The Effect of Corporate Governance Mechanism on the Quality of Earnings Among Nigeria Money Banks in Nigeria free essay sample

This paper examines whether corporate governance mechanism variable – Board Size, Board Composition, Ownership Concentration, Institutional Shareholders, Dividend Payment, Firm Size have significant impact on the quality of earnings of Nigerian deposit money banks as measured by modified (McNicols and Wilson, 1998), (Gred and Clarke, 2004) and (Chang, 2008) model of specific industry discretionary accruals as against (Dichow and Dichev, 2002), though widely accepted but is hardly industry specific. Secondary data are extracted from the annual reports of 15 banks that form the sample of the study within the period between 2006-2011. Multiple regression was used as a tool for analysis. The result reveals that corporate governance mechanisms affects earnings quality of Nigerian money deposits banks. All the corporate governance examined are positive except for the control variable firm size signifying that none of the explanatory variables is inversely related to quality of earnings amongst Nigerian money deposits banks. It is therefore recommended that amongst others that shareholders of Nigerian DMBs to ensure the inclusion of about 50% outside directors in the board and ensuring a good quantum of both institutional and block holders in the equity holdings of the banks. We will write a custom essay sample on The Effect of Corporate Governance Mechanism on the Quality of Earnings Among Nigeria Money Banks in Nigeria or any similar topic specifically for you Do Not WasteYour Time HIRE WRITER Only 13.90 / page Keywords: Earnings, Discretionary, accruals, manipulation, monitoring, quality. Introduction The rising number of corporate failures, scandals and crises such as Enron, WorldCom, Global Crossing, HIH Insurance, Ansett, Pan Pharmaceuticals, Lever Brothers, Cadbury, and Afri bank has precipitated the growing interest on the governance structures of firms by academics, practitioners, the investment community, regulatory agencies, policy makers, national and multilateral government bodies and host of other stakeholders (Tsegba, 2011). Corporate governance is about building credibility, ensuring transparency and accountability as well as maintaining an effective channel of information disclosure that would foster good corporate performance (Matama, 2008). Income smoothing and earnings quality popularly called earnings management can be regarded as two of the attractive and challenging issues in studies related to accounting because investors pay attention to amount of income as an important factor in decision making. It has long been recognized that financial statements play an important role in assessing managers’ performance by the board of directors outside investors and external regulators. It is therefore, not unlikely that managers will manipulate financial reports in order to produce a good image of themselves and the firms that they manage (Shehu and Abubakar, 2012). There are a lot of qualitative empirical studies exist on the relationship between corporate governance mechanisms and earnings quality. However, quantitative studies supporting the existence of a link between corporate governance mechanisms and earnings quality are relatively scanty and inconclusive. Besides the scanty nature of quantitative literature, most of the existing ones are more concerned about the overall quality of corporate governance mechanisms rather than particular features or practices of such governance. In addition, most of the studies are cross-sectional in nature. There is none of these studies that examine either the overall or particular features of corporate governance mechanisms and earnings quality of the banking industry in isolation. The peculiar and sensitive nature of the banking industry as well as the reforms it has continued to undergo indicates the need for special attention. This study attempts to address that omission by examining the effect of corporate governance mechanisms on the earnings quality of Nigerian money deposit banks. The relationship between corporate governance mechanisms and earnings quality is important to the extent that good corporate governance builds confidence in the minds of existing and potential investors as well as other stakeholders of a bank. This in turn creates confidence in the banking industry. The implications of these on the economy as a whole are also obvious. Economic growth will be more sustainable, capital market will be boosted and become more developed and an egalitarian and corrupt-free society will be built. All these are essential for sustained economic growth and development. This study also contributes to the growing body of quantitative research on corporate governance and earnings management. The spate of well publicized corporate failures around the world call for high-quality financial reporting, which gives a proper record of stewardship, provides details of real costs of services and in which the informed individual can have confidence. Moreover, the recent number game amongst banks in Nigeria raises additional concern on the need for financial reports that meet today’s requirements (Bello, 2005). Earnings management