The effect of corporate governance on financial performance of companies within the UK
This study aims to examine the role of corporate governance on financial performance of firms listed on London Stock Exchange. In order to tackle this aim, this research measures the effect of diversity, board independence, and size of the board on the financial performance of non-financial companies within FTSE 150 the period of 2013 to 2017. The data of 570 observations were analysed using Fixed Effects Model, Randon Effect Model, Hausman Testing Process, and Breusch-Pagan Testing Process. The findings indicated board independence has significant positive impact on financial performance of companies. Also, the findings suggested a slightly weak significant negative relationship between size and the performance. However, board gender diversity has inconclusive effect on the performance. In regard to limitations, this study adopts Tobin’s Q as a tool for measuring financial performance of the companies. This would neglect other important tools such as Return on Assets and Return on Equity. Moreover, the time of this study was 5 years. This can be extended to eight or 10 years which would increase the reliability of the findings. Also, the observations excluded all financial companies which are governed by different regulations. In regard to future studies, Modelling the impacts of Brexit and the opportunities or challenges it presents to corporate governance would be a promised idea of research to look further.
Chapter One: INTRODUCTION
Corporate governance has a number of interpretations but the general idea revolves around ensuring that the interests of shareholders are safeguarded to ensure maximum return on their capital. It can be defined as the system by which a business entity is controlled and directed (OCED, 1999). The structure outlines the sharing of rights and responsibilities by the various stakeholders by spelling out the rules and procedures on the decision making process. The most important aspect in ensuring the protection of investor rights is the board of directors. This is crated to ensure that the rights of the investors are safe guarded and aspects such as board size, independence or even frequency of meetings have all been used as proxies for measurement of corporate governance.
Financial performance of a company is a measure on the return of the capital of the shareholders and is an indication of the financial health of a company. It is measured by use of various proxies that can either be market based or accounting based. Studies have shown that there is a link between corporate governance and the financial performance of companies such as those by Jensen (1986), Coles et al., Hafsi (2013) among others.
This study attempts to look into the relationship between the financial performance of companies and the corporate governance structures in the UK. Studies have shown that companies with a strong governance structure tend to have more success compared to those that do not (Coles et al., 2008). The UK has been undergoing a number of changes in both the economic and political aspects such that the changes in regulation governing these changes needs to be studied and the impact assessed for a proper understanding for businesses. The FRC has been responsible for a number of changes that are associated with various aspects such as board independence and diversity and according to the UK Board Index (2017), there has been more emphasis on revised auditing and ethical standards alongside the improvement of gender, foreign and ethnic representation in boards of listed companies.
In 2016, the government announced a review of the female representation in leadership positions in companies with a target of about a 25% representation in the FTSE 100 boards, which was achieved, and the focus has now changed with the push being emphasized on this percentage for the FTSE 150 companies. This push will be one of the things the study will attempt to analyse to see if there is a relationship between it and the financial performance of FTSE 150 companies.
1.1. Problem Statement
The push by the FRC on the presence of non-executive independent directors into the boards of many UK companies alongside the new proposition of having a 25% representation of females on boards of FTSE 150 companies warrants an investigation as to whether these changes will have a positive or negative influence on the financial performance of companies. According to the Cranfield (2017) report, the inclusion of women in the board of companies will have an effect of improving the productivity of companies and is thus a vital aspect to be considered over the next decade. The report argues that employment of women is at its highest and that if the participation of women and men could be equalized, the GDP of UK could increase by up to 10% by the year 2025. Such assertions are important to investigate empirically which is why the study is necessitated and will try to see how female representation in the board alongside the independence and board size have an effect on the financial performance of FTSE 150 companies.
1.2 Research Objectives
The study will investigate if the changes in the UK governance code in terms of board gender diversity, independence and board size will have an effect on improving the financial performance of listed FTSE 150 companies. In doing so, the study attempts to address the following objectives;
- Investigate the relationship between gender diversity and the financial performance, as measured by Tobin’s Q, on listed non-financial FTSE 150 companies.
- Investigate the relationship between board independence and the financial performance, as measured by Tobin’s Q, on listed non-financial FTSE 150 companies.
- Investigate the relationship between board size and the financial performance, as measured by Tobin’s Q, on listed non-financial FTSE 150 companies.
1.3 Significance of the study
This study will help various stakeholders in understanding the effects of certain corporate governance aspects on the financial performance of companies. One such stakeholder would be regulators who would need to understand the relationship of how various factors affect performance so as to beef up the rules and regulations to avoid incidences of financial misconduct of companies which has over the years shown to have an effect of eroding the value of public companies. Furthermore, the study will aid companies in understanding how the inclusion of females in their board will affect their performance and form a basis for incentivizing companies on adhering to the suggestion of the FRC and this can help the companies improve their performance and even increase likelihood of access of capital on the global scene as this is something that has been a global interest. Additionally, the results of the study will help form a basis for further research on the area of corporate governance thereby help improve the body of knowledge available to further this field and enhance understanding on the possible relationships between gender diversity and other relevant corporate governance variables.
1.4 Summary of Chapter
This chapter provides a background introduction of the topic and what has been done in this area and highlights the problem statement that makes the study vital. It looks at a number of objectives the study will look into which is the effect of ender diversity, board independence and size on the financial performance of companies as measured by Tobin’s Q. The chapter then highlights the reasons why the study will be significant to the various stakeholders such as regulators, companies and other academic research on the area.
Chapter Two: LITERATURE REVIEW
This Chapter is divided into two subsections. Subsection 2.1 which will highlight the main theories adopted in corporate governance (agency theory, stewardship theory, and stakeholder theory). This includes the definition of each theory and its pros and cons and their application in the UK rules of corporate governance. Subsection 2.2 which explain the development of the adopted hypotheses in this research. In this subsection, various studies from different contexts will be explored so as to show the various distinctive and contradicted results. This is done through exploring the theoretical association between each independent variable (gender diversity, board independence, and board size) on the one hand and financial performance on the other hand. Then, this research will explore distinctive studies in various developing and developed states that explored this relation between independent and dependent variables.
