Many years ago Islamic Finance was an unknown system; but now it has expanded to become a distinctive and fast growing segment of the International Financials markets. Islamic Finance in general and Islamic banking in specific become main players in the financial world. More than 200 Islamic Banks operate in over than 70 countries concentrated in the MENA region and many western countries (Hassan & Lewis, 2007). Islamic Banking became popular during the financial crisis started in 2008 because they were not affected directly by collapse in the credit markets. According to the Banker (2009), the assets held by Islamic banks (including the Islamic windows of conventional banks) increased to $ 822bn from $639bn in 2008, an annual growth rate of 28.6% compared to a more modest 6.8% for conventional banks.
Features of Conventional banking
The conventional banking theories assume that banks earn profits by purchasing deposits from the depositors at a low-interest rate, then reselling those funds to the borrowers at higher interest rate, based on its competitive advantage at gathering information and underwriting risk (Hassan T., Mohamad S.& Bader M.K., 2009). Therefore, conventional banks make profits from the spread between the interest rate received from borrowers and the interest rate paid to depositors. The most well-known approaches to explain banking function process are the production and intermediation approaches. In the production approach, banking activities are described as the production of services to depositors and borrowers. The intermediation approach views bank as an intermediaries of financial services and assumes that banks collect funds (deposits and purchased funds with the assistance of labor and capital) and transform these into loans and other assets. The deposits are treated as inputs along with capital and labor and the volumes of earning assets are defined as measures of output.
Features of Islamic Finance
Islamic banking performs the same intermediary function but does not receive a pre-determined interest from borrowers and does not pay a predetermined interest to the depositors; the amount of profits is based on the profit-sharing agreements with the depositors and also with the borrowers (Hassan T., Mohamad S. & Bader M.K., 2009). Islamic financial principles have evolved on the basis of Shariá law, which forbids payment or receipt of Riba - the payment or receipt of interest (Obaidullah, 2005). Financing principles are governed by Islamic rules on transactions "Figh Al-Muamelat" and follow both profit and loss sharing (PLS) and non-PLS arrangements (such as leasing contracts). Freedom from Al-Qimar (Gambling) and Al-Maysir (Unearned Income) In addition to the prohibitions on interest, Islamic banks also face other restrictions - such as the use of many derivatives products, because according to Shariá all contracts should be free from excessive uncertainty "Gharar" (Obaidullah, 2005).
Table (1). A Comparison of Islamic and Traditional Banking
Literature Review
To start our research, we appealed to some articles in newspapers and journals of erudition professionals in banking sphere. A lot of professors have a very significant look on differences in Islamic and Conventional banking sphere and their profitability. We found very interesting articles about different measures of profitability, efficiency and risk between Islamic and Conventional banks.
For the Saudi Arabia, Obaidullah (2005) describe the products, processes and mechanisms that are using in the Islamic financial services industry. The text also focuses on how financial products and services should be designed and offered in this industry, given the need for full Shariah compliance. Instead of presenting facts and figures that quickly become obsolete, the Obaidullah (2005) describes in his book how financial products and processes develop as solutions to problems and as responses to profit opportunities and need for Shariah compliance.
For the same country, Rashwan (2010) estimate the differences in Business framework, Interest charging, Interest on deposits, Zakat (religious tax), Risk sharing in equity financing, Customer relationships. Also we used information from his article about MANOVA. We knew that MANOVA is designed to analyze the relation between different dependent variables (outcomes) simultaneously and that's why it is called multivariate. It has the power to detect whether groups differ along a combination of dimensions and it is a two stage test in which an omnibus test is first performed before more specific procedures are applied to tease apart group differences. To make a statistical data, he used MANOVA, which is calculated from complicated model. The selected study period is 2007, 2008 and 2009. These years are the dependent variables (DVs) for each of the four indicators: ROAA, ROAE, NL/TA and LLR/TL. The independent variable (IDV) is constructed as dummy variable where IBs and TBs were assigned 1 and zero respectively. All the variables are standardized using the Z-scores to have a normalized pattern to satisfy MANOVA assumption that each dependent variable should be a multivariate normal.
