African Development Review, Vol. 31, No. 4, 2019, 409–422
Do Shareholders Fight for Firm Value or Market Value in Emerging
Markets? The Mediating Effect of Board Structure Dynamics
on Dividend Policy Decision
Daniel Ofori‐Sasu*, Joshua Yindenaba Abor and Sydney Quaye
Abstract: The paper examines the mediating effect of board structure dynamics on the relationship between dividend
payout and shareholders’ wealth at the firm level and market level. Panel regression models are used by applying data from 27
listed firms in Ghana between 2010 and 2017. We find that board structure dynamics have a direct effect on shareholders’
wealth at both levels. However, the results reveal new evidence that board structure dynamics play a mediating role on the
relationship between dividend policy decision and shareholders’ wealth only at the market level. We find that independent
directors and CEO duality significantly reduce market value of shareholders through dividend payout decision. However,
independent directors and longer CEO tenure in office mediate a positive effect on the relationship between dividend policy
and shareholders’ wealth at the market level. Thus, the mediating effect of board structure dynamics, particularly independent
directors and CEO tenure, are important in predicting a positive relationship between dividend policy and shareholders’ wealth
at the market level than at the firm level. Therefore, regulatory bodies and investors should provide strong board structure
dynamics that serve as a mediating mechanism for prudent dividend policy decisions that add value to shareholders’ wealth.
1. Introduction
Dividend policy decision has been an important predictive power used to forecast the wealth of shareholders (Ofori‐Sasu et al.,
2017). Board structure, which involves the collective role of board members and board characteristics, influence dividend
payout decision and also builds investors’ confidence (Fuzi et al., 2016). Moreover, the various compositions of board
structure are key determinants of dividend policy (Alias et al., 2016). Thus, dividend policy decision and characteristics of
board structure are critical for enhancing shareholders’ wealth. Studies show that directors on the board take prudent decisions
on dividend payout in order to influence shareholder wealth. However, in countries where there is weak investor protection,
the key questions are: (1) do dividend policy decisions have a direct effect on shareholders’ wealth or does dividend policy
interact with board structure dynamics to impact shareholders’ wealth?; (2) does the mediating effect of board structure
dynamics influence the relationship between dividend policy and shareholders’ wealth at the firm level or at the market level?
In emerging markets like Ghana, the recent folding up of some firms, including some listed financial institutions, has
clearly shown the negative consequences associated with weak internal board structure mechanisms. Thus, the
opportunistic behaviour from managers, as well as agency interactions at various levels of the firm affects both dividend
payout decision and shareholder wealth either at the firm level or market level. Moreover, successions of unexpected
corporate failures have raised explosive concern among researchers (Krechovska and Prochazkova, 2014) regarding
corporate board decisions about dividend payout. Evidence from developing literature shows conflicting results between
various measures of board structure characteristics and dividend policies in Africa (Kyereboah‐Coleman, 2007) and other
*Daniel Ofori‐Sasu (corresponding author), Department of Finance, University of Ghana Business School, P.O. Box LG 78, Legon‐Accra, Ghana;
e‐mail: doforisasu@yahoo.com, dofori-sasu@st.ug.edu.gh. Joshua Yindenaba Abor, Department of Finance, University of Ghana Business School,
P.O. Box LG 78, Legon‐Accra, Ghana. Sydney Quaye, Department of Renewable Natural Resources, Kwame Nkrumah University of Science and
Technology, Ghana.
© 2019 The Authors. African Development Review © 2019 African Development Bank. Published by Blackwell Publishing Ltd,
9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA. 409
410 D. Ofori‐Sasu et al.
emerging economies (Hussein and Venkatram, 2013). Thus, the mixed results are associated with weak corporate
governance, contextual differences, institutional framework as well as differences in the measurement of variables in
emerging countries. Moreover, the mediating effect of board structure dynamics on the relationship between dividend
policy and shareholder wealth at the firm level or market level has not been empirically analysed and discussed from the
emerging economies perspective. This paper focuses on the Ghanaian context to investigate the interactive effect of board
structure dynamics and dividend policy on shareholders’ wealth.
Dividend policy is one of the strategic financial decisions that lie in the hands of the board of directors. The decision
whether to pay out a dividend to shareholders or retain company profit for future gains cannot be ignored among corporations
since dividend policy has a positive effect on shareholder wealth (Ofori‐Sasu et al., 2017). In addition, the board of directors
plays a significant role in influencing dividend policies in order to improve shareholders’ wealth. An effective board structure
mechanism that ensures a prudent dividend policy decision is evidenced in the attainment of shareholder value (Ranti, 2013).
This paper empirically examines two related hypotheses. First, board structure dynamics (board independence and CEO
duality) decrease the market value of shareholders but no relationship exists between these dynamics of the board structure and
firm value of shareholders. Second, dividend policy increases shareholder wealth at the market level when interacted with
board structure dynamics, but does not improve shareholder wealth at the firm level. Although, a previous study has tested the
relationship between dividend policy and shareholder value at the firm level (Ofori‐Sasu et al., 2017), there is a dearth of
studies that has examined the effect of dividend policy on shareholder wealth at the market level as well as the interaction
effect of board structure dynamics on the influence of dividend policy decision on shareholder wealth. The paper is conducted
in the context of emerging markets, particularly in Ghana, where the dividend policy framework and board structure dynamics
are relatively undeveloped and the firms are the main drivers of economic activities. From an economic policy perspective, it is
important to understand whether board structure dynamics mediate the relationship between dividend policy decision and
shareholder wealth at the firm level or market level. In this study, dividend policy and board structure dynamics, as well as
introducing an interaction term between these dynamics and dividend policy decision were used to predict the outcome of
shareholders’ wealth generated from the firm or in the market.
The rest of the paper is organized into four sections. A literature review of related studies is contained in Section 2, and Section 3
discusses the data and methodology. The empirical results are contained in Section 4 and Section 5 concludes the study.
