Profit efficiency development of Islamic Banking using the stochastic frontier approach

JEL: E44, G21, P34, Z12 Abstract This study aims to analyze the efficiency of Islamic banking with a stochastic frontier analysis (SFA) on 10 Islamic banks from 2011 to 2018. The research variables consist of input variables total profits, third party funds and personnel expenses and output variable total financing and murabahah receivables. This research method uses SFA to measure the efficiency of Islamic banks and the influence between variables. The results of this study indicate the average efficiency of Islamic banks has not reached 100% or at 82.24%. Variable third party funds, personnel expenses have a positive effect on total profits while total financing and murabahah receivables have a negative effect on total profits.


Introduction
Islamic banking annually experiences growth rates and market share of Islamic banking shows high performance and high market levels. The Chairman of OJK (Indonesia Financial Services Authority) revealed that the performance of Islamic banking reflected by the position of Islamic banking assets which recorded growth of 20.65% annually or year on year (yoy) at the end of February 2018 to Rp 429.36 trillion. Meanwhile, financing is still moving to a double-digit level of 14.76% yoy to Rp 289.99 trillion. In the first two months of 2018, Wimboh explained that there had been an increase in accounts to 560 Islamic banking accounts from the end of December 2017. This was also supported by an increase in the number of sharia commercial bank (BUS) and sharia business units (UUS). In February 2018, there are 13 sharia commercial banks, 21 sharia business units and 167 sharia rural banks (BPRS). This growth is supported by sharia capital which is classified as good, reflected by the sharia public CAR ratio of 18.62% and nonperforming financing of 4.31% which is still maintained below the 5% threshold. Islamic bank liquidity is still relatively high from the threshold. It can be seen from the position of third party funds that raised up 16.1% yoy to Rp 339.05 trillion (Kompas, 2019). According to Adiwarman in 2018 there will also be two more BUS (Sharia commercial bank) going up to BOOK III (Indonesian bank level). Additional assets from the establishment of state-owned Islamic banks, mergers and conversions could increase the share of the Islamic banking market by around eight percent (Republika, 2019).
Aside from the performance ratio which is the impetus of the National Islamic Finance Committee (KNKS) on the growth of Islamic banking, the initiation of the formation of large Islamic state-owned banks, zakat integration, and the development of halal lifestyles has an impact on Islamic banking. The positive impact is also brought by the development of the role of waqf through Sharia microfinance institutions (LKMS) 'Bank Waqf'. From the performance and programs implemented by the government, it is expected to obtain a high market share in the community and eliminate the community's assessment of Islamic banking which is the same as conventional banking. The efforts presented above are in line with the objectives of implementing the AEC (ASEAN Economic Community) program which has been implemented since 2015. The ability of Islamic banks in responding to the AEC has been well implemented in terms of high growth of Islamic banking assets and capital and human resources that are able to compete in the global market. The problems of Islamic banking before AEC were found in sharia human resources which had been covered up a lot by conventional human resources which were still scientifically very minimal, especially in the field of sharia. The improvement in HR was followed by the addition of Islamic commercial banks.
Islamic banking performance can also be seen from the efficiency of Islamic banks, the more efficient a bank, the more liability the bank has. Based on the results of previous studies, Sharia commercial bank shows a high level of efficiency Rahmawati (2015), Wahab (2015), Hosen & Muhari (2014). However, it contradicts with the results of the previous study which shows that Islamic commercial banks have a low efficient rate compared to Islamic rural bank (Effendi, 2016) and conventional banks (Oktavi, 2018).
The profit efficiency of Islamic banks is on average 93.41% below 100%, this shows a good and efficient performance in Islamic banks Suhel (2011). Overall, the results show that Islamic Bank Malaysia has an average profit efficiency of 67.34%. BPRS has a high level of efficiency (Effendi, 2016;Nuryartono et al., 2012). BPRS efficiency is still very volatile (Naufal & Firdaus, 2017). Other research results on 5 BPRS, BPRS Central Syari'ah Utama are inefficient while BPRS Dana Amanah, BPRS Dana Mulia, BPRS Harta Insan Karimah were inefficient in the first quarter of 2016 (Ramadhan et al., 2017).
Research of Tahir & Haron (2010) suggests that the Islamic banks of Africa, the Far East and Central Asia, Europe and the Middle East are relatively better at monitoring cost and profit efficiency. Efficiency of Conventional versus Islamic Banks the cost and profit efficiency of 80 banks in 21 of the Organization of Islamic Conference (OIC) countries shows that there is no significant difference between the efficiency of conventional banks and Islamic banks (Mohamad, et al., 2008). The results of research by Eisazadeh et al (2012) shows that banks in the Middle East and North Africa, The Middle East and North Africa can operate efficiently if the bank is able to save 20 percent of the total cost factors that affect efficiency production. Mongid and Muazaroh's (2017) using the SFA method of banking efficiency in Indonesia, Malaysia and Singapore is better than banking efficiency in Thailand. In general, efficiency in banks is influenced by bank size, crisis dummy, profitability, capital adequacy, total assets, and problem loans.
The efficiency of Islamic banking was previously measured by banking ratios such as operating expenses and operating income (BOPO). Based on these measurements shows that banks are considered not to have contributed greatly in increasing economic growth. According to the Secretary General of the Indonesian Islamic banking Association (Asbisindo) Achmad K Permana, 2018 is not a good year for Islamic banking to grow aggressively where the assumption of economic growth set by the government next year is not far from achieving national growth in 2017, which is still in the range of five percent.
From the above background this study aims to analyze the efficiency of Islamic banks with parametric input and output components and do not use bank ratio. The difference between this study and the previous researchers who also used parametric measurement, namely the efficiency of this study, was conducted with an intermediate approach based on the total profit of Islamic banks.

