Model Risk Under CECL: A Consumer Finance Perspective

We examine the challenges of economic forecasting and model misspecification errors confronted by financial institutions implementing the novel current expected credit loss (CECL) allowance methodology and its impact on model risk and bias in CECL projections. We document the increased sensitivity to model and macroeconomic forecasting error of the CECL framework with respect to the incurred loss framework that it replaces. An empirical application illustrates how to leverage simple machine learning (ML) strategies and statistical principles in the design of a nimble and flexible CECL modeling framework. We show that, even in consumer loan portfolios with tens of millions of loans, like mortgage, auto, or credit card portfolios, one can develop, estimate, and deploy an array of models quickly and efficiently, and without a forecasting performance penalty. Drawing on more than 20 years of auto loans data and the experience from the Great Recession and the COVID-19 pandemic, we leverage basic econometric principles to identify strategies to deal with biased model projections in times of high economic uncertainty. We advocate for a focus on resiliency and adaptability of models and model infrastructures to novel shocks and uncertain economic conditions.