Comparative Analysis of Banking System Profitability in Western Balkan Countries

The aim of this paper is to analyze the profitability of the banking sector in the Western Balkan countries. (Note 1) This paper reviews return on assets (ROA) as an indicator of profit and return on equity (ROE) as an indicator of profitability in the banking systems of the respective countries, as well as some other macroeconomic variables that influence them. The main objective of this study is to identify the specific and macroeconomic variables of this industry, that have an impact on the profitability of commercial banks operating in the Western Balkan countries during the 2008-2015 period. Specifically, this paper addresses external indicators (gross domestic product, remittances, foreign direct investment, unemployment), and industry and bank specific indicators (assets, loans, loan-to-deposit ratio, non performing loans and interest rates) that affect the profitability of the banking system in respective countries. Therefore, according to the data generated during the research and the literature review, the profitability of banks measured by the ROA and ROE indicators, regarding the analyzed countries, turns out to be extremely low, especially compared to EU countries where they strive.


Introduction
Financial sector is vital to the economy of any country, with banks comprising most of the sector, especially in developing countries, where the capital market is not strong enough (Matthew & Laryea, 2012). During their operation as intermediary institutions, banks enable the transfer of financial assets from surplus units to deficit ones. Through this process they generate profit. Hereinafter, the paper focuses on the profitability of the banking sector in the Western Balkan countries, as well as the indicators that affect it during the 2008-2015 period. All these countries have undergone a transformation process from a centrally planned economy to an open market economy. All of them, apart from Albania, are from ex-Yugoslavia and have experienced a very painful process of separation with plenty of serious consequences. Today, all these countries aspire to join the EU and are at different stages of the accession process. Financial systems of all these countries are dominated by commercial banks so it can be concluded that these countries have banking-based (bank centric) financial systems. This further highlights the importance of addressing the profitability of the banking system in these countries. The purpose of the paper is to identify and analyze the various indicators that impact the profitability of the banking sector in the six Western Balkan countries.
Indicators of profitability have been debated for years in different countries and still there are no universally accepted findings for these variables. This is due to the fact that countries have different economic, political, financial, as well as business environments.
According to Gungor (2007), the micro variables include bank size, capital, risk management, expenditure management, securities and non-performing loans, while the macro variables include inflation rate, interest rate, GDP growth, and the rate of taxation. Molyneux and Thorton (1992) were among the first to analyze the banks' profitability in 18 European countries for the 1986-1989 period. During their analysis, they found a positive correlation between return on equity (ROE) and interest rates in each country.
Similarly, Bourke (1989) was amongst the first to study the impact of internal indicators in banking performance: net income before and after taxation compared to total assets, as well as indicators of capital and bank reserves, which have a high impact on profitability indicators.
Subsequently, Demirgüç-Kunt and Huizinga (1999) analyzed the banking sector in 80 different countries, using regression analysis. These authors find a positive correlation between the inflation rate and the bank's profitability. Kosak and Zajc (2006) evaluate the determinants of efficiency in the new EU member states. They find that ROA and ROE have a positive relationship with the efficiency On the other hand, Saeed (2014), while analyzing banks in UK concludes that the inflation rate negatively impacts the bank's profitability. Ramlall (2009) and Sufian (2009) find a positive relationship between bank size and profitability. The larger the bank, the greater its profitability and vice versa. This in fact illustrates an example of economies of scale. Contrary, Hannan andPrager (2009), Kosmidou (2008) and Luo (2003) argue that large banks and profitability have a negative relationship and that small banks can generate more profits because they have lower spending and more efficient performance.Deietrich and Wanzenreid (2010) study the main determinants of profitability for the Swiss banking industry. In their empirical study, they used data from 453 banks, for the observation period 1999-2008. Generated results show that well-capitalized banks are more profitable. Whereas, Curak et al. (2011) applying the Dynamic Panel Analysis method in 16 banks from the Macedonian banking system, in the 2005-2010 period, examines the internal as well as external indicators that influence the banking profitability in this country.
The following section analyzes the macroeconomic and microeconomic factors that affect the performance of banking systems in the Western Balkan countries in the 2008-2015 period.

Macroeconomic Factors
The macroeconomic indicators that will be examined are GDP, inflation, unemployment remittances and FDI, for the 2008-2015 period, on each respective country. The following section reviews the annual increase of GDP (%) in Western Balkan countries during the 2008-2015 period.

GDP (annual, %)
This section presents the GDP increase, in Western Balkan Countries,

Inflation
The following section analyzes the inflation rate in WB countries, throughout the years.

Unemployment
Unemployment is a major problem in all WB countries, especially, in Kosovo (2015: 32.9%), Macedonia (2015: 26.1%) and B &H (2015: 27.6%). This section presents the unemployment rate in these countries. The average unemployment rate for this country was 14.94%, which in fact was 9.52% lower than the average unemployment rate for the entire comparative period for all other WB countries.