2.1. Major corporate governance theories
2.1.1. Agency theory
Alchian and Demsetz (1972) were the first to develop the agency theory based on the economic theory and Jensen and Meckling farther developed this theory in 1976. This theory is framework depicting the ideal relationship and responsibilities between the principals like the shareholders and the agents like the executives or managers. This theory focuses on the ownership of the shareholders that is retained by them as principals but the operations are relegated to the agents appointed by the principals (Yusoff and Alhaji, 2012). The main contribution made by the theory is the division of the parties related to the company into two groups, namely principals and agents, where the interests of the groups are seen as pear if not outright conflicting (Lan and Heracleous, 2010).
This theory points out the difference between the theoretical concept where the agents must act to protect the interest of the principals but does not always coincide with h decisions made by the agents in reality (Mustapha and Che Ahmad, 2011). The agents interests like the self-interest, opportunistic, behaviour and other factors are outlined in the theory as the main factors that might cause conflict of interest between the agents and the principals. However, the main development caused by the theory is the separation of the ownership and control. The suggestion made by Holmstrom and Milgrom (1994) regarding the stopping of the incentive payment and paying fixed wage is therefore inadequate in addressing the issues foreseen by the theory regarding corporate misconduct. The positivist approach is comparatively more successful where the rules set by the principles would need to be followed by the agents, which would significantly provide more individualistic solution for the different scenarios (Mitnick, 2015.). Thus, in the opinion of Miller and Sardais, (2011), this theory helps understand the relationship between shareholder and executives in corporate decisions.
The agency theory focuses on the differences between the interests of the principals and the agents, which is acknowledged by the initiative to align the goals of the two parties. According to Siebels & zu Knyphausen‐Aufseb, (2012), this factor, while insignificant in case of family firms where the family members take management positions, however, take huge importance where the agents are employed. This suggests accountability in the employees that is governed by rewards and punishments based on the achievements. The corporate governance needing to meet the needs of the shareholders while retaining their objectivity in face of individual interests of the agents is the ore outlines as an objective need of the corporate governance in the theory (Westphal and Zajac, 2013). Within the context of the UK, the agency theory is implemented in some issues such as empowering stakeholders a voice while doing general meetings as mentioned in Part 9 of the Companies Act 2006. Also, the UK rules (e.g. section 423 of Companies Act 2006) oblige the companies to do information disclosure. These measures would put parameters on the rights of agents so as to limit their rights and increase their responsibilities.
2.1.2. Stewardship theory
The stewardship theory developed by Davis, Schoorman & Donaldson (1997) is based on the sociology and psychology that states, “A steward protects and maximises shareholders wealth through firm performance because by so doing, the steward’s utility functions are maximised”.
In this theory, the stewards or appointed executives are responsible for the protection of the interests of the shareholders as that is their sole purpose. Thus, the theory focuses on the role of the management and executives instead of the individual interests of the executives. This perception aligns the company success with the personal achievements and interests of the executives (Filipovic, Podrug and Kristo, 2010). Unlike the agency theory where the personal motivations of the executives are based solely on economy, the stewardship theory focuses on the autonomy of the executives based on trust rather than economic or other incentives. This theory, therefore, outlines the position of the employees to maximise shareholder profit by inducing autonomous behaviour instead of controlling and monitoring. According to Glinkowska and Kaczmarek, (2015), the firm performance, when connected to the individual reputations of the executives, automatically aligns their interests with the shareholder, as profit maximization is one of the main indicators of the success of the firm. Thus, the successful career of an executive as a steward relies on their achievements on behalf of the shareholders (Okpara, 2011). The Japanese employment model is a prime example of this phenomenon as the personal ownership of the job role motivates the employees to give their best effort (Hernandez, 2012). This has also been aligned with the interest of the executives to gain reputations among the shareholders so that they are trusted with higher responsibility in the future.
The reduction of cost and conflict by allocating the responsibility of the chairperson to the CEO is, therefore, part of the practical application of the theory where the role of the executives as stewards are emphasised. According to Van Puyvelde et al., (2010), the individual success of the two theories is limited compared to that joint application.
Within the context of the UK, legitimacy and accountability of the board of directors are important facets of the application of stewardship theory. According to Companies Act 2006, section 414A, the directors are obliged to do “strategic report” which covers issues such as using financial main indicators so as to analyse the performance of the company, explaining the directors’ strategy in depth as well as showing their professionalism.
2.1.3. Stakeholder theory
The stakeholder theory was initially theorised in 1970 and developed gradually into a cohesive framework including a broad range of stakeholders and their accountabilities and responsibilities in 1984 by Freeman. Stakeholder theory had been developed by combining the sociological and organisational aspects of the corporate responsibility. This theory is probably the most loosely defined of the governance theories and resemble a combination of disciplines including political theory, philosophy, economics, ethics and law into a research framework.
The stakeholders can be defined as “any group or individual who can affect or is affected by the achievement of the organization’s objectives”. That directly opposes the relationships that were conceived in the agency theory. This theory divides the responsibility of the executives towards several groups in the organisation other than the shareholders (Filipovic, Podrug and Kristo, 2010). The suppliers, employees and shareholders all were included in the stakeholders groups whose interest the executives were required to protect. While Christopher, (2010) suggests that, the stakeholder theory helps recognise the various parties requiring the attention of the management while others argue that all of the groups enter a business with their own objectives that might conflict or align with one another’s. While the intermittent goals of the various stakeholders might be different, the main objective is the creation of wealth which is the main objective also outlined by the other corporate governance theories (Tricker and Tricker, 2015).
Freeman (1984) opines that the interrelation of the various stakeholder groups and their interests concern both process and outcomes for the firm performance and the individual performance which transcend the economic view regarding corporate governance. Donaldson & Preston (1995) also states that the theory as practical as the arioso needs of the stakeholders are seen as equal and one groups interests are unable to dominated the others like the other theories.
2.2. Development of Hypotheses
2.2.1. Gender Diversity and Company performance
188.8.131.52. The theoretical association between a firm’s financial performance and gender diversity
The board diversity has been one of the most explored dimensions of corporate governance in light of the failures in the world, which clearly makes the distinction between the demographic diversity and structural diversity (Kaczmarek, 2017). The behavioural theory of corporate governance supports the notion of the gender diversity as important aspect of the board composition in the various literatures as the social perception of the company. Diversity archetypes that have been the topic of the numerous researches also support the concept (Moreno-Gómez, Lafuente and Vaillant, 2018). The heterogeneous nature of the board under the various archetypes of diversity is seen as one of the most valued asset as it is understood higher level of diversity allows better understanding of the market and the segmentation (Joecks, Pull and Vetter, 2013). This is also conducive to higher level of creativity in the solutions to agency problems, which also supports the gender diversity affecting the firm performance. The upper echelon theory also emphasises the importance of gender diversity as beneficial to decision-making, which have influenced numerous studies in this matter (Adusei, Akomea and Poku, 2017).
