elliott, W.B., Hobson, J.L. and Jackson, K.E. (2011), “Disaggregating management forecasts to reduce investors' susceptibility to earnings fixation”, The Accounting Review, Vol. 86 No. 1, pp. 185-208.
European Union (2014), Regulation (EU) No 596/2014 of the European Parliament and of the Council of 16 April 2014 on Market Abuse (Market Abuse Regulation) and Repealing Directive 2003/6/EC of the European Parliament and of the Council and Commission Directives 2003/124/EC, 2003/125/EC and 2004/72/EC, Official Journal of the European Union, Brussels, EU, available at: http://data.europa.eu/eli/reg/2014/596/2021-01-01.
Financial Conduct Authority (2021), Disclosure Guidance and Transparency Rules Sourcebook, FCA, London.
Financial Conduct Authority (2018), Listing Rules, FCA, London.
Financial Conduct Authority (2020), Warning Notice Statement 20/2, FCA, London, available at: https://www.fca.org.uk/search-results?search_term=carillion (accessed 30 May 2021).
Financial Reporting Council (2021), FRC Announces Formal Complaint against KPMG and Others Regarding Allegedly False And/or Misleading Information provided to the FRC in Carillion and Regenersis Audit Quality Inspections, FRC, London, available at: https://www.frc.org.uk/news/september-2021/frc-announces-formal-complaint-against-kpmg-and-ot (accessed 10 September 2021).
Francis, J., Schipper, K. and Vincent, L. (2002), “Expanded disclosures and the increased usefulness of earnings announcements”, The Accounting Review, Vol. 77 No. 3, pp. 515-546.
Gibbins, M., Richardson, A. and Waterhouse, J. (1990), “The management of corporate financial disclosure: opportunism, ritualism, policies, and processes”, Journal of Accounting Research, Vol. 28 No. 1, pp. 121-143.
Gibbins, M., Richardson, A. and Waterhouse, J. (1992), The Management of Financial Disclosure: Theory and Perspectives, Research Monograph 20, The Canadian Certified General Accountants Research Foundation, Vancouver, BC.
Green, S.E. Jr (2001), Rhetoric and the Institutionalization of Takeover Defenses in the S&P 1500 from 1975–1998, Doctoral dissertation, Harvard University, Boston.
Guan, Y., Lobo, G.J., Tsang, A. and Xin, X. (2020), “Societal trust and management earnings forecasts”, The Accounting Review, Vol. 95 No. 5, pp. 149-184.
Guo, W., Sengul, M. and Yu, T. (2020), “Rivals' negative earnings surprises, language signals, and firms' competitive actions”, Academy of Management Journal, Vol. 63 No. 3, pp. 637-659.
Herrick, J.A. (1992), “Rhetoric, ethics, and virtue”, Communication Studies, Vol. 43 No. 3, pp. 133-144.
Higgins, C. and Walker, R. (2012), “Ethos, logos, pathos: Strategies of persuasion n social/environmental reports”, Accounting Forum, Vol. 36 No. 3, pp. 194-208.
Hossain, M., Islam, M.T., Momin, M.A., Nahar, S. and Alam, M.S. (2019), “Understanding communication of sustainability reporting: application of symbolic convergence theory (SCT)”, Journal of Business Ethics, Vol. 160, pp. 563-586.
House of Commons (2018a), Business, Energy and Industrial Strategy and Work and Pensions Committees, Carillion, 16 May 2018, available at: https://publications.parliament.uk/pa/cm201719/cmselect/cmworpen/769/769.pdf (accessed 5 December 2019).
House of Commons (2018b), available at: https://www.parliament.uk/business/committees/committees-a-z/commons-select/work-and-pensions-committee/inquiries/parliament-2017/carillion-inquiry-17-19/publications/ (accessed 25 February 2020).
Hursti, K. (2011), “Management earnings forecasts: could an investor reliably detect an unduly positive bias on the basis of the strength of the argumentation?”, The Journal of Business Communication, Vol. 48 No. 4, pp. 393-408.
Huse, M. (2007), Boards, Governance and Value Creation: the Human Side of Corporate Governance, Cambridge University Press, Cambridge.
Hutton, A.P., Miller, G.S. and Skinner, D.J. (2003), “The role of supplementary statements with management earnings forecasts”, Journal of Accounting Research, Vol. 41 No. 5, pp. 867-890.
Hyland, K. (1998), “Exploring corporate rhetoric: metadiscourse in the CEO's letter”, The Journal of Business Communication, Vol. 35 No. 2, pp. 224-245.
Joyce, Y. (2020), “Building trust in crisis management: a study of insolvency practitioners and the role of accounting information and processes”, Contemporary Accounting Research, Vol. 37 No. 3, pp. 1622-1657.