Remittances
Remittances as an indicator are unique in the way they impact the economies of the WB countries. Remittances in all WB countries are mainly used for consumption and less for investment. On average, for the analyzed countries, the highest level of remittances is injected into the economy through unofficial channels. The level of remittances injected into the economy through official channels of WB countries (as% of GDP), is presented below: From the above data, it can be noted that the largest percentage of remittance growth over the years, in report to GDP, is reflected in Kosovo with an average growth of 16.09%, closely followed by Albania with 11.54%, and B and H with 11.51%. On the other hand, countries experiencing a single digit remittances growth, for the comparative period, were Serbia with an average of 9.01% and Macedonia with an average of 8.62%. A low remittance growth in report to GDP is reflected in Montenegro with an average of only 3.78% for the entire comparative period.

Foreign Direct Investment (FDI) Net Inflows (% of GDP)
FDI as percentage of GDP are presented in the following table: From the table above data it can be concluded that FDI reflect diversity in Western Balkan countries. The largest average FDI inflow (as% of GDP) is reflected in Macedonia at a rate of 17.81% for the entire comparative period 2008-2015, which is 9.25% above the average of the respective countries in the comparative period. This is followed by Albania, with an average FDI inflow for the entire comparative period at 9.10%, which is 1.54% higher than the average of the respective countries for the 2008-2015 period.
Kosovo and Serbia have roughly the same average growth over the entire comparative period, at a rate of 6.35% and 6.06% respectively, which is 1.2% and 1.5% lower than the average for all WB countries.
The lowest average growth in FDI inflow for the whole period is reflected in Montenegro with a growth rate of 3.46% and B&H with only 2.56% average growth over the entire comparative period 2008-2015.

Banking System Assets
The participation of banking system assets in GDP in WB countries (in %) is presented in the table below:

Loans in Banking System
Lending is a very important profitability indicator in the banking systems of the WB countries, as it generates most of the revenues. Therefore, the quality of approved loans is a very important indicator of banking performance, which naturally also affects the profitability of the banking system. Problems presented as a result of delays in loan repayment not only affect the performance of the banking sector, but also have a major impact on the economy.
The following section presents the participation of loans in GDP, in WB countries (as %). participation in GDP is also reflected in B&H at 60.99%, which is above the average of the comparative period for all countries by 11.22%. Loans (as % of GDP) in Serbia were at the 50.20%, which is only 0.53% above the average of the region for the comparative period.
While, other countries had lower loan rates (as % of GDP) than the average of the entire region for the comparative period, Albania with 39.81%, Macedonia with 35.38% and Kosovo with 34.16%.

Lending-Deposit Interest Rate Spread
The interest rate spread which represents gross profit, is an important indicator which largely impacts on banks' profitability. Therefore, the following section presents the interest rate spread of the respective countries. 6.40 5.93 6.26 6.33 6.08 6.14 Source: The Global economy.com The IMF. Author's calculation. Annual reports of National banks of respective countries, by year. Author's calculation.

Participation of NPL in Total Loans of WB Countries (as %)
Throughout the years NPL as an indicator has a great importance in addressing the profitability of the banking system. The participation of NPLs in the total of approved loans is presented below: Macedonia, NPL also experienced an increase from 6.7% in 2008 to 10.9% in 2013. The average NPL participation on loans in this country for the comparative period is 9.56%, which is 3.4% below the average of the region.
The lowest level of NPL participation on loans is in Kosovo, where their rate varies from 3.4% in 2008 to 8.5% in 2013. The average NPL for the comparative period is 6.25%. On this regard, Kosovo, is the only country from Western Balkans, where NPLs are at a single digit value. Compared to the average of the region, for the entire comparative period, the level of NPLs in Kosovo is lower by 6.71%. This mainly due to the fact that banks in Kosovo have governed very conservative policies as well as enormous demand for collateral, which in certain cases exceeded double the loan value.

Bank Specific Indicators
The performance of banking systems in the Western Balkans countries for the 2008-2015 period is analyzed through two key indicators, return on assets (ROA) and return on equity (ROE).
The fluctuations of these two indicators over the years for the respective countries are presented in this section:

