SMES CORPORATE GOVERNANCE MECHANISMS AND BUSINESS PERFORMANCE: EVIDENCE OF AN EMERGING ECONOMY

This paper empirically examines the nexuses between SMEs governance mechanisms [board size (BS), board composition (BC), chief executive officer duality (CEOD), chief executive officer tenure (CEOT), board meetings (BMET), gender diversity (GEND), firm size (SZ) and firm age (AGE)] and business performance (BP) [ROA and Tobin’s Q]. The study deployed panel data multivariate regression via fixed effect for its analysis. By using annual reports of 124 Ghanaian SMEs selected on the basis of data availability, covering 2010-2019, the paper explored SMEs governance-performance-connexion by following the methodologies of researchers/scholars in extant literature. Findings/Results indicates that, there exists positive relationships among CEOT, BMET, SZ and AGE and BP. Nevertheless, BS, BC, CEOD and GEND depicted negative relationships with BP. Findings showed there are mixed results vis-à-vis governance mechanisms and BP. Findings further connote that; Ghanaian SME sector have distinctive attributes and may respond differently to governance mechanisms. Stakeholders will be abreast of the happenings in the Ghanaian SME sector for improved governance mechanisms. This paper contributes to the body of knowledge in extant literature on corporate governance and BP in the SME sector from an emerging economy’s perspective.


INTRODUCTION
Attention directed towards corporate governance (CG) issues of contemporary organizations in recent times has been rekindled. This is predominantly due to the conspicuous collapse of a number of large corporations, which Ghana has had its fair share owing to the collapse of some indigenous Banks [Construction Bank, The Royal Bank, UT Bank, Capital Bank, UniBank, Beige Bank, Sovereign Bank, GN Bank and Heritage Bank] during August 2017-January 2019 Ghana banking crises. General non-adherence to proper CG mechanisms has been mentioned as one of the chief grounds of the failure of these banks by Ghana's central bank (www.bog.gov.gh). These and other global corporate scandals as well as the apparently meagre performance of businesses on the Africa continent have given rise and stimulus to CG in the subregion (Badele and Fundeanu, 2014).
CG has been a rising field in management inquiry and an anteriority on the policy agenda in advanced economies and the concept is steadily gaining grounds as a priority phenomenon in developing economies, Ghana inclusive. Extant literature largely supports the assertion that, good CG positively influence BP (Sarpong and associate, 2013;Frimpong and associate, 2015;Nyarko and associate, 2017). To this end, CG has become a very expedient mechanism for running businesses in today's global market. In view of this, organizations are striving to inculcate the sense of governance into their business structures. The Cadbury Committee (1992) defines CG as the system through which organizations are directed and controlled.
These elements of set systems, governs the interrelationships among shareowners, management and other stakeholders (Sila and Associate, 2016). CG ensures that, organizations have proper decision-making procedures at all times, so as to guarantee the welfare of all industry players with the utmost aim of maximizing the firm's value. CG deals with the superintendence and demanding accountability from personnel that supervise and monitor the day-to-day affairs of the organization. Governance at the corporate level includes the procedures in which an organization's objects are established and implemented in the framework of the supervisory and business settings. CG deals with processes/practices that ensures that an organization is run in such a manner that results in the attainment of its set objects, while safeguarding the interest of all stakeholders in the organization.
In Ghana, CG has been gaining roots through the interventions by some stakeholders namely Ghana Institute of Directors (IoD-Ghana) and Commonwealth Association of Corporate Governance (CACG). IoD-Ghana in 2001 undertook a research, which showed a growing recognition of proper CG implementation by organizations in Ghanaian communities. Nevertheless, official CG systems are comparatively not prevalent, even though some legislations provide for CG systems for Ghanaian businesses. journal.ump.edu.my/jgi ◄ These statutes include Ghana's Companies Code 1963 (Act 179). This law regulates and governs all companies incorporated in Ghana. Again, Ghana's Securities and Exchange Commission (SEC) in 2003 introduced CG rules/guidelines by with all Ghanaian companies were to comply in addition to the commission's own law i.e., the Securities Industry Law, 1993 (PNDCL 333) now Securities Industry (Amendment) Act 2000, (Act 590). Also, there is the Ghana Stock Exchange (GSE) Listing Rules, which has contributed to governance in Ghana via its regulation of businesses found on the GSE. CG and BP association has been an issue/topic of extensive interest of study. On one hand, finance models substantially estimate a direct association between the two phenomena, whilst research findings are heterogenous and inconclusive. A number of the scientific inquiries have yielded varied outcomes. A number of researcher's findings indicates that CG influences BP positively (Sarpong and associate, 2013;Frimpong and associate, 2015;Nyarko and associate, 2017) while others have found either the reverse (Bathala and Rao, 1995;Hutchinson, 2002) or non-existence of any association (Prevost and associate, 2002;Park and Shin, 2003;Singh and Davidson, 2003;Young, 2003).
The scientific research findings have been varied depending on the nations of interest, industries and organizations under consideration. This shows that, the influence of CG on BP is mixed and provisional on several variables, such as the CG variables utilized as explanatory variables, performance measurement variables used as prognostic variables and the econometric estimation technique deployed for the empirical investigations. Also, most scientific inquiries in the field of CG have been centered on large and listed companies in advanced economies which reveals homogeneous organizational environments/settings, but the phenomenon is gradually trickling down into emerging market economies Ghana inclusive (Eisenberg and associate, 1998;Bennett and Robson, 2004). It is therefore highly imperative to empirically assess CG mechanisms in the SMEs sector from the view point of emerging market economies like Ghana. This is because most SMEs in developing economies are owned by families, close allies and friends and they are likely to ignore the practice of proper CG principles. Therefore, an attempt to bring good CG practices to the door steps of SMEs in an emerging market economy is a good call as this would have a good impact on value creation through business entrepreneurship. Again, notwithstanding the accretion of literature on the effect of CG on BP, there appear to be limited research conducted on the SMEs sector in general and Ghana in particular even though some attempt has been made to research into the subject area in Ghana, which is an astounding gap generated by extant literature.
For instance, Tornyeva, and Wereko (2012) examined the association between CG and the BP of Ghanaian insurance companies, whilst Arthur (2016) investigated the extent to which SME have adhered to good CG principles in the formulation of organizational structures. Ofoeda (2017) also investigated the influence of CG mechanisms of non-bank financial institutions (NBFIs) on their profitability in Ghana. Boachie (2021) assessed the mediating role of ownership on the nexuses between CG and BP of Ghanaian banks; Musah and Adutwumwaah (2021) assessed CG mechanism influence on the BP of Ghanaian rural banks; Gakpo and Kwakye (2021) conducted a study to find out whether good CG practices adherence affects Ghanaian State-Owned Enterprises (SOEs) financial performance among other studies.
Very little scientific studies exist to assess the linkages between the two phenomena with specific emphasis on SMEs in developing/emerging market economies in general and Ghana in particular. This research is motivated by this observable lacuna. The present scientific inquiry is therefore undertaken to assess the association between CG and BP focusing on the SME sector of Ghana. Considering the nature of SMEs operations and their predominant informal clientele base, they are widely-open to higher risk. Therefore, the deployment of good CG mechanisms to mitigate the risk levels of SMEs and improving their performance as well as ensuring their survival is highly imperative. This current research seeks to empirically assess the effect of CG mechanisms on the performance of Ghanaian SMEs as an attempt to add to the academic/research literature from the viewpoint of an emerging/developing market economy. The focal research questions for this empirical study are as follows: i. Do SMEs employ/implement good CG principles/practices in Ghana? ii.
What are the CG mechanisms/principles/elements being practiced by SMEs in Ghana? iii.
Do the CG principles influence SMEs performance in Ghana? iv.
What is the relationship/association between CG mechanism and BP of Ghanaian SMEs? The remaining part of this paper is organized as follows: The second section is a review of related literature. The third section is research methodology. The fourth section is results/findings. The fifth section is discussion. The sixth section is conclusion.

