The current issue and full text archive of this journal is available at

www.emeraldinsight.com/1086-7376.htm

Corporate
Ownership structure, corporate governance and
governance and firm performance firm performance
Evidence from an African emerging market
301
Sabur Mollah
School of Business, Stockholm University, Stockholm, Sweden
Omar Al Farooque
School of Business, Economics and Public Policy, University of New England,
Armidale, Australia, and
Wares Karim
School of Economics and Business Administration,
Saint Mary’s College of California, Moraga, California, USA

Abstract
Purpose – In spite of an abundance of corporate governance literature across the world, the
Botswana corporate sector is lacking. The purpose of this study is to investigate the relationship
among the ownership structure, board characteristics and financial performance to determine the role
of corporate governance in the performance behavior of companies listed in such an emerging market
in Africa as Botswana.
Design/methodology/approach – Ordinary least square (OLS) models are applied to Botswana
Stock Exchange listed firms over the period 2000-2007 to determine the role of corporate governance in
the performance behavior of companies listed in an emerging market.
Findings – The empirical evidence shows distinct nature of corporate governance behavior
among alternative performance measures used, in particular, between accounting-based/hybrid and
market-based measures.
Practical implications – Such diversified findings provide the policy-makers with insights to take
appropriate measures on corporate governance and stock market development in order to ensure their
efficiency.
Originality/value – The approach of this study is different from the other available literature as
it captures all types of shareholdings together in addition to other corporate governance and
firm-specific predictable variables.
Keywords Corporate governance, Ownership structure, Financial performance, Emerging markets,
Botswana, Performance management
Paper type Research paper

I. Introduction
The tenets of corporate governance and its augmented forms both in external and
internal mechanisms are designed to protect shareholders as well as stakeholders,
provided proper institutional and market oriented support are in place. This has not
Studies in Economics and Finance
JEL classifications – G32, G34 Vol. 29 No. 4, 2012
The authors acknowledge the financial support from the Office of Research and Development pp. 301-319
q Emerald Group Publishing Limited
(ORD), University of Botswana for this research. The authors thank Asma Mobarek for her 1086-7376
valuable comments on the earlier version of the paper. DOI 10.1108/10867371211266937
SEF been the case in corporate bankruptcies, scandals, failures and stock market crashes,
29,4 all over the world in the early 2000, which contributed to a dysfunctional corporate
culture and categorized firms as myopic, paranoid, bureaucratic and depressive.
Following the massive fallout, practitioners, investors, regulators, and researchers
have raised the question in (silent) wonder: “what went wrong?” This troubled
corporate world signals to all interested parties that corporate governance needs to be
302 scrutinized. Moreover, recent credit crunch has shaken the confidence of investors
irrespective of their respective geographical boundaries. It refuels the debate for the
need of good corporate governance by noting that over the last 50 years, the role and
function of corporate ownership, company board and regulatory bodies have changed
beyond recognition. This raises the question whether corporate governance is
sufficiently decentralized (i.e. control of companies is not centralized in the hands of
management) and/or distributed among different stakeholders so that management
is monitored and controlled properly. In other words, whether shareholders and other
stakeholders are taking part in the corporate governance so that corporate practice
remains unbiased, efficient and goal oriented. Because, there is evidence that firms
categorized by inefficient corporate governance have delivered inferior returns to
shareholders than firms characterized by efficient corporate governance (von Nandelstadh
and Rosenberg, 2003).
Most of the studies assume that managerial ownership of shares is representative of
managerial control and thus used as sole explanatory factor in determining firm
performance without considering holistically the ownership pattern of the rest of the
firm’s shares. It means that the concentration of managerial share ownership is
considered, but the degree of control the ownership confers is not. There are numerous
studies focusing on either the relationship between ownership structure and
performance or the relationship between corporate governance and performance. In
fact, there are plenty of studies conducted on the US and the UK markets in identifying
the influence of corporate governance and ownership structure on firm performance
but a very little is done in African; therefore, the needs of more research in the
emerging markets in Africa are well recognized (Wai and Patrick, 1973).
Nevertheless, Botswana market is one of the most promising emerging stock
markets in Africa. The Botswana Stock Exchange (BSE) is the third largest stock
exchange amongst the Southern African Stock Exchanges (SADC). In Botswana, the
pace of industrialization has suffered in the past due to factors like, excessive
import dependency and lack of local entrepreneurs. However, with the spree of
privatization and the success of stock markets in many of the Southern African
Newly Industrialized Countries (NICs), Botswana presents an optimistic ground for
effectively developing and utilizing the capital market for industrial financing. With
respect to the securities market in Botswana, foreign and national experts have
undertaken a few studies (Charles and Keith, 1999; Swami and Matome, 2001). All of
these studies mainly focused on the market structure of BSE. Importantly, Botswana
market is largely developed after 1990s, which is considered to be green filed in the
capital market research, particularly in the Southern African region. Nonetheless,
there is no recognized study found on the issues around corporate governance,
ownership structure, and firm performance in the Botswana market; therefore, the
existing evidence is lagging behind in identifying the relationship between corporate
governance, ownership structure and firm performance.
 This paper examines whether differences in ownership structure (i.e. diverse Corporate
ownership pattern/type) across firms can explain their respective performance governance and
differences in an emerging economy like Botswana where board characteristics
variables have also been tested concurrently to observe whether they exert any influence firm performance
on firm performance. Specifically, we have attempted to resolve some interesting
questions as follow: does ownership matter? If it does, then which ownership type is
more effective than the others in maximizing firm value? Can ownership be an 303
alternative method in controlling agency cost in the context of an emerging economy? In
this quest, we have investigated the relationship among the ownership structure,
corporate governance variables and firm performance measures. Our objective is to find
out whether diversification of corporate ownership can lead to improving financial
performance of the firm along with other existing corporate governance variables which
ensure efficient allocation of investment, risk sharing, improved financial disclosure and
transparency, etc. Our work differs from prior studies in finding the relationship
among these three corporate issues at a time. Since there is dearth of research in
Botswana in this area, the empirical evidence of this study definitely contributes to the
existing literature in revealing the relationship among ownership structure, corporate
governance and firm performance.
The remainder of the paper is organized as follows: Section II contains the extensive
literature review on ownership structure, corporate governance and firm performance.
Section III focuses on model development and data. Empirical results and analysis are
described in Section IV and finally concluding remarks are observed in Section V.