involves the manipulation of earnings by companies using financial statement elements that are largely at the discretion of the managers to achieve divergent personal goals. These elements are peculiar to industries depending on their nature of operations and external regulatory framework. Researchers such as (Dockery and Herbert, 2000); (La Porta, Lopez-de-Silanes, (Yakasai, 2001); (Detomasi, 2002); (Fort and Schipani, 2003); (Bai, Liu, Lu, Song and Zhang, 2005); (Barako and Tower, 2006); (Achua, 2007); (Okike, 2007) and (Shehu and Abubakar, 2012) have identified that accruals arising from depreciation are used to manipulate earnings in manufacturing companies, claim loss reserve in insurance and loan loss provision in banking. Loan losses has been identified as one of the most important factors that lead to bank failures and its provision has a direct impact on reported earnings (Grey and Clarke, 2004). The issue of earnings quality arises because financial reports may incorporate adverse information about future cash flows in a more or less timely fashion (Francis et al, 2003). According to (Ball and Shivakumar, 2002), earnings quality is about timely loss recognition that requires estimates of future cash flows from assets or outflows for liabilities. In (Basu’s, 1997) view, accountants have tendency to require a higher degree of verification to recognize good news as gains than to recognize bad news as losses. He consequently relates earnings quality to the accounting concept of conservatism, which supposes that earnings reflect ‘bad news’ more quickly than ‘good news’. In theory, one measure of earnings quality is the relation between current accruals and cash flow (Jindrichovska and Kuo, 2000). Thus, prior researches document an association between earnings pattern and earnings quality (Hunt et al, 2000; Francis et al, 2003). Relating this measure to banking, earnings quality becomes a function of DLLP and earnings pattern. Accordingly, banks with abnormal DLLP are considered as having low quality earnings while banks with normal DLLP are deemed to have high earnings quality. Incidentally, there also, exists an inverse relation between smooth and increasing earning pattern and earnings quality: the higher the smooth earning pattern the lower the earnings quality (Francis et al, 2003). Shehu and Abubakar, 2011) opined that loan loss provision is an expense on the income statement which signifies managers’ assessment of expected future losses. This means that an increase in loan loss provision reduces net income, while a fall in loan losses increase net income. Since it is the result of managers’ assessment of the likely loss that the company would incure should the borrower fail to repay his obligations as at when due, the provision for it is considered to have two (2) p ortions: non-discretionary and discretionary portions. Non-discretionary is a function of specific quality determinants in the loan portfolio- non-accrual loans, renegotiated loans, loans past due over 90 days, specific analyses on troubled large credits, usually implying internal grading system. This means that the non-discretionary portion is the provision that is based on fair and objective analyses of the firm’s economic conditions. While the discretionary portion are those accruals that largely depend on the outcome of the managers’ future expectations of uncertain events. The components of it are both quantitative and qualitative. Grey and Clarke (2004) pointed that the qualitative components include political, economic, geographical and political factors, while the quantitative are statistical analyses of loans not individually analyzed for special reserve and therefore are largely at the discretion of managers. The reasons why banks manipulate earnings are supported by three arguments of signaling argument, income smoothing or earnings quality argument and capital management argument (Zhou and Chen, 2004). The signaling argument suggests that banks use discretionary loan loss provision to insinuate that earnings will be high in subsequent periods (Wahlen, 1994: Liu and Ryan, 1995: Beaver and Engel, 1996). Contrary to the signaling argument, earnings quality argument holds that managers increase the provision for loan losses in periods when earnings are high, under the assumption of income smoothing (Beatty, Chamberlain amp; Moglio, 1995: Collins, Shacklford amp; Wahlen, 1995: Rivard, Bland amp; Morris, 2003). This implies that earnings quality in this area improves bank’s cash flows, capital adequacy, market value and overall performance. While the capital management argument suggests that since increase in loan loss provision increases regulatory capital, management exercises discretion over its provision (Ahmed, Takeda amp; Thomas 1999, Beatty et al. , 1995). Regardless of the industry and the strings attached, managers’ discretionary behavior to achieve personal gains undermines the shareholders’ wealth maximization objective of the firm. Consequently, therefore, this paper examines the influence of corporate governance mechanisms on the quality of earnings among Nigerian Deposit Money Banks (DMBs). In order to achieve this, it is posited that corporate governance mechanisms- board size, board composition, ownership