effects of banking sectoral factors on the profitability ...€¦ · specific factors had a...
TRANSCRIPT
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
1
EFFECTS OF BANKING SECTORAL FACTORS ON THE PROFITABILITY OF
COMMERCIAL BANKS IN KENYA
Tobias Olweny
Lecturer Department of Commerce and Economic Studies
JKUAT Kenya
Themba Mamba Shipho
Masters in Banking (Student)
Kenya School of Monetary Studies
Kenya
ABSTRACT
The first objective of this study was to determine and evaluate the effects of bank-specific factors; Capital adequacy,
Asset quality, liquidity, operational cost efficiency and income diversification on the profitability of commercial
banks in Kenya. The second objective was to determine and evaluate the effects of market structure factors; foreign
ownership and market concentration, on the profitability of commercial banks in Kenya. This study adopted an
explanatory approach by using panel data research design to fulfill the above objectives. Annual financial
statements of 38 Kenyan commercial banks from 2002 to 2008 were obtained from the CBK and Banking Survey
2009. The data was analyzed using multiple linear regressions method. The analysis showed that all the bank
specific factors had a statistically significant impact on profitability, while none of the market factors had a
significant impact. Based on the findings the study recommends policies that would encourage revenue
diversification, reduce operational costs, minimize credit risk and encourage banks to minimize their liquidity
holdings. Further research on factors influencing the liquidity of commercials banks in the country could add value
to the profitability of banks and academic literature.
Keywords: Assets Quality, Banking Sectoral Factors, Bank-specific factors
1.1 BACKGROUND TO THE STUDY
The stream of bank failures experienced in the USA during the great depression of the 1940s prompted considerable
attention to bank performance. The attention has grown ever since then (Heffernan, 2005). The recent global
financial crisis of 2007/2009 also demonstrated the importance of bank performance both in national and
international economies and the need to keep it under surveillance at all times. Arun and Turner (2004) argued that
the importance of banks is more pronounced in developing countries because financial markets are usually
underdeveloped, and banks are typically the only major source of finance for the majority of firms and are usually
the main depository of economic savings (Athanasoglou et al, 2006).
There are many aspects of the performance of commercial banks that can be analyzed. This study focuses on the
profitability performance of commercial banks in Kenya. Aburime (2009) observed that the importance of bank
profitability can be appraised at the micro and macro levels of the economy. At the micro level, profit is the essential
prerequisite of a competitive banking institution and the cheapest source of funds. It is not merely a result, but also a
necessity for successful banking in a period of growing competition on financial markets. Hence the basic aim of
every bank management is to maximize profit, as an essential requirement for conducting business.
At the macro level, a sound and profitable banking sector is better able to withstand negative shocks and contribute
to the stability of the financial system. Bank profits provide an important source of equity especially if re-invested
into the business. This should lead to safe banks, and as such high profits could promote financial stability (Flamini
et al, 2009). However, too high profitability is not necessarily good. Garcia-Herrero et al (2007) observed that too
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
2
high profitability could be indicative of market power, especially by large banks. This may hamper financial
intermediation because banks exercising strong market power may offer lower returns on deposit but charge high
interest rates on loans. Too low profitability, in turn, might discourage private agents (depositors and shareholders)
from conducting banking activities thus resulting in banks failing to attract enough capital to operate. Furthermore,
this could imply that only poorly capitalized banks intermediate savings with the corresponding costs for sustainable
economic growth.
The banking environment in Kenya has, for the past decade, undergone many regulatory and financial reforms.
These reforms have brought about many structural changes in the sector and have also encouraged foreign banks to
enter and expand their operations in the country (Kamau, 2009). Kenya‟s financial sector is largely bank-based as
the capital market is still considered narrow and shallow (Ngugi et al, 2006). Banks dominate the financial sector in
Kenya and as such the process of financial intermediation in the country depends heavily on commercial banks
(Kamau, 2009). In fact Oloo (2009) describes the banking sector in Kenya as the bond that holds the country‟s
economy together. Sectors such as the agricultural and manufacturing virtually depend on the banking sector for
their very survival and growth. The performance of the banking industry in the Kenya has improved tremendously
over the last ten years, as only two banks have been put under CBK statutory management during this period
compared to 37 bank-failures between 1986 and 1998 (Mwega, 2009).
The overall profitability of the banking sector in Kenya has improved tremendously over the last 10 years. However
despite the overall good picture a critical analysis indicates that, not all banks are profitable. For example the small
and medium financial institutions which constitute about 57 % of the banking sector posted a combined loss before
tax, of Ksh 0.09 billion in 2009 compared to a profit before tax of Ksh 49.01 billion posted by the big financial
institutions (CBK, 2009). The huge profitability enjoyed by the large banks vis-a-avis the small and a medium bank
indicates that there are some significant factors that influence the profitability of commercial banks. Flamini et al
(2009) and other several studies have shown that bank profitability is influenced by bank-specific factors and
industry specific factors. However, these studies were based on data from other countries and their findings may not
be applied to the local banking sector. Locally, to the researcher‟s knowledge, no studies have been done to
determine the key factors that influence the profitability of commercial banks. The aim of this study then was to
close this gap in knowledge by investigating the factors, within the banking sector that influence the profitability of
commercial banks in Kenya.
1.2 Research Objectives
The general objective of this study was to determine and evaluate the effects of banking sectoral factors on the
profitability of commercial banks in Kenya. Specific objectives derived from the general objective of the study
were as follows;
(i) To determine and evaluate the effect of bank-specific factors on the profitability of commercial banks in
Kenya
(ii) To determine and evaluate the effect of market factors on the profitability of commercial banks in Kenya
2.1 THEORIES AND MODELS OF BANK PROFITABILITY
Studies on the performance of banks started in the late 1980s/early 1990s with the application of two industrial
organizations models: the Market Power (MP) and Efficiency Structure (ES) theories (Athanasoglou et al, 2006).
The balanced portfolio theory has also added greater insight in to the study of bank profitability (Nzongang and
Atemnkeng, 2006).Applied in banking the MP hypothesis posits that the performance of bank is influenced by the
market structure of the industry. There are two distinct approaches within the MP theory; the Structure-Conduct-
Performance (SCP) and the Relative Market Power hypothesis (RMP). According to the SCP approach, the level of
concentration in the banking market gives rise to potential market power by banks, which may raise their
profitability. Banks in more concentrated markets are most likely to make „abnormal profits‟ by their ability to
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
3
lower deposits rates and to charge higher loan rates as a results of collusive (explicit or tacit) or monopolistic
reasons, than firms operating in less concentrated markets, irrespective of their efficiency (Tregenna, 2009). Unlike
the SCP, the RMP hypothesis posits that bank profitability is influenced by market share. It assumes that only large
banks with differentiated products can influence prices and increase profits. They are able to exercise market power
and earn non-competitive profits.
The ES hypothesis, on the other hand posits that banks earn high profits because they are more efficient than others.
There are also two distinct approaches within the ES; the X-efficiency and Scale–efficiency hypothesis. According
to the X-efficiency approach, more efficient firms are more profitable because of their lower costs. Such firms tend
to gain larger market shares, which may manifest in higher levels on market concentration, but without any causal
relationship from concentration to profitability (Athanasoglou et al, 2006). The scale approach emphasizes
economies of scale rather than differences in management or production technology. Larger firms can obtain lower
unit cost and higher profits through economies of scale. This enables large firms to acquire market shares, which
may manifest in higher concentration and then profitability.
The portfolio theory approach is the most relevant and plays an important role in bank performance studies
(Nzongang and Atemnkeng, 2006). According to the Portfolio balance model of asset diversification, the optimum
holding of each asset in a wealth holder‟s portfolio is a function of policy decisions determined by a number of
factors such as the vector of rates of return on all assets held in the portfolio, a vector of risks associated with the
ownership of each financial assets and the size of the portfolio. It implies portfolio diversification and the desired
portfolio composition of commercial banks are results of decisions taken by the bank management. Further, the
ability to obtain maximum profits depends on the feasible set of assets and liabilities determined by the management
and the unit costs incurred by the bank for producing each component of assets (Nzongang and Atemnkeng, 2006).
The above theoretical analysis shows that MP theory assumes bank profitability is a function of external market
factors, while the ES and Portfolio theory largely assume that bank performance is influence by internal efficiencies
and managerial decisions. Several models of the banking firm have been developed to deal with specific aspects of
bank behavior but none is acceptable as descriptive of all bank behavior. Some of these approaches are: univariant
analysis, multiple discriminant analysis, multiple regression analysis, canonical correlations analysis and neural
network method. Olugbenga and Olankunle (1998) noted that a major limitation of the univariant analysis approach
is that it does not recognize the possibility of joint significance of financial ratios, while the canonical correlations
method precludes the explicit calculation of marginal value of independent variables on the dependent variable. Nor
can the significance of individual explanatory factors be ascertained. They noted that multiple regression approaches
correct for these limitations and they produce comparable results to the discriminant analysis method.
Bakar and Tahir (2009) evaluated the performance of the multiple linear regression technique and artificial neural
network techniques with a goal to find a powerful tool in predicting bank performance. Data of thirteen banks in
Malaysia for the period 2001-2006 was used in the study. ROA was used as a measure of bank performance and
seven variables including liquidity, credit risk, cost to income ratio, size, concentration ratio, were used as
independent variables. They note that neural network method outperforms the multiple linear regression method but
it lacks explanation on the parameters used and they concluded that multiple linear regressions, not withstanding its
limitations (i.e. violations of its assumptions), can be used as a simple tool to study the linear relationship between
the dependent variable and independent variables. The method provides significant explanatory variables to bank
performance and explains the effect of the contributing factors in a simple, understood manner. This study adopted
this approach together with the correction analysis to determine the effects of banking sectoral factors on bank
profitability in Kenya.
