You are on page 1of 11

African Journal of Business Management Vol.5 (27), pp.

11199-11209, 9 November, 2011


Available online at http://www.academicjournals.org/AJBM
DOI: 10.5897/AJBM11.1957
ISSN 1993-8233 2011 Academic Journals

Full Length Research Paper

Determinants of capital adequacy ratio in Turkish


Banks: A panel data analysis
Ahmet Bykalvarc1* and Hasan Abdiolu2
1

Faculty of Health Sciences, Selcuk University, 42030 Konya, Turkey.


Faculty of Economics and Administrative Sciences, Balikesir University, 10200, Bandirma, Balikesir, Turkey.

Accepted 26 September, 2011

The purpose of this study is to investigate the determinants of Turkish banks' capital adequacy ratio
and its effects on financial positions of banks covered by the study. Data are obtained from banks'
annual reports for the period 2006 - 2010. Panel data methodology is used in this study and analyzes
relationships between independent variables; bank size (SIZE), deposits (DEP), loans (LOA), loan loss
reserve (LLR), liquidity (LIQ), profitability (ROA and ROE), net interest margin (NIM) and leverage (LEV)
and a dependent variable which is capital adequacy ratio (CAR). The results of the paper indicate that
LOA, return on equity and LEV have a negative effect on CAR, while LLR and return on assets positively
influence CAR. On the other hand, SIZE, DEP, LIQ and NIM do not appear to have any significant effect
on CAR.
Key words: Capital adequacy ratio, Turkish banks, panel data analysis.

INTRODUCTION
Financial markets have changed shape as the providers
of financial services have extended their scope of
activities. However, in recent decades financial entities
commonly go bankrupt with disastrous consequences for
individuals and society since a bankrupt limited liability
company is not responsible for losses exceeding its
financial resources (Jong and Madan, 2011). In recent
years, banking crises have become increasingly common
and increasingly expensive to deal with. Prudential
regulation of banks is supposed to prevent or at least to
reduce the frequency of such crises. The group of issues

*Corresponding author. E mail: asalvarci@selcuk.edu.tr. Tel:


+90 507 379 29 49. Fax: +90 332 241 62 11.
Abbreviations: BRSA, Banking Regulation and Supervision
Agency; BAT, Banks Association of Turkey; SIZE, bank size;
DEP, deposits; LOA, loans; LLR, loan loss reserve; LIQ,
liquidity; ROA and ROE, profitability; NIM, net interest margin;
LEV, leverage; CAR, capital adequacy ratio; LSDV, leastsquare-dummy-variables; EGLS, estimates generalized least
squares; DW, Durbin-Watson.

that fails under the heading of bank capital adequacy has


received a great deal of attention from regulators,
bankers and academics in recent years and is likely to
continue as a subject for debate for many years to come
(Morgan, 1984). The Basle Accord was partly in response
to a series of international bank failures and concern over
unequal national capital standards.
Commercial banks are legally required to maintain
adequate capital funds. The primary function of bank
capital is to provide resources to absorb possible future
losses on assets. How much capital should a commercial
bank have? The standard capital adequacy arguments
sound like a broken record. Regulators always seem to
want more capital and bankers always want less. Both
sides need well-defined goals for establishing a capital
adequacy strategy and both sides should be taking a
broader view of the costs that are relevant in setting that
strategy. From the bank stockholders viewpoint, the
function of capital is to earn a satisfactory rate of return.
Commercial banks are legally required to maintain adequate capital funds. Any workable standard for measuring
capital adequacy should be expressed in terms of the
function of bank capital (Stegall, 1966). The rules on

11200

Afr. J. Bus. Manage.

minimum capital requirements have followed an international harmonization guided by the Basel Committee,
starting from 1988, date in which the first agreement was
issued (Brogi, 2010).In order to prevent bank failures and
protect the interest of the depositors, it is necessary to
require banks to maintain a significant level of capital
adequacy. Because of its importance, the western
banking system has established internationallyrecognized capital regulations, which are formulated by
the Basel Committee on Banking Supervision. The Basel
Committee on Banking Supervision is a committee of
banking supervisory authorities that was established by
the central bank governors of the Group of Ten countries
in 1975. It sets out the details of the agreed framework
for measuring capital adequacy and the minimum
standard. The basic idea of the 1988 Basel Capital
Accord has two parts. First, it established the definition of
capital and distinguished between its core elements (Tier
1) and supplementary elements (Tier 2). The 1988 Basle
Accord explicitly considered only credit risk. It required
internationally active banks to hold a minimum capital
equal to 8% of risk adjusted assets, with capital
consisting of Tier I capital (equity capital and disclosed
reserves) and Tier II capital (long term debt, undisclosed
reserves and hybrid instruments). In 1988, the Basel
Committee introduced capital adequacy regulation that
has been adopted by more than 100 countries (Jacobson
at al., 2002 and Chami and Cosimano, 2003). In Basel II
to revise the 1988 Accord has been to develop a
framework that would further strengthen the soundness
and stability. Insufficient risk dependence and the
possibilities of arbitrage in the present regulations are
important reasons behind the Basel Committees revision
of the capital adequacy rules. Its revision can also be
seen as a natural consequence of the rapid
developments in recent years in credit risk management
and credit risk measurement and the banks greater
readiness and ability to quantify credit risk (Jacobson at
al., 2002). The Committee is also retaining key elements
of the 1988 capital adequacy framework, including the
general requirement for banks to hold total capital
equivalent to at least 8% of their risk-weighted assets; the
basic structure of the 1996 Market Risk Amendment
regarding the treatment of market risk and the definition
of eligible capital. Basel II is more risk sensitive than the
1988 Accord higher than allowed for in this framework
(elik and Kzl, 2008; Thampy, 2004). Basel Capital
Accord covers not only the calculation of capital
adequacy ratio but also other supporting issues like
sound supervisory processes and market discipline.
The aim of this paper is to investigate empirically
determinants of capital adequacy ratio in Turkish banks.
Capital adequacy ratio is analyzed based on annually
data from 2006 to 2010 by nine bank specific variables.
The bank specific variables used in this study are bank
size, deposits, loans, loan loss reserve, liquidity,
profitability, net interest margin and leverage. In the

analyses panel data method was used.


