Credit Loss Model

Abstract

Due to concerns from financial institutions and other organizations against the delay of recognizing credit loss, the Financial Accounting Standards Board (FASB) issued a new standard titled “Current Expected Credit Loss (CECL).” FASB requires organization to measure the expected credit loss at the reporting date based on historical experience, current conditions, and reasonable and supportable forecasts. This standard was issued to replace the “Incurred Loss Model” and to accommodate early recognition of credit losses in a timely manner. However, would these organizations be able to determine the credit loss line item? What factors and variables would they have to put into consideration to estimate credit loss at a given point in time? The objective of this project is to devise a statistical model for financial institutions that gives an estimated credit loss amount. We proposed three new approaches to derive the estimate; the best of these alternatives will be selected. The theoretical model is based on a Cobb-Douglas elasticity structure. Preliminary findings indicate that the elasticity has significance at .01 level. As a result of the significance, last year's elasticity can be used to predict this year's expected loss. This approach is worthy of developing into comparisons by region, size and individual bank quarterly forecast analyses. The goal of this research is to derive the best loss estimation model for prior to the effective date of the st