The interplay of board and audit committee quality governance in shaping disclosure quality in the UK market Open Access

This paper examines the effect of board and audit committee quality governance effectiveness on financial disclosure quality in the less regulated “comply or explain” environment of the UK^[1]. In the literature, albeit predominantly from an institutional context where corporate governance is heavily regulated (Beasley et al., 2009; Brown et al., 2011), the oversight of financial reporting is documented as an important role of the board of directors and of audit committees, which is duly recognised in governance codes internationally. The UK Corporate Governance Code (2018) stipulates that the board of directors should present a balanced and understandable assessment of the company’s position and prospects ^[2]. Similarly, audit committees are expected to monitor the integrity of the financial statements of the company and any formal announcements relating to the company’s financial performance and to review significant financial reporting judgments contained in them (FRC, 2018). Stakeholders also expect that both the board and audit committee can play their roles in overseeing the financial report consistent with financial report standards and requirements of related regulations, leading to the quality of the reports for making decisions of related stakeholders. The market supposes that when the board and audit committee contain quality characteristics, they must execute their roles with quality, eventually resulting in better-quality financial reports that make the stakeholders rely on those with confidence.

Although prior research on board governance and financial reporting is extensive, in this paper we focus on whether board and audit committees with expected quality contribute to the quality of financial disclosure. In addition, we check whether the effect of board and audit committee effectiveness on disclosure quality is moderated by the firm’s size or not. Moreover, to test complementary or substitutive relationships, we investigate the interaction effect between the audit committee and board effectiveness on disclosure quality and then re-check the moderating role of firm size. These issues have not been adequately investigated in the extant literature in a comprehensive manner using a composite index for board and audit committee quality effectiveness, on the one hand, and proxies for disclosure quality, on the other hand. Our study addresses these research gaps in the context of the UK’s “comply or explain” based governance structure, as well as offering rigorous methodological advancement. Such comprehensive investigations with a new methodological approach are very limited in the UK governance and disclosure quality research arena, thus making our study unique in offering new findings in a complex and endogenously related governance environment.

Our paper makes a number of important contributions to the literature. Firstly, focusing on the effect of monitoring on financial disclosure, we rely on two measures of disclosure quality, i.e. forward-looking disclosure score and receipt of investor relations award. Existing literature has not examined audit committee and board of directors’ effects on disclosure quality using forward-looking narratives and/or the receipt of Investor Relations Award by companies as proxies. While there is no disclosure quality proxy perfectly free from bias in the literature, our motivation for focusing on forward-looking disclosures and the Investor Relations Award derives from limiting the information asymmetry gap through reliance on analyst forecasts that are either directly or indirectly linked to our selected disclosure quality proxies. These proxies, unlike others used in the literature, are viewed as sophisticated and effective in disseminating the company’s financial information disclosure. Secondly, we examine using a composite measure of effectiveness whether audit committees that meet or exceed the regulatory norm (in our case, the recommendations of the Smith Report (2003) on audit committees – i.e. composed of independent directors, meet at least three times a year, composed of at least three members and have at least one member with financial expertise) are associated with financial reporting quality.

Thirdly, our paper provides indirect evidence on the joint effects of the audit committee and board of directors’ qualities on disclosure quality, in the sense that in examining the effect of audit committee effectiveness, we control for the board of directors’ characteristics. This is important particularly given that the literature recognises that audit committees and board of directors may be viewed as either substitute or complementary control mechanisms in the governance process (Brown et al., 2011; Katmon and Farooque, 2019, 2020). Fourthly, unlike prior literature, we reflect on potential moderating factors (i.e. firm size), the double interaction term effect of board and audit committee effectiveness, and lastly, the triple interaction term effect of board effectiveness, audit committee effectiveness and firm size on disclosure quality proxies. Lastly, our paper adds to the global debate about the effectiveness of corporate governance by providing evidence from a “comply or explain” setting and complements the literature based on mandatory settings. Given both board of directors and audit committees have oversight responsibility for financial reporting, examination of their monitoring effects on disclosure quality is a particularly important contribution to existing literature. In sum, our paper makes incremental contributions to accounting and organisational changes literature through addressing the limitations regarding the use of diverse disclosure quality proxies, composite measures of board and audit committee effectiveness and their interaction term effect (double), differential size effect of sample firms and their moderating effect/interaction term effect (triple), as well as robust methodological approaches. Our findings suggest some country-specific regulatory adjustments in corporate governance to accommodate the changing needs of listed companies, investors and other associated stakeholders at large in the post-crisis period.

The remainder of this paper is structured as follows. In Section 2, we provide a brief overview of the relevant corporate governance and disclosure literature and then develop our hypotheses for the relationship between board and audit committee effectiveness and disclosure quality. Section 3 outlines the research method, and the model used to test the hypotheses. In Section 4, we report the findings of our empirical tests. Lastly, Section 5 provides a brief summary of our key findings and concludes the paper.

