Corporate Governance and Firms Financial Performance in the United Kingdom

The objective of this study is to examine empirically the impact of good corporate governance on financial performance of United Kingdom non-financial listed firms. Agency theory and stewardship theory serve as the bases of a conceptual model. Five corporate governance mechanisms are examined on two financial performance indicators, return on assets (ROA) and Tobin's Q, employing cross-sectional regression methodology. The conclusion drawn from empirical test so performed on 252 firms listed on London Stock Exchange for the year 2014 indicates a positive or a negative relationship, but also sometimes no effect, of corporate governance mechanisms impact on financial performance. The implications are discussed. Thereby, so distinguishing effects due to causes, we present a proof that, when the right corporate governance mechanisms are chosen, the finances of a firm can be improved. The results of this research should have some implication on academia and policy makers thoughts.


Introduction
The aim of this study is to examine the impact of "good" corporate governance on financial performance of firms in the United Kingdom. Turnbull (1997) defines corporate governance as all the influences affecting the institutional process, including those pointing to the controllers and/or regulators, involved in organising the production, sale of goods and services. According to Ehikioya (2009), corporate governance is concerned with processes and structures through which members interested in the firm take active measure to protect stakeholders' interest.
Corporate governance has become more relevant in contemporary times as companies grow and expand both in developed and emerging economies (Freeman, 1983(Freeman, , 2010. As companies expand, they use local raw materials, employ local workforce, sell to the community, pay taxes, etc., that supposedly benefit the community. In addition, recent corporation scandals have been blamed mainly on "bad" corporate governance. (It is almost a daily occurrence to hear news upon scandals ruining corporations.) Consequences of firms' failure are huge; they can be felt in every aspect of society. For instance investors' capital can be wiped out overnight, job losses can occur, etc. (Mallin, 2016).
There is another side to the story: interest groups known as stakeholders' activities can also affect the corporation. For instance, if some society is discontent with the operations of the corporation, it may react negatively towards the firm. Thus, one can boycott its products. As a result, companies may modify their "usual governance", now focusing on social friendly issues departing from idea of shareholders primacy, -when activities are mainly geared toward maximising shareholders aims (Rodriguez-Fernandez, 2016). In addition, there is some evidence to suggest that investors are willing to pay high premium for shares of firms perceived to have a good corporate governance structure (Clarke, 2007). This affirms why corporate governance mechanisms can be considered related to the financial performance of firms.
Over the past decades, there have been many academic researchers investigating links between corporate governance and firm financial performance. Most of these academic researches point out that good corporate governance has a positive impact on firm's financial performance (Stanwick and Stanwick, 2002); however, other researchers have a different view (Donaldson and Davis, 1991;Jensen and Meckling, 1976;Ehikioya, 2009).
Firms require investors' funding to undertake expansion projects. There is evidence to suggest that corporations that improve good corporate governance mechanisms are able to increase the firm's value by 10% to 12% (Stanwick and Stanwick, 2002). The argument is that, before investors think of investing in corporation, they take into consideration the firm corporate governance mechanisms. According to Weir (1997), a firm for which corporate governance structure is seen as "undesirable" has to struggle to get loans, for example. Mallin (2016) points out that before investors commit their funds to investment activity, they consider indicators like insider shareholder, audit committees, board independence, board size, CEO duality, etc., all related to the corporate structure of the firm. In response, firms are now begun to design programmes of good corporate governance that would be attractive to providers of funds.
Yet, according to Cadbury (2000), corporate governance difficulty arises because of separation between shareholders of the business and its control in response to a system by which corporations are directed and controlled. Sometimes, an agent (manager) may have some opposing interest to that of the principal (shareholder) (Jensen & Meckling, 1976). The problem of conflict of interest can occur as a result of asymmetric information resulting from imperfect contractual agreement between managers and shareholders. Such an information can serve as an incentive to managers to pursue self beneficial business projects at the detriment of shareholders. In addition, the board of directors may find that their business interests collide with their fiduciary duties.
One role of corporate governance is to manage these conflicts between the principals and the agents. Good corporate governance, therefore, should have strong internal mechanisms to manage various interest groups, whence to reduce high agency cost; this was of course discussed already along time ago by Rose-Ackerman (1973) or by Fama (1980).
In the post-Enron financial turmoil in Asia and WorldCom in USA, there was a shift of corporate governance focus from its traditional grounds of agency conflicts to ethical issues such as accountability, transparency, disclosure, and reporting. The public demand for corporate accountability, following the high profile corporate scandals stimulated policy makers, academics, and public/private sectors to strengthen the effort of good corporate governance in corporations (Mallin, 2016). According to Aguilera (2005), the Sarbanes-Oxley Act 2002 for instance was enacted in USA in light of institutional contingencies to ensure that boards of directors adhere to best practices of corporate governance guidelines such as disclosure and honest reporting in corporations. Good corporate governance is to be centred on core principles of accountability, transparency, fairness and responsible management.
Addressing these concerns through business decision-making process has not only benefited investors, but also employees, consumers, and communities by strengthening their voices at general assembly meetings (Gill, 2008).
However, the recent financial crises in 2008 has reinvigorated the debate again as to whether good corporate governance (positively or negatively) influences firms financial performance at all. To help provide unbiased judgement into good corporate governance and the impact on firms' performance, we will research corporate governance mechanisms in UK firms. A random sample of UK firms listed on London stock exchange for 2014 will be selected, thereby avoiding sectorial bias.
A complete theoretical framework based on agency and stewardship theories will aid in answering the research question (Donaldson and Davis, 1991).
Corporate governance mechanisms such as insider shareholder, board size, board independence, CEO duality, and Audit committee meetings will be used in the study. This is in line with studies by Ehikioya (2009) andChristensen et al. (2010). Financial and market performance of the firm will be here captured using following proxies, respectively: (i) Return on assets (ROA); (ii) Tobin's Q (Perfect and Wiles, 1994;Tejersen et al., 2016). These variables will be controlled using firm's size and leverage. We will test a random sampling of 252 firms listed on London Stock Exchange, different from other prior studies in which the sample is mostly picked-up from FTSE100 companies.
Our report is structured as follows: here below, Section 2 contains some literature review.
Section 3 outlines the methodology indicating hypotheses and a description of the variables. Section 4 contains the description of the quantitative results so obtained with a statistical significance discussion.
Section 5 translates such findings into practical considerations, examining all variables. The final section (6) is reserved for concluding remarks and recommendation about future research directions.