concentration; audit committee, institutional shareholding and dividend have no significant impact on quality of earnings among Nigerian DMBs. It is the constant fear in the banking industry in spite of introduction of new code of corporate governance to enhance the efficiency of the industry practices, and the recognition of the use of discretion by bank managers as well as earnings manipulation that make this work apt and imperative. This study contributes to the sparse of literature that studied the relationship between corporate governance and earnings quality. It also extended to the financial firms by raising and discussing issues on corporate governance mechanism and earnings quality of banks using discretionary loans loss provision in the circumstance of emerging economies like Nigeria. The paper is structured as follows. Section two reviews related literature on corporate governance mechanisms and earnings quality and theoretical framework. Section three is methodology and model specification. In section four, the results and findings of the study are presented and discussed. Finally, section five deals with conclusion and recommendations. 2. 1 Literature Review and Theoretical Framework Literature on corporate governance and quality of earnings is reviewed. Specifically the study concentrates on governance mechanisms of dividend, audit committee, board size, board composition, ownership concentration and institutional shareholding. The theoretical framework that underpins the study is then presented and supported. Earnings quality as the altering of financial statements through the use of judgment in structuring transactions to either mislead the firm’s stakeholders about the true economic picture of the firm or to achieve some contractual benefit that is based on accounting numbers (Healy and Wahlen, 1999). (Schipper, 1989), opined that earnings quality is the deliberate intervention in financial reporting process to achieve personal goals. Earnings quality is the manipulation of financial statement by managers, using accounting choices, estimates and methods, to achieve some objectives that are largely in conflict with the underlying economic status of the firm. Different incentives to manage earnings are widely discussed in the literature, (Bhat, 1996), linked it to the attempt to enhance shareholders’ value and to maximize executive compensation through income smoothing and earnings quality respectively. Income smoothing, occasional big bath, living for today and maximization of variability are identified by (Koch amp; Wall, 2000). (Chang et al. 008) note three incentives to manage earnings: capital market motivation, which includes initial public offerings, seasoned equity offerings, management buoyant plans and plans for mergers to meet earnings forecast, to smooth earnings, management compensation motivation, debt agreement or job security and laws and regulations such as import regulation, antitrust laws, also can serve as incentives. Managers use discretionary accruals for opportunistic earnings quality (Cornet et al. 2009) by attaining some level of performance and affecting stock prices to enhance managers’ wealth through restricted stock returns. . 1. 1 Board Size and Earning Management There are a lot of empirical researches that have documented that board size is related to earning management. The evidence on the role of board size is inconclusive. Yermack (1996) and Eisenberg, Sundgren, and Wells (1998) demonstrate that smaller boards are associated with manipulative accounting. In the analysis of 131 different study-samples with a combined sample size of 20,620 observations (Dalton, Daily, Johnson and Ellstrand (1999) documented a positive and significant relation between board size and income smoothing. These conflicting results provide no conclusive relationship between earnings management and board size. A smaller board may be less encumbered with bureaucratic problems and may be more functional. Smaller boards may provide better financial reporting oversight. Alternately, a larger board may be able to draw from a broader range of experience. In the case of earnings management, a larger board may be more likely to have independent directors with corporate in financial experience. If so, a larger board might be better at preventing earnings management. Small boards are less effective monitors and are easier for CEOs to influence (Jensen, 1993) and (Lipton and Lorsch, 1992). 2. 1. 2 Board Composition and Earnings Quality Board independence simply refers to non-executive external directors, who do not represent any particular shareholder interest and hold no special business interest with the bank, relative to total number of directors on the board (Shehu and Abubakar, 2011). Studies on impact of board composition on earnings quality have produce varied results. Cornet et al. 2007), examine the impact of corporate governance and pay-for-performance on earnings quality. By means of 100 largest firms in the U. S. as ranked by Samp;P between 1994-2003, they find that the presence of independent outside directors reduce earnings quality. Aggregate accruals were used to proxy for earnings quality. (Cornet et al. , 2009), investigate how corporate governance mechanism