Following major financial reporting scandals – which to a large extent have been attributed to poor governance oversights – new rules have been implemented in many countries, especially in the USA and the UK, to improve the quality of corporate governance (Beasley et al., 2009; FRC, 2018) ^[3]. However, significant variation in governance requirements and overall approach to governance exists between countries and firms (Doidge et al., 2007) – the USA, for instance, relies extensively on legislative and/or mandatory enforcement (e.g. the Sarbanes–Oxley Act and (associated) SEC pronouncements), while others such as the UK and Australia adopt a voluntary “comply or explain” approach. Again, extant literature argues that corporate governance can lead to an effective mechanism for reducing managers’ propensity to manipulate earnings (Assenso-Okofo et al., 2021; Gerged et al., 2023). Corporate governance mechanisms will strengthen a company’s internal control and provide increased monitoring for a firm to reduce opportunistic behaviour and information asymmetry (Aben et al., 2021; Akmal et al., 2022; Laksmana, 2008).

Agency theory views an audit committee and boards of directors as proficient monitoring agents in the company and a dexterous instrument in improving the quality of financial reporting (Kaawaase et al., 2021; Herli et al., 2021; Katmon and Farooque, 2017). The UK Corporate Governance Code stipulates that audit committees with appropriate quality should at least comprise of three members, meet at least three times a year and that all members must be independent, with at least one member having recent and relevant financial experience. As such, we expect that the effective audit committee will help to reduce agency problems by conveying credible information to users. The UK Corporate Governance Code also requires firms to have a balanced number of independent and non-independent directors on the board; therefore, the board’s decision-making is presumed to be substantially free from conflict of interest ^[4]. In this regard, we expect independent directors will behave like “professional referees”, whereby they will make independent decisions without compulsion from executive directors. The above points are consistent with the prediction of increased disclosure because of the audit committee and board of directors’ increased monitoring of financial disclosure.

Significantly, the UK Corporate Governance Code, updated in 2024, has been extended to enhance the roles and responsibilities of the board of directors and the audit committee. This extension, which goes beyond the “comply or explain” regime, promotes outcome-based governance. This shift is designed to bolster the effectiveness of the board’s and audit committee’s oversight mechanisms, fostering increased stakeholder trust and a more optimistic outlook for the future. Outcome-based governance is a form of corporate governance reporting that presents the approach to governance taken by the board and related committees within firms and demonstrates the impact of governance decisions on strategy, objectives and the long-term viability of the firms (FRC, 2024; Rees and Briône, 2024; Turner, 2025). Under the updated UK Corporate Governance Code, the principles have been revised to make the board responsible for establishing and maintaining its effectiveness. This includes a robust annual review of effectiveness, a process that provides a comprehensive verification of the monitoring and review functions. This review covers all material controls, including financial, operational, reporting and compliance controls, offering a thorough and reassuring assessment of the governance outcomes (FRC, 2023, 2024; Turner, 2025). As a result, firms are not only encouraged but also empowered to report the results of their good governance practices. This reporting, when accompanied by high-quality explanations, aligns with the comply and explain regime, fostering a sense of active contribution to the governance process.

Research on disclosure quality and corporate governance in the UK is very limited and tends to focus on particular governance mechanism and/or on a specific aspect of corporate disclosure (Li et al., 2008). A number of US studies have examined the relationship between corporate governance and disclosure quality. Baek et al. (2009) find that outside directors and institutional ownership are positively related to disclosure. Laksmana (2008) documents that independent boards provide more details about compensation practices, and then compensation disclosure reduces information asymmetry (measured by bid-ask spread and return volatility). In Europe, Cerbioni and Parbonetti (2007) found that board-related variables strongly influence the quantity of information disclosed. Bozzolan et al. (2009) examine the effect of forward-looking disclosures on analysts’ forecast properties for a sample of European companies cross-listed on NYSE in 2002. In Australian studies, Kent and Stewart (2008) report that board meetings, audit committee meetings and auditor types are positively related to disclosure. Lim et al. (2007) report a positive relationship between board independence and voluntary disclosure. A number of other studies have focused on governance and disclosure in Asian countries (Ghazali and Weetman, 2006), similar to the European and African studies (Abdelmoneim and El-Deeb, 2024; Cerbioni and Parbonetti, 2007; Bozzolan et al., 2009; El-Deeb et al., 2024). However, the studies on governance and disclosure based on Asian setting also lack of investigate the effect of the audit committee and board monitoring.

To summarise, the large body of existing international evidence suggests a positive link between disclosure quality and audit committee quality or board/governance quality. However, some prior studies also provide evidence of either a negative or no association between them. In this regard, several limitations can be enumerated for conflicting findings in the literature, such as focus on a particular aspect of disclosure, a particular aspect of governance, such as board characteristics, use of diverse disclosure quality proxies, differences in sample countries/industries/companies/years, methodological approaches, etc. Again, some evidence tends to relate to country-specific regulatory developments in corporate governance that have taken place in many countries in the aftermath of global financial crisis. Our study addresses some of these limitations and aims to extend the existing literature.