Return on Assets (ROA) and Return on Equity (ROE)
Banks' profitability, measured through two key indicators (ROA and ROE), displays a heterogeneity in all WB countries that conduct business in an extremely difficult business environment. The following table illustrates the fluctuations of these two indicators for all WB countries over the years. From the above data it can be concluded that: In Albania, ROA varies between 0.22% in 2015 and 1.68% in 2008. The average value for the analyzed period for this country is at 0.93%, which in fact is 0.27% higher the average for all Western Balkan countries. On the other hand, ROE varies between 1.77% in 2015 and 19.95% in 2008. The average value for ROE during the comparative period in Albania is at 8.95%, which is 3.35% higher the average of the region.
In B&H, ROA varies between -0.46% in 2010 and 0.97%. The average for the comparative period is 0.42 % which is below the average of the region for 0.24%. While ROE, in this country varies between -4.02% in 2010 and 6.89% in 2012. The average for the analyzed period is 3.52% which is lower than the average of the region for 2.08%.
Performance indicators in Kosovo display better values compared to all other WB countries for the comparative period. Thus, ROA display a positive value and varies between 0.70% in 2012 and 2.90% in 2015. The average for the comparative value is 1.64% which is approximately 1% higher than the average of all WB countries.
On the other hand, ROE, in Kosovo displays satisfactory performance and for the comparative period it varies between 7.2% in 2012 and 26.4% in 2015. The average for the analyzed period is 15.7% which is higher than the average of the region by 10.10%.
The performance indicator, ROA, in Macedonia varies between 0.40% in 2011 and 1.72% in 2008. The average for the period is 0.25% which is 0.29% higher than the average of the region. While ROE varies between 3.43% in 2011 and 13.08% in 2008. The average for the comparative period is 7.62% which is 2.02% higher than the average for all WB countries.
In Montenegro, ROA displays poor performance and varies between -4.34% in 2010 and 0.77% in 2014. The average for the analyzed period, for this country is negative at -0.66%, which is below the average of the region by 1.32%. Furthermore, even ROE displays poor performance and varies between -23.36% in 2010 and 5.52% in 2014. The average for the comparative period is 5.06% which is 10.66 lower than the average for all WB countries.
Meanwhile in Serbia, ROA varies between the interval of -0.19% in 2014 and 1.78% in 2008. The average for the period is 0.71 which is above the average of the region by only 0.05%. ROE on the other hand, varies between -1.99% in 2013 and 7.32% in 2008. The average for the comparative period is only 2.88% which is in fact below the average of all WB countries for this period by 2.72%.
From the above analysis it can be concluded that both performance indicators (ROA and ROE) in Western Balkan Countries, for the comparative period (2008-2015) on average have shown better performance in the period before global financial crisis than the period after it and that some of these countries still have to fully recover from the crisis in question. Based on performance indicators (ROA and ROE) the banking system in Kosovo has shown the best results.
A few of the reasons why this is the case are explored in this paragraph. Kosovo has had little to no access in international markets. Commercial banks in this country have governed with very conservative policies. Collateral requirements in this country have surpassed double the value of the amount requested for loan. Furthermore, other than the borrower two other guarantors were required for any approved loan, an issue that burdens the client but secures the high value approved loans.

Conclusions and Recommendations
Financial systems in the WB countries are dominated by banks located in most of the countries with foreign capital. The assets of foreign-owned banks in these countries vary between 70-90% of total banking sector assets in these countries. The profitability of banks in WB countries, which carry out their business in a challenging environment, is addressed through two key indicators (ROA and ROE). Based on the literature review as well as the data analyzed in the Western Balkan countries, for the observed period 2008-2015, using the comparative method, it can be concluded that these indicators differ and are dependable on the variations of a number of internal and external variables. Therefore, for a specific period of time, some variables for a particular group of banks, in one country, may have a high statistical significance over ROA and ROE, while for the other period it may not be significant at all.
From the analysis so far and as the literature also suggests, it can be concluded that non-performing loans have high statistical significance with a negative relationship on bank profitability. The high level of NPLs in most of these countries continues to jeopardize the profitability of the banking system.
The average values for profitability indicators in Albania for the comparative period 2008-2015, are 0.93% for ROA and 8.95% for ROE. In BiH, the average for of the comparative period are 0.42% for ROA, while ROE is at 3.52%. Performance indicators in Kosovo show the highest values in comparison to all other WB countries. The average ROA, for the comparative period is 1.64%, while ROE is at 15.70%. In Macedonia, ROA for the analyzed period is at 0.95%, ROE on the other hand is at 7.62%. In Montenegro, performance indicators are as follows: ROA with -0.66% and ROE with -5.06%. In Serbia, the average for the analyzed period is 0.71% for ROA and 2.88% for ROE.
Based on the conducted analyzes and also as the literature suggests, non-performing loans and reserve provisions have high inverse statistical significance on banks profitability. Countries with high level of NPL display lower values for profitability indicators and vice versa. Therefore, as a conclusion, the generated results indicate that the banking systems of WB countries for the 2008-2015 period have been characterized by many fluctuations in terms of performance, business risk management and problems related to non-performing loans.
Based on the so far analysis, regarding the banking systems of all Western Balkan countries, specific recommendations can be given, as follows: -Improvement in legal infrastructure so that execution of collateral for bad loans is easier and faster. This would enable banks to access new financial means for new loans and would enable revenue generation, thus also influencing the profitability of the banking system of the respective countries.
-Further distribution of loans into different sectors would lower the risk exposure of banks, since in the majority of WB countries, the sectors which benefit most of the loans are trade and construction, which carry higher business risk. Therefore a diversification of loan portfolio is required in order to lower the risk exposure to bad loans from one sector only.
-To improve human and technical capacities should as well as the structure of business risk management, thus of credit risk.
-Expanding the range of banking products and services would increase revenue generating sources, thus also impacting the bank's profitability.
-Further supervision of banks would affect the management of commercial banks to behave cautiously over business risk exposure.
-Commercial banks should consider the possibility to diversify their investment portfolio, especially in financial derivatives, which would enable the improvement of profitability indicators.
-The utilization of best practices from banking systems of developed nations would also affect the improvement of profitability indicators.