How to Estimate the Impact of CECL on ALLL

Over the past several years, I’ve spoken to a number of credit union leaders who have a basic question.  What will the impact of CECL be on my credit union’s ALLL?  Before implementing a credit union on CUBI.Pro’s Analytics CECL calculator, I will ask the client what they expect their CECL forecast will be.  The responses range from accurate to highly inaccurate.  I suspect those that are inaccurate are the result of a hopeful outlook and a lack of awareness of what FASB’s new rules are meant to accomplish.  The challenge will be, as more credit unions begin to implement CECL methodologies, gaining acceptance of the resulting forecast if the expected forecast is skewed.  This article is meant to demonstrate how a credit union can arrive at a reasonable estimation of what its forecast should be so that they are better able to evaluate the effectiveness and accuracy of forecast calculators, whether developed internally or externally.

The first thing to understand about forecasting loan loss allowances is that actual loss experiences will not be affected by what is forecast.  In other words, if the credit union makes no changes in underwriting guidelines and risk distribution, charge-offs will continue to be roughly the same as they have always been relative to outstanding balances.  Therefore, the new method for forecasting won’t change actual loss experiences.

The second thing to understand about the new forecast methodology is that, yes, it is intended that the credit union recognize the full risk of loss at the time a loan is originated.  Today, we allocate for losses at the time it appears that the loan will default.  The obvious advantage to the current method is that we are able to manage the amount set aside at the time it is most likely to impact our profitability.  The downside to the current method is that these impacts can only be ‘managed’ in a deteriorating portfolio for a short period of time before large losses become inevitable and the credit union is under-funded in their allowance account.  CECL is intended to prevent these scenarios by forcing an honest recognition of real risk at origination.  The immediate impact of this change to a credit union is going to be varied, based upon the individual credit union.

So then, how would one estimate what the impact might be for their credit union?  Allow me to demonstrate how you might make a general estimate that you could use to validate a CECL model.

I’ve analyzed a large number of credit union loan portfolios over the last decade.  Generally, the average life of a consumer loan portfolio is somewhere between 28 and 32 months.  Of course, lines of credit typically have a longer life than a personal loan, but on average, I rarely see average loan life of a portfolio greater than 35 to 36 months.  Because so many credit union portfolios are weighted heavily into auto loans, the averages come in as stated above.  For this example, we will use 30 months.  For your own calculation, you may wish to take loan types into consideration.

Because, as previously noted, charge-offs are not going to change necessarily because of a change in forecasting,  our past charge-off history provides us with some guidance in establishing an estimate.  Let’s say your credit union has consistently charged off $2MM per year, on average over the past three years and your portfolio has remained stable over the same period of time.  Using these parameters, your monthly charge-off amount will be $166,667 per month.

Now that you have the average loan life and the monthly charge-off amount, you can calculate an estimated loan-life loss amount.  By multiplying $166,667 by 30 months, you come up with a total of $5MM.  When implementing a new CECL calculator, you should estimate that the new methodology should provide you with an ALLL calculation somewhere around $5 Million.  That may or may not be good news depending upon how you are currently managing your ALLL account today.  Let’s see how this change might impact a credit union in real life.  For this, we are going to look at two different credit unions.

Credit Union A (CUA) is a roughly sixty-million dollar credit union with an average outstanding loan balance of $43MM between 2016 and 2018.  CUA’s average loan life is 24 months.

	2016	2017	2018	Average
Total Loans	$38,000.00	$45,000.00	$46,000.00	$43,000.00
Allowance	$220.40	$162.00	$345.00	$242.47
Net Charge-Off	$120.60	$146.80	$314.80	$194.07

Credit Union B (CUB) is a $250MM credit union with an average outstanding balance of $191 Million between 2016 and 2018.  CUB’s average loan life is 29 months.

	2016	2017	2018	Average
Total Loans	$200,000.00	$194,000.00	$180,000.00	$191,333.33
Allowance	$2,200.00	$2,522.00	$2,124.00	$2,282.00
Net Charge-Off	$1,985.80	$3,232.30	$2,045.00	$2,421.03

Now let’s calculate an estimate of ALLL under CECL methodology using the calculation provided above.

CUA

$194 / 12 * 24 months = $388

CUA’s new estimated ALLL is $388 Thousand.

CUB

$2421 / 12 * 29 = $5,850.75

CUB’s new estimated ALLL is $5.851 Million.

One could imagine that these two credit unions will have different reactions to the estimate. CUA will be pleasantly surprised that their allowance is not much more than what they have recently been setting aside.  Their new predicted allowance is $388 Thousand compared to an actual of $345 Thousand.  However, CUB will more than double their current allowance.  Why is the difference so significant?  There are some obvious answers to that question.

First, under current practices, ALLL calculations can be more art than science, if you will.  You can observe that CUA is actually setting aside more than one year’s total losses in their allowance.  This could be because they are acknowledging rising losses as a percentage of outstanding loans, or because the portfolio itself is growing.  CUB is setting aside less or near their annual total losses.  CUB is likely more optimistic about their future performance based upon recent changes in underwriting and a shrinking portfolio, even though losses as a percentage of the outstanding portfolio are rising.  But one other significant factor is the difference in the average loan life.  Because CECL uses a loan-life forecast, it must be considered in any estimate.

This example is intended to demonstrate how a credit union’s expectations of how the new CECL methodologies will impact their ALLL are highly subjective, dependent on current strategies.  For that reason, one should consider that new methodologies may produce a significantly different result than what is currently being calculated.  That doesn’t mean that the new model is wrong, it is simply an indication of how different your current methodology is from the new methods prescribed under CECL.  But, to be clear, simply asking other credit unions what their expectations are is not a good approach because you will have little awareness of their current processes.

The good news is, if the change is significant, you have the first year to take the full blow.  After that, you should find that your actual allowance funding should be little different than what it is today on a monthly or quarterly basis.  The reason for that is again, barring significant changes in loan quality, your actual loss experience shouldn’t be significantly different.

CUBI.Pro works with credit unions across the country to support a data-driven decision culture and provide meaningful business intelligence that allows CU leaders to make better and more timely business decisions.  If your credit union is looking to enhance its business intelligence for decision making, visit www.cubi.pro for more information or email michael.cochrum@cubi.pro.