Firms’ Financial Performance and Corporate Board Diversity: Evidence from Kenya **

The basic motivation of owners of capital is to maximize their wealth by enhancing value of the firm. The objective of the agent on the other hand may be varied including enhancement of personal  income, tenure prestige, continuity in employment, and increased bargaining strength for future income maximization (Clarke, 2004; Monks & Minow, 2008). This divergence of interests between  the agent and the principal usually leads agents to engage in those activities which maximize their chosen interest of continued employment. This is more pronounced in situations where controls, ratifications and sanctioning of managerial decisions are not effective, either by the board or due to the fact that the shareholders are not empowered enough to effectively control management. To remedy such misalignments of interests, a number of governance mechanisms aimed at aligning the interests  of the agents and the principals are employed (Clarke, 2004; Monks & Minow, 2008). One mechanism is for the owners to choose a board of directors to intermediate the two contrasting interests and ensure that the interests of the shareholders are met through the management decisions. In certain cases some of the owners sit on these boards to directly take care of the shareholders’ interests. In other instants, a completely neutral board is appointed to perform the oversight (monitoring) role (Monks & Minow, 2008). The board therefore performs two key functions; monitoring (oversight) and advisory. These functions are closely related in that for one to advice, there must be some degree of monitoring. Conversely for one to monitor, there must be some degree of advising. The core aim of these functions is to safeguard the interest of the shareholder which is; to maximize their wealth (Falaye, Hoitash, & Hoitash, 2011; Jensen & Meckling, 1976; Monks & Minow, 2008). The oversight role requires independent boards. Board independence is inversely correlated with tenure. The longer the tenure, the less independent, as other interests of the board members begins overshadow their monitoring role. In order to prevent such an eventuality, boards are appointed on fixed renewable terms. The shareholders reserve the right to change the boards at the annual general meetings in cases of underperformance or at the expiry of their terms in office. On the other hand, advising is directly correlated with tenure. Advisory role requires boards that are familiar with the company, industry and well endowed with insight and experience (Masulis & Mobbs, 2011).

The main issue therefore, is how to make the boards more effective in their monitoring role so as to maximize the shareholders’ wealth. This role is at the heart of the Agency theory. The boards’ advisory role which argues that a board brings in useful experience and networks and saves costs through timely advice and exploitation of its vast network is at the heart of the Resource Dependence theory of corporate governance. These two theories are at the centre of the debate on the role of director’s heterogeneity on financial performance. On the one hand, there are schools of thought who argue that the monitoring function would ensure less wastage and a maximization of the shareholders’ interest. Another school of thought argues that what is crucial, is the advice and networks that the diverse board brings with it which is crucial to maximizing shareholders’ value. This is achieved through a cost reduction as timely decisions are made and wastages reduced due to experience and the subsequent efficiency gained in decision making ( Fracassi & Tate, 2012).

There are various cases of corporate failures which points at ineffective boards. In some of the cases, the management decisions raise the fundamental question on the integrity, competence and effectiveness of the corporate boards. Mumias sugar factory’s wores is an indictment on corporate governance in Kenya. In the company’s water bottling plant in which over 220 million Kenyan shillings was invested, the project made no profit and the project was neither its core business nor part of the company’s by-products. It was however approved by the board. In a study by Oyieke (2002), Kenya airways was a case of successful privatization, however less than two decades later; the airline is a loss maker in dire need of government bail-out of Kshs. 4.2 billion. The problems of the airline points wrong strategic decisions which were approved by the board. Haco Tiger Brand managing director with the approval of the board invested Kshs. 100 billion in the Mavoko housing project which has turned out to a bad investment decision. This decision had the board’s approval. At Telcom, the ex-staff wants the board investigated for failing to pay kshs 3.2 billion severance settlements. The largest share holder at British American Insurance Company (BRITAM) is accused of fraud, money laundering and embezzlement. At Uchumi supermarkets it is the second time the supermarket chain is seeking huge amounts of money for bail out (BD, 2015). All along the financial reports indicated a robust enterprise. These cases points squarely to corporate boards that have failed to execute their primary mandate to monitor and advise management. The corporate landscape in Kenya is replete with such costly, poorly thought through and badly executed white elephants and begs the question as to the competency, and commitment of these board to monitor and advice these organizations.