Research Method
This research is a quantitative study using secondary data. Quantitative research is an objective research approach, including the collection and analysis of quantitative data and using statistical testing methods (Hermawan, 2005). In this study a number of hypotheses were tested statistically so that they could be classified as testing hypothesis studies. Testing hypothesis research is research that tries to explain the nature of a particular relationship/influence, see certain differences in several groups, or the independence of two or more factors in a situation (Hermawan, 2005).
Population is a collection of objects that want to know their characteristics both from the results of measurements or qualitative and quantitative calculations. The population in this study are all Sharia Commercial Banks in Indonesia. The sample in this study is a Sharia Commercial Bank whose financial statements are available during the period of 2011-2018. The purposive sampling method with complete financial statement criteria is available in the year of the researcher, namely the statement of financial position, income statement and statement of changes in capital during the observation period. The measurement of variables in this study is described in Table 1.

Stochastic Frontier Approach (SFA)
The SFA method was developed by (Berger & Mester, 1997). Frontier 4.1 software is used to estimate the cost function using the panel data method on the Stochastic Frontier Approach (SFA) parametric approach. The standard stochastic cost frontier function has the following general form (log): Where Ci is total cost of bank n; Xji is input j on bank n; Yji is output k on bank n; while ei is error term. ei consist of 2 functions: where ui are error factor which can be controlled; and vi are random error factor which can be controlled. V is assumed to be distributed normally N(0, σ2v) and u distributed halfnormally, |N(0, σ2v)| where uit = (ui exp(-h(t-T))3 and h is parameter which can be estimated. Cost efficiency is derived from a cost function by the shape of general equity (log) as follow: By using the form of the stochastic cost frontier equation the cost equation can be written as follows: where C is total cost or cost efficiency; w is input amount; y is output amount; while u and v are error term. So, cost efficiency can be written as follows CFFn = = Efficiency values calculated using the SFA method are percentage forms. Percentage that shows efficient intent is a percentage with a weight of 100%. The closer it is to 100%, the more efficient the banks are in using their inputs to produce maximum output.

Result and Discussion
This study uses the data of Islamic banking financial statements that are published on each of the websites of Islamic banks and financial services authorities (OJK). Islamic bank financial reports were obtained from 2011 to 2018. Based on OJK Islamic banking statistics the number of sharia commercial banks is 14 Islamic banks, in this study the sample of this study uses 10 sharia commercial banks which have annual financial reports that are regularly published on each website Islamic commercial banks. Measurement of the efficiency of Islamic commercial banks is done by the SFA approach which is analyzed by the intermediation approach. The intermediation approach views banks as intermediaries that convert and transfer financial assets from surplus units to deficit units (Haddad, 2003). The level of efficiency is analyzed from the profit function model with the dependent variable total earnings, independent variables in the form of inputs consisting of third party funds (P1) and personnel expenses (P2) and output variables namely total financing (Q1) and murabahah receivables (Q2). Measurement of efficiency in the frontier model is done by changing each variable in the form ln. Furthermore, the SFA output as a result of the efficiency of Islamic commercial banks, in the second analysis by looking at the effect of each independent variable on total earnings as the dependent variable.
Descriptive statistic of Islamic banks: In Table 3