The audit committee offers manpower, efforts and contribution to improving the firm’s disclosure quality. The Smith Report (2003) recommendation is that the audit committee must at least comprise three members. Audit committee members who are independent of management are expected to demand high financial reporting quality. When audit committees are composed of independent non-executive directors, they will be able to exercise more power over management in demanding higher disclosure quality. Extant literature documented a positive relationship between audit committee size and independence with quality of financial reporting (Mardessi, 2022; O’Sullivan et al., 2008; Raimo et al., 2021). An audit committee must be active. Meeting frequency can be a signal of audit committee diligence and has been associated with higher-quality financial reporting (Beasley et al., 2009). The expertise of the audit committee can function as a mechanism to reduce any irregularities and financial misstatements in financial reporting (Kent et al., 2010; Mardessi, 2021). Katmon and Farooque (2020) reveal a significantly positive influence of audit committee member expertise on disclosure quality.

Drawing upon the recommendations in the Smith Report (2003), we argue that for audit committees to be effective, they must at least exhibit four characteristics. Firstly, the audit committee must comprise of at least three members. Secondly, all members of the audit committee must be independent non-executive directors. Thirdly, at least one membership of the committee must have relevant financial expertise^[5]. Fourthly, the audit committee must meet at least three times a year or meet at least four times annually, as suggested by the Blue-Ribbon Recommendation (BRC, 1999; Smith Report, 2003). These qualitative characteristics of the audit committee were recommended for all UK-listed companies, irrespective of their size and industry differences, reflecting independence, diligence and expertise.

As per the UK Corporate Governance Code, the board should establish an audit committee of independent non-executive directors, with a minimum membership of three, or two in the case of smaller companies. The committee, as a whole, shall have competence relevant to the sector in which the company operates (FRC, 2018, 2023). The audit committee’s primary role is to maintain the company’s financial integrity through effective monitoring and review, which is equally applicable to all UK-listed companies, irrespective of their size and industry differences. With an outcome-based perspective, in demonstrating audit committee effectiveness, the audit committee practices and results covering all aspects of control should be presented, along with an explanation and disclosure of information to the related stakeholders.

In examining the influence of the audit committee on financial reporting, we construct a variable called “audit committee effectiveness”, i.e. audit committee comprised of at least three or four members based on related recommendations, all the audit committee members are independent non-executive directors, at least one member of the audit committee has financial expertise and the audit committee meets at least three times a year^[6]. As such, we expect audit committee effectiveness is to be associated with higher disclosure quality. Our first hypothesis is thus:

The board of directors is the main force for monitoring and control over firms. Prior literature, such as Cornett et al. (2009), Baek et al. (2009) and regulators including the UK Corporate Governance Code (2018); Sarbanes–Oxley Act (2002) recognise the need for independent directors on boards as watchdogs to control opportunistic behaviour by managers and ensure that board decisions are always aligned with shareholder’s interests. Previous studies document a positive association between board independence and disclosure quality (Lim et al., 2007; Kaawaase et al., 2021). However, Katmon and Farooque (2020) provide strong evidence of positive reciprocal relationships between board independence and disclosure quality. Similarly, from the regulatory point of view, having a non-executive chairman is more favourable than having an executive chairman. Within agency theory, the presence of a non-executive chairman might be viewed as more credible, especially when handling important matters such as “monitoring, disciplining and compensating senior managers” (Barako et al. 2006, p. 111). Al-Gamrh et al. (2020) argue that although board independence is used to monitor the operation of the firm to improve firms’ performance and protect the interest of minority shareholder, emerging countries with low investor protection can use it to enforce political agenda rather than economic gain only. Brown et al. (2011) note that the role of the non-executive chairman is crucial in the dissemination of information. The UK Corporate Governance Code (2018) stipulates that the chairman must be an independent director on the date of appointment, and it is expected that the presence of a nonexecutive chairman increases the quality of a firm’s disclosure. Non-executive chairmen play an important role on boards by “ensuring the board activities are carried out with due diligence and information is provided to directors on a timely basis” (Brown et al., 2011, p. 113).

Additionally, the UK Corporate Governance Code underscores the significant impact of governance practices on the long-term success of the firm. The board’s strong independence, achieved through the separation of the chairman and CEO, and the balance between executive and independent non-executive directors (FRC, 2018, 2024), is a key aspect. Equally important is the board’s responsibility to share relevant information with stakeholders, demonstrating clear oversight and independence, which is equally applicable to all UK-listed companies, regardless of their size and industry differences. This transparency not only informs stakeholders but also plays a significant role in building trust in the board’s and executives’ decisions and practices.

