Sunday, October 22, 2017

TDR or Not to TDR …Much Ado about Nothing?

Posted by Amit June 13, 2011 3:20pm

Photo Credit: shuttermon

I hear from a “few” people that tackling and managing Troubled Debt Restructuring (commonly known as “TDR”) is becoming confusing based on new accounting pronouncements and ever changing regulatory emphasis. We can argue all day whether we are labeling a loan a TDR properly or not, but perhaps we’re missing the main objective of the renewed attention to this area by regulators and accountants.   At the end of the day, the issue really becomes the Allowance for Loan Loss Reserve.  I don’t want to come off as a one trick pony who keeps talking about the ALLL all the time, but it’s true. The main objective is to make sure that we are identifying ALL impaired loans.  If we identify all of the impaired loans, then we can measure the impairment and adjust the ALLL accordingly.  So let’s talk about it. Of course it goes without saying that just because we’re going to talk about it here does not necessarily mean that I am going to be able to offer any meaningful suggestions that will make it easier for you. But let’s talk.

The Delinquent Borrower

There are generally two types of borrowers that can be involved with a potential TDR situation. The first type is a borrower that has already demonstrated its inability to pay back by being consistently late on payments.  The Bank has already identified this borrower through its robust independent loan review and internal monitoring process.  And as this borrower’s inability to repay has deteriorated, so has its credit rating.  Once this borrower has made it to the Bank’s nonaccrual list or has been risk rated as Substandard, the Bank has done its job by assessing whether the loan is impaired or not. If it is deemed to be impaired, the Bank has performed a ASC310 (FAS 114) review and set up an appropriate reserve.  So far it’s pretty simple.  By now, we all know how to evaluate and set up an appropriate reserve for an impaired loan.

Now, as part of the Bank’s workout process and reaching out to the borrower, the Bank renegotiates the terms of the loan by giving the borrower some concession in the hopes that it can resurrect the borrower and the loan. The accounting impact of any restructuring for this borrower would be reviewed under ASC 310 (FAS 114).  But we had already done this previously when the loan was deemed to be impaired.  So for a real estate collateral dependent loan, there probably won’t be much of a difference in your reserve calculation since the prior review (assuming that restructuring is within a short period of the prior impaired loan assessment review).  For a loan where you have to evaluate the impact of future cash flows, there may be more of a difference between the calculation based on the new terms of the restructured loans and your prior calculation performed when you had deemed the loan as impaired and made an estimate of the anticipated potential future cash flows.  Thus, for a borrower who has already been identified as a potential credit risk through the Bank’s internal assessment process, there should not be a significant financial impact when the Bank restructures the terms of the loan and classifies the loan as a TDR.  

The “Good” Borrower

All regulatory guidance state that institutions must follow GAAP in evaluating the ALLL provision.  It’s important to note that GAAP does not state that an institution needs to predict its future losses as part of its evaluation of the ALLL.  The objective of an ALLL provision is to assess the amount of losses that are present in the portfolio as of the assessment date. Thus, the provision breaks down the calculation between two types of borrowers.   The ASC 310 (FAS 114) calculation is for those borrowers who we have been identified in our portfolio as being potentially credit risky and the ASC 450 (FAS 5) portion is for those borrowers that we know are going to incur a loss to the Bank, but we cannot, at this point, identify who they are. For the ASC 450 general pool the implication is that an event has already occurred relating to a borrower which will impact his/her ability to repay the loan, but which the Bank is not aware of yet.  Thus we are not trying to predict future impairments of credits, rather, we are trying to assess the impact of loss relating to credit impairments that have already occurred in the portfolio but which we don’t know about.  Thus, we are trying to measure the loss which is related to a borrower who has already lost his/her job today and which will impact his/her ability to repay three months from now.  Technically, if we knew all of our borrower’s supervisors and asked them to let us know when they fire their employee or if we were in contact with all tenants of our borrowers, and they let us know when they vacate their premises so that we could become aware of any vacancies as they occur, we wouldn’t need a FAS 410 calculation. We could evaluate a provision on an individual basis through the ASC 310 methodology.  But since we don’t, we need to estimate.