Theoretical Background
It has been recalled that a difficulty for nowadays implementing some "good" corporate governance resides in the possibly conflictual relationship between the shareholders and the board of directors. This has been addressed by both agency theory and stewardship theory (Donaldson and Davis, 1991).

Agency Theory
The agency theory details the relationship between the managers (agents) and the shareholders (principals) (Donaldson and Davis, 1991). It seeks to resolve divergent interests between management of the organisation and the owners, prescribing ways of resolving such conflicts, like delegating a decision-making authority to the agents who manage a project.
Along the agency theory, corporations stand a chance to increase financial performance if cost is minimized. The agency cost can be seen as a value loss by shareholders because of divergence in interests of managers and owners (Jensen & Meckling, 1976). In addition, agency costs are captured in the stock market that affects the company's share prices. Therefore if agency cost is properly managed, it can help for improving shares value, that is, it improves the overall financial performance of the firm. According to Jensen and Meckling (1976), agency costs is measured as the sum of monitoring costs, bonding costs and residual costs. Therefore, in order to reduce the agency cost the corporate governance mechanism should unravel causes of these conflicts, whence the need for grasping the "agency theory". The effective corporate governance mechanisms control should encourage managers to act in the best interest of the principal (Allen and Gale, 2001).
There is an assumption in the agency theory that, where there is a welldeveloped market, corporate controls are absent. The consequences lead to market failures, non-existence of the markets, moral hazards, asymmetric information, incomplete contract and moral selection. Various studies however, have suggested that proper monitoring, healthy market competitions, control of executive pay, prudent debt sourcing, efficient board of directors, markets for corporate control and concentrated holdings can help resolving the agency problem (Bonazzi and Islam, 2007). The supporters of agency theory argue that, the role of CEO and chairperson should be assigned to separate individuals.
This will ensure proper check and balances between CEO and the chair person (Gillan, 2006).

