Do audit committee attributes improve integrated reporting quality? Evidence from European Companies listed on STOXX EUROPE 600 INDEX

This paper aims to shed light on the relationship between audit committee attributes and the integrated reporting quality (IRQ). This study uses panel data and multiple regression. The outcomes almost entirely conrm the expectations relating to the impact of audit committee attributes on the IRQ. In particular nancial expertise and meetings are shown to have a signicant positive impact on IRQ. However, we do not nd a signicant association between key aspects of the audit committee function, such as audit committee independence and IRQ. The ndings show that the legal traditions of the country in which the company operates have no effect on the inuence of audit committee features on integrated reporting quality. The ndings also demonstrate that the civil law system (CIVLAW) has a positive and signicant impact on integrated reporting quality (IRQ) at p = 0.002. This indicates how companies operating in countries with civil law provide higher-quality integrated reports than companies operating in countries with different legal systems. This paper was designed to investigate the impact of audit committee attributes on the integrated reporting quality of European companies belonging to the STOXX EUROPE 600 index. For a more reliable estimation of integrated reporting quality, the measures proposed by ASSET4 were used. This paper attempts to ll the gap in the literature by theoretically and empirically investigating the role of audit committee characteristics. This study tested the effects of audit committee independence, audit committee nancial expertise, and audit committee meeting frequency as determinants of integrated reporting quality. The analysis was conducted on a sample of 213 European companies belonging to the STOXX EUROPE 600 index for the period 2008-2020. We nd that the educational background and experience in nance, and the annual activity level of the audit committee were shown to have à positive effect on the IRQ. On the contrary, this study highlighted that the percentage of independent members of the audit committee does not affect the IRQ.


Introduction
Among the various mandatory and voluntary modes of corporate reporting, "integrated reporting (IR) has rapidly developed as a new accounting practice to help companies understand how they create value and be able to effectively communicate this to external stakeholders" (Perego et al., 2016). Thus, the audit committee is a topical subject because the spread of the practice of this committee is a relatively recent feature of the way in which boards of directors are organized. From an agency theory perspective, the audit committee, through its oversight, control, and monitoring functions, supports the stakeholders. Therefore, the contributions concerning the determinants of integrated reporting quality are limited. This signi cant gap in literature makes it necessary to analyze the key factors that can explain the quality of integrated reporting (Chouaibi et al., 2021).
The board of directors and its committees are a relevant corporate governance mechanism in monitoring managerial actions (Fama and Jensen, 1983). Therefore, an effective corporate governance structure would lead to greater value creation in organizations, while providing a basis for IR to provide quality information (Cooray et al., 2020). In recent years, both academics and standard setters have called for improvements in the audit committee con guration.
The economic theory that puts corporate governance issues into perspective is primarily agency theory. It considers that the agency relationship that exists between the principal (the shareholders) and the agent (the managers) involves asymmetries of information of two kinds: adverse selection and moral hazard. According to Gomez (1996), asymmetric information is at the origin of the contractual relationship. Thus, Audit committee is viewed as a monitoring tool to bridging the gap of information asymmetry which in turn results in minimizing agency costs (Altawalbeh, 2020).
The literature views the audit committee as a new institutional mechanism of corporate governance, set up as a nal level of control over the quality of the audit process and reporting. Moreover, the originality of our research is justi ed by the scarcity of studies that have addressed the relationship between audit committee effectiveness and integrated reporting quality. This is a subject that has gained momentum following the various nancial scandals that have discredited the quality of the nancial information published in the nancial statements. In a climate where there is a loss of con dence in nancial reporting, various regulatory texts have converged to rede ne the governance bodies of the company, with the establishment of the audit committee within the board of directors. This body is now responsible for assisting the board in its responsibilities regarding the quality of nancial information before its nal approval and communication. In view of these considerations, we will ask the following question: Is the legal presence of the audit committee in listed European companies an effective control mechanism for improving the quality of integrated reporting?. This paper aims to ll the important gap by analyzing the effect of three audit committee attributes (independence, meeting frequency and nancial expertise) on the IRQ, in regard to a selection of 213 European companies belonging to the STOXX EUROPE 600 index between 2008-2020.
In terms of contribution, the present study provides empirical evidence on the relationship between audit committee characteristics and the quality of integrated reporting released by European companies. Indeed, this paper may be useful for companies and stakeholders as it adds to the existing integrated reporting literature by analyzing the audit committee role in ensuring the consistency of integrated reporting, the external assurance of which is still an undiscovered research eld. In the same vein, managers, shareholders, and politicians will be particularly interested in these ndings. As a result, while determining the best reporting strategy (increasing risk estimation, stock volatility, and boosting long-term shareholder value and reputation), stakeholders should examine the accuracy of disclosure.

