Impression management at board meetings: accountability in public and in private

Abstract

Purpose

To demonstrate transparency and accountability, the three boards in this study are required to meet in public in front of an audience, although the boards reserve confidential issues for discussion in private sessions. This study examines boardroom public accountability, contrasting it with accountability in board meetings held in private. The study adopts Erving Goffman's impression management theory to interpret divergences between boardroom behaviour in public and private, or “frontstage” and “backstage” in Goffman's terminology.

Design/methodology/approach

The research observes and video-records three board meetings for each of the three boards (nine board meetings), in public and private. The research operationalises accountability in terms of director-manager question-and-answer interactions.

Findings

In the presence of an audience of local stakeholders, the boards employ impression management techniques to demonstrate accountability, by creating the impression that non-executive directors are performing challenge and managers are providing satisfactory answers. Thus, they “save the show” in Goffman terms. These techniques enable board members and managers to navigate the interface between demonstrating the required good governance and the competence of the organisations and their managers, while not revealing issues that could tarnish their image and concern the stakeholders. The boards need to demonstrate to the audience that “matters are what they appear to be”, even if they are not. The research identifies behaviour consistent with impression management to manage this complexity. The authors conclude that regulatory objectives have not met their transparency aspirations.

Originality/value

For the first time, the research studies the effect of transparency regulations (“sunshine” laws) on the behaviour of boards of directors meeting in public. The study contributes to the embryonic literature based on video-taped board meetings to access the “black box” of the boardroom, which permits a study of impression management at board meetings not previously possible. This study extends prior impression management theory by identifying eleven impression management techniques that non-executive directors and managers use and which are unique to a boardroom context.

Citation

Pernelet, H.R. and Brennan, N.M. (2023), "Impression management at board meetings: accountability in public and in private", Accounting, Auditing & Accountability Journal, Vol. 36 No. 9, pp. 340-369. https://doi.org/10.1108/AAAJ-09-2022-6050

1. Introduction

The English National Health Service (NHS) regulator introduced a requirement for its boards to meet in public, arguing that transparency is a key pillar of good governance, although there was no evidence as to whether this was the case. These regulations are similar to the US “sunshine” laws, introduced to enhance the transparency of “open meetings” (as opposed to “closed meetings”) (Piotrowski and Borry, 2010). As is evident from the NHS report introducing the regulations (NHS Commissioning Board, 2012), from the title onwards, public accountability is the objective of this transparency. The Report (p. 16) states that good governance requires “transparency and public accountability” to be built in. The Report argues that “transparency [is] important for public accountability … to promote confidence between the CCG [Clinical Commissioning Group] and its staff, patients and the public” (NHS Commissioning Board, 2012, pp. 23–24).

The three boards in this study lead independent English NHS statutory clinical commissioning groups. Clinical commissioning group boards are required to demonstrate good governance to public observers (or local stakeholders). Thus, each board meeting is held in public, although the boards discuss confidential issues at meetings held in private. Thus, uniquely, the boards execute both external (i.e. in public) and internal (i.e. in private) accountability in separate fora. We explore how accountability operationalised as director-manager question-and-answer interactions in public shape non-executive directors' behaviours.

Does transparency improve accountability? Kosack and Fung (2014) observe that the relationship between transparency and accountability is increasingly ambiguous. They argue that such ambiguities temper the potential for transparency to improve governance. Roberts (2009) suggests that transparency, as a form of accountability, is “ambivalent” and that transparency is an illusion. Quattrone (2022, p. 547) characterises transparency as a paradox, and argues that seeking transparency “makes us blind”, and “makes notions of values, measurement and accountability more fragmented”. We study accountability in public and in private and thus examine whether transparency improves accountability.

A key challenge facing researchers is discovering how boards work and what happens when boards are in session (Stiles and Taylor, 2002). Zattoni and Pugliese (2019, p. 107) observe that few studies explore the concept of accountability in governance, from the perspective of board accountability processes and their execution by boards of directors, despite its centrality, citing Brennan et al. (2016) as an exception. Researchers' inability to open the “black box” (Lawrence, 1997) to observe boards in action (Leblanc and Schwartz, 2007) has the consequence that “board workings continue to be … hidden behind closed doors” (Halton, 2013, p. 423). The purpose of this study is to contribute to an understanding of boardroom behaviours by contrasting board meetings held in public, as required by regulation, with those held in private. Observing boards in both public and private fora provides the study with a level of behaviour and interactions not considered in prior boardroom research.

Goffman (1959) posits that people's behaviour changes in front of an audience. This raises the question of whether the behaviour of the three boards changes on the “frontstage” in Goffman terms during their public board sessions. We share Roberts et al.’s (2006, p. 278) observation that formal face-to-face accountability processes (at annual general meetings (AGMs)) are “augmented and subverted by a variety of other mechanisms”. Given our public/private meetings' context, we consider the “other mechanism” most appropriate for study to be Goffman's (1959) impression management, which he portrays as a performance on the frontstage. While researchers have used impression management to study a variety of topics, it has rarely been applied to study boards of directors. We conceptualise the behaviour in public of the directors of the three boards and their managers as creating the best possible impression for the audience that they are performing (good) governance. However, if the board members engage in impression management behaviour, this raises questions about the efficacy of the regulations intended to improve transparency and accountability.

We observe, video-record and audio-record three sequential (but for one) board meetings in public and in private at each of the three boards (nine board meetings in total). Video- and audio-recording board meetings, a relatively novel methodology in corporate governance, enables the application of an impression-management lens to the study of boards, previously not possible due to lack of access. Our findings provide insights into directors' and managers' use of impression management techniques unique to a boardroom context.

Our objective is to explain the divergences Pernelet and Brennan (2023a) identified in director-manager question-and-answer interactions in the public board sessions compared with the private sessions. Our research question is: How can divergences in director-manager interactions at board meetings in public and in private be conceptualised through an impression-management lens? A Goffman impression-management lens enhances our understanding of the nuanced and subtle behaviours at play in boardrooms not considered in the prior literature. We assume that the divergences are associated with the presence of an audience which affects boardroom behaviour in public. We argue that, in public, board members collaborate with managers acting as “colleague groups” (Goffman, 1959, p. 221) to create the impression that governance is being performed. We suggest that the real accountability happens in private.

We make three contributions to the literature. First, we provide evidence for the first time on the effect of transparency regulations (“sunshine” laws) on accountability by obtaining access to board meetings held both in public and in private. The public/private meetings provide an opportunity to identify and interpret changes in boardroom behaviour and associated approaches to accountability in public in front of an audience. Second, we contribute to corporate governance research into board interactions (Samra-Fredericks, 2000a, b; Macus, 2008; Van Ees et al., 2009), which are a key component of board process. Our board-process focus is accountability in terms of challenge in the boardroom, which we operationalise as director-manager question-and-answer interactions. By analysing the interaction patterns reflecting the asking-and-answering of questions at board meetings in public, compared with in private, this research explores how these interactions contribute to the boards' public performance of corporate governance. We respond to Samra-Fredericks’ (2000b, p. 311) recommendation “to focus upon the talk of the board […] in order to gain a deeper understanding of the behavioural dynamics of corporate governance”. Our in vivo methodology also contributes to the work of Bezemer et al. (2014, 2018), Pugliese et al. (2015), Nicholson et al. (2017) and Veltrop et al. (2021), who study a variety of influences on director behaviour during board meetings. Finally, our paper uses an impression-management lens to explain the public/private divergences found, which enhances our understanding of the nuanced and subtle behaviours in boardrooms. We extend prior impression management theory by identifying eleven impression management techniques non-executive directors and managers use unique to a boardroom context.

Our paper is structured as follows. We commence Section 2 with our theoretical positioning of the study and a review of the prior relevant literature. Section 3 describes our research methodology, while Section 4 presents our findings. We draw some concluding observations in Section 5.

2. Theory and prior research

This section discusses Goffman's impression management theory and its suitability for our study, followed by prior research on boards in action and the limited research applying Goffman's theory to contexts similar to ours. We acknowledge that prior research is based on boards in action on the backstage in Goffman terms, whereas our research focuses on boards in action in public, on the frontstage. We discuss the relevance of impression management to accountability.