Underpinning Theory
The Theoretical foundations/underpinnings for the extant research in CG include the following: agency theory (AT), stewardship theory (SWT), resource dependency theory (RDT), stakeholder theory (SKT), transaction cost theory (TCT), political theory (PT). This present research was based on the SWT of CG. The SWT, which originated from the social science disciplines of psychology and sociology was initiated by Davis (1991 &1993). Davis, Schoorman, Donaldson and Davis (1997) defined a steward as a person who keeps and make the best use of shareowner's capital via BP, and by so doing, he/she attains intrinsic satisfaction. SWT is substitute to agency theory vis-a-vis managerial motivation. It suggests that, stewards safeguard and make the best of shareowner's wealth via organizational Performance. Stewards are organizational top leaders working for the owners, shield and achieve returns for the owners. The stewards are pleased and inspired when business success is accomplished. It emphasizes on the locus of employees/executives to act more independently so that the shareowners' rewards are maximized. The workers assume ownership of their tasks journal.ump.edu.my/jgi ◄ and perform them assiduously. The model contends that shareowners' interests are maximized by stockowners tenure of the roles of board chairperson and CEO (Donaldson and Davis, 1991).
The theory suggests that, it concentrates on the part of executive directors to analyze link with BP. The inside directors are not opportunistic loafers. Their focus is to perform efficiently and effectively and to be good care-takers of the assets under their control. The framework asserts that, with respect to top leadership motivation, there is no hidden dispute/trouble. Administrators, act as accountable managers of shareholders wealth they manage (Davis and associate, 1997). The humanhood framework as an agent is rooted on a stewardship principle. The rationale behind is that, the focal purpose of stewards is to accomplish the goals/aims of the firms. This conduct is eventually advantageous to owners in relation to improved prices of share and return on investment. The framework suggest that directors and management are one, collective stewardship team. Directors fundamentally aid management. Stewardship theorists anticipate a substantial relation between the firm growth and the progress of stockowners. Stewards/managers/employees conduct themselves in the best interest of shareowners because, it satisfies stewards intrinsically, steward's reputation as decision makers is protected Daly (2003), it serves as a means of managing stewards' career to be seen as effective stewards (Fama and Jensen, 1983) and lastly it enables stewards to re-enter the market when their contracts expire (Shleifer and Vishny, 1997).
Most SMEs in Ghana are privately owned with only four (as at August 2021) SMEs on the Ghana Alternative market (GAX) -[another market run by the Ghana Stock Exchange (GSE) that provides SMEs with the opportunity to raise external capital both equity and debt for their business operations]. In addition, SMEs in Ghana have few employees who are mostly related to the owners and their family members or close associates constitute the board of directors. As such managers acts as stewards and perform in the best interest of the firm by maximizing the value of the firm and by so doing managers are intrinsically motivated. Source: Adopted from Davis (1991 &1993)