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
4
2.2 Factors Influencing Bank Profitability
In accordance with the above theories and models, many studies have introduced some useful variables in the profit
function of commercial banks to shed light on key factors that make a difference in bank profits. Such studies are
not without ambiguity especially with regard to the measurement of the variables and the results reported thereafter.
However there is general agreement that bank profitability is a function of internal and external factors. Koch (1995)
observed that the performance differences between banks indicate differences in management philosophy as well as
differences in the market served. Athanasoglou et al, (2006) concurred and argued that profitability is a function of
internal factors that are mainly influenced by a bank's management decisions and policy objectives such as the level
of liquidity, provisioning policy, capital adequacy, expense management and bank size, and the external factors
related to industrial structural factors such as ownership, market concentration and stock market development and
other macroeconomic factors.Though most of the studies on bank profitability are based on developed countries
especially the USA and Europe, a couple of studies focusing on developing countries (Naceur (2003), Flamini et al
(2009), Sufian and Chong (2009)) have also used more or less the same variables to study the determinants of bank
profitably.
To identify the relevant factors influencing commercial bank profitability in Kenya, this study concentrated on bank-
specific factors based on the CAMEL framework and market structural factors; ownership and market concentration.
CAMEL is a widely used framework for evaluating bank performance. The Central Bank of Kenya also uses the
same to evaluate the performance of commercial banks in Kenya. Ownership and Market concentration are chosen
because the ownership structure of banks in Kenya has somewhat changed over last decade. More foreign banks
have expanded their operations in the country thus changing the structure of the banking industry.
2.3 The effect of Bank-specific factors on Bank Profitability
Several studies (Elyor (2009), Uzhegova (2010)) have used CAMEL to examine factors affecting bank profitability
with success. CAMEL stands for Capital adequacy, Asset quality, Management efficiency, Earnings performance
and Liquidity. The system was developed by the US Federal Deposit Insurance Corporation (FDIC) for “early
identification of problems in banks‟ operations” (Uzhegova, 2010). Though some alternative bank performance
evaluation models have been proposed, the CAMEL framework is the most widely used model and it is
recommended by Basle Committee on Bank Supervision and IMF (Baral, 2005).
2.3.1 Capital Adequacy and its effect on Profitability
Capital adequacy refers to the sufficiency of the amount of equity to absorb any shocks that the bank may
experience (Kosmidou, 2009). The capital structure of banks is highly regulated. This is because capital plays a
crucial role in reducing the number of bank failures and losses to depositors when a bank fails as highly leveraged
firms are likely to take excessive risk in order to maximize shareholder value at the expense of finance providers
(Kamau, 2009).
Although there is general agreement that statutory capital requirements are necessary to reduce moral hazard, the
debate is on how much capital is enough. Regulators would like to have higher minimum requirements to reduce
cases of bank failures, whilst bankers in contrast argue that it is expensive and difficult to obtain additional equity
and higher requirements restrict their competitiveness (Koch, 1995). Beckmann (2007) argue that high capital lead
leads to low profits since banks with a high capital ratio are risk-averse, they ignore potential [risky] investment
opportunities and, as a result, investors demand a lower return on their capital in exchange for lower risk.
However Gavila et al (2009) argues that, although capital is expensive in terms of expected return, highly
capitalized banks face lower cost of bankruptcy, lower need for external funding especially in emerging economies
where external borrowing is difficult. Thus well capitalized banks should be profitable than lowly capitalized banks.
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
5
Neceur (2003) using a sample of 10 Tunisian banks from 1980 to 2000 and a panel linear regression model, reported
a strong positive impact of capitalization to ROA. Sufian and Chong (2008) also reported the same results after
examining the impact of capital to the performance of banks in Philippines from 1990 to 2005. The banking sector
in Kenya provides an interesting case to examine the impact of capital because the minimum statutory requirement
has been upgraded to Ksh, 1billion in 2012. Capital adequacy is divided into Tier I and Tier II. Tier I capital is
primary capital and Tier II capital is supplementary capital, but this study will focus on total equity of the banks as
opposed to the minimum requirements.
2.3.2 Assets Quality and its effect on Profitability
Credit risk is one of the factors that affect the health of an individual bank. The extent of the credit risk depends on
the quality of assets held by an individual bank. The quality of assets held by a bank depends on exposure to specific
risks, trends in non-performing loans, and the health and profitability of bank borrowers (Baral, 2005). Aburime
(2008) asserts that the profitability of a bank depends on its ability to foresee, avoid and monitor risks, possibly to
cover losses brought about by risks arisen. Hence, in making decisions on the allocation of resources to asset deals, a
bank must take into account the level of risk to the assets.
Poor asset quality and low levels of liquidity are the two major causes of bank failures. Poor asset quality led to
many bank failures in Kenya in the early 1980s. During that period 37 banks collapsed following the banking crises
of 1986-1989, 1993-1994 and 1998 (Mwega, 2009). According to Waweru and Kalani (2009) many of the financial
institutions that collapse in 1986 failed due to non-performing loans (NPLs) and that most of the larger bank-
failures, involved extensive insider lending, often to politicians.The CBK measures asset quality by the ratio of net
non-performing loans to gross loans. However Koch (1995) argues that a good measure of credit risk or asset quality
is the ratio of loan loss reserve to gross loans because it captures the expectation of management with regard to the
performance of loans. Hempel et al (1994) observed that banks with high loan growth often assume more risk as
credit analysis and review procedures are less rigorous, however returns are high in such loans indicating a risk and
return trade-off.
Kosmidou (2008) applied a linear regression model on Greece 23 commercial banks data for 1990 to 2002, using
ROA and the ratio of loan loss reserve to gross loans to proxy profitability and asset quality respectively. The results
showed a negative significant impact of asset quality to bank profitability. This was in line with the theory that
increased exposure to credit risk is normally associated with decreased firm profitability. Indicating that banks
would improve profitability by improving screening and monitoring of credit risk.
2.3.3 Liquidity Management and its effect on Profitability
Another important decision that the managers of commercial banks take refers to the liquidity management and
specifically to the measurement of their needs related to the process of deposits and loans. The importance of
liquidity goes beyond the individual bank as a liquidity shortfall at an individual bank can have systemic
repercussions (CBK, 2009). It is argued that when banks hold high liquidity, they do so at the opportunity cost of
some investment, which could generate high returns (Kamau, 2009). The trade-offs that generally exist between
return and liquidity risk are demonstrated by observing that a shift from short term securities to long term securities
or loans raises a bank‟s return but also increases its liquidity risks and the inverse in is true. Thus a high liquidity
ratio indicates a less risky and less profitable bank (Hempel et al, 1994). Thus management is faced with the
dilemma of liquidity and profitability. Myers and Rajan (1998) emphasized the adverse effect of increased liquidity
for financial Institutions stating that, “although more liquid assets increase the ability to raise cash on short-notice,
they also reduce management‟s ability to commit credibly to an investment strategy that protects investors” which,
finally, can result in reduction of the “firm‟s capacity to raise external finance” in some cases (Uzhegova, 2010).
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
6
In Kenya the statutory minimum liquidity requirement is 20%. However, according to CBK Bank Supervision
Annual Report (2009), the average liquidity ratio for the sector was 39.8% in 2009, 37.0 % in 2008, and way above
the minimum requirements. This has baffled many financial analysts as to how could banks withhold such amount
of cash in a credit needy economy such as Kenya (Kamau, 2009). The CBK attributes this to the banking industry‟s
preference to invest in the less risky government securities, while Ndung‟u and Ngugi (2000) as cited by Kamau
(2009) attributes this liquidity problem to the restrictions placed on commercial banks at the discount window,
coupled with thin interbank market, a high reserve requirement and preference of government securities. Thus given
the above foregoing analysis, the given Kenyan banking sector provides an interesting case to assess the effects of
liquidity on profitability.
2.3.4 Operational Costs Efficiency and its effect on Profitability
Poor expenses management is the main contributors to poor profitability (Sufian and Chong 2008). In the literature
on bank performance, operational expense efficiency is usually used to assess managerial efficiency in banks.
Mathuva (2009) observed that the CIR of local banks is high when compared to other countries and thus there is
need for local banks to reduce their operational costs to be competitive globally. Beck and Fuchs (2004) examined
the various factors that contribute to high interests spread in Kenyan banks. Overheads were found to be one of the
most important components of the high interests rate spreads. An analysis of the overheads showed that they were
driven by staff wage costs which were comparatively higher than other banks in the SSA countries.
Although the relationship between expenditure and profits appears straightforward implying that higher expenses
mean lower profits and the opposite, this may not always be the case. The reason is that higher amounts of expenses
may be associated with higher volume of banking activities and therefore higher revenues. In relatively
uncompetitive markets where banks enjoy market power, costs are passed on to customers; hence there would be a
positive correlation between overheads costs and profitability (Flamini et al, 2009). Neceur (2003) found a positive
and significant impact of overheads costs to profitability indicating that such cost are passed on to depositors and
lenders in terms of lower deposits rates/ or higher lending rates.
