The new loan loss accounting rules for CECL and IFRS 9 require thousands of organizations to learn about modeling. Likewise, accountants and others in finance are now required to learn about statistical modeling concepts. This book is intended to define terms in a manner consistent with decades of academic literature on statistical modeling and hopefully reduce some of the noise and confusion just around definition of terms. It may also serve as a useful guide to analysts new to the field tasked with IFRS 9 compliance, the international loss accounting rules, and credit risk modeling in general.
Each chapter of this book is a term that one might encounter when discussing creating lifetime loss forecasting models for CECL or IFRS 9. Not every term is a model, and some models listed are being mentioned only to explain why they are not likely to be used for loss forecasting. The CECL guidelines and subsequent FAQs have given examples of modeling techniques. Some people new to loss forecasting have assumed that those are all the available or applicable methods. This book is meant in
part to dispel that misconception.
The definitions and descriptions provided here are meant to provide an intuitive understanding across a range of modeling techniques. Mathematical derivations are kept to a minimum. The references listed will provide all the necessary details for an eager analyst.
Credit Unions can expect at least a 22% increase and Community Banks a 59% decrease in reserves as a result of CECL?
Deep Future Analytics (DFA) and Prescient Models (PM) recently conducted a joint study across 103 CECL clients to determine how much their loss reserves could change if CECL were adopted today. The results were presented in a 10/17 webinar which we summarize in this article.
Modeling Accuracy Since loan modeling is such a complex animal, the first question you should always ask is: “Can I trust the numbers?” For this, Dr. Joe Breeden, world renowned credit analyst and author of the “Living with CECL” book series, takes you through how advanced loan modeling works and the data and equations behind it. I provided a “stripped down” version of this below for those less familiar with loan modeling.
For the study we used a MultiHorizon Survival Model. This highly advanced method of computing credit risk is actually a two-step approach which adds to overall accuracy and application of the data. When you take multiple factors like Lifecycle, Environmental, and Credit Quality the information must be handled correctly or the results will be erratic. Therefore, this approach accounts for the changes in variable significance across multiple horizons. For example, delinquency is a very strong indicator of PD in the early months but then becomes less important. All in all, as we look at lifetime losses it is important to have a model that will stand the test as we look multiple years into the future.
Among community banks, the most common result (median) was a significant reduction of -59% in reserves. For credit unions the model reveals an increase in reserves of 22%. These estimates are before any Q-Factor adjustments. While the banks may be able to add Q-Factors to bring the new requirements more in line with existing practices, the credit unions will have a much more difficult time “selling” a negative Q-Factor to their auditors.
Next, we took the client data and further stress tested using the FRB Severe scenario for a near-term recession. As a result the CECL loss reserves see a significant increase with the banks going from a -59% up to +69%, and credit unions from +22% up to +107%.
Product Results To understand the differences by product, we are taking our model’s 12 month horizon and comparing to a lifetime horizon. This will measure the number of years of coverage you need under CECL, compared to what you would’ve had before.
The median ratio of CECL / 12Mo. loss rate for the individual products is essentially the average life of the loan. For example, we computed the average life of Consumer Loans to be 1.59, thus resulting in a 59% increase in reserves of the current 12 month horizon. For Auto loans the result was an increase of 135%, 55% for Consumer Line, 78% for Credit card, 140% for Residential RE, and 342% for HELOC. The HELOC is sort of an interesting case because it cannot be canceled. This poses many issues in how we manage loss estimates and this actual estimate could actually be understated.
On the commercial side, we show increases for all products as well: Ag Loans (197%), Ag Lines (2%), C&I Loans (93%), C&I Lines (140%), CRE Loans(251%), and CRE Lines(310%)
There are still plenty of institutions today that have yet to prepare for the new CECL standards. Based on these results we feel the time is now to find out your institution’s new reserve requirements so you can make the appropriate course corrections today to ensure a smooth transition tomorrow.