184.108.40.206: The empirical association between a firm’s financial performance and the gender diversity on board in developed countries
A qualitative study on UK companies with female directors revealed that there are no significant relations with the number of female representation of the board with the financial performance of the firm (Gregory‐Smith, Main and O’Reilly III, C. A. (2014). On the other hand, the study on a sample of 612 European banks from 20 European countries by De Cabo, Gimeno and Nieto, (2012), it has been found out that while there is no relation between the board composition or female representation on firm performance there is positive relations between the risk-averse decision-making with higher female representation on the board. According to a study conducted on a large sample of Spanish companies with female board members correlates the female representation with positive financial performance using Blau and Shannon indices (Campbell and Vera, 2010). This also reveals that the present negative correlation is negligible compared to the positive impact (Hoogendoorn, Oosterbeek and Van Praag, 2013). The study by Boulouta, (2013) indicates that the positive correlation of firm performance might be dependent on the higher CSP that is generated through female influence in board decisions.
220.127.116.11: The empirical association between a firm’s financial performance and the gender diversity on board in developing countries
The gender diversity in the composition of the board of directors has had mixed impact on the various firms in the developing countries as the various studies reveals different levels of impact (Low, Roberts and Whiting, 2015). A study on 53 commercial banks in 17 emerging Middle-Eastern North African (MENA) countries during 2000-2013 revealed that the impact of the gender diversity and board size had negative impact on the performance of the relevant organisations (Anwar, 2014). The study by Walt and Ingley (2003) on gender diversity on performance of the firm indicates that the possible impact of the gender diversity on firm performance might depend on the social perception of the gender in the region and the individual capability of the directors. The study conducted on the companies listed in Bursa Malaysia between the year 2008 and 2009 in both main and ACE markets indicate a positive correlation between the two factors based on the various advantages that are offered in the board decision-making capacity based on gender diversity (Julizaerma and Sori, 2012).
18.104.22.168. Gender diversity in corporate governance regulations in the UK
There is on-going research to understand the exact impact of the board gender diversity on firm performance but the issue has gained momentum in the developed countries regarding UK. The public pressure can be represented in the UK corporate governance code (UKCGC) (Post and Byron, 2015). According the 2015 statistics of the Fortune 1000 companies, the female representation in the boards of the listed companies were less than 20% that is indicative grave inequality in the field (Tricker and Tricker, 2015). The corporate governance code, on the other hand, suggested at least 25% representation of female members in the boards of the FTSE 350 companies that indicates the extent of the inequality (Terjesen, Aguilera and Lorenz, 2015). However, this needs to be stated that the movement is based more on representation, which is not dependent on the financial benefits to be gained by the move. Based on the above discussion, the following hypothesis will be tested:
H1: The gender diversity on the board has definite positive associate with the financial performance in UK listed firms.
2.3.2. Relations between board independence and financial performance
22.214.171.124. Theoretical understanding of the relations of the financial performance with the board of independent directors
The board of directors has always been considered crucial for the firms’ wellbeing and there has been a plethora of literature on the topic. The board independence has been related with the composition of the board and the independence of the directors with the effective decision-making ability of the board ion many occasions. From the theoretical background for the phenomenon, the research adopts multiple theories as basis of the understanding of the board effectiveness and it relates to the nature of the board and its independence. The agency theory is supportive of the notion that the higher independence of the board effectively reduces problems in the agency. The main duties of the independent directors are based on their evaluation of the CEO and other executives’ performance in conjunction with the chairperson. Therefore, the composition of the board of independent directions is essential for a proper evaluation of the firm performance (Smith and Garnier, 2008). A very high number of independent director, on the other hand, has proven to be detrimental to the firm performance by restricting the decision-making capability of the managers and thus negatively affect the financial performance. The nonexecutives or independent directors can be considered additional resources of the firm, which is considered them for their experience, reputation and contacts. Thus, a balance has to be reached so that the independent directors have the maximum monitoring power over the firm’s decision while not effectively reducing the power of the management.
126.96.36.199. Empirical association between the independent directors and the financial performance of firms in developed nations
The relationship of firm performance and the number of independent directors on the board of the firm is unclear in the case of developed countries (Smith and Heshusius, 2006). However, some positive effect of the higher number of independent directors on the firm performance can be observed. A study conducted by Millstein and MacAvoy between 1991 to 1995 on 154 US firms show positive connection between the return of investment and the number of independent directors on board. The study by Weir et al. (2002) also considered 311 UK based firms where the number of independent directors has shown positive impact on attracting a higher number of investors. A recent study on 187 UK and US forms between 2000 to 2007 has shown significant positive relation of the higher number of independent directors with the rising of firm value (Muller-Kahle et al., 2014). On the other hand, the proportion of the independent directors to executives has shown negative impact on the financial performance of the firm in developed nations (Agrawal and Knoeber, 1996). This result can be disputed, as a contemporary study on 250 UK firms has revealed no such relation between the two factors (Vafeas and Theodorou, 1998).
188.8.131.52. Empirical association between the independent directors and the financial performance of firms in developing nations
The relationship between the number of independent directors and firm financial performance is unclear in the case of developing countries based on the opinion of Denis and Sarin (2009). There are several studies that provide contradictory information regarding the firm performance and the independent directors in a specific firm. While a study of 24 Tunisian listed firms between 2000 and 2005 show positive impact of independent directors in firm performance, a study on a sample of 318 Chinese listed firms between 2006 to 2008 show negative results on firm performance with higher independent directors (Hui, 2012). There are many examples that support this contradictory evidence form the various studies. In a study by Mangena and Tauringana (2007) on 118 Zimbabwe firms between 2002 and 2003, the findings showed positive interrelation of firm performance with number of independent directors, which were mostly based on attracting foreign investors. A study between 2008 and 2012 on 151 South African firms also shows positive relation of the factors of firm’s financial performance with higher number of nonexecutive independent directors (Muniandy et al., 2015). An extensive study by Sun and Ma (2014) on 7,405 firm-year observations also show a negative impact of the firm financial performance from the higher number of independent directors that were conducted between 2000 and 2005. Another study on 93 Nigerian firms between 1996 and 1999 did not reveal any such negative effect (Sanda et al., 2010).
