Declaration/Statements Page DECLARATION This work has not previously been accepted in substance for any degree and is not being currently submitted in candidature for any degree. Signed………………………………………(Candidate) Date………………………………………… STATEMENT 1 This work is the result of my own investigations, except where otherwise stated. Where correction services have been used, the extent and nature of the correction is clearly marked in a footnote(s). Ot he r sources are acknowledged by footnotes gi vi ng expl icit ref er ences. A bibliography is appended. Signed………………………………………(Candidate) Date………………………………………… STATEMENT 2 I hereby give consent for my work, if accepted, to be available for photocopying and for inter-library loan, and for the title and summary to be made available to outside organisations. Signed………………………………………(Candidate) Date………………………………………… 1
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Pasiouras & Kosmidou (2006) and Deitrich & Wanzenried (2009). In the present
work, following the treatment in the mentioned literature, I use a linear regression
model for estimating the effect of various determinants on profitability. The
determinants of profitability used in this work are divided into two groups. The first
group involves eight bank-specific variables characterizing liquidity, bank capital,
credit risk, productivity growth, operating expenses, size, population density of the
region where the bank is situated as well as two dummy variables characterizing the
bank category (Cantonal, Raiffeisen or savings). The second group involves
macroeconomic factors that can affect bank profitability: GDP rate and inflation rate.
The bank profitability was measured in terms of return on average assets (ROAA) and
return on average equities (ROAE). In this study, I employed the data from 16 Swiss
banks covering a period between 2003 and 2008. I restricted the analysis to three
categories of banks, namely Cantonal, Raiffeisen and regional savings banks.
The results of the regression analysis show that five variables show a statistically
significant correlation with bank profitability. These variables are equity to assets ratio
(capital), cost to income ratio (operating expenses), logarithm of the total assets by the
price index (bank size), population density, as well as the macroeconomic variable
inflation. Equity to assets ratio, bank size and population density show a positivecorrelation with bank profitability variables ROAA and ROAE. On the other hand,
cost to assets ratio and inflation show a negative. The current findings are broadly in
agreement with previous studies. In addition, by comparing the resulting regression
coefficients standard deviations of the relevant variables with the mean value ROAA
(as explained in Chapter 6), I analyze the economic significance of the variables for
bank profitability. I find that only equity to assets, population density and bank size
size on profitability is valid only up to a particular limit. Berger et al (1987) have also
argued that very large banks face scale inefficiencies.
The other important determinant of profitability, considered in a large number of
papers, is bank’s expenses. Bank expenses are linked with the notion of efficient
management. Some of the literature suggests that expenses-related variables should be
included in the cost part of the profit function. Bourke (1989) suggest that reduction of
expenses improve the efficiency and as a consequence increase profitability, implying
a negative relation between profitability and operating expenses. In addition,
Molyneux & Thornton (1992) and Bourke (1989) emphasize a positive correlation
between quality of management and profitability.
Yet another important aspect of the banking business affecting profitability is risk
management. Poor asset quality and low level of liquidity are the two main reasons of
bank failures. Bank’s risks can be divided into liquidity and credit risks. Liquidity risk
arises when banks are unable to meet their liquidity liabilities because of unforeseen
withdrawals of deposits, and can be considered as determinant of bank’s profitability.
Bourke (1989) finds a positive relationship between the level of liquidity and
profitability. In contrast, Molyneux & Thornton (1992) report an opposite results. The
credit risk arises when a customer defaults on a loan or any other type of financial
contracts. The changes in the credit risk can dramatically affect the performance of the
bank and its loan portfolio. There is a significant amount of literature that concentrates
on the relationship between credit risk and profitability in the banking sector. The
studies that include Duca & McLaughlin (1990), Miller & Noulas (1997) indicate
negative effect of credit risk on profitability. These studies show that an increased
exposure to credit risk typically leads to a decline in the bank profitability. Miller &
Noulas (1997) argue that the higher the risk loans held by a financial institution, thehigher is the possibility of unpaid loans and consequently lower the profitability.
Several works argue that leverage is an important factor in explaining bank
performance. In general, leverage refers to the use of debt to supplement investment.
Molyneux (1993) emphasizes that higher level of equity would reduce the cost of
capital, leading to an increase in profitability. Furthermore, an increase in capital can
increase expected earning via reducing the expected cost of financial distress. Bourke
(1989), Molyneux & Thornton (1992) and Goddard et al. (2004) find a positive
relationship between capital ratios and bank profitability.
2.3 External determinants
Let us move to external determinants of bank profitability. The external variables are
usually divided into the two groups. The first group describes characteristics of the
market, such as concentration, ownership status, competition and industry size. The
second group of variables describes the macroeconomic environment, and includes
interest rate, inflation or cyclical output.
2.3.1. Market Characteristics
All investigations of the structural effect on bank’s profitability typically start with the
statement of two alternative hypotheses: Efficient-Structure (ES) and Market-Power
(MP) hypotheses. The Efficient-structure hypothesis suggests that financial institution
can observe higher profits if they have superior management and productiontechnologies that have lower cost. The increase in managerial and scale efficiencies
lead to higher concentration and consequently to the higher profits. The Market-Power
hypothesis states that increase in market power leads to monopoly profits. More
specifically, Relative-Market-Power hypothesis claims that only firms with large
market shares and well-differentiate products can achieve monopoly power and earn
non-competitive profits.
Numerous studies, such as Berger (1995 a), Smirlock (1985), Berger & Hannan
(1989) and Frame and Kamerschen (1997) analyze the profit–structure relationship in
banking and test the two hypotheses. Frame & Kamerschen (1997) reject the notion
that profits are result of superior x-efficiency (a version of the ES) and conclude that
only market power leads to super-normal profits. However, Smirlock (1985), who
directly tested these competing hypotheses, has provided evidence in favor of ES
hypothesis. Smirlock (1985) used data from a sample of 2700 banks from the 10th
Federal Reserve District in USA during a period between 1973-1978, showed a
positive relationship between market share and profitability. However he did not find
evidence for correlation between market structure and profitability. Similar
investigations were conducted by Berger (1995a), who showed that managerial
efficiency not only raised profits, but also leads to market share gains and
consequently increased concentration. Berger (1995) emphasizes that the positive
relationship between concentration and profits may be a spurious result due to
correlation with other variables. Consequently, Berger (1995) claimed that, taking into
account other factors, the role of the concentration is small. In contrast, other authors,
such as Bourke (1989) and Short (1979), suggest that increased concentration is not a
result of managerial efficiency, but rather reflects an increasing deviation from
competitive market structure leading to a high level of profitability. Hence, as
suggested by these authors, concentration can positively correlate with profitability.
