Photo Credit: Johnny Berg

How do we define an adequate level of ALLL reserve? Alas, that is the new $64 question in the banking industry. The ALLL Reserve is apparently like a priceless art piece, which is subject to significant scrutiny and different interpretations.   

Based on recent regulatory emphasis, the changes in the economic climate, the uncertainty in the real estate market and the economy have all cumulated into the most confusing and erratic environment regarding the establishment of a bank's ALLL Reserve. The regulatory scrutiny has been tremendous.  Of course, along with such scrutiny brings an element of inconsistency which is derived primarily by the regulatory desire to increase the ALLL Reserve.  Such desire is often in direct conflict with the Bank's, whose main objective is to not only maintain an adequate reserve balance but also to preserve earnings and capital levels.

The 2006 Inter-agency Policy Statement on Allowance for Loan and Lease Losses combines with other regulatory and accounting directives to establish guidelines and requirements for the accounting of projected loan losses and determining loan loss reserves.

So what's the problem? Where does this subjectivity come from? The problem is that these requirements and guidelines are themselves pieces of art subject to varying interpretations. One would think that all of these regulations would be able to create a concise and simple formula that all parties could be content with.  But then we'd have nothing to complain about.  Now at least we can blame someone else who doesn't see the glass as half full and insists that it's half empty.

The directives require banks to implement a consistent and verifiable method for determining ALLL - and be able to demonstrate that their process results in a reasonable and justifiable loan loss reserve. The Inter-agency policy requires banks to approach the ALLL reserve calculation using two distinct perspectives, the guidelines of using Financial Accounting Standards (FAS) No. 5 (Accounting for Contingencies, for grouped loans) and FAS No. 114 (Accounting by Creditors for Impairment of a Loan, for individual loans).

To complicate matters further, the Financial Accounting Standards Board (FASB) recently released the Accounting Standards Codification.  So now we all we have to get used to saying ASC 450 for FAS 5 and ASC 310 for FAS 114. This happened, of course, just as we all discovered that there's indeed a FAS114. I didn't know they went that high. So I'll try to use them going forward in this blog, although I can't promise that I will be able to correctly. 

But as is any good art form, the beauty is in the eye of the beholder. The issue in many cases that I see is not even the dollar amount of the ALL reserve, but being able to substantiate it.

The ASC 450 (I got it right, FAS 5) is perhaps where the greatest challenge with respect to documentation comes in.  The guidance tells you the following:

• The Bank should group its loans according to type or homogeneous pools of similar risk characteristics. That's easy enough. The practice seems to allow the use of call report categories. Unless a Bank has maintained better historical records, it is difficult to get prior historical quantitative data relevant to charge-offs or delinquency trend by any other sub categories.

• The Bank must calculate its historical loss rate, based on the more recent history, for each of these loan pools. Of importance here and throughout its process to determine ALLL, the Bank must document the method it uses to establish its loan pool historical loss rates. Of course I modified the guidance here a bit. The Interagency guidance doesn't really say that a bank must use the more recent history. This is just something the regulators started to enforce on their own. The understanding being that since there have been more charge-offs in recent history compared to the past, the use of the recent two year history (as opposed to a five year history) would yield a higher charge-off rate, hence a higher reserve.

I wonder what will happen when the charge offs start to decline - will we have to revert back to a five year history? Hmmm!

• The Bank must adjust those historical loss ratios for internal and external qualitative factors. This is where I think most banks fumble. This is not so easy to justify, quantify or document. I can only suggest that a bank read the Interagency guidance. It is actually pretty clear in identifying the qualitative factors that should be used. Convert them into a questionnaire of sort, and assess a risk rating for each factor. The resulting risk rating of each factor can then be used to dial up or down the historical loss percentage.

• The Bank must be able to identify loan loss trends as well as additional issues or factors that affect the quality or risk assessment of its loan pools. I read this as being a requirement to further alter the adjusted historical loss rate by an assessment of the delinquency or concentration risk trend. Regulators have also mandated that classified for each homogenous pools be further segregated, and those loans be reserved using an even higher adjusted historical loss percentage. This last part I couldn't find anywhere in the Interagency guideline.

I'm sure it's there somewhere. I need glasses but have just been pretending that I don't. When I get my new glasses and find it, I'll let you know.

• The Bank must calculate the ALLL for each loan pool to determine its total amount of loan loss reserves. This is the easiest part. Once we have the adjusted historical loss percentage we can easily apply it to the outstanding loan balance for each loan. Wait! Do I include the off balance sheet items and apply it to the portion of loan balances for which I have already performed an ASC 310 (yup, it's FAS114) review?

Off balance sheet items (i.e. unfunded credit lines, loan commitments, etc.) - if you follow the interagency and GAAP guideline, the answer would be no.  The regulatory emphasis says yes.  So I would say YES.

In the case of loans for which an ASC 310 impairment measurement review has been performed (these would be loans which you have determined are impaired and have measured the potential impairment, even if it is none) - if you follow the Interagency guideline the answer is no. The regulatory emphasis says maybe, it depends.

So what do I say? In a bit of defiance, I say "no".  The inconsistency from the regulators, I believe, stems from their confidence level on the bank's ASC 310 review process. The stronger and more objective the process, the more they are inclined to agree with me (imagine that). We'll discuss more of that in Part 2 of this masterpiece of a blog (I know you can't wait).

So to summarize, the ASC 450 reserve balance is derived by:

• Take your more recent historical loss percentage for each pool of loans (most likely two years, but that could change at any time) and adjust it by the risk assessment for the qualitative factors listed in the Interagency policy.
• The adjustment for these qualitative factors needs to be documented through some sort of an internal/external assessment
• The adjusted historical percentage after the adjustment for the qualitative factors can be further adjusted for delinquency and concentration risk for each loan pool (or may not if you don't want to, just a suggestion)
• But you should, even though not required to according to the Interagency policy, further escalate the adjusted, adjusted historical loss percentage for the classified portion of the loans for each loan pools.
• Now we have the adjusted, adjusted and further adjusted historical loss percentage which we can apply to the outstanding loan pools.
• You should include the off balance sheet items for each loan pool (you don't have to if you don't want to, but you can fight with the regulators yourself)
• I wouldn't include the loans for which you have performed a valid ASC 310 impairment review. I would make sure that my review process for ASC 310 is robust. How does one accomplish that? We'll touch upon that in Part 2

So now you can be the new Picasso.  Good Luck.

(Read Part 2, "Cracking the Code: How to Develop the Right FAS 114 Methodology" here)

Comments