2.1 Goffman's impression management theory

Goffman's (1959) frontstage and backstage metaphor is especially appropriate to illuminate the contrasts between the public and private settings of the board meetings we study. Goffman (1959) “explored ways in which people present an image of how they think their audience wishes to see them in face-to-face interaction” (Solomon et al., 2013, p. 197). Goffman's (1959) dramaturgical analysis, or theatrical metaphor, illustrates how:

the individual in ordinary work situations presents himself and his activity to others, the ways in which he guides and controls the impression they form of him, and the kind of things he may and may not do while sustaining his performance before them (1959, p. xi). [1]

Individuals use various practices and techniques, including “the arts of impression management” (1959, p. 208), to ensure the success of the performance. The concept derives from social psychology, which views impression management as driven by social relations characterised by an anticipation of an evaluation of conduct (Merkl-Davies and Brennan, 2011). During board meetings in public, stakeholders observe non-executive directors evaluating managers' conduct, by means of question-and-answer interactions. We consider Goffman's impression management theory to be an appropriate theoretical lens to interpret the director-manager interactions at the three boards. Each board meeting occurs in two parts: in public, “frontstage” before an audience, and then in private, or “backstage” [2]. Goffman characterises backstage as having various functions within organisations, including where actors can relax and step out of character. However, it can also be a “safe place” (Goffman, 1959, p. 114), or a place that serves as a “protect space” where issues can be openly discussed (Mair and Hehenberger, 2014, p. 1188). Goffman (1959, p. 121) suggests that one of the most interesting times to observe impression management is when the performer is either moving from the backstage to the frontstage, or vice versa, “for at these moments we can detect a wonderful putting on and taking off of character.” As the in-private board sessions occur immediately before/after the meetings in public, our context is especially suitable for observing people “putting on and taking off of character”. In public, non-executive directors (NEDs) create the impression of performing their expected role of challenging management. Goffman (1959, p. 8) submits that social interaction is an “information game - a potentially infinite cycle of concealment, discovery, [and] false revelation.” Interactions at board meetings are also an “information game”, whereby non-executive directors question senior management to overcome information symmetry (Brennan et al., 2016), and perform their monitoring role.

Goffman (1959) devotes his penultimate chapter to the arts of impression management, drawing together the attributes required to successfully stage a character. Trevino (2003, p. 36) explains:

At the heart of all these dramaturgic practices […] is the art of impression management […] the activity through which a performer endeavors to present a favourable impression of herself before others. It is through impression management that the performer avoids ‘creating a scene’ that can cause her embarrassment and disrupt the social order of the interaction.

Thus, performers use specific processes to ensure that the best performance and outcome emerge (Shulman, 2017).

2.1.1 Impression management: preventive and defensive practices/techniques

Impression management techniques underpin the effects of the performance. Goffman (1959, p. 13) divides these techniques into “preventive practices” used to avoid embarrassments and “performance disruption” (Goffman, 1959, p. 208/212) and “corrective” techniques used “to compensate for discrediting occurrences that have not been successfully avoided” (Goffman, 1959, p. 13), and therefore which still have the potential to disrupt the “show”. Such disruption threatens to reveal to the audience that things may not be quite as they seem. These two categories may be used either defensively or protectively. Goffman (1959) categorises impression management techniques as defensive when individuals use them to save their own performance, and as protective when they use them to save someone else's performance.

Goffman (1959) describes the attributes of defensive and protective impression management practices, each of which comprises many techniques by which a team of individuals safeguards its performance. Defensive practices require: first, dramaturgical loyalty between team members whereby they perform enthusiastically, maintain the party line and protect team secrets; second, dramaturgical discipline, which requires performers to perform their role properly and cover-up quickly for mistakes made by other team members; and third, dramaturgical circumspection, which requires individuals to carefully stage their performance, including by “keeping close to the facts” to safeguard their show (Goffman, 1959, p. 220).

Protective practices comprise tactful actions, by one or more individuals, which help save another's performance. Goffman (1959, p. 229) describes “the tactful tendency of the audience and outsiders to act in a protective way … to help the performers save their own show” and provides examples such as ignoring embarrassing moments in the performance, or accepting the excuses provided for them. Table 1 summarises these impression management techniques, which we draw from Goffman's (1959) Chapter 6 describing the arts of impression management.

2.2 Prior research on boards in action

Prior boards-in-action research (Watson et al., 2021; Watson and Ireland, 2021) considers board processes and board interactions. Board processes refer to boards-of-directors’ decision-making activities (including related aspects such as the quality of board meetings), the formality of board meetings and the interactions between executives and non-executive directors (McNulty and Pettigrew, 1999). Board processes rely on social interactions between participants in board meetings (Macus, 2008). Forbes and Milliken (1999, p. 489) emphasise that “researchers and practitioners alike are seeking to better understand the processes and behaviours involved in effective corporate performance”. Forbes and Milliken (1999, p. 492) suggest that a key dimension of board effectiveness is the ability of directors to work together in a cohesive way, and this dimension is reflected by director interactions in the boardroom (Pugliese et al., 2015). They add that board effectiveness is likely to depend on social-psychological processes, especially those that relate to group participation and interaction, the exchange of information and critical discussion. Bezemer et al. (2014) develop this thinking, suggesting that board performance is the result of director behaviours and group processes in and around the boardroom. While at a surface level their two boards appeared to follow similar processes, at a deeper level they find that factors at multiple levels of analysis explain differences in board interaction patterns, revealing the complex and nested nature of boardroom discussions.

A major challenge for corporate governance researchers analysing board processes is that it is difficult to study boards directly, generally for reasons of confidentiality. Zahra and Pearce (1989, p. 324) note that “… the relatively low number of empirical investigations into board processes is explained in part by the difficulty in securing access to boards to observe process over time … This is … one of the most challenging areas for future research in the contribution of boards”. Several authors have since reiterated that research is needed into the dynamics of what actually goes on in the boardroom, which generally remains a “black box” (Lawrence, 1997).

Some limited empirical research into the behaviour of boards has begun to open the “black box” to shed light on board roles, behaviour and relationships. Using board video-recorded observations, Pugliese et al. (2015) find patterns in director interactions, which are dynamic and multidimensional. They highlight three key dimensions of board interactions: speaking time, turn-taking and silence. They find that director interactions change with the nature of the agenda items, the board climate and board meeting arrangements. Where there is a more formal and disciplined board process, board members contribute to the debate “serially” and often at the chair's request. Less-formal board processes demonstrates spontaneous interactions, where directors interject and contribute without being asked to do so. Only a few directors lead discussions, but that the directors contributing to discussions vary according to agenda items.

Roberts and Scapens (1985, p. 447) describe accountability as “the giving and demanding of reasons for conduct”, i.e. the asking and answering of questions. Roberts (2001) develops the notion of “socializing forms of accountability” in a board-of-directors’ context, which he describes as face-to-face accountability between equal parties. He exemplifies such accountability as the responsibility jointly and severally shared by executive and non-executive directors. Nicholson et al. (2017) describe the board of directors as a transition point from external accountability (to external stakeholders) to internal accountability (whereby the board of directors, as the apex accountability body, holds management to account). Nicholson et al. (2017) explore board accountability routines. They find that boards engage in clear, recurrent accountability routines, with board members playing different roles in the routines depending on the issues before the board, allowing both directors and managers to hold each other to account. They observe a wide range of interactions. They conclude that the process of who is being held to account and who is holding them to account varies with agenda items.

2.2.1 Research on boards of directors using insights from Goffman

Research on boards of directors using Goffman is sparse. Christensen and Westenholz (1999, 2000) mobilise Goffman's theory in their study of Danish boards of directors and view the board as a theatre ensemble staging a performance for an audience of managers, consultants, auditors, and occasionally shareholders and stakeholders, such as creditors and employees. Christensen and Westenholz (2000, p. 1308) consider strategic decision-making by Danish companies' boards of directors, where employee representatives are board members. They believe Goffman's (1959) art of impression management suitable for the study of boards which comprise “relational analysis at the collective level”. Decision-making is about whether performers and audiences participate in a show that provokes enthusiasm from both parties. Again, they do not develop their analysis via impression management theory. Using interviews with board directors of publicly-traded small- and medium-sized companies, Charas and Perelli (2013) explore the link between board dynamics and governance, classifying director behaviours between formal board meetings (frontstage) versus outside-the-boardroom meetings (backstage).

2.2.2 Research on meetings-with-management using Goffman

Prior research has used Goffman's work to study question-and-answer (Q&A) interactions. The socially-oriented meetings-with-management stream of literature refers to meetings between financial analysts, advisers to investors and fund managers, and managers of publicly-traded companies that those analysts evaluate on behalf of their clients. This literature divides into two strands: public meetings, such as annual general meetings, and private meetings between analysts and management (Abraham and Bamber, 2017). The meetings-with-management can also be a combination of face-to-face and remote meetings, whereby analysts are invited to a physical meeting with management, which is also broadcast online to a wider audience of investors.

Drawing on the concept of the prison panopticon (a central observation tower placed within a circle of prison cells so all cells can be observed), from Foucault's (1977) theory of surveillance, Abraham and Bamber (2017) find that “the gaze of surveillance [from an online audience] transforms the Q&A into a dramaturgical encounter, where impression management techniques are important” (Abraham and Bamber, 2017, p. 15). Participants shift from being the target of surveillance to monitoring agents. When analysts ask questions, they move frontstage where management performs. Consequently, self-promotion may then impact the encounter as analysts perform their role asking questions for the benefit of their clients: “[Analysts] ask questions because they are advertising the fact that they're there … interrogating the CEO” (Abraham and Bamber, 2017, p. 23).