2. Literature Review
2.1 Theoretical Review
Shareholders are interested in maximizing their wealth either at the firm level or market level. In corporate finance, there
are two main concepts underlying shareholders’ wealth maximization, that is, whether corporate board structure and
dividend policy decision are relevant or irrelevant to shareholders’ value maximization. These concepts can be explained
by theories in corporate finance (Abor, 2007). In corporate governance literature, the behaviour of the board of directors
and the structure of the board are explained by agency theory; stakeholder theory (Hill and Jones, 1992); stewardship
theory (Donaldson and Davis, 1991); and resource dependency theory (Pfeffer and Salancik, 1978). Agency theory occurs
when agents prioritize their own benefits at the expense of shareholders, which eventually affects value maximization of
shareholders. An agency problem can be caused by information asymmetry where there is less information disclosure
between shareholders and managers. For agency theory, shareholders’ value is expected to fall when there is a conflict of
interest between managers and owners of the firm. Donaldson and Davis (1991) argued contrary to the agency theory and
propounded the stewardship theory that there is no agency problem since the CEO has the ultimate responsibility to
perform well. Shareholders have the interest of maximizing their wealth. Thus, stakeholder theory advocates that the
interest of the firm lies on stakeholders and that the firm must be managed in the interest of these stakeholders. The
resource dependency theory explains how the external resources of firms affect the behaviour of the firm. Companies are
able to achieve their objectives based on the various resources they depend on. The non‐executive directors of the board
are able to use resources, particularly information that is not internally available to the firm and at the same time favours
shareholders’ interest.
On the other hand, the quest to maximize shareholders’ wealth is influenced by existing theories that explain the relevance
of dividend policy. Jensen and Meckling (1976) indicated that agency theory enables managers to pay high dividends to
© 2019 The Authors. African Development Review © 2019 African Development Bank
Mediating Effect of Board Structure Dynamics 411
shareholders and also reduce the internal cash flows subject to both management discretion and board decision. Another theory
underlying dividend policy is the ‘pecking order theory’, where firms prefer internal sources of funds at a cheaper cost over
debts, and would prefer debt over external equity because of the higher cost associated with equity funds. The residual theory
reflects dividend policy where firms would prefer ploughing back profits (retained earnings or residual earnings) after
identifying growth opportunities from positive net present value projects, rather than paying out dividends. In relation to bird‐
in‐hand theory, dividend policy reflects share price appreciation where insiders are forced to allocate some profits or all profits
for dividend payment decision purposes (i.e., retaining earnings for capital gains or paying a dividend to shareholders).
Dividend payout is less risky to capital gains from price appreciation because a dividend is paid immediately it is declared,
while capital gains reflect in the future (Bhattacharya, 1979). Tax‐preference effect theory indicates the relevance of taxes to
dividend policy decisions. Thus, favourable tax treatment of capital gains over dividend yield forces investors to prefer a low
dividend payout to a high dividend payout. This implies that firms should keep low dividend payouts to avoid attracting a high
tax payout. In terms of dividend irrelevant theory, Miller and Modigliani (1961) argued that investors are indifferent between
capital gains and the dividend payout decision, especially in a perfect capital market. The signalling theory states that
‘dividend policy decisions signal or communicate information about firm's future to potential investors’ (Lintner, 1956) and
therefore, an increase in dividend payments reflects high cash flows in the future.
2.2 Board Structure Dynamics and Dividend Policy Decision
As defined by Mayer (1997), ‘ways of aligning the interest of investors and managers and ensuring that firms
operate efficiently for the benefit of investors’ is referred to as corporate governance. However, the emergence of
corporate governance was motivated by the financial crisis that hit the Asian Stock Markets, partly because of weak
corporate governance among the firms in South East Asia. This triggered the development of more robust corporate
governance measures that restored the confidence of shareholders. The composition of board structure depends on size of
board members, CEO duality, CEO tenure in office, and board independence. These characteristics of board structure
are what this paper uses as the dynamics of board structure. Outside directors act independently to protect the interest
of shareholders. Thus, a high proportion of outside directors may positively affect shareholders’ value (Nicholson and
Kiel, 2007). Dalton et al. (2003) claim that larger board size increases board diversity and knowledge, and eventually
improves shareholder wealth. Gul and Leung (2004) posit that CEO duality, a situation where the CEO doubles as the
board chairperson and executive director, results in ineffective board supervision.
The payment of dividend is considered as a source of cash flows to shareholders. Dividend policy helps managers as well as
board members to decide how much each shareholder should receive in relation to dividend payments (Ross et al., 2008).
When a firm generates profits, part of the profit could be retained and the other part could also be paid to shareholders as
dividend (Abdulkadir et al., 2016). The decision is made by the board of trustees of the company, in accordance with dividend
policy. They can also decide to pay all as dividend or plough back all profits into the firm. The price of a share is attributed to
the amount of dividend distributed to shareholders, either as cash or share dividend (Adelegan, 2003). Dividend policy
decisions may impact the value of the firm in two ways, either through retained earnings, which causes growth or appreciation
in the value of the firm, or through a dividend payment to induce investors to trade the stocks of dividend paying firms. Thus,
the value of firms that trade in the stock market are influenced by dividend policy.
3. Empirical Literature
Empirical studies have shown the effect of board structure characteristics on the dividend policy decision. Alias et al. (2016)
examined the relationship between board structure, free cash flow and dividend per share, using 361 non‐financial listed firms
in Malaysia over the 2002–2007 period. From their findings, dividend per share increases when board structure is
characterized by a large number of independent directors. Moreover, independent directors and free cash flow on dividend
were positively and significantly related. They found that an increase in board size decreases dividend per share due to agency
cost. Further, CEO duality and dividend per share were negatively and significantly associated. Studies also show that a
stronger board governance structure pays out higher dividends (Adjaoud and Ben‐Amar, 2010). In terms of the relationship
between board structure and shareholders’ wealth, some studies (e.g. Dzingai and Fakoya, 2017) highlight that board size and
© 2019 The Authors. African Development Review © 2019 African Development Bank
412 D. Ofori‐Sasu et al.
independent boards positively affect firm value; CEO duality has a negative impact on firm value; and CEO tenure in office
influences firm value. However, these studies measured shareholders’ value using accounting ratios at the firm level. Ofori‐
Sasu et al. (2017) examined the effect of the dividend policy decision on shareholders’ value of listed companies in Ghana
using market value per share held by shareholders, and they found that dividend per share improved shareholders’ value of
listed firms in Ghana.