Stewardship Theory
Unlike the agency theory that suggests that the role of CEO and chairperson should be separated, the stewardship theory argues that both roles should be combined. The stewardship theory suggests that directors are able to achieve organisational objective of shareholders by maximising their utility rather than self-serving. Some available empirical evidence supports the side of this argument of stewardship theory (Donaldson and Davis, 1991).
Moreover, stewardship theory predicts that allowing managers to work with discretion can encourage them to work better. Scholars on this side of the debate concur that managerial behaviour is not only driven by financial reward but also requires discretion to enable them to maximise the shareholders' value.
In addition, stewardship theory stresses that the concern of managers for their reputation and their career intended progression compel them to act in the interest of shareholders; therefore, agency cost will be minimized (Donaldson and Davis, 1991). There is a psychological side of the argument that managers are able to give up their best when they have job satisfy. Clarke (2004) points that allowing managers to take decisions on their own without having to go through bureaucratic processes improve job satisfaction that contributes towards the overall financial performance of the firm.
Besides, Fama and Jensen (1983) agued that managers have greater access to specific insider information, about the going concern of the organisation, than independent directors. Therefore, managers are expected to have acute knowledge of the operations of the company that will help them make well informed decisions. In that line of thought, the stewardship theory suggests that a low number of independent directors is ideal for companies (Donaldson and Davis, 1991;Christensen et al., 2010). In addition, the stweardship theory affirms that insider-dominated board of directors is more effective in achieving the organisational objective because of finer accessibility to information and technology. Finally, the stewardship theory maintains that the CEO essentially wants to work well rather than opportunistically exploits the system, -as also suggested by the agency theory (Donaldson, 1990).

Insider Shareholder
Insider shareholder is a term used to describe a director or senior officer of a corporation who owns some shares of a corporation, -usually more than 10% of the voting shares (Jensen and Meckling, 1976). According to Jensen and Meckling (1976), the size of the shareholding by the insider has effects on the general financial performance of the firm. Jensen and Meckling (1976) observed that a rise in insider shareholding by insiders does reduce the agency cost. The logic behind this finding was that managers who own significant shares of the company would not invest in destructive or excessive high-risk projects.
Therefore, by principle, managers will prudently invest in projects that are likely to reap high returns.
Several studies show indeed that increasing the proportion of insider shareholding beyond an optimal point reduces financial performance. For instance, Fama and Jensen (1983) point that such an increase can result in managerial entrenchment. Recently, Gupta and Sachdeva (2017) tested a comprehensive data set of hedge funds on financial performance of firms with much or little insider shareholding, -using multiple linear regression models. It is found that firms with insider investment perform better than others. The findings also support the view that increases in insider shareholding, up to an optimal point, about 20% of shares, (could) increase returns.
McConnell and Servaes (1990) had also found that an increase in insider shareholder increases the firm's performance, but beyond 40% to 50% a decline in firm's performance occurs. Yet, Agrawal and Knoeber (1996) reported that, insider shareholder predictive effects disappear when additional corporate governance mechanisms are included -in a single ordinary least square (OLS) regression.

Board Size
The theories of economics show that the board of directors plays an important role in the corporate governance structure of corporations (Fama and Jensen, 1983). The concern of shareholders has to do with whether the board of director is capable to monitor/control managers to act in the interest of the owners. The general notion is that companies that have a large board size are likely to have effective supervision that can improve firm performance. Anderson et al. (2004) and Williams et al. (2005) argued that a large board is likely to possess specialised skills prerequisite for efficient toward better performance. Haniffa and Hudaib (2006) also obtained a positive relationship between board size and financial performance.
Another hypothesis about a small board size inducing a better performance has been presented by researchers arguing that limiting a board size rather improves communication and decision-making (Lipton and Lorsch, 1992;Jensen, 1993;Yermack, 1996;Christensen et al., 2010;Akshita and Sharma, 2015). Lipton and Lorsch (1992) suggested that a board member number should not exceed 10. Yermack (1996) discovered an inverse association between board size and market valuation measured by Tobin's Q. In this respect, Akshita and Sharma (2015) discovered an interesting finding that a large number of board directors is considered to be an expensive affair for a firm, thus affecting firm's performance.