A BRIEF OVERVIEW OF TURKISH BANKING SYSTEM
The banking sector forms a great part of the Turkish
financial system in its dynamic economy. Most of the
transactions and activities of money and capital markets
are carried out by banks. Most State banks were
established to finance a particular industry such as
agriculture, but private banks generally have close
connections to large industrial groups and holdings.
In Turkey, first banking activities started in early 1800s
with the so-called money-changers and the Galata
bankers. During this period, all quasi-banking activities
were carried out by money-changers, and The Galata
bankers consisted mostly of the ethnic-minorities in
Istanbul. With the deterioration of the Ottoman Empires
financial situation after the Crimean war, the Empire
needed external financial support. It was during this
period when representatives of several foreign banks
came to Istanbul with the purpose of extending credits to
the Empire at high interest rates. The Ottoman Bank
(Osmanl Bankas) was established in 1856 with its head
office in London and served as the Central Bank until the
1930s.The Central Bank, founded in the early 1930s, has
the usual central bank responsibilities, such as issuing
banknotes, protecting the currency, and regulating the
banking system and credit. The Central Bank also
finances the governments budget deficits and makes
loans to public and private banks. But after 1983 the
Central Bank began to reduce lending and stepped up its
supervisory functions.
Before 1980 there were only 4 foreign banks in Turkey,
but their number grew rapidly during the 1980s as
liberalized conditions and today there are almost 50 of
them. During these years a series of reforms were
adopted to promote financial market development;
interest and foreign exchange rates were liberalized, new
entrants to the banking system were permitted and
foreign banks were encouraged to operate in Turkey.
All banks in Turkey are subject to the Banks Act and to
the provisions of other laws regarding to banks. The new
Law brought the Banking Regulation and Supervision
Agency (BRSA) into life to safeguard the rights and
benefits of depositors. The Banks Association of Turkey
(BAT) is the representative body of the banking sector in
Turkey established for protecting and promoting the
professional interests of its members.Table 1 shows the
summary of the number of banks, branches and
personnel in Turkish banking sector. As of 2010, there
are a total of 49 banks operating with 9935 branches and
189.783 personnels in Turkey.
LITERATURE REVIEW
Many researchers have worked on capital adequacy. A

Bykalvarc and Abdiolu

11201

Table 1. Number of banks, branches and personnel in Turkish banking sector.

Number of banks
State Deposit Banks
Private Deposit Banks
Banks within the structure of SDIF
Global Capital Deposit Banks
Development and Investment Banks
Participation Banks
Number of Branches
Deposit Banks
Development and Investment Banks
Participation Banks
Number of Personnel
Deposit Banks
Development and Investment Banks
Participation Banks

2006
50
3
14
1
15
13
4
7.302
6.904
42
356
150.966
138.599
5.255
7.112

2007
50
3
12
1
17
13
4
8.122
7.658
42
422
167.760
153.212
5.361
9.187

2008
49
3
11
1
17
13
4
9.304
8.724
44
536
182.665
166.326
5.307
11.032

2009
49
3
11
1
17
13
4
9.581
8.968
44
569
184.205
167.063
5.340
11.802

2010
49
3
11
1
17
13
4
9.935
9.296
42
597
189.783
171.974
5.395
12.414

Source: BRSA, BAT.

number of researchers have provided insights, theoretical


as well as empirical, into the capital adequacy. A brief
overview of the studies focusing on developed and
emerging capital markets is presented in this paper.
Modigliani and Miller (1958) state that in a world with
perfect financial markets, capital structure and hence,
capital regulation is irrelevant. Hahn (1966) analyzes
factors determining adequacy of capital in commercial
banks. Analysis of the real factors influencing the quantity
and quality of capital is similar to a study in econometrics
with size, growth, and profitability as the independent
variables and capitalization as the dependent variable on
the first level. Independent variables on the first level
become dependent variables on the second level with
other independent variables influencing them. Quantitative and qualitative influences exist on both levels of
analysis. Capitalization is a function of size and growth
factors which vary in their influence according to growth
conditions and policies affecting structure. Profitability is
also a function of size and growth. Although, profitability
fluctuates in the short run with given levels of
capitalization, in the long run levels of capitalization
adjust to achieve competitive rates of return regardless of
differences in size and growth. The principle that secular
changes in capitalization occur through changes in
capital rather than through changes in deposits and
assets is established by multiple correlation analysis for
banks in the United States for the period 1953-1962.
On the second level of analysis, two sets of independent variables influence size, growth, and profitability.
One set includes local influences. The most pervasive set
of independent variables on the second level of analysis
includes the nature of the economy, the environment
provided by government, and the nature of the banking
system.