Over the last decades, empirical research on bank efficiency in Arabic countries appears relatively scarce. During the last 10 years, the concept of Islamic banking has likewise developed to cover activities of other types of financial institutions including insurance, investment and fund management companies. Moreover, to take advantage of Islamic financial instruments, many conventional banks in GCC countries have added Islamic banking services to their regular banking operations. (Srairi, 2009). In response to globalization and deregulation, decision makers in GCC countries over the past decade have implemented various measures to enhance the credibility o the banking sector and improve its performance and efficiency. According to Srairi (2009), we compare the efficiency levels of banks between country and type of bank (conventional versus Islamic banks). Finally, we read about how he used the model of Battese and Coelli (1995) to estimate the sources of inefficiency. Using a Tran slog function with three input prices, two outputs and eight country-level variables, he find that the price coefficients of the cost function are all positive and the elasticity of the cost of labor is greater than the elasticity of the cost of fund. This suggests that banks in GCC countries should control personnel expenses more than interest expenses in the case of the increase of prices. Regarding country-specific factors, GCC banking efficiency with high levels of per capita GDP and degree of concentration seem to be associated with higher banking costs. In contrast, banking systems with higher ratios of capital to total asset, loan to deposit, density of demand and population tend to have lower costs. In addition, country-level that increase profit efficiency are per capita GDP, financial depth, capital ratio, and degree of monetization and concentration.
Existing study (Hassan, Mohamad, & Bader, 2009) in this area are classified into two groups. The first group includes studies that assess the performance of Islamic banks using traditional financial ratios. The second group of studies focuses on banks' efficiency and utilizes frontier analysis approaches rather than traditional financial ratios. In their article Hassan, Mohamad, and Bader (2009) had shown argument of 2 approaches. To collect the data they used the efficiency measure tool, which is presented by Data envelopment analysis. It is used to empirically measure productive efficiency of decision making. Efficiency can be measured by using a weighted average of the outputs and a weighted average of inputs. The results in article Hassan, Mohamad, and Bader (2009) indicate that there is a slack in the usage of resources across all banks, as measured by the efficiency results of the average bank.
To measure profitability according to Izhar and Asultay (2007), they chose Augmented Dickey-Fuller test. This study has attempted to empirically investigate the determinants of profitability in the case of an Islamic bank. Regression analysis was applied to examine which variables are actually significant in determining the profit of an Islamic bank, in this case, Bank Muamalat of Indonesia. According to Izhar and Asultay (2007), their results from Dickey-Fuller test shown that the negative relationship between loans over total assets and profitability indicator as found in this article (Izhar & Asultay, 2007) indicates that the Islamic bank portfolio is heavily biased towards short-term trade-based financing loans. As such, these loans are low risk and only contribute modestly to bank profits.
In the paper that we used (Abedifar, Molyneux, & Tarazi, 2011) we found how to investigates risk and stability features of Islamic banking using a sample of 553 banks from 24 countries between 1999 and 2009. In article (Abedifar, Molyneux, & Tarazi, 2011) they describe that Islamic banks that are small, leveraged and based in countries with predominantly Muslim populations have lower credit risk than conventional banks. In terms of insolvency risk, small Islamic banks also appear more stable than similar sized conventional banks; however, we find no significant difference between large Islamic and conventional banks. Results, which we analyze Abedifar, Molyneux, and Tarazi (2011) on interest rate risk show that interest income and expense, as well as loan quality, of Islamic banks are less responsive to domestic interest rates; however, the sensitivity of the stability of Islamic banks to domestic interest rates is not significantly different. Thanks to article we understand how estimate the risk exposure that include credit risk, insolvency risk and interest rate risk. The first is credit risk analysis ratios. It is loan-loss reserves/gross loans (LLRGL) - proxy for credit risk (Loan Risk). As a robustness check: Impaired loans/gross loans (ILGL) and the ratio of loan-loss provisions / average gross loans (LLPAGL). Second one is insolvency risk analysis. For insolvency risk analysis, we employ the Z-score measure which is widely used in the literature as a stability indicator, where E(ROA) is the expected return on assets, ETA is the equity to asset ratio and SD(ROA) is the standard deviation of ROA. And the third is Interest rate risk analysis.
Shaikh (2009) also testing the hypothesis, like us, using the descriptive analysis, ANOVA. He explains that strong relationship between Islamic and Conventional banks shows that both Islamic and Conventional banks are more profitable and the risk procedures in Islamic banks are adequate to mitigate their largely equity - based investments and give their customers adequate return which are comparable with conventional banks. If specific country rate risk is incorporated, the results can be compared with Islamic banks in other countries. This paper concludes that equity - based business of Islamic banks posing a slightly more risk than conventional banks is well mitigated by Islamic banks through their effective and adequate distinct risk management procedures.
Research question
Many years ago Islamic Finance was an unknown system; but now it has expanded to become a distinctive and fast growing segment of the International Financials markets. Islamic Finance in general and Islamic banking in specific become main players in the financial world.
