This research is used to investigate the efficiency levels in the banking sectors in various Arabian countries, while considering the role of economic development and different financial reforms taking place, and how these reforms impact the banking industry performance and efficiency (Al-Jarrah, 2007). The comparison of these financial institutions to American or western culture nations shows how these reforms can help or cripple banking industries around the world. Efficiency estimation techniques can be categorized into two groups including parametric and nonparametric methods. Data Envelopment Analysis (DEA) is an example of a nonparametric method of analysis, which is believed to be a better form of analysis compared to parametric methods because it is able to measure bank efficiency as it related to the ability of the other to characterize the frontier technology in a simple mathematical form with the ability to accommodate non-constant returns to scale (Al-Jarrah, 2007).
DEA constructs the frontier of observed input-output ratios through linear programming techniques, because it depicts how a particular decision making unit(DMU), which operates relative to other DMUs in the sample by providing a benchmark for the best practice technology based on the individual experience of banks or branches in the provided sample (Al-Jarrah, 2007). Linear programming is used to determine the best possible outcomes of an experiment or mathematical model like DEA. DEA has the ability to estimate the efficiency under the assumption of constant return to scale and the various variable returns to scale, but factors like imperfect competition and constraints in finance have the ability to cause a DMU to not operate at its optimal scale, and as a result the use of the CRS specification confuses measures of technical and scale efficiency (Al-Jarrah, 2007).
Analysis
There are three main types of bank performance approaches including production, intermediation, and profitability. Production approaches treat banks as institutions using capital and labor for production for various kinds of banking services (Thagunna & Poudel, 2013). This approach is best to consider banking institutions at the local level. Intermediation considers an individual branch or entire banks through the transferring of money borrowed from deposits, into the money lent to borrowers (Thagunna & Poudel, 2013). This process is beneficial to banking institutions, because the interest given to the deposits is significantly lower than the interest charged to borrowers.
The profitability approach is similar to the production approach. The outputs of profitability are a profit oriented model and the outputs are maximized (Thagunna & Poudel, 2013). This approach is useful in the banking sector or industry because it depicts the growth and assets acquired by individual branches or entire banks. Profitability makes it easier for banking decision makers to make educated decisions and comparisons within their institutions, to make determinations as to which products or services are more productive for the different communities the banks operate within.
Hypothesis
Null Hypothesis: Islamic banks performs better than their traditional bank counterparts
Alternative Hypothesis 1: Islamic banks do not perform better than their traditional bank counterparts.
Alternative Hypothesis 2: Half of the Islamc banks are more efficient than traditional banks.
Input and Output Variables
The inputs for the DEA experiment are the total deposits, interest expenses, and operating non-interest expenses. The outputs include total loans including loans, advances, and bill purchases, interest incomes, and operating non-interest income. It is important to use homogeneity conditions for DMUs in the model in order to encourage the use of DEA for banking firms with similar resources and operations providing similar products or services (Thagunna & Poudel, 2013). This is important because it makes it clear it is not possible to make comparisons with individual samples without similar attributes to make relevant or valid comparisons.
The variables used for the data set include K for inputs, M for outputs, and N for firms. The observations of the set of input and output can be represented by column for input which would be vector xi, the output vector yi, and the sets would look like xik and yim (Al-Jarrah, 2007). The inputs matrix is X=[KxN] and the output matrix is Y=[MxN], which takes into consideration the data is for all of the firms which were previously defined as N.
DEA proposes a variable return to scale and an output oriented model, which uses the radial form, which is based on technological efficiency where the most efficient firms or linear combination of firms produce as much or more for every output, which are the given inputs, or uses as little or less of the input, which are the given outputs (Al-Jarrah, 2007). Every firm wants to have as much input as possible, while maintaining the outputs as low as possible. The efficient frontier is composed of these un-dominated firms and the piecewise linear segments, which connect the set of input or output combinations of the firms, which yields a convex production possibility set (Al-Jarrah, 2007).