2.3.5 Diversification of Income and its effect on Profitability
Financial institutions in recent years have increasingly been generating income from “off-balance sheet” business
and fee income. Albertazzi and Gambacorta (2006) as cited by Uzhegova (2010) noted that the decline in interest
margins, has forced banks to explore alternative sources of revenues, leading to diversification into trading
activities, other services and non-traditional financial operations. The concept of revenue diversifications follows
the concept of portfolio theory which states that individuals can reduce firm-specific risk by diversifying their
portfolios. However there is a long history of debates about the benefits and costs of diversification in banking
literature. The proponents of activity diversification or product mix argue that diversification provides a stable and
less volatile income, economies of scope and scale, and the ability to leverage managerial efficiency across products
(Choi and Kotrozo, 2006). Chiorazzo et al (2008) noted that as a result of activity diversification, the economies of
scale and scope caused through the joint production of financial activities leads to increase in the efficiency of
banking organizations. They further argued that product mix reduces total risks because income from non-interest
activities is not correlated or at least perfectly correlated with income from fee based activities and as such
diversification should stabilize operating income and give rise to a more stable stream of profits (Uzhegova, 2010).
The opposite argument to activity diversification is that it leads to increased agency costs, increased organizational
complexity, and the potential for riskier behavior by bank managers. Kotrozo and Choi (2006) mentioned that
activity diversification results in more complex organizations which “makes it more difficult for top management to
monitor the behavior of the other divisions/branches. They further argued that the benefits of economies of
scale/scope exist only to a point. The costs associated with a firm‟s increased complexity may overshadow the
benefits of diversification. As such, the benefits of diversification and performance would resemble an inverted-U in
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
7
which there would be an optimal level of diversification beyond which benefits would begin to decline and may
ultimately become negative
Using annual bank level data of all Philippines commercial banks Sufian and Chong (2008) found a positive
relationship between total non-interest income divided by total assets, a proxy for income diversification and bank
profitability. Uzhegova (2010) using a HH index of interest income, commissions, fee income, trading income, non-
interest income and other operating income found empirical support of the idea that banks involved in
diversification activities expect some benefits. While Kotrozo and Choi 2006, using a similar index found that
activity diversification tends to reduce performance compared to banks more focused in their activities.
2.4 The Effects of Market Structural Factors on bank profitability
2.4.1 Ownership and its Effects on Profitability
Claessens and Jansen (2000) as cited by Kamau (2009) argued that foreign banks usually bring with them better
know-how and technical capacity, which then spills over to the rest of the banking system. They impose competitive
pressure on domestic banks, thus increasing efficiency of financial intermediation and they provide more stability to
the financial system because they are able to draw on liquidity resources from their parents banks and provide access
to international markets. Beck and Fuchs (2004) argued that foreign-owned banks are more profitable than their
domestic counterparts in developing countries and less profitable than domestic banks in industrial countries,
perhaps due to benefits derived from tax breaks, technological efficiencies and other preferential treatments.
However domestic banks are likely to gain from information advantage they have about the local market compared
to foreign banks.
However the counter argument is that unrestricted entry of foreign banks may result in their assuming a dominant
position by driving out less efficient or less resourceful domestic banks because more depositors may have faith in
big international banks than in small domestic banks. They cream-skim the local market by serving only the higher
end of the market, they lack commitment and bring unhealthy competition, and they are responsible for capital flight
from less developed countries in times of external crisis.(Bhattachrya,1994)
The ownership structure of banks in Kenya has changed over the last few years. For example according CBK Bank
Supervision Annual report of 2000 and 2008, in the year 2000, there were five banks in which government had a
significant ownership, but in 2008, the number had reduced to three banks. During the same period the number of
locally incorporated foreign banks increased from four to eight, while the number of branches of foreign-owned
banks decreased from seven to five. This shows that there is now less state involvement in the industry and more
foreign banks have been allowed to expand their operations in the country. Kamau (2009) used a sample of 40 banks
in Kenya from1997-2006 and linear regression method to analyze factors of X-inefficiencies. The results showed
that an increase in the degree of foreign ownership in Kenya is associated with a reduction of cost X-inefficiencies,
suggesting that the degree of foreign-owned banks influences the performance of the local banking sector.
2.4.2 Market Concentration and its Effect on Profitability
The market power theory, as it was discussed under bank performance theories, posits that the more concentrated the
market, the less the degree of competition (Tregenna, 2009). According to Nzongang and Atemnkeng (2006) high
degrees of market share concentration are inextricably associated with high levels of profits at the detriment of
efficiency and effectiveness of the financial system to due decreased competition. Secondly, since commercial banks
are the primary suppliers of funds to business firm, the availability of bank credit at affordable rates is of crucial
importance for the level of investments of the firms, and consequently, for the health of the economy. In situation of
increased concentration, the possibility of rising costs of credits is reflected by a reduction of the demand for bank
loans and the level of business investments. The effect multiplies many folds in as much as bank management
capitalizes on the market share concentration factor.
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
8
However there is a long held view that market power is necessary to ensure stability in banking. Banks that are
profitable and well-capitalized are best positioned to withstand shocks to their balance sheet. Hence banks with
market power, and the resulting profits, are considered to be more stable Northoctt (2004). Large banks with market
power have typically been viewed as having incentives that minimize their risk-taking behavior and improve the
quality of their assets (the screening theories). Keeley (1990) as cited by Northoctt (2004) argues that the rise in
bank failures in the United States during the 1980s was due in part to an increase in competition in the banking
industry. Flamini et al (2009) noted that if high returns are the consequence of market power, this implies some
degree of inefficiency in the provision of financial services. In this case it should prompt policymakers to introduce
measures to lower risk, remove bank entry barriers if they exist, as well as other obstacles to competition, and
reexamine regulatory costs. But bank profits are also an important source for equity. If bank profits are reinvested,
this should lead to safer banks, and, consequently high profits could promote financial stability.
Tregenna (2009) using a sample of USA commercial banks and savings institutions from 1995 to 2005 and a linear
regression panel model, found robust evidence that concentration increases profitability in USA banks and then
concluded than the high profitability of banks in the USA before the 2007/2008 financial crisis was not earned
through efficient processes, but through market power and the profits were not reinvested to strengthen the capital
base of the financial institutions. Nzongang and Atemnkeng (2000) examined the effects of concentration to the
profitability of Cameroonian commercial banks from 1987 to 1999. Unlike Tregenna (2009), who used the
concentration ratio of the 3 largest banks in the USA to model market concentration, Nzongang and Atemnkeng
(2000) used the Herfindahl-Hirschman index to measure market concentration in Cameroon. The results indicate
that market concentration power is of paramount importance in the determination of bank profitability.
The banking sector in the Kenya looks very competitive judging by number of local and foreign banks in the
industry. CBK Bank Supervision Report (2009) as 31 December 2009 there were 44 commercial banks, 13 of which
are foreign-owned. However Beck and Fuchs (2004) noted that most customers in Kenya below the top tier of
corporate and wealthy borrowers face a non-competitive banking market and are often effectively tied to one bank,
with very high switching costs hence the interest rate spread and margins in the country.
The review of literature has revealed that bank profitability can be influenced by bank-specific factors and external
factors. Bank-specific factors are those factors within the direct control of managers and can be best explained by
the CAMEL framework, while external factors include industry-specific and macroeconomic factors. This study
focuses only on industry-specific factors as external factors. The review of literature also revealed that the multiple
linear regressions method is the most used in modeling the relationship between bank profitability and its factors.
The relevant interrelationships among bank-specific factors and market specific factors and their impact on bank
profitability, as revealed by the reviewed of literature, are depicted in the conceptual framework (Figure 2.1).
Finally, it is clear from the reviewed literature that few local studies have been dedicated on this particular area of
bank performance and that studies that have attempted to do so have tended to study each factor of performance to
the exclusion of other factors.
2.5 Conceptual Framework
The conceptual schema of the relation between the independent variables and dependent variable distilled from the
literature review by the researcher is shown on Figure 2.1 below. It assumes that the relationship between the
independent variable and dependent variables is linear.
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
9
Figure 2.1: Schematic Diagram showing relationships between variables
Independent Variables
Dependent Variable
Affects
3.1 RESEARCH DESIGN
The main objective of this study was to determine and evaluate the effects of banking- sectoral factors on the
profitability of commercial banks in Kenya. This study adopted an explanatory approach by using panel research
design to fulfill the above objective. The advantage of using panel data is that it controls for individual
heterogeneity, less collinearity variables and tracks trends in the data something which simple time-series and cross-
sectional data cannot provide (Baltagi, 2005).
3.2 Target Population
The population of this study comprised of all licensed commercial banks in Kenya between the period of 2002 and
2008. As at 31 December 2008, there were 43 registered commercial banks comprising of 14 large banks and, 29
small and medium banks (Appendix 1).
3.3 Sample Design
All the banks were considered for this study. However, commercial banks which discontinued or started their
operation in the middle of the period under review were not considered. As results, out of the 43 commercial banks,
38 (88%) banks formed the sample of this study. The 38 banks compromised of 13 large banks and 25 small and
medium banks.
3.4 Data Collection
The study employed secondary data. The data was collected from the Central Bank of Kenya and Banking Survey
2009. The banking Survey is an annual publication that publishes annual financial statement of all banks in Kenya
covering a period 10 years, while the Central Bank of Kenya publishes annually, major financial indicators of the
sector.