Board Independence
Both agency theory and stewardship theory predict different outcomes depending on the board composition.
According to the agency theory, the board of directors can monitor effectively if these are independent from the management (Fama and Jensen, 1983;Beasley, 1996;Christensen et al., 2010). The argument is that incentives exist for outside directors to protect their reputation that motivate them to exercise decisional control (Fama and Jensen, 1983;Christensen et al., 2010). Beasley (1996) argues that where there are non-executive directors on the board financial statement fraud is unlikely to occur. Yekini et al. (2015), employing content analysis and panel data set from UK FTSE350 companies, discovered a significant relationship between board independent and information disclosure measured by the proportion of nonexecutive directors. Their research shows that firms with non-executive directors are more likely than others to disclose information which can improve company performance. Rosenstein and Wyatt (1990) argued that the proportion of independent directors has a positive impact on company's share price and financial performance. Both Yekini et al. (2015) and Rosenstein and Wyatt (1990) support the view of agency theory that non-executive directors can improve company performance because of ability to monitor managers.
In contrast, the stewardship theory argues that inside directors have in-depth knowledge of the company which makes them aware of valuable resources that improve firm performance (Donaldson, 1990).
Other scholars argue in support of stewardship theory that, inside directors are trustworthy stewards of firms' resources and improve company performance because of information asymmetry (Donaldson and Davis, 1991;Nicholson and Kiel, 2007). Agrawal and Knoeber (1996) and Klein (1998) discovered a significant negative association between the number of independent directors and performance of firms.

CEO duality
In some companies, a CEO may have two functions; he serves as chairperson of the board of directors and as executive manager (Elsayed, 2007). Corporate governance guidelines presume that when a CEO is also the chairperson of the board, this leads to concentration of power (ASX, 2007). The primary concern of CEO duality is that, managerial domination of the board of directors can lead to dubious control of meeting's agenda (Firstenberg and Malkiel, 1994). In this regard, CEO/chair may decide to send information that serves personal interest only to the board of directors. Consequently, in corporations where there is a lack of strong monitoring of corporate governance mechanism, management can rather pursue their self-interest (Fama and Jensen, 1983).
According to Lorsch and MacIver (1989) In contrast, the supporters of stewardship theory maintain that duality of CEO/chairperson should rather lead to superior firm performance. Stoeberl and Sherony (1985) argue that duality of CEO allows clear-cut leadership direction for strategy formulation and implementation, that is good for business. In addition, other scholars have said that because powers reside in one person uncertainty with regards to the identity of the person taking responsibility of decision is reduced (Christensen, et al., 2010). Therefore, companies can achieve superior performance when there is duality of CEO. Cannella and Lubatkin (1993) documented a positive association between CEO duality and ROE. Boyd (1995) and Essen et al. (2013) came out with same conclusion.

Audit committees
The role of the audit committee is to ensure that the integrity financial reporting of the corporation meets corporate governance council standard. It also ensures compliance of entities such as mandatory disclosures (Davidson et al., 2005) Kent and Stewart (2008) discovered that the quantity of disclosure was positively related to frequency of board and audit committee meetings held.
However, there is some conflicting evidence from other scholars work. Klein (1998) discovered that the presence of audit committees do not have any effect on the quality of accounting performance measures. Vafeas and Theodorou (1998) also find no evidence to support that a relationship exists between performance and the "board structure (director affiliation and ownership, chairman affiliation, and committee composition)".

Overview of previous studies relevant to UK
Concerning corporate governance mechanisms on UK's corporation, there are mixed results. Guest (2009) indicated that board size has a strong negative impact on profitability, Tobin's Q, and share returns. According to Guest (2009), UK boards play a weak monitoring role; therefore any influence of large board size is likely to reflect the malfunction of the advisory board. In short, Guest's study supports the argument that a large board size is a hindrance to good communication and effective decision making.
Florackis (2005) discovers the existence of "non-linear impact of managerial ownership and managerial compensation on company performance". He finds a strong evidence that managerial ownership and managerial compensation can work as alternative mechanisms in mitigating agency costs and, therefore, generating good financial performance. Weir et al. (2002) analyzed "the relationship between internal and external corporate governance mechanism on performance of UK firms within the context of Cadbury Committee's Code of Best Practice". They discovered a "weak relationship" and documented that there is no evidence to support that "firms on top or bottom performance deciles have different corporate governance characteristics". Weir et al. (2002) also raised an argument that it will not be right to impose a corporate governance mechanism on a firm given that market for corporate control is known to be a set of effective means for reducing the agency cost. The Weir et al. (2002) work supports the view that CEO shareholding can cause entrenchment resulting in poor firm performance.
Mura (2007) also documented a weak relationship between non-executive director shareholding and firm performance. But, he discovered that, the size of independent directors have a positive impact on firm performance.
Recently, Al-Najjar (2017) discovered that board size and board independence have a significant impact on the pay of CEO and firm performance for several UK firms.