184.108.40.206 Board Independence in Corporate governance regulations in the UK
The UKCGC states , “The board and its committees should consist of directors with the appropriate balance of skills, experience, independence and knowledge of the company to enable it to discharge its duties and responsibilities effectively”. (F. R. Council, 2012)
This is one of the main principle stated in section B1 regarding the board composition, which indicates the need for balance between the executive and non-executive directors on the board. However, these are only suggestions as the companies are given ample leeway in explaining their reason behind any irregularity in the appointment of the board members and the board composition based on the various diversity archetypes (Knyazeva, Knyazeva and Masulis, 2013). Thus, the board independence is given high importance in the code as the company might decide to deviate from the code if their circumstances dictate.
Based on the above discussion, this hypothesis will be tested:
H2: Board Independence has a positive associate with on the financial performance of a company.
2.3.3. Relation between board size and financial performance
220.127.116.11. Theoretical association between board size and financial performance
The theoretical support or the concept of the relation between firm performance and the board size is indecisive according to Upadhyay et al., (2014). While the board size has been related to a higher firm performance on some occasions (Yawson, 2006), the increased expenses associated with a larger board are also true. Jensen and Meckling, (1976) correlated this higher level of expense with reduced profitability from the increased agency costs. The agency theory contradicts these factors as it clearly supports a positive relationship between the firm performance and larger board size. Multiple studies based on the agency theory relates the higher number of board members with the ease and efficiency in the decision-making process, which makes this a positive factors for firm performance (Pearce and Zahra, 1992; Goodstein et al., 1994; Yawson, 2006; Dalton et al., 1998; Jiraporn et al., 2009). The previous point of the attracting more investors and increasing stakeholder confidence with higher number of independent board members is reinforced by Pfeffer (1973) and Ntim and Soobaroyen (2013).
18.104.22.168. Empirical association between board size and financial performance in developing nations
There are different findings of the studies conducted on the topic in the developing countries. This shows that there has been no support for any clear relations either positive or negative between the board size and firm performance in the developing countries according to empirical literature. A report based on a study on 452 US firms between the 1984 and 1991 reveal that the rise in the number of the board members cause a rise in expenses and the resultant impact on the firms profitability effectively reduces the value of the firm (Yermack, 1996). Similarly, another study on 879 Finnish listed firms from 1992 to 1994 reveal similar results where the board size negatively affects the ROA, which is directly related to the firms’ financial performance (Eisenberg et al., 1998). This similarly supported by the study conducted on 2,890 listed firms in the US between the years 1996 and 2004, where the board size has been directly related with lower Tobin’s q ratio and return of assets in the annual accounting (Cheng, 2008). Another large sample report by Guest, 2009 that considers the financial performance of 2,746 UK listed firms from 1981 to 2002, supports the previous correlation by showing a negative impact of the board size on the Tobin’s q ratio and share returns. Another study by Upadhyay et al. (2014) between 2000 and 2003 on 1500 US firms also found a clear negative relationship between the size of board and the ROA. The findings of Afrifa and Tauringana, (2015) also show the similar interrelations while studying 234 UK firm performances from 2004 to 2013.
A number of studies that reported a positive relationship between the board sizes and the factors that indicate the performance of a firm are also present in literature (Lehn et al., 2007). A study on 8,165 US-listed firms’ performance tracked between 1992 and 2001 by Coles et al., (2008) revealed a positive relationship between the board size and firm performance. A contemporary on the same period conducted on 1,618 listed firms in US by Wang (2012) also support the positive relationship between the board size and the growth of a firm. Denis and Sarin (2009) related this phenomenon with the larger boards decision-making capability which reduces for each member as the board size increases, effectively increasing the business risk-taking capability of the firm. Therefore, there is no doubt about larger board having better decision-making capabilities.
22.214.171.124. The empirical association between board size and firm financial performance in developing countries
Generally, there are compilation of the results of studies regarding the board composition and size and its affect firm performance on the developing countries put together from the previous research. According to Kamla and Roberts, (2010), unlike the developed countries, there is little proof of any relation between the board size and firm performance in developing nations. The sample report of 347 Malaysian firms listed from 1996 to 2000 showed a positive relations between the board size and the firm’s ROA which resulted in the increasing value of the firm (Haniffa and Hudaib, 2006). A similar study on the Indian listed firms shows similar interrelationship that correlated the larger board size with the higher financial performance of the firms (Jackling and Johl, 2009). A study in Pakistan, on 14 listed commercial banks from 2008 to 2012 also showed significant positive relations between larger boards to firm financial performance (Malik et al., 2014).
There are also studies that contradict the findings of the above studies as the instances of board sizes interfering with operations of the firm and resulting in underperformance is also common. A study by Mashayekhi and Bazaz, (2008) on 240 Iranian firms in 2005-2006 clearly indicated negative relations between the higher board size and performance. The study by Sanda et al., (2010) on 93 Nigerian firms between 1996 and 1999 also reveal negative impact of the larger board size on the growth. According to the researcher, the larger board opens up more possibilities for lower ROA, EPS, and ROE based on the findings from the study. Hui (2012) discovered a negative relation between the ROE and board size in the study. However, a similar study on 81 South African firms from 1999 to 2005 showed no such relation between board size and financial performance (Mangena and Chamisa, 2008).
126.96.36.199 Board Size in Corporate governance regulations in the UK
The specific regulations regarding the nominations committee is the main factor in the UKCGC that governs the appointment of the board members (Larcker and Tayan, 2015). The guidance offered, therefore, is based on the neutrality of the board and does not specify any size for the board. The statement, that the appointment of the board should be sufficient, does not specify any number but focuses on the composition of the board (Hsu and Wu, 2014). The relative ratio of executive and non-executive members and the fair and appointment processes to retain the objectivity has therefore clearly given more importance than the size (Liao, Luo and Tang, 2015). Thus, the board is free to elaborate on their composition based on their specific situation. However, take into consideration the amount of compensation for the directors on board, this would negatively impact the financial performance of the company. Also, based on the theory of stewardship theory, which was mentioned in the UK, which focuses on the autonomy of the executives and their moral motivation, the following hypothesis would be:
H3: Board Size has a negative associate with on the financial performance of a company.
