5 July International Accounting Standards Board 30 Cannon Street, London EC4M 6XH United Kingdom. Dear Board Members:

Transcription

1 5 July 2013 International Accounting Standards Board 30 Cannon Street, London EC4M 6XH United Kingdom Dear Board Members: Consejo Mexicano de Normas de Información Financiera (CINIF), the accounting standard setting body in Mexico, welcomes the opportunity to submit its comments on the Exposure Draft ED/2013/3 on Financial Instruments: Expected Credit Losses (the ED) issued for exposure in March Set forth below you will find our comments on the topics included in the ED. Overall comments The recognition of expected credit losses on financial instruments is clearly a step forward to better account for financial assets and is welcomed in Mexico by the financial sector. However, there are different points of view regarding the mechanics of how to achieve that goal. The main concern that has been expressed by the banks is that the model of two buckets to classify financial instruments would not permit a gradual recognition of the expected credit losses. They believe that forecasting the default on financial instruments that may experience a substantial increase in the credit risk over the next 12 months is difficult, but feasible, but determining up front the expected credit loss over the full life of the instrument would face several hurdles. The first challenge to forecasting a substantial increase in credit risk is the definition of default. There can be several indicators of default. One would be a one-month lag in payment. A second would be a larger delay in payment. Another would be failure to comply with certain covenants established in the loan agreement, which could even occur before there is a default in payment of principal or interest. Mexican banks favor a mechanic of aggregating the indicators of default, giving to each of them a certain weight in measuring the expected credit losses considering also the expected severity of the losses. When several indicators are present, there would be a compounded effect that would increase substantially the expected credit loss and, if the indicators continue or increase, the total expected credit losses over the life of the financial instrument would be almost fully recognized when the financial asset is measured individually due to a substantial increase in credit risk. A critical feature to be considered when compounding the default indicators is that these be determined through the loan cycle and not just at the moment the allowance is determined. In other words, a possible downturn in the economic cycle should be considered even if the economy is currently performing well. 1

2 This would allow for a seamless provisioning of expected credit losses, without a jump from the first to the second bucket. In discussions held with officers of several banks, they indicated that this is what the Basel II guidelines are striving to achieve. In this regard, we wonder if what should be included in the standard is the basic principle of recognizing expected future credit losses, and allow the regulators of each country to set up the methodology, based on Basel II Guidelines and adhering to the principle of recognizing such expected future credit losses. We also believe that a significant effort should be made to attain convergence with the FASB, to avoid the existence of two similar, but different, models. Another issue that was unanimously rejected by the banks is the computation of interest on the net amortized cost of a financial instrument (after deducting to the gross amount the impairment losses that have been recognized). They oppose recognizing the interest on the net amortized cost for two reasons. The first one is a practicality issue, since the systems of the banks would need significant modifications to be able to cope with this requirement. However, the more fundamental reason is that, once a loan has shown significant signs of deterioration in its credit risk, such as lack of payment of principal and interest, recognizing interest revenue would be equal to recognizing assets whose collection is doubtful, when revenue may never actually be realized. Set forth below you will find our responses to the specific questions you raise. Question 1 Do you agree that an approach that recognises a loss allowance (or provision) at an amount equal to a portion of expected credit losses initially, and lifetime expected credit losses only after significant deterioration in credit quality, will reflect: (i) (ii) the economic link between the pricing of financial instruments and the credit quality at initial recognition; and the effects of changes in the credit quality subsequent to initial recognition? If not, why not and how do you believe the proposed model should be revised? Do you agree that recognising a loss allowance or provision from initial recognition at an amount equal to lifetime expected credit losses, discounted using the original effective interest rate, does not faithfully represent the underlying economics of financial instruments? If not, why not? There is agreement in Mexico on recognizing a loss allowance at an amount equal to the portion of initial expected credit losses; however, the consensus is that the allowance need not be for the full life of the financial instrument, but rather a minimal floor, based on certain indicators. Thereafter, to have seamless provisioning, an additional loss allowance would be added as new indicators arise. Based on the severity of the indicators, some may be compounded and result in a higher allowance. The effects of changes in credit quality subsequent to initial recognition would have to take into consideration new indicators of default that, when aggregated and compounded, would lead to a larger allowance, which could approach the lifetime expected losses when the loan is measured separately on such basis, due to a significant increase in credit risk. Regarding the economic link between the pricing of financial instruments and the credit quality at initial 2