Empirical Evidence
Empirical inquiries on CG and BP have generated myriad and inconclusive results. This can be attributed to the deployment of diverse CG measurement variables depending on the authors interest, the operating market of interest, country specific effects, methodological approach as well as the utilization of different performance measurement, since performance is dependent on the efficacy of the entire CG system and not on individual mechanisms (Ofoeda, 2017). CG scholars have resorted to address such issues by broadening the CG mechanism and inculcating more variables into the CG principles. (Bhagat and Black, 2000;Weir andassociate, 2002, Ahmed et al., 2021). Information from extant literature indicate that CG influences market/operating variables of BP (Foong et al., 2019). Example, Brown and Caylor (2004), asserts that proper CG results in the decrease in managers discretion and control with respect to investor's funds allocation. Shleifer and Vishny (1997), also opine that, well-governed businesses yield good BP. Conversely, Gompers and associate (2003), Bebchuk and Cohen (2005) and Bebchuk and associate (2009) all are of the view that, firms that generate high Tobin's Q are those with stronger stockholder rights, indicating that well-governed firms have better market values. In the context of SMEs, CG is more related to the distinctive roles of the shareholders as owners and managers either of directors or other officers (Abor & Biekpe, 2007). Researches have proven that, good CG may have positive impact on SME performance (Abor & Biekpe, 2007). Boachie (2021) assessed the mediating role of ownership on the nexuses between CG and BP of banks in Ghana. Deploying a multiple linear regression methodology on a panel data set of 414 observations on 23 banks over an eighteenyear period, the study found that, audit independence, CEOD, non-executive directors and banks size depicted a positive effect on BP. The study's results further showed that foreign ownership has an interrelating impact between CG and profitability.
Musah and Adutwumwaah (2021) investigated the effect of CG mechanism namely BS, board independence, GEND and CEOD on the financial performance of Ghanaian rural banks and reported that there was a direct but insignificant association between CEOD and profitability. Findings indicated further a direct relationship between BS and profitability even though that of ROA was insignificant. Further, board independence was found to be a significant cause of BP and an inverse association between GEND and BP. journal.ump.edu.my/jgi ◄ Gakpo and Kwakye (2021) conducted a scientific inquiry to ascertain whether good CG practices adherence affects Ghanaian State-Owned Enterprises (SOEs) financial performance. Findings revealed that there were some direct and negative associations between the four explanatory variables and performance. Nevertheless, most of the associations were statistically insignificant. Improvement in CG was indirectly achieved via interactions with other factors. Badu and Assabil (2021) explored the connexon between BC and financial information value relevance in Ghana and observed that a larger proportion of outside directors is related with lower value of firm. Findings further indicated that BS was directly associated to firm value, while CEOD was inversely related with firm value. Fiador and Sarpong-Kumankoma (2021) investigated the influence of CG elements on bank loan portfolios quality by utilizing a panel-corrected standard error estimation model with the most current 11-year data spanning 2006 -2016 on some banks in Ghana. Results from the study revealed that CG is important in the Ghanaian banking sector and plays a focal role in ensuring low loan loss. The study further indicated that having more members on the board with varying level of banking acumen, boards with mostly outside members and CEOD the chair of the board can be harnessed to ensure low loan loss. Involvement of the female gender on boards appears to diminish from excellent BP, depicting the impression of minimum effort in the Ghanaian financial industry.
Agyei-Mensah (2021) assessed the influence of board attributes on business' investment decisions by analyzing financial statements of businesses listed on the GSE spanning 2014-2018 and deploying multivariate linear regression analysis. Study findings shows that the percentage of outside directors and financial gurus on the board are inversely correlated to firm investment. Study results further suggest that non-executive directors and financial gurus on the board can assist businesses minimize over investment and enhance investment efficacy.
Puni and Anlesinya (2020) explored the impact of CG mechanisms indorsed by Ghana's SEC on BP. By using panel data multivariate linear regression methodology for 38 listed Ghanaian firms covering the period 2006-2018, to ascertain the compliance and contribution level of the CG variable indorsed by the SEC to BP, the study revealed that the inside and outside directors' presence on board of businesses enhance BP. Likewise, BS, BMET and shareowner ownership structure mostly had a positive influence on BP. Nevertheless, board committee's presence largely resulted in an adverse effect on BP whereas CEOD had no effect on BP.
Ahulu and MacCarthy (2019) assessed the impact of CG mechanisms on the market value of firms in Ghana. By deploying panel data multiple linear regression analysis on thirty-one firms listed on the Ghana stock exchange covering the period 2009 to 2018 the study found that CG mechanisms accounted for 84.9% of the variation of a firm's market value for the period under consideration. Additionally, the research depicted a significant association between CEOD, outside director, BS and firm's profitability and value. The study recommended the two-tier board structure of governance to improve firm's profitability and value.
Ofoeda (2017) investigated the effect of CG principles of non-bank financial institutions on their profitability in Ghana and noted that, there exists positive relationship among BS, audit committee size, meetings of the audit committee and profitability. However, BC, GEND, BMET and audit committee independence show an inverse association with NBFI performance. Arthur (2016) investigated the extent to which SME have adhered to good CG principles in the formulation of organizational structures, everyday decision-making processes and more importantly, in organizational operations. Findings revealed that, most SMEs do not have CG structures in place to guide their operations. Tornyeva, and Wereko (2012) examined the association between CG and the BP of Ghanaian insurance companies, utilizing panel data regression analysis. The study's results depict that large BS, board skill, management skill, longer serving CEOs, size of audit committee, audit committee independence, foreign ownership, institutional ownership, dividend policy and annual general meeting are positively correlated with the financial performance of Ghanaian insurance companies.
The adoption and implementation of good CG mechanisms, necessitates an appreciation of the variables that affect its implementation. This aids firms to choose appropriate CG variables applicable to individual firm's specific requirements. The role of CG practices vis-à-vis BP is well-established in theory; an empirical assessment of the same is limited and inconclusive so far especially focusing on SMEs in developing economies, Ghana inclusive. On the basis of this, this study explored the outcome of CG practices on BP focusing on the SME sector in Ghana.

