ANALISIS FAKTOR-FAKTOR YANG MEMPENGARUHI PROFITABILITAS PADA BANK UMUM SYARIAH DI INDONESIA PERIODE TAHUN 2010-2014

Abstract

The process of the development of Islamic banks in Indonesia is very impressive, reaching an average asset growth of more than 65% of the year, within the last five years. Therefore, the role of Islamic bank in supporting the national economy is expected to be increasingly significant. Profitability is one indicator that is of considerable concern in observing the development of Islamic bank. Profitability can be measured by using a ratio Return On Assets (ROA). ROA is the ratio that indicates the bank's financial performance benchmarks in net income from total assets. The lower ROA ratio means Islamic banks have a low level of profits and more problematic position of these banks in terms of asset utilization. This study aimed to analyze the factors that influence ROA Islamic bank in Indonesia. This study using Capital Adequacy Ratio (CAR), Liquid Assets to Total Assets (LTA), market share of financing, and Net Interest Margin (NIM) as the independent variable and profitability ratio measured by ROA as the dependent variable. The sample in this study was obtained from a purposive sampling method and obtained five Islamic Banks that qualify as research sample is BCA Syariah, BRI Syariah, Bank Syariah Mandiri, Bank Mega Syariah and Bank Muamalat Indonesia in 2010-2014. This study using secondary data obtained from semi-annual financial statements that can be downloaded from the official website of each bank and the official website of Bank Indoneisa. The analytical method used in this study is Multiple Linear Regression Analysis with a significance level of 5%. The result showed that simultaneously CAR, LTA, market share of financing and NIM have significant effect on ROA of Islamic banks. Partially, variable CAR has negative significant effect, while the LTA, market share of financing and NIM have positive and significant effect on ROA of Islamic banks. Results of regression estimation showed the predictive ability of the model 52.2%, while 47.8% were influenced by other factors outside the model that has not been covered in this research.