Corporate Governance Quality and Audit Quality in Malaysia

This paper examines the impact of corporate governance quality on audit quality in Malaysia. The sampling frame is 457 Malaysian non-finance listed companies, over the periods 2003 to 2007 (pre-2007 Code period) and 2008 to 2012 (post -2007 Code period), consisting of 2,285 observations for each period. This study uses pooled ordinary least square (OLS) to test the research hypotheses and model. The results show that the effectiveness of the audit committee (AC) has no significant influence on audit fees in the preand post-2007 Code period, and the effectiveness of the board has no significant influence on audit fees in the pre-2007 Code period, although it has significant influence on audit fees in the post-2007 Code period. The results suggest that the existing corporate governance framework in relation to AC has limitation in its governance role on audit process. Our study contributes to existing literature conducted in the US, the UK and Australia where their institutional settings are different from that of Malaysia. In addition, our study is based on the 2007 Code’s recommendation which contributes to the previous research conducted in Malaysia and provides an insightful evidence to the regulator on the corporate governance regime in Malaysia.


Introduction
In Malaysia, the board of directors (board) of public-listed companies have a fiduciary duty to act in the interest of the company.The board is required to establish an independent audit committee (AC) to implement and support the oversight function of the board.The governance role of the AC is to ensure that the interest of the shareholders are properly protected through the oversight of financial reporting and external audit processes.The governance role of the board and AC are governed by the Listing Requirements of Bursa Malaysia Securities Berhad (previously known as Listing Requirements of Kuala Lumpur Stock Exchange (KLSE)) [1,2], [3][4][5] and Malaysian Code on Corporate Governance [6,7].
In the past two decades, high-profile financial cases pertaining to corporate governance scandals continuously plagued in Malaysia, such as Perwaja Steel Bhd, Aokam Perdana Bhd, Idris Hydraulic (M) Bhd, Transmile Group Berhad, Axis Incorporation Bhd, Megan Media Holdings Berhad, Linear Corp Bhd, Silver Bird Group Bhd and Malaysia Airline Systems [8,9].These cases have caused a loss of public confidence in the role of auditors in preventing fraud.In addition, investors had lost confidence in the Malaysia market during the 1997/1998 Asian Financial Crisis [10].Consequently, the corporate governance framework in Malaysia was reformed.The regulators, the public and investors are now closely scrutinised on corporate governance in Malaysia.
A number of studies have examined the relationship between corporate governance quality and audit quality, particularly on the governance role of the board and AC, in relation to their oversight of the external audit process.These studies are based on settings in the US [11,12,13], the UK [14,15,16], Australia [17] and Malaysia [18,19,20].
Previous studies conducted in the US, the UK and Australia did not provide conclusive evidence on the relationship between audit fees and the effectiveness of the board and AC.[15] and [12] found that audit committee independence has a significant and positive relationship with audit fees.However, this finding contrasted that of [11,14,17].The results found in [12] on the relationship between audit committee expertise and audit fees also contrasted with the findings of [13,16,15,17].Similarly, [15] and [17] found that the frequency of audit committee meetings has a significant and positive relationship with audit fees did not support the findings of [11] and [12].In the study of [12], there was no evidence to show that board independence has a significant relationship with audit fees; this result was not supported by [13,15,17].The differences opinions in the US, the UK and Australia could be due to the country variations of corporate governance requirements.
Previous studies in Malaysia have shown that the effectiveness of the board and AC variables were tested separately [18,19,20], with data prior to the implementation of the 2007 Code [18,19] and the study period covering a short time frame, i.e. three years or less [18,19,20].Furthermore, the findings of these studies on the relationships between the audit fees and the effectiveness of the board and AC were inconclusive.[18] found that board independence, audit committee expertise and the frequency of audit committee meetings are significant and positively related to audit fees.This findings contrasted that of [19] and [20] who found board independence and frequency of audit committee meetings have no significant relationship with audit fees.
Our study examines the relationship between corporate governance quality and audit quality.We use the effectiveness of the board and AC as proxies of corporate governance quality and audit fees paid to the external auditor as the proxy of audit quality.The effectiveness of the board and AC are the best practices in corporate governance of the Code, the 2007 Code and corporate governance guide.The sample consists of 457 Malaysian non-finance public-listed companies which had traded their shares on the main board of Bursa Malaysia from 2003 to 2012.We divided the ten-year data into two periods, i.e., pre-2007 Code period and post-2007 Code period, to test the relationship between the audit fees and the effectiveness of the board and AC before and after the introduction of the 2007 Code.
Our study shows that the effectiveness of the board and AC have no significant influence on audit fees in the pre-2007 Code period.In the post-2007 Code period, the effectiveness of the board is significant and positively related to audit fees; however, there is no evidence to show that the effectiveness of AC has significant influence on audit fees in the post-2007 Code period.This suggests that the existing corporate governance framework particularly on the board has an influence on the quality of audit process, but corporate governance framework in relation to audit committee has limitation in its governance role on audit process.Our results partially support the regulatory initiatives intention at enhancing role and responsibilities of the board and AC in order to improve the audit process.
Nevertheless, this study contributes to the literature in two ways.Firstly, our data is based on the recommendations of the Code, the 2007 Code and corporate governance guide with a longer time-frame study period from 2003 to 2012 which enabled us to provide further empirical evidence that lending support to the latest corporate governance principles and best practices in enhancing the audit process.And secondly, our findings provide an insight evidence to the regulators, policy-makers, industry players and investors which shows that the principles and best practices of 2007 Code have its limitations in enhancing the audit process, alerting that such limitations need to be addressed in order to provide and sustain a sound corporate governance mechanism.
The next section of this paper provides the hypotheses development followed by the research methods, results and discussion.The last session of this paper provides a conclusion of the study.