It is obvious that the dividend payout decision is relevant and impacts shareholders’ wealth (Sarwar, 2013); board structure
characteristics also influence shareholders’ value (Dzingai and Fakoya, 2017) and the structure of corporate board influences
dividend policy decision — which translates into impacting shareholders’ wealth (Amidu, 2007). Aspects of board structure,
namely board duality, independence and size, have been found to promote corporate decisions in the interest of shareholders
(Brickley et al., 1997). However, it appears that the mediating effect of board structure on the relationship between dividend
payout decision and shareholders’ wealth maximization is scarce in empirical literature from the perspective of emerging
economies. That is, if board structure dynamics have a role to play by influencing dividend policy decision, does this role
affect the firm value or market value of shareholders?
The study looks at how different dynamics of board structure interact with the dividend policy decision to influence
shareholders’ wealth and whether the impact is concentrated on shareholders’ wealth at the firm level or market level.
4. Data and Methodology
The study examines the mediating effect of corporate board structures on the relationship between dividend policy decision
and shareholders’ wealth. The study used a panel data of 27 firms listed on the Ghana Stock Exchange market for an 8‐year
period (2010–2017). This is because the dataset for the selected listed companies were available for the specified period in the
study. Data for the macroeconomic variables were obtained from Bank of Ghana Data scope for the specified years. The study
employed the random effect estimation model because firms included in the sample were randomly selected, the entity's error
term is uncorrelated with predictor variables, and the listed firms existed before the start of the period under study. Moreover,
the study used the Hausman test to confirm whether the random effect model was appropriate (Table 1).
4.1 Empirical Estimation
The model specifications employed in the study are shown below.
1. Mediating effect of board structure on the relationship between dividend policy decision and shareholders’ wealth at the
market level:
MVAit = β0 + β1DIVPOLICYit + β2BINDit + β3BODSIZEit + β4CEODit + β5CEOTit
+ β6 (BIND∗DPS)it + β7 (BODSIZE∗DPS)it + β8 (CEOD∗DPS)it + β9 (CEOT∗DPS)it + β10FIRMSIZEit
+ β11OVHDEXPit +β12FIRMAGEit + β13FIRMRISKit + β14FATit + β15TAXt + β GDPgrowth + β17INTrate16 t t
+ β18INFratet + μit (1)
2. Mediating effect of board structure on the relationship between dividend policy decision and shareholders’ wealth at the
firm level:
ROAit = α0 + α1DIVPOLICYit + α2BINDit + α3BODSIZEit + α4CEODit + α5CEOTit
+ α6 (BIND∗DPS)it + α7 (BODSIZE∗DPS)it + α8 (CEOD∗DPS)it + α9 (CEOT∗DPS)it + α10FIRMSIZEit
+ α11OVHDEXPit +α12FIRMAGEit + α13FIRMRISKit + α14FATit + α15TAXt + α16GDPgrowtht + α17INTratet
+ α18INFratet + εit (2)
© 2019 The Authors. African Development Review © 2019 African Development Bank
Mediating Effect of Board Structure Dynamics 413
Table 1: Summary of variables
Variable Measurement Expectations
Dependent variables
Shareholders’ wealth Market level: market value (Marketvalueofequity)it – (Bookvalueofequity)it Expected to increase
added (MVAit) as firm value
increases
Firm level: Firm value (ROA ): Net Profitit it Positive as net profits
Total Assetit increase
Independent variables
Dividend policy Dividend per share DIV POLICY = Dividend Paidit ≥ 0 (>0 if firm paid Positive/negative
decision (DIV POLICY ), which is a Number of sharesitit
dummy variable dividend; 0 if firm did not pay dividend in a
particular year)
Board structure Board independence (BINDit) Dummy variable: 1, when the number of non‐ Positive/negative
dynamics executives members are more than the number of
executive members, and 0 otherwise
Board size (BODSIZEit) Total number of board members Positive/negative
CEO duality (CEODit) Dummy variable: 1, if CEO is also the chairperson Positive/negative
/vice chairperson of the board, otherwise 0
CEO tenure (CEOTit) Number of years a CEO has been in office Positive/negative
Mediating variable Interaction terms between board Board independence multiplied by dividend per Positive/negative
structure dynamics and share (BIND⁎DPS)it
dividend policy Board size multiplied by dividend per Positive/negative
share (BODSIZE⁎DPS)it
CEO duality multiplied by dividend per Positive/negative
share (CEOD⁎DPS)it
CEO tenure multiplied by dividend per Positive/negative
share (CEOT ⁎DPS)it
Control variables
Firm specific Firm size (FIRM SIZEit) Natural log of total assets Positive/negative
variables Overhead Overhead Expensesit Negative/positive
costs ((OVHD EXP)_it) Total Assetsit
Firm age (FIRM AGEit) Number of years since its incorporation Positive/negative
Firm risk (FIRM RISKit) σROA Positive
Tangibility (FATit) Fixed Assetsit Positive/negative
Total Assetsit
Corporate tax (TAXit) Corporate tax per year Negative
Macroeconomic GDP growth (GDPgrowth ) GDPt − GDPt−1t Positive
variables GDPt−1
Interest rate (INTratet) Interest rate per year Positive
Inflation rate (INFratet) Inflation rate per year Positive/negative
5. Empirical Results and Discussion
The study presents descriptive statistics of the variables. Summary statistics and Pearson's correlation are used to screen and
test the reliability of the dataset. These are presented in order to ensure consistency, efficiency, reliability and robustness of
findings.