Chapter Three: RESEARCH METHODOLOGY
3.1. Introduction to the chapter
Research methodology highlights the different effective processes, techniques and methods that are being adopted in order to carry out a successful research analysis. In this particular chapter, a detailed idea about the sample size and the research design used for this study purpose is discussed (Chalevas, 2011). The chapter depicts the suitable methods of collecting the appropriate data and information related to this study. Moreover, certain experimental models will be looked into, which will then explore the relationship between corporate governance and financial performance. Lastly, panel data methodology will be examined as the regression model to use when trying to establish the relationship between corporate governance and financial performance.
3.2. Selected research population and sample selection
In this research project, the population for this study will be selected from the UK firms listed under London Stock Exchange in an effort to investigate the relationship between corporate governance and financial performance. The study, in its choice of sample, will examine a number of criterion, key of which will be the availability of the data online. The first criterion should be examining the yearly reports of the selected listed firms for the period between 2013 and 2017 available on the official website of the London Stock Exchange (Bhagat et al, 2008). The second criterion would be examining the financial reports of the firms and the details of stock market for the same time period which can be available on websites such as Thomson Reuters (Munisi & Randøy, 2013).
In doing so, the sample selected for this study will be the FTSE 150 non-financial companies data for the period of 2013 to 2017. This is the targeted sample that would be easily available from Thomson Reuters and other online sources. When running the analysis, it will be important to only choose data sets that are complete and in the data analysis chapter, it will be mentioned how many observations were found to be complete. The choice of non-financial companies only in the analysis is because financial companies like banks are governed by different regulations which is quite different from the ones governing non-financial companies such as the capital requirements and the regulatory adherence they have to exhibit.
There were several reasons which described the importance of looking at these reports of the listed firms. Firstly, those satisfied the balanced analysis of the collected panel data (De Andres & Vallelado, 2008). Panel data helps in conducting both cross-sectional observations and time series as well as minimising the negative effects of the multicollinearity problems (Ntim et al., 2012). Also, the degrees of freedom get improved considerably by using the panel data for this study. Moreover, with the help of this data, the possible endogeneity problems can be reduced which may arise from the heterogeneous nature of the firms thus ensuring that the financial performance of the firms is improved by establishing cross-sectional with the corporate governance activities (Cremers & Nair, 2005). The results of this study could be effectively compared with the findings of the pre-established literature by using both cross-sectional observations and time series data.
3.3. Research design
The different variables of this study will be discussed in this section which investigated the correlation between corporate governance and financial performance. In other words, the way in which the changes in corporate governance activities influence the financial performance of the UK listed firms as explained herein. As a matter of fact, the variables that had been explored in this case are Board Size, Board Independence and Gender Diversity.
Financial performance of the firms = f [board size + board independence + gender diversity]
All these variables are independent; however the firms’ financial performance is a dependent variable.
This design will be attemting to answer three key hypothesis;
- The gender diversity on the board has definite positive associate with the financial performance in FTSE 150 UK listed firms.
- Board Independence has a positive association with the financial performance of a company.
- Board Size has a negative association with the financial performance of FTSE 150 companies.
3.4. Determining the corporate governance variables
It is crucial to judge a well-organised director’s board with respect to the financial performance of the firms (Gujarati & Porter, 2003). As mentioned in the previous section, three independent variables of corporate governance which have been considered for this research study were Board Size, Board Independence and Gender Diversity. The following table 3.4 depicts the detailed determination of these variables based on the previous literature studies:
Table 3.4: Details of used variables in the case of corporate governance
|Performance of Firm||
|Tobin’s q||TQ||For this calculation market value of total equity and book value of total debt is required which are related to book value of asset.|
|Diversification on the basis of gender||DIV||The ratio of female executives in the board of directors|
|Independence of Board||INDB||= the ratio between presence of independent non-executive directors in board and the total number of directors|
|Size of Board||BS||Number of directors in the board|
|Leverage||L||The ratio of total liabilities to total assets of a company|
3.5. Determining and evaluating the performance of the companies
This study has chosen one measure because of certain reasons. First, it must be denoted that there is not much agreement made on the suitable measure that should be chosen for determining the financial performance of a company (Mangena et.al, 2012). Thus, in this study, prime emphasis had been laid upon the Q-ratio which is commonly used in corporate governance studies (Renders et.al, 2010). Also this ratio helped in drawing steady comparison between the findings of previous research studies relating to the effect of corporate governance changes on firms’ financial performance (Haniffa & Hudaib, 2006; Tsamenyi et al, 2007).
The Tobin’s q value is used to determine the ratio of the market value of a company as well as the replacement costs of the firms (Ammann et.al, 2011). This q value is being calculated for determining the difference between the net value of the total assets of the company and total equity according to the books of accounts, plus the net market value divided by the total assets of the company (Albassam, 2014). With the help of the Tobin’s q value, the importance of corporate governance strategies for increasing the shareholders’ market returns in the company can be easily and effectively determined (Jackling & Johl, 2009).
However, the Tobin’s q has being criticized to a considerable extent by different researchers on certain occasions (Bruno & Claessens, 2010). For example, many criticised that it is not necessary that a firm would have high market value because of the effective management of assets. In many cases, it has been perceived that a firm’s market value is higher although the assets of the firm are not managed properly. These assumptions based on certain firms because the others to get overvalued (Guest, 2009). Tobin’s q ratio is also marked based on the biased evaluation of the assets of the companies. Nevertheless, sudden economic crisis might negatively affect the financial markets since the market value is related with the global economic conditions.