We argue that when the board has a majority of independent directors and is chaired by an independent non-executive director, it signals “board committee effectiveness” that the board is effective to promote high standards of disclosure quality. As such, we expect board effectiveness to be associated with higher disclosure quality. Our second hypothesis is thus:

Acknowledging that the sample includes a wide range of firms from the FTSE350, we argue that the relationship between audit committee effectiveness, board effectiveness and disclosure quality might be moderated by the firm’s size. According to the agency theory, large firms are subject to higher public scrutiny, hence motivating the managers to provide a higher quality of disclosure, which potentially reduces the agency cost (Watson et al., 2002, p. 297). Moreover, large firms are more resourceful than small firms; hence, they are in a position to provide better information (Buzby, 1975). Furthermore, large firms are normally involved in various projects that require higher demand for various information (Lehn et al., 2009). Besides, large firms also have a greater tendency to provide more information to attract potential investors (Donnelly and Mulcahy, 2008). Previous literature demonstrated that there is a positive relationship between firms’ size and the quality of disclosure (Chow and Wong-Boren, 1987; Hossain et al., 1994; Singhvi and Desai, 1971; Inchausti, 1997). Hence, we predict that firm size moderates the relationship between audit committee effectiveness, board quality effectiveness and disclosure quality in large firms. Our third hypothesis is thus:

There is a lack of research in the literature examining the complementary or substitutive nature of board and audit committee quality governance effectiveness on disclosure quality. Given that both audit and board committees are effective monitoring tools, the present study intends to observe their complementary or substitutive relationships with disclosure quality. A complementary link indicates when the association shows a positive relationship between the interaction term and disclosure quality. This confirms that the roles of audit and board committee quality effectiveness are aligned with improving disclosure quality. On the contrary, a substitutive role implies a negative direction of the relationship between the interaction term and disclosure quality. Given that both H1 and H2 suggest positive relationships of audit and board committee effectiveness with disclosure quality, our fourth hypothesis is thus:

In addition, we also expect that firm size moderates the complementary link between board and audit committee effectiveness, and disclosure quality is more pronounced in large firms than for smaller firms. Our fifth hypothesis is thus:

We focus on two measures of disclosure quality – forward-looking disclosure and the receipt of the Investor Relations Magazine Award (both of these are explained below) ^[7]. These proxies have been used in the accounting literature, and they are admittedly far from imperfect, so our use of them in the relatively less regulated “comply or explain” environment of the UK is appropriate. As noted by Cerbioni and Parbonetti (2007), disclosure quality is a complex and highly debated issue in the literature that remains unresolved, thus appearing as a limitation. As such, it is impossible to find a disclosure quality proxy that is completely free from bias and/or subjectivity such as management earnings forecast or other disclosure index. Further, disclosure quality proxies typically reflect a decrease in the information asymmetry gap (Grüning and Ernstberger, 2010), relating to financial analysts who are viewed as sophisticated users of a company’s disclosure and effective disseminators of the company’s information (Gavious, 2007). This is because the analyst forecast is directly and indirectly related to our first and second proxies above, respectively, when projecting a firm’s earnings per share.

For the first proxy, forward-looking disclosure, our use of annual reports as a source of firm disclosure quality is consistent with prior literature (Bozzolan et al., 2009). Forward-looking information is one of the value-relevance information elements in the sense that it increases analyst forecast accuracy, associated with the share price. Such disclosure is embedded with predictive information about a company’s activities that make it possible to reduce earnings uncertainty in the future. Following the literature, we use “selected keywords” relevant to forward-looking to measure the quality of narrative forward-looking disclosure. Because the keyword selection was developed based on synonyms of forward-looking keywords that are widely used in annual reports and analyst reports. Forward-looking disclosure score (FLSCORE) was derived from the number of forward-looking sentences in the annual report. We used computerised content method analysis, whereby specific forward-looking keywords were detected in the companies’ annual reports using N6 software. The list of forward-looking keywords was adopted from Hussainey et al. (2003). Examples of forward-looking keywords include accelerate, anticipate, await, envisage, estimate, eventual, expect, forecast, forthcoming, outlook and predict. Using the search function, we entered forward-looking keywords one by one in N6 to produce a search result for each keyword by company. The total score represents the number of forward-looking sentences in the companies’ annual reports. We performed the validity test by reading 30 sentences for several keywords randomly selected from the search result produced by N6 and found that 96.60% of the sentences referred to the forward-looking information. This test confirms the ability of N6 to detect forward-looking information in documents imported into the software. Although bias cannot be fully eliminated, it is, however, relatively very low (around 3.4%). Moreover, the use of computerised content analysis provides a credible score with high comparability and consistency compared to traditional content method analysis (Hussainey et al., 2003).