Hypotheses development
Audit committee is viewed as a monitoring mechanism and serves to support the oversight function of the board.There are studies based on agency theory which suggested that the board that emphasis a high quality of monitoring will demand a higher quality audit resulting in greater audit effort by the auditor and resulting higher fees [15,27].This study used the effectiveness of audit committee and the effectiveness of board as proxies for corporate governance quality.

The effectiveness of audit committees (ACE) and audit fees (AF)
A large number of international and national studies in previous years have addressed whether ACs are effective in terms of providing better quality of audit or improved internal control structure.Empirical evidence documented conflicting results on the impact of the effectiveness of the board and AC on the audit fees and the direction of the association between these variables.Consistent with previous studies [18,15,20], and based on the best practices in corporate governance of the Code, the 2007 Code and the corporate governance guide, we claim that audit committee independence, frequency of audit committee meetings, audit committee with financial expertise and audit committee size contribute to the effectiveness of AC.
An independent AC demands a wider scope of audit from the external auditors and willing to support external auditors during scope intervention with management [12].Audit committee with financial expertise has a better understanding on the auditing issues, risks, and the audit procedures that auditors propose to address auditing issues and risks [21,22,12].ACs who meet regularly have a better understanding of current auditing issues and have a positive influence on the scope of audit at different stages of the audit [12].Company with larger AC is able to enhance its status and increased resources which may increase its effectiveness in fulfilling its monitoring role [15].An effective AC in assuring the quality of audit process demands additional audit procedures from the external auditor, therefore increasing audit fees [14,12,15].Based on these observations, we posit that an effective AC demands wider scope of audit from the external auditor, leading to a higher audit fees paid to the external auditor.Therefore, we offer our first hypothesis as follows: H 1a .Companies with an effective AC are likely to be charged with higher audit fees in pre-2007 Code period.H 1b .Companies with an effective AC are likely to be charged with higher audit fees in post-2007 Code period.To be truly effective, a board needs more independent [11,23] and diligent [11] directors on the board.Previous studies have found that a higher percentage of independent directors on the board helps to reduce the frequency of fraudulent financial reporting [24] and lower the occurrence of earnings overstatement [25].[26] suggested that an effective board meets frequently.The board that meets frequently shows better diligence in discharging its responsibilities and improving its governance role of monitoring the financial reporting process [11].Consistent with the best practices in corporate governance that recommended in the 2007 Code, we also included the board duality as one of the measurement for an effective board [7].Board duality requires higher control of risk and higher audit effort and leading higher audit fees paid to external auditor [15].

The effectiveness of the board (BOD_E) and AF
Our study is based on the previous studies [11,23], the best practices in corporate governance of the Code, the 2007 Code and the corporate governance guide to determine the effectiveness of the board (Table 2).Following the discussion above, we posit that an effective board will demand additional audit procedures from the external auditor to improve the financial reporting process, thus increasing the amount of audit fees paid to the external auditors.Therefore, we tender the following hypothesis: H 2a .Companies with an effective board are likely to be charged with higher audit fees in pre-2007 Code period.H 2b .Companies with an effective board are likely to be charged with higher audit fees in post-2007 Code period.