Table 2 shows the summary statistics of the variables employed in the shareholders’ wealth maximization models estimated
between 2010 and 2017 of listed firms in Ghana. The summary statistics is employed to screen for outliers which have the
potential to affect the efficiency, consistency and biasness of the estimated coefficients. However, the summary statistics do
not show any evidence of outliers. Table 3 reports the Pearson Correlation Coefficient matrix to check for possible
multicollinearity between the explanatory variables. For multicollinearity to occur, the correlation coefficient between two
© 2019 The Authors. African Development Review © 2019 African Development Bank
414 D. Ofori‐Sasu et al.
Table 2: Summary statistics
Variable Mean Std. dev. Min. Max. VIF
MVA 17.8616 1.9858 12.1115 22.6282 –
ROA 0.08006 0.1647 −0.7066 0.6581 –
DIV POLICY 0.0894 0.3409 0 1 1.29
BIND 0.7216 0.4495 0 1 1.40
BODSIZE 8.0455 2.2034 3 13 1.65
CEOD 0.0625 0.2428 0 1 1.41
CEOT 2.9432 1.7129 1 7 1.76
OVHD EXP 0.1731 0.3239 −0.1728 3.5774 1.34
FIRM SIZE 18.1215 2.3011 12.5436 22.2546 2.54
FIRM RISK 0.1485 0.3784 0.0063 2.5396 1.47
FIRM AGE 36.0284 21.7999 5.00 104 1.31
FAT 0.6819 3.3125 0.00005 41.4867 1.07
TAX 0.1619 0.4426 −3.3263 1.7113 1.19
GDP Growth 5.4426 3.0219 1.68 11.25 2.06
INTrate 14.3155 4.2888 8.9 23 1.93
INFrate 12.3060 3.8614 8.73 19.25 1.53
Notes: Significant level: p‐value < 1% (***): VIF mean= 1.62.
variables should be 0.7 or more (Kennedy, 2008). As shown in Table 2, we observe no presence of the multicollinearity
problem among the variables as confirmed by a mean variance inflation factor (VIF) of 1.62 and each of the variables has a
VIF below 10 (see Table 3). The study proceeds to analyse the interactions between board structure, dividend policy decision
and shareholders’ wealth.
5.1 Regression Results and Discussion
The regression results depict how the interaction between board structure dynamics and the decision to pay dividend impact
shareholders’ wealth. We employ the Hausman test to choose whether the fixed effect model or the random effect model is
appropriate. From Table 3, a chi2 statistic from the Hausman test is not significant; indicating that the random effect estimation
is preferred to the fixed effect. The robust standard error was used to correct for heteroscedasticity and autocorrelation while
year trend dummies were employed to control for endogeneity and omitted variable biases. To begin, the study examines the
relationship between board structure dynamics and shareholders’ wealth at the firm level and at the market level (see Table 4).
From Table 4, the negative and significant relationship between board independence and shareholders’ wealth suggests that
increasing the number of non‐executive directors on the board will increase board diversity, which leads to a situation that
makes it difficult for independent directors to have cohesiveness in decision making due to large size (board independence),
thus decreasing shareholders’ wealth at the market level and firm level. Board size was positively and significantly related to
shareholders’ wealth at the market level and firm level which supports the work by Hussein and Venkatram (2013). CEO
duality was negatively and significantly associated to shareholders’ wealth at market level but had no significant effect on
shareholder wealth at the firm level; which implies that the presence of CEO duality decreases the wealth of shareholders at the
market level but does not influence the wealth of shareholders at the firm level, which supports the work by Kyereboah‐
Coleman (2007). CEO tenure in office has a negative and significant relationship with shareholders’ wealth in the market but it
has a positive relationship with shareholders’ wealth at the firm level. The result indicates that CEOs who stay longer in their
position, focus more on firm value creation (Abor, 2007) rather than creating value at the market level; thus, decreasing the
shareholders’ wealth.
From Table 4, we show that board structure dynamics that are statistically significant to shareholders’ wealth at the market
level include board independence, board size, CEO duality and CEO tenure whereas board independence, board size and CEO
tenure are statistically significant to shareholders’ wealth at the firm level. We observe that board structure dynamics have an
inconsistent relationship with shareholders’ wealth at both levels.
© 2019 The Authors. African Development Review © 2019 African Development Bank
Mediating Effect of Board Structure Dynamics 415
© 2019 The Authors. African Development Review © 2019 African Development Bank
Table 3: Pearson's correlation matrix
DIV OVHD GDP
POLICY BIND BODSIZE CEOD CEOT FIRM‐SIZE EXP FIRM AGE FIRM RISK Growth INTrate INFrate TAX FAT
DIV POLICY 1.0000
BIND −0.2554 1.0000
BODSIZE 0.1357 0.2782 1.0000
CEOD −0.0669 −0.2586 −0.1335 1.0000
CEOT −0.1827 −0.0058 −0.0826 0.0773 1.0000
FIRM SIZE 0.1534 0.3035 0.5253 −0.3878 0.1257 1.0000
OVHD EXP −0.0480 −0.0937 0.0400 0.3176 −0.0518 −0.3818 1.0000
FIRM AGE −0.1732 −0.0330 −0.0807 0.2156 0.0836 −0.1493 0.0211 1.0000
FIRM RISK −0.0570 −0.2366 −0.2469 0.0997 0.0378 −0.4181 0.1527 −0.1940 1.0000
GDP Growth 0.0535 0.0631 −0.0416 −0.0482 0.3581 0.0908 −0.1116 0.0480 0.0419 1.0000
INTrate −0.0699 0.1144 0.0526 −0.0318 0.4425 0.1606 −0.0830 0.0492 −0.0321 0.0635 1.0000
INFrate −0.0217 −0.0102 0.0589 0.0099 −0.1538 −0.0334 0.1300 −0.0269 −0.0960 −0.4994 −0.2491 1.0000