Santomero and Watson (1977) show that too tight a


capital regulation lead banks to reduce their credit offer
and, as a result, give rise to a fall in productive investment (Barrios and Blanco, 2003). They argue that, from
societys viewpoint, the optimal level of capital for the
banking system should be determined by the point at
which the marginal public returns to bank capital
(decreased chances of failure and decreased chances of
disruption of the payments system) exactly equal the
marginal public costs of bank capital (the opportunity cost
of diverting capital from other productive uses). Under
some legal and political structures, however, regulators
may not consider the social costs and, therefore, will
require more capital in the system than society may
desire. Short provides an excellent graphical presentation. They address the question of capital adequacy
macro/banking system level (Morgan, 1984). Marcus
(1983) explained the dramatic decline in capital to asset
ratio in U.S commercial banks during the last two
decades. He hypothesized that the rise in nominal
interest rates might have contributed to the fail in capital
ratios, time series-cross section estimation supports the
hypothesis regarding the interest rate.
Jeff (1990) gave an overview of capital requirements
for banks and financial institutions showing that there was
no difference in capital standards for these two types of
financial institutions. He said that capital adequacy was
reflected in asset size as a proxy of a well-managed
bank. This gave benefits for strongly capitalized banks
according to poorly capitalized banks which will sell
assets to raise capital. In I990's capital adequacy has
become the major benchmark for financial institutions.
And it was considered as a primary measure for safety
and soundness. Jeff was also regards the return on
assets ratio as a primary measure of a well-managed

11202

Afr. J. Bus. Manage.

bank. Bensaid et al. (1995) consider the capital requirement in the context of both adverse selection and moral
hazard. Adverse selection arises as the quality of the
banks assets is private information to the banks owners,
and moral hazard arises as the banks profit depend on
the unobservable effort chosen by the banker. Song
(1998) examined Korean banks responses to the Basel
risk weighted capital adequacy requirements implemented in 1993. The author concluded that the higher
capital requirements were generally effective because
Korean banks generally did not much utilize cosmetic
adjustments to increase their capital ratios.
Reynolds et al. (2000) studied financial structure and
bank performance from 1987 to 1997. Financial performance ratios which were used as dependent variables
(capital adequacy, liquidity, profitability, and loan
preference) were regressed to structural variables (bank
assets, net income, administrative expenses and time).
They have examined the financial structure and
performance of banks in eight East and Southeast Asian
Countries. They found that profitability and loan
preferences increases with size, but capital adequacy
decreases with size, so large banks have smaller capital
adequacy ratios, and profit is directly related to capital
adequacy. And as management (given by administrative
expenses) increased from a small size, the capital
adequacy ratio fill to a minimum, then increased as
management became larger and larger. Yu (2000)
documented bank size, liquidity and profitability are the
main determinants of bank capital ratio in Taiwan. The
author summarized that large banks in Taiwan have
much lower capital ratios than the small banks which is
consistent with the previous study where the large banks
feel that they are too big to fail. The author also
suggested that the banks mainly use internal source of
capital, this contributes that more profitable banks tend to
have higher capital ratios. The remarkable finding of this
paper is the relationship between the equity-to-asset ratio
and the liquidity ratio is significantly positive for small
banks, but significantly negative for medium size banks.
Aggarwal and Jacques (2001) reported that US banks
increased their capital ratio without increases in credit
risk. They concluded that the prompt corrective action
positively and significantly affected capital ratio in both
high capital and low capital banks with a faster speed of
adjustment in undercapitalized banks. Rime (2001)
examined the Swiss banks capital and risk behavior. The
author adopts a simultaneous equations approach to
examine whether Swiss banks which close to the
minimum regulatory standards tend to increase their
capital ratio. He suggests regulatory pressure has a
positive and significant impact on capital ratio. However,
there is no evidence of capital requirement has significant
impact on the banks risk taking behavior. Saunders and
Wilson (2001) suggested that the relationship between
charter value and capital structure decisions is
procyclical. Their regression results showed that during

economic booms situation, high charter value banks


posses a higher capital ratio. Nevertheless, during
economic recessions, higher charter value banks uphold
higher losses of charter value. The most important finding
of this paper is that charter value may not able to lessen
the amount of risky activities that banks involved.
Morrison and White (2001) state that capital adequacy
requirements are then useful mainly in restricting bank
size to be small enough to avoid moral hazard problems.
Such regulation can be looser the better is the regulators
reputation for auditing ability. This also suggests that
capital regulation can be looser in economies where
accounting procedures are more transparent. Tanaka
(2002) analyzes the effect of bank capital adequacy
regulation on the monetary transmission mechanism. The
results suggest that using a general equilibrium framework and a representative bank, the model demonstrates
that the monetary transmission mechanism is weakened
if banks are poorly capitalized or if the capital adequacy
requirement is stringent. Moreover, it predicts that Basel
II may reduce the effectiveness of monetary policy as a
tool for stimulating output during recessions.
Ghoshi et al. (2003) found that Indian public sector
banks have not resorted to assets substitution across the
risk-weight categories by substitute low risk government
securities for high risk loans in order to meet their capital
requirement. This show that capital regulation does
influence the banks decisions making. Chen (2003) reviews the situation and regulation of the capital adequacy
of state commercial banks in China. He finds that while
government support is proved to be the invisible treasure
of state banks, capital enhancement is always desired
and the most practical method is to use subordinated
debt to increase their supplementary capital. Chami and
Cosimano (2003) show that the overemphasis by regulators and market practitioners on Tier I or equity capital as
the relevant constraint for banks is not necessarily
supported by the Basel Accord. They show, in contrast,
that the Basel Accord intended for total capital-a
minimum of 8%- and not for equity capital to be the
binding constraint. Navapan and Tripe (2003) explained
that comparing banks return on equity is one way of
measuring their performance relative to each other. They
asserted that the proposition that there should be a
negative relationship between a banks ratio of capital to
assets and its return on equity may seem to be selfevident as to not need empirical verification. They found
negative relationship between capital and profitability
exists. Thampy (2004) shows the impact of capital
adequacy regulation on loan growth. Since loans have
the highest risk weight, a capital constrained bank would
want to conserve its capital by allocating fewer assets to
loans. This trend becomes more severe as the capital
constraint becomes binding which is the case for banks
with less than the required capital level. However, for
banks with high capital adequacy ratios, there is little
impact on loan growth. He also shows that in a capital