Sample
The sample consists of the Malaysian non-finance listed companies that had traded their shares on the main board of Bursa Malaysia for the years from 2003 to 2012.These companies are selected because they have gone through the changes of Malaysia corporate governance framework in 2000 (The Code) and 2007 (2007 Code).We split the study periods into two, i.e., 2003 to 2007 (pre-2007 Code period) and 2008 to 2012 (post-2007 Code period).In the earlier period, corporate governance practices in Malaysia were based on principles and best practices of the Code and in the latter period were based on the principles and best practices of the 2007 Code.Consistent with previous studies, we have excluded finance listed companies in this study as they are governed by a different regulatory body [27,28].All data were hand collected from the annual financial statements available in the website of Bursa Malaysia.The sample size is 457 companies after further eliminating the companies with missing data.Table 1 provides the sample size and number of firm-year observations by industry.

Dependent variable
Consistent with recent studies on audit fees [13,19,15], our dependent variable is the statutory audit fees paid to external auditors, measured as the natural logarithm of audit fees (lnAF).

Independent variables
ACE and BOD_E are the main independent variables in this study.ACE and BOD_E are measured based on the best practices in corporate governance of the Code, the 2007 Code and the corporate governance guide.The best practices in corporate governance in relation to AC include audit committee independence, frequency of audit committee meetings, audit committee with financial expertise and audit committee size.ACE is a dichotomous variable with 1 denoting (a) all AC members are non-executive directors with majority of independent directors, (b) at least one member is a member of accounting association or body, (c) at least three members in AC [6,7], and (d) AC members meet at least four times per year [29] and 0 otherwise.The best practices in corporate governance pertaining to the board consist of board duality, board independence and frequency of board meetings.BOD_E is a dichotomous variable equal to 1 when (a) the roles of chairman and chief executive officer (CEO) are not combined, (b) at least one-third of the board membership are independent non-executive directors [6,7], and (c) board members meet at least five times per year [29] and 0 otherwise.

Control variables
In addition to the two independent variables, we control for the effects of other variables on audit fees that were found in previous studies [15,30].The control variables include LOSS, lnTA, LEV, BIG4, lnSUBs, NSH5_S.LOSS is an indicator variable equal to 1 if the company is making loss during the year and 0 otherwise.lnTA is measured as the natural logarithm of total assets.LEV is the leverage level of the company, measured by the ratio of long-term debts to total assets.BIG4 is an indicator variable equal to 1 if the company is audited by big-four auditor and 0 otherwise.lnSUBs represents the complexity of the company, measured as the natural logarithm of number of subsidiaries.NSH5_S is the total number of shareholders holding 5% or more shares in the company.We also control for the director's remuneration (DR) on audit fees as the 2007 Code recommends that public-listed companies pay sufficient remuneration to attract and retain directors [7].Hence, we expect director's remuneration as one of the determinants of audit fees.DR is measured as the natural logarithm of remuneration paid to the directors.Consistent with previous studies [28,15], we include year and industry dummies to control for the unobserved effects during the sample period and the industry effect respectively.

Regression model
The regression model use to test the hypotheses in this study is as follows:

Correlations
Panel A and Panel B of Table 3 report on the correlation matrix for the relevant variables for the pre-2007 Code period and post-2007 Code period respectively.There is no multicollinearity problem with other experimental variables and control variables as the correlations are all below 0.7 [31].We also calculate the variance inflation factor (vif) for all variables and the results show that all vif values are less than 10.0 suggesting that multicollinearity is not a substantive issue in this study [31,32].