Their findings also show that being under surveillance impacts how questions are both asked and answered, entailing impression management. First, analysts may withhold questions due to the surveillance technologies, because they do not want to share their proprietary insights with their competitors. Second, analysts may be concerned about asking questions as these may lead to social humiliation, for example, because the question demonstrates their lack of understanding. Third, analysts do not want to damage their relationship with management by departing from a ritual code by asking questions that might lead management to lose face. Management may also resist responding to questions to avoid disclosing a discrepancy between appearance and actual activity. Thus, Abraham and Bamber (2017) illuminate the relevance of Goffman's (1959) impression management to question-and-answer sessions in public. Drawing on Goffman (1959), they identify one impression management technique – the practice of greeting analysts to the meeting using the person's name. Pre-existing social relationships between analysts and managers prompt greetings and farewells. Abraham and Bamber (2017) describe the greetings and farewells as small acts of deference and demeanour (Goffman, 1959), ensuring that analysts and managers do not destroy each other during the question-and-answer sessions.

Solomon et al. (2013) provide insights into impression management in private social and environmental reporting (SER) meetings and find that “both investors and investees employ Goffmanesque, staged impression management as a means of creating and dissembling a dual myth of social and environmental accountability” (Solomon et al., 2013, p. 195). Illuminated by Goffman's framework, Solomon et al. (2013, p. 196) contend that a private SER meeting is an “empty encounter” rather than a mechanism of accountability. They identify impression management as the tool both parties use to construct a dual myth of social and environmental accountability within meetings, rendering private reporting processes “impotent and empty” (Solomon et al., 2013, p. 196), while perpetuating a myth of accountability. Solomon et al. (2013, p. 202) contend that the dual myth is a “frontstage play [that is, a frontstage focus on environmental, social and governance during the SER meetings] against a backstage backdrop of financial prioritisation”. They summarise the meetings as “involv[ing] performance in an ‘empty theatre’, since both investors and investees are only concerned about the performance in the theatre ‘next door’ which is where the money is being made” (2013, p. 206). Thus, “rather than monitoring their investees during the performance, the investors appear to be … a passive uncritical audience” (2013, p. 210).

Corrigan (2018) video records open-to-the-public (for the purpose of democratic accountability) municipal budget meetings, and considers the meetings from the perspective of visual impression management tactics (skilful use of imagery). Corrigan (2018, p. 25) contends that “in municipal budget-making, impression management involves … creating a ‘monopoly of credibility’ … and when … effective it creates an aura of authenticity”. The audience of citizens attending the meetings is constrained by the meetings’ choreography which limits participation.

Johed and Catasús (2018) use the concept of face-work (Goffman, 1967) to consider the role of auditors at AGMs through a Goffman lens, and how auditors prepare backstage for the meetings. Auditors undertake backstage preparation with management prior to their annual performance frontstage, where their role is that of the independent auditor conveying trust to shareholders. Auditors are usually successful in preventing backstage issues from becoming discernible to shareholders (the audience). However, when incidents threaten to challenge the auditors' and management's face (or disrupt the show, using Goffman (1959) terminology), auditors act to save their own face as well as that of management. We highlight this “protective manoeuvre” (Goffman, 1967, p. 16) because it overlaps with the protective practices, which Goffman (1959) identifies as an impression management technique, as we note in Section 2.1.1 earlier. Johed and Catasús (2018, p. 370–371) consider this protective action by the auditors, and submit that where actors are part of a ritual (as during an AGM) they

… emotionally identify with others … and each participant involved in the ritual comes to believe that the ritual is a collective concern that carries a collective stake … this is what motivates others to save a person’s face … By stepping in and sustaining the ritual, the actors reinforce the myth,

This analysis of the rationale behind saving another's face sheds further light on protective impression management techniques that Goffman (1959) describes, which are applicable to the question-and-answer sessions at the board meetings in public that we observe.

2.3 Accountability and impression management

A tension exists between accountability and impression management (Frink and Ferris, 1998; Stapleton and Hargie, 2011). Social situations requiring accountability prompt impression management (Roussy and Rodrigue, 2018). Impression management may undermine accountability (Martins et al., 2020). In a study of legitimacy following a crisis, Conway et al. (2015) find evidence in support of impression management behaviour rather than the discharge of accountability. Ammeter et al. (2004) develop a theoretical model of role theory, trust, and accountability. They suggest that individuals may use impression management techniques to increase the trust others have in them, thus lowering accountability. They argue that trust may serve as a proxy for accountability. Hall et al. (2017) argue that accountability and impression management share common roots in terms of symbolic interactions. Individuals create meanings using interpersonal interactions such that behaviours are representations or symbolic. In these interactions, individuals anticipate the views of others and conform their behaviours to elicit desired responses from those salient others. This perspective reflects Spira's (1999, 2002) view of audit committees (board-of-directors’ sub-committees) as ceremonial.

To summarise, prior research has used Goffman's dramaturgical theory to explain performances of a variety of accountability-related meetings held in public and in private, such as general meetings, meetings with analysts, public budget meetings and private meetings between investors and investees. However, Goffman's dramaturgical theory has hardly been used in a board-of-directors’ context and provides an opportunity to develop our understanding of boardroom behaviours and how boards work. Specifically, we analyse accountability in terms of challenge in the boardroom, operationalised as board question-and-answer interactions, and how non-executive directors and managers perform and respond to challenge in front of a public audience, to create the impression that non-executive directors are performing challenge and managers are providing satisfactory answers.

3. Research questions and Methodology

This section describes our research question and methodology.

3.1 Research questions

Our research question is: How can divergences in director-manager interactions at board meetings in public and in private be conceptualised through an impression-management lens?

3.2 Population and selection of sample

Having previously served on an NHS statutory clinical commissioning group board, the first author had met directors of two clinical commissioning groups in the London area [3]. She approached the chairs of those boards (Board A and Board B) and received permission to video- and audio-record the public and private sessions of three board meetings each. She was then introduced to a third board, Board C, which also gave her permission to observe and record three of its meetings in public and private. Each of the three boards had almost identical governance structures, shared London-centric challenges, and were at similar stages in their organisational lifecycles, thus controlling for potential external sources of variation resulting from random sampling.

3.3 Data collection

Data sources comprised full verbatim transcripts of video- and audio-recordings of both the public and private sessions of three board meetings at each of the three boards Secondary data sources were documentary and were only used occasionally, for example, to provide the context for an agenda item if necessary or to understand better the facts underpinning board discussions. These comprised the board agenda and board papers from each meeting.

The three boards held their board meetings in accessible locations so that members of local communities across the respective boroughs could observe the meetings. However, public attendance varied by board. For Board A, members of the public were consistently present to observe the meetings and actively ask questions or make comments. The numbers attending varied from 5 to 12 people. Some individuals were regular attendees. For Board B, although no member of the public attended the public sessions, these meetings were videoed and streamed in real-time onto the organisation's website, for any members of the public who wished to watch the meeting live or at a later date. The recordings remained available on the website and could be easily accessed. For Board C, no members of the public attended the meetings that we observed, although seating and information was available in anticipation of public attendees. Individuals due to present to Board C, or other NHS representatives, occasionally sat in the public seating area for parts of the meeting, but not as members of the public.

A risk with observing boards is that the observation changes behaviour. The first author who conducted the recordings is confident that her presence did not affect boardroom behaviour. Board members did not pay much attention to her, generally did not address her (even though she knew some of them), and walked past her after meetings without speaking to her. None of the board members spoke to, or acknowledged, her during either the in-public or in-private sessions. When the meetings ended, she occasionally exchanged small talk with some board members, but she was not asked about the study.

3.4 Data analysed and data analysis

Pernelet and Brennan (2023a) develop a concept-driven director-manager question-and-answer typology/analytical framework and associated coding scheme, comprising six/48 question categories/subcategories and eight/69 answer categories/subcategories. Together, the challenge/questions and its response/answers comprise the most important interaction in boardrooms as they underpin directors' and managers' accountability. They identify divergences in director-manager question-and-answer interactions in the public board sessions compared with the private sessions. We summarise the data in Table 2. This shows the length (in minutes), the number of questions asked/answered at each board meeting in public and private and per hour and the transcript word count for each meeting. Eight board meetings occurred over five months in 2019 (January, March and May 2019), with one board meeting in September 2018. The length of the board meetings in public ranged from 83 min (1 h and 23 min) to 140 min (2 h and 20 min), while the length of the board meetings in private ranged from 16 min to 76 min (1 h and 16 min). On one occasion, Board C held its private session before the public board meeting. The nine board meetings produced over 24 h of observations and recordings totalling between 3,500 and 18,000 words per board meeting. We subsequently analysed the director-manager question-and-answer interactions in the transcripts manually.