Methodology
In this section, we illustrate the statistical analysis of empirical data for the variables and their indicator involved in this study. We use multiple regression models with research hypotheses. In addition, we use cross-sectional regression analysis to test empirical data, again in accordance with prior literature as this study is for one year; see Rodriguez-Fernandez (2016) and Watsham and Parramore (1997) supporting the use of cross-sectional data to test variables on one year. We have used a software data analysis package in excel to test the data along a multivariate analysis to obtain descriptive statistics of the total variables, such as mean, standard deviation, minimum, maximum, coefficient of variation, skewness and kurtosis.

Timeframe and Statistical Analysis Model
Next, we use the correlation method to estimate the relationship between independent, dependent, and control variables. Multiple linear regressions are "finally" employed to test corporate governance mechanisms on firms' financial performance.

Research and Sampling Design
The study uses the cross-sectional data method to test a sample of firms listed on London Stock Exchange for the year 2014 (LSE, 2014). The study is restricted to listed firms because they are expected to adhere to set regulation standards. In addition, listed firms are likely to prepare their accounting figures in compliance with international accounting practice (Ehikioya, 2009). We stress that we excluded financial institutions because they are subjects to different regulations from non-financial firms, whence may lead to outliers . In fact, mentioned scholars in prior reports have done likewise (Gust, 2009;Rodriguez-Fernandez, 2016). We have not taken into account the possibility of cross share holding (Rotundo and D'Arcangelis, 2010; D'Arcangelis and Rotundo, 2015;Cerqueti et al., 2018).

Hypotheses Development
After reviewing the literature from the prior studies, five hypotheses emerge: H1: Companies with large insider shareholding are those with superior financial performance (Jensen and Meckling, 1976;McConnell and Servaes, 1990).
H2: Companies with large board size achieve superior financial performance (Anderson et al., 2004).
H4: Companies with CEO duality achieve less financial performance (Firstenberg and Malkiel, 1994) H5: Most high financial performance companies are those with high frequency audit committee meetings (Kent and Stewart, 2008)

Description of Variables and Measure Indicators
This section covers descriptions of variables used in the study. These include dependent, independent and control variable. In addition, we will indicate measurement and proxies use to measure variables of corporate governance mechanism and their relationship with financial performance; see Table I. According to Core et al. (2006) operating profit measured by ROA is a better measure when examining the relationship between financial performance and corporate governance. For example, ROA is not affected by leverage, extraordinary items, and other discretionary items. In addition, other researchers (Brown and Caylor, 2009;Muth and Donaldson, 1998) have used ROA as a measure of accounting. Based on these factors and previous studies, we use ROA in this study.
Secondly, the forward-looking financial market measure Tobin's Q is used in this study. This is consistent with the efficient market hypothesis established by Malkiel and Fama (1970) where Tobin's Q was used to capture existing assets and future growth potentials of the company. Tobin's Q also captures investors' expectations to future events, including evaluation of current business strategies (Rose-Ackerman, 1973; Demsetz and Villalonga, 2001;Ehikioya, 2009;Christensen et al., 2010;Rodriguez-Fernandez, 2016). Let us describe the dependent variables:

(i) Return on Assets (ROA)
The Return on Assets (ROA) gives an indication of how best the assets of a company is utilised to generate profit. The ROA is calculated by dividing annual earnings of the company by its total assets.