II. Literature
Corporate governance mechanisms are market, institution and legal settings that
protect outside investors from opportunistic behavior of managers or controlling
shareholders. In the absence of such protection, asymmetries of information and
difficulties of monitoring suffered by outside investors enable managers to misallocate
and expropriate corporate resources, often at the expense of minority investors and the
long-term firm performance. von Nandelstadh and Rosenberg (2003) provide evidence
that firms categorized by inefficient corporate governance have delivered inferior
returns to shareholders while firms characterized by efficient corporate governance
have been valued higher. Shaheen and Nishat (2005) also relate corporate governance
to firm performance and suggest that firms with relatively poor governance are less
profitable, less valuable and pay out less cash to their shareholders. Black et al. (2003)
report evidence that corporate governance is an important factor in explaining the
market value of Korean public companies. Therefore, efficient corporate governance
provides better firm performance, while poor corporate governance leads to bad
financial performance. Bhasa (2006) explains that corporate governance within the
conflict of interests’ framework is subject to behavioral motivation of those who run
the corporations. The profoundness of conflicts of interests lies in the locus of control
by the managers, owners, institutional investors or with the markets.
It is well documented that corporate governance system varies according to the
ownership structure of the corporate sector. At one end of the spectrum, there are
companies where which ownership is dispersed among small shareholders, while control
is concentrated in the hands of managers (Berle and Means, 1932). The dispersed
shareholding is observed in countries with “common law” legal system – USA,
SEF UK (La Porta et al., 1999) – where the Anglo-Saxon corporate governance system relies
29,4 on sophisticated legal protection of investors from appropriation by managers.
Generally, voting on important internal (e.g. election of the board of directors) and
external (e.g. mergers and liquidations) corporate matters is the main means of control
(Easterbrook and Fischel, 1983). Hence, the enforcement of voting rights is the key issue
of the Anglo-Saxon corporate governance system. While, in the other end of the
304 spectrum, there are companies with concentrated ownership of large investors
(Grossman and Hart, 1980; Shleifer and Vishny, 1986). In such companies, managers’ act
at the dictate of the controlling shareholder(s) or debtor(s). The concentrated ownership
is common for countries where it is quite costly for small investors to exercise both their
control and cash flow rights. Large investors enjoy economies of scale and reduce
traditional free rider problem. The continental Europe and Japan experience corporate
governance conducted by large investors (La Porta et al., 1999).
The empirical analysis of documents shows us two conflicting ideas for testing
ownership structure and performance relationship based on the effect of direct
shareholder monitoring on performance. The first group argues that the positive
relation between ownership and performance is a result of incentive alignment and this
alignment can be achieved if managers pursue interest of general shareholders. The
second group looks for a negative relation between ownership and performance, often
called the entrenchment argument, where concentrated owners focus less on firm
performance and instead maximize their own utility (Morck et al., 1988; McConnell and
Servaes, 1990). Minority shareholders will, in this case, have no effect because of the
presence of dominating owners. This argument in many ways is closely related to the
Leibenstein X-efficiency arguments where the negative relation is based on the cost of
capital (Fama and Jensen, 1983b). When an investor can no longer optimize the portfolio
of assets, but must decrease liquidity in order to obtain high ownership concentration,
the firm performance is to decrease as a result. Nevertheless, another group argues
for incentive alignment and entrenchment argument at different levels/degrees of
ownership stake (Morck et al., 1988; Stulz, 1988; Farooque et al., 2007).
The discussion about the relationship between ownership structure and corporate
performance is based on theoretical as well as empirical arguments. A number of studies
have sought to evaluate empirically the link between managerial share ownership and
firm performance. The results, however, are not uniformly in agreement. Demsetz (1983)
argues that there should be no relationship between ownership structure and firm
performance. Pursuing this argument empirically, Demsetz and Lehn (1985) find no
significant correlation between profit rates and various measures of ownership
concentration. Tsetsekos and DeFusco (1990) report no significant differences in the
returns on the various portfolios constructed according to managerial share ownership.
In contrast, Mehran (1995) provides evidence of a positive relationship between
managerial equity ownership and firm performance. Welch (2003) supports that firm
performance is statistically dependent on managerial ownership for Australian firms.
Similarly, Wruck (1988) finds a strong and positive link between the change in
ownership concentration and firm performance. A good number of empirical studies also
suggest positive relation between the concentration ownership and corporate
performance. For German corporations, Gorton and Schmid (1996) find that block
holders improve company performance. Hiraki et al. (2003) show that insider ownership
is positively related to firm value in Japan. In Japanese corporations, large shareholders
replace badly performed managers more often than dispersed ones (Kaplan and Minton, Corporate
1994). On the contrary, Agrawal and Mandelker (1990) and Loderer and Martin (1997) governance and
report a negative correlation between managerial ownership concentration and firm
performance. Tam and Tan (2007) also suggest that ownership concentration is firm performance
negatively related to performance in Malaysia. Sarkar and Sarkar (1999, 2000) provide
evidence on the role of large shareholders in monitoring company value. They find that
block-holdings by directors’ increases firm value after a certain level of holdings. Short 305
and Keasey (1999), Dwivedi and Jain (2002), Davies et al. (2005) and Farooque et al. (2007)
document non-linear relation between managerial ownership and firm performance.
A few studies on other types of ownership show mixed results on their relationship
with firm performance. Steiner (1996) and Xu and Wang (1999) find a positive correlation
between performance and institutional equity holdings, while Black (1998) suggests that
shareholder activism by institutional investors has little effect on firm performance.
Sarkar and Sarkar (1999, 2000) do not obtain any evidence of active governance from
institutional investors. With regard to government ownership, Boardman and Vining’s
(1989) reveal negative performance effect of state ownership. Xu and Wang (1999)
also find that firm profitability is either negatively correlated or not correlated with
government shareholding. Regarding impact of foreign equity ownership, Sarkar and
Sarkar (1999, 2000) highlight that foreign equity ownership has a beneficial effect on
company value. Dwivedi and Jain (2002) provide evidence that a higher proportion of
foreign shareholding is associated with increase in market value of firms in India. But
Kumar (2002) concludes that the foreign shareholding in Indian firms does not influence
the firm performance significantly. This is in sharp contrast with other existing studies
with respect to India and other developing countries.
Finally, Dwivedi and Jain (2002) report that public shareholding has a negative
association with performance. Despite the above inconclusive evidence on ownership
structure patterns/types/variables, Clark (2008) argues for a causal relationship
between the presence of a principal owner and the decision to contribute to ideological
politics. He proposes ownership structure should be included as a variable in future
studies of corporate political behavior.
Regarding the board characteristics variables, there is mixed evidence in the empirical
literature linking board size to corporate performance. One group of researchers predicts
board size to have a positive association with firm performance (Pearce and Zahra, 1992;
Dwivedi and Jain, 2002) while another group has shown a negative relationship (Yermack,
1996; Hermalin and Weisbach, 2003). Meanwhile, yet another group has arrived with a
non-linear or an inverted “U” shaped relationship (Vafeas, 1999; Golden and Zajac, 2001).
Bigger boards are expected to have representation of people with diverse backgrounds
and thus expected to bring knowledge, wider perspective and intellect to the board.
Board size is also associated with presence of more outsiders, who foster more careful
decision-making policy in firms. This is because the reputation cost for outside directors
in case of a firm’s failure is likely to be high. Their reputation, however, is not enhanced
by an equal measure with the firm’s success. On the flip side, larger groups also suffer
from a problem of diffusion of responsibility or “social loafing”, wherein individuals
discount the likelihood that others will detect their poor contributions. Dahya and
Mcconnell (2005) report that company appointed directors to conform with the standard
of calling for at least three outside directors for publicity traded corporations exhibited
a significant improvement in operating performance both in absolute and relative
SEF terms to various peer-group benchmarks. They also find increase in stock prices around
29,4 announcement that outside directors are included in consistent with this
recommendation. However, they do not endorse mandatory board structure. But the
evidence indicates that such a mandatory board structure is associated with an
improvement in performance in UK companies. Again, the presence of audit committee,
where non-sponsor director chairs the committee, ensures proper disclosure and greater
306 accuracy of financial statements and is expected to positively relate to corporate
performance. If any sponsor director acts as the chairman in the audit committee, the
person may pursue to hide his/her own improper actions and some material information
to investors.