This is consistent with the traditional structure-conduct-performance paradigm.
Molyneux & Thornton (1992) also found support for this view by taking a sample of
European banks in eighteen countries between 1986-1989. The theoretical links
between the measure of concentration and market power (MP) have been investigated
by a number of works. For instance, Cowling & Waterson (1979), Dansby & Willig
(1979) and Novshek (1980) have shown that a Cournot oligopoly would generate
equilibrium price-cost margins.
The general view on competition and market structure relationships is based on
monopoly power hypothesis. In line with this hypothesis, more concentrated markets
tend to be more collusive with banks achieving monopolistic profits since they operate
with wider margin of intermediation. These arguments are referred to as the
“Structural Model” since they are based on the structure of the banking market.
However, “Non Structural Model” have also been developed, to tackle the theoreticaland empirical deficiencies of the structural model. The works of Rosse & Panzar
(1977) and Panzar & Rosse (1982, 1987) elaborate on the “non structural model”,
developing the so-called H-statistic, which was later used for the examination of the
competitive structure of the banking sector. The Panzar and Rosse (P-R) model
measures competition and investigates the competitive conduct of banks without
employing any information about structure of the market. The P-R methodology was
widely used in literature for investigating the relationship between competition,
concentration and profitability. Coccorese (2004), in his investigations of Italian
banking industry, used Rosse-Panzar test and to determine the link between observed
level of local banking competition and macroeconomic performance in a given region.
Al-Muharrami et al (2006) investigated the market structure of the GCC (Gulf
Cooperation Council) banking industry during the period 1993-2002 and evaluated the
monopoly power of banks. By using P-R model, they found a mixed bag of
competitive and monopolistic competition as well as monopoly within the GCC
economies. Their findings emphasized that the banking market in GCC, except of
Saudi Arabia and Kuwait, have some way to go in developing a competitive structure.
Several works investigate the relationship between bank regulation and
supervision on the one hand and profitability on the other. Pasiouras et al. (2007),
investigated 677 commercial banks operating in 88 countries covering the period 2000-
2004, and found empirical evidence of correlation of cost and profit efficiency with
regulation and supervision around the world.
Another question, which has been analyzed in the literature, is related to the
ownership status of the bank and its connection to profitability. According to the Short
(1979) the government ownership of the banks is correlated inversely with
profitability. The work of Barth et al. (2004) also emphasizes the relationship betweengovernment ownership and bank efficiency. In contrast, Molyneux & Thornton (1992)
have shown a significant positive correlation between return on capital and
government ownership, suggesting that state-owned banks generate higher returns on
capital than their competitors from the privet sector. Thus, the debate on this question
remains open.
2.3.2. Macroeconomic characteristic
Let us move to the second group of external determinants of bank profitability, which
deal with macroeconomic control variables. The bank profitability is very sensitive to
the changes in macroeconomic conditions, in spite of the trends in the industry
directed at diversification and large use of financial engineering to manage risk
associated with business cycles. In general, when the economy is doing well, the
banks are encouraged to lend more funds and charge higher margins, by improving the
Switzerland has one of the most stable economies in the world and a per capita gross
domestic product (GDP) that is higher than in most of the developed countries of the
world. Furthermore, Switzerland has a low budget deficit and unemployment levels.
Its currency, the Swiss Frank (CHF), has traditionally held a reputation for stability.
The inflation rate in recent years was kept below 1%, which is lower than the figures
for the EU and USA. The interest rate has also been quite low, due to a high rate of
savings and a large inflow of foreign money. Switzerland traditionally attracts a lot of
foreign investors and has reputation of safe place to invest because of its policy of
long-term monetary security and bank secrecy.
The impressive economic accomplishments are in part the result of a strong link
between industry and international trade as well as achievements of the service
industry. Many of the Swiss companies have expanded beyond their domestic market,
making Switzerland an important player in world trade. The most important trading
partner for Switzerland has traditionally been the EU. Approximately two-thirds of all
exported goods and four-fifth of all imports are traded with Europe. In Europe,
Switzerland’s most important trading partner has been Germany. Swiss economy is
based on production of export-oriented machinery and electronic equipment, such asgenerators, turbines, textile and tool machines, mills, watches as well as other
innovative high-tech products. Apart from this, Switzerland produces sizable amounts
of agro-chemicals and pharmaceutical drugs.
A major role in the Swiss economy is played by the financial services (banking,
insurance, investment). According to the “Swiss Banking Sector” Compendium
Edition 2006, the financial sector whose net output was 68 bn CHF
and private bankers. Different groups of banks are fully or partially specialized. In
order to better understand the banking system of Switzerland, it is useful to provide a
brief description of each type.
Let us begin the description from the category of big banks. The big banks cover
all areas of financial activities. For this reason, they are key figures in most sectors of
the domestic market as well as being major players in the international financial
markets. The two main “big” banks are UBS AG and the Credit Suisse Group. They
account more than 50% of the balance sheet total of all banks in Switzerland. Both
banks have broad branch networks inside the country and abroad. UBS AG is the
leading bank in Switzerland for individual as well as corporative clients, and is also
orientated globally. UBS is the world’s leader in the wealth management and an
important international player in investment banking and securities business. The
Credit Suisse Group can be characterized as a globally-active provider of financial
services. It offers various services such as financial advice to private clients, insurance
companies and as a financial intermediary. It serves global companies and institutions
as well as public corporations.
The cantonal banks are state-owned banks and the major part of their capital isowned by the canton. Nowadays there are a total 24 Cantonal banks in Switzerland.