(ii) Tobin's Q
The Tobin's Q is a ratio of market value of company outstanding stock and debt divided by replacement cost of the company's assets ("book value") (Christensen et al., 2010). =

Independent Variables
The corporate governance mechanisms recalled in the empirical framework section are going to be the independent variables of this study. They are:

(i) Insider Shareholding
The Insider shareholding refers to any director, corporate officer or institutional investor who owns at least 10% of the total shares of a corporation (Jensen and Meckling, 1976). Insider ownership is measured as the percentage of company outstanding shares owned by such insiders: The Company board size refers to the number of members on the board. There is some evidence to suggest that a large board size results in better decision making than a small board size thereby leading towards high financial performance (Williams et al., 2005).
The Independent board refers to outside board directors who are not affiliated to top executives of the firm (Fama and Jensen, 1983).

(v) Audit Committee Meetings
Audit committee meetings occur when the board of directors charged with the responsibility of financial reporting and disclosure of information for the company. It is argued by scholars that the frequency of audit committee meetings is strongly related to the performance of a company. The logic is that regular meetings will mean that more information can be obtained and disclosed (Christensen et al., 2010). (2002) and Essen et al. (2013). These variables are estimated through:

IS
Insider-shareholding, measured as the proportion of shares owned by insiders.

BS
Board-size, measured as the number of board of directors on company's board.

IB
Independent board, measured as proportion of independent board on company's board.

CD
CEO duality, measured as a Function of board chair person combined with CEO, CEO = 1 if CEO is also chairperson, otherwise = 0.

AC
Audit committee meetings, measured as a function of the number of audit committee meetings held.

FS
Firm size measured as the logarithm of the firm's total assets.
LG Leverage, measured as total assets/total shareholders' equity.

Data Analysis and Discussion
This section contains the discussion of the empirical data use in the study. A correlation analysis is employed to show the relationship between corporate governance variables and either ROA or Tobin's Q. A regression analysis is presented to show how independent corporate governance mechanisms can either positively or negatively affect the dependent variables ROA or Tobin's Q.

Descriptive Statistics
The usual statistical characteristics, including mean, minimum, maximum, standard deviation, coefficient of variation (Cov.), skewness (Skew.), and kurtosis (Kurt.) are reported in Table II for the dependent, independent and control variables for a sample of 252 firms listed on London Stock Exchange data, extracted from Bloomberg.
ROA has a large standard deviation showing that the data is largely spread around the mean, whence the Coefficient of variation = 189%. This points to a high variation in the accounting-based performance among the UK's firms. For the Q ratio, the closeness of the mean and standard deviation signifies that the market-based performance among UK firms are closely netted. This is reflected in the relatively low coefficient of variation = 92%.
With reference to the independent variables, which represent corporate The statistical results for audit committee meetings held among UK firms show a low standard deviation=1.80 for the mean= 4.50, giving a low coefficient of variation = 40%. Notice that the (AC) data ranges from 0 (!) to a Max.= 14.
With reference to the control variables, the logarithm of assets has a mean = 3.3 and a standard deviation = 0.70 giving a low coefficient of variation = 21%.
The minimum value = 1.10 and the maximum = 5.55. Here the data clusters around the mean, which implies a low size variation of such firms. The leverage shows a mean of 2.95 and standard deviation of 4.89 giving a coefficient of variation = 166% in data spread. The data ranges from -33.59 to 46.96. Such a negative leverage implies that the cost of borrowing is greater than the return on investment. Thus, variations in firms' debt are rather consequent as confirmed by the 166% coefficient of variation value.
About the skewness, apart from ROA=-1.92 and IB=-0.49 that are negatively skewed, the remaining variable distributions present a positive skewness, indicating that the tail of these is on the right. Also in terms of kurtosis, with the exception of variables CD=59.21 and LG=44.92 that have heavy tail or outliers in the data distribution, the remaining variables have light tails or few outliers.
These features point to an "interesting" random selection.