III. Data and model

All the 19 companies listed in Botswana stock market have been considered as the
primary sample of this study over the period of 2000-2007. Daily market prices of firms
and Botswana market Domestic Index (DCI) for the same period are collected from BSE
price quotations. Company specific data have been collected for the annual reports of
the sample companies for the period of 2000-2007. For the information on the
ownership structure and corporate governance practices in Botswana, we have
conducted a survey over 19 companies from November 2007 to April 2008.
The paper attempts to investigate the effect of ownership structure and corporate
governance on firm performance. Firm performance is measured using
accounting-based measures (return on asset (ROA) and return on equity (ROE)),
market-based measure (market capitalization) and hybrid-type measure (TOBIN’S Q)
for the Botswana firms. Sponsor director shareholding is considered to test incentive
alignment, entrenchment, and agency hypotheses. Sponsor director shareholding help
board making better judgment due to the inside information and, particularly, enhance
firm performance because of incentive alignment hypothesis. On the other hand,
entrenchment hypothesis suggest that sponsor shareholding destroys firm value
because directors take negative NPV projects and concentrate on the directors’
remuneration maximization, which creates agency problems too between shareholder
and manager (Jensen and Meckling, 1976). If the sponsor director shareholding reject
incentive alignment hypothesis, we will find a positive relationship and this
relationship, in particular, will reject agency hypothesis too. Otherwise, the results will
reject entrenchment hypothesis. Public shareholding is also considered to test incentive
alignment, entrenchment, and agency hypotheses. If public shareholding and firm
value have positive relation, this rejects incentive alignment and agency hypotheses
but opposite happens due to the negative relationship.
Institutional and foreign shareholdings are considered as the value enhancing
parameters in the paper. These variables are also tested for corporate governance
systems of the listed firms in Botswana. Higher institutional and foreign ownership are
considered to be better monitoring mechanisms of the listed firms and eventually
enhances firm value. However, chairing audit and executive committees, and board
size are the corporate governance parameters of this study. Bigger board ensures better
governance and hence value of the firm. Chairing audit and executive committees by
the sponsor director are tested against entrenchment hypothesis. A significant
relationship between these variables and firm value rejects entrenchment hypothesis
and hence destroy the firm value.
 Building on the review of extant literature discussion above, the main null Corporate
hypotheses of the study is to be tested as follows:
governance and
Ho1. Ownership structure of Botswana firms has no influence on firm performance firm performance
measures.
Ho2. Board characteristics of Botswana firms have no influence on firm
performance measures. 307
We use multiple regression models to determine the relationship between ownership
structure, board characteristics and financial performance. Our basic model is as follows:
Performanceit ¼ a þ bðownershipÞi þ gX i þ lZ i þ 1it
where:
Performanceit ¼ performance of the ith firm at time period “t”.
Ownership ¼ ownership structure of the ith firm.
Xi ¼ firm’s specific variable of the ith firm.
Zi ¼ corporate governance variables of ith firm.
1it ¼ error term.
We have built the above model into two stages: at the first stage, we have taken
ownership structure/pattern/type variables and some firm specific control variables
while at the second stage, we have added some corporate governance variables, in
particular board characteristic variables to the original one. Based on existing literature,
we have used both accounting-based and market-based performance measures as the
dependent variable such as, ROA, ROE, and log of Market Capitalization (LnMktCap),
TOBIN’S Q. Similarly, for ownership structure, we have included five variables –
Sponsor Holdings, Government Holdings, Institutional Holdings, Public Holdings
and Foreign Holdings. Finally, beta of the firm, size of the firm and industry dummy are
used as controls. So, our model 1 looks like the following:
Model 1
ROA=ROE=LnMktCap=Q ¼ a þ b1 SPONSOR þ b2 GOVT þ b3 INSTIN
þ b4 PUBLIC þ b5 FOREIGN þ g1 BETA þ g2 SIZE
þ g3 INDDUMY þ 1it
Now, we extend model 1 with the inclusion of three corporate governance/board
characteristics variables. Number of directors, BOARDSIZE, audit committee headed by
sponsor director, CHAR_AC and the executive committee headed by sponsor director,
CHAR_EC, are included in the extended model 2 below as corporate governance
variables. Thus, our final model is[1]:
Model 2
ROA=ROE=LnMktCap=Q ¼ a þ b1 SPONSOR þ b2 GOVT þ b3 INSTIN
þ b4 PUBLIC þ b5 FOREIGN þ g1 BETA þ g2 SIZE
þ g3 INDDUMY þ l1 BORDSIZE þ l2 CHAR_AC
þ l3 CHAR_EC þ 1it
SEF It is noted here that we have also tested “fixed-firm effects” of both models to observe
29,4 whether there is any influence of particular year of the data set or unobserved firm
heterogeneity. Therefore, year dummies (YDUM) are used in each model separately.