In terms of executive remunerations, the corporate landscape in Kenya defies the expectations especially for a developing country. The executive pay for the top ten listed companies at the NSE is  at par with their profit growth at 21%. These directors earn 1.4% of their companies’ net earnings (NSE, 2011, ROK,2014). The directors (CEO and CFO) of Safaricom collectively earn over 200 million a month (BD, 2015). In the embattled Kenya Airways, the Chief Finance and Chief executive officers split Kshs. 85 million a month (BD, 2015). Most of these top managers own substantial shares hence earning handsome dividends as well.

These incidences points to boards that are either out-rightly misplaced or whose interests are variance with those of the share holders. The results of this poor corporate governance are erosion of investor confidence, colossal loss of investor’s funds and a down grading of Kenya’s global competitive ranking from 106 to 144 (ROK, 2014).

The study therefore aims at investigating the relationship between the board experience and corporate financial performance in Kenya between 2001 and 2010. This broad aim is achieved through specifically analyzing the relationship between boards’ tenure, boards’ education and past experience, and boards’ entrenchment behavior (tenure^2) on performance variables.

A basic proposition is that the composition of the board affects the way the board performs its core functions of monitoring and advising. The boards’ heterogeneity affects performance by influencing management dynamics (Clarke, 2004; O​k​p​a​r​a​,​ ​2​0​1​1). A heterogeneous board is crucial as it brings to the organization rich human capital, diverse experience, diverse social capital, diverse abilities and views, different behavioral perspectives, diverse networks, and different socio-psychological orientations which are crucial in group dynamics. These contributions increases with diversity of past experience which brings to bear the rich background and exposure necessary for a competitive business environment and the cumulative experience in a particular board (tenure) which improves the quality of decisions and reduces the length of time it takes to make a decision (Fracassi & Tate, 2012; O​k​p​a​r​a​,​ ​2​0​1​1; P​i​t​e​l​i​s​,​ ​2​0​1​2; Rhode & Packel, 2010; Rose, 2007; Yangmin & Cannella, 2010).

There is however no single theoretical perspective on the nature of the relationship between the boards’ heterogeneity and financial performance. There is an amalgamation of economic, human relations, sociological, political, and organization theories which discusses some aspects of these relationships. The most relevant theories reviewed in this paper are the resource dependency and human capital theories.

This study integrates both the human capital theory and resource dependancy theory to come up with a relevant theoretical framework(Hillman, Cannella, & Paetzold, 2000). Theoretically, the argument for board diversity (heterogeneity) is mixed. There are a number of theoretical arguments in favour of a mixed board. (Carter, D’Souza, Simkins, & Simpson, 2010) argued that a more diverse board is able  to make decisions based on the evaluation of more alternatives compared to a homogeneous board which suffers from “group think”. This diversity may not necessarily be achieved primarily from compositional effect, but also from time effect. Board diversity therefore increases the firms’ capacity to link with global and domestic markets, expand access to global and domestic talent pools, and strengthen social capital. Those with political experience and networks can help firms deal with government bureaucracy hence saving on substantial costs (Goldman, et al., 2009).

The litrature on board heterogeneity does not conclusively handle heterogeneity in a growing democracy, ethnically fragmented society where board composition does not only consider experience and qualification, but also affirmative actions. The literature is also devoid of cases where interlocking boards include members who serve in both public and private company boards. In Kenya, most of the board chairs are previous powerful CEOs (60%), such boards may defy the theoretical and empirical position on monitoring. Where the board has a number of prevoius top level management even if tenure is short, monitoring will be optimized.