Bykalvarc and Abdiolu

constrained environment banks will reduce the supply of


loans. It also provides an explanation for the high
proportion of investments held by banks. The reduction in
the supply of loans is greater for banks which are
inefficient. The impact of higher capital standards on the
supply of bank credit in the economy would have a
greater impact in economies which have a bank
dependent or dominated financial system as opposed to
a capital markets dominated system.
Asarkaya and zcan (2007) analyzed the determinants
of capital structure in the Turkish banking sector. They
proposed an empirical model in order to identify the
factors that explain why banks hold capital beyond the
amount required by the regulation. They used a panel
data set that employs bank-level data from the Turkish
banking sector covering the period 2002 2006 and
estimated the model with generalized method of moments. The findings of their study suggested that lagged
capital, portfolio risk, economic growth, average capital
level of the sector and return on equity are positively
correlated with capital adequacy ratio and share of
deposits are negatively correlated with capital adequacy
ratio. Ahmad et al. (2008) reported new findings on
determinants of bank capital ratios in Malaysia. This
study presents a positive relationship between regulatory
capital and banks risk taking behavior. The study also
observes that capital requirement regulations introduced
in 1996 was ineffective whereas those mandated in 1997
are proved successful in the financial crises period. Also,
the study finds inconsistency with developed country
literature where results shows that bank capital ratios not
to been motivated by bank profitability. Toby (2008)
intended to determine the effects of bank liquidity
management practices (monetary policy outcomes) on
industry asset quality, measured with the proportion nonperforming loans in the loans portfolio. He also
investigated the effects of capital adequacy regulation on
selected bank asset quality and efficiency measures. He
found that the use of the minimum liquidity ratio is
irrelevant in controlling industry non-performing loans.
The cash reserve ratio is a more effective tool in controlling the level of non-performing loans in the industry
as a whole and the distressed banks in particular. As the
ratio of equity to loans advances increases, he should
expect the classified loans ratio to decrease and

CAR =

11203

asset quality to rise, and vice versa. Under regimes of


rising equity-to-total-assets ratio, he should expect the
loan loss reserves ratio to fall, and vice versa.
Mathuva (2009) finds that bank profitability is positively
related to the core capital ratio and tier 1 risk based
capital ratio. The study, using the return on assets and
return on equity as proxies for bank profitability for the
period 1998 to 2007, also establishes that there exists
negative relationship between the equity capital ratio and
equity. Ho and Hsu (2010) examine the relation between
firms financial structures and their risky investment
strategy in Taiwans banking industry. Their first result
demonstrates that the restrictions on capital adequacy
ratio have indeed affected firms risky investment
strategies, as market share and leverage are positively
related. Second, the firm performance is significantly and
positively related to firm size, leverage and financial cost.
Finally, the regression results show that financial
structures for banking firms are positively related to the
states of business cycle.
DATA AND METHODOLOGY
Data description and variable definitions
The purpose of this study is to investigate the determinants of
Turkish banks' capital adequacy ratio and its effects on financial
positions of banks covered by the study. This study used secondary
data and the data get from annual reports of the sample banks.
Data directly took from the commercial banks balance sheet
statement, profit and loss statement and from notes to account.
Time study period is five years from 2006 to 2010. In Turkey have a
population of commercial banks of 32 banks comprising 3 stateowned banks, 11 privately-owned banks, 1 bank under deposit
insurance fund, 11 foreign banks founded in Turkey and 6 foreign
banks having branches in Turkey. The study excluded bank under
deposit insurance fund and foreign banks having branches in
Turkey. After this selection, our final sample contains 24 banks.
Panel data methodology is used in this study and analyze
relationships between bank specific variables [bank size (SIZE),
deposits (DEP), loans (LOA), loan loss reserve (LLR), liquidity
(LIQ), profitability (ROA and ROE), net interest margin (NIM) and
leverage (LEV)) and a dependent variable which is capital
adequacy ratio (CAR)].
The total capital requirement requires a total risk-weighted capital
adequacy ratio of 8 per cent is used as the proxy for bank capital
adequacy ratio in this study. CAR is calculated according to the
Formula 1, presented as below:

Shareholders' Equity
Amount Subject to Credit Risk + Amoun Subject to Market Risk + Amount Subject to Operational Risk
(1)

Nine bank specific variables, that are hypothesized to influence


CAR, are examined. These bank specific variables are SIZE, DEP,
LOA, LLR, LIQ, ROA, ROE, NIM and LEV. Their selection
criteriaand a priori expectations of expected relationship with bank
capital adequacy ratio are referred to previous country bank
studies.

Explanatory variables and hypotheses


Bank size (SIZE)
The natural logarithms of total assets are used as a proxy of banks
size. Banks' size is important because of its relationship to bank

11204

Afr. J. Bus. Manage.

ownership characteristics and access to equity capital. Bank access


to equity capital may reflect a relative importance of bankruptcy cost
avoidance or managerial risk aversion. Jackson et al. (2002)
propose that the large banks wish to keep their good ratings and
therefore have considerable market-determined excess capital
reserves. However, Gropp and Heider (2007) and earlier Shrieves
and Dahl (1992) found that a banking organizations asset-size is
an important determinant of its capital ratio in an inverse direction,
which means that larger banks have lower capital adequacy ratios.
This may occur because firm size might serve as a proxy for a
banking organizations asset diversifications which reduces their
risk exposure. Therefore, we hypothesize either a positive or
negative relationship between bank size and capital adequacy ratio.
H1: Bank size has no statistically significant impact on banks capital
adequacy ratio.