IV. Results and analysis

To identify the impact of ownership structure and board characteristics on the firm
308 performance in Botswana, the final regression models is tested for Ramsey’s (1969)
RESET test and White (1980) test to check heteroskedasticity problem. But both the
tests are unable to reject the hypothesis of homoskedasticity. Moreover, Durbin-Watson
is found close to two for cross-sectional pooled data, which signals no autocorrelation
problem of the models.
Table AII in the Appendix presents a detailed structure of data which reflects the
mean, minimum, maximum and standard deviation of both dependent and independent
variables. ROA varies between 2 0.26 and 22.22 with the mean of 0.4529 while mean
ROE is 1.22. Similarly, mean LnMktCap and TOBIN’S Q is, respectively, 5.44 and 0.0029.
With regard to ownership structures, average institutional shareholding is 36.76 that
varies between 0 and 97.84 and indicates the highest concentration in company’s equity.
The second largest concentration group is sponsor director holdings having mean value
of 9.88. Minority and foreign shareholdings, respectively, are average ownership of 8.79
and 11.42 of the total outstanding shares of listed firms. Beta of the firms disperse from
2 0.56 to 1.12 that indicates the risk association with the market index, the negative beta
indicates that the particular stock is negatively correlated with the overall market,
i.e. when the overall market moves downward, the negative beta stock moves upward
and vice versa. This type of stock may also help the investors to balance the portfolio in
the time of overall market downturn. Finally, firm size varies from 9.54 to 19.97 with an
average of 14.
The Pearson’s correlation matrix in Table AIII in the Appendix shows the expected
relationship of all the independent variables with ROA, ROE, LnMktCap and TOBIN’S
Q with few exceptions. However, the correlation matrix also shows the correlation
between the independent variables is either low degree or moderate degree, which
suggests the absence of multicolinearity. As suggested by Bryman and Cramer (1997),
the Pearson’s correlation between each pair of independent variables should not
exceed 0.80; otherwise independent variables with a coefficient in excess of 0.80 may
be suspected of exhibiting multicollinearity. Moreover, the collinearity diagnostics,
provided by SPSS including collinearity statistics (tolerance and variance inflated
factor “VIF”), condition index and variance proportion, support the Pearson’s
correlation coefficients and document. Therefore, the results produce no evidence of
multicollinearity problem in the regression models.
Table AIV in the Appendix reports the regression results from multiple regression
analysis of model 1. The results indicate that in terms of model explanatory power and
F-statistics, the dependent variable LnMktCap (market-based performance measure)
shows stronger relationship with independent variables than dependent variables
ROA and ROE (accounting-based performance measures) and TOBIN’S Q
(hybrid performance measure). The F figure is statistically insignificant for ROA at
9.80 percent without year dummies and 23 percent with year dummies, that is, more
than 5 and 10 percent level. The coefficient of multiple determinations for ROA, i.e. R 2
is 10.6 percent without year dummies and 14.8 percent with year dummies,
respectively, variation of dependent variable can be explained by the joint variation of Corporate
independent variables, but whenever the degrees of freedom has been adjusted the governance and
overall model becomes least weaker (denoted by adjusted R 2, i.e. only 4.5 and 3.2
percent, respectively). Both ROE and TOBIN’S Q also reveal similarity as found in firm performance
ROA. However, ROA is relatively better than ROE and TOBIN’S Q as performance
measure, which shows foreign shareholding has gotten significant positive
performance effect. With regard to the model validity with performance measures, 309
similar trend is observed in the extended model 2 (Table AV) for ROA, ROE and
TOBIN’S Q as measures of performance. So, it can be conferred that accounting-based
and hybrid measures of performance are less relevant to be explained by ownership
structure variables. Demsetz and Villalonga (2001) report a correlation between
accounting profit rate and TOBIN’S Q at about 0.6 by pointing a choice between the
estimates of what management has accomplished and the estimate based on what
management will accomplish. While accounting profit rate based on accounting data is
biased in a number of situations (e.g. changes in accounting practices, tax rules
and extraordinary short-run fluctuations in income), TOBIN’S Q is also be sensitive
to problems with valuation of, e.g. intangible capital. Thus, encountering problems
in ROA, ROE and TOBIN’S Q as measures of performance, we have relied on
market-based performance measure, i.e. LnMktCap in our subsequent discussion on
Tables AIV and AV.
In Table AIV, the model is statistically significant (i.e. F-statistics) with moderate
level of adjusted R 2 when LnMktCap is used as dependent variable for both with and
without year dummies. In case of ownership structure variables, both SPONSOR and
GOVT show negative performance effect. This implies that both sponsor or government
ownership is value destructive, i.e. market capitalization of the firm decreases for any
increase in their proportion of share ownership to the firm’s total shares. On the other
hand, INSTIN and PUBLIC reveal positive but insignificant effect on performance.
However, FOREIGN shows positive impact on performance at 10 percent level of
significance without year dummies and positive insignificant with year dummies. With
regard to control variables, industry dummy (INDDUMY) variable provides positive
relationship with performance which indicates that market capitalization is more
industry specific. In terms of year influence, year 2007 has positive performance effect.
From this discussion, we argue that firm performance cannot be fully explained by only
ownership variables and firm specific variables, i.e. there are some important corporate
governance variables that might influence firm’s performance. This is captured by our
extended model 2 to including some important board characteristics variables. This
expanded model produces better results in explaining firm’s performance as presented
in Table AV in the Appendix.
In Table AV, we rely on market-based performance measure, i.e. LnMktCap rather
than ROA, ROE and TOBIN’S Q, as mentioned earlier for both with and without year
dummies. Similar to Table AIV results, both SPONSOR and GOVT show negative
performance effect, i.e. value destructive. But unlike Table AIV results, both INSTIN
and FOREIGN have changed their sign from positive to negative where INSTIN has
appeared to have negative significant relation with performance and FOREIGN has
negative significant relation with firm performance only when year dummies are
included. These findings are consistent with Agrawal and Mandelker (1990), Loderer
and Martin (1997), Tam and Tan (2007), Black (1998), Boardman and Vining’s (1989),
SEF Xu and Wang (1999) and Kumar (2002), respectively. On the contrary, PUBLIC has
29,4 emerged to have positive significant impact on performance for both with and without
year dummies. Thus, it is evident that only increased minority shareholdings can
improve financial performance of listed companies in Botswana stock market. This
finding is in contrast to the finding of Dwivedi and Jain (2002). The negative significant
performance effect of SPONSOR, GOVT, INSTIN and FOREIGN indicate the presence