Theoretical Background
Our intention is to provide a detailed explanation of the related accounting phenomenon, the subject of our study, using the following three main theories: the agency theory, the signaling theory, and the resource dependency theory.
The agency theory. Is based on the existence of con icts of interest between shareholders and managers, and different types of agency costs are associated with these con icts (Jensen and Meckling, 1976). Indeed, it is based on the asymmetry of information that is at the origin of this con ict, as con rmed by Gomez (1996): "the asymmetry of information is therefore at the origin of the contractual relationship". The manager is perceived as opportunistic by his ability to use this information differential in his favor. The opportunism of the agents consists in concealing part of the information from the principal, in particular at the time of the signature of the contracts, which makes them incomplete, and therefore imperfect. It is the lack of transparency of information that allows the opportunism of managers and encourages them to manipulate the information according to their strategies. To this end, Gomez (1996) notes: "the opportunism of the actors involved is due to the asymmetries of information that they exploit". This is why controls must be put in place in order to manage the ow of information as well as possible and to guarantee the convergence of interests between the stakeholders. Control mechanisms are analyzed according to an internal/external typology (Charreaux, 1997). From the perspective of agency theory, boards of directors act as internal control mechanisms that help align the interests of shareholders and those of managers (Fama and Jensen, 1983).
Signaling theory. Is based on the fact that information is unequally shared by the actors in a company and even if it were shared by all, the same information would not be perceived in the same way (Charreaux, 1997). The managers of a company thus have more information than investors. They make decisions that they consider favorable in order to convince the market; to do this, they use signals. A signal is a transfer of information to the actors in the rm in order to reduce the original information differentials, but its validity is subordinated to its cost (Spence, 1973). The emission of a signal is interpreted as additional information provided to the market with the aim of in uencing investors and provoking the reaction of the share price (Noe and Rebello, 1996). Thus, a positive signal from a company's management can be used to anticipate better future performance and lead to an increase in the company's share price and vice versa. The managers of companies in a good nancial situation have an interest in providing information to investors in order to distinguish themselves from rms in a bad position.
Resource dependency theory Developed by Pfeffer (1973) and Pfeffer and Salancik (1978), emphasizing the important role played by the board of directors in accessing resources that would enhance rm performance. The fundamental suggestion of resource dependence theory is the need for environmental links between the rm and the external resources needed to survive. This means that boards of directors are an important mechanism for absorbing critical elements of environmental uncertainty in the rm. Thus, environmental linkages could reduce the transaction costs associated with environmental interdependence. The organization's need to demand resources leads to the development of exchange relationships between organizations. In addition, the unequal distribution of resource requirements leads to interdependent organizational relationships. In summary, resource dependence theory provides a compelling rationale for creating linkages between the rm and its external environment through independent advice, as rms that create links could improve their survival and performance (Yusoff and Alhaji, 2012). Hence, the resource dependency theory suggests that larger audit committees are better able to provide greater resources and authority to effectively carry out their responsibilities (Allegrini and Greco, 2013).

Hypotheses Development
Because of the asymmetry of information between owners and managers, the board of directors plays a central role in safeguarding the interests of shareholders. Indeed, in order to minimize agency costs, the board assumes a supervisory role over the management through the approval of the company's strategy, the monitoring of the control system, and the maintenance of reliable accounting records. Given its various responsibilities, the board of directors generally delegates its oversight function to the audit committee. The audit committee can enhance the quality of the audit process by facilitating the disclosure of irregularities. One of the purposes of IR is to ensure transparency on nancial and non nancial information, such as strategy, business models, risks and opportunities, governance, compensation, and sustainability. This increased transparency is likely to be used as a governance tool to prevent corporate opportunism.
Certain attributes of the audit committee make it more e cient to perform the above tasks in the disclosure process, leading to better information in the integrated reports. The unique assumptions regarding the effects of individual audit committee attributes on integrated reporting quality are elaborated below.