Experience is acquired through human capital (knowledge) accumulation. Knowledge and skills can be classified into two dimensions: functional (education and training) and organization specific (tenure). Effective boards require that directors have a set of functional skills or relationship skills with the external networks to obtain information and analyze business issues relevant for advising and also acquire sufficient knowledge of the firms operation to be able to oversight management (Forbes & Milliken,1999). Organization-specific skills and knowledge relates to the possession of detailed information and deeper understanding of operations and internal management issues. These skills are acquired cumulatively based on the tenure in a specific board (Sahghal, 2013; V​a​f​e​a​s​,​ ​2​0​0​3). Directors need to acquire both knowledge and skills to be able to make consistent decisions and contribute positively to the advising and monitoring roles.

Experience can be modeled as a function of the initial stock of education and past experience gained in similar positions or directorship in other firms, and time spent on a particular board (tenure)

Where EP_i is experience in firm I, k_i  is the stock of education and past managerial experience of the board members prior to joining the present board, including interlocking directorship, $\beta_1$ the effect of tenure of the board on experience, T tenure is the average number of years the members have been on the current board, T ² is the average change of boards behavior $\beta_2$ is the effect of change of boards behavior on experience.

The notion of board independence assumes that directors will monitor and advise management to the same degree over the directors’ tenure. However management friendliness hypothesis suggests that seasoned directors are more likely to befriend management and hence be less likely to play their monitoring role on management effectively as a result of entrenchment (Hwang & Kim,  2009;  V​a​f​e​a​s​,​ ​2​0​0​3).This occurs as relationships forms between management and boards overtime which tends to compromise the boards’ independence, in which case directors may qualify as independent, yet fail to exercise that independence due to the “friendliness” that have developed overtime with management (Huang, 2013; Hwang & Kim, 2009). When terms are limited, the tendency could be less monitoring towards the end of the term or alternatively more vigilant towards the end of term, hence monitor keenly if they expect an extension of their term. Whichever the direction, the behavior is determined by the individual directors post term expectations. In this respect the members’ age could be crucial. Those still below the ceiling age for qualifying as directors, but have finished the mandatory term in a given organization, may play it safe and not rock the boat as they expect future appointments. Those finishing their mandatory term and have reached the ceiling age may monitor more closely as they want to leave behind a legacy (Adams & Ferreira, 2007; Guner, Malmender, & Tate, 2008; Hwang & Kim, 2009).

Knowing the external environment and having varied industry and environmental experience is also critical in the performance of the strategic advisory role of the board. In this regard, a board needs to have a deep knowledge of the industry and have functional networks (Fracassi & Tate, 2012). Such knowledge is acquired through the diversity of prior experience before joining the board. Board tenure is a function of board composition (compositional effect) and time effect. Board tenure can change due to a change in board composition( Erhardt, Werbel &Shrader 2003). Board tenure could also change due to passage of time (time effect) (Huang, 2013). In the monitoring function, there is a substitution between the board as a monitor and the shareholders as a monitor. Where the shareholders are empowered, boards’ monitoring functions are performed by the shareholders. However, firms with weak shareholders’ right, benefits more from stronger monitoring by the board.

The role of powerful CEOs cannot be overlooked in all these as they may influence the composition  of the board. In this process of influence, their aim is to have a ‘cozy’ board. This is very prevalent in cases where the CEO is an owner, a major shareholder or in case of public corporation where the CEO is well connected politically (Adams, Almeida, Ferreira, 2005; Hambrick & Fukitomis 1991). In this case, the CEO would wish to shift the composition of the board to favor them by weakening the monitoring function by preferring long tenured board members. A board with members from professional associations such as bankers, accountants, lawyers, may not be free from the requirements of these external bodies. Hence the discharge of their monitoring and advising function is moderated by their professional bodies (Guner, et al., 2008).

