The Impact of Financial Liberalization on Islamic Bank's Efficiency in Indonesia

The aim of this research is analyze the state of efficiency demonstrated by foreign banks, as aggressor, as compared to that of the local banks – government banks and Islamic banks during the period 2009-2012. Data Envelopment Analysis (DEA) with intermediation approach is used as the main tool of analysis. Input variable chosen are deposit, asset, and personnel expenses. Meanwhile loan and income are chosen as output variable. To find out the difference of efficiency between foreign banks and islamic banks, this study use parametric different test of Independent Sample T-test. The study indicates that foreign banks are less efficient if compared to the local banks, especially Islamic banks and Sharia Business Units. It is also evidenced that there is no significant difference between the efficiency of foreign and Islamic banks during the period 2009-2012.DOI: 10.15408/aiq.v7i2.1680


Introduction
Increase in foreign participation in the financial sector of emerging market brings benefits the host countries in the form of better technology and financial systems as well as from increased competition stimulated by new entrants. Does foreign banks always prominent? Claessens et al (2001) highlights that foreign banks earn greater profits than domestic banks in developing countries, but the opposite in developed countries, because foreign banks adopt better technology compared to banks in the developing countries. e banking system plays an important role in the economic development of any countries. As intermediary institutions, banking industries are required to have good performance (Sutawijaya and Lestari, 2009). Among successful indicators of bank performance is efficiency which refers to a measure of the deviation between actual performance and desired performance (Mester, 2003). As a consequence, a bank should minimize waste and eliminate unnecessary operational costs in order to be efficient. Managements of banks are aiming at rigorously cut costs in order to improve efficiency and ultimately profitability. Efficiency brings benefits in the form of positive spread (Karimzadeh, 2012). us it is a kind of survival strategy for financial institution in today's nogrowth worlds. e measurement of economic efficiency lies on determining the extent of technical and possibly allocative efficiency in an organization. Efficiency can be measured using parametric approaches and non-parametric approach (Deb, 2012). e parametric technique uses a pre-specified functional form for efficiency frontier, which may result in an inaccurate efficiency measurement (Berger and Humphrey, 1997). Economists have employed frontier efficiency measurement techniques to measure the productive performance of banks. Frontier efficiency measurement techniques use a production possibility frontier to map a locus of potentially technically efficient output.
is study employs the non-parametric frontier method, Data Envelopment Analysis (DEA) to examine whether the foreign banks are relatively more efficient than the Islamic banks. e empirical results reveal that increasing foreign bank entry is associated with increasing profitability, growing overhead expenses and rising non-performing loans in Indian public sector banks (Kalluru and Bhat, 2009). erefore, it is interest to investigate whether that condition will be happened as well as to the State-owned banks, and especially Islamic banks (BUS) and Sharia Business Units (UUS) in Indonesia.
e purpose of this study is to analyze the state of efficiency demonstrated by foreign banks compared to the local banks, which are government banks and Islamic banks. Evan, Hofler and Payne (2006) state that foreign banks have substantially better efficiency scores than all categories of domestic banks. Other researcher also examined that foreign banks are more efficient than the local banks in developing countries, and the foreign bank entry has a rather negative impact on the local bank efficiency (see Fathi, 2010). However, different researcher finds that foreign banks are not more efficient than Islamic banks (see Saif andYaseen, 2005, Sufian andHabibullah, 2010). To fill the research gap, this study examines to test theory related to banks' efficiency in foreign banks and Islamic banks.

Literature Review
DEA is a popular technique to evaluate bank efficiency performance and productivity improvement which combining all input and output data of the bank in a single measures (Deb, 2012). erefore, the DEA efficiency score for a specific firm is defined not by an absolute standard but is relative to the other firms under considerations (Tahir et al, 2009). e efficiency of the banking technique is measured by calculating the ratio between output and input. DEA will calculate banks using n inputs to produce outputs of different m (Miller andNoulas, 1996 in Sutawijaya andLestari, 2009). Bank efficiency is measured as follows: DEA assumes that each DMU will have weight that maximize of efficiency ratio (maximize total weighted output/total weighted input) (Muharam and Pusvitasari, 2007). Assumption of this efficiency ratio made DEA research is using orientation output in calculating efficiency technique, but both of these assumptions would have obtained the same results ( Sutawijaya and Lestari, 2009: 58). An DMU can be said as efficient relatively if it dual value equal to 1 (efficiency value of 100 percent). Contrarily when the dual value is less than 1, then DMU is considered not efficient relatively (experiencing inefficient) (Silkman, 1986;Nugroho, 1995 in Huri andSusilowati, 2004).
Measurement model engineering approach banks on the assumption frontier is divided into two types, namely (Sutawijaya and Lestari, 2009): First, DEA CCR model ( Charnes -Cooper -Rhodes , 1978). e assumption that is used in this model is Constant Return to Scale (CRS). Some linear program transformed into ordinary linear program in term of primal or dual, as follows: Maximization: Function limitations and constraints: e efficiency of each bank is calculated using linier program to maximize the amount of output that weighted from the s bank. Constraints weighted sum of inputs must be equal to one for s banks, while constraints for all banks are weighted output minus the weighted sum of inputs must be less or equal to 0. is means that all banks will be in or under reference of frontier performance which is straight line that cuts axis origin (Insukirdo in Sutawijaya and Lestari, 2009).
Second, BCC DEA models (Bankers, Charnes and Cooper, 1984). e assumption that is used in this model is Variable Return to Scale (VRS). Some linear program transformed into ordinary linear program in term of primal or dual, as follows:

Maximization Function limitations and constraints:
Where U o can be a positive or negative value.
Banker, Charnes and Cooper with the method of variable returns to scale (VRS) finally is known as CCR model (Charnes-Cooper-Rhodes) and BCC (Banker-Charnes-Cooper) (Amrillah, 2010). CCR assumes the existence of CRS. e mean of assumption CRS is that the proportional changes at all levels of the input will produces the same proportional change in the output level. Meanwhile, BCC assumes the existence of VRS. e mean of VRS assumption is that all the units that will produce measurable changes at various levels of output and the assumption that the scale can affect the efficiency of production. In this study will only use the input orientation approach assuming CRS, because the approach of input orientation to find out how much reduction in inputs that to be done by the DMUs that have poor performance for improving its performance.
Determining input-output which are used to calculate efficiency is one of important phases either when using parametric method or non parametric method. Input variable is the one which if increases, the other factors being constant, the efficiency unit would decreases and if decreases, the efficiency would increase and output variable is the one which if increases, the other things being equal, the efficiency would increase and vice versa (Nia et al, 2012).
In conducting analysis of banks' efficiency, there are five main approaches for defining inputs and outputs, namely: the production approach, the intermediation approach, the operational approach, the profitability approach and the modern approach (Nia et al, 2012). Since the intermediation approach has been used extensively in determining the inputs and outputs of the bank industry, therefore this study adopts this approach. e intermediation approach accounts for the cost of resources used and price (value) of outputs produced (Cook, et al, 2000). Intermediation approach constitutes a better instrument to study efficiency, and gives a more accurate image of how efficiently a bank is using its resources to generate profit Berger and Humphrey 1997, Taylor et al, 1998in Cook, et al, 2000.

Methods
Data employed in this research was secondary data consisted of deposit, asset, personnel expense, loan and income of Foreign banks, State-owned Banks, Islamic Banks and Sharia Business Units in Indonesia. is secondary data were obtained by using observation method on e analysis technique used is descriptive and comparative methods. e data analysis method of this research is data envelopment analysis (DEA) using Banxia Frontier software. us, three inputs under the intermediation approach: deposit, assets, personnel expenses and two outputs: loan and income are utilized to investigate efficiency of foreign banks and Islamic banks.

Discussion
A unit (bank) said to be efficient if it obtained a score of 100 percent. Banks with 100 percent efficiency rate is said to be efficient. Furthermore, the most efficient banks will become a benchmark for other units. According to the result of calculation DEA analysis with Constant Return to Scale (CRS) assumption by using Banxia Frontier software, it can be concluded that from the total of ten foreign banks as sample, only four foreign banks that can achieve efficiency score of 100 percent in 2009. Meanwhile, foreign banks are not efficient but have efficiency scores above average in 2009 was Citibank and Deutsche Bank with the efficiency of each bank is equal to 99.2 percent and 98.6 percent.
However, the efficiency score of foreign banks in 2010 declined, which only two foreign banks was perfectly efficient. While inefficient foreign banks but have efficiency scores above average in 2010 were the Bank of China Limited (98.1 percent), Citibank (94.6 percent), and e Bank of Tokyo Mitsubishi UFJ (99.2 percent). In 2010 foreign banks that concluded inefficient was Bank of America, Deutsche Bank AG., Standard Chartered Bank, Hongkong & Shanghai Banking, and the Royal Bank of Scotland.
In 2011, only two foreign banks can achieve efficiency score of 100 percent, which were Deutsche Bank AG and e Bangkok Bank Comp. Ltd. en the bank with the lowest efficiency score in 2011 was e Royal Bank of Scotland.
In 2012, the most efficient foreign bank is Deutsche Bank, JP Morgan Chase Bank, e Bangkok Bank Comp. Ltd. and e Hongkong & Shanghai Banking. It can be concluded that foreign bank that can maintain the efficiency score of 100 percent during 2009-2012 was e Bangkok Bank Comp. Ltd. In table 2., it explains that efficiency scores of foreign banks compared to the islamic banks (BUS) and Sharia Business Units (UUS). Islamic banks and Sharia Business Units was the banks that attained efficiency score of 100 percent during 2009-2012. As well as it occured to the Bank of America N.A. ird level bank that had efficiency score in above average was Bank of China Limited, which the bank can achieved 95 percent (2009) (2010), they states Islamic banks currently exist in all parts of the world and it is seen as an alternative system, which has a lot to offer.
For foreign banks and local banks are already efficient, in order to maintain its performance by minimizing the input variables at the same time increasing the output variable resulting in productivity, but with due regard to fairness is primarily concerned with the cost of labor input variables. For the banks that have not yet reached an efficiency score of 100 percent are expected to pay more attention to variables that are still not used optimally so as to avoid waste of resources and thus increases the efficiency of the bank.
For the Indonesian Bank Regulator and the OJK, to create rules that supports the development of the financial industry, especially Islamic finance especially in banking. Because Islamic Banking Finance, is a nascent industry in the State of RI that the majority Moslem has huge potential. For investors are expected to be able to see bank efficiency as consideration in investing their fund in the banks observed. Further research who wants to carry out the similar research is suggested to use DEA efficiency analysis with VRS (Variable Return to Scale) so that all units measured will make change at various outputs, which technology and production scale will influence efficiency level.