Internal and External Factors Effects towards Return on Assets in Indonesian Foreign Exchange Sharia Bank

This study aims to analyze the influence of internal factors such as financing and third-party funds, as well as external factors including inflation, BI, and exchange rate against Return On Asset (ROA) at Indonesian Foreign Exchange Sharia Bank. Furthermore, the library research was used and quantitatively described. Also, using the total sampling technique, the research team obtained six sample banks, namely Bank Mega Syariah, Muamalat Indonesia, Panin Dubai Syariah, BNI Syariah, BRI Syariah, and Bank Syariah Mandiri. The data include Balance Sheet and Profit, Loss Financial Statements, as well as the published reports on inflation, BI, or exchange rates obtained in a quarterly period from 2015 to 2019. According to the F-test, there was a significant influence between the independent variables (Financing, Third Party Funds, Inflation, BI, and Exchange Rates) on the dependent (ROA). Meanwhile, the t-test showed there was no significant effect of partial financing, Third Party Funds, BI rate (X ), and Kurs (X ) on the ROA. Therefore, this study showed there was an increase in ROA 4 5 because the society involves in financing due to high inflation.


INTRODUCTION
Profitability is a company's ability to generate profits (Marginingsih, 2018) by using total existing assets (Swandayani & Kusumaningtias, 2012). There are eight criteria to measure profitability level, namely Gross Profit Ratio, EBITDA (Earnings Before Interest, Taxes Depreciation, dan Amortization), Operating Profit Margin, Net Profit Margin (NPM), Cash Flow Margin, Return on Assets (ROA), Return on Equity (ROE), and Return on Invested Capital (ROIC) as stated in the circular letter of Bank Indonesia No.6/23/DPNP on May 31, 2004.
The higher the ROA (Gumanti, 2011), the more the profit (Syah, 2018), and the better the its position in terms of asset used (Lalujan, Pelleng, & Tumbel, 2016). Based on Bank Indonesia Circular SE No.13/24/DPNP in October 25th 2011, the ideal minimum ROA requirement is 1.5%. In other words, when the profit made is below the stipulated value, the bank is not optimal in managing its assets.
The research team used ROA as profitability indicator. This is because Bank Indonesia prioritizes profitability as measured by assets whose funds mostly come from public deposits. It is also the most objective measurement method based on available accounting data. Therefore, the amount of ROA can reflect series results of company policies, especially banking. Also, it is a measure of the financial performance in obtaining profit before tax, as generated from the total assets (Swandayani & Kusumaningtias, 2012).
Determinant factors of profitability are classified as internal and external (Ernayani, Moorcy, & Sukimin, 2017). The internal factors include fundraising, as well as the managements of capital, liquidity, and cost. Meanwhile, the external are beyond management control, such as competition, regulation, concentration, market share, ownership, scarcity of capital, money supply, inflation, interest rates, foreign exchange rates, economies of scale, and bank size (Syah, 2018 December 2019, which was below 10%, hence classified as mild. Inflation is classified into four levels, namely mild, moderate, severe, and hyperinflation. The mild occurs when increase in price is below 10%, moderate between 10%-30%, hyperinflation between 30%-100% in a year. Meanwhile, hyperinflation or uncontrolled inflation occurs when price increases are above 100% a year (Hambarsari & Inggit, 2016 14,017.00 (BI, 2020).
In carrying out its operational activities, banks as an intermediary institution cannot be Third Party Fund (TPF) is an internal factor affecting profitability, which has a positive and significant effect on bank performance (ROA). This means the more TPF the bank can acquire, the higher its performance (ROA) (Sudiyatno & Suroso, 2010). This has a positive significance as the most influential variable to profitability (Anggreni & Suardhika, 2014). In contrast to previous studies, Muliawati and Khoiruddin (2015) stated that the variable has no significant effect on the profitability of Sharia Commercial Banks. This is in line with Sihombing and Yahya (2016) that there was no significant effect of TPF on the profitability of Islamic banking in Indonesia. Furthermore, Third Party Funds have no effect on the profitability of Islamic Commercial Banks in Indonesia during 2012-2015 (Fitriana & Musdholifah, 2017). Dasari and Wirman (2020) also showed there was no effect on profitability.
Beside internal factors, in the implementation of its operational activities, banks cannot be separated from the influence of economic development conditions on macroeconomic theory. In this case, they arise from external factors (government and central bank policies) whose main objective is to maintain monetary stability, such as balance sheets, payment, national income (gross domestic and gross national products), as well as the rates of economic growth, inflation, unemployment, foreign exchange, money supply and interest (Swandayani & Kusumaningtias, 2012). As an intermediary institution, banks are vulnerable to inflation risk. When a country has high inflation, it will result to an increase in consumption, which will consequently affect the pattern of saving and financing in the society. This change will directly impact the operational activities of banks, where the amount of funds collected from the public will decrease, and have an impact on performance in earning income and generating profits (Sukirno, 2006).
According to Dwijayanthy and Naomi (2009), there was a significant relationship between the inflation and exchange rates on profitability, while the BI rate was insignificant.
Furthermore, Hidayati (2014) showed that the variable rate of inflation and exchange rate had a significant influence on the profitability of Islamic banks. Syah (2018) proved there was a significant effect of BI on ROA. Moreover, Swandayani and Kusumaningtias (2012) stated that the variables of inflation, interest rates, foreign exchange rates, and the money supply gave a significant effect on the ROA of sharia banking in Indonesia. However, Ady (2020) showed the exchange rate had a negative or no significant effect on the profitability in Indonesian banks.
Macroeconomic and internal factors are variables that affect banking, among which inflation is insignificant (Istan & Fahlevi, 2020). According to Sukirno (2013). This variable gave a negative effect on profitability. The higher the inflation rate, the lower the ROA ratio.
Basically, an increase in this variable indicates economic growth, but it is detrimental for longer period. Therefore, the high rate causes the prices of domestic goods to be relatively more expensive than imported ones. Also, the real value of savings or third-party funds can decline with increase in financing since people withdraw funds from banks due to expensive local goods. This is in line with the increase in exchange rates as a result of increment in domestic goods prices, and cheap imports.