Analysis of the Efficiency Level of Sharia Commercial Bank
This study uses the Stochastic Frontier Approach (SFA) method to determine the level of earnings efficiency at Islamic commercial banks. Data processing using Frontier software 4.1. Following are the results of earnings efficiency with the SFA method of Islamic banks   Rp 17,314,492,247. This shows the use of personnel expenses that are higher than the output obtained so that the output that can be achieved is only 1.2% with a profit of Rp.74,492,188. Based on the results of the average efficiency of Islamic commercial banks, it can be classified the level of efficiency based on the standard deviation. The grouping of BUS earnings efficiency values into five categories using quartile percentages ± standard deviations (Rahmawati, 2015)  Based on the grouping of Sharia Commercial Bank (BUS) it can be concluded that the overall category of BUS shows less efficient. This shows the problems of Islamic banks as a whole so that there needs to be an increase in market share of financing to the community one of them by making new products based on technology. The high level of efficiency of Islamic banks is also shown in the research of Rahmawati (2015), Wahab (2015), Hosen & Muhari (2014), Suhel (2011) and Mohamad et al (2008).
The Influence of Input-Output Variables on the Total Profit of Sharia Commercial Banks  (2019) Data processing using software frontier 4.1. The form of the equation of profit efficiency levels as follows.
In the input variable, namely third party funds (lnP1) regression coefficient 0.045, this shows that if third party funds have increased by 1%, the total profit will increase by 0.045%. This indicates that the High Third Party Fund (DPK) of Sharia Commercial Bank (BUS) gives an increase in the amount of BUS profit where the higher DPK, the profit toal will also increase. In the t-test, DPK does not affect the total profit where the t-ratio of 0.24120 is smaller than t-table 1.99045. The results of this study are in accordance with the research of Suhel (2011) andEffendi (2016).
The personnel load input variable (lnP2) regression coefficient 0.007 shows that if the personnel load increases by 1%, the average total BUS profit will increase by 0.007%. This shows the use of personnel expenses incurred by the BUS for employees in accordance with the expected performance of the BUS so that the profit earned increases annually. Statistically the personnel expense has t-ratio of 0.63029 smaller than t-table 1.99045 where the personnel expense has no effect on the total BUS profit.
In the output variable that is total financing (lnQ1) the regression coefficient of -0.003 shows that if the total financing has increased by 1%, then the total profit will decrease by 0.003%. This shows that the financing output channeled has a high level of non-performing financing so that the total profit has decreased. Total financing does not affect the total profit where the value of t-ratio 0.23681 is smaller than t table 1.99045. The results of this study are in accordance with the research of Suhel (2011), Effendi (2016, Rahmawati (2015), and Wahab (2015).
In murabaha receivables variable (lnQ2), the regression coefficient of -0.008 shows that if murabaha receivables increase by 1%, the total profit will decrease by 0.008%. Murabahah receivables less contribute to total BUS profit, this is due to low economic activity in which murabaha financing is one type of financing that is much in demand by the public that should be getting bigger murabahah receivables hence the high profits earned. Murabaha receivable variable does not affect the total profit with a value of t ratio of 0.48734 smaller than t table 1.99045. The results of this study are in accordance with the research of Hosen & Muhari (2014), Rahmawati (2015), Wahab (2015), and Naufal & Firdaus (2017).

Conclusion
The results of the profit efficiency of Islamic banks with SFA method for 2011-2018 averaged 0.82249 or 82.24%. The profit efficiency level of Islamic banks that are close to 100% is found in BCA syariah banks, Maybank syariah and Jabar Banten syariah banks as compared to sharia commercial banks below 100% at Muamalat Indonesia Bank, Victoria Syariah Bank and Mega Syariah Bank.
Regression results show that third party funds and personnel expenses have a positive effect on total profits. The use of personnel expenses incurred by Islamic banks for employees in accordance with the expected performance so that the profit earned is increasing every year. Total financing and murabahah receivables have a negative effect on total profits. The acquisition of murabahah receivables is less contributing due to the low economic activity where murabahah is one of the financing that is much in demand by the public so that if the murabahah receivables are lower the profit obtained is also small.