Business Performance
The standard/criteria for performance measurement depends on the researcher's interest and the justification basis (Akinlo, 2012). A careful scrutiny of extant literature shows that, researchers/scholars have measured BP using two main measures namely, financial (accounting/market-based performance measures [profitability, liquidity, efficiency, gearing and market values indicators] and non-financial measures [operational measures, quality measures, overall competitive position and customer satisfaction indicators] (Essel, 2021). This is done via either undertaking objective analysis of company's annual financial statements or by gathering subjective data/information from survey respondents based on their individual opinions on BP. This present study measures BP via extracting financial information/data from the annual reports of the 124 SMEs under study, focusing on one key profitability measure i.e., return on assets (ROA), and one key market value indicator i.e., Tobin's Q as used by journal.ump.edu.my/jgi ◄ researchers in extant literature like Ofoeda (2017)  The following hypotheses are formulated based on the extant related literature reviewed above on CG mechanisms and firm performance: H1: There is a positive relationship between BS and BP H2: There is a positive relationship between BC and BP H3: There is a negative relationship between CEOD and BP H4: There is a positive relationship between CEOT and BP H5: There is a negative relationship between BMET and BP H6: There is a positive relationship between GEND and BP H7: There is a positive relationship between SZ and BP H8: There is a positive relationship between AGE and BP

METHODOLOGY
This section details the sampling technique deployed to arrive at the sample size, discusses the data sources and model specification for the empirical investigation.

Target Population
This present scientific inquiry centers on Ghana with the main characteristics of interest been all SMEs operating within the metropolis of Accra and Tema in the Greater Accra Region of Ghana whose operations involve adherence to governance practices in pursuit for profit.
Out of a list of 287 SMEs (study population/universe) registered with the Association of Ghana Industries (AGI)-a voluntary business association and the Ghana Enterprises Agency (GEA) formely National Board for Small Scale Industries (NBSSI)-a government ageny operating under the auspices of the Ministry of Trade and Industry mandated by the Ghana Enterprises Agency Act, 2020 (Act 1043) to technically and financially support the growth of SMEs in Ghana, 124 were in good standing with complete data. This figure therefore constituted the sampling frame for the study.

Sample Selection and Sample Size
For the purposes of this study, all 124 SMEs who were in good standing was used as the sample size.