They are semi-governmental organizations with a state guarantee of liabilities. The
banks in this group vary in size and business activities. The smaller banks are
orientated towards domestic market and traditional banking, focusing mainly on
savings and mortgages, while the larger institutions are involved in a broader array of
financial activities. In spite of the fact that they are closely connected with the state,
cantonal banks must upkeep commercial principles in their financial activity, while
their chief target is to promote the canton’s economy. The balance sheet of cantonal
banks normally ranges from 2bn to 80bn CHF. In total, Cantonal banks in Switzerland
account for around 30% of banking business and have a combined total balance sheet
of more than 300bn CHF.
Regional and saving banking is exemplified by smaller universal banks that focus
mostly on domestic, traditional banking with emphasis on lending/deposits business.
The activity of banks is typically restricted to one region or small geographical areas.
Their advantage lies in familiarity with local circumstances and the regional business
cycle together with proximity to customers. As of 2005, regional banks reported a
balance sheet total of 43bn CHF and employed approximately 2,400 staff.
The Raiffeisen banks are small independent banks that are located in rural areas.
They are organized in the form of cooperatives and are mostly orientated towards
mortgage lending. The Raiffeisen banks are affiliated to the Swiss Union of
Raiffeisen. The function of this union is to coordinate banks activities, provide on-the-
ground framework conditions for the business activity as well as supporting local
businesses in various activities, such as advising clients and selling banking services,
thereby helping the banks to concentrate on their core business. Furthermore, the
Union is responsible for the risk management of the participating banks and serves as
their strategic leadership. The Raiffeisen cooperation of banks takes a leading position
among retail banks in Switzerland and its market shares have significantly increased
during the past few years. This group has the largest branch network in Switzerland.
There are in total 390 Raiffeisen banks with 1154 branches all together.
Commercial banks are usually medium size universal banks which combine
commercial and mortgage loans with brokerage and portfolio management activities.
Stock exchange banks are smaller in size in comparison to commercial and deal withonly brokerage and portfolio activities. Their balance sheet only partially reflect their
activity.
Foreign banks are financial institutions which operate under the Swiss banking
law, but their capital is owned by foreigners. These banks vary in size and activity.
For example, some qualify as universal, while other concentrate only on trade or
financial market activities. Foreign banks are mostly European in origin,
predominantly EU (50%), as well as Japanese (20%).
with private matters, such as inheritance or divorce, if the client wants to keep
information confidential.
An important aspect, which makes Swiss bank accounts very attractive for
foreigners, is that Switzerland does not collect any taxes from non-residents who open
accounts in Swiss banks. However, there are three exclusions to this rule. The first one
is Swiss withholding tax, which covers 35% of withholding tax paid on dividends and
interests by Swiss companies. The second exception is US citizens. A US citizen
cannot invest in US securities from their Swiss bank account before he/she reports it
to the IRS (Internal Revenue Service). Despite the fact that this taxation makes a
Swiss bank account less attractive for US citizens, they continue to hold Swiss
accounts, by following a specific strategy. The American clients use their US accounts
to invest in US securities and use their Swiss account in order to diversify in to the
other (non-US) investments. Finally, the third exception from the Swiss taxation is EU
residents. The clients who live in the European Union have to pay a withholding tax
on the interests from certain investments in the range between 15% and 35 % (Swiss
Banking Sector 2006).
3.4 Future of the Swiss banking system.
Recent economic trends have added pressure on banking costs, together with
regulatory changes and more exacting demands on the part of customers, they have brought about serious changes in the banking market. In addition, market structure,
product innovation, distribution organization, and changing competitiveness of
Switzerland in global economy are also leading to significant changes in the Swiss
banking industry.
According to the study «The Swiss Banking Industry in the Year 2010» (Bernet &
Monnerat), which has been conducted jointly by Accenture and the Swiss Institute of
Banking and Finance at the University of St.Gallen, some of the trends of Swiss
can be seen as a corollary of the Modigliani-Miller theorem, which states that under a
certain market price process in the deficiency of taxes, bankruptcy cost and
asymmetric information and in an efficient market the value of a firm is unaffected by
the way how the firm is financed.
4.2 Independent variables
4.2.1 Bank-specific determinants of profitability
In this work I use eight bank-specific variables in determining bank profitability:
liquidity, bank capital, credit risk, productivity growth, operating expenses, size,
density of population for the regions where banks situated, and two dummy variables
indicating cantonal and Raiffeisen banks. In order to better understand the relation
between these variables and profitability, it is useful to provide a detailed description
of each variable.
Let us begin with the liquidity variable. Banks can fail to meet their obligations
without essential liquidity. Liquidity risk for the bank is associated with an excess
demand for repayment of deposits over liquid resources of a bank. By definition,
liquid reserves consist of assets that can be converted into cash quickly without loss.
As is well known, banks’ main earnings come from accepting deposits and making
loans based on these. However, in order to be protected from the liquidity risk, banks
should keep part of their funds in the liquid assets. As a result they have less funds to
create loans and as a consequence have lower rate of return than if they used all of the
deposits to make loans. Thus, holding a larger amount of liquid assets is associated
with lower rate of return (Matthews 2008). In this work I use the ratio of liquid assetsto total assets (LATA) as a measure of liquidity. The liquidity ratio measures the
extent to which banks can quickly liquidate assets to cover short-term liabilities. In
general, the higher the value of this ratio, the larger number of liquid assets, held by a
bank, and the larger the margin of safety that bank possesses to cover short-term
debts. Consequently, larger amounts of the liquid assets are associated with lower
rates of return. A number of studies have investigated the relationship between
liquidity and profitability. For example, Molyneux & Thornton (1992) report negative
correlation between the two. In contrast, Bourke (1989) finds a positive relationship
between the level of liquidity and profitability. In explaining the results, Bourke
(1989) argues that perhaps depositors accept lower interest rates in return for greater
security provided by liquidity.
I use the ratio of equity to assets (EA) as a proxy for the bank capital strength.
Financial institutions that have a higher capital to assets ratios are expected to be
much safer in comparison with institutions with lower ratios. Moreover, the higher the
equity to assets ratio the lower is the need for external funding. Banks with low value
of the capital ratio (less equity) are riskier and have higher return (in compensation for
the greater risk), than financial institutions that are better capitalized. In their analysis
of the effects of equity to assets ratio on bank profitability, Dietrich & Wanzenried
(2009) present a two-fold argument. On the one hand, since debt is a cheaper form of
finance, one can expect that the capital ratio will affect bank profitability negatively.