Correlation between variables
This section on correlations will help to determine whether there is multicollinearity among any of the variables. We noticed that prior researchers have raised concerns of possible multicollinearity among variables which could thereby distort the estimates of the regression results. In addition, because this research considers data for only one year period, there is no heteroscedasticity problem (Alin, 2010;Koop, 2008;Gujarity and Porter, 2009).  According to Gujarity and Porter (2009), a correlation above 0.8 signals a possible evidence of multicollinearity in the data set. The results in Table III indicate that multicollinearity is unlikely to be a problem here. Nevertheless, there are mixed results: several variables are negatively correlated to ROA, whereas others have positive correlation with ROA. "Interestingly", the correlation is negative for Concerning Tobin's Q and corporate governance mechanisms (Table IV), again, multicollinearity is unlikely to be a problem as none of the variables has a correlation above  LG -0.09 -0.03 0.00 0.05 -0.01 0.06 0.10 1.00

Regression Analysis
In this section, the multiple linear regressions models, see Sect. 2.5, are used in order to establish the impact of corporate governance mechanisms on the two response variables, ROA and Tobin's Q. The statistical results are given in Table V and Table VI respectively. However, additional change in AC, will result that ROA on average would decrease by -0.8 %, because of the negative coefficient.
In addition, R 2 =0.12, see Table V, implies that only 12% of all the independent and control variables explain the effects on the dependent variable ROA, whence 88% of ROA behaviour has to be explained by other independent variables not included in this study.  Significant level: 0.000 Therefore, one can make a prediction that for a unit increase in BS, holding all other factors constant, the Q ratio on average will increase by 19%. However, all things being equal, a unit increase in FS will have a corresponding average decrease equal to -113 % of the Q ratio.
An important feature should be emphasized: the independent variable IS is not significant in either models. This means that insider shareholding does not influence financial performance (measured through ROA and Tobin's Q). This confirms the findings of Agrawal and Knoboer (1996). However, this finding disagrees with Jensen and Meckling (1976). Therefore, our first hypothesis that companies with a large insider shareholding are those with superior financial performance can be rejected.
A contrario, BS is statistically significant in both models, implying that a large board size could improve financial performance (ROA and Tobin's Q). This finding is consistent with Anderson et al. (2004). Therefore, our second hypothesis is acceptable.
Concerning IB, which presents a statistically significant effect in ROA, one concludes that additions to the board of independent directors will improve financial performance. This is consistent with Beasley (1996) and Donaldson (1990). One could consider that the third hypothesis can be accepted concerning ROA. However, in terms of Tobin's Q, IB is statistically insignificant: the independent board of directors has no effect on financial performance. One should reject the third hypothesis for Tobin's Q.
In terms of CD, there is a lack of statistical significance for both dependent variables (ROA and Tobin's Q), suggesting that it does not matter whether there is a dual role or a separation of CEO and chair role: the financial performance remains unaffected. This finding is inconsistent with Firstenberg and Malkiel (1994) who suggest that companies with CEO duality do not perform well financially. Therefore, the fourth hypothesis is rejected.
Regarding AC, a statistical significance is obtained for ROA but with a negative coefficient. This suggests that increasing the frequency of audit committee meetings impacts negatively on the financial performance (ROA). This finding disagrees with that of Kent and Stewart (2008). However, in terms of Tobin's Q no statistical significance is found, which suggests that the frequency of audit committee meetings lacks some predictive ability on the financial performance (Tobin's Q). This finding is consistent with Weir et al. (2002). Therefore, the fifth hypothesis is also rejected.

Discussion
The above results explain why there is a controversy in this field of study. The statistics so obtained provide mixed results, depending on the Model. There are corporate governance mechanisms that have no statistical significance; some have positive, and others have a negative statistical significance on estimating financial performance, we stress, using ROA or Tobin's Q.
The findings show that, insider shareholding has insignificant influence on both ROA and Tobin's Q. This supports the findings of Agrawal and Knoeber (1996). The implication is that whether managers own many or a few shares in a company is irrelevant for the financial performance. This means, there should be no hindrance to pay executive bonuses in shares instead of salary, in order to increase insider shareholding.
One field of controversy is board size and its impact on financial performance.
The outcome of the study indicates a positive statistic significance of board size on the two financial performance ratios (ROA and Tobin's Q). It is seen that increasing the size of the board improves financial performance contrary to the argument of Jensen (1993). The findings in this study, however, support the argument of Anderson et al. (2004) that large boards help in proper allocation of committee work for enhancing growth and financial performance. This argument supports the views of Fama and Jensen (1983) who argued that the role of the board involves monitoring managerial behaviour, which is likely to be more effective with a large board size. In this respect, one can follow Williams et al. (2005) arguing that financial markets place a high premium on large board size, perceived to be better resourced for monitoring and or skills transfer abilities.
Both Donaldson (1990) and Beasley (1996) argued in favour of a high proportion of board independent members. They documented that companies displaying high proportion of board independent directors achieve a high financial performance. For instance, Donaldson (1990) stressed that those companies command creditability in accounting, whence investors seeing this have a favourable opinion. The outcome of our study suggests that there is a statistical significance of an independent board on ROA, as in Donaldson (1990) and Beasley (1996). However, an insignificant test result is discovered for board independent influence on the Q ratio, as in Fosberg (1989 imagination and studies are needed by researchers about this theoretical deficit.