Our dataset comprises 418 questions and 510 answers (see Table 2). For this paper, we focus on parts of the data that reveal divergences in director-manager question-and-answer interactions between the meetings in public and in private. For example, in Quotation 1, a NED questions a third party presenter in the public session (marked 2 in Quotation 1). This is followed in the subsequent in-private board session by criticism of the third party presenter's responses, thus demonstrating contrasting behaviour in public and in private. Quotation 1 also shows a NED's frontstage challenge (marked 2 in Quotation 1), followed by a continuation of the NED's challenge in the private session, backstage (marked 11 in Quotation 1). In the frontstage part, the manager responding deploys a corrective and defensive impression management technique, and provides a combination answer which deploys seven elements (marked 3–9 in Quotation 1) to respond to the NED's question. The combination answer is a corrective technique, which comprises two or more subcategories of answer (to one question) provided by one individual. In this case, it provides a combination of as many answer categories as possible, with the objective of stopping further difficult questions. It was effective in this example, distancing the manager from the plan's weaknesses, to enable the show to go on (see further discussion in Section 4.3.7).

Protectively, to prevent the performance from being further disrupted, NED 1 in Quotation 1 chooses not to ask a follow-up question in public but does so in private following the public session (marked 10 in Quotation 1). Other NEDs then join the discussion backstage, introducing further challenge, although they had not asked any questions in public (marked 13, 14, 15 in Quotation 1). We conceptualise this contrasting behaviour using an impression-management lens.

4. Findings

We commence this section by highlighting the divergences we identified between director-manager question-and-answer interactions in public and private. We then consider the defensive and protective impression management techniques used by the boards in the interactions.

4.1 Divergences between interactions in public and private

As shown in Table 3, there are five divergences (identified with the symbols ①②③④⑤) between director-manager question-and-answer interactions in public and private. First, the questions NEDs most often ask of management in public comprise simple questioning offering mild challenge①. In contrast, there is a much greater frequency of moderate challenge① questions in private. The lack of moderate challenge in public seems to be partially counterbalanced in private. Issues that are too sensitive or embarrassing to probe in public are deflected during the public sessions but are then openly discussed in private. Second, supportive② and optimistic comments② only occur in public. Third, combination questions③ (where two or more question subcategories are asked by one individual in one question) are more frequent in public (81 incidents) versus in private (23 incidents). All the supportive comments and most optimistic comments occur as part of combination questions in public, and precede a more substantive question. Fourth, moderate challenge questions often result in combination answers④ (where two or more subcategories of answer (to one question) are provided by one individual). There are 63 combination answers in public and only 27 in private. In addition, moderate challenge questions often result in collaborative answers (34 in public, 14 in private) (untabulated) (which involve several parties adding to each other's answers to respond to a question), which generally inhibits further questioning (untabulated but see Quotation 10). Finally, CEOs answer nearly twice as many questions in private as in public⑤, and a greater proportion in public are contributions to collaborative answers (18 out of 40, compared to 12 out of 75 answers in private) (untabulated). Management as a group responds to the largest number of questions in public (untabulated). This is as expected from those tasked with presenting reports to the boards, and then participating in the following Q&A session. In private, the CEO responds to the majority of questions, while contributing less than the managers in public.

4.2 Impression management techniques used by the boards in this study

We suggest that impression management techniques explain the key divergences between the interactions in public and private. Board members endeavour to protect the organisational narrative in public, while in private they try to deal with issues that they do not wish to publicise. Earlier in Table 1, we summarised Goffman's preventive and corrective/defensive and protective impression management techniques. In Appendix 1, we develop and specify eleven impression management techniques which apply in a board-of-directors’ context. Appendix 1 reflects Goffman's categories of dramaturgical loyalty, dramaturgical discipline and dramaturgical circumspection introduced earlier in Section 2.1.1/Table 1. As Goffman (1959, pp. 220–221) suggests, the performance will not “be judged by the results it achieves but by the degree to which … skills [of questioning] have been proficiently applied.”

4.2.1 The trade-off between performing challenge and protecting information

Board members use impression management techniques to manage the trade-off between demonstrating the board's good governance and the requirement to protect information which might reveal inconsistencies between the organisational narrative and reality. The boards need to demonstrate to the audience that “matters are what they appear to be” (Goffman, 1959, p. 17). Defensive or protective techniques can be used to either prevent “disruptions” (Goffman, 1959, p. 208) to the performance, or to correct the performance when a disruption has occurred. A disruption challenges the organisational narrative and could, for example, comprise a question from a NED identifying areas of concern within a report.

4.3 Defensive and protective impression management techniques

We summarise what we mean by the eleven impression management techniques used in the questions and answers in Appendix 2. Table 4 summarises the frequency of nine of our eleven impression management techniques in public and in private. Preventive techniques include NEDs' use of ❶mild challenge questions, ^❹supportive and optimistic comments when questioning managers, ^❺questioning (one's) own understanding and ^❻combination questions. Corrective techniques include ^❼combination answers, ^❽collaborative answers, ^❾close down debate, ^❿agree with the questioner and acknowledge the challenge. We use quotations to illustrate these impression management techniques. While ^❷dramaturgical circumspection: Keeping close to the facts (Quotation 3) and ^❸exercise tact (Quotation 4) featured in our dataset, we only counted question-and-answer moments. We discuss each of the eleven impression management techniques in the sections to follow.

4.3.1 Mild challenge ^❶

The first public-private divergence we identify for interpretation is that most of the questions asked by NEDs in public comprise simple questioning offering mild challenge and these questions result in clear answers. Table 4 and Table 5 show that mild challenge, i.e. simple questioning with no follow-up, is more frequent in public. From an impression management perspective, NEDs asking mild challenge questions is a preventive technique deployed to prevent or avoid disruptions to team performance, which can result from a challenging question. The NEDs use preventive techniques, first, as a defensive technique to defend their own performances as they question managers, and second, as a protective technique to protect the performances of the managers they question. First, the NEDs defend their performance as questioners by asking relevant, yet straightforward questions, offering, for the most part, mild challenge which avoids disruptions. Thus, while they fulfil their expected role of challenging management by asking questions, they also use the defensive technique of dramaturgical loyalty to the overall performance by not “exploit[ing] their presence in the front region … to stage their own show” (Goffman, 1959, p. 214). They could “stage their own show”, for example, by asking hard, probing questions which could either cause embarrassment to the manager being questioned, or contradict the boards'/chairs' optimistic narrative. Instead, most of the challenging questions are asked in private, backstage. We find (see Table 5) that 40% of questions in private offer moderate challenge (i.e., constructive or coherent challenge), compared with 16% of those in public, reflecting NEDs' reluctance to constructively challenge in public.

NEDs also use mild challenge as a protective impression management technique whereby the questioner asks questions in a protective way to support the overall team performance. By asking straightforward questions, those “being inspected may put on an exemplary show” (Goffman, 1959, p. 232) and provide answers which easily satisfy the questioner. Each board member depends on the dramaturgical cooperation of the others/managers to foster “a given definition of the situation” (p. 83). The interactions indicate to observers that the NEDs provide challenge by asking questions, and the managers respond with clear answers, demonstrating transparency and thus governance-in-action. In Quotation 2, the manager responds to a mild challenge question with a clear answer comprising a simple explanation. These relatively simple, mildly challenging questions reinforce the impression that NEDs are questioning managers and managers are providing clear answers.

4.3.2 Dramaturgical circumspection: keeping close to the facts ^❷

By asking straightforward questions, the NEDs also exercise the defensive technique of dramaturgical circumspection by “keeping close to the facts“ (Goffman, 1959, p. 220), thus safeguarding their show (i.e. as questioners of the managers). However, this approach prevents them “from staging a very elaborate one” (p. 220). Of the questions asked in public, 84% (untabulated) offer mild or no challenge. Despite this, the audience can appreciate that questions are being asked, even if they are the types of question Spira (2002) identifies as anticipated or ceremonial, or for which a response may have been prepared in advance (Spira, 2002, p. 140).

Quotation 3 and Quotation 4, which occur following presentation of a board report, illustrate keeping close to the facts. Quotation 3 exemplifies a NED exercising the defensive technique of dramaturgical circumspection (Goffman, 1959, p. 220), by keeping close to the facts through asking a simple question with no follow up.

4.3.3 Exercise tact ^❸

Goffman (1959, p. 167) observes that “to ensure that communication will follow established, narrow channels, each team is prepared to assist the other team, tacitly and tactfully, in maintaining the impression it is attempting to foster. Goffman (1959, p. 14) further emphasises the role of “tact”, explaining that “few impressions could survive if those who received the impression did not exert tact in their reception of it”. The interaction in Quotation 4 illustrates a NED using, first, defensive impression management concerning his performance by keeping close to the facts and asking a simple question with no follow-up. He then receives a selective, partial response from the manager. However, he then uses the protective impression management technique of tact, by accepting the excuses provided to protect the manager's performance, and does not further probe for a more complete answer. Instead, the NED moves onto a new simple question addressed to the finance director. In our study, many straightforward questions offering mild challenge are linked to the financial reports, and the finance director generally provides clear factual answers.