affects earnings quality at large publicly traded U. S. companies for th e period between 1994 2002. Large independent boards constraint earning quality was found by the study. Roodposhti and Chashmi, 2010) find that for the period between 2004-2008 in Iran, using 196 firms listed on Tehran Stock Exchange, revealed a negative association between board independence and earnings quality. On the contrary, (Hashim and Devi, 2008) examine the relationship between board independence, CEO duality and accrual management in Malaysia. Using 200 top non-financial companies listed on Malaysian Stock Exchange, they find that large percentage of independent executive directors is associated with higher income-increasing earnings quality. Also, (Shah, Zafar and Durrani, 2009), study the relationship between board composition and earnings quality in Pakistani listed companies for the period between 2003 and 2007. They find no significant relationship between board composition and earnings quality. Yet, all studies exclude financial firms, the inclusion of which might have yielded different result. (Macey and O’Hara, 2003) opined that governance structure is industry specific and there is a systematic difference between the governance of different industries 2. 1. 3 Ownership Concentration and Earnings Quality Ownership concentration, which is also referred to as blockholders. It is the proportion of shares (usually more than 5%) owned by a certain percentage of shareholders. There are arguments that the higher the number of shares owned by the blockholders, the more the pressure on managers to act in conformity with shareholders interest (Sanda et al. , 2005). (Ramsy and Blair, 1993) opined that large ownership concentration has more incentives to manage earnings because the expected benefit from equity holding in the firm outweighs the cost associated with monitoring managers If this is true, then we expect ownership concentration to be inversely related to earnings quality. Some researchers observe that high ownership concentration beyond a certain level may lead to abuse of power, which could be detrimental to the value maximization goal of the firm (Sanda et al. , 2005). Inconsistent results were yielded on the relationship between ownership concentration and earnings quality. (Roodposhti and Chashmi, 2010) find a negative relationship between ownership concentration and earnings quality, while they used 196 firms listed on Tehran Stock Exchange as their sample for the period between 2004-2008, to examine the effect of board composition and ownership concentration on earnings quality. In the same vain, (Klai and Omri, 2011), investigate the impact of corporate governance on financial reporting quality in Tunisia. The study used 22 listed firms for the period between 1997-2007. They find that ownership concentration is negatively associated with earnings quality. Conversely, Using top 200 listed non-financial companies, (Hashim and Devi, 2008), examine the association between board independence, CEO duality and accruals management. They find that ownership concentration is associated with high income-increasing earnings quality. Besides the exclusion of financial firms from all the studies mentioned above, economic differences of nations calls for an investigation in of similar problems in an economy like ours. 2. 1. 4 Institutional Shareholding and Earnings Quality Institutional shareholders have both the incentive and power to compel managers to act in consonant with value maximization objective of the firm. (Shehu, 2011) note that institutional ownership has emerged particularly in the banking sector as a tool for protecting minority interest. We, therefore, expect that institutional shareholding and earnings quality will be inversely related. (Cornet et al. , 2007), investigate how governance structure and incentive based compensation influence firm performance when measured performance is adjusted for earnings quality. The study used top 100 firms rated by Samp;P in U. S. , they find that earnings quality is significantly reduced by institutional shareholders whether institutional shareholders is measured based on the proportion of shares owned by all institutional shareholders or by institutional involvement in the firm. This finding is an extension of (Klein, 2002). The study might have revealed different result if carried out in the Nigerian context. (Shehu, 2011) examine the interaction between corporate governance and financial reporting quality in deposit money banks. Using all 21 banks quoted on NSE for the period between 2007-2009, the study reveals a positive and significant relationship between institutional shareholding and earnings quality. The major drawback of this research is that it uses (Dichow and Dichev, 2002) model, which though is widely accepted but is hardly industry specific. The presence of institutional investors with substantial shareholdings restrain managers from engaging in income increasing discretionary accruals when companies have high free cash flow, however, when there is no free cash flow agency problems. 2. 1. 