3 recognition, the consensus is that it does certainly exist. However, an adequate pricing does not mean the credit losses have been avoided by such pricing. Some loans will have collateral, and the quality of the collateral varies. Also, there are other characteristics that have to be considered at initial recognition, such as the loan to value ratio in mortgage loans. If at initial recognition the value of the collateral is significantly higher than the amount of the loan, the risk indicator would be minimal. However, if the value of the collateral is similar or less than to the amount of the loan, the risk indicator to consider at inception would be higher. This may be the case where the collateral is similar to the amount financed in mortgage loans or car loans, when the down payment is insufficient, in which case a minimal initial allowance should be recognized, based on this risk indicator. However, one practitioner believes that the use of indicators of default that may be forecasted and compounded would not lead to comparable information among entities and believes that the model proposed by the IASB is operational and would adequately reflect the expected credit losses. Recognizing a loss allowance upon initial recognition at an amount equal to the lifetime expected credit losses does not faithfully represent the underlying economics of the financial instrument, because there are insufficient indicators at that moment to estimate lifetime expected credit losses. Also, there will not be an adequate matching of revenues of expenses, since interest has not begun to be earned. Question 2 (c) Do you agree that recognising a loss allowance (or provision) at an amount equal to 12-month expected credit losses and at an amount equal to lifetime expected credit losses after significant deterioration in credit quality achieves an appropriate balance between the faithful representation of the underlying economics and the costs of implementation? If not, why not? What alternative would you prefer and why? Do you agree that the approach for accounting for expected credit losses proposed in this Exposure Draft achieves a better balance between the faithful representation of the underlying economics and the cost of implementation than the approaches in the 2009 ED and the SD (without the foreseeable future floor)? Do you think that recognising a loss allowance at an amount equal to the lifetime expected credit losses from initial recognition, discounted using the original effective interest rate, achieves a better balance between the faithful representation of the underlying economics and the cost of implementation than this Exposure Draft? The ED establishes that indicators of default will determine which financial instruments will fall in default during the next 12 months to provide an allowance for their lifetime expected credit losses. We believe there are no indicators on how to reliably determine lifetime expected credit losses at the moment the credit is granted. The experience of Mexican banks is that a financial instrument will not show signs of expected losses at any particular moment, but instead it will gradually accumulate those signs. One of them could arise when the credit is granted if the ratio of the loan to the value of collateral is very close in mortgage or car loans. Therefore, the best way to determine expected credit losses is to gradually take those signs of expected losses into consideration, based on historical experience and provide for expected credit losses. Most of the Mexican banks and practitioners believe that the approach to account for expected credit losses proposed in the ED will not achieve a better balance between faithful representation and the 3

4 underlying economics, since it will not capture the gradual deterioration, which is common in financial instruments, and will recognize extemporaneous effects, generally too late, which is what happened in 2008 in some countries. Gradually recognizing expected credit losses will avoid those sudden jumps in the allowance. In this regard it would be necessary that the methodology consider expected losses through the cycle and not just at a certain point in time, to adequately recognize them. This will be a forecasting issue that may be difficult to implement, but it is feasible. The comments from the bankers are that what Basel II is pursuing is a gradual and anticipated recognition of expected credit losses, and that models implemented by banks pursuant to Basel II guidelines have been working satisfactorily. However, not all of the models are forecasting expected losses through the cycle, and this will be needed to have comparative information. However, one practitioner believes that the model of the ED is adequate, is in line with Basel II guidelines, and is a practical way of determining expected credit losses that would result in comparable financial information for expected credit losses. Question 3 Do you agree with the proposed scope of this Exposure Draft? If not, why not? Do you agree that, for financial assets that are mandatorily measured at FVOCI in accordance with the Classification and Measurement ED, the accounting for expected credit losses should be as proposed in this Exposure Draft? Why or why not? Regarding financial instruments measured at amortized cost, there were no objections among the Mexican bank officers. Some bank accountants questioned why it was necessary for those financial instruments measured at fair value through OCI to determine impairment, when the market value would already consider impairment. However, the bank credit risk managers indicated that markets do not always anticipate credit deterioration until there is public information about an event of default. Notwithstanding, it would be difficult to gather the necessary information to measure deterioration for these instruments, which will have to be analyzed individually, starting from their credit rating. Guidelines will be needed, since the price vendors that supply such data to the banks do not always consider increase in credit risk. Regarding commitments and financial guarantees, there is consensus that these should also be included in the scope of the standard. In the case of leases, there is general consensus that these should be included in the scope, as all the financial leases are in fact a financing with the leased assets as collateral. Question 4 Is measuring the loss allowance (or a provision) at an amount equal to 12-month expected credit losses operational? If not, why not and how do you believe the portion recognised from initial recognition should be determined? The operational problem is identifying the loans for which the credit quality will significantly deteriorate 4