On the other hand, better capitalized financial institutions are safer and therefore gain
profits even at times of economical downturn. This fact in turn increases the
creditworthiness of bank, therby reducing its funding cost and external funding. Thus,
capital ratio might also have a positive effect on profitability. As a Consequence, the
direction in which equity to assets ratio affects banks profitability is not well
determined and I hope to shed some light on this question in present work.
The next parameter I investigate is credit risk. Credit risk can be characterized as
the possibility of loss due to default of debtor (Matthews (2008)), for example when a
customer defaults on the loan or other type of financial contract. The default on loans
can take form of failure to pay interest payments or principal on maturity. In this work
I use the ratio of loan loss provisions to loans (LLPL) as a proxy of bank’s credit
quality. To evaluate this ratio, the loan loss provision was taken from the bank’sincome statement and the loans were taken from its balance sheet. Loan loss
provisions are a non-cash expense for the banks to account for possible future losses
on loan defaults. Banks expect a default on a certain percentage of loans, and record
this percentage as an expense when calculating pre-tax income. This serves as a
guaranty of solvency and capitalization for the banks in case of defaults. The value of
loan loss provision increases with the riskiness of the loans that bank makes. Banks
making small number of risky loans normally have a lower level of loan loss provision
than banks taking higher risks. Changes in the credit risk can dramatically affect the
performance of the bank and changes in its loans portfolio. Studies, which include
Duca & McLaughlin (1990) and Miller & Noulas (1997), indicate a negative effect of
credit risk on profitability, indicating that an increased exposure to credit risk leads to
a decline in the bank’s profitability. Miller & Noulas (1997) argue that the higher the
risk loans held by a financial institution, the higher is the possibility of unpaid loans,
and consequently lower is the profitability. Dietrich & Wanzenried (2009) and
Athanasoglou et al. (2005) have shown a negative correlation between profitability
and credit risk. Hence, in this work I expect a negative correlation between
ROAA/ROAE and loan loss provisions to loans ratio. However, it worth pointing out
that, banks specialising in riskier loans would in general make greater loss provision
and expect a higher return for the greater risks. In these cases there may be no relation
between actual return and loss provision, or even a positive correlation if banks are
risk averse.
Banks improve profitability by monitoring the credit risk through forecasting the
future levels of risk. Stiglitz & Weiss (1981) have shown that, in equilibrium, a loan
market may be characterized by credit rationing. Banks that make loans are interested
in two factors, the interest rate they receive on loans and their riskiness. However, the
interest rate that banks charge on loans could in turn affect the riskiness of the loans.
In particular, it can affect riskiness, by sorting borrowers whereby more riskier
borrowers deal with higher interest rate (the adverse selection effect), and also
possibly by affecting the action of borrowers making borrowers prefer riskier projects
with higher interest rate instead of the less riskier ones (the incentive effect) (Stiglitz
& Weiss (1981)). Thus, banks may not compensate an increase in risk by increasing
the rate of interest. In fact, the dependence of profitability on interest rate has the formshown on figure 2. The rising rate of interest has two opposite effects on bank loan
revenue. The first effect is positive, which means that revenue increases with rise in
interest rate. The second effect is negative, implying a decrease in the expected
revenue as the interest rate increases, due to the increase in risk of default. After the
certain point the second factor will exceed the first one and the total expected profit
diversification and increase of operational efficiency with the economy of scales, it
has been shown in a large number of works that a larger size has a positive affect on
bank profitability. However, in the case of extremely large banks, the impact of size
on profitability could be negative due to the agency cost, bureaucratic processes and
other negative aspects of large organizations. There is a large amount of literature
investigating the correlation between bank’s size and profitability. For example,
Bourke (1989) examined the correlation of bank size and profitability using 90 banks
between 1972 and 1981 in twelve countries and territories such as Australia, USA
(California, Massachusetts, New York), Canada, Ireland, England, Wales, Belgium,
Holland, Denmark, Norway and Spain. Smirlock (1985) used over 2700 unit state
banks operating in the seven-state area under the jurisdiction of the Federal Reserve
Bank of Kansas City between 1973 and 1978. Both Bourke (1989) and Smirlock
(1985) found a positive relationship between size and bank profitability. On the other
hand, Gibson (2001) who investigated profitability of European banks argued that
positive effect of large bank’s size on profitability holds only up to a particular limit.
In the present work, I use dummy variables to describe bank category. As was
described in Chapter 3, Swiss banks can be divided into several categories (big banks,
cantonal banks, regional and savings banks, Raiffeisenkassen banks, commercial
banks, consumer loan banks, stock exchange banks, other banks, foreign, and private
bankers). In the current investigation I will use only banks from the Cantonal,
Raiffeisen and savings groups. In order to specify the category to which a particular
bank belongs, I shall use two dummy variables, one for the Cantonal banks and one
for the Raiffeisen, CN and RFF , respectively. The banks for which both of the dummy
variables are zero will be considered as savings.
The last bank specific variable involved in the present analysis is density of population (DEN) for the region where the bank is situated. The functioning of
Cantonal, Raiffeisen and saving banks is limited to a geographically small area.
Switzerland is officially separated into 26 cantons that have varied area, density of
population and per capital income. All these factors may have an impact on the
profitability of corresponding banks. For instance, the cantons with higher density of
population have more bank activities than areas with less population. For this reason,
one can expect a positive correlation between density of population and profitability.
market transaction will rise during economic growth, leading to a growth in interest
margins. Consequently, revenue rise faster than cost of lending thereby increasing
profits, during periods of economic growth. Demirguc-Kunt & Huizinga (1999) found
a positive correlation between the business cycle and bank profitability. Similarly,
Athanasoglou et al. (2005) find a positive effect of GDP growth rate on bank
profitability in the Greek banking industry, although the cyclical output was found
significant for profitability only in the upper phase of the cycle.
Another macroeconomic variable that was used in the current research was
inflation. In economy inflation means a rise in the general level of prices of goods and
services over a period of time. With a rising price level, each unit of currency has the
ability to purchase fewer goods and services, and therefore inflation has a negative
impact on purchasing power of money. To measure inflation one uses the inflation
rate, which is calculated as the annualized percentage change in the general price
index. High level of inflation and uncertainty about inflation in the future discourage
investments and savings and may negatively affect the bank’s lending to the private
sector. Moderate inflation has positive aspects as well, since it mitigates the effects of
economic recession (however there are some exclusion, for example supply shocks
and stagflation of 1970s) and acts as a debt relief by reducing the level of real debt.
