Preparing for the Storm: What Should Banks Be Thinking About If Credit Quality Worsens?

By Joseph Jalbert June 11, 2024

While many institutions and industry experts anticipated a credit downturn during the turbulent years of the coronavirus pandemic, high levels of credit losses have yet to materialize.  That said, there have been recent signals and concern that credit quality may start to crack.  Credit card and auto loan delinquencies have risen to pre-pandemic levels.  Loans secured by office space continue to be of paramount concern.  And, of course, there are concerns about the repayment capacity of borrowers whose loans are due to reprice at rates considerably higher than what they are currently paying.

With this backdrop, it is critically important that institutions have the proper processes and controls in place to ensure deterioration in credit quality is identified timely and expectations of credit losses are appropriately captured in financial reporting.  What are some steps institutions can take now to be best prepared?

Track Those Modifications

 For most institutions, along with the adoption of the current expected credit losses standard (“CECL”) came the elimination of the concept of Troubled Debt Restructurings, or what we affectionately know as TDRs.  While TDRs are unlikely to be missed, institutions will need to establish mechanisms to track their much wordier counterpart: loans modified to borrowers experiencing financial difficulty. 

 In a credit environment where institutions have low levels of modifications, this may not be a tall task; however, this would be an opportune time for institutions to review their processes and systems to ensure they are able to capture complete and accurate modification data for internal and external reporting purposes, particularly if volume increases.  This is information that is of significant interest to various stakeholders including the Board of Directors, regulators and auditors.  Management will also need this information to properly assess expected credit losses, as discussed below.

Evaluate Potential Credit Losses

 Most financial institutions went live with their CECL models not much more than a year ago and have yet to see how the model will react in a stressed credit environment.  One solution to this is stress testing, where management models stress scenarios in order to understand the impact of severe economic downturns on credit losses.  Management can also use the results of these tests to assess the resilience of the institution’s capital and loss-absorbing capacity.

Another valuable tool for management to leverage in order to fully understand their CECL model is sensitivity analysis.  This process helps management to identify particularly sensitive assumptions within the model and understand how changes to those assumptions will impact the outcome.  For example, by how much would the target reserve for credit losses increase or decrease if prepayment rates were adjusted?  Running various scenarios allows management to determine whether the model result changes commensurately with expectations or whether recalibration of the model is necessary.

Under the CECL guidance, codified in Accounting Standards Codification (ASC) 326, institutions are required to continuously evaluate and manage their loss drivers to ensure a continued correlation with actual loss experience.  Management should regularly ask, “Do(es) the loss driver(s) we selected continue to be the most relevant, or are there other variables we should be considering?”  Many of our clients, for example, selected unemployment, GDP, or a combination thereof as their loss driver(s).  While this may have been perfectly reasonable when CECL was first implemented, management is required to defend these decisions on an ongoing basis.  This will require the collection of high-quality data on loan performance, borrower characteristics, macroeconomic factors and historical loss experience.

BNN Note: many readers may be using a software solution for their CECL model, but vendor overreliance should be avoided.  While your vendor may provide an updated loss driver analysis on an ongoing basis, don’t assume this is the case.  We strongly recommend institutions maintain a checklist of model-related items that should be updated, or, at a minimum, evaluated and at what frequency.  This will help ensure critical updates don’t fall through the cracks.

ASC 326 further requires institutions to evaluate expected credit losses on individual loans when they possess unique characteristics as compared to identified pools within the portfolio.  This requires institutions to develop processes to identify loans that meet the criteria for individual evaluation and can include automated systems to flag loans based on size, risk rating changes, or other relevant metrics.  Such guidelines, however, should not be so rigid as to exclude loans that clearly exhibit unique risk characteristics but simply did not meet one or more criteria.

BNN Note: This remains a relatively new process for many institutions which will likely undergo change as time passes.  What won’t change, however, is the importance of documentation to support management’s decision-making process for identifying and evaluating individual loans.  We recommend institutions periodically reassess their portfolio to determine whether additional loans should be individually evaluated based on additional criteria and ensure these decisions are memorialized.

Controls, Controls, Controls

Underpinning all the above must rest a sound control environment that allows for the capturing of critical data necessary to properly evaluate credit losses.  Institutions should continuously evaluate controls surrounding data integrity and completeness, model governance, key model inputs and assumptions, and monitoring and reporting.

BNN Note: Continuously is the operative word here.  Controls are seldom set-it-and-forget-it as systems, processes and people change over time.  As this occurs, key stakeholders will want assurance that documented processes are similarly evolving and continue to be adhered to.

Be Prepared

 A culture of strong credit monitoring is critical in any environment, but its importance becomes particularly acute when credit quality softens.  Identifying potential problems early allows institutions to take proactive measures to mitigate risks before they escalate to significant financial losses.  These measures also allow management to better evaluate expected portfolio losses and incorporate these estimates into a model that can stand up to challenges from stakeholders.

Institutions that instill a controls mindset and continuously challenge existing processes will find themselves better positioned to weather whatever storm may come.