Deposits (DEP)
Share of deposit is a ratio of total deposits to total assets. Deposits
are generally considered cheaper sources of funds compared to
borrowing and similar financing instruments (such as financing by
bond or syndication and securitization loans) for banks (Kleff and
Weber, 2003). When deposits increase, banks should be more
regulated and controlled to guarantee the depositors rights, and to
protect a bank from insolvency. If depositors cannot assess
financial soundness of their banks, banks will maintain lower than
optimal capital ratios. Optimal capital ratios are those that banks
would have observed if depositors could have assessed their
financial positions properly. But if depositors can assess a bank's
capital strength, a bank will maintain a relatively strong capital
positions because greater capital induces depositors to accept
lower interest rates on their deposits. Asarkaya and zcan (2007)
found a negative sign between share of deposit and capital
adequacy ratio.
H2: Share of deposit has no statistically significant impact on banks
capital adequacy ratio.

bad financial situation. Blose (2001) found that reserve of loan


losses caused a decline in capital adequacy ratio. Hassan (1992)
and Chol (2000) also argued a negative relationship between
capital adequacy ratio and loan loss reserve.
H4: Loan loss reserve has no statistically significant impact on
banks capital adequacy ratio.
Liquidity (LIQ)
A liquid asset to customer and short term funding are included to
proxy bank liquidity. Angbazo (1997) states that as the proportion of
funds invested in cash or cash equivalents increases, a bank's
liquidity risk declines, leading to lower liquidity premium in the net
interest margins. Therefore, an increase in bank liquidity (high LIQ)
may have a positive impact to capital ratio.
H5: Liquidity has no statistically significant impact on banks capital
adequacy ratio.
Profitability (ROA and ROE)
In this study return on assets and return on equity are used as a
proxy for profitability. In general, banks have to rely mainly on
retained earnings to increase capital. Profitability and the capital
adequacy ratio is most likely positively related, because a bank is
expected to have to increase asset risk in order to get higher
returns in most cases. Gropp and Heider (2007) found that more
profitable banks tend to have more capital relative to assets. Thus,
a positive relationship is expected between profitability and capital
adequacy ratio.
H6: Return on assets has no statistically significant impact on
banks capital adequacy ratio.
H7: Return on equity has no statistically significant impact on banks
capital adequacy ratio.

Net interest margin (NIM)


Loans (LOA)
Share of loans is a ratio of total loans to total assets. This ratio is
important because of its relationship with diversification and the
nature of investment opportunity set. It measures the impact of
loans in assets portfolio on capital. When risk increases, depositors
should be compensated for loss so capital adequacy ratio should
increase. Mpuga (2002) found a positive relationship between
capital adequacy ratio and share of loans. Therefore, a positive
relationship is expected between share of loan and capital
adequacy ratio.
H3: Share of loan has no statistically significant impact on banks
capital adequacy ratio.

Loan loss reserve (LLR)


Loan loss reserve defined as a valuation reserve against a bank's
total loans on the balance sheet, representing the amount thought
to be adequate to cover estimated losses in the loan portfolio. We
consider loan loss reserves to gross loans ratio as a proxy of bank
risk as this ratio may indicate the banks financial health. A negative
impact of loan loss reserve in capital could mean that banks in
financial distress have more difficulties in increasing their capital
ratio. In contrast, a positive effect could signal that banks voluntarily
increase their capital to a greater extent in order to overcome their

Net interest margin is defined as the ratio of net interest income to


average earning assets. It is a summary measure of banks' net
interest rate of return. While it is well known that the net interest
margin is a significant element of bank profitability, however the
effects of market interest rate volatility and default risk on the
margins are not well recognized. The net interest margins are set
by banks to cover the costs of intermediation besides reflect both
the volume and mix of assets and liabilities. More specifically,
adequate net interest margins should generate adequate income to
increase the capital base as risk exposure increases (Angbazo,
1997). The charter value which discussed in introduction predicts a
positive relationship between bank management quality and bank
capital. However, bank management may reduce the capital
cushioning if the default risk is very low. As a result, a negative
relationship is expected between net interest margin and capital
adequacy ratio.
H8: Net interest margin has no statistically significant impact on
banks capital adequacy ratio.

Leverage (LEV)
The final bank specific variable is the bank leverage factors which
proxy by the total equity to total liabilities ratio. Shareholder will find
high leveraged banks are more risky compared to other banks,

Bykalvarc and Abdiolu

11205

Table 2. Bank specific variables and predicted signs.

Bank specific variable


Bank Size (SIZE)
Deposits (DEP)
Loans (LOA)
Loan Loss Reserve (LLR)
Liquidity (LIQ)
Profitability (ROA and ROE)
Net Interest Margin (NIM)
Leverage (LEV)

Predicted sign
+/+
+
+/+
+ and +
+/+

Table 3. Descriptive statistics of study variables.

Variable
Mean
Median
Maximum
Minimum
Std. Dev.
Skewness
Kurtosis
Jarque-Bera
Probability
Observation

CAR
0.200
0.170
0.713
0.118
0.091
2.832
12.721
632.874
0.000*
120

SIZE
22.973
23.160
25.742
19.692
1.676
-0.107
1.832
7.055
0.029**
120

DEP
0.616
0.625
0.879
0.123
0.138
-0.973
4.840
35.858
0.000*
120

LOA
0.523
0.564
0.762
0.036
0.160
-0.933
3.337
17.993
0.000*
120

LLR
0.042
0.037
0.193
0.000
0.031
1.737
7.977
184.210
0.000*
120

LIQ
0.578
0.536
1.359
0.140
0.232
0.616
3.385
8.330
0.016**
120

ROA
0.016
0.016
0.055
-0.027
0.013
0.151
4.882
18.167
0.000*
120

ROE
0.129
0.136
0.344
-0.247
0.098
-0.486
4.409
14.656
0.001*
120

NIM
0.039
0.038
0.073
-0.085
0.018
-3.215
22.015
2014.607
0.000*
120

LEV
0..168
0.141
0.969
0.071
0.116
4.703
29.628
3987.645
0.000*
120

Note: Asterisks (*) and (**) denote the null of normality was rejected at 1% and 5% significance levels respectively.

therefore this increase required rate of return of the shareholders.


