The Impact of Audit Committee on Financial Performance of Insurance Firms in Nigeria

This study examined the effect of audit committee size on the financial performance of insurance companies in Nigeria between the year 2004 and 2015. The study used secondary data obtained from the annual report, National Insurance Commission Facts Books and Nigeria Stock Exchange Facts Books of fifteen selected insurance companies listed on the Nigeria stock exchange. The collected data were analyzed using descriptive statistics and regression analysis. The result revealed a significant negative correlation between audit committee and financial performance. The study recommended that the regulatory authorities focus more on other appropriate measure like competence (financial expert in committee) and independence of the committee that will ensure check and balance framework in the audit exercise and hence positive performance.

for a large size audit committee members with a diverse professional judgment and experience but not to unmanageable size. However, small audit committees that is not large enough e.g with one or two members could be regarded as weak because it will be easy for management to exert pressure on them to gain its support when there is any dispute with the auditor.. Empirical studies show mixed result on the effect of audit committee size on performance. Xie, Davidson and DaDalt (2003) and Bédard et al. (2004) find no significant association between audit committee size, measured by the number of directors on the committee, and performance management. This is also supported by the study of Abbott, Park and Parker (2004), which find negative relationship between audit committee size and reporting quality. However, Abdul Rahman and Ali (2006) study the extent of the effectiveness of the audit committee size in reducing earnings management of Malaysian quoted companies. Their study reveals positive association between audit committee size and performance management. This indicates that a certain minimum number of audit committee members may be vital to firm performance. Carcello et al. (2002) study found gap between what audit committees say they are doing and what their charter mandated. Even though this gap may be due to numerous reasons including liability concerns, it raises the general issue of transparency and accountability with respect to activities of the audit committee.

Audit Committee and Corporate Governance Disclosure
Study relating to importance of Audit Committee independence was conducted by DeZoort and Salterio (2001) with the result revealing greater support for auditor in an accounting dispute case with increased independent director experience and audit knowledge using 68 audit committees in their sample. They found out that there is no affect for level of accounting knowledge. DeZoort and Salterio (2001) examined the effects of Independence director experience; audit knowledge on Support for management or auditor and discovered that there is a better support for auditor in an accounting disputes situation with increased independent director experience and audit knowledge. The latter finding may have been due to the non-technical, generic nature of the accounting issue at hand. Audit Committee's may be firmed primarily for cosmetic reasons to make it appear to outside stakeholders that the company desires monitoring of financial reporting and controls.
Fogarty and Kalbers (1998), using data from their earlier study in 1993 discovered weakness association between Agency Theory factors and Audit Committee effectiveness. Collier and Gregory (1999) used 142 UK companies listed on the London Stock Exchange with result showing little support for the findings of Menon and Williams (1994) that confirm positive relationship between the Big 6 and leverage and Audit Committee activity.
Survey of Haka and Chalos (1990) concluded that Audit committee chairs are consistently different from other groups on what should constitute full disclosure (with audit committee members wanting greater disclosure). Krishnamoorthy (2002), established in their investigation that it is important to distinguish between the "form" (meeting regulatory requirements) and the "substance," (the effectiveness) of audit committees. Management exerts a significant influence over the quality of the interactions between the audit committee and the external auditor.. Dechow, Sloan and Sweeney (1996), result shows that firms subject to AAERs (accounting and auditing enforcement releases) are more likely to influence earnings to achieve lower cost of external financing. Beasley (1996) study evaluates the effect of corporate governance mechanism on fraud and discovered that larger percentage of outside members on board reduces fraud likelihood.
In their subsequent study, Beasley, Carcello, Hermanson and Lapides (2000), found that fraud companies in all 3 industries studied are less likely to have independent audit committees, outsiders on the board and to have an internal audit function. Abbott et al. (2000) was able to conclude that Independence together with activity of the Audit Committee is associated with a lower incidence of AAER. Klein (2002) report of studies carried out on corporate governance mechanism and abnormal accrual in the books established negative relationship between board/audit committee independence and abnormal accruals. In their subsequent study, Abbott et.al. (2000) findings reveals that misstatement is affected by presence of a completely independent audit committee holding a minimum number of meetings while not significant are expertise and Characteristics of the Board of Directors . Wild (1996), result suggest that the audit committees provide a useful oversight mechanism for the financial reporting process and that this increased oversight results in improved earnings quality.
Abbott and Parker (2000) study concluded that Audit committees that are both independent and active are positively associated with selection of an industry specialist. Beasley and Petroni (2001) in their survey examining the percentage of outside directors on the board of 165 firms confirm that the likelihood that a specialist Big-6 auditor is selected increases with the percentage of outside directors on the board, but the outside board membership percentage has no impact on the choice between a non specialist Big-6 and non Big-6 Auditor.

Research Methodology 3.0.1 Study Population and sample
The population for this study consists of all the 35 insurance companies listed on the floor of Nigeria stock  Vol.10, No.14, 2019 exchange market out of which purposive sample was used in selecting the fifteen (15) listed insurance companies. These insurance companies were chosen because of their consistency on stock exchange market listing during the period of the study.

Data Gathering Method
Secondary data derived from the annual report of the insurance companies listed in the Nigerian Stock Exchange (NSE) during the twelve years period of 2004 and 2015 was used for the study with other related materials particularly the National Insurance Commission (NAICOM) and the Nigerian Stock Exchange database and Fact Books.