5 Dividend Payment and Earnings Quality While many studies ignore this variable as corporate governance mechanisms, this study consider dividend payment as a n important parameter that measures the overall efficiency of the board. A board that has high frequency of dividend declaration may force earnings managers to have less discretion in manipulating earnings. This is because higher earnings will attract high dividends leading to free cash flow. Larger free cash flow payout reduces managers’ ability to make bad investment (Jensen, 1986). Likewise, high payment obliges managers to raise additional capital via stock market there by being exposed to specialist, financial analyst, investment bankers, regulatory authorities and the press (Goergen, 2007). From these views, apart from the fact that high dividend is a signal to management effectiveness; it serves as a disciplinary mechanism in limiting management discretion over cash flow. In absence of priority we hypothesized that firms with high level of dividend history will have less level of unethical accounting practice that has to do with earnings misrepresentations (Bello, 2013). 2. 2 Theoretical Framework and Model Development In order to link corporate governance with earnings quality, the study first looks at the theories that induced earnings manipulations. Two prominent are opportunistic and desirous. The first theory which embedded the philosophy of this paper is opportunistic tendency of managers to engage in unethical in absence of good governance structure. Secondly, from corporate legal point of view board are to act as trustees of shareholders. Infact they are like operating shareholders directly overseeing the affairs of management. Agency theory, postulates an inevitable conflicts. Whereas, managers will be targeting better performance for short term gains, the interest of shareholders would be that of long term benefit of capital appreciation and return (dividend). Within the agency framework, it is both logical and inescapable that management behavior will be self serving (Amat, 1996). The end result will be that of managers manipulating earnings. These two theories; agency theory and opportunisms theory provide a complete framework for understanding corporate governance and earnings quality. 3. 1 Methodology The research design is inclined to use ex-post financial data because of its empiricism as well as practicality. The period of the study is six (6) years (2006 to 2011) both years inclusive. The period is considered more appealing because all DMBs for which relevant financial data from 2006 to 2011 is obtained would have survived the distress syndrome of the early 1990s and 2005 consolidation saga in Nigeria. We believe also that the six-year period would provide an adequate time series of data (observations) to realistically identify Nigerian DMBs that have been managing earnings over a number of years. This is consistent with Michelson et al (2000) idea that adequate time series studies captures incidence of smoothing, whereas one period studies reflect attempts to smooth. The data is extracted from the consolidated annual reports of the sampled banks sourced in the Nigerian Stock Exchange fact book 2010/2011. The population of the study is all the 25 DMBs that survive consolidation exercise as at 31st December 2006. The ample is drawn using criteria of complete data availability in which 15 banks automatically formed the sample of the study. Two steps Panel regression is used as tool of analysis because it satisfies our purpose of predicting and explaining relations between variables and also providing residuals of the LLP model to represent the explained variable in the second regress ion model. A pre-requisite for the use of any variant of earnings quality detection models is computation of the smoothing instrument. Theoretically and in practice, the smoothing instrument relevant to earnings quality studies in the banking industry is the LLP. Previous studies have investigated earnings quality instruments such as dividend income, changes in accounting policies, pension costs, extraordinary items, investment tax credit, depreciation and fixed charges, and many others (Kamarudin et al, 2001). However, (Kanagaretnam et al, 2001), specifically conclude that banks use LLP and charge-offs to smooth income. 3. 1. 1 Estimation of Discretionary Loan Loss Provision and Variables Measurement Now, in order that we use LLP in detecting earnings quality, we need to estimate the LLP made by sample banks in the industry. Nonetheless, we begin by making distinction between regulatory LLP as per the requirements of the PGs and discretionary loan loss provisions (DLLP) that is used for earnings quality. The distinction is useful because, where banks provide only in compliance with the PGs (non-discretionary LLP), it would be wrong to conclude that it uses LLP for earnings quality. The DLLP therefore contains an element of provision in excess of PGs requirements. It follows that to conclude that banks smooth their reported earnings; there must be evidence of DLLP in the financial reports. Stated differently, the presence of DLLP is a prima facie pointer to the possibility of earnings quality behaviour among the banks. (McNichols and Wilson, 1998), (Grey and Clarke, 2004) and (Chang, 2008) adapted measuring earning management in banks with discretionary loan loss provision. Consistent with