Audit committee independence and integrated reporting quality
One of the principal characteristics that can in uence the audit committee's ability to supervise and monitor is its independence ( independence, and multiple terms have a positive effect on integrated reporting quality. They also nd that the frequency of meetings and nancial expertise of the audit committee have no effect on the level of voluntary disclosure. Therefore, we introduce the following hypothesis: There is a positive association between the audit committee independence and integrated reporting quality.

Audit committee nancial expertise and integrated reporting quality
The audit committee's competences and individual members' expertise, according to existing literature, improve its e ciency (Velte, 2018). The appropriate expertise of the members of the environmental audit committee is associated with greater transparency of environmental and sustainability-related information as this committee presents a guarantor of the quality of internal control and the reliability of environmental information (Jamoussi and Jarboui, 2018).
In the same vein, Huang and Thiruvadi (2008) show that there is a substantial link between audit committee accounting expertise and fraud prevention. On the contrary, Appuhami and Tashakor 2020) found that the educational background and experience in nance of the members of the audit committees have no effect on the quality of the integrated reports.
The ndings of Cohen et al. (2013) indicate that combining industry expertise with accounting competence can increase the audit committee's e ciency in monitoring the nancial reporting process. According to Sultana et al. (2015), audit committee members with nancial expertise, prior audit committee experience, and independence are related with shorter audit delays. On the basis of these developments, we can make the following hypothesis:

Hypothesis 2
Audit committee expertise has a signi cant positive impact on integrated reporting quality.
Audit committee meeting frequency and the integrated reporting quality According to Khlif and Samaha (2016), audit committee meetings improve internal control quality. In addition, Li et al. (2012) found that participating at least four audit committee meetings per year is signi cantly and positively connected to the level of voluntary disclosure. They argue that having audit committee meetings on a regular basis improves the committee's monitoring ability. Appuhami and Tashako (2017) investigated the link between audit committee size, frequency of meetings, committee independence, and gender diversity and voluntary corporate social responsibility disclosure. They found that audit committee size, frequency of meetings, committee independence, and gender diversity have a signi cant positive impact on the level of corporate social responsibility disclosure.
As previously noted, the IR's complexity requires extensive oversight and monitoring. In this regard, a higher frequency of audit committee meetings, as well as improved supervisory and monitoring functions for this body, could help to improve the quality of the process of acquiring and representing information in an integrated and quality manner. The rationale for including nancial experts is to help other members understand the auditors' judgments and to discern the substance of disagreements between the external auditors and management (Mangena and Pike, 2012). Thus, audit committees that meet regularly, according to Karamanou and Vafeas (2005), have more time to e ciently perform the function of monitoring disclosure processes. In the same vein, Altawalbeh (2020) found that more audit committee meetings in a regular basis would enhance the effectiveness of the monitoring function ability to perform its monitoring function. Accordingly, the hypothesis to be empirically tested is: Hypothesis 3 There is a positive association between audit committee meeting frequency and integrated reporting quality.

Method
Sample construction and data collection Panel A of Table 1 provides an overview of the sample-selection process.

Measures
Dependent variable: Integrated reporting quality (IRQ) The dependent variable is the degree of consolidated integrated reporting which is calculated using the Thomson Reuters Datastream ASSET4. It represents the quality of the information contained within the integrated reports and will be measured by the ASSET4 Data Stream 'CGVS' item. CGVS is de ned in Thomson Reuters' ASSET4 database as "the commitment and effectiveness of the company's management in creating an overall vision and strategy that integrates nancial and non-nancial aspects. It re ects a company's ability to convincingly demonstrate and communicate that it integrates economic ( nancial), social and environmental dimensions into its daily decisionmaking processes" (Thomson Reuters). This score takes values between 0 and 100 (Chouaibi et al., 2021).

