Credit Portfolio Risk Analysis: Loss Modeling and Estimation of Credit Insurance Premiums Ellys Agustina (a), Sapto Wahyu Indratno (b*), Elisa Murti Dewi (a)
a) Department of Actuarial Sciences, Faculty of Mathematics and Natural Sciences, Institut Teknologi Bandung, Jl. Ganesa No.10, Lb. Siliwangi, Bandung, 40132, Indonesia
b) Statistics Research Division, Faculty of Mathematics and Natural Science, Institut Teknologi Bandung, Jl. Ganesa No.10, Lb. Siliwangi, Bandung, 40132, Indonesia
*saptowi[at]itb.ac.id
Abstract
Credit disbursement in Indonesia has surged significantly, reaching IDR 7,090 trillion according to OJK data. Despite this increase, the Non-Performing Loan (NPL) rate of 2-3% indicates a substantial risk of defaults. Consequently, credit insurance is vital to mitigate potential financial losses and maintain economic stability in Indonesia. This research models losses due to default within a credit portfolio and determines appropriate insurance premiums. Credit risk is modeled through default events using an indicator function, and the frequency of default events is modeled using a Probability Generating Function (PGF). A large credit portfolio is divided into several exposure groups through discretization. Losses are expressed in terms of PGF, with the loss probability function using the PGF outcomes of event frequencies. This approach simplifies the calculation of large loss probabilities with the Panjer algorithm. The research^s findings indicate that expected losses can be calculated as the first derivative of the PGF loss distribution, while unexpected losses are determined using Value at Risk (VaR). Capital reserves are calculated as the difference between expected and unexpected losses. The credit insurance premium model is developed based on expectation, standard deviation, and variance principles. A generic simulation was conducted to check the effectiveness of the proposed model. The experiments show that the approach can be implemented easily and provides reasonable results.