prior studies (e. g. , Kim and Kross, 1998; and Kanagaretnam et al, 2001), the beginning balance of nonperforming loans, change in non-performing loans and change in total loans to estimate the non-discretionary component of LLP is used. Because the beginning balance of nonperforming loans (NPL) is usually positively related to LLP, therefore, with a higher level of beginning nonperforming loans, banks will have to make a higher LLP. In addition, change in nonperforming loans (CHNPL) in the current period to have a positive effect on LLP is expected. The sign of the coefficient of change in the value of loan deflated by beginning loans (CHLOAN) is also positive. An increase in loan portfolio will most likely result in an increase in LLP. Equation (1) provides an estimate of the non-discretionary LLP: LLPit = 0 + 1NPLit + 2CHNPLit + 3CHLOANit + it[1] Where, LLPit = provision for loan losses deflated by beginning loans; NPLit-1 = beginning of period nonperforming loans deflated by beginning loans; CHNPLit = change in the value of nonperforming loans deflated by beginning loans; and CHLOANit = change in value of loans deflated by beginning loans. In equation (1) above, the independent variables account for the non-discretionary component of LLP, and consequently, the DLLP is given by the residual term. In order to explain the cross-sectional differences in the level of DLLP, it requires a two-stage analysis where the first stage explicitly models the non-discretionary portion of LLP using a model as per equation (1). In the second stage, the residual from the first stage regression, representing the discretionary portion, is subsequently used as the dependent variable. A drawback of this stepwise estimation procedure is that, it systematically underestimates the absolute value of the regression coefficients in the second stage (Kanagaretnam et al, 2001). Hence, to alleviate this potential problem, analysis using a single regression model is conducted including the three variables used in equation (1) to explicitly account for the non-discretionary component of the LLP. The empirical model is given thus: LLPit = 0 + 1EBTPit + 2L/DEPit + 3 WELLit + 4 LASSETit +5 CHNPLit + 6 NPLit-1 + 7 CHLOANit + it †¦Ã¢â‚¬ ¦ [2] Where, EBTPit = earnings before tax and provisions deflated by beginning assets; LLPit = provision for loan losses deflated by beginning loans; L/DEPit = ratio of loans to deposits; WELLit = a dummy variable which equals 1 when industry capital ratio is well above the legal requirement (i. e. , when the total risk-based capital ratio exceeds 10% and the tier 1 risk-based capital ratio exceeds 8%), and equals 0 otherwise; LASSETit = the natural logarithm of total loan assets; CHNPLit = change in the value of nonperforming loans deflated by beginning loans; NPLit-1 = beginning of period nonperforming loans deflated by beginning loans; and CHLOANit= change in value of loans deflated by beginning loans. The first three variables (EBTP, L/DEP, and WELL) explain cross-sectional differences in DLLP; the third variable (LASSET) is a control variable and the last three variables (CHNPL, NPL, and CHLOAN) account for the non-discretionary component of LLP. The corporate governance variables of the study- board size, board composition, ownership concentration, dividend and institution shareholding are measured below: Board size (BS) is the total number of directors in the board Board composition (BC) is the ratio of independent or outside directors to total board size. Ownership concentration (OC) is the percentage of shares owned by blockholders (more than 5%). Institutional shareholding (IS) is the ratio of equity share owned by institutional investors to total number of shares issued. Dividend payment (DP) is the total amount of dividend paid. Firm size (FS) is the control variable which is the natural log (1n) of total assets. Such control is necessary because the bigger the bank, the larger the expected agency problem it will experience. Grey and Clark, 2004) note that large banks likely to avoid using discretionary loan loss provisions to manipulate earnings. A lot of researchers controlled for firm size in corporate governance studies including (Sanda et al. 2005), (Dabo and Adeyemi, 2007) and (Roodposhti and Chasmi, 2011). The regression model for testing the hypothesis of this study is presented below: LLPit = ? it + ? 1BSit + ? 2BCit + ? 3OCit + ? 4ISit + ? 5DPit + ? FSit + eit LLP= Loan Loss Provision BS=Board Size BC = Board Composition OC = Ownership Concentration IS = Institutional Shareholding DP = Dividend Payment FS = Firm Size 4. 1 Result and Discussion The analysis begins with a range of descriptive statistics on dependent variable and independent variables with mean, standard deviation, minimum and maximum presented below: Table 1: Summary of Descriptive Statistics | BS| BC| OC| IS| DP| FS| Mean| 14. 5444| 0. 2644| 0. 3177| 0. 5608| 2. 6556| 11. 2100| Std. Dev. | 2. 65703| 0. 11349| 0. 10243| 0. 5267| 0. 33000| 0. 11000| Minimum| 9. 00| 0. 0| 0. 12| 0. 43| 10. 37| 68. 97| Maximum| 20. 00| 0. 50| 0. 60| 0. 68| 4. 15| 83. 