3.6. Panel data
Through this panel data methodology correlation coefficient between two important variables like performance of the organisation and corporate governance approaches taken by firm. Different research works are done on the basis of this panel data methodology. Software is used under this panel data technique for checking the regression value. Observation of time related to different entities relate to the panel data analysis (Bai et.al, 2004). Two types of dimensions are normally used for the data analysis purpose; one is time series and other is cross-sectional (Bai, 2009). Panel data methodology is appropriate over here as it basically focuses on different organisations and their performances over time (Raj and Baltagi, 2012). Through this approach, researcher could be able to analyse various data points and degree of freedom also could be observed significantly (Shan & McIver, 2011). This process also helps to make clear segregation among explanatory variables. Panel data methodology has always been effective and some of the advantages of panel data have been presented below in a detailed manner: (i) Individual differences or changes could be understood through panel data. Panel data always makes segregation among companies, any individual or any nation. Panel data always considers this as separate entity whereas other techniques like cross section or time series couldn’t able to make differentiation among all. Variables could be easily controlled by Panel data (Hsiao, 2014). (ii) As per analysis panel data is instructive. More efficient outcome could be obtained through panel data analysis as it generates more viable data. Variables are not co-related with each other over here. It gives different options and unlike time series and cross sectional, this has a different identification parameter for different variables (Tsamenyi et al, 2007). This is the reason; more intact and significant evaluation could be obtained through this process. (iii) The small firms mainly produce panel data which is gathered from the companies, individuals or family units. This is because more variables can be measured from small entities which help to avoid potential prejudices (Hsiao, 2014), (iv) The panel data helps in identifying and quantifying the exact influence over the research study which cross-sectional data and time series methods cannot provide (Frees, 2004), (v) It is used in developing complex behavioural models effectively by employing cross-sectional data or time series (Bai, 2009). However, there are certain limitations perceived in panel data which are not present in pure time series analysis (Wooldridge, 2010), (vi) Adjustment dynamics also could be understood properly through the panel data. Some problems are also occurred through cross sectional techniques which could be overcome through this process (Wooldridge, 2010).
When the panel data was examined econometrically, the observations were not presumed to be independently distributed over time (Wooldridge, 2009). Hence, it is important to employ special and effective models and methods so that the collected panel data can be analysed properly. The parameters of the model remain constant by applying simple ordinary least squares (OLS). Thus, it can be hereby said that the panel estimation method helps in eliminating the heterogeneity biasness by involving either of the fixed effects model or the random effects model (Baddeley & Barrowclough, 2009).
3.6.1 Fixed effects model
This could be stated as the regression procedure which has the ability to control non measurable variables. Fixed effects model helps to find out the time-invariable changes which are unobserved and on the basis of equations these are calculated (Allison and Waterman, 2002). Through this model correlation coefficient between both observed and unobserved variables could be understood. At the time of correlation, impact of unobserved variables cannot be controlled (Allison, 2009).
If correlation coefficient value will be zero then fixed effects model might not be effective. There are some drawbacks also present of this fixed effects model. In this case, constant time independent variables couldn’t be observed through the regression model (Baltagi et.al, 2005). As per analysis it is found that sometime errors are occurred high in fixed effects model compared to random effects. As a consequence, p value goes higher. Accounting observers have been facing trouble in this process due to its unsteady regression value and less control over the fixed effects. Sometime, this procedure cannot be able to measure the coefficient values of the variables (Allison, 2009). Average of each individual present in time invariable factor becomes zero and this doesn’t give any suitable outcome in the research procedure (Allison and Waterman, 2002). For this particular reason multiple arguments and contradictions present on this particular model.
3.6.2 Random Effects Model
As per research and findings, random effects model have various advantages; one of the prime advantages is time constant variables (Drukker, 2003). These variables are given permission in this model which is a drawback for fixed effects model (Drukker, 2003). Here, variables are set for a specific time period; still as per this model unobserved factors don’t have any kind of relation with explanatory factors. This process or approach is used when one individual is selected from a large group of people for getting any kind of ideologies or perceptions (Torres-Reyna, 2007). Random Model is also not giving any kind of help in handling unobserved factors and this approach believes there is no such type of bonding or correlation between observed factors and unobserved factors (Wooldridge, 2009). As per analysis and understanding it could be evaluated that random effects model has the ability to give more precise and concrete information compared to Fixed Effects Model. Moreover, this model is too easy to handle and understand. Prime goal of this research paper is to find out an approximate result and for this reason this model would be helpful and effective. For this particular reason here Hausman and Breusch-Pagan LM testing procedure are implied to understand the model.
3.6.3 Hausman Testing process
In every testing process some estimators are present and this process is applied when both fixed and random model has considerable differences. Analysis stated that estimator present in Random model is more efficient compared to Fixed Model but it needs to be checked whether this supports Hausman testing or not (Arellano, 2003). After testing if it is found this not supports the Hausman testing then Fixed Method Modeling is used instead of Random Modeling. Through the discussion some differences could be established between Random effects modeling and fixed effects model and these differences are based on some benefits and drawbacks (Hsiao, 2014). This difference could be found out through a statistical testing process. Through this comparison testing it could be understood, random effect is suitable or restrictive fixed effects can be applied for getting good outcome (Allison, 2009). One of the testing procedures is null hypothesis testing that comes under Hausman testing process. According to this testing outcome, approximations of both fixed and random effects model is same (Wooldridge, 2010). There are some differences present between these two on the basis of viewpoints. Fixed effects model has the viewpoints that every single unobserved variable has an approximate relationship with explanatory factors where random effects model doesn’t support this statement (Croissant and Millo, 2018). Random effects modeling don’t have this type of pre-assumption procedure. There are two hypothesis have been mentioned here which could be important in this type of research process,
H0: As per this null hypothesis, explanatory factors are related to unobserved variables chosen for the research process.
H1: This could be stated as alternative hypothesis which denotes explanatory factors don’t have any kind of relationship with unobserved variables.
To justify the null hypothesis (H0), Random effects model must be applied whereas for the justification of H1, fixed effects model would be good. After this Hausman Test needs to be done. If the p value would be below 0.05 then automatically H0 will be rejected and as per criteria H1 would be chosen. In this particular case fixed effects model is the ideal option for analysis. On the other hand, if p value goes higher from 0. 05 then random models will be applicable instead of fixed model. In this case H1 would be rejected and H0 will be accepted. On the basis of this context, random effects model would be good for analysis (Croissant and Millo, 2018).
Hausman test is highly required for managing performance and along with that it always maintain a good corporate governance practice. Hasuman test will never be applicable if Prob>chi2=0.2450 is found. For this reason H1 will be rejected instead of H0. For this particular study random effect model will be highly beneficial. Here, fixed effect model will no longer be effective and regression analysis will be required. There are two types of regression analyses present like pool regression and random regression model. Through Breusch-pagan test, it could be understood which regression model will be effective for the analysis purpose.
