SECU Lending Performance: A Little Struggle Over Transparency...

   Better Things To Do?

There was a locking of horns over SECU lending yesterday in the comments [link]. Makes for some mindless reading, apologize for wasting your time. 

Here's the condensed version.  Much of the controversy in this blog since the arrival of the "new/new" in 2021, has been over changes in SECU lending policies and practices. Not surprising, since at heart, lending is what SECU is all about. 

Centralization, abandonment of local, branch decision-making, risk-based lending, the elimination of credit committees, failure of effective collections are among the changes questioned. Weakened financial performance, with greatly elevated loan losses and delinquency, seem to justify the concern.

✅ The latest brouhaha centers around the level of delinquency on SECU loans. At 12/31/2024, SECU 60+ day delinquency surpassed $1 billion ($1.15 billion) for the first time in history. But by 3/31/2025 the 60+ day delinquency had dropped an incredible "$591 million (down 51%!)" to $556 million. Simply asking how that was achieved set off a range war of opacity. Not sure why.

Eventually it was pointed out that many SECU loans had been "modified" (a process called "troubled debt restructuring" or TDR) between 12/31/24 and 3/31/2025. Basically SECU borrowers who have experienced legitimate financial hardships have their loans "modified" and brought current. A fresh start, a great process, again called a "TDR".

😎 By the way, a TDR is the best argument ever against risk-based lending and for member credit committees. If it is quite reasonable to listen to and help members who have financial problems after a loan is made, why isn't it equally reasonable to listen to and help members who have had legitimate financial problems before a loan is made?

✅ So everything's good now? Hard to say, here's why:

At 12/31/2024:   60+ day delinquency:  # loans - 27,000   balances - $1.15b

At   3/31/2025:   60+ day delinquency:  # loans - 17,000⬇ balances - $ 556m⬇

At 12/31/2024:   SECU TDR's                  # loans - 13,700   balances - $256m

At   3/31/2025:   SECU TDR's                  # loans - 16,400⬆ balances - $479m⬆ 

❋ SECU appears to have modified and brought current only @ ⬆2,700 member loans (16,400 - 13,700) in an amount of $223 million ($479m - $256m), while the total # of delinquent loans dropped by ⬇10,000 (27,000 - 17,000).

 ❋ TDR balances increased by only ⬆$223 million ($479m - $256m), while total delinquent balances declined by ⬇$590+ million ($1.15 b - $556m).

😎 Something still doesn't seem to quite add up. Even after subtracting for the TDR's, how were the # of delinquent loans reduced by 7,300 (⬇27%!) and delinquent balances by @$370 million (⬇32%!) in just 90 days? 

 Hope the answer is: "Great management!"

 

Incredible results? 

 

 

SECU Analytics: "We're #3! Don't Worry, Be Happy! ... And Be Quiet!"

   

     An anonymouse integrity check!

 
✅ From a commenter: "I'm offering you a free, and admittedly optional integrity check".

"Recall that the Callahan analyst, in content you published on this blog, questioned your comparison set, and suggested that comparing a 50 billion+ CU to all CU's > $500 million [the "peer group" is defined by NCUA - not me!] wasn't very meaningful or appropriate."

"Here are the 12 month rolling loss rates through 1Q 2025 for the 10 largest CU's ranked by asset size.

1) Navy 2.48%
2) SECU 0.66%
3) PenFed 1.76%
4) SchoolsFirst 1.05%
5) BECU 0.59%
6) Golden 1 0.76%
7) America First 0.99%
8) Alliant 1.04%
9) Mount. America 1.25%
10) Randolph Brooks 0.57%"
 

"* SECU has the third lowest loss rate, and is only 9 basis points higher than RB, who has the lowest loss rate at .57%."

 

✅ The commenter would like to compare SECU's loan loss and delinquency performance against the top 10 credit unions which is reasonable and that comparison looks good. 

😎 The SECU loan analytics folks seem increasing desperate to explain away the lack of performance since 2021 - they claim it was caused by inflation, soaring interest rates, adjustable rate mortgages, a leap year, wrong "peer" group, inaccurate official data, Mike Lord, and those unreliable members! 

 

Of course, we all understand that data can be viewed in many different ways. How does one determine which data is most important? The best answer is to focus on the data which affects SECU members the most.  Does the SECU member care about Navy's loss rate or Randolph Brooks'? The answer is "No, not at all" - it doesn't affect the SECU members in any way!

 

✅ The following info does directly affect SECU members - in a large, hard dollar way:

 

😎 The SECU charge-off rate for 2021 was  .20%, which was $50 million.

     The SECU charge-off rate for 2022 was  .35%, which was $95 million.      

     The SECU charge-off rate for 2023 was  .62%, which was $197 million.

     The SECU charge-off rate for 2024 was  .68%, which was $233 million.

❋ Charge offs for first quarter 2025 were $69 million, which if "annualized" (x 4) would mean a projected $276 million in losses for 2025.  

✅ And then there is this:

😎 The SECU 60+ day delinquency on the March, 2021 Call Report was $210 million. 

     The SECU 60+ day delinquency on the March, 2022 Call Report was $216 million. 

     The SECU 60+ day delinquency on the March, 2023 Call Report was $470 million.

     The SECU 60+ day delinquency on the March, 2024 Call Report was $700 million.  

     The SECU 60+ day delinquency on the December, 2024 Call Report was $1.15 billion!

     The SECU 60+ day delinquency on the March, 2025 Call Report was $556 million.*

 

 ❋ That is an improvement! SECU delinquency levels are now a little over twice as high as in 2021 ($210 million), rather than over three times as high at March, 2024 ($700 million).

 

✅  The "LA" folks proclaimed the following: "* Delinquency dropped $591 million, down 51% from year-end! "

 

As a little "integrity check", hope that remarkable 51% drop in delinquency would be explained.  How was that accomplished? What new innovative collection techniques were adopted in the last three months? Will equally dramatic improvements occur in the second quarter?

😎 Lets hope an "explanation of integrity" is forthcoming and that the 51% drop is real. Why? First because $1.15 billion is a big number!  And, because if 51% drop is not real and you add it back, SECU's delinquency ratio would double...

New math? ... or just "new/new"?