From an impression management perspective, these interactions fall within the ambit of preventive techniques, whereby NEDs defend their performances as questioners and protect those of managers as responders by keeping close to the facts when asking questions. Goffman (1959, p. 209) also asserts that there are some facts that could discredit the situation being performed. They could be “negatively-valued characteristics that everyone can see but no one refers to.” Preventive impression management techniques are used to avoid these facts coming to the audience's attention.

4.3.4 Questions starting with supportive comments^❹

One of the clearest divergences between questions in public and private is the NEDs' use of supportive comments before asking managers more substantive questions. At board meetings in public, we observe the chairs using positive/optimistic rhetoric to set the “definition of the situation” (Goffman, 1959, p. 83). For example, the first agenda item on all boards comprises the “Chair and CEO update”, which is generally verbal, and used by chairs to provide a strong and optimistic start to the meeting, idealising (Goffman, 1959) the organisation's performance, and setting the scene for the rest of the performance. Quotation 5 is illustrative. The chair introduces the topic in positive terms at the start of the meeting, additionally using the dramaturgical opportunity to idealise (Goffman, 1959) her organisation, using optimistic rhetoric, reinforcing the organisation's image to idealise its activities.

Supportive comments are a protective impression management technique designed to soften a substantive question and protect the performance of the manager being questioned. They are also preventive, as the intent is to prevent embarrassment when asking a more challenging question. Quotation 6 demonstrates a supportive comment before asking a substantive question when a NED questions the content of a report. When asking a question that can disrupt the recipient's performance within the team, the questioner praises the content of the presentation which has the effect of moving the focus away from the question.

Although Quotation 7 is not a question, it illuminates the approach where positive aspects are emphasised before substantive comments are made. A NED suggests how the board might make its views known to an underperforming hospital. Here, the use of “blah, blah” indicates that the NED does not believe in the positive aspects that he is suggesting, further emphasising the use of such supportive comments to create a positive impression in public at variance with the NED's negative perception revealed in private.

4.3.5 Questioning (one's) own understanding ^❺

Questioning (one's) own understanding has a similar effect to supportive comments and is also more frequently used in public than in private. Combined with a more substantive question, it is also a preventive and protective impression management technique. Before asking the substantive question, this technique directs the focus from the question onto the questioner, and thus minimises the potentially negative effect of the question on the responder. As Goffman (1967) notes, “before engaging in a potentially offensive act, he [the questioner] may provide explanations as to why the others ought not to be affronted by it” (Goffman, 1967, p. 17). Quotation 8 illustrates questioning (one's) own understanding, directing the focus from the question onto the questioner as a preventive, protective technique “I got a bit lost reading the third bullet point about mental health, because it seems, if I have read it correctly” …. See also Quotation 1 (marked 1 in Quotation 1) “– if I have understood it correctly – so please correct me if I am wrong”.

4.3.6 Combination questions ^❻

Combination questions in public mostly begin with the questioner using a supportive comment, an optimistic comment or questioning (one's) own understanding, before asking a more substantive question. Quotation 9 illustrates an optimistic comment (marked 1 in Quotation 9) as the first part of a combination question comprising two questions (marked 2 and 3 in Quotation 9).

4.3.7 Combination answers^❼

We highlight a further public-private divergence whereby moderately challenging probing questions mostly result in combination answers. Goffman's (1959) dramaturgical perspective suggests that such combination answers are corrective impression management techniques used by those answering questions to prevent their performance from being disrupted by challenging questions. In Quotation 1 earlier, the plan for children's services is presented to Board A for approval and introduced as “very important for the whole borough.” In the questions and answers which follow, one NED asks a probing question (marked 2 in Quotation 1), using the protective impression management technique of questioning his own understanding to soften it (marked 1 in Quotation 1) before asking the probing question (marked 2 in Quotation 1). However, his probing is highly pertinent. He questions why this “very important” plan only addresses 35% of the children it sets out to serve. The expectation would be that the plan should explain what happens to the other 65% (marked 11 in Quotation 1). As a “single note off-key can disrupt the tone of an entire performance” (Goffman, 1959, p. 52), not including this critical point in the report should have been addressed in the manager's answer to the NED. Instead, the CEO provides a combination answer which deploys seven elements (marked 3–9 in Quotation 1) to respond to the NED's question. Goffman argues that interaction is like an “information game” (Goffman, 1959, p. 8) in which individuals selectively disclose, exaggerate, conceal and/or neglect specific information), which Quotation 1 exemplifies.

4.3.8 Collaborative answers^❽

A further public-private divergence that we highlight is that moderately challenging questions often result in collaborative answers, which generally prevent further questions. The collaborative answer is a corrective defensive impression management technique deployed to compensate for the possible disruption of the team performance resulting from a challenging question. When asking questions of the board, members of the public attending the public session employ moderate challenge. [4] This level of challenge often leads to a collaborative answer from board members, who together provide complementary answers. Quotation 10 illustrates collaborative answers. In Quotation 10, a member of the public “punch[es] holes in the management's story” (Abraham and Bamber, 2017, p. 22). This question relates to the questioner's participation in the design of a new model of care (marked 1 in Quotation 10). Several Board A participants (NED, manager and chair) co-operate in providing a collaborative answer to robustly deflect the challenge. Given the strength of the member of the public's challenge, the team uses the defensive practice of dramaturgical discipline (Goffman, 1959, p. 216), to manage the potential disruption of their performance resulting from this question. Dramaturgical discipline (Table 1) requires a performer to have the agility cover up quickly for mistakes made by others. In this case, assertions made in the presentation to the board contrasted with the questioner's personal experience.

The collaborative answer is a corrective impression management technique delivered through cross-team collaboration, as shown in Quotation 10 (NED, Manager, Chair). Punching holes in management's story (as the member of the public [4] in Quotation 10 tries to do: “all your fine words”) (marked 1 in Quotation 10) threatens the organisation's claim that patients' views are sought when designing models of care, which is a statutory requirement. This challenge mobilises a NED (marked 2 and 3 in Quotation 10), the relevant manager (marked 4 and 5 in Quotation 10), and the chair (marked 6 in Quotation 10). Together, they deploy a corrective and defensive impression management technique – the collaborative answer – to rebut the assertion jointly. They attempt to reduce the discrepancy between the self-proclaimed actions of the organisation and the reality (Boiral, 2016). At no point do they ask the questioner about his experience or directly respond to his comment. Furthermore, the member of the public's question (marked 1 in Quotation 10) and the answers (marked 2, 3, 4, 5 and 6 in Quotation 10) to his question were subsequently toned down in the minutes of the meeting.

These examples in quotations 1–10 address four of the five interactions in Section 4.1 that we identify as having the clearest public-private divergences. We demonstrate that these divergences result from the use of impression management techniques in public.

4.3.9 Close down debate^❾

The quotations earlier illustrate how interactions can be used to prevent or deflect challenge and ensure a harmonious performance. We address the fifth divergence, which relates to the dominance of CEOs backstage. In public, CEOs answer 40 questions (often participating in collaborative answers, which are corrective impression management techniques), while in private they answer 75 questions across all boards (Table 4). This divergence indicates that CEOs adopt a much higher profile away from public scrutiny. We highlight three further impression management techniques. These are ^❾close down debate, ^❿agree with the questioner and acknowledge the challenge.

Closing down debate is a further example of corrective and defensive impression management facilitated via the chair. The technique enables the chair, having identified unwelcome challenge, to exercise dramaturgical discipline (Table 1) and quickly avoid answering the question or engaging in further discussion. In Quotation 11, the CEO supports the chair (marked 3 in Quotation 11) and together they “cover up on the spur of the moment for inappropriate behaviour on the part of [their] team-mates” (Goffman, 1959, p. 216) (in this case, by a NED (marked 1 in Quotation 11)) by the chair and the CEO suggesting that the issue raised would be discussed at some other time (marked 2 and 4 in Quotation 11). By expressing concern and challenging the external context, including the long-term failure of the health service, the NED challenges the definition of reality promoted by the chair in public earlier in the meeting (marked 1 in Quotation 11). According to Goffman (1959), this challenge is a potential “performance disruption” (Goffman, 1959, p. 208/212) or an “inopportune intrusion”, which is a “source of embarrassment and dissonance” (p. 210) especially given that the meeting is broadcast onto the Internet. This interaction illustrates how chairs and CEOs can manage a threat to the performance, in this case, by referring to board process (marked 2 and 4 in Quotation 11). A corrective impression management technique is immediately deployed to rebut the NED's criticism of the NHS plan. First, the chair uses his position as primus inter pares (first amongst equals) in the boardroom to close down the debate (marked 2 in Quotation 11). Second, the CEO provides a combination response (marked 3 and 4 in Quotation 11), turning their joint response into a collaborative answer (marked 2, 3 and 4 in Quotation 11) to save this part of the performance from further questions. Furthermore, in the two board meetings which follow, discussion of the new NHS plan is minimal and only occurs in the private sessions. The board never has the “opportunity for us to have lots of discussions.” (marked 4 in Quotation 11).

