Best practices for banks when implementing q-factors

Now that current expected credit losses (CECL) requirements are in effect and the calculations are in place, many banks are focusing on finalizing their qualitative factors (q-factors) frameworks including how to calculate and apply the adjustments within their new models for calculating allowances. As call report deadlines approach, there’s still time to review and recalculate q-factors.

In the past, when banks were calculating their q-factors, they often times were subjective in the weight and risk level they gave to different quantitative and qualitative factors. They could say, for example, if a recession were to happen, their reserves would be affected by an amount bank leadership determined it should be, and if they had a leadership crisis or some other pressure, they could say it would be affected by this other percentage that they decided it should be. Typically these adjustments were varying amounts of basis points, which would not have to be backed by historical or external third-party data.

With CECL, financial institutions must incorporate “reasonable and supportable forecasts” into estimations of their lifetime expected credit losses, meaning they have to develop a quantifiable method to establish the risk ratings for the q-factors they are using for their allowance calculation. In other words, they have to quantify the qualitative.

The first step in quantifying the qualitative is to ascertain its maximum loss scenario. Many banks will look at the previous financial crisis to calculate this maximum loss amount in order to use that as a “ceiling” for resolving how much to allocate to q-factors. Then it can consider which q-factors its operations. Previously, most banks were applying all nine regulatory qualitative factors. However, with the introduction of forecast-based CECL models as well as the requirement to incorporate “reasonable and supportable forecasts” into the model, many banks have used their CECL implementations to reevaluate the nine factors and gauge if any are either currently being incorporated into the calculation or if any do not pertain to the institution any longer.

Using relevant internal or external series data, the selected q-factors will then be assigned a risk level based on how risky they are to the bank at that time. Many banks are incorporating “low,” “moderate,” and “high” risk environments when developing their Q factor framework. “Low” risk environments would translate to a small percentage of Q factor adjustment or many times none at all. As risk increases based on the qualitative factor, the bank would move from “low” to “moderate” to “high” which would continue to increase the Q factor basis points added to its CECL estimate. We recommend the bank look to historical data when determining these risk levels. For instance, what was the highest level of delinquencies that the bank sustained? This metric can then be used as the “high” risk level for the trends in delinquencies qualitative factor. On the other hand, where does the bank typically operate in a normal environment for delinquencies (i.e., a five-year average may be 25 basis points of its total loan portfolio)? This can then be considered its “low” risk environment. “Moderate” risk would then be somewhere in the middle.

This type of analysis should be completed for each of the qualitative factors that the bank will use and be supported by internal or external data. From here, banks will evaluate each of their qualitative factors and the metrics used for each will resolve where the bank lands as far as a risk level. With the combination of the “max loss” scenario as well as the risk level, the bank’s qualitative adjustment will automatically calculate.

For example, a bank establishes the allowance ceiling for its max loss scenario is 3% based on how it performed during the last financial crisis. In current economic times, the model calculation estimates a lifetime loss amount of around 1%. The bank would have a Q factor range of 2%, meaning that if all q-factors were in a “high” risk level, the qualitative factor would equal 2% plus the 1% for the model, equaling a total of 3% which is consistent to how the bank performed in the last high-risk scenario. Putting one factor to practice, let’s say the hypothetical bank lost its team of lenders who had been there for years. Because one of the q-factors assesses the experience, ability and depth of the relevant management and staff, that will put that Q factor in a much higher risk rating than it has been before. But that’s only one of the factors, the bank would continue this exercise across all of the qualitative factors.

To ensure accuracy and avoid any double counting of risks, banks should check to see if any aspects of q-factors already reside in their models. Sometimes they are already forecasting things like the unemployment rate. In that case, a bank could double count that risk if it incorporates it separately into its calculation and also accounts for it in its qualitative factors.

Although putting CECL in place has given banks some additional work, it’s also given them clearer forecasts supported by specific well-documented and defensible data.

Baker Tilly’s CECL specialists have been addressing new developments with articles and webinars. For more information on CECL, visit our resources page, including webinar recordings that address q-factors in particular. Join us for our next CECL webinar on April 19, “Crossing the CECL finish line with a model validation”.