Bank-specific Factors
Capital adequacy
Assets Quality
Liquidity management
Operational cost
efficiency
Income diversification
Market Structure Factors
Foreign Ownership Structure
Market Concentration
Bank Profits
ROA
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
10
3.5 Data Analysis
The collected data was analyzed using descriptive statistics, graphs, correlations, multiple linear regression analysis
and inferential statistics.
3.5.1 Descriptive Statistics and Relationship Analysis
Mean values and graphs were used to analyze the general trends of the data from 2002 to 2008 based on the sector
sample (38 banks), large banks sample (13 banks), and small and medium banks sample (25 banks). Scatter plots
and a correlation matrix were used to examine the relationship between the dependent variable and explanatory
variables.
3.5.2 Operationalization of the Study Variables
This section presents the measurements that were used to operationalise the study variables before the application of
the linear multiple regression analysis
Table 3.1: Operationalization of the study variables
Variable Measurement
Profitability Ratio of profit before tax to total assets.
Bank-Specific variables
Capital
Adequacy Ratio of total equity to total assets
Asset
Quality
Ratio of non-performing loans to gross loans. Higher ratio indicates poor asset quality
Liquidity Ratio of liquid assets to total liability deposits.
Operational
Cost
efficiency
Ratio of operating costs (staff wages and administrative expenses) to net operating income (net
interest income, net foreign exchange income, net fees and commission, and other income).
Higher ratio indicates inefficiency
Income
Diversificati
on
1-(HHI of net interest income, foreign exchange income, commissions and fees, and other
income). Index ranges from 1 to 0. Where 1 indicates complete diversification, 0 indicates
complete focus
Market Factors
Foreign
Ownership
Ratio of foreign annual assets held by foreign banks to total annual banking sector assets
Market
Concentrati
on
HH index of the annual deposits of all commercial banks in the market. Index ranges from 10,000
to 0. Indicating an uncompetitive market to a competitive market
3.6.3 Multiple Linear Regression Analysis
A multiple linear regression model and t-statistic were used to determine the relative importance of each
independent variable in influencing profitability. The t-statistic was used to test the two hypotheses at a maximum of
10% significance level. The multiple linear regressions model is shown on equation 1 below. This model was run
using Eviews 5. The analysis was based on the sector sample (38 banks), large banks sample (13), and small and
medium banks (25 banks).
ROAit = Ci+α1CAPit +α2ASQit+α3LIQit+α4CIRit+α5RDIit + β1FGNt +β2MKTt + ei…… (1)
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
11
Where;
ROAit = Profitability of bank i at time t
CAPit = Capital adequacy of bank i at time t
ASQit = Asset quality of bank i at time t
LIQit = Liquidity of bank i at time t
CIRit = Operational cost efficiency of bank i at time t
RDIit = Income diversification of bank i at time t
FGNt = Market foreign ownership structure at time t
MKTt = Market Concentration at time t
Where t = 2002….2008, Ci = constant for each bank (fixed effects), α= bank specific factors coefficients, β= Market
factors coefficients.
The above model can three forms: pooled model, fixed effects model and random effects model. Pooled model
assumes homogeneity in the study units, while the other two assume heterogeneity in the study units. Kennedy
(1998) and (Baltagi (2005) argued that the fixed effects model is suitable if the data exhaust the population, the
study is focusing on a specific set of N firms and the inference is restricted to the behavior of these sets of firms. The
foregoing argument suggested that the fixed effects model would be suitable for this particular study. To test for
suitability of the fixed effects model, the F-statistic was used (Gajarati, 2007). The null hypothesis of the F-statistic
is that the study units are homogeneous and as such the pooled model is better, while the alternative is that the study
units are heterogeneous and therefore they cannot be pooled. The F-statistic is given as follows;
Where RRSS is the restricted sum of residual squares (pooled model) and URSS is the unrestricted sum of the
squares (Fixed effects model). N is the number of cross-section, T is the number of time periods and K is the number
of parameters to be estimated. The null hypothesis is accepted when the test statistic is less than the appropriate
critical value. Rejection of the null hypothesis leads to the acceptance of the fixed effects model.
The pooled model was estimated and a RRSS of 1146.206 was obtained. The fixed effects model was estimated to
get URSS of 842.31. Applying the above formula, an F statistic of 2.145 with 37 and 220 degrees of freedom was
obtained. The F-statistic critical value of 37 and 220 degrees at 1% is 1.710 and as such the null hypothesis of
homogeneity was rejected at 1% significant level hence the fixed effects model was used. However in the sample of
large banks, and small and medium banks sample the pooled model was used.
3.6.3 Model Assumptions and Data properties
The following diagnostic tests were carried out to ensure that the data fits the basic assumptions of linear regression
models;
Normality: Descriptive statistics were taken to examine the distribution of data. Upon examination the Skewness
and Kurtosis of the data it was clear that most of the variables were close to normal distribution.
Multicollinearity: Schindler and Cooper (2009) suggested that a correlation above 0.8 between explanatory
variables should be corrected for. To ensure that none of the explanatory variables were highly correlated to each
other, a correlation matrix was used and none of the variables were highly correlated to each other. After all, one
advantage of panel data models is the ability to control for multicollinearity.
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
12
Heteroscedasticity: Finally, The model was estimated in Eviews assuming cross-section heteroscedasticity (Whites‟
cross-section weights) to control for the possible effects heteroscedasticity in the error variance (Gujarati, 2007)
4.1 FINDINGS AND DISCUSSIONS
4.2 Trend Analysis of Profitability and Banking-Sectoral Factors
This section of the study aimed at establishing the general trend of profitability and the seven banking-sectoral
factors in the Kenyan banking sector from 2002 to 2008.
4.2.1 Trend Analysis of Profitability
Table 4.1 reports the mean scores of ROA from 2002 to 2008. The mean of score of ROA for the whole sector was
1.4% and rose to 2.4% in 2008 showing an increase of 71.4%. For large banks ROA was 1.5% in 2002 and rose to
4% showing an increase of 166.7%. ROA for small and medium was 1.4% in 2002 and rose to 1.8% by only 28%.
Table 4.1: The Annual Mean Scores of Profitability from 2002 to 2008
Variable Category 2002 2003 2004 2005
2006 2007 2008
% ∆ since
2002
Sector (%) 1.4 1.5 1.3 1.7
2.2 2.8 2.4 71.4
ROA Large banks (%) 1.5 2.5 2.3 2.9
3.6 3.8 4 166.7
Small & medium (%) 1.4 1.1 0.8 1.2
1.6 2.4 1.8 28.6
Source: Research Data, 2010
The reported results in table 4.1 mean that the profitability of the sector increased from 2002 to 2008. In the banking
industry, ROA of more than 1.5% indicates good performance (Flamini et al, 2009). Therefore this means the
performance of the sector was comparable to international standards. This is very important for the development of
this country as banks play a very important role of financial intermediation. However analysis by bank size indicates
that, large banks enjoyed more profit increase than small and medium banks during this period. From 2002 to 2008
the average profitability of the large banks increased by 166.8%, while for small and medium banks increased by
only 28.6%. This lends support to the argument that the local banking market is largely dominated by larger banks.
4.2.2 Trend Analysis of Bank-specific Factors
This section analyses the average performance of the banking sector in terms of the five bank-specific factors
between 2002 and 2008 and the mean scores are reported in 4.2. The mean value of CAP for the whole sector from
2002 to 2008 was 18%, for large banks was 12.23% and for small and medium banks was 20.66%. The mean score
of ASQ for the sector was 16.43%, for the large banks 12.12% and for the small and medium banks was 18.19%.
The average mean value of LIQ, a proxy for liquidity, was 43.08% for the sector, for large banks it as 41.07% and
for small and medium banks it was 44.97%. CIR which represents operational costs was 65.84% for the sector,
57.66% for large banks and 69.17% for small and medium banks. Lastly the mean score of RDI, which measures the
ability of banks to generate revenue from different sources, was 0.48 for the whole sector, 0.53 for large banks and
0.46 for small and medium banks.
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
13
Table 4.2 Aggregate Mean Scores of Bank-specific Factors
Variables
Banking Sector
(%)
Large
banks
(%)
Small & Medium
banks
(%)
CAP 18.22 12.23 20.66
ASQ 16.43 12.12 18.19
LIQ 43.08 41.07 44.97
CIR 65.84 57.66 69.17
RDI 0.48 0.53 0.46
Source: Research Data, 2010
The capital adequacy results suggest that about 18% of the total assets of the sector were financed by shareholders‟
funds while the remaining 82% was financed by deposit liabilities. The high leverage is not surprising because the
business of banking is to mobilize more deposits from customers. The CBK stipulates that banks must keep core
capital of not less than 8% of total deposit. This implies that Kenyan banks on average operated above minimum
statutory levels. However an interesting observation is that small and medium banks seem to use more shareholders
funds to finance their assets that large banks as the mean of CAP for small banks is (20.66%) higher that the mean
score of CAP for large banks (12.23%). One possible reason is that the fixed minimum capital requirement of Ksh,
350 million is very high for small and medium banks relative to their growth, while it is low for large banks.
The mean ratio of assets quality (ASQ) indicates that small and medium banks had a poor loan book than large
banks as the mean value of ASQ for large banks (12.2%) was less than the mean score of small banks (18.19%).This
is not surprising because most small and medium banks do not have the capacity to invest in stringent credit risk
management practices compared to big banks. The mean score of liquidity (LIQ) shows that the sector was very
liquid, two times more than the minimum statutory liquidity ratio of 20% set by CBK. The higher liquidity ratio
indicates that banks in the country prefer to invest in safe, short-term investments than credit loans. The average
ratio of cost to income (CIR) was 65.84% an indicator that overheads are high in the local banking sector. It is even
worse for small and medium banks as the mean was 69.17% against 57.66% for large banks. Lastly the income
diversification index indicates that the revenue income of local banks was poorly diversified as the average was
0.48, with the income of large banks more and better diversified than for small and medium banks.
