Audit Committee Experts and Mandatory Disclosure Transparency of Malaysian Government Linked Companies

Literature Review and Hypotheses Development

Transparency has been described as one of the approaches which is more direct and observable to assess the quality of financial reporting. The stakeholders and a firm would gain benefits from enhanced transparency as it will act as a panacea in eradicating the information gap between both parties. From agency theory perspective, audit committees serve as an intermediary between both parties of which effective aud committees would lead to transparency of mandatory disclosure (Eyenubo et al., 2017). The inclusion of audit committees seems beneficial as they possess two characteristics that allow them to perform their monitoring function effectively, namely expertise and independence (Islam et al., 2010).

However, it is critical to formulate the right combination of audit committee expertise as most of the studies simply measured the expertise as those who have working experience either in accounting or finance (see Salleh et al., 2017). Hence, the new measures on expertise have been introduced, namely academic qualification (see Nelson & Devi, 2013) and training (see Azmi et al., 2013). Regarding experience, a broader definition is employed by concentrating on accounting financial experts, finance experts and supervisory experts (see Kusnadi et al., 2016).

Population, Sample and Data Collection

A quantitative-positivist research philosophy has been adopted to answer the research questions. Data of this study is based on the GLCs listed on the Bursa Malaysia and involved in GLC Transformation Program (GLCT), which is known as G20. Currently, G20 is a group of 17 GLCs. After eliminating two delisted companies and five GLCs in the finance industry, the final sample of 10 listed GLCs for the year 2009 until 2017 are analysed. The panel data technique is applied, resulting a total of 90 firm-year observations. In the process of collecting data, this study used secondary data from the corporate reports, consisting of annual reports, integrated annual reports and sustainability reports.

Development of Mandatory Disclosure Index

The development of mandatory disclosure index begins with the selection of reliable mandatory items, by referring to the existing disclosure indices from the previous studies (see Akhtaruddin, 2005; Hossain, 2008) as well as the established checklist instrument from a Big-4 accounting firm (Ernst & Young and KPMG) and legal requirements from BMLR and Companies Act 2016. The selection of items covers the whole content of the corporate reports by considering the availability and relevancy of information. In total, a complete list of 110 mandatory items is identified. Table 01 shows the distribution of 110 mandatory items across the corporate reports.

The items of information are numerically scored on an unweighted and a dichotomous basis. The unweighted approach is employed based on the assumption that each item is considered equally important, thus given the same scores (Akhtaruddin, 2005). The items of information are numerically scored on a dichotomous basis either by assigning one (1) score if disclosed in the corporate reports, or zero (0) score if otherwise. If a particular item is irrelevant to a sample company, a non-applicable (NA) score is given. Next, disclosure scores for each sample firm is calculated as follows and it is expressed in ratio:

$$\mathit{M}\mathit{A}\mathit{D}\mathit{T}\mathit{R}\mathit{A}\mathit{}=\mathit{}\mathit{A}\mathit{C}\mathit{T}\mathit{D}\mathit{}/\mathit{}\mathit{M}\mathit{X}\mathit{T}\mathit{D}$$

Where:

MADTRA= The value of the mandatory disclosure transparency score for each firm.

ACTD= The number of items each firm actually disclosed.

MXTD= The expected maximum number of items disclosed by each firm.

Regression Model and Variables Defined

The Ordinary Least Squares (OLS) regression model is employed to investigate the relationship between audit committee experts and mandatory disclosure transparency. The following regression equation is estimated for this study:

$$MADTRA=\beta 0+\beta 1ACCEX+\beta 2FINEX+\beta 3SUPEX+\beta 4ADVAQ+\beta 5TRAIN+\beta 6INDEP+\beta 7ACSIZ+\beta 8TNURE+\beta 9FIRPE+\beta 10FILEV+\beta 11FISIZ+\epsilon$$

Table 02 summarises all variables used in the model above.

Descriptive Statistics and Correlations

Table 03 displays the descriptive statistics for all variables. The results for mandatory disclosure transparency score indicate the highest score achieved by a firm is 97.3% and the lowest score is 86.9%. The mean of mandatory disclosure transparency score is reasonably high, which is 93.4% out of the 110 items. In spite of this achievement, this value poses the question as to why the GLCs’ mandatory disclosure transparency score is yet to achieve a full score, where Malaysia is apparently under a regulated reporting environment. Several clarifications are possible. First, it may be due to inadequate enforcement mechanisms (Abdullah et al., 2013). Second, some companies may have disclosed items by using other media such as company’s website and/or press releases, where these media do not form part of this study’s data collection.