In Quotation 12, using the board meeting agenda, the chair similarly closes down a challenge from the Healthwatch representative, which threatened the narrative that patients' views are included in the redesign of healthcare services, which is a statutory requirement. Healthwatch questioned the lack of clear links between engagement with patients and the health services that are then ostensibly designed in response to their feedback (marked 1 in Quotation 12). The chair quickly closes down debate, but at the same time makes a positive comment, “good point” (marked 2 in Quotation 12), to the Healthwatch representative. The chair uses the temporal constraints of the board agenda to manage the challenge (marked 3 in Quotation 12). Here, this enables him to move quickly from an agenda item that could threaten the performance to the next item (“any other business”), which he can control from the start.

4.3.10 Agree with the questioner^❿

^❿Agree with the questioner or acknowledge the challenge, can be used as a corrective and defensive impression management techniques by individuals providing the answer. It partly redirects the focus of the question onto the notion of challenge. Such answers can also be protective. By agreeing with the questioner, the managers indicate that the questioners are offering a good challenge, thus appearing to fulfil their governance role. It may also have the effect of discouraging further questions. Agree with the questioner is illustrated in Quotation 13 and in Quotation 1 (earlier) (“but yes, I agree with you”) (marked 9 in Quotation 1).

4.3.11 Acknowledge the challenge

When managers are asked moderately challenging questions, one frequent response is to acknowledge the challenge. This impression management technique is used nearly twice as often in public as in private (e.g. “That is a very good question …” (marked 3 in Quotation 1); “and I take your point” (marked 3 in Quotation 10)). We show that acknowledging the challenge is a routine response to a challenging question. Quotation 14 indicates a performative approach to the requirement to challenge. Following a presentation that the board receives positively, one NED indicates in public that he nevertheless needs to think of a challenging question to ask. This comment suggests that challenging or questioning is a necessary ritual, irrespective of the situation, and illustrates Spira's (1999, p. 248) “ceremonial aspect” of challenge.

Acknowledge the challenge can be used as a corrective and defensive impression management technique by the individual answering the question. We describe the use of acknowledge the challenge as a corrective impression management technique compensating for a difficult question being asked by refocusing the question onto the notion of the challenge itself. This impression management technique is defensive impression management, where the manager uses dramaturgical discipline to reduce the impact of the question. As Goffman (1959, p. 216) notes, the disciplined performer will be prepared and can discount (or deflect) a disruptive event, to remove its importance. Thus, by acknowledging the challenge, the manager promotes the notion that the challenge is a good one (marked 3 in Quotation 1), and the importance of the substantive question is reduced.

Sometimes a responder encourages the questioner to continue to challenge (see Quotation 15). This response emphasises that challenge is an expected part of the board's performance, and is overtly recognised in some of the board meetings in public.

Acknowledge the challenge is an impression management technique with several effects. First, the public audience can appreciate that the NEDs challenge the managers and that the managers respond to that challenge. Hence, challenge is performed and, from the perspective of the audience, “the satisfactory handling of searching questions … reflects well on all concerned, enhancing reputations …” (Spira, 1999, p. 248). The second effect of acknowledge the challenge is to refocus the interaction away from the question posed, and instead on to the notion of challenge itself, consequently distracting from what may be seen as a threat to the performance. Goffman (1967) offers an insight into this technique, suggesting that the recipient of the challenge can still attempt to maintain the fiction that no threat has occurred to his face or performance because of the question, by acting as if no event has occurred at all. Alternatively, the recipient acknowledges that an event has occurred, but shows that it does not threaten the performance (Goffman, 1967, p. 17). Futrell (1999, p. 499) describes the effect of this type of interaction as “attempting to maintain the appearance of a cooperative and participatory atmosphere” adding that it requires “interactional agility.”

The third impression management effect, linked to the second, is that acknowledge the challenge can assist the recipient in avoiding answering the question (see Quotation 15 and 16). Quotation 15 illustrates acknowledge the challenge, where the Healthwatch representative asks a mild challenge question about the long waiting times for hospital appointments (marked 1 in Quotation 15). Such challenge has the potential to discredit the narrative of competence and improvement the board portrays. As Goffman (1959, p. 51) suggests, an audience has a “tendency to pounce on trifling flaws as a sign that the whole show is false”. Therefore, the CEO uses the notion of challenge to avoid answering the question (marked 2 in Quotation 15). By acknowledging the challenge, the CEO re-focuses the discussion on the need for challenge, distancing the organisation from the question about waiting times without offering a response apart from a self-serving platitude (marked 3 in Quotation 15).

Similarly, in Quotation 16, the manager uses acknowledge the challenge to avoid giving a clear answer to a NED's searching question.

In Quotation 17, a NED critically questions the information in a proposal presented to the board. This acknowledgment of the requirement to challenge management (marked 1 in Quotation 17), coupled with recognition that it has taken place (marked 3 in Quotation 17), shows one way in which a board can “play out” (Abraham and Bamber, 2017, p. 22) challenge for the audience.

5. Conclusion

We demonstrate that the boards exhibit the key characteristics of Goffman's dramaturgical framework. The presence of an audience of stakeholders leads board members (NEDs and managers) to employ a range of preventive and corrective impression management techniques, during their interactions. The boards need to demonstrate to their audiences that “matters are what they appear to be” (Goffman, 1959, p. 17), even if they are not. These techniques enable board members and managers to manage the trade-off between demonstrating the good governance they are expected to implement, while protecting information that could lead an audience to question both the organisation's governance and competence.

Adopting Goffman's impression-management lens in a different context to prior research has enabled us to develop insights into impression management techniques unique to a boardroom context. We show how these impression management techniques are used to create the appearance that non-executive directors are challenging management. However, the challenge of management in public is not so robust that it risks disrupting the show, in Goffman (1959) terms. Provoked by transparency regulations, boards' use of impression management compromises accountability, operationalised as question-and-answer interactions in public board meetings. Pernelet and Brennan (2023b) extend this Goffman analysis to consider how governance is performed frontstage and backstage, illuminating the performativity of governance.

Six times per year, the boards hold their meetings in public to demonstrate governance and transparency, while simultaneously working together to ensure that they do not disclose issues that could tarnish the organisational narrative that they are improving local healthcare services. Impression management is the tool of choice to manage this complexity. We opened our paper with the question: Does transparency improve accountability? We find UK NHS transparency regulations (“sunshine” laws) undermine accountability, which findings are consistent with prior research (e.g., Roberts, 2009; Quattrone, 2022). Conversely, Roberts (2001) commends closed private face-to-face unitary board meetings as creating a dynamic of openness and engagement leading to a dialogue conducive to accountability. Mandating transparency compromises open dialogue and engagement. However, in the context of contracts with employees and contractors, Fahn and Zanarone (2022) conclude that transparency creates accountability. They encourage more research on when and how transparency benefits organisations.

We call for more evidence-based regulation. The Regulator assumed that requiring boards to meet in public would improve transparency. Rather, we find it changed boardroom behaviour, leading to less not more transparency. Similarly, Roberts (2018) highlights the different ways individuals and groups such as boards of directors respond to external (capital markets') pressures created by transparency. We argue that the boards we study respond to the external pressure from having to meet in public by keeping up appearances. Nicholson and Cook (2009, p. 312) develop a model to explain the behaviour of boards of directors. They challenge the assumption that increased transparency reduces director self-interest. They describe the assumption that increased transparency leading to better boardroom behaviour is “old ways of thinking.” Similarly, Janning et al. (2020) characterise transparency in a board-of-directors’ context as an illusion, a false solution intended to restore trust and higher standards of moral behaviour. We concur with their observations. Thus we conclude that transparency may impede accountability.

Our findings are based on the three public-sector boards in our study. It is not our intention to provide generalisable findings. Rather, we believe our unique public/private meetings' context provides an opportunity to reveal changes in boardroom behaviour in the two contrasting settings. Our findings may be applicable to contexts in which more typical boards meet in public in front of an audience, such as AGMs. Prior research on such meetings does not as yet study directors' behaviour at those meetings. Detailed records of company AGMs are available nowadays because, increasingly, AGMs are recorded. This permits researchers to analyse the words spoken at AGMs and also their choreography, performance and body language of the participants, including the directors. The behaviour of the parties participating in AGMs and their interactions deserve more attention, not only directors, and shareholders, but also auditors (e.g. Johed and Catasús, 2018), the company secretary, public relations advisors, etc. Much board activity occurs outside the boardroom, but this activity is not observed in our study.

Our novel methodology needs to be tested in other board contexts and in other industry sectors. Our paper focusses on directors' and managers' question-and-answer patterns. Our unique dataset of video-recordings provides future opportunity to analyse not just the words used during board meetings but also non-verbal behaviours in boardrooms. A question can be interpreted in many different ways, depending on how it is being asked (e.g. tone, facial expression, openness of body and so forth). Non-verbal behaviours play a role in dramaturgical performances. Goffman (1967) would be a useful starting point to extend our analysis to non-verbal behaviours, while Choudhury et al. (2019) provide a potential methodology.