Results
The output oriented CCR model resulted with banks scoring less than one or one. When a bank scored less than one, there are not producing at the same level of output as the banks scoring one, which have the same level of input (Thagunna & Poudel, 2013). At the same time, it is clear the banks scoring less than one have the opportunity to increase their output to score a one in the future.
DEA is a valuable tool, which can be used for strategic policy, and operational decision problems, especially in the service and non-profit sectors (Thagunna & Poudel, 2013). This tool allows the decision makers to make comparisons, which can be used later to make changes to strategies, policies, and operational decisions. Similar to other industries, it is the strategies, policies, and operational decisions, which can make the difference between an entity being efficiency and performing optimally, and not.
This approach provides an analytical quantitative benchmarking tool for measuring relative efficiency, but revealing the best-practical frontiers by analyzing the results received from the model and each DMU (Al-Jarrah, 2007). As the results are complied, it is important to make the comparison between the different types of banking sectors, which provide the variations, which make it impossible to compare small banks to larger banks, or banks who primarily work in loans, while others focus more closely to mortgage and refinancing.
Limitations
The main limitation to this research is the reliance on secondary sources for applicable data. This is due to the fact the data collected on various banks is from annual reports. The problem with these reports as data is the method in which the information is collected, which is a form of self-reporting. Not to say banks are always looking to mislead customers and potential investors, but the possibility the numbers provided can be altered to make the financial institution seem more productive than it really is.
Another disadvantage of the use of DEA to measure the efficiency of banks is the varying dimensions’ different banks can have (Al-Jarrah, 2007). Additionally, too many dimensions can make it more difficult to make comparisons between different entities. Western banks are more interested in their profitability, which does not always equal to efficient. In finance, profit rules, but at the same time, it is also more beneficial for various banking institutions to invest into their local communities, which will allow for more growth, because the community members will be more willing to deposit more than they are borrowing.
Some dimensions make it seem a bank is more efficiency, than it really is, because there is not a similar entity to make true comparisons to. Which means the results are not able to be compared to, which makes the values or the inputs and outputs irrelevant (Al-Jarrah, 2007). This is also why the self-reporting is a tricky method for data collection, because it can alter the validity of the constrained variables.
Findings and Conclusion
In conclusion, the first two hypotheses are rejected, because it is difficult to determine with any certainty any banks are more efficient than others, because western culture banks and Islamic banks function in different ways, which makes the DEA analysis difficult to make definite comparisons. At the same time, the results did depict at least 45% of the Islamic banks are just as efficient as the western culture banks and within their local communities (Thagunna & Poudel, 2013). This reality, allows for the acceptance of the third hypothesis, because at least half of the banks are as efficient.
DEA analyzing is more useful when analyzing one bank instead of banks in different communities, or different countries, because it is better make comparisons within separate branches instead of separate financial institutions (Thagunna & Poudel, 2013). This is most likely due to the fact Islamic nations are still building their monetary policies and accounting requirements. Once again this raises the issue of the annual financial reports coming from the Islamic banks, because there is no consistency within the Islamic communities around the world, or with the policies and practices of the western culture banks.
It is clear there needs to be a set of consistency standards or conditions to ensure the efficiency measurements are correct and verifiable (Al-Jarrah, 2007). This is why the parametric approached like DEA are useful for efficiency measurements, because it allows the researcher to focus on the specific measurements and conditions they are focused on, to minimize the effect of information or data, which is not valid to the experiment at hand. The best solution to make better comparisons of efficiency of banks is to use multiple approaches, like adding ANOVA to make comparisons on an international level instead of being limited to branches within a bank, or different sized banks within a community.
References
Al-Jarrah, I. M. (2007). The use of DEA in measuring efficiency in Arabian banking. Banks and Bank Systems, 2(4), 21-30.
Thagunna, K. S., & Poudel, S. (2013). Measuring Bank Performance of Nepali Banks: A Data Envelopment Analysis (DEA) Perspective. International Journal of Economics and Financial Issues, 3(1), 54-65.