The foregoing analysis shows that the profitability of the sector improved during the period under review, but large
banks were dominant. Furthermore, the performance of small and medium banks in terms of asset quality,
operational cost efficiency and income diversification was poor compared to the large banks, while in terms capital
adequacy and liquidity they were comparable to the larger banks.
4.1.3 Trend Analysis of Market Factors
In this section, the study sought to analyze the trend of foreign ownership and market concentration in the banking
sector from 2002 to 2008. The degree of foreign ownership is given by the percentage ratio of total assets held by
foreign-owned banks to the total assets of the banking sector in each year, while market concentration is given by
the HHI index using total deposit for each bank in each year in figure 4.1 below.
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
14
Figure 4.1: Analysis of Market Concentration and Foreign Ownership
Source: Research Data, 2010
The Figure above shows that foreign-owned banks, though small in number (about 27%) controlled about 50% to
40% of total assets in the sector between 2002 and 2008. This demonstrates the influence of foreign banks in the
sector and seems supports the counter argument that unrestricted entry of foreign banks may result in their assuming
a dominant position by driving out less efficient or less resourceful domestic banks because more depositors may
have faith in big international banks than in small domestic banks.
The HHI index, on the other hand indicates that the market structure is moving from high concentration to low
concentration. An index above 1800 represents a highly concentrated industry, which indicates the presence of
oligopoly (Kamau, 2009). Therefore the concentration of the local banking market is exhibiting a loose oligopoly. A
highly concentrated market results in market power and less competitive strategies which lead to high interest
margins.
4.3 The Relationship between Profitability and Banking-Sectoral factors
It is always important for a researcher to assess the general relationship between two variables before subjecting
them to a linear regression analysis to ascertain whether they are linearly related or not. This section therefore aimed
at establishing the relationship between the profitability of commercial banks and the seven explanatory variables.
4.3.1 The Relationship between Profitability and Capital adequacy
The results presented in figure 4.2 indicate that the capital ratio (CAP) is positively related to return on assets
(ROA), the profitability measure. The coefficient of correlations is 0.176 which indicates that the relationship may
not be very strong. However it is clear that the weak positive relationship is due to the two extreme banks, Eco Bank
and Oriental Bank which had relatively sufficient capital levels but posted poor profitability results. These results
provide reasonable evidence to the consistent view that, the higher the capital levels, the higher the profitability.
Generally a bank that depends more on leverage will experience more volatile earnings and this also affects the
credit creation and liquidity function of the bank.
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
15
Figure 4.2: The Relationship Analysis, Profitability and Capital Adequacy
Source: Research Data, 2010
4.3.2 The Relationship between Profitability and Assets Quality
Figure 4.3 presents the relationship between assets quality or credit risk and profitability. It is clear from the this
figure that there is a negative and strong relationship between poor assest quality and profitability as the plots are
clustered strongly aroung the trend and the coefficient of correlation is -0.71. This means banks which fail to
monitor their credit loans tend to be less profitable than those which pay particular attention to assets quality.
Again, as it was observed under desctriptive statistics, the small and medium banks (Oriental, Eco bank, City Finace
bank) that had the highest ratio of non-performing loans to gross loans are associated with low profitability. This is
inline with the theory that increased exposure to credit risk is normally associated with decreased bank profitability
(Kosmidou, 2008).
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
16
Figure 4.3: Relationship Analysis, Profitability and Asset Quality
Source: Research Data, 2010
4.3.3 The Relationship between Profitability and Liquidity
In the literature review, the divergent views regarding the relationship and the effect of liquidity on profitability was
explored. Furthermore, the descriptive analysis in section 4.1 above showed that local banks prefer to invest in short
term liquid assets as demonstrated by the high liquidity ratios. Figure 4.4 shows a correlation coefficient of 0.176
between profitability and liquidity, indicating a positive correlation between the two variables.
With the exception of Eco bank, Oriental and City Finance, there is evidence that liquid banks are associated with
better profitability. Notably, National bank of Kenya was the least liquid bank, Habib Zurich and Habib bank Ltd
were highly liquid. These findings seem to be against the argument that liquidity has a negative effect on
performance (Kamau,2009), but they seem to support the counter-argument that illiquidity force banks to borrow
from the money market expensive funds, or to prematurely liquidate their long-term investments at „fire prices‟ to
cover their immediate cash needs, thus reducing their profitability (Elyor,2009). However the, such results need to
be read with caution given the relatively weak coefficient of correlation.
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
17
Figure 4.4: Relation Analysis, Profitability and Liquidity
Source: Research Data, 2010
4.3.4 The Relationship between Profitability and Operational Costs Efficiency
The nature of the relationship that exists between operating costs and profitability is presented in Figure 4.5. The
coefficient of correlation(r) of -0.76, suggests a strong negative correlation between profitability and Operational
costs. These findings are not surprising, as the issue of high operative costs was covered extensively in the literature
review and the descritive analysis showed that operating costs are higher in the sector. For example Oriental Bank
had a high ratio of operating cost to income and as a results made an aggregate loss of about -6%, whilst Standard
Charterd was amongst the lowest and made a profit of about 4%.
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
18
Figure 4.5: Relationship Analysis, Profitability and Operational Costs efficiency
Souce: Research Data, 2010
However an important findings is that the local market seems to be competitive. In relatively uncompetitive markets
where banks enjoy market power, costs are passed on to customers; hence there would be a positive correlation
between operating costs and profitability (Flamini et al, 2009).
4.3.5 The Relationship between Profitability and Income diversification
The descriptive analysis in section 4.2 showed that revenue diversification in the sector is average, with large banks
showing a higher diversification index that small and medium banks. Figure 4.5 displays the relationship between
proftability and diversification of income and the coefficient of correlation is 0.26. indicating that the more banks
generate their revenue from different activities, the more profitable they become.
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
19
Figure 4.6: Relationship Analysis, Profitability and Income Diversification
Source: Research Data, 2010
As discussed above, large banks seem to be well diversifed in terms of their income than small banks. The same
pattern is being observed in Figure 4.5 above. The Bank of Baroda, and Oriental is the least diversifed, while most
of the big banks ( KCB, Barclays, Citibank, and Standard chartered bank) appear to be the most divesfied in terms of
revenue. This relationhip supports the argument that product mix reduces total risks because income from non-
interest activities is not correlated or at least perfectly correlated with income from fee based activities and as such
diversification stabilizes operating income and gives rise to a more stable stream of profits
4.3.6 The Relationship between of Profitability and Market factors
The aim of this section was to establish the relationship between the market factors; degree of foreign onwership,
market concetration and profitability and the results are reported in table 4.3. The coefficient of correlation between
proftability (ROA) and the degree of foreign ownership (FGN) is -0.116, while the coefficinet of correlation
between markert concetration (MKT) and profitability is -0.128.
Table 4.3: Correlation matrix of Profitability and Market Factors
ROA FGN MKT
ROA 1.0000 -0.116 -0.128
FGN -0.116 1.0000 0.800
MKT -0.128 0.800 1.0000
Source: Research Data, 2010
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
20
This means that bank foreign ownership is negatively associated with profitabilty which does not supports the
argument that foreign banks usually bring with them better know-how and technical capacity, which then spills over
to the rest of the banking system. The negative correlations of market concetration and profitability on the otherhand
is also not in support of the market power hypothesis. In both cases the coefficients of correlation is weak
suggesting a weak relationship.
4.3 Regression Results for the Effects of Banking-Sectoral Factors on Profitability
The above relationship analysis has shown that all the variables are somehow related to profitability. The aim of this
section is to explore in detail the above relationships by using regression analysis which is more robust that the
scatter plot analysis. The regression results are reported as follows; Section 4.3.1 reports the summary statistics of
the regression model, Section 4.3.2 and 4.3.3 reports the regression results in terms of the specific objectives of the
study. The detailed Eviews results are found in appendix 2.
4.3.1 Summary Statistics of the Regression Model
Table 4.4 reports the summary statistics of the regression model. The second column gives summary statistics of the
regression model based on the full sample. The R2 of the sector sample regression was 0.948, F-statistic was 90.79
with p-value of 0 and the DW statistic was 1.877.
Table 4.4: Summary Statisitics of Regression Model
Statistic Sector Large banks Small & Medium banks
R-squared 0.948 0.79 0.71
Adjusted R-squared 0.937 0.77 0.70
F-statistic 90.79 45.34 59.59
Probability(F-stat) 0.000 0.000 0.000
DW statistic 1.877 1.12 1.39
Source: Research Data, 2010
The R2 is a measure of the goodness of fit of the banking-sectoral factors variables in explaining the variations in
bank profitability. This means the variables jointly explain about 95% of the variation in the profitability of banks.
Thus these variables collectively, are good explanatory variables of the profitability of commercial banks in Kenya.
The null hypothesis of F-statistic (the overall test of significance) that the R2 is equal to zero was rejected at 1% as
the p-value was sufficiently low. Secondly the D.W. statistic was about 1.88 implying that there was no serious
evidence of serial correlation in the data
4.3.2 Regression Results for the Effects of Bank-Specific factors on Profitability
The first objective of the study was to determine and evaluate the effects of bank specific factors on profitability.