Data Analysis
This research used a combination of primary and secondary sources of data. Raw data was gathered from the 124 SMEs on CG issues via interviews and administration of questionnaires. This was performed in two stages. The first stage involved interviewing top management leadership of the 124 SMEs to ascertain CG issues in the selected firms. The second stage involved administering questionnaires to the management of the SMEs as a means of validating CG issues that came up during the interview phase. This aided the researchers to settle on CG mechanism such as BS, BC, CEOD, CEOT, BMET, GEND, SZ and AGE for this study.
The study also utilized financial information obtained from final accounts of the 124 SMEs involved in the inquiry. The researchers deployed abstraction methodology to collect financial information from the annual reports of the 124 selected SMEs. The study deployed panel data framework for its analysis in view of its superior benefit of permitting more data points. The data for this study spans 2010 to 2019 (times series), across 124 SMEs (cross-sectional). journal.ump.edu.my/jgi ◄

Model Specifications and Definition of Variables
The econometric model specification adopted for this current study follows Biekpe, 2006a, Miyajima andassociate (2003), Wen and associate (2002) and other similar studies deployed in extant finance/economic literature is given as: Where represents business performance (BP) variables, ROA and Tobin's Q of SME i at time t, i = 1, 2, 3………124 SMEs used as outcome/criterion/prognostic/response/dependent variable. Git is a K -dimensional vector of CG focal/main explanatory/predictive/independent variables, Xit represents the vector of firm level (control) variables. βo represents the model's intercept/constant, β1-8 is the coefficients of Git and Xit variables of SME i at time t and Ɛit is the residual/error/stochastic term decomposed into = + , where µi is the firm's specific effect, and vit is a random term. Depending on the behavior of the model's error term, the panel data analysis can be conducted in different ways. The serial correlation and heteroskedasticity conditions of the whole model also influences the choice of analysis to apply. The pooled ordinary least squares (POLS), fixed effects (FE), random effects (RE), generalize least square (GLS) and dynamic panel (DP) models are the notable econometric methodological techniques employed in extant literature.
The authors performed Hausman Specification Test (HST) to ascertain the appropriate/best methodology to use to run the regression model. The HST revealed FE is apposite to use to run the regression model. The results revealed that the hypotheses regarding the test was constant under both the null and the alternate hypothesis for the FE. Put differently, the HST rejected the null hypothesis (H0) which stated that RE model was appropriate but failed to reject the alternate hypothesis (H1) which stated that FE model was appropriate. The regression results tables i.e., tables 3 and 4 shows the HST. Following empirical studies in extant literature, the econometric model (equation 1) for this study is specified (broken down) as follows: The measurements of variables of study are presented in Table 1.  (2017) It must be noted that, the empirical model specified in equation (2) did not include any of the audit committee variables/parameters i.e., audit committee size, audit committee independence, and audit committee meeting times. This is because, most of the SMEs in Ghana are not listed on the GAX, where the SMEs are required to set up audit committees as part of the listing/operating requirement. Currently, (as at October 2021), there were only four (4) SMEs listed on the GAX. In addition, those non-listed SMEs do not institute audit committees on their own. They only prepare final accounts (Annual Reports) and submit copies to the Registrar of Companies and other statutory organs of state as required by law. journal.ump.edu.my/jgi ◄

FINDINGS AND DISCUSSIONS
This fourth section discusses the empirical results of the study by first presenting the summary descriptive statistics, followed by correlation matrix and then the empirical regression results. Table 2 presents a summary descriptive statistic of the prognostic and explanatory variables for the study. The average profitability in terms of ROA of the 124 SMEs operating in the Accra/Tema metropolis of Ghana is 19 per cent. The implication is that investors (both equity-holders and debt holders) who made capital available to these SMEs made on the average of 19 per cent return on their investments. Majority of the SMEs on the average seems underperforming so far as Tobin's Q as a performance measure is concern with a mean ratio of 0.65. This suggests, a greater number of the SMEs did not break-even.

Descriptive Analysis
The SMEs under study had on the average, a board size of about 5 members and even though they are a mix of executive and non-executive board members, majority are executive board members, meaning the board seriously lack independence as, only 15% of the board members are selected outside of the firms, a situation which leaves much to be desired. This is not in line with the best corporate governance practices and may affect the board's effectiveness and hence BP.

Correlation Analysis
The Author performed correlation test to ascertain whether multicollinearity problem exist and also bring out the associations/relationships between the prognostic variable and the explanatory variables as well as determine the correlations among all the explanatory variables. With respect to multicollinearity, the focal reason for conducting this test was to circumvent a situation where two or more independent variables with high correlation would be included in the regression model. Table 3 depict the correlation matrix with ROA and Tobin's Q as the criterion variable against the other CG predictive variables.
The correlation matrix for the CG regression model depicts that, all the explanatory variables are eligible to be part of the regression model because the correlation between them was not high. All the explanatory variables had a less than 0.6 correlation figures and this is considered satisfactory as indorsed by Hair and associate (2016).
The correlation matrix depicts positive correlation between the tenure of SMEs CEOs (CEOT) and BP (ROA and Tobin's Q) as well as board activity intensity i.e., frequency of board meeting (BMET) and BP (ROA and Tobin's Q). This same direct correlation is reflected in the two control variables i.e., SMEs size and age and BP (ROA and Tobin's Q). The remaining explanatory variables (board size, board composition, CEO duality and gender diversity) showed a negative correlation with BP (ROA and Tobin's Q). journal.ump.edu.my/jgi ◄ Note: ***, ** and * represents significant at 1%, 5% and 10% level, respectively.