So why do I think this is relevant to TDRs? Hmm, I don’t know yet, but I know it is. As I said initially, there are generally two types of borrowers that are involved in a TDR related transaction.  The one is the delinquent borrower which we discussed previously, and then you have the borrower that is current on its obligations and perhaps has just lost his job or a tenant in his building.  This is the borrower who creates all of the confusion and problems when it comes to TDR.  This is a borrower that approaches the Bank and identifies himself as one that might have some problems repaying the Bank based on the previously set terms. Now, this is a borrower for whom the Bank was previously accounting for in its ASC 450 reserve.  We know that there are several other borrowers like him that are having problems, but we don’t know who they are until they either come forward to tell us or until they start falling behind in their payments. This is a “good” borrower because he is coming to us and telling us his problems. But now that the Bank has identified him as a potential problem, the issue becomes, “do I lump him in the same category as the other horrible borrowers that are dead beats and not paying us, or do I treat him differently?”

I think this is where fundamentally most banks go awry. GAAP doesn’t really care whether he is an honest borrower who comes forward and tells you his problems or one that demonstrates his problems through late payments. The accounting treatment for both is basically the same. When a borrower comes forward and explains his difficulties, and requests an adjustment to the original terms of the loan, in essence, a concession from the bank, this is a borrower we need to evaluate for impairment purposes under ASC 310.

Now you’re saying, “but wait, ASC 310 review is performed for loans which are deemed to be impaired.”  And the definition of an impaired loan is where it is probable that a creditor will be unable to collect all amounts due (according to the contractual terms of the loan agreement).  In this case, you suggest that the bank will be able to collect all amounts due, it’s just the borrower needs some help getting there. Is there not a difference between a small and a short term concession to the borrower versus one that is a significant concession? The answer is no, not really.

I think the concept of TDR, in this case, may be easier to comprehend if one thinks not of the actual modification of the loan terms and trying to label it as FDR or not TDR, but rather, think of it first in terms of the credit risk of the borrower. This is a borrower who has come to the Bank and expressed a concern about his financial capacity.  If you have identified a specific individual risk in the portfolio, GAAP says, measure it.

I think what had added to prior confusion was the issue of materiality of the concession given in determining whether the restructuring qualifies as a TDR. Regulators considered “minor tweaking” of interest rate concessions and adding missed payments to end of loan terms as not being significant and, therefore, not be constituted as conditions meeting the criteria for classifying a loan as a TDR.  This was supported by a lack of specific GAAP guidance. Thus, a combination of regulatory lax over the years, and lack of clarity in GAAP in definition of TDR, has added to the confusion of such classification.  During years of economic growth, the “minor tweaking” on loans that were not classified as TDRs, did not have a significant impact on a bank’s financial position.   However, as the economy contracted and worsened, many banks, unfortunately, started to disguise the credit risks of their borrowers as “minor tweaking” to loans.  In defense of the regulators (it’s not often that I find myself on this side), they start questioning the minor adjustments and inconsistent with their prior approach started to classify these loans as TDR. During such time, banks looking for support under GAAP found gaps in the accounting literature.  

Accounting Profession to the Rescue

In April, 2011 FASB issued a new guidance (ASU 2011-02), “A Creditor’s Determination of Whether a Restructuring Is a Troubled Debt Restructuring.”  Effectively, this guidance restricts significantly the ability of restructuring to be not deemed as a TDR.  The basic test for determining a TDR is a two prong test.  First, the borrower is experiencing economic difficulties and two that the bank grants the borrower a concession.  Thus, the overall objective is to determine if the borrower is potentially impaired.  The new FASB guidance, along with regulatory consensus, allow institutions to provide modifications in interest rates or other loan terms that are not deemed concessions as long as they are done to remain competitive with market conditions. However, some banks have used this exception to not characterize a loan modification as a TDR without adequately assessing the borrower’s financial capacity. The new guidance does specify modification in payment terms that may be deemed as insignificant and not constitute as TDR.  Basically, what this means is that the FASB realized that the intent is not to simply to provide a criteria to determine whether the modification constitutes a TDR but to align the modification with the economic substance of the modification. What good is classifying something as a TDR if the impairment analysis demonstrates no or nominal impairment and, therefore, no adjustment to the ALLL.