5 over the next 12 months. This may be very subjective. Risk managers believe there should be a methodology that would demonstrate that this is possible. Bank officers, both in the accounting and risk areas, believe that operationally it will be very difficult to identify, on a statistical basis, the specific loans that will show significant increase in their credit risk. Banks will need very robust databases to accomplish this task. Some banks that report to a foreign parent have such databases and may be able to easily implement the model. Also, once having determined the loans that will have significant deterioration, it will be necessary to determine the expected loss over the life of the loan, which will be even more subjective since there will be insufficient information to gauge severity of default in loans for which the only evidence is a statistical probability of default. Question 5 (c) (d) (e) Do you agree with the proposed requirement to recognise a loss allowance (or a provision) at an amount equal to lifetime expected credit losses on the basis of a significant increase in credit risk since initial recognition? If not, why not and what alternative would you prefer? Do the proposals provide sufficient guidance on when to recognise lifetime expected credit losses? If not, what additional guidance would you suggest? Do you agree that the assessment of when to recognise lifetime expected credit losses should consider only changes in the probability of a default occurring, rather than changes in expected credit losses (or credit loss given default ( LGD ))? If not, why not and what would you prefer? Do you agree with the proposed operational simplifications, and do they contribute to an appropriate balance between faithful representation and the cost of implementation? Do you agree with the proposal that the model shall allow the re-establishment of a loss allowance (or a provision) at an amount equal to 12-month expected credit losses if the criteria for the recognition of lifetime expected credit losses are no longer met? If not, why not, and what would you prefer? A significant increase in credit risk after initial measurement would undoubtedly require an increase in the loan loss allowance. However, it would be very difficult to determine an expected credit loss at an amount equal to lifetime expected credit losses. The increase in credit risk could mean that the probability of default has increased. However, it may still be less than 100%. Also, the severity of default could be in a certain range, say 50% to 60%. This could help to measure the expected credit losses at such point in time. However, it may not be the same expected credit loss over the life of the loan, unless the forecasting of the expected loss is made through the cycle, considering in the good years future possible events, that could result in adverse changes in the economic environment. Therefore, it would be very difficult to predict the future changes in probability and severity of default. We believe there should be additional guidance on how to determine both the probability of default and the severity of default (how much will not be recovered). In this case we do not agree that the changes in expected credit losses to be incurred should not be considered, as the severity of default is the second part of the equation to determine the total expected credit losses. Regarding the operational simplifications, these seem to be based more on rules than on a principle, and the policy of the IASB has always been to focus on principles rather than on rules. Further, it will 5

6 not be applicable to all financial instruments, since many of them do not have a credit rating. It may be applicable to sovereign debt and to some corporate debt. Regarding the past due rule of 30 days, it may represent a probability of default for certain types of loans but not for other types of loans where different delinquency patterns should be considered. Regarding the re-establishment of a loss allowance at an amount equal to 12 months of expected credit losses if the criteria for recognizing lifetime expected credit losses is no longer met, we believe that the sudden reduction of an allowance would not capture all the features of the probability and severity of credit risk. The Mexican bankers indicated that they would prefer following a methodology of gradually decreasing the probabilities of default or the severity of default to adjust the allowance, rather than making an abrupt decrease in the allowance. Question 6 (c) Do you agree that there are circumstances when interest revenue calculated on a net carrying amount (amortised cost) rather than on a gross carrying amount can provide more useful information? If not, why not, and what would you prefer? Do you agree with the proposal to change how interest revenue is calculated for assets that have objective evidence of impairment subsequent to initial recognition? Why or why not? If not, for what population of assets should the interest revenue calculation change? Do you agree with the proposal that the interest revenue approach shall be symmetrical (ie that the calculation can revert back to a calculation on the gross carrying amount)? Why or why not? If not, what approach would you prefer? There is consensus that interest should not be calculated on the net carrying amount of the loan, i.e. its amortized cost less the related allowance for loan losses. The operational problems that this would create to identify for each loan the allowance that was provided to compute the interest would be enormous and significant changes in accounting systems would have to be implemented. In addition, there is an objection based on the principle that this would be equal to recognizing assets whose collection is doubtful, when revenue may never actually be realized. The Mexican Banks indicated that their experience shows that when a loan has shown a significant increase in its credit risk and would be classified in the second bucket, the probabilities of collecting additional interest are significantly lower than the amount of principal that would be collected. Therefore, they oppose recognizing interest revenue for problem loans. Whenever this practice is followed, disclosure should be made of the amount of unaccrued interest to which the entity would have been entitled. If a specific allowance is provided once there is objective evidence of impairment, such as indicators of delinquency, the question is whether additional interest will be collected. The consensus of Mexican banks and most practitioners is that the non-accrual of interest, which has been a longstanding practice of Mexican banks, should continue. However, one practitioner believes that the proposed model is adequate. Another practitioner indicated that the proposed model would not be consistent with the revenue recognition model, which requires that revenue to which the entity expects to be entitled should not be adjusted for credit risk. Notwithstanding, these practitioners recognize that past experience indicates that once a loan shows significant credit risk, the amount of future interest that will be collected is little or nothing and, therefore, the non-accrual of interest is an adequate practical expedient in these circumstances. 6