These effects were investigated by testing the hypothesis that;
‘Bank-specific factors affect the profitability of commercial banks significantly in Kenya’
The multiple linear regression and t-statistic results used to test this hypothesis are reported in Table 4.5. The
coefficient of CAP is 0.076 with a t-statistic of 5.464 in the main sample, 0.034 and t-statistic of 1.840 in the large
banks sample and 0.054 and t-statistic of 4.672 in the sample of small and medium banks. The positive coefficients
mean an increase in capital leads to an increase in profitability and the high t-statistic value indicates that the impact
is statistically significant at 1 % test level.
ASQ has a negative beta of -0.048 with a t-statistic of -5.087 in the main sample, coefficient of -0.028 and t-statistic
of -2.877 in the sample of large banks, coefficient of -0.056 and t-statistic of -6.24 in the sample of small and
medium banks. This means poor asset quality leads to lower profitability to all banks. This negative impact is
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
21
significant at 1% test level. The effect of Liquidity (LIQ) to ROA is 0.009 with t-value of 2.095 in the main sample,
0.003 (0.573) in the sample of large banks and 0.010 (1.734) in the sample of small and medium banks. This means
an increase in liquidity leads to an increase in profitability. This impact is significant at least, at 10% test level in all
samples. However the coefficient is weak, implying a weak positive impact.
The results for CIR are as follows; in the main sample the impact is -0.068(-16.972), in the sample of large banks is
-0.075 (-10.14) and in the sample of small and medium banks the impact is -0.063 (-10.348). This means operational
costs inefficiency leads to poor profitability. The effect is more on large banks than in small and medium banks and
it is significant at 1% test level in all samples
Table 4.5: Regression Results for the effects of Bank-Specific factors on Profitability
Sector Large banks
Small
& Medium banks
Variables Coefficient Coefficient Coefficient
C
4.935***
(7.017)
6.187***
(4.547)
3.619***
(2.688
CAP
0.076***
(5.464)
0.034*
(1.840)
0.054***
(4.672)
ASQ
-0.048***
(-5.087)
-0.028***
(-2.877)
-0.056***
(-6.269)
LIQ
0.009**
(2.095)
0.003
(0.573)
0.010*
(1.734)
CIR
-0.068***
(-16.972)
-0.075***
(-10.14)
-0.063***
(-10.348)
RDI
0.017**
(2.456)
0.026***
(3.249)
0.028***
(2.506) ***
Significant at 1%; **
Significant at 5%; *Significant at10%; t-statistic in brackets
Source: Research Data, 2010
Finally, the impact of RDI is 0.017 (2.456) in the main sample, 0.026 (3.249) in the sample of small and medium
banks and 0.026 (2.506) in the sample of large banks. This means income diversification or product mix leads to
increased profitability. This impact is statistically significant at least, at 5% test level. Clearly the above results
indicated that all the bank-specific factors had a significant impact on the profitability of banks during the period
understudy at least, at 10% test level. This means that Bank-specific factors affect the profitability of commercial
banks significantly.
4.3.3 Regression Results for the Effects of Market Factors on Profitability
The second objective of the study was to determine and evaluate the effects of market factors on profitability.
Market factors do not significantly affect the profitability of commercial banks significantly in Kenya
Table 4.6 reports the results for the effects of market factors on profitability. The impact of foreign ownership
(FGN) is 0.004, with t-statistic of 0.215 in the main sample, -0.035 and t-statistic of -0.842 in the sample of large
banks, and 0.046 with a t-statistic of 1.186 in the sample of small and medium banks. The effect is positive in the
sample of small and medium banks, but negative in the sample of large banks and it is statistically insignificant in all
samples.
With regard to market concentration, the effect is -0.001, t-statistic of -0.708 in the main sample, clearly
insignificant, weak coefficient and not in support of the SCP hypothesis. In the sample of large banks, it has a weak
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
22
positive coefficient of 0.001, and statistically insignificant. And in the sample of small and medium banks it is
negative and insignificant as well.
Table 4.6: Regression Results for the Effect of Market factors on Profitability
Sector Large banks
Small
& Medium banks
Variables Coefficient Coefficient Coefficient
FGN
0.004
(0.215)
-0.035
(-0.842)
0.046
(1.186)
MKT
-0.001
(-0.708)
0.001
(0.535)
-0.002
(-0.962) ***
Significant at 1%; **
Significant at 5%; *Significant at 10%; t-statistic in brackets
From the above results it is evident that market factors had little effect on the profitability of banks during this
period. The low t-statistics and high p-values (i.e. not significant at least at 10% test level) of both variables indicate
that the null hypothesis of the t-value; that the true population coefficients are equal to zero is not rejected
4.4 Discussion of Results
The scatter plot analysis and multiple regression analysis have shown that bank-specific factors are not only related
to the profitability of banks, but they also influence the profitability of commercial banks in Kenya significantly.
Elyor (2009 argued that well capitalized banks have a stronger revenue generating capacity and can collect more
deposits. The analysis revealed that capital adequacy is the most robust and important factor influencing profitability
in the sector. The results showed that a 1% increase in capital adequacy could result in 0.076% increase in
profitability. This was statistically significant at 1% (5.464) confidence level. The same statistically significant and
positive impact was found in the sample of small and medium banks, and large banks. Similar results were also
found by Neceur (2003) when evaluating the determinants of bank profitability in Tunisia. Suffian and Chong
(2008) also reported the same results after examining the impact of capital on the performance of banks in
Philippines.
This result means banks should focus on improving their capital levels in order to improve their profitability. This
will enable the banks, not only to be cushioned against exogenous shocks, but also to take full advantage of business
opportunities as they come and increase their profitability in process. Thus this finding provides support to the
argument that well capitalized banks face lower cost of bankruptcy and lower need for external funding especially in
emerging economies where external borrowing is difficult and costly. It also provides evidence that supports the
CBK‟s move to gradually increase capital levels by 2012.
Operational costs efficiency was also found to be the next critical factor influencing profitability. The study found
that a 1% increase in operational costs could results in 0.068% decrease in profitability and this finding was
statistically significant at 1% (-16.972) level. Flamini et al (2009) and Neceur (2003) also found the same results for
SSA and Tunisian banks respectively. The importance of efficient overhead management cannot be over emphasized
in this study. The descriptive analysis of this factor showed that operating expenses are as high as 65.84% of
operating income on average in the sector, 69.17% small and medium banks and 57.66% large banks. It is therefore
obvious that a lot needs to be done to reduce staff wage costs and administrative costs in the sector to improve
profitability. The strong negative impact of CIR indicates that banks are not able to pass all their operating cost to
customers which may be an indicator of the competiveness and lack of market power in the sector.
Asset quality showed a negative effect of -0.048, statistically significant 1% level, meaning a 1% increase in the
asset quality ratio (indicating deteriorating asset quality), could lead to 0.048 % reduction in profitability. The effect
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
23
was the same in the sample of small and medium banks, and large banks. These results are consistent with previous
findings by Kosmidou (2008) and Flemini et al (2009). This means local banks need to improve their processes of
screening credit customers and monitoring of credit risk especially the small and medium sector as the descriptive
analysis showed that small and medium banks have a poor loan book (18.19%) compared to large banks
(12.12%).This is an important indicator because local banks have had serious problem with non-performing in the
past which led to collapse of many banks. Again these results provide support for the CBKs move to establish the
credit bureau reference as this is expected to go a long way in helping banks reduce the rate of bad loans in the
industry and thus improving profitability.
Another important finding after assets quality is income diversification. This factor had a positive effect of 0.017
(2.456), statistically significant at 5% confidence level. It was statistically significant in large banks and in the small
and medium banks. Investigating the relative importance of bank specific factors on the profitability of banks
operating in developed, advanced emerging, secondary emerging and frontier markets, Uzhegova (2010) also found
that income diversification leads to increased profitability. This means banks that diversify their source of revenue
between, interest income, fees and commissions, foreign exchange activities and other, are profitable than those that
largely depend on a single source of income. This is line with argument that diversification provides a stable and
less volatile income, economies of scope and scale, and the ability to leverage managerial efficiency across products
(Chiorazzo et al, 2008)
Finally the effect of liquidity was 0.009 (2.095) statistically significant at 5% significance level, indicating that
liquidity positively influences profitability. The implication of this finding is that investing in short-term, less risky
securities like government bonds leads to increased profitability. However, all the coefficients were very weak
implying a weak impact. Thus these results support the risk and return theory. The descriptive statistics analysis
showed that liquidity in the sector is well above statutory limits and that small and medium banks are more liquid
that large banks. This finding suggests that these funds are underutilized. Past studies regarding the effect of
liquidity on profitability are mixed but these findings are consisted with Kosmidou et al (2008) and Ghazali (1999
However to the contrary, market factors do not have any significant influence to the profitability of banks in Kenya.
The impact of foreign ownership in the sector was positive (0.004) but not statistically significant (0.215). The
results were almost the same in all samples indicating that foreign ownership is not a critical factor of profitability in
the sector and as such a public policy to encourage the presence of foreign banks may, therefore, not yield any
advantage in terms of bank profitability. This finding is diametrically against the argument that foreign banks bring
with them better know-how and technical capacity, which then spills over to the rest of the banking system and thus
improve profitability (Jansen, 2000; Kamau, 2009). Flamini et al (2009) obtained similar results and they concluded
that foreign-owned banks face the same local conditions as local banks, with regard to risk and the performance of
the domestic economy.