Suggestions

Based on my experience too often community banks look primarily only at a borrower’s payment history as an indicator of financial capacity.  If a borrower has been making payments regularly, conventional wisdom dictates that it’s a good borrower and making a concession to drop the interest rate in line with “market rate” should not be a TDR.  If you can support that the borrower can obtain the same lower rate from another source AND support the fact that the borrower’s financial condition has not deteriorated in any significant way, I agree it’s not a TDR. To avoid confusion and regulatory inconsistency, my suggestion would be to first and foremost examine the financial condition of any borrower that approaches you for a concession. Obtain all relevant financial information to make an assessment of the borrower’s credit risk.  If the assessment indicates that there has been deterioration or likelihood of default has increased based on the loan’s original terms, then you have a bad loan which you need to protect and work with your borrower in minimizing any potential loss to the bank.  To me, it doesn’t really matter if we call it a TDR or not.  A TDR is just a label.  If we use this TDR label to disguise and postpone the inherent credit risks by not measuring it effectively now, it will come back to haunt us. 

Not too often, concessions or modifications are handled by the Chief Lending Officer (“CLO”) of a community bank. Unfortunately, they are not often well versed on accounting consequences of such decisions (needless to add that not many of us are these days), and there may be hestiation on their part to label a borrower with whom they have a relationship and who has been paying well but needs some help during a bad time, as potentially impaired. So what we should do is to make sure that all modifications’ requests are centralized in the Bank and that all decisions in this regard are discussed and approved by relevant parties within the Bank. A CFO who is also responsible for the ALLLL calculation should not be surprised about a loan modification which was not properly labeled, or which an impairment analysis was not performed during a regulatory exam. Based on recent regulatory exams, we are learning the need to document all decisions and to ensure that all parties within the institution are involved in all modifications. If we decide that we’re not going to classify a modification a TDR, what we’re really saying is that this is not an impaired loan.  If that’s the case, then let’s make sure we document our decision and how we got there. The reality, unfortunately, is not that the regulators have increased their scrutiny in this area, they certainly have and they should, but that banks have taken this area very lightly. The regulators have learned from the banks that they have closed that not too often what was deemed by banks as minor concessions to help a borrower in need were actually serious credit risks which were not adequately analyzed, labeled or measured.   Community bankers need to, unfortunately, learn the same message.

So once we classify a loan as a TDR and perform an ASC 310 (FAS 114) analysis and measure any potential impairment to properly modify our ALLL, how do we classify this loan?  Do we have to place it on a nonaccrual status? GAAP does not address classification of loans and regulatory guidance and actual practice has been inconsistent.  The basic rule is if the loan was paying consistently prior to the restructuring and continues to pay on the restructured terms, there should be no reason to place the loan on a non-accrual list.  Having said that, I have heard from banks where their regulators have insisted that such loans be placed on a non-accrual list or be classified as Substandard.   While not having been part of such discussions, I believe what the regulators are really saying in such situations is that they do not have confidence in the bank’s restructured terms.  In other words, they are suggesting that, based on the borrower’s financial capacity analysis, the modified prepayment terms are in excess of the borrower’s ability to repay. Thus while the Bank may have done a good job in renegotiating the terms, those terms do not support the borrower’s ability to repay.  On the other hand, if the bank’s analysis supports the new terms and the borrower demonstrates the continued ability to pay, then there should be no reason to classify such loan as a Substandard or place it on the non- accrual list.  Of course, don’t tell them that you read it here.