Regression Assumptions Testing
The authors ensured that the standardized multivariate linear regression model satisfy all the assumptions [Normality, autocorrelation, endogeneity and heterogeneity] necessary under multiple linear regression analysis to avoid a situation where an assumption would be violated which will result in biased/spurious outcomes. The assumption testing guarantee that the model is fit for regression. The regression assumption testing is presented in Table 4.

Regression Analysis
Multivariate linear regression analysis is utilized to explore the association between CG variables (BS, BC, CEOD, CEOT, BMET, GEND,) and other control variables (SZ and AGE) and BP (ROA and Tobin's Q) of the 124 SMEs in Ghana. The results of the FE balanced panel data regression are shown in Table 5 with ROA as outcome variable and Table 6 with Tobin's Q as prognostic variable. journal.ump.edu.my/jgi ◄ Using ROA as the dependent variable, the regression results indicate that, the explanatory variables explained 67.51% of the variance in the outcome variable. The F-statistics demonstrates satisfactory validity of the estimated models i.e., the balanced panel regression model has good fit. The regression outcome vividly depicts that, there exist a mixed result between the governance explanatory variables and ROA. The regression results with Tobin's Q as dependent variable indicate that, the explanatory variables explained 41.62% of the variance in the outcome variable. The F-statistics demonstrates satisfactory validity of the estimated models i.e., the balanced panel regression model has good fit. The regression outcome vividly depicts that, there exist a mixed result between the governance explanatory variables and Tobin's Q.

