In a recent webinar I participated in with SAS we discussed Economic Impact Analysis (EIA). While EIA is similar in concept to stress testing, its main goal is to allow credit unions to move quickly to evaluate economic changes to their portfolio—such as those brought about by a crisis like the COVID-19 pandemic.
There are four main components to EIA.
- Portfolio data: At a minimum this needs to be segment level with loss performance through time. If needed, this data could be augmented with external data
- Scenarios: Multiple economic possibilities are necessary to help assess timing and magnitude of potential, future loss
- Models or methodologies: These are required to link scenarios to the portfolio to forecast loss amounts
- Visualization of results: This is essential to clearly understand the portfolio loss behavior. While tables are useful, nothing illustrates odd behavior better than a picture (e.g., a box plot or tree map or histogram).
A credit union looking for a practical approach for getting started should consider the following steps:
- Start with segment level data instead of account level. This should reduce the common complexities that arise when sourcing and cleaning account level data.
- Develop segment level models or methodologies to capture the impacts of macroeconomic changes. These can be simple provided they incorporate relationships to macroeconomic elements.
- Create multiple scenarios. The more the better. Different scenarios will provide different insights in how the portfolio reacts to changing macroeconomic environments.
- Execute models and explore results. This is where (I believe) the fun begins. Be curious – change portfolio assumptions (e.g., growth or run-off), and scenarios, to see how losses will react.
Now is the time to act, to gain an understanding about the economy’s impact on one’s portfolio. But it is worth mentioning this is also an investment into the future. As mentioned earlier, EIA has its roots in stress testing. By creating an EIA process now, a credit union not only better positions itself to build a robust stress test platform but also has the foundation to tackle CECL.
To view the webinar on demand, please visit NAFCU.
Jonathan Leonardelli, FRM, Director of Business Analytics for the Financial Risk Group, leads the group responsible for model development, data science, and technical communication. He has over 15 years’ experience in the area of financial risk.