Consequently, the high leveraged banks may find raising new
equity difficult due to the high cost of equity capital. Ultimately, the
high leveraged banks may hold less equity than low leveraged
banks. Therefore, a positive relationship is expected between
leverage and capital adequacy ratio.
H9: Leverage has no statistically significant impact on banks capital
adequacy ratio.
Table 2 presents the summary of the selected bank specific variables that affect the capital adequacy ratio. The expected relationship between the bank specific variables and the bank capital
adequacy ratio also indicated.

capital adequacy ratio by using a multivariate panel regression


model. This model was useful and suitable because the research
focus lied in examining the contemporaneous relationships between
capital adequacy ratio and bank specific variables. Based on both
theoretical and empirical literature reviewed, this study hypothesize
the model between CAR and nine bank specific variables, namely
SIZE, DEP, LOA, LLR, LIQ, ROA, ROE, NIM and LEV. The
hypothesized model is represented as follows:

CAR = f SIZE, DEP, LOA, LLR, LIQ, ROA, ROE, NIM, LEV (2)
In order to see whether the above identified bank specific variables
could explain CAR, the multivariate panel regression model is
formed:

Econometric model
This study examined the effects of bank specific variables on

CARit = 0 + 1 SIZEit + 2 DEPit + 3 LOAit + 4 LLRit + 5 LIQit + 6 ROAit + 7 ROEit + 8 NIM it + 9 LEVit + it
In the above equation
variables while

it

is constant and

is coefficient of

is the residual error of the regression. The

multivariate panel regression method is used to compute the


estimates of the regression model stated above and all estimations
have been performed in the econometrical software program
EViews 5.1 whereas the ordinary calculations in Excel.

(3)

EMPIRICAL RESULTS
Various descriptive statistics are calculated of the variables under study in order to describe the basic characteristics of these variables. Table 3 shows the descriptive
statistics of the data containing sample means, medians,
maximums, minimums, standard deviations, skewness,

11206

Afr. J. Bus. Manage.

Table 4. The pairwise correlation matrix for dependent (CAR) and explanatory variables.

CAR
SIZE
DEP
LOA
LLR
LIQ
ROA
ROE
NIM
LEV

CAR
1
-0.363*
-0.373*
-0.773*
-0.234**
0.665*
0.275*
-0.039
-0.611*
0.503*

SIZE

DEP

LOA

LLR

LIQ

ROA

ROE

NIM

LEV

1
0.362*
0.122
0.131
-0.322*
0.362*
0.648*
0.183**
-0.425*

1
0.266*
0.181**
-0.354*
-0.121
0.171
0.165
-0.542*

1
0.046
-0.698*
-0.274*
-0.120
0.506*
-0.309*

1
-0.089
-0.087
-0.034
0.196**
-0.135

1
0.162
-0.150
-0.401*
0.501*

1
0.824*
-0.121
0.326*

1
0.096
-0.173

1
-0.264*

(*) and (**) indicate significance at 1 and 5% respectively.

kurtosis as well as the Jarque-Bera statistics and


probabilities (p-values).
As it can be seen from the Table 3, all the variables are
asymmetrical. More precisely, skewness is positive for
five series, indicating the fat tails on the right-hand side of
the distribution comparably with the left-hand side. On
contrary, SIZE, DEP, LOA, ROE and NIM have a
negative skewness which indicates the fat tails on the
left-hand side of the distribution. Kurtosis value of all
variables also shows data is not normally distributed
because values of kurtosis are deviated from 3. The calculated Jarque-Bera statistics and corresponding pvalues are used to test for the normality assumption.
Based on the Jarque-Bera statistics and p-values this
assumption is rejected at 1% level of significance for
CAR, DEP, LOA, LLR, ROA, ROE, NIM and LEV and
rejected at 5 percent level of significance for SIZE and
LIQ.
The dependent and independent variables are tested
for multicollinearity based on a simple correlation matrix.
As depicted in Table 4, all of them are have no
collinearity problem.
Having concluded that none of the bank specific
variables are highly correlated and no multicollinearity
amongst these variables exist; the effect of bank specific
bank variables on the capital adequacy ratios is
examined by the Panel Data estimation. The regression
results of panel data are reported in Table 5.
The dependent variable (CAR) is the capital adequacy
ratio. Model I and Model II correspond to cross-section
fixed effects, that is, least-square-dummy-variables
(LSDV) or fixed effects and cross-section random effect
models respectively (Since the errors are expected to be
correlated, we used panel estimates generalized least
squares (EGLS) in order to get efficient estimates). The
models are estimated using a panel of 120 observations
for the period 2006 to 2010 derived from 24 Turkish
banks. The estimated coefficients are also assigned for
the ith banks with the aim of capturing the influence of
specific characteristics of each individual bank.

Then, we extended the regression results in order to


select which model is better; fixed effects or random
effects model. A central assumption in random effects
estimation is the assumption that the random effects are
uncorrelated with the explanatory variables. One
common method for testing this assumption is to employ
a Hausman (1978) test to compare the fixed and random
effects estimates of coefficients (Baltagi, 2001;
Wooldridge, 2002). The intention is to find out whether
there is significant correlation between the unobserved
individual specific random effects (i) and the regressors.
The result of Hausman test based on chi-squared statistic
as reported in Table 6 suggested that the corresponding
effects are statistically significant, hence the null
hypothesis is rejected by our data and fixed effects model
is preferred.
Finally, the analysis of the regression results is as
follows. In both models, the result for LOA, LLR, ROA
and ROE reveal a consistent signs and significant
relationship with CAR. However, further analysis will be
based on fixed effects model (Table 5, 2nd column).
As Table 5, 2nd column reports that the values of
adjusted R Square (0.8146) suggest that model serves its
purpose in determining the effect of bank specific
variables on capital adequacy ratio. In other words,
81.46% variability of the capital adequacy ratio can be
explained by the SIZE, DEP, LOA, LLR, LIQ, ROA, ROE,
NIM and LEV. Before analyzing the coefficients, one
should look at the diagnostics of regression. In this
matter, Durbin-Watson (DW) statistic can show us the
serial correlation of residuals. As a rule of thumb, if the
DW statistic is less than 2, there is evidence of positive
serial correlation. The DW statistic in our output is 1.8606
and this result confirms that there is no serial correlation.
With computed F-value of 17.3348 (p<0.000) for the
panel data regression, we reject the null that all
coefficients are simultaneously zero and accept that the
regression is significant overall.
As it can be seen from the Table 5, SIZE, DEP, LIQ
and NIM do not have any effect on capital adequacy ratio

Bykalvarc and Abdiolu

11207

Table 5. Panel regression results (Dependent variable: CAR).