The main purpose of our project is to compare Islamic Banking system to Conventional one; their profitability and insolvency risk exposure using modern business models. We are going to test profit compatibility of two types of banks with respect to risk management analysis. The model is derived from the paper "Risk management in Islamic and Conventional banks: A Differential Analysis" written by Shaikh S. A. The degree of relationship between profitability of Islamic and conventional banks will be checked by ROE and ROA. As risk management alone can not control more than 50% of relationship, the co-relational coefficient (R) will be taken as 0.5.
I Hypothesis testing
H0: Co-relational coefficient for Return on Equity>0.5
H1: Co-relational coefficient for Return on Equity <0.5
II Hypothesis testing
H0: Co-relational coefficient for Return on Assets>0.5
H1: Co-relational coefficient for Return on Assets <0.5
The insolvency risk exposures will be compared according to Z-score measure that is derived from the paper "Risk in Islamic Banking" written by Abedifar P., Molyneux P., A Tarazi A.. We will test insolvency risk exposures according to average means of Z-score measures for two types of banks.
III Hypothesis testing
H0: Mean of Z-score measures for Islamic banks > mean of Z-score measure for conventional banks
H1: Mean of Z-score measures for Islamic banks < mean of Z-score measure for conventional banks
Data
For our project we used quarterly financial statements for the years 2008-2011 for 3 conventional banks: Delta Bank (Data is taken from Kazakh Stock Exchange, JSC "Delta Bank", quarterly financial statements for the years 2008-2011, http://www.kase.kz/ru/emitters/show/NFBN), KazInvest Bank (Data is taken from Kazakh Stock Exchange, JSC "KazInvest Bank", http://www.kase.kz/ru/emitters/show/KIBN) and Bank Center Credit (Data is taken from Kazakh Stock Exchange, JSC "Bank Center Credit", quarterly financial statements for the years 2008-2011http://www.kase.kz/ru/emitters/show/CCBN), and 3 Islamic banks: Abu Dhabi Islamic Bank (data is taken from bank's official site, quarterly financial statements for the years 2008-2011, http://www.adib.ae/financial-results), Bank Muamalat Malaysia Berhad (data is taken from bank's official site, quarterly financial statements for the years 2007-2011, http://www.muamalat.com.my/corporate-overview/financials/2011.html), Al Baraka bank Lebanon (data is taken from bank's official site, quarterly financial statements for the years 2008-2011, http://www.barakaonline.com/default.asp?action=article&id=150).
Method of collection - we downloaded necessary financial statements from banks' official web sites and Kazakh Stock Exchange. Then we input the Total Assets, total Equity and Net Income data for all banks, and made calculations of Return on Assets and Return on Equity.
Overall the number of observations taken is 42. The more observations are taken, the more reliable are the results of the models. The number of 42 observations is not so much but we have looked at quarterly performance of banks for 4 years.
Our choice of Kazakhstani banks is explained by our desire to learn banking sector in our country. As we are future specialists of this industry, this project is beneficial for us.
Methodology
For testing the hypothesizes that we introduced in research question, we used Excel software, and such data analysis tools as Descriptive statistics, Correlation analysis and Regression analysis.
Table 1
Model Summary
ROE Conventional
ROE Islamic
Mean
0,029574731
0,035771469
Standard Error
0,006016389
0,003877189
Standard Deviation
0,038990659
0,025127059
Sample Variance
0,001520271
0,000631369
Range
0,213964364
0,133585116
Minimum
-0,017073056
-0,007385027
Maximum
0,196891308
0,126200088
Count
42
42
Table 2
Model Summary
ROA Conventional
ROA Islamic
Mean
0,003903391
0,003922472
Standard Error
0,000669709
0,000503412
Standard Deviation
0,004340209
0,003262481
Sample Variance
1,88374E-05
1,06438E-05
Range
0,018100036
0,014814293
Minimum
-0,001930922
-0,000429138
Maximum
0,016169115
0,014385155
Count
42
42
Descriptive statistics are used to describe and discuss data sets of two types of banks more generally and conveniently. The results of descriptive analysis for ROA and ROE are the following:
The means of ROE and ROA that is average and the most common measure of central tendency is higher in Islamic banks. That means that Islamic banks are more profitable (Shaikh S. A., 2009)
Standard error is the standard deviation of the mean which is lower for Islamic banks in both cases.
Standard deviation shows how much variation or "dispersion" there is from the average shows how much variation or "dispersion" there is from the average. As Islamic banks have lower standard deviations, the data points tend to be very close to the mean. And in conventional banks the data points are spread out over a large range of values.