TAX 0.0542 0.0523 −0.0052 0.1028 −0.0847 −0.0391 0.0739 −0.2818 0.2170 −0.0788 −0.0228 −0.0005 1.0000
FAT −0.0186 0.0552 0.1032 −0.0216 −0.0677 0.1060 −0.0350 0.0658 −0.0057 0.0686 −0.0564 −0.0503 −0.0170 1.0000
416 D. Ofori‐Sasu et al.
Table 4: Effect of board structure dynamics on shareholders’ wealth
Market value (MVA) Firm value (ROA)
Fixed effect (1) Random effect (2) Fixed effect (3) Random effect (4)
BIND −1.4685*** (0.5014) −1.4561*** (0.5168) −0.0781*** (0.0264) −0.0797*** (0.0280)
BODSIZE 1.4481*** (0.5467) 1.2586** (0.5471) 0.0267 (0.0251) 0.0094** (0.0049)
CEOD −3.7500*** (0.7108) −3.7143*** (0.8593) 0.2021 (0.2088) 0.1762 (0.2082)
CEOT −0.3736*** (0.1456) −0.4031** (0.1727) 0.0019 (0.0045) 0.0131*** (0.0033)
FIRM SIZE 0.4471*** (0.0748) 0.4661*** (0.0751) 0.0007 (0.0112) 0.0032 (0.0126)
OVHD EXP 0.3597** (0.1580) 0.3219** (0.1636) 0.0597 (0.048) 0.0553 (0.0503)
FIRM AGE 0.0067 (0.0101) 0.0069 (0.0093) 0.0017** (0.0008) 0.0018** (0.0008)
FIRM RISK 5.7693** (2.926) 6.1648** (3.1267) −0.2496 (0.2536) −0.1432 (0.2755)
GDP growth 0.0306 (0.0546) −0.0073*** (0.0027)
INTrate 0.0211 (0.0476) −0.0061** (0.0030)
INFrate 0.0741** (0.0349) −0.0022 (0.0033)
_cons 5.4628 (2.5159) 4.6436 (2.5942) −0.0715 (0.2797) 0.0428 (0.1903)
R‐square 0.3494 0.3627 0.2071 0.2396
Hausman test
Chi2 (9) 4.75 4.32 1.53 5.52
Prob > chibar2 0.8559 0.6649 0.9092 0.7864
Notes: Number in parentheses are t‐statistics. Significant levels: *** 1%, ** 5% and * 10%.
Further, we analyse the mediating effect of board structure dynamics on the relationship between dividend policy decision
and shareholders’ wealth maximization at the firm level and at the market level by using the random effect model. We run five
different models for the first and second equations above to check for robustness and consistencies in the signs (positive and
negative) attached to the coefficient of the regressors.
Board structure dynamics directly affect shareholders’ wealth (see Table 4), however, the random effect estimation results
in Table 5 reveal that the dynamics of board structure are not statistically strong in predicting shareholders’ wealth when
dividend policy decision and the interaction terms were introduced in the models. From Table 5, we observe that board size
and CEO tenure in office had no significant relationship with shareholders’ wealth. Board independence was negative and
significantly linked to market value in model 6, but the relationship was inconsistent across the estimation models. The results
show that board structure dynamics had no significant relationship with shareholders’ wealth at the firm level. However, CEO
duality was negative and significantly associated with shareholders’ wealth at the market level and the relationship was
consistent.
The results show that shareholders’ wealth is maximized at the market level when the dividend policy decision is taken. We
observe that the dividend policy decision positively and significantly affects shareholders’ wealth at the market level.
However, the decision to pay a dividend has no significant effect on shareholders’ wealth at the firm level. Moreover, the
results provide evidence to support that the dividend policy decision predicts shareholders’ wealth at the market level, rather
than firm level. The first lag of dividend policy was positive and significantly associated with shareholders’ wealth. This
suggests that the previous year's decision to pay out a dividend to shareholders improves shareholders’ wealth at both the
market and firm levels. This supports the residual theory where insiders of a firm prefer to retain their profits for capital gains.
The results give the indication that in the current year, firms in Ghana prefer to retain their profits for capital appreciation in
order to increase shareholders’ wealth at the market level. Thus, in the current year, the dividend policy decision will not
significantly influence shareholders’ wealth at the firm level since insiders of the firm prefer ploughing back rather than paying
out a dividend.
Introducing dividend policy and the interaction term (BIND * DIV POLICY) improves the marginal effect and the
significance level associated with board independence. Thus, the mediating effect of board independence (i.e., interaction term
between board independent and dividend policy) on the dividend policy decision causes a negative and significant relationship
between board independence and shareholders’ wealth at the market level. The implication is that non‐executive directors will
prefer paying out dividend rather than ploughing back earnings for future gains — leading to a decrease in shareholders’
wealth at the market level. However, the interaction term between dividend policy and board independence has a positive and
© 2019 The Authors. African Development Review © 2019 African Development Bank
Mediating Effect of Board Structure Dynamics 417
© 2019 The Authors. African Development Review © 2019 African Development Bank
Table 5: Random effect estimation: Mediating effect of board structure dynamics in predicting the relationship between dividend
policy decision and shareholders’ wealth
Shareholders’ wealth (market level) Shareholders’ wealth (firm level)