Our second proxy for disclosure quality is the receipt of the Investor Relations Magazine Award (IRAWARD) by companies, consistent with the literature (Agarwal et al., 2008; Boesso and Kumar, 2007; Bushee and Miller, 2010). In contrast to the forward-looking disclosure (FLSCORE) based on information from the annual report and classified as an “internal proxy”, IRAWARD is an “external proxy” for disclosure quality that depends on the analyst’s perceptions of a firm’s investor relations activities in a year. The Investor Relations Magazine commissions annually an independent research firm to obtain nominations from market analysts and fund managers for firms that have performed the “best” in distinct categories of investor relations over the previous 12 months ^[8]. Recipients of the Investor Relations Magazine Award are selected based on the highest ranking/score by the respondents on the company’s investor relations activities. The respondents comprise sell-side analysts, buy-side analysts and portfolio managers in the UK^[9]. The Investor Relations Magazine Award covers many important investor relations components, including narrative reporting, corporate literature, internet reporting, virtual conferencing, disclosure practice, corporate social responsibility practice, annual reports, analyst meetings and briefings, as well as explanations about shareholder value. Also, recipients of the Investor Relations Magazine Award are considerably varied and are not biased towards large firms. Analysts are key players in the capital market, and they understand the value of the information provided by the company. So, the consensus among market analysts and fund managers in the selection of the Investor Relations Magazine Award recipient might be regarded as an alternative reasonable and more credible proxy to a self-developed disclosure index, which depends on the subjective judgement of researchers. Since IRAWARD relies on analysts’ perceptions of a firm’s disclosure policy, who are the key players in the capital market with expertise in evaluating a firm’s disclosure policy and covering many important investor relations components, while not being biased towards large firms, it is less likely to be influenced by external factors.

Our research is based on 170 matched-pair firms of total of 340 firms as presented in Table 1. We selected firms who were either winners or first runner-up (second position) for the Investor Relations Magazine Award in the year 2019, 2020, 2021 and 2022 to represent firms with high-quality disclosure ^[10]. Since the aim of the study is to focus on quality disclosure, for sample selection, we rely on the “external” measure of disclosure quality (IRAWARD) rather than the “internal” measure (FLSCORE) to make it free from bias and managerial discretion in disclosure. Moreover, both award-winning and first runner-up companies are recognised by market analysts for producing quality disclosure of financial and non-financial information compared to non-award-winning companies, including first runner-up companies. As such, non-award-winning companies are not included in the sample of this study. Now, after selecting winners and runner-up companies, we excluded firms operating in the financial and highly regulated industries because they are subject to specific regulations ^[11]. Consistent with Boesso and Kumar (2007) who also used a match-paired sample using Investment Relations Magazine Award winners in the USA, we applied stringent criteria to control for the year, industry and size effects. In composing our control sample, we matched firms using the following criteria:

1. same year and industry under observation;

2. closest total assets; and

3. not nominated as a winner or runner-up in the years under observation.

Also, given that the Investor Relations Magazine Award for a particular year is based on the assessment of a firm’s investor relations activities in the preceding year, we have used lagged data in our study.

The list of winners and nominated firms for the IR Magazine Award was obtained from the Investor Perception Study research report, produced by the Cross Border Group Ltd (IR Impact, 2025). Award data was requested directly using email from the event organiser at the Cross Border Group. The forward-looking information data was manually extracted from companies’ annual reports. Similarly, corporate governance attributes were also manually collected from these annual reports. We used Datastream to obtain financial data relating to control variables.

In examining the main effects on disclosure quality, including forward-looking disclosure (FLSCORE) and receipt of the Investor Relations Magazine Award (IRAWARD), our empirical model is specified as follows.

In the regression model, in addition to the main variables of interest, we also control for a number of governance-related variables that have been incorporated in prior literature. In particular, consistent with prior studies, our regression model includes board meetings, substantial ownership, auditor type and analyst following variables. Additionally, we control for other firm-specific variables, including firm size, profitability, leverage, firm growth, industry dummies and year dummies:

where:

FLSCORE = Total number of forward-looking statements in the annual report

IRAWARD = Value 1 if receive winner or first runner-up for disclosure award, 0 for non-winner

ACQUALITY = Value 1 if ACSIZE >/=3, ACIND = 100%, ACMEET>/=3, and ACEXP>/=1, Otherwise 0.

ACQUALITYBR = Value 1 if ACSIZE >/=3, ACIND = 100%, ACMEET>/=4, and ACEXP>/=1, Otherwise 0.

ACSIZE = Number of audit committee member (1 = if audit committee members =>3, 0 = otherwise)

ACIND = Audit committee independence (1 = if all audit committee members are independent, 0 = otherwise)

ACMEET = For ACQUALITY, number of audit committee meetings (1 = if audit committee meeting => 3, 0 = otherwise). For ACQUALITYBR (as suggested in Blue Ribbon Recommendation), number of audit committee meetings (1 = if audit committee meeting => 4, 0 = otherwise)

ACEXP = Audit committee members with expertise (1 = if audit committee members with financial literacy is => 1, 0 = otherwise)^[12]

BODQUALITY = Value 1 if BODIND>/= 50% and BODDUALITY = 1, Otherwise 0.