3.6.4 Breusch-pagan testing process
Pool regression is normally tested by Breusch-pagan testing procedure (Maddala and Lahiri, 1992). This testing procedure is used conditionally for checking the continuous correlation rose due to random effects model (Baltagi et al, 2005). If the value of Breusch-pagan is high then random effect model implies in OLS model. OLS model stands for ordinary least square (Baum et al., 2003). Through this testing procedure, two hypotheses could be raised;
H0: Pooled regression process would be effective.
H1: Random regression model would be effective.
In the case of significant p value H0 will be rejected and H1 will be accepted. In the case of Breusch-pagan test, Prob>chibar2=0.0000. In this case H0 is neglected and H1 will be automatically accepted. After having this overall discussion, it could be stated that Breusch-pagan test and Hausman test support H0 which indicates random effect model is best for the analysis of data in this particular paper.
Chapter Four: DATA ANALYSIS AND DISCUSSION
This chapter looks into the general analysis of the various statistical aspects of the variables and tries understanding the implication of the results obtained from the analysis. It looks into the summary statistics of the variables used and analyzes the results in line with some of the hypothetical assumptions highlighted in the literature review chapter. The analysis performed looks into the descriptive statistics thereafter looking into the correlation and regression analysis and the impact of these results examined.
4.2 Descriptive statistics
It is important to have an understanding of some of the general statistics such as the mean, standard deviation and the quartiles of the data before looking at the other results. This is because an understanding of these qualities will be able to give us the general view of how the data looks like before delving into the specifics of how the models yielded result and the significance of such results to the topic in hand. Muller-Kahle et al. (2014) looked at the various implications of having different combinations of high or low mean and standard deviation and some of the highlights that their study mentioned is that data with low standard deviation is more reliable as it implied less uncertainty with regards to how the data is collected and will be indicative of a stable market environment.
The number of observations that had complete data sets is 570 so this is what is used in the analysis and the years covered by the data started from 2013 to 2017 and looked into FTSE 150 companies and each year had 114 complete observations. The choice of these observations is based off the fact that financial companies in the UK operate under slightly different regulations as opposed to non-financial companies in regards to various aspects such as financial reporting or capital requirements hence the use of non-financial companies’ data. The table below summaries the key features of the data used.
Table 4.1: Table showing the summary statistics of the variables used for analysis
The mean of the Tobin’s Q came to about 1.2806 and it is key to note that the standard deviation is really high which is indicative of the fact that there the return of the FTSE 150 companies varied a lot as measured by this variable. The gender diversity had a mean of 0.2296 over the 5 years which is close to the 25% target that the FRC had earlier suggested but far from the 20202 target of a 33% representation in the board by 2020. There were also companies that had 0 representation of females on their board though it seems highly unlikely his to be the case in reality and could just be an error in terms of the data collection process. The independence stood at 0.5961 which is close enough to the two thirds representations that have been suggested over the years and in fairness, the last two years had a 79% representation of independent board members which suggests that companies have been adhering to the requirements. The average board size came to around 10 and the standard deviation is not that substantial and the implications of this number will be discussed further in later sections. The leverage of the companies had an average close to 60% which is to be expected and really does not say much other than the companies are not overleveraged and it is used as a control variable to streamline the companies.
4.3 Correlation analysis
This attempts to look at the correlation between the variables used in the analysis in an attempt to see if there are going to be problems of serial correlation between variables. This value usually ranges from -1 to 1 and is a representation of the direction and strength of the correlation between the variables. A value of 1 shows a very strong positive correlation between variables while a value of -1 shows a strong negative correlation between the variables. The sign of positive or negative shows the direction while the value shows the strength. The table that follows shows a summary of the Pearson correlation matrix;
Table 4.2: Table showing the correlation matrix of the variables used in study (* <0.1, ** <0.05, *** <0.01, **** <0.001)
|Variable||Gender diversity||Board Independence||Board Size||Leverage|
|Board Independence||0.3385 ****||1||
|Leverage||0.1214 ***||-0.0531||0.2032 ****||1|
This table is important since when the value of the correlation is above 0.8, the multicollinearity problem is usually encountered and as such the variables with such high values will have to be joined together in some way. Fortunately for our study, there is no correlation factor that is above 0.8 which was indicative of no multicollinearity issues in regard to the independent variables used in the study. The largest correlation factor is that between gender diversity and board independence which is also very significant and this is to be expected since the relationship can exist due to the fact that members of the board that were female were most likely also independent directors. The other large value is that between size and leverage but this relationship has no economic sensibility. There is a negative correlation between diversity and board size which could be due to the fact that there aren’t many female representations in large boards and the other negative relationship is between independence and leverage which seems to suggest that when the board is more independent, the company becomes less leveraged. This seems to be in line with literature since independent directors will normally improve the protection of investor interests and a reduction of leverage can be seen to be a good thing hence showing better protection for shareholders (Nguyen, 2016).
4.4 Analysis of the regression models
For the regression, a panel data methodology is used as earlier suggested and both the fixed and random effects models were run on the data. As earlier mentioned, there exist tests to help decide which of the two models to be picked and the study ran the Hausmann test to help in identifying which between the two models is going to be appropriate for the data available. The result of the Hausmann test is in favor of the Random Effects model and the details can be found in the Appendix and the equation of the regression by Random effects took the form of;
This is the model that guided the answering the research objectives and Random effects controlled for the unobserved heterogeneity of the companies when the value is constant and related to the independent variables. The table below summaries the coefficient values for the above regression model.
Table 4.3: Table showing the values of the regression coefficients (* <0.1, ** <0.05, *** <0.01, **** <0.001)
|Board Independence||1.7109||0.000 ****|
|Board Size||-0.1358||0.000 ****|
The table indicates a positive relationship between gender diversity and board independence while it shows a negative relationship between board size and leverage. Board independence and board size are the significant ones while leverage is highly insignificant and diversity is just slightly insignificant.
H1: The gender diversity on the board has definite positive association with the financial performance in FTSE 150 UK listed firms.
The results show that indeed the gender diversity does have a positive relationship with financial performance; however, the p value is 0.149 which is higher than the normally accepted 0.1% significance level. This therefore implies that under the 0.1% significance level, the relationship is not supported and thus the relationship is nonexistent at this significance level and cannot be included in the regression equation.