DISCUSSIONS
Whereas CEOT, BMET, SZ and AGE depicted positive (direct) relationship with profitability (ROA), BS, BC, CEOD and GEND revealed negative (inverse) relationship with profitability (ROA). Except for the GEND variable, all the other seven explanatory variables had statistically significant relationships with profitability (ROA). BS depicting a significant inverse (negative) relationship with profitability (ROA), indicates that, the smaller the BS, the better the SME profitability journal.ump.edu.my/jgi ◄ (ROA). This implies that small BS perform better and are much more efficient than large BS, which may lead to the hegemony by the CEO of the firm. Large BS seems to be shoddier than a small board. Large boards may be deficient in collaboration/cooperation, pliability and communication, which may surpass the advantages of having large boards. Small BS is more associated with the quality of monitoring management activities which ensures that management decisions are in the best interest of the firm.
Similarly, BC (board independence) had a significantly negative (inverse) association with ROA (BP), suggesting that, the more external (non-executive) board members there are on the boards of SMEs, the poorer they perform. This suggests that, the significance of external (non-executive) directors with respect to their outside experience vis-à-vis excellent financial/legal basis is unseen in the Ghanaian SMEs performance. The possibility of this assertion lies in the fact that, executive (inside) directors have access to information that is pertinent to evaluating managerial know-how and the tactical desirability of ingenuities. As such they are in the position to distinguish between genuine or illicit causes of business adversity.
In addition, executive directors are more likely to have firm-specific know-how, whereas non-executive directors are more likely to have general knowledge. In order for non-executive directors to perform their supervisory/monitoring and advisory role effectively/efficiently they must sustain substantial costs in respect of the attainment of firm-specific knowledge/capability which possibly explains the meagre performance of the SMEs. In addition, non-executive directors' cost of acquisition and maintenance is high and may far surpass the advantages obtained from running a board with many non-executive directors.
This outcome is in line with the theory of stewardship, which opines that maximum firm performance is aligned with superior inside director's composition as these executive directors, acting as stewards perform their duties in the best interest of shareholders but in sharp contraction to the agency theory perspective which argues that in order to solve the agency problem, firms must engage more non-executive directors independent of management stimulus in order to accomplish all-out performance. This result is consistent with empirical work by Agrawal & Knoeber (1996); Conyon and Peck (1998); Weir and Laing (2000); Ezzamel and Watson (2002); Haniffa and Hudaib (2006) but inconsistent with findings of Baysinger & Butler (1985), Brickley & James (1987), Weisbach (1988), Rosenstein & Wyatt (1990), Byrd & Hickman (1992), and Brickley and associate (1994), Kyereboah-Coleman and Biekpe (2006a), Abor and Biekpe (2007), Nanka-Bruce, (2009) who find positive correlation between BC and BP. The findings of this research therefore reject H2 With respect to CEOD, the study result revealed that, there exist a significantly negative/inverse relationship between CEOD and ROA. This means that, when SMEs CEOs doubles as chairpersons of the board of directors, SMEs BP/profitability/ROA drops. This is an indication that, the Ghanaian SME governance structure typology, supports the separation of CEO management powers from the overall supervisory, monitoring and controlling powers of the board which has the overall intention/motive of eradicating conflict of interest problems associated with agency problem and agency cost.
This finding is in line with empirical inquiries which have found out that, the one-tier board configuration typology leads to enhanced BP as evident by the scholarly works of Berg & Smith (1978), Fama and Jensen (1983), Bickley & Coles (1997), Sanda and associate (2003) but inconsistent with the findings of Donaldson and Davis (1991) and Lin (2004) who find positive association between CEOD and BP. The separation of CEO and board chairperson reduces the tautness between board members and managers thus positively impelling BP in the Ghanaian SME industry. The findings of this research therefore support H3.
CEOT and ROA revealed a statistically significant and positive/direct relationship. This suggests that, the longer the CEOs serve, the more knowledgeable and experienced they become, leading to improved BP. In addition, guaranteeing the career safety of the CEOs motivate them to take capital intensive decisions that have lasting bearing on BP. It is also highly imperative that, in assessing the performance of the CEO, a more wide-ranging approach is deployed in order to prevent the CEOs from focusing solely on immediate income/revenue/profits but also recognize future benefits to be derived as a result of the CEO's strategic long-term decisions taken. This thus stimulate a CEO's performance psychosomatically in view of the likelihood of the recognition of the investment decisions being beheld by him/her. CEOs experiencing such scenarios have high tendencies to be hands-on and high-minded in their decisions in view of the psychosomatic stimulus. The study's result is in conformity with the findings of (kyereboah-Coleman, 2007) but contradict the findings of (Abor, 2006). The findings of this research therefore support H4.
The study findings also revealed that, a positive and statistically significant association exist between BMET and ROA. This means that, the more the board meets (i.e., high board activity intensity represented by frequent board meetings), the better the performance of the SMEs which is reflected in the potency of the monitoring, supervision and controlling role-played by the board. BMET is recognized as an imperative means of enhancing the efficiency of the board of directors. It is opined/claimed that, BMET are vital means in which board members acquire firm-specific knowledge that assists them in their monitoring, supervisory and controlling role. The regularity of BMET is stated as impelling BP positively. This finding is consistent with the works of Davidson and associate (1998) and Godard and Shatt (2004) but at loggerheads with the findings of Jensen (1993), Johl (2006), (kyereboah-Coleman (2007) and Ofeoda (2017). The findings of this research therefore reject H5. journal.ump.edu.my/jgi ◄ The study's results also revealed that, a negative relationship exist between GEND and ROA, but this relationship is not significant. This indicates that, as female board members increase, the shoddier the SMEs perform. This means, a vastly female dominated board suggest that, SMEs performed poorer than their male counterparts. Working mothers/females aside their formal career duties also perform household functions which is time-demanding; therefore, they may have limited time available for performing their professional duties as compared to their male counterparts which may have account for their meagre show-down on the board. This is because the Ghanaian culture requires the feminine gender to take care of the home, i.e., child caring and other household chores irrespective of whether the female is a corporate working woman or not, thereby making them inefficient in performing their professional functions. This result is inconsistent with the findings of Kang and associate (2010) but consistent with Ofoeda (2017) empirical inquiry. The findings of this research therefore reject H6.
The study controlled for SME size and age to ascertain their impact on ROA and they both recorded significant and positive relationship with ROA. This implies that medium firms with staff strength ranging between 30 -99 employees performed better than the small firms with staff strength between 20-29. SMEs which have been in existence for long (older firms) also performed better than SMEs which have not been in existence for long (newer firms). This is because firms gain experience as they stay for long in the industry. They are able to learn from their mistakes and perform well. The larger and older SMEs who have established excellent rapport with their trade creditors/suppliers, clientele base and financial institutions, may have benefited from trade credit facilities from their dependable suppliers and may had had easy access to external debt funding in the form of loans and overdraft facilities for their business operations. Medium and older firms are more varied and hence have lesser difference of income, rendering them able to withstand extraordinary debt ratios. Financial institutions and trade creditors are less hesitant to advance credit facilities to medium and older firms in view of the perceived lesser risk levels attributes.
Contrary, micro and newer firms may find it comparatively deer to resolve discrepancies in information with lenders, thus, may present lesser debt ratios. This finding is consistent with Abor (2006) and Ofoeda (2017). The findings of this research therefore support H7 and H8.
The empirical regression result of the association between the governance explanatory variables and Tobin's Q as dependent variable is presented in Table 4. Again, just like the ROA, the empirical results vividly depicts that there exists varied result between the governance predictive variables and Tobin's Q. Again, Whereas CEOT, BMET, SZ and AGE depicted positive association with Tobin's Q, BS, BC, CEOD and GEND revealed negative association with Tobin's Q. All the explanatory variables had statistically significant association Tobin's Q with the exception of GEND. The empirical results indicate that the smaller the BS, the better the Tobin's Q. This is in line with studies such as Jensen (1993), Lipton & Lorsch (1992), Yermack (1996) that support the view that smaller BS are better for business output. Small BS is more associated with the quality of monitoring management activities which ensures that management decisions are in the best interest of the business. H1 is therefore rejected.
Similarly, BC had a negative association with Tobin's Q inferring that when non-executive directors on the board are more, performance tend to dwindle. This finding is in conformity with the findings of Agrawal & Knoeber (1996) who opines that boards increased for political motives usually leads to too many non-executive directors, which is irrelevant so far as performance is concern. This however opposes other researches by Brickley & James (1987), Weisbach (1988), Byrd & Hickman (1992), and Brickley and associate (1994) on non-executive directors. H2 is therefore rejected.
With respect to CEOD, the findings indicates that when the role of the CEO and that of the chairperson of the board is vested in a single personality, it has an adverse impact on performance i.e., Tobin's Q falls. This finding is in line with the findings of scholars/researchers like Berg & Smith (1978), Bickley & Coles (1997) who are of the view/opinion that when the CEO doubles as the board chairperson it results in leadership facing conflict of interest problems and hence and agency cost arises. H3 is thus supported.
Regarding CEOT, the study's finding indicates that when the CEO stays in office for quite a long time, its boost performance, as the considerable long stay of the CEO enable him/her to learn on the job, gain experience leading to improved Tobin's Q.
This finding is consistent with Kyereboah-coleman (2007), thus support H4. The statistically significant and positive sign of BMET signifies that, the more the board meets, it is reflected in a better Tobin's Q value.
However, H5 is thus rejected. Regarding GEND, the study's finding reported an insignificant and negative association with Tobin's Q inferring that women participation on the board of SMEs was not felt or women did not involve themselves in board activities. This present study suggests that the SZ and AGE have positive influence on Tobin's Q and are also statistically significant. This can be attributed to the fact that SZ and AGE enhance performance and it's reflected in an improved Tobin's Q. H7 and H8 are thus supported.