BODIND = Percentage of independent directors on the board [excluding chairman]

BODDUALITY = Status of the board chair (1 = non-executive, 0 = executive)

BODSIZE = Number of board members

BODMEET = Number of board meetings held during the year

SUBSOWN = Total percentage of shares held by substantial shareholders (i.e. 3%/+)

BIG4 = Auditor a Big4 firm, including KPMG, PWC, EY and Deloitte (Big4 = 1, Non-Big4 = 0)

ANALYST = Number of analysts following

PROFIT = Return on assets

LEV = Debt to asset ratio

SIZE = Natural log of total sales revenue

GROWTH = Market to book value ratio

INDUSTRY = Industry dummies

YEAR = Year dummies

We incorporated several control variables in our model. We control for board size (BODSIZE) due to the fact that a large board offers greater information related to the firm’s performance from the perspective of technology, product, regulation, etc. (Lehn et al., 2009). We include board meeting (BOARDMEET) since Karamanou and Vafeas (2005) found that high board meetings increased earnings forecast updates than firms with fewer board meetings. We also controlled for firm size (SIZE) in line with Nelson et al. (2010) since large firms have a greater tendency to provide better disclosure transparency than small firms because they are in need of more external capital to attract potential investors (Donnelly and Mulcahy, 2008, p. 420). Regarding profitability (PROFIT), prior studies recognise that profitability potentially creates incentives for managers to provide more disclosure (e.g. Nelson et al., 2010). Managers in highly profitable firms are encouraged to provide better disclosure, given that profitable firms have more information to disclose about the projects they are involved in (Li et al., 2008).

We followed Nelson et al. (2010) in the inclusion of large audit firms (BIG4) in our model. Large audit firms are more competent in consulting their clients and providing higher-quality information in their annual report (Wallace et al., 1994). With respect to the analyst following (ANALYST), Yu (2008) argues that analysts motivate firms to supply accurate information to users. As such, monitoring by financial analysts has the potential to influence managers to provide better quality disclosure. This is in line with Lang and Lundholm’s (1996) finding that there is a positive link between analyst following and disclosure quality. We control for leverage (LEV) in line with Nelson et al. (2010) since firms with high leverage tend to produce higher-quality information to mitigate their condition in the eyes of their creditors (Wallace and Naser, 1995). We included SUBSOWN in the model since substantial shareholders have the power to demand greater disclosure from the firms (Heflin and Shaw, 2000). GROWTH has been controlled since firms with high growth opportunities have more projects and planning, hence possess greater information to disclose than low-growth firms (Khurana, Pereira, and Martin, 2006).

To control for the variation of year effects, a year dummy was also introduced in the model, as in Nelson et al. (2010). Boone et al. (2007) claim that controlling for industry effects is able to cater for the heterogeneity factor, given that each industry shares “similar production technology and market conditions” (p. 76). Hence, industry level information significantly influences a firm’s disclosure environment (Piotroski and Roulstone, 2004). Therefore, in our study, industry variation was captured using industry dummies, as in Beekes and Brown (2006).

Table 2 presents the descriptive statistics for disclosure quality proxies forward-looking disclosure (FLSCORE) and receipt of the Investor Relations Magazine Award (IRAWARD), corporate governance and control variables. The descriptive statistics show that the mean forward-looking disclosure (FLSCORE) is 103.91, whilst the maximum value is 401 and the minimum value is 11. As we match-compare samples between firms with and without disclosure quality, there is a wide gap from the minimum-to-minimum forward-looking disclosure score. This leads to a standard deviation of forward-looking disclosure score showing a high value of 67.89. The high score in our study indicates that the extent of forward-looking disclosure has substantially increased over time. Given that our Investor Relation Awards (IRAWARD) variable is dichotomous (1 = winner, 0 = non-winners), the mean is an unsurprising 0.5. With respect to the audit committee characteristics of the firms, the mean audit committee size (ACSIZE) shows that 95.6% of the audit committees in the sample are comprised of three or more members, while 91.5% of audit committees have at least one member with recent financial expertise (ACEXP). The mean (median) for audit committee independence (ACIND) is 90.6% (100%) and about 95.3% of audit committees meet (ACMEET) at least three times a year. Further, the descriptive statistics show that 79.7% and 60.3% of the sample fully comply with the audit committee effectiveness (ACQUALITY) and audit committee effectiveness based on criteria suggested in the Blue Ribbon Recommendation (ACQUALITYBR), respectively.

About 74.4% of the sample companies show board effectiveness (BODQUALITY), with 57.53% independent directors on board (BODIND), and 86.2% companies chaired by a non-executive director (BODDUALITY), the average number of board members (BODSIZE) 9.62 and board meeting (BODMEET) 8.703. The mean (median) percentage of substantial shareholdings (SUBOWN) is 30.31% (29.13). The majority, 97.3%, of the firms in the sample are audited by large audit firms (BIG4), while the mean number of analysts following (ANALYST) is 14.78. With respect to the control variables, the mean profitability (PROFIT) is 7.07’; mean leverage (LEV) is 25.52; mean firm size (SIZE) is £569.81m and the average market to book value (GROWTH) is 3.86. For determining in case of endogeneity problem concerns, ranked-portfolio of board independent (PORTFOLIOBOD) and industry-average board independent, excluding firm i (INDBODIND) are included as an instrumental variable. INDBODIND is a proxy for industry-level board independent with the range from 46.42 to 73.07.