Explanatory variable
CONSTANT
SIZE
DEP
LOA
LLR
LIQ
ROA
ROE
NIM
LEV
N x T = 24 x 5
2
R
Adj. R2
F-Stat.
Durbin-Watson Stat.

Model I:
{Cross-Section (Fixed Effects)}
0.5094(0.3658)
-0.0093(0.0151)
0.0269(0.0724)
-0.1829***(0.0925)
0.3485***(0.1957)
0.0317(0.0438)
4.0319**(1.6017)
-0.4419***(0.2429)
-0.5284(0.3613)
-0.2228**(0.0916)

Model II:
{Cross-Section (Random Effects)}
0.6542***(0.0789)
-0.0107*(0.0033)
0.0001(0.0332)
-0.3254*(0.0386)
-0.3448*(0.1278)
0.0054(0.0263)
3.1358*(1.0266)
-0.3096**(0.1402)
-0.9060*(0.2574)
-0.0254(0.0629)

120 (balanced)
0.8644
0.8146
17.3348*
1.8606

120 (balanced)
0.7699
0.7512
40.9177*
1.3546

Values in parentheses are the standard errors. *, ** and *** denote significant level at 1, 5 and 10%, respectively.

Table 6. Hausman test for correlated random effects.

Test cross-section random effects


Cross-section random

Chi-Sq. Statistic
37.941159

Chi-Sq. d.f.
9

Probability
0.0000

H0: i are uncorrelated with Xit; H1; i are correlated with Xit.

at 1, 5 and 10% significant levels, respectively. An


interesting point in the results of the study is the
relationship observed between LOA and CAR. The effect
of LOA on CAR is statistically significant at 10% level as
expected, but with the wrong sign. High amount of loan
loss reserve is commonly signifying a high risk because
the bank expects the loans will default. This also implies
that the worse the financial health of the bank, the higher
is the banks capital adequacy ratio. The coefficient of
LLR is statistically significant at the 10% level and has a
positive sign. This explains that banks in financial distress
have more difficulties in decreasing their capital ratio.
LEV has a negative coefficient and statistically significant
at 5% level. This means the low leverage bank which has
a high total equity to total liabilities ratio will hold more
equity capital. It is consistent with our initial priori
because low leveraged bank may not find raising new
equity difficult and thus hold more equity than high
leveraged banks. ROA has a significant and positive
effect on capital adequacy ratios. The coefficient of ROA
shows that a one unit increase in profitability increases
the bank capital by 4.0319 unit according to the Table 5.
On the other hand, ROE has a significant and negative
effect on CAR. The coefficient of ROE shows that a one
unit increase in profitability decreases the bank capital by

0.4419 unit according to the Table 5.


To sum up our results, LOA, LLR, ROA, ROE and LEV
seem to effect CAR. On the other hand, SIZE, DEP, LIQ
and NIM do not appear to have significant effects on
capital adequacy ratios. Hypothesis testing results are
summarized in Table 7.
Conclusions
The main objective of the present paper is to investigate
empirically determinants of CAR in Turkish banks and its
effects on financial positions of banks covered by the
study. This study used secondary data and the data
gotten from annual reports of the sample banks. Data
directly taken from the commercial banks balance sheet
statement, profit and loss statement and from notes to
account. Time study period is five years, from 2006 to
2010. Panel data regression is used in this study and
analyzes relationships between bank specific variables:
SIZE, DEP, LOA, LLR, LIQ, ROA, ROE, NIM and LEV
and a dependent variable which is CAR.
The results of the paper indicate that LOA, return on
equity and LEV have a negative effect on CAR, while
LLR and return on assets positively influence CAR. On

11208

Afr. J. Bus. Manage.

Table 7. Summary of hypotheses testing.