With regard to market concentration, the main hypothesis underlying this factor is the SCP hypothesis which
postulates that market concentration has a positive impact on the performance of banks, indicating that large banks
are able to exercise market power. The market concentration index showed that the local banking industry is moving
from high concentration to low concentration. The overall regression results showed that market concentration had a
negative effect on profitability. However in the sample of large banks the effect was positive indicating that large
banks may have been able to exercise market power in line with SCP hypothesis. In the sample of small and medium
banks the effect was negative indicating that market concentration was not beneficial to these banks. However in all
samples the coefficients were weak and were statistically not different from zero. Clearly these results failed to
support the Structure-conduct-performance or market power hypothesis. This might mean concentration is less
beneficial in terms of profitability to the Kenyan commercial banks than competition (Kosmidou, 2008)
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
24
4.6 Conclusion
The main objective this study was to determinate and evaluate the effects of banking sectoral-factors on the
profitability of commercial banks in Kenya. Two specific objectives were derived from the main objective. The first
specific objective was to determine and evaluate the effects of bank-specific factors expressed within the CAMEL
framework. The second objective was to determine the effects of market structure factors; foreign ownership and
market concentration. Panel data from 2002 to 2008 of 38 commercial banks was analyzed using multiple linear
regressions method. From the discussion of the findings above, it can be concluded that the bank-specific factors are
the most significant factors influencing the profitability of commercial banks in Kenya than market factors. The
study revealed that profitable commercial banks are those that strive to; improve their capital bases, reduced
operational costs, improve assets quality by reducing the rate of non-performing loans, employ revenue
diversification strategies as opposed to focused strategies and keep the right amount of liquid assets. Indeed the
descriptive analysis of these factors by bank size showed that large banks perform better than the small and medium
banks hence the superior profitability performance. Thus it can be concluded that profitability in the Kenyan
banking sector is largely driven by managerial decision than market factors.
4.7 Future Research
The study sought to investigate factors that influence profitability of commercial banks in Kenya. However the
variables used in the study were not exhaustive. Future research could incorporate macroeconomic variables such as
GDP, inflation and exchange rates. Also a study on the factors influencing the liquidity position of commercial bank
in the country could add great value to the performance of local banks and academic literature.
REFERENCES
Athansasoglou, P., Brissimis, S. & Delis, M. (2006). Bank-Specific, Industry-Specific and Macroeconomic
Determinants of Bank Profitability. Journal of International Financial Markets, Institutions and Money.
[Online] 121-136. Available from: http://ssrn.com/abstract=1106825. [Accessed: 03/06/2010]
Aburime, U. (2008). Determinants of Bank Profitability: Company-Level Evidence from Nigeria. [Online].October
2008. Available from: http://ssrn.com/abstract = 1106825. [Accessed: 10 June 2010]
Abu Bakar, N. & Tahir, I. M. (2009). Applying Multiple Linear Regression and Neural Network to Predict Bank
Performance. International Business Research. [Online]. Vol.2 No.4 . Available from:
www.ccsenet.org/journal.html. [Accessed: 20/06/2010].
Bobáková, V. (2003). Raising the Profitability of Commercial Banks. XI 4/2003. [Online]. Available from:
www.nbs.sk/img/documents/biatec/bia0403/2125.pdf. [Accessed: 9 June 2010]
Baral, K. J. (2005). Health Check-up of Commercial Banks in the Framework of CAMEL: A Case Study of Joint
Venture Banks in Nepal. The Journal of Nepalese Business Studies.[Online] Vol II No.1. Available
from:http://www.nepjol.info/ index.php/JNBS/ article/viewFile/55/483. [Accessed: 20/06/2010]
Berger, A. (1995). The Profit-Structure Relationship in Banking-Test of Market Power and Efficient-Structure
Hypothesis. The Journal of Money, Credit and Banking. [Online] 27 (2). Available from:
http://www.jstor.org/stable/2077876. [Acessed: 26/06/2010]
Beck, T. & Fuchs, M. (2004). Structural Issues in the Kenyan Financial System: Improving Competition and Access.
W/P 3363. [Online]. July 2004. Available from: http://siteresources.worldbank.org. [Accessed: 20 June 2010]
Baltagi, B.H. (2005). Econometric Analysis of Panel Data. John Wiley & Sons Publish. Chichester.
Central Bank of Kenya (2009). Bank Supervision Annual Report 2009. Central Bank of Kenya, Nairobi.
Cooper, D. C., & Schindler, P. S. (2009). Business Research Methods. 9th
edn.Tata McGraw-Hill. New Delhi
Choi, S. & Kotrozo, J. (2006). Diversification, Bank Risk and Performance: A Cross-country Comparison.
[Online]. October 2006. Available from: http://ssrn.com/abstract=1013430. [Accessed: 15 August 2010]
Dougherty, C. (2007). Introduction to Economics. 3rd
edn. Oxford University Press Inc. New York
Elyor, S. (2009). Factors Affecting the Performance of Foreign Banks in Malaysia. A Thesis Submitted in Partial
Fulfillment of the Requirements of Universiti Utara Malaysia for the Degree of Master of Science (Banking).
[Online]. October 2009. Available from:www.ep3.uum.edu.my/1760/1/ Saidov_Elyor_Ilhomovich.pdf
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
25
Flamini, V., McDonald, C. & Schumacher, L. (2009). The Determinants of Commercial Bank Profitability in Sub-
Saharan Africa. WP/09/15. [Online]. January 2009. Available from:
www.imf.org/external/pubs/ft/wp/2009/wp0915.pdf. [Accessed: 26 May 2010]
Gavila, S. & Santabarbara, D. (2009). What Explains the low Profitability in Chinese Banks.W/P30. [Online]
March2006. Available from: http://ssrn.com/abstract= 1413123 [Accessed: 24 May 2010]
Ghazali, M. (2008). The Bank-Specific and Macroeconomic Determinants of Islamic Bank Profitability: Some
International Evidence. [Online] November 2008. Available from:
http://dspace.fsktm.um.edu.my/bitstream/1812/752/1/abstract.pdf [Accessed: 23 May 2010]
Guranti, D. N. (2007). Basic Econometrics. McGraw-hill. New York
Heffernan, S. (2005). Modern Banking. John Wiley & Sons. London
Hempel, G. H., Simonson, D. G. & Coleman, A.B. (1994). Bank Management. 4th
edn. John Wiley & Sons. New
York
Kosmidou, K. (2008). The determinants of banks „profits in Greece during the period of
EU Financial integration. Journal of Managerial Finance. [Online] 34 (3). Available from:
http://www.emeraldinsight.com. [Accessed: 05/06/2010]
Kamau, A.W. (2009). Efficiency in the Banking Sector: An Empirical Investigation of Commercial Banks in Kenya.
A thesis submitted in partial fulfillment of the Requirements of Nairobi University for the Degree of Doctor of
Philosophy. Nairobi: University of Nairobi
King, G.R. & Levine, R. (1993). Finance and Growth: Schupter Might be Right. The Quarterly Journal of
Economics. [Online]. 108 (3). pp. 717-737. Available from: www.jstor.org [Accessed: 01/06/2010]
Kennedy, P. (1998). A guide to Econometrics. 4th
edn. Blackwell. Malden
Koch, T.W. (1995). Bank Management. 3rd
edn. The Dryden Press. London
Mathuva, D. (2009). Capital Adequacy, Cost Income Ratio and the Performance of Commercial Banks: The Kenyan
Scenario. International journal of Applied Economics and Finance. [Online]. ISSN 199-0886. Available from:
http://www. ScienceAlert.com. [Acessed: 29 May 2010]
Mwega, F.M. (2009). Global Financial Crisis: Kenya: Discussion series. Paper 7. [Online]. May 2009. Available at
www.odi.org.uk/resources/download/3312.pdf. [Accessed: 25 May 2010]
Naceur, S. (2003). The Determinants of the Tunisian Banking Industry Profitability: Panel Evidence.
[Online].October 2003. Available from: www.mafhoum.com/press6/ 174E11. pdf [Accessed: 29 May 2010]
Ngugi, R. (2001). An empirical analysis of interest rate spread in Kenya. [Online]. May 2001. Available at
http://www.aercafrica.org/documents/rp106.pdf. [Accessed: 25 May 2010]
Ngugi, N., Amanja D. & Maana L. (2006). Capital Market, Financial Deepening and Economic growth in Kenya.
[Online]. Available from: http://www.csae.ox.ac.uk/ conferences/2009-EDiA papers/513-Isaya.pdf [Accessed:
27 May 2010]
Northcott, C. (2004). Competition in Banking: Review of Literature. WP 2004-24. [Online]. June 2004. Available
from: http://www.bankofcanada.ca/en/res/wp/2004/wp04-24.pdf [Accessed: 7 June 2010]
Nzongang, T. & Atemnkeng, J. (2006). Market Structure and Profitability Performance in the Banking Industry of
CFA countries: the Case of Commercial Banks in Cameroon. [Online] May 2006. Available from:
http://www.jsd-africa.com/Jsda/Summer_2006/PDF [Accessed: 28 June 2010]
Okpara, G. C. (2009). A Synthesis of the Critical Factors Affecting performance of the Nigerian Banking System.