310 of high agency problem in Botswana stock market. Of these four ownership structure
variables, GOVT has minimum role due to a very low proportion of average
shareholding as shown in Table AIII. Therefore, we argue that agency conflict lies with
SPONSOR, INSTIN and FOREIGN as the largest ownership concentration groups and
both entrenchment and strategic alliance/conflict of interest hypothesis applies to
them. It is of course a bit puzzling to observe that FOREIGN shareholding is having a
value destroying relation despite the expectation that they could improve corporate
governance mechanism viz-a-viz firm performance by reducing agency costs.
Apart from ownership structure variables, the relationship between board
characteristics variables and firm performance in Table AV is rather interesting as
well as conflicting to some extent. Of the three variables, BOARDSIZE is found to have
positive but insignificant relation with performance for the year dummies both with
and without. On the other hand, the variable CHAR_AC reveals positive significant
performance effect at 1 percent level of significance while CHAR_EC negative
significant performance effect at 10 percent level of significance only when year
dummies are not included. That is, when audit committee is chaired by the sponsor
director, then the firm performance improves and when the executive committee is
chaired by the sponsor director, the performance decreases. Both the committees are
headed by a sponsor director of the firm but shows conflicting relationship with firm
performance. While the former is associated with alignment of interest, nevertheless
the latter is linked to entrenchment and related agency costs for the firm. One reason
for CHAR_AC to have positive significant performance effect is its direct involvement
with external and internal auditors, who are supposed to have more independence than
other insiders in the firm.
Lastly, in regards to control variables, similar to Table AIV, INDDUMY reveals
positive significant performance effect in Table AV while the rest is showing positive
but insignificant relation. This signifies that firm performance is more industry
specific than firm size or firm level risk. In terms of year influence, year 2007 has
positive performance effect which is also similar to Table AIV.
Thus, from the above discussion, we confer that Ho1 is fully rejected, as all
ownership structure variables except one (PUBLIC) are showing negative significant
relation with LnMktCap as the measure of market-based performance. This implies
unworthy presence of agency conflict with entrenchment and/or strategic
alliance/conflict of interest effect, which to some extent mitigated by the increased
stakes of minority shareholders by dint of their monitoring capacity. Again, our Ho2
is partially rejected, as one (CHAR_AC) out of three board characteristics variables
has appeared to have shown positive significant relation with LnMktCap as the
measure of performance. Finally, we observe no strong presence of fixed-firm effects
or unobserved firm heterogeneity in explaining firm performance. The changes,
observed during the sample period, are having either a minimum significance or of no
significance at all.
V. Conclusion Corporate
This study examines whether firm performance is influenced by different ownership governance and
structure and board characteristics variables for all listed firms in Botswana stock
market for the period of 2000-2007. In the developed regression model, alternative firm performance
performance measures are used as dependent variable (e.g. ROAs, ROE, LnMktCap
and TOBIN’S Q) while ownership structure variables (e.g. sponsor directors’
ownership, government ownership, institutional ownership, public ownership and 311
foreign ownership) and important corporate governance parameters (e.g. board size,
chairing audit committee and executive committee by the sponsor director) are
considered as the explanatory variables of interest. In addition, a number of control
variables (e.g. industry dummy, beta, firm size and year dummies) are also included in
the prescribed model.
We have provided empirical evidence that accounting-based and hybrid
performance measures are not fitted for the prescribed model (i.e. ROA, ROE and
TOBIN’S Q). It is market-based performance measure (i.e. LnMktCap) that can explain
the role of ownership structure and board characteristics variables on firm’s financial
performance. It implies that Botswana corporate sector has been advanced to market
oriented system in developing appropriate corporate governance mechanisms and
mitigating existing agency conflicts. Despite the importance of accounting numbers,
our findings denote that investors are less attractive to accounting-based and/or hybrid
performance measure(s), which is prone to accounting manipulations of earnings
management or smoothing. Rather, investors put value on free market operation based
on available information about the listed firms.
On the market-based performance measures as dependent variable (i.e. LnMktCap),
our findings suggest that all major ownership concentration groups (e.g. SPONSOR,
INSTIN, GOVT and FOREIGN) are destructive to firms’ financial performance and
value except minority shareholdings (e.g. PUBLIC), which is consistent to the tenets of
agency theory (i.e. conflict between majority and minority owners). It is dispersed
ownership that improves firm performance and mitigate agency conflicts in the
corporate sector of Botswana stock market. In addition, chairing audit committee and
executive committee also play vital role as corporate governance mechanism to
influence firm’s performance in both positive and negative directions, respectively.
Such opposite response to performance is a bit puzzling though audit committee has a
direct link to external and internal auditors to have positive performance effect or at
the least a compensating balance of having negative performance effect of SPONSOR
shareholdings. On the other hand, executive committee’s role is consistent with the
perspective of agency theory. Finally, we have also observed that firm performance is
industry specific in Botswana rather than firm specific risks and size and there is
no persistent presence of fixed-firm effects or unobserved firm heterogeneity to explain
firm performance.
The limitation of this research is the small sample size (population of 19 firms for
eight years) from a relatively small BSE. Despite the limitation, this paper significantly
contributed to literature on the issue in African emerging markets because the data set
used in the research is a brand new hand collected and survey data, which is unique.
This study is focused on the unidirectional relationship between ownership structure
and other corporate governance variables and firm performance measures and it has
not captured endogeneity/simultaneity of ownership and performance relationship due
SEF to small sample size. As such, we expect that any future research in the field needs to
29,4 overcome these limitations. A useful extension of the analysis would include additional
policy variables measuring changes in the market conditions such as adoption of IFRS
as accounting standard, trade or tax policy changes, etc. to see whether ownership
structure changes noticeably or not. If so, to what extent and why? Do companies in
emerging markets actually raise substantial equity finance? Who are the buyers of this
312 equity? If they are dispersed minority shareholders, why do they buy equity despite the
apparent absence of minority interest protections? These key questions, however, are
left for further research in the days to come.