The other question that gets asked a lot as well is assuming that a TDR is placed on the non-accrual list (because the borrower had either previously demonstrated a weak payment history or because of the deterioration of the financial capacity of the borrower), how long of a satisfactory payment history does the borrower need to demonstrate before being moved out of the TDR and non-accrual lists.  There is definitely an inconsistency and vagueness in this area.  There is the OTS guidance which states that a six month history should be sufficient as long as all other information about the borrower demonstrates his ability to maintain the modified terms of the debt service. A recent regulatory exam of an OCC regulated bank, which is a client of mine indicated that the examiners insisted that the borrower demonstrate at least one year satisfactory payment history. If the borrower is continuing to pay the question really becomes when we can release the ASC 310 portion of the reserve that we have allocated for this borrower. Whether you continue to classify the loan as a TDR or not is really irrelevant.  Other than disclosures in the footnotes of financial statements and window dressing of percentages, there is really no more of a financial impact whether the loan is a TDR or if a loan is still deemed impaired.  So, even if we were to take the loan out of TDR because the borrower has demonstrated a satisfactory payment history for the past six months, how do you classify the loan? If the financial information of the borrower has not improved much, can we really justify that it is not impaired?  So the underlying issue is the borrower’s credit risk rating. 

What the inconsistency from the regulatory exam is perhaps really telling us is that maybe we are too focused again on the payment history of the borrower in determining whether it should continue to be classified as a TDR and/or non-accrual.  What perhaps we really need to do is to assess the improvement, if any in the borrower’s financial capacity and the strengthening of his ability to repay. If we can justify the movement of the credit risk rating for the TDR to be better than Substandard and support the fact that it is no longer impaired, then I think the regulators, too, would concede.

The concept of TDR analysis is consistent, in my opinion, with that of an impaired loan analysis.   In one of my previous blogs when I discussed the portion of the ALLL calculation which is based on ASC 310 (FAS 114), I noted that the hardest part of this calculation is the starting point. The calculation is intended to measure the amount of impairment on all impaired loans.  But the confusion and inconsistency often comes from defining what constitutes an impaired loan.  In that writing, we discussed the need to bring some objectivity to this process by using the loan credit risk rating (i.e., Substandard and Doubtful) and non-accrual status. If we want to minimize regulatory exam surprises and also bring some objectivity to the TDR process, I would recommend that all loan modifications be subjected to an impaired loan analysis and be measured for potential impairment. If we measure the potential impairment and set up the right ALLL in accordance with ASC 310 for the loan, does it really matter if we fail to classify it as a TDR? TDR or not TDR: who cares? Let’s get the ALLL right and then we’re set. It seems to me that it’s a much ado about nothing.

 

Comments

First | Previous | Showing comments 6 to 6 of 6
Visitor
Re: TDR or Not to TDR ...Much Ado about Nothing? Reply #1 on : Fri June 17, 2011, 12:54:29
Its great to get a practitioners' insight to compliment the dryness of FASB releases and SEC filings. Very helpful piece.

Ok, so I have no doubt that ASU 2011-02 will drive a surge in reported TDR's in the third quarter among community banks. A few banks that I follow have shown TDR spikes from zero to 3-5% of loans in recent quarters, assumedly due to early adoption of the new guidance. However, these banks aren't providing any ALLL/charges on their new TDR pools. And from my understanding, the new guidance doesn't change how impairment is measured. Could we see a big surge in TDR's, with very little accompanying ALLL provision? As you say, "Let's get the ALLL right and then we're set." The evidence from early adoptors is that ALLL is not getting set correctly, even with tighter TDR definitions.

There seems to be enormous disparity in the quality of accounting among smaller community banks right now. I'm looking at certain banks with 5% nonaccruals, another 5% accrual TDR's, who have taken cumulative provisions for credit losses of 1% since the beginning of 2008, because their impaired assets are only marked down to 98 cents on the dollar.

Thanks for your thoughts. Brendan.
Reply to this comment Quote this comment

Add a comment

  • Required fields are marked with *.

If you have trouble reading the code, click on the code itself to generate a new random code.