Inflation affects both the cost and revenue of financial institutions. The correlation
between inflation and bank profitability crucially depends on whether inflation is
anticipated or unanticipated (Perry (1992)). Anticipated inflation implies the ability of
banks and businesses to predict inflation and and therefore take steps to protect
themselves from its effects (lenders demand higher nominal interest rates). In the case
of the anticipated inflation, banks can adjust interest rate, and this can result in
revenues increasing faster than cost, leading to a positive impact on profitability.Bordes et al. (1991), investigating the profitability of anticipated inflation in banking,
show the mechanism through which anticipated inflation leads to higher profitability
of banks. In anticipation of inflation banks raise the nominal interest rate charged on
loans, leading to a rise in the equilibrium return on the bank assets relative to the
equilibrium interest that banks pay on deposits, consequently leading to an increase in
bank profitability. This, happens because banks can anticipate inflation in a better way
that their clients. Conversely, in case of unanticipated inflation banks will be slow to
Expanding equation (1) to reflect the selected variables (which are reported in the
Table 2) the models are organized as follows:
t it t t t
t t t t t t
RFF CN DEN INF
GDP TACIR LLPL EA LATAc ROAA
,,10,9,8,7
,6,5,4,3,2,1 ln
ε β β β β
β β β β β β
+++++
+++++++=
(2)
t it t t t
t t t t t t
u RFF CN DEN INF
GDP TACIR LLPL EA LATAa ROAE
,,10,9,8,7
,6,5,4,3,2,1 ln
+++++
+++++++=
γ γ γ γ
γ γ γ γ γ γ
(3)
Where ROAA and ROAE are main measure of bank profitability and are calculated as
Net Income/Total assets and Net Income/Total Equity respectively. LATA is the
liquidity variable measured as liquid assets/total assets. EA is the proxy for the bank
capital strength and is measured as the ratio equity/total assets. LLPL is ratio of loan
loss provisions to loans and was used as a proxy of bank’s credit quality. CIR is
expenses management variable, measured as cost to income ratio. TA is bank size
determinant, measured as total assets/price index, and ln TA is the logarithm of thisratio. GDP stands for the growth rate of gross domestic product, and INF is inflation
rate for the corresponding period. DEN is the density of population in the
corresponding region. CN and RFF are the two dummy variables that indicate the
bank category (Cantonal and Raiffeisen).
In the current work I also investigate models in which only bank specific factors
such as liquidity, capital, capital risk, operating expenses management, bank size,
population density in cantons and category of bank (i.e. endogenous factors) are
considered. These models are specified as follows:
Note: ROAA: Net Income /Total assets; LATA: Liquid assets/total assets; EA: Equity/total
assets; LLPL: Loan loss provision/loan; CIR: cost-income ratio; TA: Total assets/ price index;lnTA: log of total assets/price index; DEN: population density in cantons; CN and RFF:
dummy variables of category of banks – cantonal and Raiffeisen; GDP: growth domestic
product rate; INF: current period inflation rate; LM: Lagrange Multiplier.
Let us firstly begin by looking at details of data presented in Table 6. The table
reports the regression results of the estimation of model (2) (containing all variables)
and model (4) (containing only bank specific variables) for ROAA (return on average
assets) taken as the profitability variable. The first set of four columns show the
results when only bank specific variables including the two dummy variables are
considered. The final four columns contain estimation results for all variables,
including macroeconomic factors. One can notice that there are no significant
differences in results for the values of the regression coefficients and corresponding
significance level for the two models. It is also noticeable, that the explanatory power
(in terms of adjusted R ²) decreases insignificantly from 0.661469 to 0.647098 with
the insertion of additional macroeconomic factors. Thus, for the sample used in this
work, external factors do not substantially influence the overall descriptive power of
the models. However, if one employed longer time series, the difference between the
two models might become significant. In general, the relation between the additional
macroeconomic variables and performance of banks can be useful for policy
decisions. The value of Lagrange Multiplier in the BP test is 11.56 for bank specific
model and 4.54 for the all-variable model, both of which are less than the
corresponding critical values, indicating that both models do not have
heteroscedasticity.
I begin the analysis of the regression results presented on Table 6 with the
liquidity variable (LATA). The correlation coefficient for this variable is positive in
both the models indicating that LATA might have a positive effect on bank
profitability (measured by ROAA). However, the value of probability for the two
models shows that there is a 27% and 33% chance respectively, that the true value of
the liquidity coefficient is zero. This implies that liquidity does not have a statistically
significant effect on bank profitability. The standard error is comparable to the
coefficient estimate, thereby making the estimate unreliable. Therefore, the positive
value of the correlation coefficient might be a result of estimation error. On the other
hand, if the sign of the correlation is determined correctly, the result would indicate
innapropriateness of a simplistic explanation. As outlined earlier, in order to be
protected from the liquidity risk bank should keep a part of their funds in the liquid
assets, and as a result have fewer funds to create loans. Hence, a larger amount of the
liquid assets can be associated with lower rates of return. In contrast, the present
results show a positive correlation. This might be a result of the fact that better run
banks make higher profits and manage risk better, which implies holding more liquid
assets thereby reducing the chances of liquidity crises. Our findings are consistent
with the findings of Bourke (1989), who has also shown a positive correlation between liquidity ratio and profitability. Bourke (1989) analyzed the internal and
external determinants of bank profitability in twelve countries in Europe, North
America and Australia.
The next variable EA is a measure of bank capital strength (equity over total
assets). Capital ratio EA has one of the largest regression coefficients in the present
study, and therefore seems to be one of the main determinants of performance of
Swiss banks. The parameter is characterized by a relatively high significance
coefficient. The probability that true value of capital variable is zero is less than 10%.