Variable
SIZE
DEP
LOA
LLR
LIQ
ROA
ROE
NIM
LEV

Sign
+
+
+
+
-

Reject null hypothesis


No
No
Yes
Yes
No
Yes
Yes
No
Yes

the other hand, SIZE, DEP, LIQ and NIM do not appear
to have any significant effect on CAR.
REFERENCES
Aggarwal R, Jacques KT (2001). The impact of FDICIA and prompt
corrective action on bank capital and risk: estimates using a
simultaneous equations model. J. Bank. Finan., 25: 1139-1160.
Ahmad R, Ariff M, Skully MJ (2009). The determinants of bank capital
ratios in a developing economy. Asia-Pac. Finan. Mark., 3(4): 255272.
Angbazo L (1997). Commercial bank net interest margins, default risk,
interest-rate risk, and off-balance sheet banking. J. Bank. Finan.,
21(1): 55-87.
Asarkaya Y, zcan S (2007). Determinants of capital structures in
financial industries: The case of Turkey, pp. 91-109 www.
bddk.org.tr/websitesi/turkce/Raporlar/BDDK.../3885makale5.pdf.
Baltagi BH (2001). Econometric analysis of panel data, 2d ed. New
York: John Wiley & Sons.
Barrios VE, Blanco JM (2003). The effectiveness of bank capital
adequacy regulation: a theoretical and empirical approach. J. Bank.
Finan., 27: 1935-1958.
BAT. www.tbb.org.tr.
Bensaid B, Pages H, Rochet JC (1995). Efficient regulation of banks
solvency. Institut d'economie industrielle, Universit des Sciences
sociales de Toulouse.
Blose LE (2001). Information asymmetry capital adequacy, and market
reaction to loan loss provision announcements in the banking
industry. Q. Rev. Econ. Finan., 14 (2): 239-258.
Brogi, M (2010). Capital adequacy and dividend policy: evidence from
Italian banks. Govern. E Organ. Delle Banche, pp. 219-248.
http://ssrn.com/abstract=1707962.
BRSA. www.bddk.gov.tr.
elik F, Kizil (2008). Basel II bank capital adequacy approach and the
Turkish banking. Dou Unv. J., 9(1): 19-34.
Chami R, Cosimano T (2003). The nature of capital adequacy
constraints under the Basel accord. Working Paper Series,
http://ssrn.com/abstract=456981.
Chen J (2003). Capital adequacy of Chinese banks: evaluation and
enhancement. J. Bank. Reg., 4(4): 320-327.
Chol G (2000). The macroeconomic implications of regulatory capital
adequacy requirements for Korean banks. Econ. Notes by Banca
Monte Dei Paschi Di Siena Sp A, 29(1): 111-143.
Ghoshi S, Nachane, DM, Sahoo S (2003). Capital requirements and
bank behavior: an empirical analysis of Indian public sector banks. J.
Int. Dev., 15: 145-156.
Gropp R, Heider F (2007). What can corporate finance say about banks
capital
structures?
Working
paper,
http://www.wiwi.unifrankfurt.de/schwerpunkte/finance /master/brown/177 .pdf.
Hahn PJ (1966). Factors determining adequacy of capital in commercial
banks. J. Finan., 21(1): 135-136.

Significance level
0.10
0.10
0.05
0.10
0.05

Hassan K (1992). An empirical analysis of bank standby letters of credit


risk. Rev. Fin. Econ., 2 (1): 31-44.
Hausman JA (1978). Specification Tests in Econometrics.
Econometrica, 46 (6): 1251-1271. http://dx.doi.org/10.2307/1913827
Ho SJ, Hsu SC (2010). Leverage, performance and capital adequacy
ratio in Taiwans banking industry. Japan. World Econ., 22: 264272.
Jackson P, Perraudin W, Sapporta V (2002). Regulatory and economic
solvency standards for internationally active banks. J. Bank. Finan.,
26: 953-976
Jacobson T, Linde J, Roszbach K (2002). The IRB approach in the
Basel Committees proposal for new capital adequacy rules: some
simulation-based illustrations. Econ. Rev., 4: 35-72.
Jeff L (1990). Capital adequacy: The benchmark of the 1990s bankers
magazine, 173(1):14-18.
Jong P, Madan DB (2011). Capital adequacy of financial enterprises.
Working Paper Series. http://ssrn.com/abstract=1761107.
Kleff V, Weber M (2003). How do banks determine capital? Empirical
evidence from Germany. ZEW Disc. Paper, pp.03-66.
Marcus AJ (1983). The bank capital decision: a time series-cross
sectional analysis. J. Finan., 38 (4): 1217-1230.
Mathuva DM (2009). Capital adequacy, cost income ratio and the
performance of commerical banks: the Kenyan scenario. Int. J.
Appl.Econ. Finan., 3(2): 35-47.
Modigliani F, Miller M (1958). The cost of capital, corporation finance
and the theory of investment. Am. Econ. Rev., 48: 261-297.
Morgan, GE (1984). On the adequacy of bank capital regulation. J.
Finan. Quant. Anal., 19 (2): 141-162.
Morrison AD, White L (2001). The role of capital adequacy requirements
in sound banking systems. Oxford Fin. Res. Center Working Paper.
FE-04. http://ssrn.com/abstract=270728.
Mpuga P (2002). The 1998-99 banking crisis in Uganda: What was the
role of the new capital requirements? J. F. Reg. Comp., 10(3): 224242.
Navapan K, Tripe D (2003). An exploration of the relationship between
bank capital levels and return on equity. Proceeding of the 16th
Australasian Finance and Bank. Conf. (AFBC03). Palmerston North,
pp. 1-15.
Reynolds SE, Ratanakomut SG (2000). Bank financial structure in precrisis east and south East Asia. J. Asian Econ., 11 (3): 319-331
Rime B (2001). Capital requirements and bank behavior: empirical
evidence for Switzerland. J. Bank. Finan., 25: 789-805.
Saunders A, Wilson B (2001). An analysis of bank charter balue and its
risk constraining incentives. J. Finan. Serv. Res., 19: 185-195.
Shrieves R, Dahl D (1992). The relationship between risk and capital in
commercial banks. J. Bank. Finan., 16: 439-457.
Song I (1998). Korean Banks responses to the strengthening of capital
adequacy requirements. Federeal Reserve Bank of San Francisco.
Working Paper (PB98-01)
Stegall DP (1966). How capital adequacy standards affect savings
deposits. Banking, 58 (11): 58-60
Tanaka M (2002). How do bank capital and capital adequacy regulation
affect the monetary transmission mechanism? Cesifo Working Paper.
799. http://ssrn.com/abstract_id=348461.

Bykalvarc and Abdiolu

Thampy A (2004). BIS capital standards and supply of bank loans.


Working Paper Series. http://ssrn.com/abstract=561723.
Toby AJ (2008). Monetary policy, capital adequacy regulation and
banking system soundness in Nigeria: empirical research findings. J.
Fin. Manag. Analy., 21 (1).
Wooldridge J (2002). Econometric analysis of cross section and panel
data. MIT Press.

11209

Yu HC (2000). Banks' capital structure and the liquid asset-policy


implication of Taiwan. Pac. Econ. Rev., 5(1): 109-114.