ANALISIS FAKTOR-FAKTOR YANG MEMPENGARUHI NON PERFOMING LOAN (NPL) DI INDONESIA (Studi Pada Bank Umum Konvesional Yang Terdaftar Di Bank Indonesia Tahun 2019-2020)

Abstract

ABSTRACT Conventional Commercial Banks are vulnerable to non-performing loans caused by the COVID-19 pandemic that has hit the Indonesian economy. Although bank management has made efforts based on a credit rating of 5C + 1 C, it is still potentially exposed to credit risk. The purpose of this study was to determine how much influence the variable Return on Assets (ROA), Capital Adequacy Ratio (CAR), Operational Costs to Operational Income (BOPO), Loan to Deposit Ratio (LDR) and Bank Size (SIZE) on Non-Performing. Loan (NPL) at a banking company listed on the IDX. This study used a sample of 15 conventional commercial banks in Indonesia for the period 2019-2020. The sampling technique used was purposive sampling with the following criteria: (1) Conventional Commercial Banks registered with Bank Indonesia for the period 2019-2020. (2) Conventional Commercial Banks that have the financial reports needed in the research for the 2019- 2020 period. Data obtained from the annual reports of each bank for the years 2019-2020. The analysis technique used is multiple linear regression and hypothesis testing using t-statistic to test the partial regression coefficient and f-statistic to test the effect simultaneously with a significance level of 0.05. Before being tested with multiple linear regression, a classic assumption test was first carried out to test the normality of the data. The results show that there is no storage from the classical assumption test. These things indicate that the data available in this study have met the requirements for use in multiple linear regression models. The regression analysis shows that partially ROAt-1 has a negative and significant effect on NPL, while LDR and CAR have a negative and insignificant effect on NPL. In addition, other variables such as SIZE and BOPO have a positive and insignificant effect on NPL. The regression estimation results show the predictive ability of the model is 65.5%, while the remaining 34.5% is influenced by other factors outside the model that have not been included in the study.