Islamic Banking Efficiency: Comparative Studies Between Malaysia and Indonesia

The purpose of this research is to measure efficiency level of sharia banking in Malysia and Indonesia countries and to analyze the factors affecting the level of efficiency of sharia banking for both countries. This reseach uses DEA (Data Envelopment Analysis) method in order to test the assumptions of Variabel Return to Scale (VRS), and uses Kolmogorov-Smirnov and Mann Whitney U-Test in order to test normality, and uses regression dummy variable of the data from the first quarter of 2011 until the fourth quarter of 2014. The research shows overall level of efficiency of sharia banking in Malaysia and Indonesia countries are fluctuating. Based on result, sharia banking in Indonesia more efficient than sharia banking in Malaysia; However, there is no significant differences among them. The reasons of this inefficiency are deposits, total financing, fixed asset, and personnel cost. However, operational income is the most efficient variabel for both countriesDOI: 10.15408/sjie.v5i1.3129


INTRODUCTION
The prospect of Islamic finance industry growth in Asia is very strong from time to time. It brings business opportunities for various industries, especially in terms of sharia banking and capital markets sector. In 2013, The Islamic finance industry in Asia consisted of sharia banks (total assets of USD 189 or 49% of the total assets of Islamic finance), sukuk (total assets of USD 177 or 45%), Islamic funds asset under management (total assets of USD 18 or 5%), and takaful (total assets of USD 3 or 1%). The largest centre of Islamic Finance in Asia is Malaysia, Indonesia, Pakistan, and Bangladesh. Islamic finance also began to grow in several Asian countries including Singapore, Brunei Darussalam, the Philippines, Hong Kong, Kazakhstan, Azerbaijan and Thailand. Islamic banking as part of the national banking system has an important rolein the economy. The role of Islamic banking is to encourage economic activity of a country then if it is not properly manage, it will trigger the economic crisis. Banking efficiency indicator is a reflection of the banking soundness and become a benchmark of public confidence. As an intermediary function where market share will be wide open. Banks were a lower in collecting third party funds (DPK) and also lower in the distribution of funds, indicates inefficient. The ability to collect public funds (DPK) well and passes well and can minimize the cost to obtain the maximum profit that the bank is efficient. Islamic banking, which grows rapidly, is charged to have a good level of efficiency in the foreign banking competition. Moreover, there will AEC 2015 and QAB 2020.
Research on the efficiency level competition among banks in ASEAN countries is very important to study as a tool to prepare the economy countries that are fused of the AEC (ASEAN economic Community) and QAB (Qualified ASEAN Banks) members, which are expected to support global economies. In order to remain in operation, each financial entity must be measurable results of its work in the form of performance. Efficiency in banking However, BOPO and NOM ratio has a weakness in measuring the efficiency i.e. difficult to generalize whether a ratio is good or bad, difficult to say whether a company is strong or weak and it does not measure cost of capital (Endri, 2008). In addition, the ratio of CAMEL also paid little attention to the efficiency factor. To overcome the existing deficiencies in measuring company performance ratio analysis, Frontier approach was developed to measure the company efficiency.
One of Frontier approach often used in analysing bank efficiency is Data Envelopment Analysis (DEA). DEA is a mathematical programming technique for measuring the efficiency of Decision Making Unit (DMU). This method has an advantage over methods parametric. The advantage in using a non parametric method is that we can identify the unit that is used as a reference. to make improvements and see the object of this study involves the Islamic banking in Malaysia and Indonesia. It is a reflection of the level of competition among two governments as well as other policy solutions can be scientific considerations for regulators.