Variance shows volatility, it is a measure of risk. So, Islamic banks have lower risk measures than conventional banks by ROE and ROA descriptive analysis.
Range measures the distance between the lowest and highest values in the data set. The Conventional banks have larger spread of data.
Table 3
Correlation between ROA values of Islamic and Conventional banks
ROA Conventional
ROA Islamic
ROA Conventional
1
ROA Islamic
-0,107256164
1
Table 4
Correlation between ROE values of Islamic and Conventional banks
ROE Conventional
ROE Islamic
ROE Conventional
1
ROE Islamic
0,017769561
1
The correlation analysis shows how two estimates of Conventional and Islamic banks move in relation to each other. The correlation between Return on Equities is positive, but not close to 1. So, they are not perfectly correlated. The correlation between Return on Assets is negative, i.e. the values of two types of banks move in opposite directions. But as the correlation coefficient is far from -1, Return on Assets are not perfectly negatively correlated.
ROA regression analysis
Table 5
Regression Statistics (ROA)
Multiple R
0,107256164
R Square
0,011503885
Adjusted R Square
-0,013208518
Standard Error
0,003283957
Observations
42
Table 6
ANOVA
df
SS
MS
F
Significance F
Regression
1
5,02024E-06
5,02024E-06
0,465511
0,498992022
Residual
40
0,000431375
1,07844E-05
Total
41
0,000436395
a) Dependent Variable: ROA Islamic
b) Predictors: (constant), ROA Conventional
ROE regression analysis
Table 7
Regression Statistics (ROE)
Multiple R
0,017769561
R Square
0,000315757
Adjusted R Square
-0,024676349
Standard Error
0,025435192
Observations
42
Table 8
ANOVA
df
SS
MS
F
Significance F
Regression
1
8,17374E-06
8,17374E-06
0,012634
0,911066617
Residual
40
0,025877959
0,000646949
Total
41
0,025886133
a) Dependent Variable: ROE Islamic
b) Predictors: (constant), ROE Conventional
The results of regression statistics:
R is the square root of R-Squared and is the correlation between the observed and predicted values of dependent variable. The value of R for Return on Equity analysis is 0.0177, that is lower than 0.5. So, we reject Null hypothesis and accept Alternative hypothesis of I Hypothesis Testing. The value of R for Return on Assets analysis is 0.107, that is also lower that 0.5. We reject Null hypothesis and accept Alternative hypothesis of II Hypothesis Testing. The results for R squared are quite low, i.e. low proportions of values of ROE and ROA in Islamic banks can be predicted from ROE and ROA values in Conventional banks.
The adjusted R-square attempts to yield a more honest value to estimate the R-squared for the population. The adjusted R-square is negative in both results, and there is a great difference between adjusted R-square and R-square because the number of observations is small and the number of predictors is large.
As the p-values are greater than 0.05, we would say that the ROE and ROA values of Conventional banks do not show a statistically significant relationship with the ROE and ROA values of Islamic banks, or that the values of Conventional banks do not reliably predict the values of Islamic banks.
Insolvency risk analysis
For insolvency risk analysis, we employ the Z-score measure which is widely
used in the literature as a stability indicator.
where E(ROA) is the expected return on assets that is derived from residual output of regression analysis, ETA is the equity to asset ratio and SD(ROA) is the standard deviation of ROA.
By the results of this model we compare means of Z-scores for two types of banks that are 32.38456 for Islamic banks and 38.69465 for Conventional banks. So, we reject Null hypothesis and accept Alternative hypothesis of III Hypothesis testing. With higher average Z-score Conventional banks are less exposed to insolvency risk and are more stable.
Limitations
For all models that we have used in our research the main limitation was usage of only two variables - ROA and ROE. For more accurate estimation of profitability there is a need to use more variables. Also, in general, statistics provides an estimate of the minimal error that might be in the measurement. The actual error can be much greater than the minimal (statistical) error. In addition, the sample size matters. If you don't have a big enough sample, you can't give a very reliable answer. Of course, with limited resources it might not be possible to collect a large sample. There's a tradeoff between sample size and reliability.
Conclusion
The relationship between Conventional and Islamic banking is not strong (ROEr=0,0177), (ROAr=0,1072). It shows that H0 for I and II hypothesis testing are rejected. The values for the ROA and ROE of two types of banks are not strongly correlated, and they do not predict performance of each other.
By the results of Insolvency risk analysis we have found that Conventional banks have higher Z-score measure. So, Kazakhstani Conventional Banks are more stable.
The project concludes that performance of Islamic banks is different from Conventional banks.