Selection of a Regression Model
The panel data regression model equation is written as: Yit = α + β1X1it + β2X2it + β3X3it + β4X4it + β5X5it + eit Furthermore, the panel test stage according to Iqbal (2015) suggested that after carrying out the test phase by choosing the panel data regression method, subsequently, the estimation model should be determined as follows 1) F Test (Chow Test), when the prob value at F is smaller than the significance, the fixed is chosen rather than the common effect, and vice versa when the prob is greater. 2) Hausman Test, when prob. Chi squares is smaller than significance, the fixed effect is chosen over random, and vice versa when prob. Chi squares is greater. 3) Lagrange Multiplier (LM) Test, when the prob value is smaller than the significance, random effect is chosen rather than common, and vice versa when the prob is greater.

Classic Assumptions
Based on the determination of the estimation method used, such as the F-Test (Chow Test), the selected model was fixed effect, where the probability F value was (0.0000) <0.05.
The same model was also chosen in the Hausman Test, with a probability F value of (0.0258) <0.05. Since the fixed effect model was used in this study, the Lagrange Multiplier test was not required. Therefore, the classical assumption test was carried out, with the results as follows.

Normality Test
Test Normality used fallow jarque method with the results outlined in the following figure Graph 2. Normality Test Results Based on Graph 2, the results were obtained in the form of a jarque-bera probability test value of 0.266221. The JB test value was greater than the significance level (0.266221> 0.05), hence, the H0 was accepted and the residual had a normal distribution.

Multicollinearity Test
The multicollinearity test results are as follows.

Autocorrelation Test
Autocorrelation Test Results are as follows.

Heteroscedasticity Test
To determine the heteroscedasticity test, the White and Glejser method was used, with the results as follows. According to the results, the smallest chi-square was (0.5753), which means that it was above the value of α (0.05). Therefore, H0 was accepted and there was no heteroscedasticity problem. The following is the heteroscedasticity test using the Glejser method, with this output; Table 4. Glejser Test Heteroscedasticity Results Based on Table 4., the smallest chi-square was (0.3544), which indicated that it was above the value of α = 0.05. Therefore, H0 was accepted with no heteroscedasticity problem.
Based on the test methods above, it can be concluded that the Chi-square value was small with the value α= 0.05, hence, there was no heteroscedasticity problem.

Determination Coefficient Test (R2)
Due to the estimation, the fixed effect model was selected, while the test results for the coefficient of determination (R2) are as follows

T-test
The partial test results can be determined by observing the prob and t statistical values of the fixed-effect model (FEM) as the selected estimation model with the following results

F Test
In estimating the data, a significance level of 5% or 0.05 was used. Based on the data processing conducted, the following hypothesis testing results were discovered Return on Assets is one of the criteria for measuring profitability level by comparing profit before tax with total assets, and it is affected by internal and external factors. Based on the results from external factors, the inflation variable gave a significant influence on ROA. This indicated that an increase in this variable can have a positive effect on increasing the Return on Assets. It can be related to inflation when prices on a certain level also rise. Therefore, to survive, people undergo financing in sharia banks. The higher the financing, the greater the ROA will be when the level of non-performance financing decreases. The increase in inflation will also have an impact on the BI rate because one of the government's ways of reducing inflation is to increase customer's interest in saving funds at financial institutions. However, in this study, there was no inflation of BI rate on Indonesian Foreign Exchange Sharia Bank since the interest principle is not used in its operations, but rather determines the profit-sharing on savings and investments.