20| Observation| 90| 90| 90| 90| 90| 90| Source: Output of data analysis using E-view The table 1 shows the average independent directors in the board composition of the Nigerian banks is 26%, board size accounted for about 14 directors, block holders and institutional shareholding averaging 56% and 32% respectively of the shares issued and N2. 66k is the average dividend paid by Nigerian banks. The control variable is a veraging 11. 1 billion naira worth of assets by the banks. The standard deviations of most of the variables are not far away from their respective means values. This indicates a favourable level of dispersion that the data is not skewed and good to produce a reliable result which is confirmed by the values of skewness and kurtosis though not reported but attached. The minimum and maximum number of both executives and non-executives directors are 9 and 20 and that of independent directors of the banks are 10% and 50% respectively. In addition, the block holders and institutional shareholders of Nigerian banks range from 12% to 60% and 43% to 68% respectively. The total assets of the banks range from 68. 97 to 83. 20 billion naira during the period of the study. Table 2: Correlation Matrix | FIQ| BS| BC| OC| IS| DP FS| FIQ| 1| | | | | | BS| 0. 25| 1| | | | | BC| 0. 21| -0. 025| 1| | | | OC| 0. 29| 0. 090| 0. 024| 1| | | IS| 0. 42| 0. 035| -0. 034| 0. 34| 1| | DPFS | 0. 170. 31| -0. 146 0. 70| -0. 015-0. 034| 0. 158-0. 071| 0. 4890. 415| 10. 74 1| Source: Output of data analysis using E-view The correlation results presented in table 2 shows that all the explanatory variables are positively and strongly associated with explained variable except institutional shareholders and dividend paid. Thus, there is a strong relationship between corporate governance mechanisms and loan loss provision of the Nigerian money deposit banks. On the other hand, most of the independent variables are negatively and not significantl y associated between them. This indicates an absence of multicolinearity between the explanatory variables of the study. The correlation matrix reveals the relationship between all pairs of explanatory variables involved in the regression model. High correlation among the independent variables point the possibility of multicollinearity (excessive correlation), a situation which distorts the standard errors of estimates and the validity of the result became questionable. The correlation coefficients showed that multicollinearity does not exist among the variables. Additionally, this study adopts further test for excessive correlation using the variance inflation factor (VIF) and tolerance values. The purpose of additional investigation is to provide adequate assurance that the research findings are robust to the model specification. Table 4:Multicollinearity Test Variable| VIF| Tolerance| BS| 1. 081| 0. 925| BC| 1. 003| 0. 997| OC| 1. 080| 0. 926| IS| 1. 407| 0. 711| DP| 1. 657| 0. 603| FS| 1. 440| 0. 694| | | | Source: Output of data analysis using E-view To formally substantiate the lack of multicollinearity between the independent variables, colinearity diagnostics are observed and that the variance inflation factors (VIF) and tolerance values indicate no multicollinearity in the data. The values for tolerance and VIF are shown in Table 4. The tolerance value and the variance inflation factor (VIF) are two advanced measures of assessing multicollinearity between the independent valuables. The variance inflation factors and tolerance values are computed and found to be consistently smaller than ten and one respectively indicating absence of multicollinearity (Neter, Kutner, Nachtsheim, and Wasserman, 1996 and Casey and Anderson 1999). In addition, the tolerance values are consistently smaller than 1. 0 thus further substantiates the fact that there is no multicollinearity between independent variables (Tobachnick, and Fidell, 1996). The following regression result of the study is presented and discussed. Table 3: Regression Results Variable| Coefficient| T-Statistic| Sig. | | BS| 0. 355| 4. 123| 0. 000*| | BC| 0. 196| 2. 360| 0. 021**| | OC| 0. 144| 3. 673| 0. 008*| | IS| 0. 186| 4. 890| 0. 002*| | DP| 0. 520| 4. 873| 0. 000*| | FS| -0. 488| -4. 908| 0. 000*| | R-sq uared| 0. 43| Adjusted R-squared| 0. 40| F-statistic| 10. 46| F-Sig| 0. 000*| Durbin-Watson stat| 1. 99| | | Source: Output of data analysis using E-view The results show that the estimated model of the study is fit because all the explanatory variables are significant in determining the dependent variable. It can also be observed that the coefficients of all the explanatory variables are positive except for the control variable firm size signifying that none of the explanatory variables is inversely related with quality of earnings among Nigerian deposit money banks. The cumulative influence of all the explanatory variables put together is able to explain the dependent variable to 40% as indicated by the adjusted R2 and the remaining 60% is controlled by other factors. Again, the value of the F- statistic 10. 46 and significant at 1% confirms that the model is well fitted. This provides evidence of rejecting the null hypothesis that corporate governance mechanisms have no significant impact on the quality of earnings among Nigerian deposit money banks. The Durbin- Watson of 1. 