Regression results
Table 4 presents the OLS regression result of lnAF on ACE, BOD_E and control variables.
The results show significant explanatory power with R 2 of 0.762 in the pre-2007 Code period and 0.764 in the post-2007 Code period.Table 4 shows that in the pre-2007 Code period and post-2007 Code period, ACE is not significantly related to lnAF.This result is in contrast with our prediction that an effective AC demands wider scope of audit from the external auditor, leading to higher audit fees paid by the company.Thus, we have to reject the hypothesis H 1a and H 1b .The result shows that ACE has no influence on lnAF.BOD_E is not significantly related to lnAF in the pre-2007 Code period.Therefore, we have to reject hypothesis H 2a .BOD_E is significantly and positively related to lnAF at 5 percent level in the post-2007 Code period.This result in the post-2007 Code period is consistent with our prediction that an effective board demands higher quality of audit services to ensure reliability of financial statements, to complement its own monitoring managerial behavior and resulting in higher audit fees paid by the company.We do not reject the hypothesis H 2b in post-2007 Code period.This suggests that only BOD_E has a significant influence on lnAF in the post-2007 Code period.The OLS regression result of lnAF on control variables shows that, in the pre-2007 Code period and post 2007 Code period, lnDR, lnTA, lnSUBs are significant and positively related to audit fees at 1 percent level.LEV shows a statistically significant positive relationship with audit fees at 10 percent level in the pre-2007 Code period.In the post-2007 Code period, LOSS and BIG4 are significant and positively related to audit fees at 1 percent level.The result in this study also shows that there are both industry effect and time effect in the research model.Yes Yes Note: Variables are as defined in Table 2; *, **, *** significant at 10, 5 and 1 percent levels.

Discussion
In this study, the results show that ACE has no significant influence on lnAF in the pre-and post-2007 Code period is in contrast with the result of [14,12,15].These studies had argued that an effective AC demands addition audit procedures in order to assure the quality of audit process, leading to an increase in audit fees, is not supported by the results of our study.The results also show that BOD_E has no significant impact on lnAF in the pre-2007 Code period.The finding of BOD_E has a significant influence on lnAF in the post-2007 Code is consistent with the US study of [11,13] and the UK study of [15] who found that an effective board helps to reduce the frequency of fraudulent financial reporting [24], lower the occurrence of earnings overstatement [25], improve its governance role of monitoring the financial reporting process [11] and enhance the quality of audit and financial reporting process [23] and therefore will demand more audit procedures from the auditors, leading to an increase in audit fees.
In this study, we also find that lnTA, lnSUBs, LEV have a significant influence on lnAF in the pre-2007 Code period, lending support to the findings by [19,15,18,15] had argued that companies with large total assets have more complex operations; companies with more subsidiaries have complex group transactions and weak internal control; companies with higher leverage will demand higher audit quality to protect themselves from business and finance risk and therefore increase the audit fees paid to the external auditor.In the post-2007 Code period, lnTA, lnSUBs, LOSS and BIG4 have significant influenced on lnAF.These results also provide support to the findings by [19,15].Consistent with our expectation, lnDR has a statistically significant and positive relationship with lnAF in the pre-and post-2007 Code.This suggests that lnDR is an important determinant of audit fees.

Conclusion
Our study has provided an overview of the impact of the corporate governance quality on audit quality in Malaysia during the period from 2003 to 2012 (i.e., pre-and post-2007 Code period), using the effectiveness of the board and AC to proxy for corporate governance quality and audit fees to proxy for audit quality.The timeframe is interesting and appropriate as the Code was revised in 2007 and the 2007 Code was aimed at strengthening the roles and responsibilities of the board and AC.The sample has focused on 457 nonfinancial companies that had listed and traded their shares on the main board of Bursa Malaysia from 2003 to 2012.There is no evidence to show that the effectiveness of AC has a significant influence on audit fee in the pre-and post-2007 Code period.The results also show that the effectiveness of the board does not has a significant influence on audit fees in the pre-2007 Code period but it has significant influenced on the audit fees in the post-2007 Code period.This suggests that the existing corporate governance framework particularly on the board has an influence on the quality of audit process, but the corporate governance framework in relation to audit committee has limitation in its governance role on audit process.Our results partially support the regulatory initiatives intention at enhancing role and responsibilities of the board and AC in order to improve the audit process.This study is subject to a number of limitations that need to be considered in future research.Firstly, the use quantitative research method and secondary data to explore the effectiveness of the board and AC may not reflect the true performance of the board and AC.We suggest to use qualitative research method to explore the performance of the board and AC on governance outcomes by including other attributes such as the board and AC's individual behavior, informal networks of the board and AC, the relationship among the board, AC and external auditors and the supervisory and compliance structure of the company.Secondly, our study did not make allowance for the political connections and ethnicity documented in previous studies conducted in Malaysia; this should be considered for future study.

Table 2
displays the definition of variables in the regression model.We use the ordinary least squares (OLS) regression model to explore the relationship between the audit fees and the effectiveness of the board and AC.

Table 2 .
Definition of Variables

Table 3 .
Pearson Correlation Matrix

Table 4 .
Regression of lnAF on ACE, BOD_E and Control Variables