Model Specification
This study employed a modified version of the econometric model of Ntim and Ossei (2006). These models are as follows; MODEL 1 Relationship between audit committee and ROE MODEL 2 Relationship between audit committee and ROA (2) MODEL 3 Relationship between audit committee and Tobin's Q The a priori expectation is such that: BACTt BOSt, DEIt, CGDIt, and AUDCOMt > 0. A positive relationship is expected between explanatory variables (BOSt, DEIt, CGDIt, AUDCOMt and BACTt) and the dependent variables (ROE, ROA and Tobin's Q). The correlation coefficient (o) will help explaining the various levels of association between the independent variables.

Descriptions of Variables and Measurement
The variables used in the model are as described and measured below: ROA = Return On Asset. This is measured as the ratio of Earning Before Interest and Tax (EBIT) to Total Asset. ROE = Return on Equity. This is measured as the ratio of Earning Before Interest and Tax (EBIT) to Ordinary Shares. Tobin's Q = Market Value of Equity + Total Debt/Total Assets BACT = Number of board meetings held during a financial year BOS = Proportion of outside directors sitting on the board. DEI = Directors ordinary shares as a percentage of total outstanding shares of the firm CGDI = Ratio of total score of the Individual company to maximum Possible score obtainable by company. AUDCOM = Number of member in the committee.

Data Analysis Method
Panel data regression analysis methodology that combined time series and cross sectional data was used to measure the degree of association between disclosure and performance    Table 4.1 report the average values of all the variables both dependent and independent alike. The table above reported average value of 27% for return on equity, 6.5% for return on asset, 110% for Tobin's Q. Given the reported average values of the three dependent variables it stands that across the 15 insurance companies sampled in the study over the period of 12 years spanning from 2004 to 2015, the average ratio of earnings before interest rate to ordinary share (return on equity) is 27%, average ratio of earnings before interest and tax to total asset (return on asset) is 6.5% while the average ratio of the sum of market value of equity and total debt to total asset stood at110%. As reported in table 4.1 the average value of AUDCOM size was 94%. The reported means value of audit committee revealed that the average audit committee for the sampled insurance companies over the period understudied stood at 94% in size using the Company and allied matter decree of 1990 as amended of maximum of 6 members for a guide. The standard deviation stood at 13.9% which is a reflection of the average dispersion of the distribution of the distribution of observations corresponding to each of the variables from the centre. The minimum and maximum return on equity (ROE) across cross sectional unit over the period covered in the study stood at -141% and 278%. For return on asset (ROA) the minimum and maximum values stood at -79.13% and 109.9%, while for Tobin's Q the minimum and maximum statistics stood at 26.56% and 488.8% respectively. The reported minimum and maximum values of audit committee stood at 67% and 100%.  (2018) From the table above for model 1, it was reported that main explanatory variable of AUDCOM and control variables except DEI showed a negative coefficient. In the table above, the AUDCOM revealed coefficient of -1.228820 with corresponding probability of 0.0141. The result showed that our main variables of AUDCOM exert a negative and significant impact on financial performance measured in terms of ROE. Moreover, increasing the size of AUDCOM by one unit will lead to decrease in performance by -1.22820. The R 2 of 0.260470 implies that about 26% of the systematic variation in financial performance measured in terms of ROE is jointly explained by explanatory variables. The P-value < 0.05 revealed significant joint impact of explanatory variables on financial performance. AUDCOM has significant impact on ROE at probability value of 0.0141.  (2018) The table above revealed that main independent variable of AUDCOM and control variables except DEI again showed a negative coefficient. The AUDCOM revealed coefficient of -0.351431 with corresponding probability of 0.0464. The result showed that our main variables of AUDCOM exert a negative and significant impact on financial performance measured in terms of ROA. The result further revealed that an attempt to increase the size of AUDCOM by one unit will lead to decrease in performance by -0.351431. The R 2 of 0.350784 implies that about 35% of the variation in performance measured in terms of ROA is jointly explained by independent variables. The P-value < 0.05 revealed significant joint impact of explanatory variables on financial performance. AUDCOM has significant effect on ROA at probability value of 0.0464  (2018) The table above for model 3 reported that AUDCOM and control variables except CGDI and DEI showed a negative coefficient. In the table above, the AUDCOM revealed coefficient of -0.905279 with corresponding probability of 0.0317. The result showed a negative and significant relationship between AUDCOM and financial performance measured in terms of Tobin's Q Moreover, increasing the size of AUDCOM by one unit will lead to decrease in performance by -0.905279. The R 2 of 0.592971 implies that about 59% of the systematic variation in financial performance measured in terms of Tobin's Q is jointly explained by explanatory variables. The P-value < 0.05 revealed significant joint impact of explanatory variables on financial performance. AUDCOM has significant impact on Tobin's Q at probability value of 0.0317.

Summary and Conclusion
Audit committee reveals significantly negative impact on performance measured in terms of return on equity, return on asset and Tobin's Q. This finding is in congruence with previous empirical studies of. Xie et al. (2003) and Bédard et al. (2004) that find negative association between audit committee size, measured by the number of directors on the committee, and performance management. This is also supported by the study of Abbott et al. (2004) that find negative relationship between audit committee size and reporting quality. The finding however, disagreed with the result of Abdul Rahman and Ali (2006) that reveals positive association between audit committee size and performance management. This indicates that a certain number of audit committee members may be vital to firm performance.

5.0.1Recommendation
Since there is negative relationship between number of audit committee member and performance, it is necessary that the code of insurance corporate governance in Nigeria release by NAICOM look at compliance to other appropriate measure like competence (financial expert in committee) and independence of the committee that will ensure check and balance framework in the audit exercise and hence positive performance.