Variables Model 5 Model 6 Model 7 Model 8 Model 9 Model 10 Model 11 Model 12 Model 13 Model 14
DIV POLICY 1.790*** 1.554*** 2.863** 1.793*** 0.448 0.00864 −0.0174 0.127 0.0206 −0.0256
(0.426) (0.346) (1.125) (0.425) (0.503) (0.0202) (0.0186) (0.0920) (0.0183) (0.0389)
DIV POLICYt‐1 0.970*** 0.932*** 0.638 0.975*** 1.218***
(0.322) (0.339) (0.743) (0.326) (0.0989)
BIND −0.451 −0.712** −0.440 −0.443 −0.409 −0.0437 −0.0724 −0.0425 −0.0177 −0.0426
(0.307) (0.291) (0.318) (0.299) (0.337) (0.0349) (0.0445) (0.0349) (0.0160) (0.0361)
BODSIZE −0.00576 −0.00452 −0.00495 −0.00579 −0.00466 0.000365 0.000502** 0.000455 0.000261 0.000393
(0.00376) (0.00302) (0.00401) (0.00377) (0.00386) (0.000321) (0.000231) (0.000302) (0.000344) (0.000341)
CEOD −1.967*** −1.932*** −1.937*** −1.934*** −1.965*** 0.00637 0.0103 0.00968 0.126 0.00642
(0.391) (0.444) (0.399) (0.373) (0.410) (0.145) (0.148) (0.145) (0.115) (0.146)
CEOT −0.204 −0.167 −0.191 −0.204 −0.203 0.00816 0.0122 0.00966 0.00937 0.00819
(0.131) (0.105) (0.128) (0.131) (0.127) (0.0146) (0.0127) (0.0143) (0.0135) (0.0146)
BIND*DIV POLICY 6.266*** 0.690***
(0.956) (0.204)
BODSIZE* DIV POLICY −0.00977 −0.00108
(0.00639) (0.000943)
CEOD* DIV POLICY −18.26 −66.43***
(22.94) (9.672)
CEOT* DIV POLICY 1.202*** 0.0307
(0.435) (0.0286)
FIRMSIZE 0.293*** 0.277*** 0.289*** 0.292*** 0.266*** 0.00687 0.00519 0.00650 0.00441 0.00618
(0.0447) (0.0534) (0.0454) (0.0456) (0.0553) (0.00580) (0.00781) (0.00592) (0.00605) (0.00642)
OVHD EXP 0.825 0.751 0.800 0.829 0.748 0.109** 0.101* 0.106** 0.124** 0.107**
(0.558) (0.457) (0.556) (0.567) (0.537) (0.0502) (0.0541) (0.0506) (0.0505) (0.0509)
FIRMAGE 0.0121*** 0.0116*** 0.0120*** 0.0120*** 0.0128*** 0.000829*** 0.000766** 0.000818*** 0.000535 0.000847***
(0.00228) (0.00318) (0.00225) (0.00226) (0.00261) (0.000261) (0.000347) (0.000269) (0.000433) (0.000254)
FIRMRISK 0.403** 0.357** 0.411** 0.403** 0.371* 0.0386 0.0335 0.0395 0.0374 0.0378
(0.172) (0.160) (0.183) (0.174) (0.218) (0.0406) (0.0394) (0.0400) (0.0468) (0.0410)
FIRMAGE2 4.515*** 4.292*** 4.497*** 4.520*** 4.460*** 0.0329 0.00832 0.0309 0.0487 0.0315
(0.834) (0.767) (0.821) (0.838) (0.810) (0.0734) (0.0680) (0.0727) (0.0751) (0.0725)
FAT −0.0156 −0.0230*** −0.0159 −0.0154 −0.0156 −0.00230** −0.00313* −0.00234** −0.00182* −0.00231**
(0.0144) (0.00892) (0.0148) (0.0145) (0.0148) (0.000998) (0.00174) (0.000956) (0.00109) (0.000993)
TAX 0.177 0.139 0.173 0.176 0.172 −0.0315** −0.0357** −0.0320** −0.0350*** −0.0316**
Continued
418 D. Ofori‐Sasu et al.
© 2019 The Authors. African Development Review © 2019 African Development Bank
Table 5 (Continued)
Shareholders’ wealth (market level) Shareholders’ wealth (firm level)
Variables Model 5 Model 6 Model 7 Model 8 Model 9 Model 10 Model 11 Model 12 Model 13 Model 14
(0.240) (0.222) (0.241) (0.241) (0.241) (0.0152) (0.0178) (0.0151) (0.0119) (0.0154)
GDP Growth −0.0459 −0.0803 −0.0445 −0.0466 −0.0760 0.00895*** 0.00516** 0.00911*** 0.00642*** 0.00818***
(0.0945) (0.101) (0.0982) (0.0943) (0.0920) (0.00219) (0.00254) (0.00260) (0.00152) (0.00257)
INTrate −0.00312 −0.0146 −0.000556 −0.00369 −0.0214 0.00363* 0.00237 0.00392* 0.00156* 0.00317
(0.0751) (0.0826) (0.0769) (0.0750) (0.0755) (0.00191) (0.00230) (0.00230) (0.000928) (0.00223)
INFrate 0.0294 0.0110 0.0310 0.0291 0.0209 0.00549*** 0.00347* 0.00568*** 0.00451*** 0.00527***
(0.0602) (0.0652) (0.0619) (0.0600) (0.0604) (0.00176) (0.00194) (0.00200) (0.000995) (0.00190)
ytrend −0.242 −0.207 −0.258 −0.241 −0.212 −0.0150* −0.0112 −0.0168* −0.0114 −0.0143
(0.196) (0.224) (0.207) (0.196) (0.194) (0.00858) (0.00711) (0.00867) (0.0103) (0.00871)
Constant 10.61*** 11.40*** 10.57*** 10.63*** 11.38*** −0.248*** −0.161 −0.253** −0.181 −0.228**
(1.438) (1.701) (1.521) (1.426) (1.535) (0.0916) (0.129) (0.0999) (0.119) (0.107)
Observations 210 210 210 210 210 216 216 216 216 216
Notes: Numbers in parentheses are t‐statistics. Significant levels: *** 1%, ** 5% and * 10%. Models 5 to 9 show the estimations for dividend policy and board structure dynamics and its
effect on shareholders’ wealth at the market level. Models 10 to 14 show the estimations for dividend policy and board structure dynamics and its effect on shareholders’ wealth at the firm
level.
Mediating Effect of Board Structure Dynamics 419
significant effect with shareholders’ wealth at the market level and firm level. This gives the indication that non‐executive
directors (board independence) encourage dividend payout decisions in order to increase shareholders’ wealth at the market
level and firm level.
Introducing an interaction term between CEO tenure and dividend policy (CEOT * DIV POLICY) does not cause a
significant effect between board structure dynamics and shareholders’ wealth. However, the interaction term (CEOT * DIV
POLICY) was positive and significantly linked to shareholders’ wealth at the market level, indicating that the mediating effect
of CEO tenure on dividend policy leads to an increase in shareholders’ wealth at the market level. This shows that CEO tenure
has no direct effect on shareholders’ wealth but has a partial mediating effect on the relationship between dividend policy and
shareholders’ wealth at the market level.