Table 1

Protective and corrective impression management techniques

Impression management technique	Preventive “Avoid embarrassment”*	Corrective “Compensate for incidents not avoided”**
Defensive techniques	Protect own performance • Dramaturgical loyalty • Dramaturgical discipline • Dramaturgical circumspection: “keeping close to the facts”***
Protective techniques	Protect others' performances • Exercise tact • Ignore embarrassing moments • Accept excuses provided

Source(s): Goffman (1959, *p. 13, **p. 212, ***p. 220)

Table 2

Board meeting data: length of meetings, questions asked, answers given, transcript words and the dataset

	Board A			Board B			Board C			Total
	January 2019	March 2019	May 2019	January 2019	March 2019	May 2019	September 2018	March 2019	May 2019
	No. Minutes
Length of public session	125	140	120	103	83	123	134	112	101	1,041
Length of private session	53	62	54	48	76	54	16	19	27	409
Total length of each board meeting	178	202	174	151	159	177	150	131	128	1,450
	No. of Questions asked
Questions asked in public	33	27	25	27	29	46	27	25	27	266
Questions asked in private	25	23	32	12	28	11	8	9	4	152
Total questions asked per board meeting	58	50	57	39	57	57	35	34	31	418
Questions per hour in public	16	12	13	16	21	22	12	13	16
Questions per hour in private	28	22	35	15	22	12	30	28	9
Average questions per hour in public: all boards										16
Average questions per hour in private: all boards										22
	No. of Answers provided
Answers in public	54	35	34	24	43	48	25	27	27	317
Answers in private	25	30	37	18	47	19	7	7	3	193
Total answers per board meeting	79	65	71	42	90	67	32	34	30	510
	No. of Words in datasets
Total dataset per board meeting	13,750	17,911	11,943	12,492	17,965	17,197	5,625	3,499	7,007
Total dataset by board			43,604			47,654			16,131
Total dataset										107,389

Quotation 1

Contrasting behaviour in public and in private

		Challenge frontstage
NED 1	1	My question is probably two-fold; one is – if I have understood it correctly – so please correct me if I am wrong
	2	we have set ourselves a target of 35% of the children in this catchment that we will provide a service to, and we are running at about 31% so we are a bit short, trying to catch up. Why such a low target? What is the primary restraint? Why isn't it 100% of the children?
Manager	3	That is a very good question …
	4	… The target was set by NHS England … and it is something in the long-term plan that they are beginning to extend …
	5	… And it does seem low, doesn't it, because you think, why aren't we looking after all the children who might have a need? …
	6	… Some children who may be having early signs of emotional distress … needs are being met by school pastoral services [or] they might be supported by … youth services
	7	… There is a whole range of other services out there and about 35% is children with a diagnosed health illness
	8	… So, I think we need to think broader going forward
	9	but yes, I agree with you it does look quite low when you look at it in those terms
		(Board A, public)
		Continuation of challenge backstage
Chair	10	Any items that people want to talk about from Part 1?
NED 1	11	We had a presentation on children's mental health. First of all, I was really disappointed with the 35%. I understand we don't know if 65% of the kids are being serviced elsewhere, but it seemed – especially for public consumption – a very unambitious target. And then we're talking about a £6 m spend on [other initiatives], and yet we're talking about – the rest of it is peanuts, an investment in the hundreds of thousands … If it's a priority for us, what sort of investment should we be putting in? Is that the constraint? Is it that the target should really be 80%, and we just need more resources and more money, or are we just living within our means? … I'd just like to make sure I understand
CEO	12	I think we should … I mean … eh … [the presenter] tried to answer the question and I agree the answer was problematic. I'll try to answer it. 35% doesn't represent everybody … it's not named people, these are just, kind of, statistical norms and that kind of stuff. So, it's a kind of rough benchmark and the 35% target has been set nationally. You [need to] invest in the 65% so they don't become the 35% … I think we need to explain that better, I think you're right, the answer wasn't good enough
NED 2	13	I wanted to pick up about quick access … that did not come across in the presentation at all. It goes back to the 35% and how we can show the pathway and really target that early intervention
NED 3	14	So that's exactly it, that's the key thing for me: how quickly do people get seen and how quickly do they get treated
Chair	15	There is something about the system not recognising the severity of the changes and … what we are seeing. The question is how do we prioritise more and have a bit more scale of ambition within the system we have got? … So we can take that away, and bring it to the clinical leads … and feed in some of this with the NHS plan
		(Board A, private)

Table 3

Divergences in types of questions and answers in public and private

	Frequency in public		Frequency in private
	No.	%	No.	%
Mild challenge ①	159	60	78	53
Moderate challenge ①	43	16	63	40
Supportive comment ②	29	12	0	0
Optimistic comment ②	11	5	0	0
Combination questions ③	81	37	23	26
Combination answers ④	63	27	27	20
Questions answered by the chief executive ⑤	40	17	75	51

Note(s): Key: Five divergences (① to ⑤) between the types of questions and answers in public and private extracted from Pernelet and Brennan's (2023a) analysis

% = Percentage of total questions/answers in public/private

Table 4

Frequency of impression management techniques

Techniques	In public		In private		Total
	No.	% of total	No.	% of total	No.	%
Preventive techniques
❶ Mild challenge (Quotation 2)	150	68%	70	32%	220	100%
❹ Supportive and optimistic comments (Quotation 5, 6, 7)	40	100%	0	0%	40	100%
❺ Questioning (one's) own understanding (Quotation 1, 8)	9	69%	4	31%	13	100%
❻ Combination questions (Quotation 9)	81	78%	23	22%	104	100%
Corrective/defensive techniques
❼ Combination answers (Quotation 1)	63	70%	27	30%	90	100%
❽ Collaborative answers (Quotation 10)	34	71%	14	29%	48	100%
❾ Close down debate (Quotation 11, 12)	3	100%	0	0%	3	100%
❿ Agree with the questioner (Quotation 1, 13)	17	81%	4	19%	21	100%
Acknowledge the challenge (Quotation 1, 10, 15, 16, 17)	10	67%	5	33%	15	100%

Quotation 2

Mild challenge

NED	I wanted to pick up on the Healthwatch report … and I am wondering what the plan was moving forward to take on board those findings and make some changes?
Manager	So, it is feeding into the GP leadership group and an action plan on the back of that is being developed
	(Board A, public)

Quotation 3

Keeping close to the facts: defensive impression management to defend the NED's role as questioner

NED	In terms of the governance [diagram], where would the conversation around activity growth and risk happen … ?
FD	So, it will happen in finance committee
	(Board A, public)

Quotation 4

Exercising tact: protective impression management to protect the manager questioned

NED	On the basis of what we know at the moment on the finances, the commissioning intentions we talked about earlier – are they deliverable in terms of affordability?
Manager	Yes, at the moment we have been able to deliver the commissioning intentions, yes
NED	Where are we on QIPP [quality improvement programme]?
FD	We are around about £1 m. … So we have made good progress …
	(Board B, public)

Quotation 5

Chair's supportive and optimistic comments

Chair

So I thought I had better start off by talking about the publication this week of the NHS long-term plan … It focuses on lots of the clinical areas that we talk about all the time here at our board meetings so I think that is really important …. We will be offering a more coordinated, more proactive and more personalised approach to care … many things we are [already] working on and doing. More on the quality of care which is at the heart of everything that we do here. Focus on supporting our workforce which is really important … About digitally enabled care, we are doing an awful lot of work on that so that is really exciting … I for one feel at the beginning of a new year, really encouraged and excited by the potential here to improve care for our patients …

(Board B, public)

Quotation 6

Supportive comment is protective impression management

NED	… it was a really interesting report and really thorough and really well laid out to make it accessible, thank you … I was wondering about the risks in terms of equalities and the potential impact on potential carers and what we are doing with the local authority in terms of supporting carers at home … ?
	(Board B, public)

Quotation 7

Supportive comments before challenge

NED	My thought is the tone of voice of all this can be quite tricky to get right. I do think that we should recognise that they have put a lot of work in blah, blah, they have made progress, we welcome the fact that they have got as far as they have; the positive stuff, not just beating them up. But I think the message has to be it is still pretty high risk, an awful lot of things need fixing
	(Board B, private)

Quotation 8

Questioning (one's) own understanding is protective impression management

NED

Could I just ask a quick question? On the QIIP [quality improvement programme], page 160, I got a bit lost reading the third bullet point about mental health, because it seems, if I have read it correctly, I am just not sure if I have, it seems to imply that we have achieved the QIIP target. Have I understood that correctly, that we have missed the opportunity to improve the quality but we have saved the money?