Summary and Concluding Remarks
This study has examined the impact of 5 corporate governance mechanisms (insider shareholding, board size, independent directors, CEO duality, and audit committee meetings) on financial performance (ROA and Tobin's Q), taking into account 2 control variables. The study covers a sample of 252 firms listed on London Stock Exchange in 2014. Two theories of corporate governance, agency theory, and stewardship theory, form the theoretical framework. The outcome of the regression results displays mixed findings similar to prior studies (Christensen et al., 2010;Rodriguez-Fernandez, 2016).
For instance, many prior studies suggest that the size of insider shareholding affects the financial performance (Jenson andMeckling, 1976: Fama andJensen, 1983;Gupta and Sachdeva, 2017). However, the outcome indicates that insider shareholding has no influence on financial performance, itself consistent with findings of Agrawal and Knoeboer (1996).
Some of the corporate governance mechanisms such as board size and independent board members exhibited predictive power on both financial performance indicators, ROA and Tobin's Q. This finding is in agreement with Christensen et al. (2010) further concluding that a strong independent board is one of the solutions to agency problem by reducing cost, thereby improving financial performance. Somewhat inconclusively, the frequency of audit committee meetings indicates some influence on the financial performance indicator ROA but no influence on Tobin's Q.
Finally, our study about CEO duality demonstrated no influence on both ROA and Tobin's Q. This finding disagrees with the conflicting prior literatures having examined the variable. The supporters of agency theory suggest a positive outcome when the role of the CEO and the chairperson is separated (Rechner and Dalton, 1991;Balatbat et al, 2004). On the other hand, supporters of stewardship theory argue for role of the CEO and the chairperson to be combined as it allows clear leadership direction that improves performance (Stoeberl and Sherony, 1985). However, our finding demonstrates a neutral cause; it does not matter whether the CEO and chairperson's role is combined, or otherwise; the outcome of financial performance remains unchanged whatever the board choice.
To help achieve a robust finding, the corporate governance mechanisms were controlled by firm size and leverage. For the firm size, the regression coefficients are negative, but have opposite values as concerns the leverage.
Notice that we have not taken into account the possibility of cross share holding even though some thought should be given on the matter (Rotundo and D'Arcangelis, 2010;D'Arcangelis and Rotundo, 2015;Cerqueti et al., 2018), investigating "cross performances".
Thus, our study, like previous studies, provides mixed findings and partial conclusions to the debate. However, it has strengthened some of the existing theoretical framework. We can conclude that companies in the United Kingdom can improve their financial market performance by adopting the right corporate governance mechanisms. We have indicated which (among others) corporate governance mechanisms influence financial performance indeed.
Thus, future researchers could explore other theories like stakeholder theory, shareholder theory, leadership cycle theory and others in order to introduce other variables in the considerations, such as board diligence or CEO tenure. In addition, other factors such as technology, global financial crises, economic conditions (booms and recessions), cross share holdings, … can be investigated, as they might likely have some impact on financial performance. Furthermore, panel data can be employed to test variables over several years with other data sizes. Brexit influence will likely attract new considerations.

Acknowledgements
This work is partially based on a dissertation submitted by MK to University of Leicester, School of Business, as a fulfilment for the M. Sc. Degree in Finance

Data Availability statement:
The data that support the findings of this study are available from [Bloomberg : https://www.bloomberg.com/professional/ ] Restrictions apply to the availability of these data, which were used under license, at the University of Leicester Library, for this study. Data are available from the authors with the permission of University of Leicester and Bloomberg