On average, 43.0% and 20.0% of audit committee members have accounting financial expertise and finance expertise, respectively. The results reflect that the GLCs comply with the requirements of the BMLR and revised MCCG on having audit committee member with relevant accounting or finance qualifications and expertise. Meanwhile, the average score for the supervisory experts is 56.0%, which stipulates that a significant number of audit committee members have supervisory expertise. Next, the mean value of 34.6 % for advanced academic qualification asserts that the audit committee have either a master’s degree or doctorates. For training, the average number of trainings attended is eight, thereby indicating that the GLCs follow the revised MCCG’s recommendations on the need to undertake continuous training.

For control variables, the vast majority of the sample firms is composed with independent directors, with the mean value of 81.0%. The average of audit committee size is 4 members, ranging from 3 to 6 members. The average audit committee tenure is around 6 years, thus fulfilling the revised MCCG’s recommendation to have a cumulative term of up to 9 years. Next, the mean value for the firm performance is 5.2%, ranging from -13.4% to 15.9%. The financial leverage ranged from 22.5% to 77.9%, averaging 54.4%, whilst the firm size which calculated by the natural log of total assets, it ranges from 8.94 to 11.45 with an average score of 10.15.

Table 04 presents the results of the correlation analysis. The MADTRA is positively and significantly correlated with four variables p < 0.01 (ACCEX, FINEX, SUPEX, FISIZ). Four variables are positively correlated with MADTRA, but not significant p > 0.05 (ADVAQ, ACSIZ, FIRPE, FILEV). Meanwhile, three variables (TRAIN, INDEP, TNURE) have a negative association with MADTRA.

Regression Results

Table 05 presents the results of OLS regression analysis examining on the association between audit committee experts and mandatory disclosure transparency. It is apparent that the F-value is 14.218 (P=0.000), which indicates that the model estimate fits the data at an acceptable level. The R-square of 66.7% and the adjusted R-square of 62.0% suggest the variations in mandatory disclosure transparency score explained by all variables are quite moderate.

As the study is concerned with the individual effect of explanatory variables on the mandatory disclosure transparency, the presence of multicollinearity is observed by using the correlation matrix and variation inflation factor (VIF). Based on the Pearson correlation analysis in Table 04 , none of the variables are found to be exceeding the critical level of 0.90 (Appuhami & Tashakor, 2017). Likewise, the VIF scores as depicted in Table 05 are found to be below the threshold value of 10 (Madi et al., 2014). Thus, these results support the lack of presence of multicollinearity. Consequently, the results from the regression analysis can be interpreted with a high degree of confidence.

In respect of expertise variables, the results show that the audit committee accounting financial experts is significantly and positively associated with mandatory disclosure transparency. Hence, the H1 is supported. This is in line with the findings by Kusnadi et al. (2016). In this respect, it can be argued that the audit committees are effective in recognising potentially misleading transactions and detecting anomalies in corporate reporting, which could enhance the corporate transparency. The findings also indicate that the audit committee finance experts is found to be significantly and positively correlated to the extent of mandatory disclosure transparency. Therefore, the H2 is supported. This result reinforces the findings from Ghafran and O’Sullivan (2017). Although finance experts normally have inadequate understanding of accounting matters, they may assist in deciding whether the accounting methods truly reflects the economic substance of transactions.

With regard to audit committee supervisory experts, the results show that it is positively associated with mandatory disclosure transparency. Thus, the H3 is supported. This result corroborates the findings in Ghafran and O’Sullivan (2017). The derivation of such expertise via experience on supervising financial reporting responsibilities and overseeing the performance of companies may motivate the corporate management to adhere with mandatory disclosure requirements. The regression results also indicate a significant and positive association between audit committee advanced academic qualification and mandatory disclosure transparency, thus supporting H4. This finding is in agreement with Yasin and Nelson (2012). In detail, the result is consistent with the agency theory argument that this expertise is an effective corporate governance mechanism that could enhance mandatory disclosure transparency.

On the other hand, the audit committee training is significantly associated with mandatory disclosure transparency, but the sign of the coefficient is not as predicted. Hence the H5 could not be supported. In terms of the control variables, the statistical result shows that audit committee tenure, firm financial leverage and firm size have significant association with mandatory disclosure transparency. Meanwhile, audit committee independence, audit committee size and firm performance show an insignificant influence on mandatory disclosure transparency.