On the other hand, although the interaction term between CEO duality and dividend policy had no significant impact
on the relationship between board structure dynamics and shareholders’ wealth at the firm level, there was a negative
and significant effect between the interaction term and shareholders’ wealth at the firm level. We observe that CEO
duality has no direct effect on shareholders’ wealth at the firm level but it has a mediating effect on dividend policy to
influence shareholders’ wealth at the firm level. This suggests that CEOs become chief executive directors and also chair
the board to take prudent dividend policy decisions that favour shareholders’ wealth at the firm level. Further, overhead
cost was positively and significantly (5 per cent) associated with shareholders’ wealth at the firm level. This indicates
that spending more on inside directors provides a form of incentives that motivate them to work efficiently to improve
firm value.
We observe that GDP growth, interest rate and inflation rate had a positive and significant association with shareholders’
wealth at the firm level. This suggests that board structure dynamics and dividend policy decisions maximize shareholders’
wealth at the firm level effectively in a macroeconomic environment. Thus, external environmental factors are important for
predicting shareholders’ wealth at the firm level.
5.2 Discussion Summary
The paper finds out whether shareholders maximize their wealth at the firm level or at the market level by employing the
random effect model to analyse the mediating effect of board structure dynamics on the relationship between dividend policy
decision and shareholders’ wealth. First, we find that board structure dynamics have a direct effect on shareholders’ wealth
maximization at both the market level and firm level. We find a stronger direct effect of independent directors, board size, and
CEO tenure on shareholders’ wealth maximization at the market level and firm level. However, CEO duality has a direct effect
on shareholders’ wealth only at the market level. Moreover, it is noticeable that board size increases shareholders’ wealth at
both levels whereas board independence decreases shareholder wealth at both levels. CEO duality and CEO tenure decrease
shareholders’ wealth at only the market level. The relationship between CEO tenure and shareholders’ wealth was not
consistent at both levels of wealth maximization.
Shareholders fight for wealth maximization at the firm level and market level but they do so through the dividend policy
decision and board structure dynamics. Thus, shareholder wealth maximization is a function of board structure dynamics and
the dividend policy decision. In our second analysis, we introduce dividend policy and interaction terms between board
structure dynamics and dividend policy to predict shareholders’ wealth. The rationale is to investigate the mediating effect of
board structure dynamics on dividend policy decisions and whether the decision to pay out a dividend translates into
shareholder wealth maximization at the market level or firm level.
Introducing dividend policy and interaction terms neutralizes the relationship between board structure dynamics and
shareholders’ wealth at the firm level. Thus, board structure dynamics do not have a direct impact on shareholders’ wealth
at the firm level when there is an interaction effect of dividend policy. We observe that the dividend policy decision
positively and significantly affects shareholders’ wealth at the market level. We find that the mediating role of board
independence causes a negative and significant relationship between board independence and shareholders’ wealth at the
market level. However, the interaction term between board independence and dividend policy increases shareholders’
wealth at the market level and firm level. The implication is that outside directors make a prudent dividend payout
decision in favour of shareholders’ wealth maximization at both levels. Moreover, the interaction term, that is, the
mediating effect of CEO tenure on dividend policy, leads to an increase in shareholders’ wealth at the market level. This
implies that CEOs stay longer in office to influence dividend policy decisions in order to increase shareholders’ wealth in
© 2019 The Authors. African Development Review © 2019 African Development Bank
420 D. Ofori‐Sasu et al.
the market. We observe that CEO duality has no direct effect on shareholders’ wealth at the firm level but it plays a
mediating role on the relationship between dividend policy decision and shareholders’ wealth by reducing shareholders’
wealth at the firm level (see Table 5, model 13).
6. Conclusion and Policy Implication
In conclusion, the study used the random effect model to provide evidence that board structure dynamics alone play an
effective role in influencing shareholders’ wealth at the market level and firm level. Our findings reveal that independent
directors and CEO duality have a direct negative effect on shareholders’ wealth at the market level. However, the relationship
between board structure dynamics and shareholders’ wealth maximization is inconsistent but the dynamics play a strong,
effective and mediating role on the relationship between dividend policy decision and shareholders’ wealth only at the market
level. Thus, board structure dynamics is more effective for predicting shareholders’ wealth at the market level in the presence
of the dividend policy decision. Specifically, we find that independent directors indirectly reduce the market value of
shareholders only but they provide a better dividend policy decision (through dividend payout) in order to increase
shareholders’ wealth at both levels. This is because the outside directors of the firms prefer earnings retention to dividend
payout in seeking shareholder value maximization at the market level but prefer dividend payout in order to maximize
shareholders’ wealth at both levels.
CEO duality reduces market value directly without a dividend policy framework, but it mediates through dividend policy to
increase firm value. Also, CEO tenure indirectly influences shareholder wealth by mediating through dividend policy to
increase market value. We can deduce from the findings that independent directors and CEO tenure mediate a positive effect
on the relationship between dividend policy and shareholders’ wealth only at the market level. This suggests that independent
directors as well as the CEO's longer stay in office influence the dividend policy decision in such a way that shareholders’
wealth will be increased at the market level.
Thus, shareholders fight for market value maximization since the mediating effect of board structure dynamics, particularly,
independent directors and CEO tenure, are important in predicting the relationship between dividend policy and shareholders’
wealth at the market level than at the firm level. Therefore, outside directors should put in place strong board mechanisms and
a dividend policy framework that increases shareholders’ wealth. A CEO, who doubles as board chairman and managing
director of the firm should focus on policies that maximize shareholder wealth, both at the firm level and market level. CEOs
should be encouraged (by incentive plan) to maintain an increasing wealth of shareholders, as long as they stay in office.
Finally, regulatory bodies and investors should provide strong and effective board structure dynamics that serve as a mediating
mechanism for prudent dividend policy decisions that add value to shareholders’ wealth. The study suggests that further
research should extend datasets to include more emerging economies in Africa with longer time periods and test the robustness
of the results. Future studies can also look at various measures of board structure characteristics and dividend policies and how
these variables interact with shareholders’ wealth at different levels from the perspective of firms across different groups in
sub‐Saharan Africa and the world at large.
References
Abdulkadir, R. I., N. A. H. Abdullah and W. C. Wong (2016), ‘Dividend Payment Behaviour and its Determinants: The
Nigerian Evidence’, African Development Review, Vol. 28, No. 1, pp. 53–63.