Note
1. Variables definitions of the model are presented in Table AI in the Appendix.

References
Agrawal, A. and Mandelker, G.N. (1990), “Large shareholders and the monitoring of managers:
the case of antitakeover charter amendments”, Journal of Financial and Quantitative
Analysis, Vol. 25 No. 2, pp. 143-61.
Berle, A.A. and Means, G.C. (1932), The Modern Corporation and Private Property, Revised
edition 1967, Harcourt, Brace & World, New York, NY.
Bhasa, M.P. (2006), “Ownership structure and firm performance: a review of literature”, Icafai
Journal of Corporate Governance, Vol. 4 No. 4, pp. 29-49.
Black, B. (1998), “Shareholder activism and corporate governance in the United States”,
in Newman, P. (Ed.), The New Palgrave Dictionary of Economics and the Law, Macmillan,
London.
Black, B.S., Jang, H. and Kim, W. (2003), “Does corporate governance affect firm value?”, paper
presented at Workshop Paper, available at: www.law.uchicago.edu/Lawecon/ workshop-
papers/black-jang-kim-stanford.pdf
Boardman, A.E. and Vining, A.R. (1989), “Ownership and performance in competitive
environments: a comparison of the performance of private, mixed and state-owned
enterprises”, Journal of Law & Economics, Vol. 32, pp. 1-33.
Bryman, A. and Cramer, D. (1997), Quantitative Data Analysis with SPSS for Windows,
Routledge, London.
Charles, C. and Keith, R.J. (1999), “An event study of Botswana, Zimbabwe, and Johannesburg
stock exchanges”, South African Journal of Business Management, Vol. 30 No. 4,
pp. 131-40.
Clark, M.S. (2008), “Firm ownership structure and corporate political behavior: a causal”,
available at: www.allacademic.com
Dahya, J. and Mcconnell, J.J. (2005), “Board composition, corporate performance and the Cadbury
committee recommendation”, available at: www.ssrn.com
Davies, R., Hillier, D. and McColgan, P. (2005), “Ownership structure, managerial behaviour and
corporate value”, Journal of Corporate Finance, Vol. 11, pp. 645-60.
Demsetz, H. (1983), “The structure of ownership and the theory of the firm”, Journal of Law
& Economics, Vol. 26, pp. 375-90.
Demsetz, H. and Lehn, K. (1985), “The structure of corporate ownership: causes and
consequences”, Journal of Political Economy, Vol. 93, pp. 1155-77.
Demsetz, H. and Villalonga, B. (2001), “Ownership structure and corporate performance”, Journal Corporate
of Corporate Finance, Vol. 7, pp. 209-33.
governance and
Dwivedi, N. and Jain, A.K. (2002), “Corporate governance and performance of Indian firms: the
effect of board size and ownership”, available at: www.ingentaconnect.com/content/klu/ firm performance
errj/2005/00000017/00000003/00006939
Easterbrook, F.H. and Fischel, D.R. (1983), The Economic Structure of Corporate Law, Harvard
University Press, Cambridge, MA. 313
Fama, E.F. and Jensen, M.C. (1983b), “Separation of ownership and control”, The Journal of Law
& Economics, Vol. 26, pp. 301-25.
Farooque, O.A., van Zijl, T., Dunstan, K. and Karim, W. (2007), “Corporate governance in
Bangladesh: link between ownership and financial performance”, Corporate Governance:
An International Review, Vol. 15 No. 6, pp. 1453-68.
Golden, B.R. and Zajac, E. (2001), “When will boards influence strategy? Inclination X
power¼strategic change”, Strategic Management Journal, Vol. 22 No. 12, pp. 1087-111.
Gorton, G. and Schmid, F. (1996), “Universal banking and the performance of German firms”,
Working Paper 5453, NBER.
Grossman, S. and Hart, O. (1980), “Takeover bids, the free-rider problem, and the theory of the
corporation”, Bell Journal of Economics, Vol. 11, pp. 42-64.
Hermalin, B.E. and Weisbach, M.S. (2003), “Board of directors as an endogenously-determined
institution: a survey of economic literature”, Economic Policy Review, Vol. 9, pp. 7-26.
Hiraki, T., Hideaki, I., Ito, A., Kuroki, F. and Masuda, H. (2003), “Corporate governance and firm
value in Japan: evidence from 1985 to 1998”, Pacific-Basin Finance Journal, Vol. 11,
pp. 239-65.
Jensen, M.C. and Meckling, W.H. (1976), “Theory of the firm: managerial behavior, agency costs
and ownership structure”, Journal of Financial Economics, Vol. 3, pp. 305-60.
Kaplan, S. and Minton, B. (1994), “Appointments of outsiders to Japanese boards: determinants
and implications for managers”, Journal of Financial Economics, Vol. 36, pp. 225-57.
Kumar, J. (2002), “Does ownership structure influence firm value? Evidence from India”, paper
presented at Conference Paper, 2005, available at: www1.fee.uva.nl/fm/Conference/
Hyderabad/Jayesh_CG.pdf
La Porta, R., Florencio, L.S., Shliefer, A. and Vishny, R. (1999), “Corporate ownership around the
world”, Journal of Finance, Vol. 54 No. 2, pp. 471-517.
Loderer, C. and Martin, K. (1997), “Executive stock ownership and performance: tracking faint
traces”, Journal of Financial Economics, Vol. 45, pp. 223-55.
McConnell, J.J. and Servaes, H. (1990), “Additional evidence on equity ownership and corporate
value”, The Journal of Financial Economics, Vol. 27, pp. 595-612.
Mehran, H. (1995), “Executive compensation structure, ownership and firm performance”,
Journal of Financial Economics, Vol. 38, pp. 163-84.
Morck, R., Shleifer, A. and Vishny, R.W. (1988), “Management ownership and market valuation –
an empirical analysis”, The Journal of Financial Economics, Vol. 20, pp. 293-315.
Pearce, J.A. II and Zahra, S.A. (1992), “Board composition from a strategic management
perspective”, Journal of Management Studies, Vol. 29 No. 4, pp. 411-38.
Ramsey, J.B. (1969), “Tests for specification errors in classical linear least squares regression
analysis”, J. Roy. Statist. Soc. B., Vol. 31 No. 2, pp. 350-71.
Sarkar, J. and Sarkar, S. (1999), The Governance of Indian Corporations, India Development
Report, Oxford University Press, Oxford, pp. 201-18.
SEF Sarkar, J. and Sarkar, S. (2000), “Large shareholder activism in corporate governance in
developing countries: evidence from India”, International Review of Finance, Vol. 1 No. 3,
29,4 pp. 161-94.
Shaheen, R. and Nishat, M. (2005), “Corporate governance and firm performance – an exploratory
analysis”, paper presented at Conference Paper, available at: http://ravi.lums.edu.pk/cmer/
Conference2005/images/Rozina%20and%20Dr.Nishat%20corporate%20Governance.pdf
314 Shleifer, A. and Vishny, R.W. (1986), “Large shareholders and corporate control”, Journal of
Political Economics, Vol. 94 No. 3, pp. 461-88.
Short, H. and Keasey, K. (1999), “Managerial ownership and the performance of firms: evidence
from the UK”, Journal of Corporate Finance, Vol. 5, pp. 79-101.
Steiner, T.L. (1996), “A re-examination of the relationship between ownership structure, firm
diversification and Tobin’s Q”, Quarterly Journal of Business and Economics, Vol. 4,
pp. 39-48.
Stulz, R. (1988), “Managerial control of voting rights: financing policies and the market for
corporate control”, Journal of Financial Economics, Vol. 20, pp. 25-54.
Swami, B.N. and Matome, T.T.K. (2001), “Development of capital market in Botswana”, The
African Journal of Finance, Vol. 3 No. 1, pp. 31-58.
Tam, O.K. and Tan, M.G. (2007), “Ownership, governance and firm performance in Malaysia”,
Corporate Governance, Vol. 15, pp. 208-22.
Tsetsekos, G.P. and DeFusco, R. (1990), “Portfolio performance, managerial ownership, and the
size effect”, The Journal of Portfolio Management, Vol. 15 No. 1, pp. 13-18.
Vafeas, N. (1999), “Board meeting frequency and firm performance”, Journal of Financial
Economics, Vol. 53, pp. 113-42.
von Nandelstadh, A. and Rosenberg, M. (2003), “Corporate governance mechanisms and firm
performance: evidence from Finland”, Working Papers 497, Hanken School of Economics,
Helsinki.
Wai, U.T. and Patrick, H.T. (1973), “Stock and bond issues and capital markets in less developed
countries”, International Monetary Fund Staff Papers 20, pp. 253-305.
Welch, E. (2003), “The relationship between ownership structure and performance in listed
Australian companies”, Australian Journal of Management, Vol. 28 No. 3.
White, H. (1980), “A Heteroskedasticity-consistent covariance matrix estimator and a direct test
for heteroskedasticity”, Econometrica, Vol. 48 No. 4, pp. 817-38.
Wruck, K.H. (1988), “Equity ownership concentration and firm value”, Journal of Financial
Economics, Vol. 23, pp. 3-28.
Xu, X. and Wang, Y. (1999), “Ownership structure and corporate governance in Chinese stock
companies”, China Economic Review, Vol. 10, pp. 75-89.
Yermack, D. (1996), “Higher market values of companies with a small board of directors”, Journal
of Financial Economics, Vol. 40 No. 2, pp. 185-211.