(Board A, public)

Quotation 9

Combination questions beginning with optimistic comment

NED	1	I very much welcome the new model of [care], I think it is a big step forward …
	2	… am I right that there are some thresholds and triggers … that if volumes exceed certain levels then the question of cost will reappear on the agenda … ?
	3	… And if so, do we have appropriate monitoring in place … how are we going to monitor that position?
		(Board B, public)

Quotation 10

Collaborative answers

Public member	1	I apologise, but I keep looking at that presentation which was presented to us earlier. Since I was one of the initial patient experience representatives … I have to tell you that the amount of patient involvement and patient experience that has been taken note of, [and] that has been involved in the processes displayed here earlier has been negligible. And all of your fine words that I hear and the intentions I hear, the practice has been deeply lacking
NED	2	Thank you for your comments. I think we have been trying harder. I will just focus on the last year where we have done a number of workshops … here with my colleagues and the engagement team and that was very successful in terms of looking at our … health pathway. So I think we are trying, and I think that model worked very well. And we are about to have another focus group where those people came along, in June …
	3	… And I take your point about how you engage and sometimes it is very difficult … But if I take the example of the mental health workshop that we have been doing in the community … I think we are on the right track, and replicating that sort of engagement in getting the views of people on the ground would be something that we could work towards
Manager	4	I think it is also important to recognise that engagement needs to change over time as we go through different processes … But as we have then moved into the assurance process, we have also had patients on the assurance panels as well as part of the process that we have gone through, so they have been on all of the assurance panels that were presented in the slide …
	5	… I think the reason why you may not have seen as many contract awards is because we are going through a piloting stage in some of these pathways before we then make a final decision about contracts award … So, I think we need to be refining engagement to make sure that it is able to influence in a way that is most useful I think as well
Chair	6	I think — keep challenging, but there have also been a number of things that have happened that you may not have been involved with as well … The diabetes meeting was in this room … and I have never seen so many people with interest in diabetes ever in terms of shaping what it would look like going forward. So I think to say that we are never engaged is probably not true. We can always do more. We can always do more but it is not because we are not trying, I think is the answer
		(Board A, public)

Quotation 11

The chair and CEO close down debate

NED	1	I agree with everything you said, but I have two concerns. For me, the absence of any evidence of sensible long-term planning on workforce seems to me to be a major concern … a lot of aspiration in this plan, and it will be critical on having the right number of key skilled people. That has been a long-term failure of the health service to get to grips with workforce planning. The other issue is that the government continues to delay its statement about relationships between social care and the health service. Again a key thing … where integration is critically dependent on getting that interface right
Chair	2	Thank you, very helpful, we are going to have lots of conversations about this
CEO	3	I just wanted to say that it did only come out all in one day, and just remember this is a high-level plan at the moment. Actually, there is lots of detail in it …
	4	So we will make sure there is an opportunity for us to have lots of discussions … over the next few meetings
		(Board B, public)

Quotation 12

The chair uses the board agenda to close down debate

Healthwatch representative:	1	Just one small comment. Maybe there is an element in trying to communicate where the patient experience has driven a change … I think it would be great if we could see presentations where it says our patients said this, and clinicians are interpreting this and delivering this. I think that … there are good examples of work, but if it is not communicated … then you can't see the link, and then you can see also where the gaps are and where more engagement needs to be done
Chair	2	Okay, good point
	3	If there aren't any other questions, any other business [next agenda item]?
		(Board A, public)

Quotation 13

Agree with the questioner

CEO	So I think you are right … where there are things we need to respond to as a system, [this] will be absolutely critical
	(Board B, public)

Quotation 14

The onus on the NED to challenge

NED	I thought it was fantastic as well and I was going to pass over in silence, but everyone here is so very positive I was just trying to figure out a question that would be challenging not congratulating …
	(Board B, public)

Quotation 15

Using the notion of challenge to avoid answering a question

Healthwatch Representative	1	We are getting feedback that [patients are] waiting a while, maybe there is a communications exercise in explaining to patients how long they have got to expect … Maybe there is a gap in the perception of the patient in understanding how long it might take to get seen …
CEO	2	So that is the stuff that makes a difference … I would suggest you keep challenging us with that … So I absolutely think that as we go forward you keep putting that challenge to us …
	3	So I think keep challenging us, and certainly … since I have been here this [organisation] has always been trying to do that better than it has in the past …
		(Board A, public)

Quotation 16

Using the notion of challenge to avoid answering a question

NED	What I am interested in is how will the patients experience integrated care? If it is across the country within two years, how is a member of the public going to see a difference? What is the plan that is different in their eyes?
Manager	I think that is a very interesting challenge and I think it will be different … in each local system …
	(Board B, public)

Quotation 17

Acknowledge the challenge

NED	1	I just want to pick up on the measurement issue and [make] a few comments
	2	I think it would be a lovely thing to reduce childhood obesity but [you should] make it clear that it is not going to happen overnight. I thought it was pretty ambitious for two years, but anyway – put some goals in there – but also state that it is not going to happen in this timeframe on a major scale
Manager	3	I think the timescale point is a really good point
	4	Whilst it is a two-year plan it isn't meant to try and say that all of these things are going to be done and dusted in two years' time
		(Board B, public)

Table A1

Impression management techniques in a board-of-directors’ context

Impression management technique (Includes Goffman's (1959) description (in quotes) where relevant)	Preventive “Avoid … embarrassments”* Avoid disruption of the performance or narrative	Corrective “Compensate for occurrences [incidents] not … avoided”** Prevent disruption of the performance or narrative as a result of a question or comment
Defensive techniques Protect/defend own performance Avoid facts being drawn to the attention of the audience	Protect own performance ❶Mild challenge: demonstrates that NEDs are asking questions and thus providing challenge, while exercising ❷“dramaturgical circumspection” by “keeping close to the facts” *** (e.g. simple question, no follow up) NEDs exercise “dramaturgical loyalty” to the performance by not “staging their own show”, e.g. by not asking tough questions. They proficiently apply the skill of questioning	Protect own performance ❼Combination answers provide a sequence of answers to prevent the performance being disrupted by challenging questions, and to discourage further questions ❽Collaborative answers provide complementary answers from different individuals to prevent the performance being disrupted by challenging questions, and to discourage further questions ❾Close down debate enables management to avoid answering the question or engage in further discussion ❿Agree with the questioner discourages further questions Acknowledge the challenge reduces the impact of the question, by redirecting the focus of the question onto the notion of challenge. Also demonstrates that the recipient has been challenged and is responding
Protective techniques Protect the performance of others (in the performance team)	Preventive Protect others' performances NEDs use ❶mild challenge: simple questions with no follow up, which allows the recipient to provide a satisfactory answer. Use ❸“tact” to accept an answer and not ask follow up/more challenging questions By NEDs using ❹supportive or optimistic comments to start a substantive question lessens the impact of the question on the recipient ❺Question own understanding as part of a combination question minimises the negative effect of the question on the recipient, by directing the focus of the question onto the questioner ❻NEDs use combination questions to precede a substantive question with a supportive and optimistic comment to protect the performance of the manager being questioned	Corrective Protect others' performances Supportive and optimistic comments as part of combination questions move the focus away from the question and minimise its effect on the manager being questioned Using ❹supportive or optimistic comments to start a substantive question moves the focus away from the question and protects the performance of the manager being questioned ❺Question own understanding as part of a combination question directs the focus away from the question onto the questioner. This minimises the negative effect of the question on the manager ❿Agree with the questioner indicates questioner is offering good challenge, and protects the questioner's performance Acknowledge the challenge indicates that the questioner is offering good challenge and performing governance

Source(s): *Goffman (1959, p. 13); **Goffman (1959, p. 213); ***Goffman (1959, p. 220)

^❶- These are the eleven impression management techniques identified in the study unique to a board-of-directors’ context

Table A2

Impression management techniques

Techniques	Objective
Preventive techniques
❶ Mild challenge	Avoid disruptions, yet maintain the appearance of challenge
❷ Dramaturgical circumspection: Keeping close to the facts	Safeguard the show
❸ Exercise tact	Ignore embarrassing moments; Accept excuses provided
❹ Supportive and optimistic comments	Soften a more substantive question
❺ Questioning (one's) own understanding	Direct the focus from the question onto the questioner
❻ Combination questions	Soften the impact of a more substantive questions that follows
Corrective/defensive techniques
❼ Combination answers	Prevent performance being disrupted by probing questions
❽ Collaborative answers	Prevent the performance being disrupted by challenging questions, and discourage further questions
❾ Close down debate	Prevent disruption of performance
❿ Agree with the questioner	Discourage further questions and indicate questioner is offering good challenge, thus protecting the questioner's performance
Acknowledge the challenge	Dramatic realisation of expectations of challenge

Notes

1. Goffman uses male pronouns throughout his work, which we continue when discussing his work.

2. We assume the audience is external stakeholders attending the meetings held in public. Some might argue that a performance continues backstage, with the non-executive directors and managers performing in front of each other.

3. We acknowledge that the first author having previously met people associated with Board A and Board B could be seen to compromise researcher independence, but it is hard to imagine a board providing researcher access without some personal contact. See Leblanc and Schwartz's (2007) observations on the difficulty of obtaining access to boards of directors.

4. In this research, we only analyse directors' and managers' question and answers. Thus, we only analyse responses to stakeholder interventions, not their interventions per se.