In a study of Swedish firms, (Thorsell & Anders, 2012) reported no significant relationship between experience and performance. They expanded the definition of experience to include external ties (interlocking directorships), managerial experience, age and venture capital ownership. (McIntyre, et al., 2007) in their study on board composition and performance, reports that experience had a positive correlation with performance. (F​i​n​k​l​e​,​ ​1​9​9​8) reports that experience as measured through board size, directors’ reputation as scholars, and directors’ financial expertise, had significant positive effect on firms’ performance. In a study by (H​o​w​t​o​n​,​ ​2​0​0​6), it reported that longer tenure among board members (the average being 7.38 years), was significantly related to firms’ performance. According to (Kang, Chen, & Gray, 2007), managerial experience had significant positive association with firms’ performance. This was due to the industry experience and problem solving skills that they bring with them. Hence their advice is taken more seriously by management and this cuts down on wastage and costs hence improving financial performance. (Bodnaruk, Kendel, Massa, & Simonov, 2008) reported that more experienced directors are more independent, wiser, use resources economically and experiment less in their decisions. This leads to timelier, less costly decisions and better performance in organizations. Hence experience has a positive association with firms’ performance. Experience is positively and significantly associated with age. Hence the older, the more generally experienced the person is (V​a​f​e​a​s​,​ ​2​0​0​3). (Coles, Daniel, & Naveen, 2008) reported that firm specific knowledge positively affects the quality of advice that boards render to management. Hence tenure is positively associated with performance. (Hambrick & Fukitomis, 1991) showed that generally managers tend to engage in more incremental and routine learning after about 2.5 years with the firm. Therefore tenure is positively correlated with performance.

Holding board composition constant, (Huang, 2013) showed that board tenure exhibits an inverted U- shaped relationship with firm’s value. (Huang, 2013), further reports that the accumulation of firm- specific knowledge results in better acquisition decisions, better disclosure of value-relevant information in financial statements and a higher likelihood of engaging in innovation. However, this effect is only up to a threshold level of board tenure. As tenure continues to increase, (Huang, 2013) argues that boards become “Zombie Boards” due to a decline in their ability to oversight management as they become cozy with management. The results seems to suggest that for additional year of tenure, the benefits of the learning dominates for ‘younger’ boards while the costs of entrenchment dominates for ‘older’ boards (Huang, 2013).

Empirical evidence shows that firms with more complex operations and firms with more intangible assets have greater advising needs. Due to such needs, these firms have larger board size and greater insider representation on the boards (Coles, et al., 2008) as their need for firm specific knowledge is higher. The knowledge used for advising is also used for monitoring as it allows boards to identify weaknesses and consider the firms’ exposure to risk in the context of its operating environment for more complex firms, the maximum firm value is reached at an average tenure of 11.2 years, while for more R & D intensive firms, the maximum firm value is reached at an average tenure of 10.4 years. The empirically observed peak value in the relationship between Tobin’s Q-ratio and board tenure is around 8 years (Huang 2013). With all control variables held constant at their relative means, for average board tenure of three years, an additional year of tenure increases firm value by an average of 0.45%, while for an average board tenure of 15 years adding one year to a board tenure decreases firm value by an average of 0.52%. These results are consistent with the interpretation that the marginal value of learning exceeds the marginal cost of entrenchment when board tenure is shorter, but that entrenchment effect dominates the learning effect as board tenure continues to increase (Huang 2013).

(V​a​f​e​a​s​,​ ​2​0​0​3) found that an individual board member may influence the others in monitoring management as the less tenured members “respect experience”. If the longest tenured member exits, there will be a change in the mean tenure and the influence pattern will change. The decision making processes and even the quality of the decision will change. This can be considered as an ‘experience shock’ to the board. Hence board dynamics changes every time a board is subjected to a shock such as exit or death of a long tenured member (Huang, 2013). Recent studies that examined how prior social relationship between CEOs and directors affect firm performance and corporate decisions, found that network ties between directors and CEOs weakens the intensity of board monitoring (Fracassi & Tate, 2012), hence tenure is negatively associated with performance. (Bill, Hasan, & Wu, 2012) found no significant relationship between directors’ tenure and firms’ financial performance for the Finnish firms they studied.