We tabulate our t-test findings in Table 3. Our result demonstrated that the IR winners have significantly higher forward-looking disclosure (FLSCORE) at p  < 0.01, greater effectiveness of audit committee (ACQUALITY, p  < 0.05) and (ACQUALITYBR, p  < 0.01) and a higher board size (BODSIZE, p  < 0.01) than the non-winners of IRAWARD. Moreover, our finding also indicates that the IRAWARD winners have more analyst following (ANALYST, p  < 0.01), a larger size of total sales revenue (SIZE, p  < 0.01) and higher profitability (PROFIT, p  < 0.05). Our results are qualitatively similar when we run the alternative analysis using the Mann–Whitney U test.

The correlation between the dependent and independent variables used in the regression analysis is given in Table 4. We carried out pairwise correlation to check the direction of the relationship among all the variables as well as to observe multicollinearity. A correlation coefficient of above 0.9 and variance inflation factor (VIF) of more than 10 indicates that multicollinearity is present (Hair et al., 2014), and this might lead to accidental significant results. Correlation coefficients in Table 4 show that the highest correlation is 0.58, which is between the number of analysts following (ANALYST) and the natural log of total sales revenue (SIZE). Further tests (untabulated) reveal that VIF is below 10 for all variables, thus confirming that multicollinearity does not pose any problem to our data.

To examine the effect of the audit committee and board effectiveness on financial reporting/disclosure quality, we first run Poisson regression that suits the model with count data when forward-looking disclosure is the dependent variable ^[13]. Model 1 and Model 2 in Table 5, the multivariate tests of the effect of forward-looking score (FLSCORE) on audit committee effectiveness, measured using ACQUALITY and ACQUALITYBR, show a significant positive influence at p  < 0.01 (coef = 0.162) and p  < 0.01 (coef = 0.1098), respectively, thus supporting our H1. These results imply that firms meet the recommended benchmark set by regulators in relation to audit committee characteristics provide better forward-looking disclosure than their counterparts. Notably, the composite board effectiveness measure (BODQUALITY) does not indicate any explanatory power on forward-looking disclosure in either model, thus not supporting our H2 here. This reason could be because, as per extant literature, board quality effectiveness also depends on board size and board meeting, in addition to board independence and board duality. Likewise, the insignificant effect of board effectiveness may be linked to the busyness of independent board members, multiple memberships, overreliance on other governance tools such as quality internal audit and audit committee and/or “comply or explain” environment in the UK board governance structure. For other governance variables, such as the number of board members (BODSIZE), board meeting (BODMEET), large audit firm (BIG4) and analyst following (ANALYST) are all positively associated with forward-looking score (FLSCORE) in both models. This implies that monitoring by internal and external governance mechanisms is crucial in improving disclosure quality. However, substantial shareholders (SUBSOWN), reported a significant negative relationship in Model 1 (coef = −0.0034, p  < 0.01) and Model 2 (coef = −0.003, p  < 0.01), thus indicated that higher SUBSOWN reduce disclosure quality. Previous literature suggest that higher substantial ownership might reduce public disclosure quality since the substantial shareholders might request and obtain more information through private channels from the company itself (Eng and Mak, 2003).

In Models 3 and 4, when we replace the board effectiveness measure (BODQUALITY) with its individual measures of board independence (BODIND) and non-executive chairman (BODDUALITY), we find that BODIND, but not BODDUALITY, appears to have a negative relationship with forward-looking disclosure (FLSCORE) in Models 3 and 4, thus “partially” rejecting our H2 here. Eng and Mak (2003) and Gul and Leung (2004) found a negative link between outside directors and voluntary disclosure, thus suggesting that firms might exchange board independent with disclosure quality as monitoring structure. Other findings in Models 3 and 4 remain similar to Models 1 and 2. Overall, after controlling for the board of directors’ quality effectiveness, we find that the audit committee effectiveness that meet recommended regulatory benchmarks (i.e. ACQUALTY and ACQUALITYBR) are associated with increased disclosure quality.

Again, in Model 5, we replace audit committee effectiveness measures (ACQUALITY and ACQUALITYBR) with their individual measures of audit committee size (ACSIZE), independence (ACIND), expertise (ACEXP) and meeting (ACMEET), our results reveal that ACIND, ACEXP and ACMEET have a positive significant association with forward-looking disclosure (FLSCORE). In addition, replacing board effectiveness measure (BODQUALITY) by board independence (BODIND) and non-executive chairman (BODDUALITY), we find BODIND remains significantly negative, while BODDUALITY is insignificant with FLSCORE, similar to Models 3 and 4 reported earlier. On the other hand, board characteristics such as board size (BODSIZE) and frequency of board meetings (BODMEET) appear to have a positive relationship with forward-looking disclosure (FLSCORE). Thus, we conclude that H1 is fully supported, and H2 is partly supported. Large audit firm (BIG4) and analyst following (ANALYST) also show a positive significant association with FLSCORE and substantial shareholders (SUBSOWN) significant negative relationship, as reported before in other models. Our findings provide support for the view that audit committees and boards play a complementary role in improving disclosure quality.

We also test for the effect of the audit committee effectiveness and board effectiveness using another proxy for disclosure quality – the receipt of the Investor Relations Magazine Award (IRAWARD). Logistic regression is applied to determine the effect on IR award, which is a dummy variable and value into two groups of figures: value 1 when the firms are the winner or the first runner-up for the award, and value 0 when the firms have no award. Similar to that for forward-looking disclosure, Table 6 shows logistic regression results in five models.

