In our latest HealthScore calculation we find some expected trends, such as a continued decline in the overall credit union HealthScore, but also an unexpected surprise: membership growth (MG). After seeing two quarters of year-over-year score declines, MG sprang to life in Q3 with a 6.25% score improvement.
Before we explore trends, I should call out one important point about our HealthScore model that people sometimes miss. Our scores, the HealthScore itself and the 17 component scores, are a reflection of the average health of the industry based on the health of its parts, a perspective we develop by first scoring the health of each individual credit union. We do not score an aggregated credit union industry balance sheet or income statement.
This is important to understand because often you will hear that the industry is larger than it has ever been, or is growing loans at a faster clip than it ever has, or is serving more people than it ever has (see this recent CU Times article as an example).
This can certainly all be true, but if you assess the health of the industry based on the aggregate, you miss the nuanced impact of the individual institutions comprising the industry. Thus, the credit union community can be the biggest and best it has ever been in aggregate, but at the same time be composed of hundreds of credit unions that are poor performers likely to contribute in the near future to the continued decline in the total number of credit unions nationwide.
The chart below illustrates this to a degree, comparing the total number of credit unions (left axis) with the health of the credit union community (right axis).
The chart takeaway is this: as less healthy credit unions go away, fewer but healthier credit unions remain.
In Q3 2022, the HealthScore saw its third straight score decline. What makes this notable is that aside from a decline in Q1 2020, the HealthScore has not experienced successive year-over-year drops since 2013. That is eight years of continued improvement.
We’ve had similar historically positive experiences with Delinquency (DL) and Charge Off (CO) scores, and in similar fashion, those positive trends have reversed for both metrics. There is much to be said about the DL score, but I want to focus specifically on the CO score in this post.
Last quarter I wrote that while the CO score still improved year-over-year, the pace of improvement had slowed greatly. The implication was that CO would turn negative, which it did this quarter – as expected.
The good news is that scores for CO have been excellent in recent years. Case in point: even with a 1.42% year-over-year score decline, the raw CO score is 7.66 out of 10. That is really strong, especially considering that the industry has seen scores as low as 4.52 (you remember the Great Recession, right!?).
Of course the question in front of us is: will it stay there? The answer is probably not.
You may recall from past posts that when we created the HealthScore model, our scoring system generally attributed a score of five to average industry performance. This means that credit unions with a CO score of 7.66 are way above average. Unless you believe that the fundamentals have changed so much that historical averages are meaningless, we’ll find our way back there. The industry average score for CO is 6.12 – unless you remove the COVID years, in which case the industry average is 5.95.
Given that the DL score experienced its second straight year-over-year decline, I’m inclined to believe that there’s no place to go but down for our CO score. And depending on how quickly the CO score falls to the mean, we could see continued pressure on return on assets (RA) scores that serve to extend its own year-over-year score decline trend (the Q3 RA score is 4.95 – slightly below the five threshold – and it has declined year-over-year for two straight quarters).
All this, despite seven straight quarters of positive year-over-year loan growth scores (LG), which would otherwise be a bright spot were it not for the overarching credit quality score concerns.
Let’s explore that positive LG trend because it is worth looking into, given its strength. Rather than focus on the year-over-year percent change, let’s explore the raw score. Remember again that a score of 5 is generally considered average based on credit union historical performance. With that in mind, take a look at the LG scores in the table below.
Quite the upward trajectory! At the end of the day, if the Fed can tame inflation without a nasty recession and corresponding job losses, it is possible that we could maintain positive loan growth while keeping DL and CO scores above average – and perhaps then see that negative RA trend positive.
Note: You may wonder why, if the score averages for credit unions historically hover around five, scores are below five for most of the chart history. This chart does not contain the entire date range we analyzed to create the HealthScore model.
Oh, and back to that 6.25% membership growth (MG) score improvement I mentioned at the beginning of this post. Quite the surprise! We had two successive quarters with year-over-year declines prior to this quarter so the positive change is welcome. That said, the industry is still well below that average benchmark score of five. We’ve got work to do.
As with most of our HealthScore posts, we like to end by highlighting the latest list of Top Performers. Sycamore Federal Credit Union (Sycamore, AL) retained the top spot. This quarter also saw three new members of the Top Ten, including Whatcom Educational Credit Union out of Bellingham, WA, DATCU Credit Union out of Denton, TX, and SRI Federal Credit Union out of Menlo Park, CA.
The full Top Ten is listed in the table below. Congratulations on your excellent quarter!
|JOLIET MUNICIPAL EMPLOYEES||IL||http://www.jolietmunicipal.com||$10,134,472||821||8.765|