7 Reverting back to a calculation of interest on the gross carrying amount should only be permitted if there is strong evidence that the loan is performing as a loan with minimal risk, i.e. when default is no longer probable. Question 7 (c) Do you agree with the proposed disclosure requirements? Why or why not? If not, what changes do you recommend and why? Do you foresee any specific operational challenges when implementing the proposed disclosure requirements? If so, please explain. What other disclosures do you believe would provide useful information (whether in addition to, or instead of, the proposed disclosures) and why? The bankers indicated that they believe such disclosures would be cumbersome and that in most cases would obscure other information that is of value. Therefore, it is a question of reaching a balance for the disclosures that should be provided to the public. This kind of disclosures would be understandable only by some specialists when analyzing the financial information of banks, who would in turn summarize their conclusions for their clients. We believe that a proper balance should be sought for the disclosures requested and that the principles that will guide the Disclosure Framework be considered, in order that only relevant information is disclosed. Question 8 Do you agree with the proposed treatment of financial assets on which contractual cash flows are modified, and do you believe that it provides useful information? If not, why not and what alternative would you prefer? We do not agree with the proposed treatment, which indicates that the credit risk of the modified financial instruments has to be compared to the original credit risk to determine if there has been a significant increase in the credit risk, to establish if the allowance should be based on a 12-month probability of default or on the expected losses over the life of the loan. We believe that, if the modification is due to renegotiation of a financial asset that was in a troubled status, there should be firm evidence that the financial asset is performing adequately before eliminating the allowance that was provided before the loan was renegotiated. This may require that performance be monitored over several periods, which could vary depending on the reasons for the modification (restructuring), how the debtor is performing operationally, and other factors. The experience of the Mexican Banks is that a majority of renegotiated loans will again show a significant increase of their credit risk and, therefore, should not be transferred back to the category of performing loans until they prove to be performing. Close monitoring will be required. In this regard several practitioners believe that all renegotiated loans should be treated as financial assets that are credit impaired. Some practitioners indicated that to avoid recognizing losses under the incurred loss model, some 7

8 banks renegotiate or grant a new loan to pay the old one. This practice will be more difficult when expected losses are recognized, provided that a renegotiated or replacement loan would remain in the category of credit-impaired loans. Question 9 Do you agree with the proposals on the application of the general model to loan commitment and financial guarantee contracts? Why or why not? If not, what approach would you prefer? Do you foresee any significant operational challenges that may arise from the proposal to present expected credit losses on financial guarantee contracts or loan commitments as a provision in the statement of financial position? If yes, please explain. There is consensus agreement with the proposals on the application of the general model to loan commitments and financial guarantee contracts. At the present time only the banks that are reporting to their parent entity either under IFRS or USGAAP (four of the five major banks in the country), are recognizing a provision for expected losses on loan commitments. It is also the case for financial guarantees, which are measured at fair value and, if a loss is expected, a provision is recorded based on IAS 37 or US GAAP. These banks indicate that, for loan commitments, they prefer to provide for the fair value of the losses rather than apply the two-bucket approach prescribed in the model, by forecasting how many loans will default in the first year after withdrawing the funds and provide for the losses on such loans over the life of the loans. For other banks, implementing the procedures prescribed in the ED will require setting up adequate procedures. At present all Mexican banks are providing for expected losses on the credit lines of credit card loans on a statistical basis, whereas a higher provision is assigned to those cases where the credit line is actively used and the cardholder is only paying the minimal amount due every month, but not the total amount due. Regarding financial guarantees, the banks that do not follow IFRS or USGAAP, consider the commission as revenue when collected, instead of amortizing it over the life of the financial guarantee. There may be some resistance to change. Question 10 Do you agree with the proposed simplified approach for trade receivables and lease receivables? Why or why not? If not, what changes do you recommend and why? Do you agree with the proposed amendments to the measurement on initial recognition of trade receivables with no significant financing component? If not, why not and what would you propose instead? Regarding trade receivables, a simplified approach geared to measure the expected credit losses of trade receivables over their lifetime would be operational for those receivables subject to normal commercial terms, which is normally up to 60 or 90 days. However, when the credit term is extended significantly, as in department stores, the same methodology for consumer loans should be followed. In 8