Regression results for Models 1 and 2 reveal that the composite measure for audit committee effectiveness, ACQUALITY and ACQUALITYBR, both are positive and significant at p  < 0.01. By contrast, the composite board effectiveness measure (BODQUALITY) remains insignificant as reported before. In Models 3 and 4, when we replace board effectiveness measure (BODQUALITY) with board independence (BODIND), non-executive chairman (BODDUALITY), we find that the firm receipt of IRAWARD increases with the presence of board independence (BODIND) but not with non-executive chairman (BODDUALITY), in Model 3 only. Further, board size (BODSIZE), but not board meeting (BODMEET), shows a significant positive relationship with IRAWARD in both Models 3 and 4. This suggests that our H2 is “partially” supported here. In Model 5 when ACQUALITY or ACQUALITYBR is replaced by audit committee size (ACSIZE), independence (ACIND), expertise (ACEXP) and meeting (ACMEET) and board effectiveness measure (BODQUALITY) by board independence (BODIND) and non-executive chairman (BODDUALITY), only the frequency of audit committee meeting (ACMEET) and independent directors (BODIND) are showing a significant positive association with IRAWARD. Board size (BODSIZE) also shows positive relationship with IRAWARD. These findings provide “partial” support of both H1 and H2. As for other governance variables, analyst following (ANALYST) is consistently showing a positive effect on IRAWARD, though that is not the case for large audit firm (BIG4). Overall, with some variations, the results support the view that audit committee and board with quality characteristics improve the quality of financial reporting in the UK’s less regulated “comply or explain” environment.

To observe the moderation effect of firm size, we further conduct regression analysis by splitting the sample into large and small firms based on median size. Table 7 presents the findings for the effect of audit committee effectiveness and board effectiveness on both FLSCORE and IRAWARD split between small and large firm samples. For large sample firms, both ACQUALITY and ACQUALITYBR have a positive significant association with FLSCORE but not with IRAWARD. Small firms consistently show insignificant result for ACQUALITY and ACQUALITYBR with FLSCORE and IRAWARD. Therefore, our H3 is “partially” supported. Overall, the findings are generally consistent with our main results in Tables 5 and 6 and support the view that the regulatory benchmark help improves financial disclosure quality.

To examine the potential complementary or substitutive effect between the board and audit committee characteristics, an interaction term (ACQUALITY*BODQUALITY) is developed to rerun the original regression models in Tables 5 and 6, with the expectation of a positive relationship with disclosure quality proxies (FLSCORE and IRAWARD), to indicate that they complement each other. Table 8 of the regression results reveal that the interaction term (ACQUALITYBR * BODQUALITY) is positive and significant for IRAWARD (Coef = 4.527, p  < 0.10) but positive and insignificant for FLSCORE (coef = 0.068). For ACQUALITY * BODQUALITY, both FLSCORE and IRAWARD shown insignificant results, though positive relationship is reported. Therefore, our H4 is “partially” supported here.

Again, when we include the interaction term in our split sample of large versus small firms in Table 9 and Table 10, the interaction term ACQUALITBR * BODQUALITY has a significant positive association with FLSCORE for the large firms in Table 9, indicating a complementary relationship between the board and audit committee effectiveness in large firms. However, it is negative and significant for small firms in Table 9. The other interaction term ACQUALITYBR * BODQUALITY remains insignificant for both large and small firms. Thus, H5 is “partially” supported here. For the IRAWARD proxy for disclosure quality in Table 10, the interaction terms are negative and insignificant for both large and small firms, so not supporting H5. Overall, these results are consistent with the view that audit committee and board effectiveness play complementary each other in improving the quality of forward-looking information in large firms. In small firms, the role of audit committee effectiveness can be substitutive to board effectiveness function in improving the firm’s quality of disclosure.

In this section, we perform several additional analyses (i.e. relating to unique sample and endogeneity problem concern) to check for the robustness of our findings.

Recognising Brown et al.’s (2011) point that firm’s corporate governance practices are unlikely to be changed over the long run and the nature of data we have used and its limitations, in particular, the utilisation of non-unique sample over the period may introduce “stickiness” bias to our findings. To address this concern, we re-ran the main models for the effects of audit committee and board characteristics on FLSCORE and IRAWARD using unique firms per period by selecting the data from the most recent year if a firm is a winner in more than one year. Our results based on a unique sample of 190 firms presented in Table 11 are qualitatively similar to our main findings reported in Tables 5 and 6. In Particular, audit committee effectiveness ACQUALITY and ACQUALITYBR still demonstrate positive significant effects on FLSCORE, while ACQUALITYBR on both FLSCORE and IRAWARD. In addition, BODQUALITY also reveals positive significant effects on FLSCORE. The reduced unique sample results provide support for the view that audit and board committees that meet the regulatory benchmark contribute to achieving better financial disclosure quality.