CONCLUSION AND RECOMMENDATION
SMEs governance germaneness cannot be over-emphasized in view of the various associated benefits that accompany proper governance practices in today's competitive global economy. The espousal of good CG mechanisms improves firm's operational pellucidity, guarantees answerability, enhances firm's profitability and aids in safeguarding shareholders interest in bringing into conformity managers-(agents) and shareholders-(principal) interest. This research explores the linkages between Ghanaian SMEs governance mechanisms and BP. The governance variables utilized in this study include BS, BC, CEOD, CEOT, BMET and GEND, while controlling for SZ and AGE. The outcomes of the study depict a mean board size of five members, 15 per cent non-executive director board composition, inferring the use of more executive directors on the board. Similarly, only 15 per cent of the board members were females. The mean CEOD was 75%, CEO tenure averaging 7 years in office and an average board meeting of 3 times per annum. The mean journal.ump.edu.my/jgi ◄ SME age was 10 years and an average SME size of 50 employees. The FE balanced panel data multivariate linear regression results depict further that, except for GEND variable, all the other explanatory variables exhibited statistically significant relationship with BP. Whilst BS, BC, CEOD and GEND exhibited negative relationship with BP; CEOT, BMET, SZ and AGE exhibited positive relationship with BP. The above-mentioned analysis indicates that, there are heterogeneous results regarding CG and BP of Ghanaian SMEs. It must however be noted that, this study's empirical findings are consistent with other empirical results in extant literature.
The outcomes of this research essentially suggest that, the espousal of CG mechanism have some vital repercussions for Ghanaian SMEs. A key outcome of a proper CG mechanism is easy access to external funding both debt and equity from investors/promoters. It must nevertheless be stated that, for the Ghanaian SMEs case/situation, managers and industrial stakeholders need to evaluate the typology of CG structures that influence BP positively, as this research revealed myriad results. This is an indication that CG practices cannot be introduced from developed countries and implemented on the Ghanaian SME sector, as the study's findings depicts that they have distinctive attributes and may respond incongruously to CG mechanisms. CG certainly encompasses a wider set of parameters including legal and economic setting, hands-on and positive practices within a firm, appropriate accounting modalities and principles, internal control mechanisms, information flow quality and content and the involvement level of ordinary (bottom-level) employees in the day-to-day decisions of the organization. The existence of these parameters lessens the complications connected to information discrepancies rendering the SMEs less risky to invest in. This paper has presented some knowledge/information on the significance of CG for Ghanaian SMEs; nevertheless, future studies is essential to further advance some knowledge presented in this research in view of the fact that this study generated heterogeneous results/findings. Likewise, more work is needed to ascertain the influence of the other CG variables that were not captured in this study [e.g., Director's skills, director's remuneration, outside block holders, promoter's equity, non-promoter's shareholding, etc.] on SMEs operations in Ghana. This would aid provide preeminent practice criterion/standard to how governance mechanisms can be meritoriously utilized within the SME industry or should be considered as a policy initiative, as the study's findings indicate that the Ghanaian SME industry provides distinctive governance mechanisms.