Kasznik, R. and Lev, B. (1995), “To warn or not to warn: management disclosures in the face of an earnings surprise”, The Accounting Review, Vol. 70 No. 1, pp. 113-134.
Laine, M. (2009), “Ensuring legitimacy through rhetorical changes? A longitudinal interpretation of the environmental disclosures of a leading Finnish chemical company”, Accounting, Auditing and Accountability Journal, Vol. 22 No. 7, pp. 1029-1054.
Liff, R. and Wahlström, G. (2018), “Failed crisis communication: the Northern Rock bank case”, Accounting, Auditing and Accountability Journal, Vol. 31 No. 1, pp. 237-260.
Lok, J. (2010), “Institutional logics as identity projects”, Academy of Management Journal, Vol. 53 No. 6, pp. 1305-1335.
Loughran, T. and McDonald, B. (2016), “Textual analysis in accounting and finance: a survey”, Journal of Accounting Research, Vol. 54 No. 4, pp. 1187-1230.
Merkl-Davies, D.M. and Brennan, N.M. (2017), “A theoretical framework of external accounting communication: research perspectives, traditions and theories”, Accounting, Auditing and Accountability Journal, Vol. 30 No. 2, pp. 433-469.
Pooley, C.R. (2018), FCA Launches Investigation over Carillion Market Updates, Financial Times, 3 January 2018. available at: https://www.ft.com/content/96d2855e-6b17-38d3-91f9-f20ce2ca63df (accessed 18 April 2020).
Roberts, J. (2014), Profit Warnings: Catching the Falling Knife, Hargreaves Lansdown, available at: https://www.hl.co.uk/news/articles/archive/profit-warnings-catching-the-falling-knife (accessed 7 May 2020).
Roberts, J., Sanderson, P., Barker, R. and Hendry, J. (2006), “In the mirror of the market: the disciplinary effects of company/fund manager meetings”, Accounting, Organizations and Society, Vol. 31 No. 3, pp. 277-294.
Rutherford, B.A. (2005), “Genre analysis of corporate annual report narratives: a corpus linguistics–based approach”, The Journal of Business Communication, Vol. 42 No. 4, pp. 349-378.
Rutherford, B.A. (2013), “A genre-theoretic approach to financial reporting research”, The British Accounting Review, Vol. 45 No. 4, pp. 297-310.
Sandell, N. and Svensson, P. (2016), “The language of failure: the use of accounts in financial reports”, International Journal of Business Communication, Vol. 53 No. 1, pp. 5-26.
Sandell, N. and Svensson, P. (2017), “Writing write-downs: the rhetoric of goodwill impairment”, Qualitative Research in Accounting and Management, Vol. 14 No. 1, pp. 81-102.
Shanahan, F. and Seele, P. (2015), “Shorting ethos: exploring the relationship between Aristotle's ethos and reputation management”, Corporate Reputation Review, Vol. 18 No. 1, pp. 37-49.
Shrives, P.J. and Brennan, N.M. (2017), “Explanations for corporate governance non-compliance: a rhetorical analysis”, Critical Perspectives on Accounting, Vol. 49, pp. 31-56.
Stanton, P. and Stanton, J. (2002), “Corporate annual reports: research perspectives used”, Accounting, Auditing and Accountability Journal, Vol. 15 No. 4, pp. 478-500.
Suddaby, R. and Greenwood, R. (2005), “Rhetorical strategies of legitimacy”, Administrative Science Quarterly, Vol. 50 No. 1, pp. 35-67.
Vignone, M.J. (2012), “Family, buildings, and wars: organizational conceptual metaphors”, OD Practitioner, Vol. 44 No. 1, pp. 34-37.
Vincent, M. (2017), Profit Warnings Come in Threes, or Fours, or Fives, Financial Times, 23 January 2017. available at: https://www.ft.com/content/a90a155a-e168-11e6-9645-c9357a75844a (accessed 30 May 2021).
Westphal, J. and Park, S.H. (2020), Symbolic Management: Governance, Strategy, and Institutions, Oxford University Press, Oxford.
White, R. and Hanson, D. (2000), “Rationality and rhetoric in the corporate world: the corporate annual report as an Aristotelian genre”, Prometheus, Vol. 18 No. 3, pp. 303-317.
Acknowledgements
The authors thank participants for their comments at the British Accounting and Finance Association 2021 annual (Online) conference, the European Accounting Association Virtual Annual Congress 2021 and the 13th Interdisciplinary Perspectives on Accounting Conference 2021. We are also grateful to two excellent reviewers for their constructive expert advice to improve the paper and to Editor, Lee Parker, for his handling of our paper.