Independent variables
Independent variables are de ned as follows: audit committee independence (AC_IND) is measured by the percentage of independent members to the total number of the audit committee members (Raimo et al., 2020). In addition, we de ne the nancial expertise of the audit committee (AC_EXP) as the proportion of audit committee members possessing professional accounting quali cations among the total number of audit committee members.
Finally, the variable denoted (AC_MEET) is de ned as the number of audit committee meetings held during the year.

Board characteristics and rm's features as control variables
The audit committee attributes are not the only factors that in uence IRQ. We control for other determinants of IRQ To control for possible variable across countries, industries and time, we also include country, industry and year dummies (the industry dummies are based on the rst digit of the SIC code). The speci c measures of all variables are described in Table 2.

Regression Model
To test the three hypotheses described in Section 2, we estimate a panel regression model where the integrated reporting quality is used as a dependent variable.
We have speci ed one econometric model for estimation. Equation (1)  YEAR, INDUSTRY and COUNTRY stand respectively for year, industry and country effects; ε it is the random error term.
Concerning the indices i and t, they correspond to rm and period components of the study. Detailed de nitions and measurements of all study variables are summarized in Table 2.
Equations are estimated using a panel data methodology, applying the ordinary least squares regression (GLS).

Results
Descriptive statistics Table 3 reports some descriptive statistics for all the analyzed variables, except the industry and year dummies used in the model analysis. All the non-indicator variables are winterized at the 1% and 99% levels to eliminate extreme observations. Regarding the debt ratio, the companies in question have an average ratio of around 29.1%, it is important to note that European companies are more indebted. In terms of rm pro tability (ROA), the mean of the rm pro tability is 0.054, while the maximum and minimum are 0.693 and-1.209, respectively, with a standard deviation of 0.075.
In addition, the companies in the sample have an average ROE of 0.117. We note that, on average, size of the companies (measured by total log assets) of around 16 with a low of 4.468 and a high of 19.745. This shows that the rms in our sample are large sizes.  Table 2.

Correlation analysis
We explore the relationship between the characteristics of the audit committee and integrated reporting quality through a Pearson rank correlation. The Table 4 shows the results. They show no evidence of an unacceptable level of multicollinearity. In fact, the highest correlation coe cient is 0.66 for ROE and ROA. According to Farrar and Glauber (1967), harmful levels of multicollinearity should not exist until the correlation coe cient reaches ±0.8 or ±0.9. In support of this, we have evaluated the potential multicollinearity among the variables also using a variance in ator factor (VIF) analysis. VIFs vary between a low value of 1.01 to a high value of 1.83. The effect of multicollinearity is not signi cant if all VIFs are less than 10 (Myers, 1990). Therefore, multicollinearity does not represent a problem in the interpretation of our ndings Given the particular nature of panel data, it is necessary to follow the order of some econometric steps and perform certain tests to lead to robust estimations. First, we point out that the xed effects model is discarded since our regression includes time-invariant dummy variables. Performing a xed-effect regression would lead to arbitrarily omitting such variables. The Breusch-Pagan LM test for randomness produces a signi cant Chi-square value. We con rm the existence of individual effects, and we run a random effect regression model. Second, we test for panel level heteroskedasticity using the Breusch-Pagan test, as shown in Table 5. According to this table, the tests applied allowed acceptance of the null hypothesis of heteroscedasticity (p-value = 0.7347).
The decision of xed and random effects rests on the result of Hausman's test. A Hausman test has been typically used to determine the consistency of the GLS estimator in static models with pooled cross-section-time-series data. This test, when run for the data of the present study, gave a signi cant result, which proves the use of random effect regression analysis for us model before regression analysis. Hausman's test was run, and for our model, a random effect was applied as per the results of Hausman's test. The results support H2 and H3, while H1 is not supported. In particular, the results for H1 the results show a positive but non-signi cant relationship between AC_IND and IRQ at p = 0.318. This result suggests that greater independence of the audit committee has no signi cant impact on integrated reporting quality. Thus, H1 is rejected.