Corresponding author
Sabur Mollah can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected]

Or visit our web site for further details: www.emeraldinsight.com/reprints
Appendix Corporate
governance and
firm performance
Proxies Calculations

Dependent variable
Firm performance Return on assets (ROA) Net profit/total assets 315
Return on equity (ROE) Net profit/shareholders’ equity
LnMktCap Natural LnMktCap
TOBIN’S Q Market value to book value of the firm
Independent variables
Ownership variables: Sponsor shareholdings Percentage of held by sponsor
(SPONSOR) directors
Government holdings (GOVT) Percentage of held by government
Institutional shareholdings (INST)
Percentage of held by institutions
Public shareholdings (PUBLIC) Percentage of held by general public
Foreign shareholdings (FOREIGN)Percentage of held by foreign
individuals and institutions
Firm specific variables: Risk parameter (beta) Dimson’s beta
Firm size Natural log of total assets
Industry dummy: Impulse dummy for industry 1 ¼ specific industry, otherwise 0
Corporate governance Board size Number of member of board of
variables: directors
Chairing internal audit committee Impulse dummy (1 if internal audit
(CHAR_AC) committee is headed by sponsor
director, otherwise 0)
Chairing executive committee Impulse dummy: 1 if executive Table AI.
(CHAR_EC) committee is headed by sponsor The variables and
director, otherwise 0) their proxies

n Mean Minimum Maximum SD

ROA 130 0.4529 2 0.26 22.22 2.15202

ROE 124 1.2205 2 0.26 59.20 6.15326
LNMKTCAP 134 5.4356 2.45 8.94 1.64991
TOBIN’S Q 124 0.0029 0 0.68 0.06103
SPONSOR 135 9.8840 0.00 86.20 22.03483
GOVT 134 0.4949 0.00 8.29 1.97154
INSTIN 135 36.7612 0.00 97.84 38.06131
PUBLIC 135 11.4148 0.00 59.45 14.21286
FOREIGN 135 8.7887 0.00 74.86 21.98444
BETA 132 0.1423 2 0.56 1.12 0.25754
SIZE 130 13.9968 9.54 19.97 2.72332
INDDUMY 135 0.3407 0.00 1.00 0.47572
BORDSIZE 135 6.7185 0.00 13.00 2.78228
CHAR_AC 135 0.4444 0.00 1.00 0.49875 Table AII.
CHAR_EC 135 0.4815 0.00 1.00 0.50152 Descriptive statistics
 29,4
SEF

matrix
316

Table AIII.
Inter-item correlation
TOBIN’S CHAR_ CHAR_
ROA ROE LNMKTCAP Q PUBLIC SPONSOR GOVT INSTIN FOREIGN INDDUMY BETA SIZE BORDSIZE AC EC

ROA 1.00
ROE 0.513 * * 1.00
LNMKTCAP 0.082 0.154 1.00
TOBIN’S Q 0.112 0.108 0.234 1.00
PUBLIC 20.072 2 0.004 2 0.178 * 0.001 1.00
SPONSOR 20.064 2 0.065 2 0.339 * * 2 0.043 0.302 * * 1.00
GOVT 20.029 2 0.042 2 0.347 * * 0.100 0.380 * * 2 0.114 1.00
INSTIN 20.027 0.087 0.136 0.112 0.071 2 0.198 * 20.153 1.00
FOREIGN 0.271 * * 0.102 0.385 * * 0.109 2 0.096 2 0.176 * 0.066 20.162 1.00
INDDUMY 0.151 0.208 * 0.648 * * 0.234 * 2 0.115 2 0.084 20.179 * 0.085 0.479 * * 1.00
BETA 0.077 0.254 * * 0.183 * 0.246 * * 0.116 0.026 20.116 0.069 2 0.113 0.151 1.00
SIZE 20.133 2 0.074 2 0.019 2 0.032 0.164 2 0.047 0.241 * * 20.073 2 0.006 20.004 2 0.070 1.00
BORDSIZE 20.006 0.001 0.364 * * 2 0.120 2 0.101 2 0.145 0.026 20.005 0.157 0.490 * * 0.136 0.184 * 1.00
CHAR_AC 0.137 0.008 0.300 * * 2 0.221 * 0.051 2 0.294 * * 0.284 * * 0.446 * * 0.449 * * 0.206 * 2 0.080 0.102 0.209 * 1.00
CHAR_EC 0.117 2 0.015 0.242 * * 2 0.045 0.027 0.261 * * 20.241 * * 20.403 * * 0.342 * * 0.402 * * 0.034 0.017 0.199 * 20.056 1.00
Model 1 Dependent variable ROA ROE LnMktCap TOBIN’S Q