9 this case financing is a substantial part of the business of a department store. Most department stores have good statistics on the performance of their receivables, which would enable them to measure expected losses based on the same probability and severity indicators as banks use for consumer loans. Some practitioners believe, however, that the two options should be allowed. Regarding financial lease receivables, these are similar to a loan with the leased asset as collateral. We do not see a valid reason to exclude them from the same methodology as that for commercial loans. We foresee problems in certain non-regulated financial businesses that grant mortgage, automobile and other type of loans. Many of these may not yet have strong databases to determine probability and severity of default. Also, there are entities that have purchased loans from banks and have issued debentures listed on the stock exchange to finance the purchase of such loans. These entities do not have statistical historic information on the performance of the loans they purchased and will have to set up a methodology to measure expected losses. It is not impossible to establish a methodology, but these entities are well behind banks in managing credit losses under an expected loss model. Some practitioners indicated that they consider that the practical expedient of not considering the time value of money in receivables of less than 12 months, included in the revenue recognition project, will lead to unduly recognizing the full revenue in the case of entities that finance their sales, prior to recognizing expected credit losses. Question 11 Do you agree with the proposals for financial assets that are credit-impaired on initial recognition? Why or why not? If not, what approach would you prefer? We agree with the proposals. When an entity purchases a troubled portfolio, it has already projected the future cash flows in order to establish a price to bid for the portfolio. Therefore, there is information on the future cash flows that will enable the entity to determine the effective interest rate. Loss allowances will have to be recognized based on the same methodology, and due to the features of these loans, each one of them will have to be monitored regarding probability and severity of default. Since there is no past experience with these portfolios, it will be more difficult to measure the loss over the lifetime of the financial instruments, but if probabilities of default are added and a severity of default is determined, an adequate allowance for expected losses may be measured. Question 12 (c) What lead time would you require to implement the proposed requirements? Please explain the assumptions that you have used in making this assessment. As a consequence, what do you believe is an appropriate mandatory effective date for IFRS 9? Please explain. Do you agree with the proposed transition requirements? Why or why not? If not, what changes do you recommend and why? Do you agree with the proposed relief from restating comparative information on transition? If not, why? 9

10 The consensus is that five years will be needed to set up proper methodology on measuring expected credit losses, considering three years to develop the model and two years for its implementation. This is based on the experience when the Mexican banks implemented expected loss models. If the expected loss models already in use are in line with the final proposals of IFRS 9, less effort will be required. Some banks or insurance entities that are subsidiaries of foreign entities have an additional advantage in this respect. Regarding the transition requirements, Mexican Banks and some practitioners believe that a retrospective application would not be realistic, since it would be impossible to determine what would have been the expected loss in prior years without using hindsight, and even so, the effect on prior years would be very subjective. The consensus was that a prospective application should be followed. In such case there would be a need to have relief from presenting comparative information. However, other practitioners believe that in the two years required for implementation, the banks could gather the information necessary to retrospectively apply the expected loss model. Question 13 Do you agree with the IASB s assessment of the effects of the proposals? Why or why not? None of the Mexican banks has assessed the effects of the proposals. Therefore it is not possible to comment on the expected effects of the proposals. We recommend performing an impact analysis before issuing the new IFRS Should you require additional information on our comments listed above, please contact Juan M. Gras at (52) ext. 105 or me at (52) ext. 103 or by at jgras@cinif.org.mx or fperezcervantes@cinif.org.mx, respectively. Sincerely, C.P.C. Felipe Perez Cervantes President of the Mexican Financial Reporting Standards Board Consejo Mexicano de Normas de Información Financiera (CINIF) Cc: Jan Engström Amaro Gomes 10