In studying the association between diversity variables and financial performance, endogeneity could pose a problem. Endogeneity will lead to biased and inconsistent parameter estimates that make reliable inferences virtually impossible. Assuming

For i=1,..,N, and t=2,…,T, with $|\alpha|$< 1. The disturbance $\eta_i$ and $\mu_{i t}$ have the standard properties, that is E($\eta_i$)=0 , E($\mu_{i t}$)=0, therefore the E($\eta_i$, $\mu_{i t}$)=0 for i=1,…,N, and t=2,…, T. Additionally, the time varying errors are assumed uncorrelated;

E($\mu_{I S}, \mu_{i t}$)=0 for i=1,…,N, and $\forall$ t$\neq$s. No additional conditions are imposed on the variance of $\mu_{i t}$, hence moment conditions used below do not require homoscedasticity.

x_it is assumed to follow an autoregressive process.:

For i=1,…,N and t=2,…T, with $|\rho|$=1. such that E($\varepsilon_{i t}$)=0 and E($\eta_i, \varepsilon_{i t}$)=0 for i=1,…N, and t=2,…T.

Two sources of endogeneity exist in the x_it process: i) the fixed-effect component $\eta_i$, has an effect on x_it through a parameter $\tau$ - 1 implying that y_it and x_it have both steady state determined only by $\eta_i$. ii) the time-varying disturbance $\mu_{i t}$ impacts x_it with a parameter $\theta$. By using lagged values of the dependent variables as instruments, the endogeneity problem is eliminated. This is because the lagged value of the dependent variable is uncorrelated with the error term but correlated with the explanatory variables in the model (Soto 2010, Wintoki, Linck, & Netter, 2012).

Therefore the study specifies a dynamic panel data model of Arellano and Bond (1991) of the form:

Where $\eta_i$ and $\lambda_t$ are individual and time specific effects respectively, xit is a vector of explanatory variables, $\beta(L)$ is a vector of associated polynomials in the lag operator and q is the maximum lag length in the model. The number of time periods available on the i th individual Ti , is small and the number of individual N is large.

Identification of the model requires restrictions on the serial correlation properties of the error termuit and/or on the properties of the explanatory variables xit . It is assumed that if the error term was originally autoregressive, the model has been transformed so that the coefficients $\alpha_i$’s and $\beta_i$''s satisfy some set of common factor restrictions. Thus only serially uncorrelated or moving average errors are explicitly allowed. The $v_{i t}$ are assumed to be independently distributed across individuals with zero mean. The xit may or may not be correlated with the individual effects $\alpha_t$, and for each of these cases they may be strictly exogenous, predetermined or endogenous variables with respect to $v_{i t}$.

In this dynamic model we apply first difference transformation as it eliminates $\eta_i$ from the transformed error term without at the same time introducing all lagged values of the disturbance $v_{i t}$ into the transformed error term. This transformation allows the use of suitably lagged endogenous variable as instruments (Arrelano & Bond 1991; Stock & Yogo, 2005). If we have a balance panel, p=1, and there are no explanatory variable nor time effects, the $v_{i t}$ are serially uncorrelated, and the initial conditions y; 1 are uncorrelated with $v_{i t}$ for t=2,…, T., then using first difference we have:

Eguations                    Instruments available

$\begin{array}{ll}\Delta y_{i 3}=\alpha \Delta y_{i 2}+\Delta v_{i 3} & y_{i 1} \\ \Delta y_{i 4}=\alpha \Delta y_{i 3}+\Delta v_{i 4} & y_{i 1}, y_{i 2} \\ \Delta y_{i T}=\alpha \Delta y_{i T}+\Delta v_{i T} & y_{i 1}, y_{i 2}, \ldots y_{i, T-2},\end{array}$

