Analysis of the Effect of Macro and Micro Variables on Non Performing Loan (NPL) Level in PT Bank Mandiri (Persero) Tbk

Bank Mandiri is one of the banks whose job is to extend credit to the public, both through individuals and groups. In the process of lending to the public, Bank Mandiri is faced with a low rate of return on funds that have been lent (non-performing loans) or non-performing loans, where the low rate of refund (NPL) is influenced by several variables, such as macroeconomic variables, such as inflation, economic growth and interest rates and microeconomic variables, such as ROA, ROE, BOPO and CAR. The amount of credit extended by Bank Mandiri during 2008-2018 experienced an increase in value / volume, but was also accompanied by an increase in Non Performing Loans (NPL) experienced by Bank Mandiri, causing a high NPL ratio, where this problem financing appeared due to the low desire of customers to pay debts according to the due date. Data collection in this research was carried out through a documentation study and observation approach with quantitative descriptive research type where the data used were data sourced from quarterly reports of independent banks in 2008-2018, as well as quarterly inflation data, data on quarterly economic growth and quarterly interest rates that were determined by Bank Indonesia. Data analysis method used is to use the Test of Goodness of Fit analysis. The results showed that simultaneous macroeconomic variables, such as inflation, economic growth and interest rates, as well as microeconomic variables, such as ROA, ROE, BOPO and CAR had a positive effect on NPL level variables. Partially macroeconomic variables, such as inflation and economic growth do not significantly influence the NPL level variable, while the interest rate variable has a positive effect on the NPL level variable. For microeconomic variables, variables such as ROA, ROE, BOPO and CAR each influence the NPL level variables. Based on the coefficient of determination test microeconomic variables have a very large influence on the level of NPL, when compared with the NPL level variable.