Abor, J. (2007), ‘Corporate Governance and Financing Decisions of Ghanaian Firms’, The International Journal of Business
in Society, Vol. 7, No. 1, pp. 83–92.
Adelegan, O. (2003), ‘The Impact of Growth Prospect, Leverage and Firm Size on Dividend Behaviour of Corporate Firms in
Nigeria’, African Development Review, Vol. 15, No. 1, pp. 35–41.
Adjaoud, F. and W. Ben‐Amar (2010), ‘Corporate Governance and Dividend Policy: Shareholders’ Protection or
Expropriation? Journal of Business Finance and Accounting, Vol. 37, No. 5 & 6, pp. 648–67.
© 2019 The Authors. African Development Review © 2019 African Development Bank
Mediating Effect of Board Structure Dynamics 421
Alias, N., M. H. Yaacob, R. A. Rahim and F. M. Nor (2016), ‘Board Structure, Free Cash Flow and Dividend per Share in
Malaysia Listed Firms: An Empirical Study of Interaction Effect’, Malaysian Journal of Society and Space, Vol. 12, No. 2,
pp. 58–67.
Amidu, M. (2007), ‘How Does Dividend Policy Affect Performance of the Firm on Ghana Stock Exchange’, Investment
Management and Financial Innovations, Vol. 4, No. 2, pp. 103–12.
Bhattacharya, S. (1979), ‘Imperfect Information, Dividend Policy, and “the Bird in the Hand” Fallacy’, The Bell Journal of
Economics, Vol. 10, No. 1, pp. 259–70.
Brickley, J. A., J. L. Coles and G. Jarrell (1997), ‘Leadership Structure: Separating the CEO and Chairman of the Board’,
Journal of Corporate Finance, Vol. 3, No. 3, pp. 189–220.
Dalton, D., C. Daily, T. Certo and R. Roengpitya (2003), ‘Meta‐Analyses of Financial Performance and Equity: Fusion or
Confusion? Academy of Management Journal, Vol. 46, pp. 13–26.
Donaldson, L. and J. H. Davis (1991), ‘Stewardship Theory or Agency Theory: CEO Governance and Shareholder Returns’,
Australian Journal of Management, Vol. 16, No. 1, pp. 50–65.
Dzingai, I. and M. B. Fakoya (2017), ‘Effect of Corporate Governance Structure on the Financial Performance of
Johannesburg Stock Exchange (JSE)‐Listed Mining Firms’, Journal of Sustainability, Vol. 9, No. 867, pp. 1–15.
Fuzi, S. F. S., S. A. A. Halim and M. K. Julizaerma (2016), ‘Board Independence and Firm Performance’, Procedia
Economics and Finance, Vol. 37, pp. 460–65.
Gul, F. A. and S. Leung (2004), ‘Board Leadership, Outside Directors’ Expertise and Voluntary Corporate Disclosures’,
Journal of Accounting and public Policy, Vol. 23, pp. 351–79.
Hill, C. W. L. and T. M. Jones (1992), ‘Stakeholder‐Agency Theory’, Journal of Management Studies, Vol. 29, No. 2,
pp. 131–54.
Hussein, S. K. and R. Venkatram (2013), ‘Corporate Governance and Firm's Value: An Empirical Analysis of Agri‐input
Firms in India’, IRACST – International Journal of Commerce, Business and Management (IJCBM), Vol. 2, No. 6,
pp. 353–61.
Jensen, M. C. and W. H. Meckling (1976), ‘Theory of the Firm: Managerial Behaviour, Agency Costs, and Ownership
Structure’, Journal of Financial Economics, Vol. 3, pp. 305–50.
Kennedy, P. (2008), ‘A Guide to Econometrics’ 6th edn. Blackwell Publishing, Oxford, UK.
Krechovska, M. and P. T. Prochazkova (2014), ‘Sustainability and its Integration into Corporate Governance Focusing
on Corporate Governance Management and Reporting’, Procedia Engineering, Vol. 69, pp. 1144–51.
Kyereboah‐Coleman, A. (2007), ‘Corporate Governance and Shareholder Value Maximization: An African Perspective’,
African Development Review, Vol. 19, No. 2, pp. 350–67.
Lintner, J. (1956), ‘Distribution of Incomes of Corporations among Dividends, Retained Earnings, and Taxes’, The American
Economic Review, Vol. 46, No. 2, pp. 97–113.
Mayer, F. (1997), ‘Corporate Governance, Competition, and Performance’, in S. Deakin and A. Hughes (eds.), Enterprise
and Community: New Directions in Corporate Governance, Blackwell Publishers, Oxford, UK.
Miller, M. H. and F. Modigliani (1961), ‘Dividend Policy, Growth and the Valuation of Shares’, The Journal of Business,
Vol. 34, pp. 411–33.
Nicholson, G. J. and G. C. Kiel (2007), ‘Can Directors Impact Performance? A Case‐Based Test of Three Theories
of Corporate Governance’, Corporate Governance: An International Review, Vol. 15, pp. 585–608.
Ofori‐Sasu, D., J. Y. Abor and A. K. Osei (2017), ‘Dividend Policy and Shareholders’ Value: Evidence from Listed
Companies in Ghana’, African Development Review, Vol. 29, No. 2, pp. 293–304.
© 2019 The Authors. African Development Review © 2019 African Development Bank
422 D. Ofori‐Sasu et al.
Pfeffer, J. and G. Salancik (1978), The External Control of Organizations: A Resource‐Dependence Perspective, Harper and
Row, New York.
Ranti, U. O. (2013), ‘Determinants of Dividend Policy: A Study of Selected Listed Firms in Nigeria’, Change and Leadership,
Vol. No. 17, pp. 107–19.
Ross, S., R. Westerfield, J. Jaffe and B. Jordan (2008), ‘Modern Financial Management’ 8th edn. McGraw Hill, New York.
Sarwar, M. S. (2013), ‘Effect of Dividend Policy on Share Holder's Wealth: A Study of Sugar Industry in Pakistan’,
Global Journal of Management and Business Research Finance, Vol. 13, No. 7, pp. 47–54.
© 2019 The Authors. African Development Review © 2019 African Development Bank