EA has a positive effect on bank profitability in Switzerland (an increase in the equity
ratio of say 0.1 would raise profitability by 0.87 percent). The regression coefficient is
comparable for the two models (2) and (3). Moreover, the error bars are typically
smaller that the coefficients indicating a robustness in determining their signature. As
was mentioned before, financial institutions, which have higher capital to assets ratios,
are supposed to be much than institutions with have lower ratios. The obtained results
have a clear theoretical justification. Financial institutions with a reliable capital
position are able to realize business opportunities more effectively and have more
flexibility and time to deal with problems, such as unexpected losses, and therefore
can achieve an increased profitability. It is also reasonable to suppose that better
capitalized banks (because they are less risky) get access to cheaper sources of funds,
or that caution implied by high capital ratio is preserved in the loan portfolio, which
also improves the rate of profit (Bourke1989). Our findings are consistent with
Bourke (1989). Similar results were obtained by Molyneux & Thornton (1992) for
European banking industry and Demirguc-Kunt & Huizinga (1999) for large panel of
banks in 80 countries.
Let us turn to the LLPL variable, which defined as the ratio of loan loss provision
to total loans. In contrast to expectations, results show positive correlation between
credit risk and profitability. The probability that there is no correlation of LLPL with
dependent variable ROAA is just 7.3%. Therefore, this variable has a statistically
significant effect on bank profitability. The standard error (0.02508) is smaller than
the coefficient (0.04552), implying robustness in determining the signature of the
correlation (a similar conclusion follows from the value of the T-statistic (1.8153)).
The positive correlation implies that higher risks result in higher margins, andtherefore higher profits. This can be the case for the banks specialising in riskier loans
that would make greater loss provision and expect a higher return for the greater risks.
In such cases there may be no association between actual return and loss provision or
even a positive one – if banks are risk averse. Similar results were obtained by
Kosmidou et al. (2005) for the UK commercial banking industry over the period 1995-
2002. In contrast, other studies show a negative correlation between the credit risk and
profitability, implying that an increased exposure to credit risks lead to decline in a
bank’s profitability. Athanasoglou et al. (2005) show this for the Greece banking
system where, as they argue, that bank managers attempt to maximize profits mainly
through policies that improve screening and monitoring credit risk. Miller & Noulas
(1997) also show similar results for the large banks in USA. However, in case of
Switzerland, Dietrich & Wanzenried (2009) found that credit risk variable does not
have statistically significant effect on bank profitability.
As expected, CIR (cost to income ratio) appears to be an important determinant of
bank profitability in Switzerland. The numbers in Table 6 show a negative, and
statistically significant (0.0046), effect of CIR on the dependent variable ROAA. The
high t-statistic value (-2.918) shows that standard error is smaller in comparison with
the value of the coefficient, indicating a confident determination of sign of the
regression coefficient. The difference between the value of coefficients when only
bank specific variables are included in the regression (-0.0086) and all variables are
included (-0.00704) is not significant and does not change main results. The negative
correlation implies that an increase in expenses reduce the profit of the Swiss banks.
Therefore, efficient cost management is prerequisite for improved profitability of
banks. These findings are consistent with previous studies Kosmidou et al. (2005) and
Guru et al. (1999) who found a negative correlation between measure of cost and bank
performance in Greece and Australia, respectively. In addition, Kosmidou &
Pasiouras (2006) show that cost to income ratio appears to be the most significant
determinant of profitability for both foreign and domestic banks in the European
Union.
Let us next consider the bank size variable lnTA. Size of the bank was measured
by logarithm of the ratio total assets over price index in order to take in to account
changes in price every year. The results show a positive (0.03652), however nothighly significant (0.2094), effect of bank size on dependent variable ROAA.
Numerical value of the coefficient does not change significantly (from 0.03652 to
0.04138) when additional macroeconomic variables are added to the regression model.
The value of t-statistics (1.2656) shows that the standard error is smaller than the
value of the coefficient and therefore it is likely that the sign of the coefficient was
determined appropriately. The large size might result in economy of scale that will
reduce the cost of accumulation and processing information. Moreover, larger banks
have a higher degree of product and loan diversification compared to the small banks.
This is especially true for the banks considered in this work: small savings, cantonal
and Raiffeisen banks, where it is not necessary to take in to account negative aspects
of large financial organizations, such as bureaucratic processes. Similar results were
obtained by Smirlock (1985) in his investigation of over 2700 unit state banks in USA
between 1973 and 1978.
The population density DEN of the canton where each bank situated has a positive
and significant effect on bank profitability in Switzerland as measured by ROAA. The
value of coefficient (7.6E-05) and significant level (0.1454) stay completely
unchanged with the addition of macroeconomic variables to the regression model. The
t-statistic shows the standard error is smaller than the value of the coefficient, thereby
indicating that the sign of the correlation was determined correctly. A positive
correlation between population density and profitability implies that with an increase
in population of a canton the number of deposits, loans and operations with the banks
increase, which leads to a rise in profitability.
The two dummy variables CN and RFF indicating the bank category do not show
a statistically significant correlation with bank profitability. These results may be a
result of narrow sampling. In current work I have used only small saving banks
situated in the rural areas and cantons. However, Dietrich & Wanzenried (2009) who
investigate 453 commercial banks in Switzerland, including the two giants, Credit
Swiss and UBS, also found no significant correlation between bank category and
profitability.
Let us now turn to the external factors related to the macroeconomic environment
in Switzerland. The GDP growth rate (GDP) shows a negative (-0.00200), but not a
statistically significant, impact on bank profitability measured in the return on averageassets ROAA. The standard error (0.00975) is significantly larger than the regression
coefficient (-0.00200), implying a significant uncertainty in determination of the effect
(a similar conclusion follows from the small value of the T-statistic (0.20554)). On
first thoughts, as was discussed in Chapter 4, one might expect a positive correlation
between the GDP growth rate and bank profitability. However, current results do not
provide substantive support to the arguments that the business cycle impacts on the
performance of banks. This could be seen as a result of narrow sampling, both in
the obtained value is comparable to the mean ROAA/ROAE, one concludes that the
given variable has a significant economic effect on profitability.
Table 9. The calculated value of variables for determination of economic
significance.