H2: Board Independence has a positive association with the financial performance of a company.
The results show that there does exist a positive relationship between board independence and the financial performance of the FTSE 150 companies and this relationship is significant at the 0.001 significance level. The relationship is also implied to be a strong one given the 1.7109 coefficient factor and this is in agreement with the hypothesis. This therefore suggests that the relationship is supported statistically and can be included in the regression equation to explain the financial performance.
H3: Board Size has a negative association with the financial performance of FTSE 150 companies.
The findings show that there does exist a negative relationship between the board size and the financial performance of the FTSE 150 companies and this relationship is significant at the 0.001 significance level. The relationship is not as strong with a coefficient of 0.1358 but it is still in agreement with the hypothesis. This therefore suggests that the relationship is supported statistically and can be included in the regression equation to explain the financial performance.
H4: Leverage has a negative effect on the financial performance of FTSE 150 companies
The findings show that there does exist a negative relationship between the leverage and the financial performance of the FTSE 150 companies and this relationship is not significant at the 0.1 significance level. This therefore implies that under the 0.1% significance level, the relationship is not supported and thus the relationship is nonexistent at this significance level and cannot be included in the regression equation.
4.6 Discussion of Findings
4.6.1 Gender Diversity
The statistical insignificance of the results suggests that we cannot include this variable in the regression equation and thus the effects of gender diversity on financial performance based on this study seem to be inconclusive. This result seems to be in agreement with the study done by Oxelheim et al. (2006) which found no significant effect of the inclusion of female directors to the board and the financial performance of those companies and Lam et al. (2013) also came to the same conclusion as this study when they looked at Chinese companies for a 5 year period ending 2011. The argument for the inconclusive results given by these studies is that the men and women are equally well educated and offer the same value to companies regardless of gender and the study by Oxelheim argued more towards looking at diversity in terms of the skill set of the directors rather than the gender.
Huang and Kisgen (2013) however, found a positive relationship between these two and the reason around it is that they postulated that men are more overconfident than women and thus the inclusion of female in the board serves to counter balance the over aggressive nature with which they may issue debt or make investment decisions thus improving the performance. The same conclusion was come to by Gonzalez (2014) who also argued from the point of reduced leverage by inclusion of women in the board hence improving performance. Conversely, there are studies that show a negative relationship and one such study is that by Cao et al. (2015) that found women in boards in Swedish companies tended to be more risk seeking compared to the men and this had a negative effect on the performance on the companies and this result is also corroborated by the study by Smith et al. (2005) which also came to the same conclusion.
Clearly there exist a lot of conflicting results in regard to the effect of females in the board and it may need further research once time has been allowed for the effect of these changes to reflect well on the data as this is still a new concept which is probably why there exist varying opinions and findings from research.
4.6.2 Board Independence
The study found a significant positive relationship that suggested a strong relationship with performance. The theory behind this is in agreement with the Agency theory since this ensures that the interests of the shareholders are better protected by there being structures to put checks on the management since the independent directors are very free from influence from any person from within the company. Having a more independent board can also serve as an advantage in securing capital from the global markets since people value companies that observe such practices as they see the companies as better suited to safeguard their interests.
Empirically, these findings are supported by a number of similar studies such as the ones by Chen (2011), Muller-Kahle et al. (2014) and Wu and Li (2015) all found a positive relationship between the two in their studies all looking at the proportion of independent non-executive directors on the financial performance measured through Tobin Q. Hambrick (2008) had results contrary to these, finding a negative relationship and so did Roudaki et al. (2015) when they investigated listed firms in New Zealand for the period between 1999 and 2011 and they used ROE and ROA as the measure for financial performance.
Most literature suggest a positive relationship between these two and empirically, the support is overwhelmingly on this view which is why the UK is in a push to ensure companies abide to this policy.
4.6.3 Board Size
The results of the regression indicate a slightly weak significant negative relationship between the performance and size and this is in line with some of the theories out there. There is an argument that smaller boards have less issues when it comes to making decisions which therefore means that the decision making process is very quick as compared to companies with larger boards and thus implementation of strategy would be quick giving the companies with smaller boards a higher chance to be successful. This theory is also supported by the study by Yermack (1996) and Wang (2012) which both looked at this effect in the US market and found results similar to our findings. Wang looked into US listed companies and his sample size was 8165 and his measure of performance was Tobin’s Q same as our study. Nguyen et al. (2014) argues that board size is one of the most important characteristic however his study did not find a significant relationship between the variable and the financial performance of the companies.
There of course have been findings from studies that argue to the contrary such as the study by Anderson et al. (2004) who argues that a larger board is more effective in limiting the powers of the CEO and also it may help in accessing a variety of skill set from the various members to better improve the workings of the company. The breadth of literature in regard to this variable is wide and varying and as such an argument can be made in favor of either camp. However the UK Board Index (2017) indicated that globally a number of 10 seemed more reasonable which is what the study had and still got a negative relationship with the financial performance.
The results of the study found a very insignificant relationship between the two and additionally, this variable is used as the control variable to adjust to the effects of varying leverage between companies. The theory around it is that an increase in the debt obligations of a company will lead to increased cash flows and there may raise agency issues with the cash flows if badly managed leading to an opportunity for declined performance. One argument in favor of the leverage could be that the banks that give credit will do so to a company that is adhering to good financial standards and as such if a company has a big debt; it implies it is more in line with regulations and can then have a good performance which makes the relationship positive. This is however disapproved by studies like Mangena et al. (2012) which found a negative association which then goes to making the case for the contrary argument. The opinions on this are mixed and no consensus exists in the breadth of literature and the results of our study suggest this.
4.7 Summary of Chapter
The chapter had a look at the descriptive statistics of the variables used in the study and then ran a correlation test that found no multicollinearity problems with the independent variables. Only 570 observations had complete data and this is what is used. The choice of these observations is based off the fact that financial companies in the UK operate under slightly different regulations as opposed to non-financial companies in regards to various aspects such as financial reporting or capital requirements. Therefore the study excluded all financial companies from analysis which reduced the observations from slightly over 1100 observations to 620 then reduced further to 570 due to incomplete and missing data. This then prompted the regression to be done with all variables and only board independence and board size were statistically significant. The results were then interpreted giving possible reasons as to the reasons behind it.