Test of Hypotheses
The results obtained regarding the impact of the audit committee independence on the IRQ extend those obtained by Li et al. (2012) and Ismail and Rahman (2011).
In addition, the results show that the educational background and experience in nance of the audit committee members do in uence the quality of the integrated reports provided by the rms (p = 0.03). Therefore, H2 is accepted. This result supports various studies (for instance: Raimo et al. (2020) and Velte, 2018). In which there is a positive and signi cant association between audit committee nancial expertise and integrated reporting quality. independence, frequency of meetings and gender diversity on the level of CSR disclosure. They also stress that the presence of a CSR or sustainability committee has a positive effect on the quality of integrated reports. They show a positive effect of CSRCOM on the IRQ at p = .096. Also, we nd that LEVE, ROA and ROE are non-signi cant associated with integrated reporting quality. Finally, Table 5 shows a positive effect of rm size, with FSIZE having a positive effect on the IRQ at p = .015 These tests achieved results are displayed in Table 6. The ndings show that the legal traditions of the country in which the company operates have no effect on the in uence of audit committee features on integrated reporting quality. The ndings also demonstrate that the civil law system (CIVLAW) has a positive and signi cant impact on integrated reporting quality (IRQ) at p = 0.002. This indicates how companies operating in countries with civil law provide higher-quality integrated reports than companies operating in countries with different legal systems.

Theoretical Contribution
The application of the agency, signaling, and resource dependency development theories in the framework of this research shows that a director's application of ethical behavior is more likely to enhance the accuracy of nonnancial information at the cultural, environmental, and social levels, while reinforcing, encouraging, and consolidating the de nition of CSR within companies. The ndings provide contradictory evidence from the agency theory perspective, carefully handling these corporate governance characteristics is essential in order to strike a balance between principal agent relationships and avoid future agency costs. Also, the paper contributes to addressing the gap that currently exists as to what is meant by IRQ. Despite the large number of studies that examine IR, there is still a lack of consensus on understanding IRQ.

Practical Implications
Several practical implications. In fact, managers and policymakers can bene t from our ndings as well. In terms of managers, our ndings show that nancial accounting knowledge on the audit committee enhances the external auditors' dependence on internal audit work, which improves integrated reporting quality. In terms of policymakers, our ndings show that having nancial specialists on an audit committee has a major impact on improving nancial reporting quality through timely disclosure.
The ndings of this paper have signi cant policy consequences. Through proper legislation, policymakers should encourage the formation of larger audit committees and audit committees with a higher percentage of independent members. They should also establish a minimum number of audit committee meetings per year. These regulations, which aim to increase the e cacy of audit committees' supervisory and monitoring tasks, would promote corporate transparency and improve the quality of integrated reports.

Future Research Directions
This study offers some signi cant ndings, they should be considered with care due to the study's limitations. To start, only three audit committee attributes that were relevant to the study topic were chosen for this investigation. As a result, future research into additional audit committee characteristics, such as audit committee tenure, is possible. Second, the study sample included non-nancial industries, which have a variety of audit committee and reporting processes. As a result, nancial organizations are excluded from all studies on the impact of audit committee attributes on integrated reporting quality. It will be interesting to see if there is a major difference between nancial and non-nancial organizations when it comes to the impact of audit committee attributes on integrated reporting quality. Third, the absence of endogeneity tests, i.e., the relationship between audit committee attributes and the integrated reporting quality, could be studied in both directions.

Conclusion
This paper was designed to investigate the impact of audit committee attributes on the integrated reporting quality of European companies belonging to the STOXX EUROPE 600 index. For a more reliable estimation of integrated reporting quality, the measures proposed by ASSET4 were used. This paper attempts to ll the gap in the literature by theoretically and empirically investigating the role of audit committee characteristics.
This study tested the effects of audit committee independence, audit committee nancial expertise, and audit committee meeting frequency as determinants of integrated reporting quality. The analysis was conducted on a sample of 213 European companies belonging to the STOXX EUROPE 600 index for the period 2008-2020. We nd that the educational background and experience in nance, and the annual activity level of the audit committee were shown to have à positive effect on the IRQ. On the contrary, this study highlighted that the percentage of independent members of the audit committee does not affect the IRQ. Declarations