Variable Calculated value Mean ROAA
CIR 0.072 0.43
DEN 0.33; 0.0044 0.43
EA 0.2 0.43
lnTA 0.1 0.43
INF 0.018 0.43
The results show that cost to income ratio CIR has a value of the statistic (0.072)
that is significantly lower than the mean ROAA (0.43). Therefore, one can conclude
that despite statistical significance, there is no sizable economic effect on bank
profitability due to CIR. The result shows that the increase of cost efficiency does not
lead to the significant increase in bank profitability.
A similar quantity was calculated for the population density variable DEN.However, instead of multiplying the regression coefficient into standard deviation, it
was multiplied into the maximum and minimum value of density in the considered
Cantons, respectively. The obtained value for most dense Canton is 0.33 and for the
least dense is 0.0044. Comparing with mean of ROAA (0.33), one can conclude that
population density variable is an economically significant. Thus, the density of
population has a significant influence on profitability. In other words, the profitability
of saving banks varies significantly depending on which region or Canton they are
situated in. This can be explained by dispersion bank efficiency and different revenue
generated in different regions.
The equity to assets variable EA shows a value 0.2. Comparing with mean ROAA
(0.43), I conclude that equity to assets variable is not only statistically but also
economically significant and has considerable positive effect the bank profitability. As
was mentioned previously, this is a direct reflection of the fact that financial
institutions that have a higher capital to assets ratio are expected to be much safer
thereby reducing their funding cost and costs of external funds.
Bank size variable lnTA gives a value 0.1, which implies a marginal economic
significance. The sign and the value of the correlation coefficient indicate that bank
size has a positive effect on bank profitability. This conclusion is in direct
correspondence with expectations based on economy of scale, as was mentioned
previously.
Finally, inflation yields a low value (0.018) for the statistic, in comparison with
the mean of ROAA (0.43). Therefore, one should conclude that despite the
statistically significant correlation, inflation does not have an important economic
impact on bank profitability. This result may be interpreted as a result of narrow
sampling. The current work concentrated only on small savings banks (Cantonal,
Raiffeisen and regional savings). These banks are not significantly influences by
macroeconomic changes, in the same manner as the big commercial banks.
To sum up, the present results show that only some of the chosen variables have
significant economical effect on profitability of small savings banks in Switzerland.
The main variables that have a significant economic impact on profitability are EA,
DEN and lnTA. All of these variables show a positive impact on bank profitability.
Thus, current study implies that capital (equity to assets), bank size (total assets) and
population density in the region where the bank is operating are the main determinants
of bank profitability for small savings banks in Switzerland. On the other hand, CIR
and INF variables, although statistically significant, do not show a strong economic
impact on bank profitability. Finally, LATA, CN, RFF and GDP do not show a
statistically significant correlation with bank profitability. The absence of statisticallysignificant correlation might be a result of narrow sampling, both in terms of the
number of banks considered and their categories, as well as a short time span of
Al-Muharrami S., Matthews K. and Khabari Y. 2006. Market Structure and Competitive
Conditions in the Arab GCC Banking System. Cardiff Business school working paper series .
Akhavein, J.D., Berger, A.N., Humphrey, D.B., 1997. The effects of megamergers onefficiency and prices: evidence from a bank profit function. Finance andEconomic Discussion Series 9, Board of Governors of the Federal ReserveSystem.
Athanasoglou, P.P., Brissimis, S.N. and Delis M.D. 2005. Bank-specific industryspecificand macroeconomics determinants of bank profitability. Journal of International Financial
Markets, Institutions and Money, Vol. 18, No. 2, pp. 121-136.
Athanasoglou P.P, Delis M. D. and Staikouras C. K,. Determinants of bank profitability in thesouth Eastern European region. Working paper № 47, available at:
http://www.bankofgreece.gr/BogEkdoseis/Paper200647.pdf , [Accessed: 1 September 2009].
BankScope database
Barth, J.R., Caprio, G., Levine, R., 2004. Bank supervision and regulation:Whatworks best? Journal of Financial Intermediation, 13 (2), 205-248.
Berger, A.N., Hanweck, G.A., Humphrey, D.B., 1987. Competitive viability inbanking: Scale, scope and product mix economies. Journal of Monetary
Economics 20, pp. 501-520.
Berger, A.N., 1995(a). The profit - structure relationship in banking: Tests of marketpowerand efficient-structure hypotheses. Journal of Money, Credit, andBanking 27, pp. 404-431.
Berger, A.N., 1995(b). The relationship between capital and earnings inbanking. Journal of Money, Credit, and Banking, 27, pp. 432-456.
Berger, A.N. and Hannan, T.H., 1989. The price-concentration relationshipin banking.Review of Economics and Statistics 71, pp.291-299.
Bikker, J.A., Hu, H., 2002. Cyclical patterns in profits, provisioning andlending of banks and procyclicality of the new Basel capital requirements. BNLQuarterly Review 221, pp.143-175.
Bordes C., Goyeau D., Melitz J., Sauviat A. 1991. The profitability of anticipated inflation in
banking, Economics Letters 37 , pp. 57-60
Bourke, P., 1989. Concentration and other determinants of bankprofitability in
Europe, North America and Australia. Journal of Banking and Finance 13,pp. 65-
Brooks C. 2008. Introductory Econometrics for Finance, Cambridge University press, pp.490-492.
Coccorese P. 2004. Banking competition and macroeconomic conditions: a disaggregate
analysis. Journal of International Financial Markets, Institutions and Money.Volume 14, Issue3, pp. 203-219.
Cowling K. and Waterson M. 1976. Price-Cost Margins and Market Structure.
Economica 43 (1976),141 pp. 267-274.
Dansby, E. and Willig, R. 1979. Industry Performance Gradient Indexes, AmericanEconomic Review, 69, pp. 249-60.
Demirguc-Kunt, A.and Huizinga, H., 2000. Financial structure and bank
profitability.World Bank Mimeo.
Demirguc-Kunt, A. and Huizinga, H., 1998. Determinants of commercialbank interestmargins and profitability: some international evidence. World Bank EconomicReview 13, pp. 379-408.
Dietrich A. and Wanzenried G.2009. What Determines the Profitability of Commercial Banks?
New Evidence from Switzerland, Working Paper Series, available at:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1370245 , [Accessed: 1 September
2009].
Duca, J. and McLaughlin M. 1990. Developments affecting the profitability of
commercial banks. Federal Reserve Bulletin, July.