METHOD
The object of this variable study includes 19 Islamic Banks (full-fledged  (2006), Yumanita (2008a, 2008b); Rusdian (2013), Warninda and Hosen (2015), Adawiyah (2015), Hosen and Rahmawati (2016) . Therefore, it is assumed that Islamic banks generate operational income, and the total financing using fixed assets, the cost of personnel, and third party fund (DPK). The assumption used is the Variable of Return to Scale (VRS) / input-oriented taken in this study because sister bank management is easier to improve performance for input in terms supervision for reducing the company's expenses. It will be easier for management to supervise the input used to increase a higher output.
DEA is a linear programming technique to assess the performance of Decision making units (DMU) or a bank in an industry operates in relation to other banks in the sample. In the DEA approach, linear programming is used to maximize the ratio between input and output (Charnes, Cooper, and Rhodes, 1978) as well as for DMUs Islamic banking industry. For DMUs banking industry that become the object of study of the entire sample input and output respectively denoted (marked) by then 'n' and 'm', where n = input = input and m = output. The bank efficiency is calculated by an equation: , for i = 1, ......, m dan j = 1, ...., n. (1) Charnes, Cooper, and Rhodes (1978) suggested that part of this linear programming can be changed to be ordinary linear programming by the following: = =1 = 1 and u i and v j ≥ 0 In the same way can be converted into a dual linear programming problem: While ɛ s is the total value (score) bank-s technical efficiency, the value 1 signifies the limit point. The equation (3) and (4)  Bank-s all located on the right side the limit or inefficient bank described as a point S. Overall, the technical efficiency (ɛ s ) is calculated by the ratio of AQ/AS.
Thus, the bank-s is to be reduced (1-ɛ s ) of input to achieve efficiency at the point Q.  However, scale efficiency can occur due to an increase (IRS) or decrease (DRS) return to scale. To get both of results, the settlement of linear programming equation (3) and (4) should be limited to the sum from 1 to N is less than or equal to 1 (≤1) or OBV. In the approach, it will be assessed thelevelof technical efficiency constant returns to scale (CRS) and the technical efficiency of variable returns to scale (VRS). CRS technical efficiency gives the assumption that if the numbers of inputs risein the amount of x, then the output also risesby X. In other words, the increase in output is proportional to the increase in input. While the VRS technical efficiency gives the assumption that an increase in output is not proportional with the increase of input, may be larger or smaller than the increase in inputs. In DEA, efficiency states that ratio between the total input andtotal output weighted. Where each unit of economic activity is assumed to freely determine the weight of each variable input and output variables that exist, as long as it is able to meet the two conditions required i.e.

Level of Islamic Bank Efficiency in Indonesia
In 2013   Source: Data processed using WDEA (VRS assumption) Based on the processed data in figure 2, the overall growth rate of the Islamic banks efficiency in Indonesia has fluctuated due to the trend of individual efficiency levels also fluctuate. During the study period, the score achieved that the highest Islamic bank efficiency was by Bank of Jabar Banten Sharia (100%) and the lowest efficiency was achieved by Bank of Muamalat Indonesia amounted to 90.67%. Based on the results of the efficiency measures that the Islamic banks, which achieved a score of 100% can be interpreted that the bank has been able to optimize all its resources and were efficient categorized bank.  Source: Data processed using WDEA (VRS assumption)    Malaysia has reached the level of efficiency in the range of 80-99%, but cannot achieve perfect efficiency score.

Combining Efficiency Analysis between Indonesia and Malaysia
From the Figure 4 shows that the average level of Islamic banking efficiency between Indonesia and Malaysia has fluctuated. Blue arrow label leds Indonesia is at the top, it shows that the level of efficiency of Islamic banking in Indonesia was more efficient than the Islamic banks in Malaysia.

Total Potential Improvement of Islamic Bank in Indonesia and Malaysia
In this study the authors tried to process the data obtained from WDEA to achieve the total potential improvement to be repaired because the inefficiency due to the occurrence of shortage or excess in the variable input (DPK, personnel burden, fixed assets) and output (total financing and operational income). From the figure 6,  Fixed assets are tangible assets include buildings, office equipment, vehicles, land, ATM machines, etc. Actually, this can be minimized with the cooperation between Islamic banks with the parent bank to reduce the cost of fixed asset, for example with a strategy ATM joint, mutual insurance, optimization of office channelling, Optimization in providing joint ATM was necessary so that customers who are in disadvantaged areas can access Islamic banking facilities without having to incur huge costs, as well as management of Islamic banking should be to educate employees on asset management in order to achieve efficient use of existing fixed assets.
Expenses for personnel burden conducted due to Islamic banking expansion. When the number of employees increased, the cost of personnel human resource rules more comprehensible by determining the minimum limit of personnel burden that must be issued.

Kolmogorov-Smirnov Test
To determine whether there is a significance difference in the level of efficiency of Islamic banking in Indonesia and Malaysia, the testing of nonparametric statistics using the software SPSS 20 is needed, where the first thing to do with examining the data on the DEA analysis normality test Kolmogorov-Smirnov to be known normally distributed data or not, when normal use independent test t-test to determine differences in efficiency results but if not normally distributed, then tested the Mann-Whitney U-test. From the results of data normality test with Kolmogorov-Smirnov, the test resulted produce value Asymp. Sig. (2-tailed) for .000 whichis less than 0.05 which means that the data were not normally distributed. Therefore, testing the significance of differences is done by using the Mann Whitney U-Test.