If the model is mean stationary, then the first differences \Delta y_{i t} will be uncorrelated with \eta_i, hence $\Delta y_{i, t-1}$ can be used as instrument in the level equations (Stock & Yogo, 2005). Hence in addition to the instruments available for the first differenced equations, we then have:

Equation                       Instruments available

$\begin{array}{ll}y_{i 3}=\alpha y_{i 2}+\eta_i+v_{i 3} & \Delta y_{i 2} \\ y_{i 4}=\alpha y_{i 3}+\eta_i+v_{i 4} & \Delta y_{i 3} \\ y_{i, T-1}=\alpha y_{i 3}+\eta_i+v_{i T} & \Delta y_{i, T-1}\end{array}$

Performance= Tobins’ Q ratio, ROA, P/B and Share price, diversity= age, experience, board size, Previous perform=lagged values of Tobin’s Q ratio, P/B, Share price and ROA. Firm size=natural log of the total assets of the company, firm=unique time invariant unobserved firm level characteristics based on firm level fixed effects in the regression estimates, time =time period of that observation. The coefficient of interest is $\beta_i$, where H0 : $\beta_i$ =0 and Ha : $\beta_i$≠0.

The study will estimate the above equations for the overall sample of 29 firms. These 29 firms’ financial performance over the period 2001-2010 in terms of the accounting variable and stock variable will be analyzed.

The overall boards are relatively large in size with a mean size of 10 members. The boards’ average age is 64 years indicating relatively older boards. The average board tenure is 8.2 years implying that most members serve more than one fixed term of 4 years. The stock of experience has an average of 4.91 years an indication that those appointed to the boards have substantial past experience. There is considerable interlocking of board members. The performance variables are normally distributed. The ROA has a mean of 0.1164 and a standard deviation of 0.1236. There is low use of assets to generate wealth in the sample firms. The Q-ratio ranges between 0.0403 and 10.79 and a mean value of 3.678 implying overvalued stocks. The P/B value has a mean of 1.3895, a minimum of 0.03 and a maximum of 9.3. The higher the ratio, the higher the premium the market is willing to pay for the company.

The study sought to find out the relationship between tenure of the board and the performance variables. This assumes that there is correlation between the study variables. The table below shows the strength and direction of such relationship.

The table above shows that there is positive correlation between ROA, tenure, interlocking directorship and stock of experience. There is however a negative correlation between tenure ^2, average age of the board, board size, and firm size. The correlation between the Q-ratio and board tenure, stocks of experience, average age, interlocking directorship and firm size is positive.Tenure^2 and board size are negatively correlated with Q-ratio. Share price is positively correlated with board tenure, stocks of experience, average age, interlocking directorship and firm size. It is however negatively correlated with Tenure ^2 and board size. Price to book value (P/B) is positively correlated with board tenure, stock of experience, interlocking directorship and average age of the board. It is however negatively correlated with tenure ^2, board size and firm size.

The negative correlation between tenure ^2 and the performance variables can be explained as resulting from the entrenchment behavior of the boards. As the boards get entrenched, they tend to play less of the oversight role as they begin to get friendly with management. This friendly relationship with management compromises the quality of monitoring and even the nature of the advice they tender to management. They tend to play it safe with their monitoring and advice as their personal interests gain prominence as opposed to the interest of the share holders. This negative board behavior tends to grow with tenure, hence the parabolic relationship between experience and performance as observed also in other empirical studies reviewed in this paper. The relationship between the board size and the performance variables is negative. This implies that the larger the board, the poorer is the financial performance. This could be attributed to the conflicting interests of the board, the heavy financial burden such a board exerts on the organization both direct and indirect; there is also the problem of group think and majority rule, which could be manipulated by the dominant interests. Tenure is positively correlated with the performance variables. This is explained by the fact that the more knowledgeable about the firm, the better the advice the board renders. When we integrate the effect of tenure and tenure^2, we get the inverted U relationship between experience and performance. This implies that there is an optimal point at which the negative entrenchment behavior outweighs the positive advising behavior. Stock of experience has a positive correlation with the performance variables. It implies that the previous experience and the networks built over time and the education qualification improves both the advising and monitoring roles of the board.