99 reveals that serial correlation will not pose a problem to the statistical result of the study. The result in respect of board size and board composition shows that both of them positively and statistically significant at 1% and 5% respectively. This implies that the more their numbers the better the quality of earnings among Nigerian DMBs. For board size, the result reveals that Nigerian banks should have a minimum of 9 and maximum of 20 executive and non-executive directors for their reported earnings to be of quality. The finding supported that of Jensen (1993) and Lipton and Lorsch (1992) who suggest that small boards are less effective monitors and are easier for CEOs to influence and contradicts those of Yermack (1996), Eisenberg et al. (1998) and Loderer and Peyer (2002). The findings related to board composition or independence is line with (Hashim and Devi, 2008; Cornet et al. , 2009), and contrary to (Cornet et al. 2007; Roodposhti and Chashmi, 2010; Shah, Zafar and Durrani, 2009). Therefore, the policy implication is for Nigerian banks to have atleast 10% and not more than 50% independent directors out of the total maximum number of directors of 20. Moreover, for institutional share holders the result reveals that institutional shareholding is positively significant in influencing the quality of earnings among the Nigerian DMBs. This implies that banks with high number of institutional holders, their managers are easily restrained to manipulate accounting numbers. Looking at the level of association between institutional ownership and loan loss provision, a positive relation emerged and supported statistically. This significant association indicates that institutional investors are a major consideration in managers aggressive earnings management strategy. This result is not surprising as it shows institutional investors in Nigerian banks are effective in constraining managerial behaviour of earnings management. Consistent with the argument that institutional investors in Nigeria create incentives for managers of their portfolio firms to manage earnings aggressively, these institutional investors focus excessively on current earnings performance (Koh, 2003). Interestingly, this study extends the findings of Shehu (2011) who used a sample of 63 firm-year observations to document a positive relationship between institutional investors and financial reporting quality in the Nigerian banking industry. It also supports Cornett et al. (2008) who used 24,005 sample of U. S. industrial firms to document a postive and robust relationship between institutional investors and firm performance even when performance is stripped of the discretionery accruals. However, it contradicts that of Dabo and Adeyemi (2009) who fail to establish a statistically significant association between institutional shareholding and managers’ opportunistic behaviour using 20 most active quoted firms on the Nigerian Stock Exchange. Moreso, it conflicts with the finding of Al-Fayoum (2010) in their sample of Jordanian industrial firms. It can therefore be concluded that large institutional shareholding in the Nigerian manufacturing firms helps to allay the agency problem and leads to the protection of minority shareholders’ interest. In addition, the result in respect of ownership concentration and earnings quality reveals that ownership concentration is positively and strongly impacting on earnings quality at 1% level of significance among Nigerian deposit money banks. This is in line with our expectation, because given the fact that in most cases the institutions are the blockhoders of the company, therefore the results of the two variables would go in the same direction. The result contradicts the findings of Roodposhti and Chashmi (2010), Klai and Omri (2011) and supports the findings of Hashmi and Devi (2008). The implication of this finding is that the concentration of equity ownership in the hands of few individuals should be encouraged by the bank regulatory authorities. An equity ownership ceiling that should be raise. The result regarding dividend payment and earning management shows that managers will decline from managing earnings to enable them pay dividend to share holders. Statistically, dividend payment influences earning quality at 1% level of significance. This finding is in line with Bello (2013), Goergen,(2007) and Jensen (1986). Finally, the control variables banks’ size significantly and inversely affected the quality of earnings of DMBs in Nigeria. Size appears to affect earnings management inversely indicating that banks with larger assets have low earnings quality since they engage more in earnings management. This may be as result of the availability of much asset may motivate the managers to discretionary take selfish decisions to benefit their personal interest. 5. Conclusion and Recommendation Boards of directors, institutional and block holdings are responsible for monitoring, evaluating, and disciplining banks’ management. Perhaps one of the most important responsibilities of the board from a creditor’s perspective is oversight of earnings quality. Consistent with this idea, it is found that board size, board composition, ownership concentration, institutional shareholdings and dividend paid are all st rongly playing a prominent role in restraining management toward earning management.