Eichengreen, B. and Gibson, H.D., 2001. Greek banking at the dawn of thenewmillennium. CERP Discussion Paper 2791, London.
Frame W.S. and Kamerschen D.R.1997. The Profit-Structure relationship inlegally protected banking market using efficiency measures. Review of Industrial Organizations 12, pp 9-22
Gibson, H.D., 2005. Developments in the profitability of Greek banks. Bank of Greece
Economic Bulletin 24, pp.7-28.
Goddard, J., Molyneux, P., Wilson, J.O.S., 2004. The profitability of Europeanbanks:a cross-sectional and dynamic panel analysis. Manchester School 72 (3),pp. 363-381.
Guru B.K., Staunton J., Balashanmugam B., 1999 Determinants of commercial bank profitability in Malaysia. In : Paper presents at the Proceedings of the 12 th Annual Australian
Finance and Banking Conference, Sydney, Australia, December 16-17.
Haslem, J.A., 1968. A statistical analysis of the relative profitability of commercial
Hassan M.K. and Bashir A.-H.M. 2003. Determinants of Islamic banks profitability. In: Paper
presents at the Proceedings of the Economic Research Forum (ERF) 10 th Annual Conference,
Marrakesh-Morocco, December 16-18.
Matthews K., and Thompson J. 2008, The Economics of Banking , John Wiley & Sons, LTD,Second edition.
Miller, S. and Noulas A. 1997. Portfolio mix and large-bank profitability in the USA.
Applied Economics 29, pp. 505-512.
Molyneux, P. and Thornton, J., 1992. Determinants of European bankprofitability: Anote. Journal of Banking and Finance 16, pp. 1173-1178.
Molyneux P.1993, Structure and performance in European banking. DoctoralDissertation, University of Wales, Bagnor.
Novshek W. 1980. Cournot equilibrium with free entry. In Review of Economic Studies, vol.
47, pp. 473-486.
Panzar, J. and Rosse J. 1987. Testing for monopoly equilibrium. Journal of Industrial
Economics 25, pp. 443-456.
Panzar, J.C. and Rosse J.N. 1982. Structure, conduct and comparative statistics. Bell
Laboratories Economic Discussion Paper.
Pasiouras, F., and Kosmidou, K., 2006. Factors influencing the profitability of domestic andforeign commercial banks in the European Union. Research in International Business &
Finance, 21 (2), pp. 222-237.
Pasiouras, F., Tanna, S., Zopounidis, C., 2007. Regulations, supervision and banks’ cost and profit
efficiency around the world: a stochastic frontier approach. available at:
http://www.business.bham.ac.uk/events/papers/F_Pasiouras.pdf [Accessed: 1 September 2009].
Pasiouras F, and Kosmidou K., 2008. Determinants of profitability of domestic UK commercial banks:
panel evidence from the period 1995-2002 . Economic, Finance and accounting, Research working
Short, B.K., 1979. The relation between commercial bank profit rates andbankingconcentration in Canada, Western Europe and Japan. Journal of BankingandFinance 3, pp. 209-219.
Smirlock, M., 1985. Evidence on the (non) relationship betweenconcentration andprofitability in banking. Journal of Money, Credit, and Banking 17, pp. 69-83.
Stiglitz J.E., and Weiss A. 1981. Credit Rationing in Markets with Imperfect Information, American Economic Review 71, pp. 393-410.
Trading Economics (Global Economic Research) web page, available at:
Bernet B. and Monnerat B.R. The Swiss Banking Industry in the Year 2010, Financial
Services | Banking, available at: http://www.accenture.com/NR/rdonlyres/42D43E61-3E43-4B82-A0AA-B7BF6B6F4AFB/0/SwissBanking2010_e.pdf , [Accessed: 1 September 2009].
Bichsel R., Blum J. 2002 The Relationship between Risk and Capital in Swiss Commercial
Banks: A Panel Study, Working Paper 02/04, Swiss National Bank,.
Bikker J. A. 1999 Efficiency and the European Banking Industry: An Exploratory Analysis toRank Countries, De Nederlandsche Bank Staff reports, No. 42.
Compendium Edition 2006. Swiss Banking Sector . available at:
http://shop2.sba.ch/11116_e.pdf , [Accessed: 1 September 2009].
Economic profile of Switzerland, available at:http://www.mydeltaquest.com/english/member_content_article.php?article_id=124,
[Accessed: 1 September 2009].
Egli D., Rime B., 1999, The UBS-SBC merger and competition in the Swiss retail banking
sector, Studienzentrum Gerzensee Working Paper , № 00/02, December.
Freuler, B., 2005. Ausmass und Ursachen des Effizienzgefälles im europäischenBankensektor: Eine parametrische und nichtparametrische Untersuchung, Dissertation
University of Basel , Basel.
Neuberger, D., Schacht, C., 2005. The Number of Bank Relationships of SMEs: ADisaggregated Analysis for the Swiss Loan Market. Thuenen-Series of Applied Economic
Theory 52, University of Rostock, Institute of Economics.
Neuberger, D., Pedergnana, M., Räthke-Döppner, S., 2008. Concentration of BankingRelationships in Switzerland: The Result of Firm Structure or Banking Market Structure?
Journal
of Financial Services Research 33 (2), pp. 101-126
Rime, B., Stiroh, K., 2003. The performance of universal banks: Evidence from Switzerland.
Journal of Banking and Finance 27 (11), pp. 2121-2150.
Rime B. and Stiroh K.J, 2001. The performance of Universal Banks: Evidence from
Switzerland. Working Paper No. 01.03, available at:http://www.szgerzensee.ch/fileadmin/Dateien_Anwender/Dokumente/working_papers/wp-0103.pdf , [Accessed: 1 September 2009].
Sheldon G., Haegler U., 1993, “Economies of Scale and Scope and Inefficiencies in Swiss
Banking,” in eds. Blattner, Genberg, and Swoboda, Banking In Switzerland , 103-134,Heidelberg: Physica.
Sheldon G. 1994, Economies, Efficiencies and Technical Progress in Swiss Banking, in eds.
D. E. Fair and R. Raymond, The Competitiveness of Financial Institutions and Centres in
Europe, 115-132, The Netherlands: Kluwer Academic Publishers.