The table below shows the result of the system GMM estimation of the relationship between the performance variables, and board experience. The table also reports the control variables and how they are related to the performance.

Tenure has a positive effect on all performance variables. This conforms to both the empirical and theoretical literature reviewed in this study (Hwang & Kim, 2009; Sahghai, 2013) and the theoretical literature as reported in (Anderson, et al., 2012; Forbes & Milliken, 1999; Hillman, et al., 2000; Schultz, Tan, & Walsh, 2010). Tenure improves the monitoring role which in turn has a positive influence on performance. Tenure^2 has a negative effect on all performance  variables.  This conforms to the hypothesis that the entrenchment effect of the board negatively affects their monitoring function (Goldman, et al., 2009; Huang, 2013; Hwang & Kim, 2009; V​a​f​e​a​s​,​ ​2​0​0​3). This means that the entrenchment effect dominates the learning effect at longer tenure (15 years) Ref appendix A-1. This is due to the cozy relationship that develops between management and the entrenched board whose other interests begins to dominate the monitoring interest. As the boards become more entrenched, they tend to overlook some of the management excesses as they seek to extract favors from management. This concurs with the empirical results of (Huang, 2013). They actually become zombie boards who adopt any proposal from management and fail to critically monitor management. The entrenchment behavior eventually compromises the monitoring role. This negative effect of an entrenched board can be strong enough to negate the positive effects of both the tenure and stock of experience. This gives rise to an inverted U shaped relationship between experience and performance with an optimal experience performance level of 7 years for Tobin’s Q- ratio.

The experienced board is less receptive to new ideas and keen to extract more individual benefits. This entrenchment cost outweighs the benefit of advice resulting from experience. The simulation in this study with different tenure of boards showed that when the simulation was run with tenure below the average (8.2 years), there was a positive association, however when the simulation was run with tenure above the average age of board tenure, there was a negative association. This concurs with the findings of (Fracassi & Tate, 2012; Huang, 2013). The simulation results for up to 15 years (three terms) concur with the empirical studies of an inverted U-relationship between experience and performance and tenure and performance as measured by Q-ratio (ref appendix A-1).

The control variables have the correct sign. Even though a larger board brings in a large pool of past experience (stock of experience) according to the resource dependency theory, the boards’ behavior of a larger boards, such as the competing interest that leads to intense lobbying and opposing positions outweighs the past experience hence board size has significant negative association with performance variables except in the case price to book value. This negative association between board size and performance variables could be attributed to the resulting agency problem (Jensen & Meckling, 1976). The size of the firm, significantly affects performance, this concurs with the theoretical and empirical literature reviewed.

The study concludes that there is a significant positive relationship between tenure and performance, and experience and performance. The positive relationship between tenure and performance implies that monitoring function improves with tenure. The boards’ entrenchment behavior as proxied by Tenure^2 is significantly negative. This implies that as the board gets entrenched; the boards’ monitoring role becomes more accommodating, hence failing to check management.

The study also concludes that the relationship between tenure and firms’ performance shows an inverted U- relationship. As tenure increases, the entrenchment behavior captured by tenure ^2 outweighs the positive monitoring role. Hence there is an optimal tenure level at between 7 and 8  years depending on the sector.

Finally, advising is negatively correlated with tenure, while monitoring is positively correlated with tenure. However, with interlocking directorship and owners sitting on the boards, monitoring and advising may not be a trade off.