European Journal of Economics and Administrative sciences. [Online] Issue 17 ISSN 1450-2887. Available
from: http://www.eurojournals.com [Accessed: 05/06/2010]
Olugbenga, S. & Olankunle, A. (1998). Bank Performance and Supervision in Nigeria: Analysing the Transition to
Deregulated Economy. RP 71. [Online]. March 1998. Available from: http://idl-
bnc.idrc.ca/dspace/bitstream/10625/22919/1/113309.pdf [Accessed: 8 June 2010]
Oloo, O. (2009). The Banking Survey 2009. Think Business. Nairobi
Sufian, F. & Chong, R. R. (2008). Determinants of Bank Profitability in a Developing Economy: Empirical
Evidence from Philippines. Asian Academy of Management Journal of Accounting and Finance. [Online] 4(2)
p.91-112. Available from:http:// www.usm.my/journal/aamjaf/vol%204-2-2008/4-2-5.pdf. [Accessed:
06/06/2010
Tregenna, F. (2009), The fat years: the structure and profitability of the US banking Sector in the pre-crisis
period.Camridge Journal of Economics. [online] 33(10) p. 609-63. Available from:
www.oxfordjournals.org/cgi/content/abstract/33/4/609. [Accessed: 06/06/2010]
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
26
Uzhegova, O. (2010). The Relative Importance of Bank-specific Factors for Bank Profitability in Developing
Economies/P 2010/02. [Online]. April 2010. Available from: http://ssrn.com/abstract=1595751. [Accessed: 10
June 2010]
Waweru, N. and Kalani, V. (2009).Commercial Banking Crises in Kenya: Causes and Remedies. African Journal of
accounting Economics, Finance and Banking Research. [Online] 4 (4). Available from,
http://globip.com/pdf_pages/african-vol4-article2.pdf. [Accessed: 28/05/2010]
Appendices
Appendix 1: List of Commercial banks as at 31 December 2008
Table 1: List of Commercial Banks
Registered banks as at 31 December 2008
N Large (Assets >Ksh.15 billion) Abbreviation
1 Kenya Commercial Bank Ltd KCB
2 Barclays Bank of Kenya Ltd BARCL
3 Standard Chartered Bank Ltd STD
4 Co-operative Bank of Kenya Ltd COP
5 CFC Stanbic Bank Ltd CFC
6 Equity Bank Ltd**
7 Commercial Bank of Africa Ltd CAFR
8 Citibank, N.A. CITB
9 NIC Bank Ltd NIC
10 National Bank of Kenya Ltd NBK
11 Diamond Trust Bank Ltd DIAM
12 I & M Bank Ltd I&M
13 Prime Bank Ltd PRM
14 Bank of Baroda Ltd BAR
Medium(Ksh 5 billion <Assets < Ksh 15 billion)
15 Imperial Bank Ltd IMP
16 Bank of Africa Ltd BAFR
17 Bank of India INDIA
18 Ecobank Ltd 10 ECO
19 Family Bank Ltd**
20 Chase Bank Ltd CHS
21 Fina Bank Ltd FIN
22 K-Rep Bank Ltd KRP
23 African Banking Corporation Ltd ABC
24 Habib AG Zurich HABZ
25 Development Bank of Kenya Ltd DEV
26 Giro Commercial Bank Ltd GIR
27 Guardian Bank Ltd GDN
28 Southern Credit Banking Corp. Ltd SCRDT
29 Gulf African Bank Ltd**
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
27
Small(Assets< Ksh. 5 billion)
30 Consolidated Bank of Kenya Ltd CONS
31 Habib Bank Ltd HAB
32 Victoria Commercial Bank Ltd VTR
33 Equatorial Commercial Bank Ltd EQTRL
34 Fidelity Commercial Bank Ltd FED
35 Credit Bank Ltd CREDT
36 Transnational Bank Ltd TRANS
37 Middle East Bank Ltd MDL
38 First Community Bank Ltd**
39 Paramount-Universal Bank Ltd PMNT
40 Oriental Commercial Bank Ltd ORTL
41 Dubai Bank Ltd DUB
42 City Finance Bank Ltd CITF
43 Charterhouse Bank Ltd **
** Not included in the sample
Appendix 2: Detailed Eviews5 Regression Results
Table 1: Regression Results, Main Sample
Dependent Variable: ROA?
Method: Pooled EGLS (Cross-section weights)
Date: 08/17/10 Time: 10:45
Sample: 2002 2008
Included observations: 7
Cross-sections included: 38
Total pool (balanced) observations: 266
Linear estimation after one-step weighting matrix
Variable Coefficient Std. Error t-Statistic Prob.
C 4.935231 0.703327 7.016975 0.0000
CAP? 0.076043 0.013918 5.463834 0.0000
ASQ? -0.048364 0.009507 -5.087166 0.0000
LIQ? 0.009427 0.004499 2.095256 0.0373
CIR? -0.068347 0.004027 -16.97155 0.0000
RDI? 0.017143 0.006981 2.455622 0.0148
FGN? 0.003864 0.017943 0.215360 0.8297
MKT? -0.000573 0.000809 -0.707602 0.4799
Fixed Effects (Cross)
_ABC--C 0.484091
_BOA--C 0.815694
_BOB--C -0.110478
_BOI--C -0.489618
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
28
_BBK--C 2.360983
_CFC--C 0.390212
_CHS--C -0.806784
_CTB--C 0.002003
_CTF--C -4.544482
_CBA--C 0.171732
_CNS--C 0.544505
_COP--C 1.045727
_CRB--C 0.625829
_DEV--C -0.163489
_DIA--C -0.139525
_DUB--C -1.821864
_ECO--C -2.855662
_EQT--C -0.310616
_FED--C -0.153557
_FED--C -0.153557
_GIR--C 0.025673
_GDN--C -0.074752
_HBZ--C 0.042821
_HBB--C 0.042530
_IMB--C 0.054841
_IMP--C 1.913436
_KCB--C 0.508912
_KRP--C 0.614073
_MDL--C -0.187432
_NBK--C 1.279710
_NIC--C 0.185694
_RTL--C -2.368763
_PMT--C 0.256951
_PRM--C -0.433202
_SCR--C 0.390778
_STD--C 1.611831
_TRS--C 2.068476
_VTR--C -0.822720
Effects Specification
Cross-section fixed (dummy variables)
Weighted Statistics
R-squared 0.947581 Mean dependent var 6.667342
Adjusted R-squared 0.937144 S.D. dependent var 7.253862
S.E. of regression 1.818616 Sum squared resid 730.9276
F-statistic 90.79585 Durbin-Watson stat 1.877006
Prob(F-statistic) 0.000000
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
29
Unweighted Statistics
R-squared 0.669846 Mean dependent var 1.911053
Sum squared resid 872.0177 Durbin-Watson stat 2.038110
Table 2: Regression Results, Large banks Sample
Dependent Variable: ROA?
Method: Pooled EGLS (Cross-section weights)
Date: 08/21/10 Time: 22:28
Sample: 2002 2008
Included observations: 7
Cross-sections included: 13
Total pool (balanced) observations: 91
Linear estimation after one-step weighting matrix
Variable Coefficient Std. Error t-Statistic Prob.
C 6.187146 1.360453 4.547858 0.0000
CAP? 0.034731 0.018876 1.839932 0.0694
ASQ? -0.028027 0.009740 -2.877421 0.0051
LIQ? 0.002842 0.004959 0.573195 0.5681
CIR? -0.075150 0.007411 -10.14009 0.0000
RDI? 0.025815 0.007970 3.238767 0.0017
FGN? -0.034958 0.041515 -0.842064 0.4022
MKT? 0.000983 0.001836 0.535471 0.5938
Weighted Statistics
R-squared 0.792703 Mean dependent var 3.905012
Adjusted R-squared 0.775221 S.D. dependent var 2.281229
S.E. of regression 1.081551 Sum squared resid 97.08950
F-statistic 45.34180 Durbin-Watson stat 1.112499
Prob(F-statistic) 0.000000
Unweighted Statistics
R-squared 0.557080 Mean dependent var 2.845275
Sum squared resid 122.5179 Durbin-Watson stat 0.950914
Table 2: Regression Results, Small and medium banks Sample
Dependent Variable: ROA?
Method: Pooled EGLS (Cross-section weights)
Date: 08/21/10 Time: 22:24
Sample: 2002 2008
Economics and Finance Review Vol. 1(5) pp. 01 – 30, July, 2011 ISSN: 2047 - 0401
Available online at http://wwww.businessjournalz.org/efr
30
Included observations: 7
Cross-sections included: 25
Total pool (balanced) observations: 175
Linear estimation after one-step weighting matrix
Variable Coefficient Std. Error t-Statistic Prob.
C 3.619236 1.346383 2.688119 0.0079
CAP? 0.054203 0.011600 4.672862 0.0000
ASQ? -0.056343 0.008987 -6.269490 0.0000
LIQ? 0.007369 0.004251 1.733594 0.0848
CIR? -0.062805 0.005785 -10.85671 0.0000
RDI? 0.028475 0.011364 2.505842 0.0132
FGN? 0.045757 0.038577 1.186146 0.2372
MKT? -0.001630 0.001694 -0.962402 0.3372
Weighted Statistics
R-squared 0.712151 Mean dependent var 3.424557
Adjusted R-squared 0.700086 S.D. dependent var 3.947106
S.E. of regression 2.161610 Sum squared resid 780.3169
F-statistic 59.02373 Durbin-Watson stat 1.398354
Prob(F-statistic) 0.000000
Unweighted Statistics
R-squared 0.525417 Mean dependent var 1.425257
Sum squared resid 1064.922 Durbin-Watson stat 1.503557