(Constant) 1.508 2.246 * 2.457 0.694 4.585 * * * 4.616 * * * 0.014 0.012

(1.380) (1.867) (0.671) (0.159) (8.162) (7.797) (0.380) (0.288)
Independent variables SPONSOR 2 0.015 2 0.014 2 0.037 2 0.027 2 0.325 * * * 2 0.311 * * * 2 0.056 2 0.045
(2 0.143) (2 0.134) (2 0.347) (2 0.256) (2 4.751) (2 4.681) (2 0.524) (2 0.420)
GOVT 0.003 2 0.002 0.035 0.039 2 0.348 * * * 2 0.343 * * * 0.020 0.031
(0.030) (2 0.018) (0.320) (0.350) (2 4.744) (4.817) (0.181) (0.278)
INSTIN 0.008 0.000 0.026 0.023 0.003 0.010 0.030 0.022
(0.081) (0.001) (0.245) (0.216) (0.052) (0.152) (0.284) (0.205)
PUBLIC 2 0.028 2 0.025 0.020 0.015 0.108 0.109 0.042 0.040
(2 0.255) (2 0.225) (0.174) (0.130) (1.463) (1.515) (0.365) (0.348)
FOREIGN 0.284 * * * 0.290 * * 0.052 0.055 0.119 * 0.110 2 0.056 2 0.061
(2.676) (2.716) (0.479) (0.495) (1.681) (1.609) (2 0.506) (2 0.552)
INDDUMY 0.000 2 0.013 0.157 0.151 0.501 * * * 0.512 * * * 0.158 0.169
(2 0.001) (2 0.118) (1.401) (1.320) (7.069) (7.445) (1.400) (1.472)
BETA 0.086 0.122 0.234 * * 0.244 * * 0.084 0.021 0.218 * * 0.198 * *
(0.939) (1.198) (2.493) (2.311) (1.355) (0.325) (2.297) (1.879)
SIZE 2 0.121 2 0.100 2 0.093 2 0.094 0.037 0.008 2 0.061 2 0.092
(2 1.321) (2 1.070) (2 0.906) (2 0.877) (0.611) (0.132) (2 0.595) (2 0.856)
YDUM2000 – 2 0.169 – 0.058 – 2 0.036 – 0.031
(2 1.500) (0.477) (2 0.494) (0.259)
YDUM2001 – 2 0.177 – 0.079 – 0.022 – 0.055
(2 1.552) (0.638) (0.307) (0.441)
YDUM2002 – 2 0.083 – 0.118 – 0.008 – 0.050
(2 0.724) (0.934) (0.110) (0.393)
YDUM2003 – – – 0.151 – – – 0.020
(1.233) (0.165)
YDUM2004 – 2 0.179 – 0.080 – 0.024 – 0.053
(2 1.536) (0.635) (0.321) (0.426)
YDUM2005 – 2 0.182 – 0.048 – 0.068 – 0.034
(2 1.586) (0.397) (0.928) (0.284)
YDUM2006 – 2 0.221 * – – – 0.072 – –
(2 1.881) (0.965)
YDUM2007 – 2 0.197 * – 0.226 * – 0.234 * * – 0.272 * *
(2 1.697) (1.885) (3.143) (2.268)
R2 0.106 0.148 0.109 0.147 0.603 0.652 0.091 0.148
Adjusted R 2 0.045 0.032 0.045 0.024 0.575 0.605 0.025 0.025
F-statistics (significance) 1.731 * 1.276 1.694 1.195 22.003 * * * 13.636 * * * 1.382 1.204
(0.098) (0.230) (0.108) (0.288) (0.000) (0.000) (0.212) (0.281)
Notes: Significant at: *10, * *5 and * * *1 percent levels, respectively; value within the parenthesis indicates t-values
firm performance

Regression results
Corporate

(model 1)
governance and

Table AIV.
317
 29,4
SEF

318

(model 2)
Table AV.
Regression results
Model 2 Dependent variable ROA ROE LnMktCap TOBIN’S Q

(Constant) 1.656 1.364 3.685 2.931 5.059 * * * 5.198 * * * 0.024 0.026

(1.410) (1.050) (0.964) (0.694) (9.119) (9.437) (0.630) (0.619)
Independent variables SPONSOR 2 0.027 20.023 20.033 20.023 20.316 * * * 2 0.302 * * * 2 0.046 2 0.037
(2 0.257) (20.218) (20.309) (20.214) (24.890) (2 5.061) (2 0.425) (2 0.342)
GOVT 0.012 20.005 0.127 0.124 20.580 * * * 2 0.593 * * * 0.113 0.122
(0.085) (20.037) (0.904) (0.866) (26.461) (2 7.139) (0.794) (0.858)
INSTIN 2 0.006 20.038 0.123 0.114 20.380 * * * 2 0.401 * * * 0.141 0.138
(2 0.035) (20.214) (0.730) (0.660) (3.399) (2 3.864) (0.828) (0.800)
PUBLIC 2 0.042 20.038 20.007 20.019 0.177 * * 0.188 * * 0.020 0.010
(2 0.371) (20.327) (20.059) (20.157) (2.470) (2.818) (0.167) (0.082)
FOREIGN 0.232 * 0.230 0.148 0.155 20.116 2 0.165 * * 0.070 0.070
(1.668) (1.618) (1.059) (1.078) (21.344) (2 2.060) (0.495) (0.491)
INDDUMY 0.027 0.032 0.254 * 0.258 * 0.567 * * * 0.548 * * * 0.247 * 0.254 *
(0.204) (0.239) (1.853) (1.850) (6.823) (7.151) (1.786) (1.828)
BETA 0.094 0.133 0.251 * * 0.290 * * 0.071 2 0.023 0.231 * * 0.241 * *
(1.008) (1.262) (2.670) (2.714) (1.229) (2 0.388) (2.447) (2.257)
SIZE 2 0.113 20.088 20.066 20.045 0.010 2 0.047 2 0.034 2 0.045
(2 1.179) (20.890) (20.640) (20.417) (0.171) (2 0.845) (2 0.332) (2 0.412)
BORDSIZE 2 0.081 20.098 20.152 20.170 0.004 0.046 2 0.122 2 0.130
(2 0.731) (20.864) (21.362) (21.483) (0.065) (0.724) (2 1.089) (2 1.138)
CHAR_AC 0.059 0.079 20.202 20.199 0.467 * * * 0.537 * * * 2 0.240 2 0.243
(0.358) (0.466) (21.250) (21.193) (4.553) (5.554) (2 1.470) (2 1.453)
CHAR_EC 0.057 0.035 20.033 20.055 20.149 * 2 0.112 2 0.060 2 0.066
(0.421) (0.252) (20.245) (20.402) (21.714) (2 1.386) (2 0.448) (2 0.479)
YDUM2000 2 20.009 2 0.011 2 2 0.077 2 0.012
(20.079) (0.096) (2 1.206) (0.100)
(continued)
Model 2 Dependent variable ROA ROE LnMktCap TOBIN’S Q

YDUM2001 – 20.015 – 0.043 – 2 0.051 – 0.048

(20.131) (0.361) (2 0.771) (0.404)
YDUM2002 – 0.086 – 0.066 – 2 0.019 – 0.023
(0.745) (0.562) (2 0.287) (0.192)
YDUM2003 – 0.177 – 0.064 – – – 2 0.040
(1.503) (0.536) (2 0.338)
YDUM2004 – – – – – 0.014 – –
(0.211)
YDUM2005 – 20.008 – 20.027 – 0.065 – 2 0.016
(20.067) (20.232) (0.998) (2 0.133)
YDUM2006 – 20.056 – 20.117 – 0.091 – 2 0.088
(20.453) (20.933) (1.371) (2 0.701)
YDUM2007 – 20.026 – 0.123 – 0.248 * * * – 0.196
(20.222) (1.009) (3.765) (1.606)
R2 0.113 0.156 0.146 0.187 0.667 0.737 0.132 0.188
Adjusted R 2 0.027 0.014 0.059 0.042 0.634 0.692 0.044 0.043
F-statistics (significance) 1.317 1.102 1.679 * 1.286 20.544 * * * 16.478 * * * 1.495 1.299
(0.224) (0.361) (0.088) (0.213) (0.000) (0.000) (0.144) (0.205)
Notes: Significance at: *10, * *5 and * * *1 percent levels, respectively; value within the parenthesis indicates the significant
firm performance
Corporate
governance and

Table AV.
319