Impact on Expected Credit Loss Model (ECL) amid COVID-19 disruptions

-The test of a Financial Crisis-driven model in times of Global Crisis

By Rahul Maharshi, Financial Services Team , (rahul@vinodkothari.com, finserv@vinodkothari.com)

The disruption throughout the globe due to the COVID-19 disease has caused tremendous uncertainty and difficulty in the financial sector. The Expected Credit Loss (ECL) model was introduced as an aftermath of the 2008 global financial crisis, to curb the loopholes of the incurred loss model and to provide a forward looking approach in the accounting of loan loss provisioning by inclusion of various credit measures. The model which is built up with the focus on credit strength of a particular exposure, is likely to get tested is these testing times.

This write-up is an attempt to explore how the ECL model should be looked given the disruption caused by COVID 19.

The approach to ECL at time of COVID-19 disruption

The ECL model requires application of judgement and differs from entity to entity. Such application of judgement and approaches are considered based on the industrial practices as well as the business environment in which the entity operates. For example, a financial institution may not have an aggressive approach in determination of ECL for the portfolios serving to the affluent sections of the economy as compared to the low credit-score portfolios.

However, conventional assumptions taken in the ECL model may not hold well in the current environment the financial sector across the world is going through. The assumptions of significant increase in credit risk (SICR), at times when there is extension in the repayment structure may be a classical example of SICR which would result in stage-shifting of the credit exposure. But the same would not hold well, in times of such a disruption, where the financial sector regulator has proposed to extend a three month moratorium for repayment of term loans, considering the disruption to the economy.

We arrive at ECL estimates with the help of three primary data points [B5.5.49] –

• Best available information about the past events;
• Best available information about the current conditions; and
• Forecast of economic conditions

In the current COVID-19 disruptions, the entities should actively consider the effects of the COVID-19 disruption as well as the measures taken by the government and regulators in the current situation as well as the prospective measures which would affect the forecast. Such measures should be considered as forward looking “macro-economic information” [B5.5.4] and accordingly be considered by the entities.

However, it is easier said than done, considering the rapid changes and updates in the current stressed environment. But if the ECL estimates are arrived at, after proper consideration of reasonable and supportable information, the same can provide better transparency to the financial statements and aid in providing assurance to the stakeholders.

Ind AS 109 is clear in the assessment of SICR, where the assessment has to be done on the changes in the risk of default occurring over the expected lifetime of a financial instrument. Such an assessment has to be done by the entity based on information that is available to it without undue cost or efforts. The grant of a payment holiday at times of the disruption would not be construed as a change in the risk of default occurring over the expected lifetime of a financial instrument.

Globally, various regulators, Central Bank of various countries have come up with publications and directives for assistance in such times of disruption. Some of the publications and initiatives have been discussed at length below:

Global response to the COVID-19 disruption for sound application of the ECL model

In light of the current COVID-19 disruption, the International Accounting Standards Board (IASB) published[1] a document intending to support the consistent application of the IFRS Standards which are applicable on approximately 120 countries. In its document the IASB has indicated that extension of payment holidays to all borrowers in particular classes of financial instruments should not automatically result in all those instruments being considered to have suffered an SICR.  The assessment of SICR has to be based on information that is available to an entity without undue cost and effort.

The European Central Bank (ECB), the apex banking authority for the 19 European Union countries has also published[2] a document as a guidance for financial institutions in careful implementation of the ECL and avoidance of excessive pro-cyclical assumptions in ECL during the COVID-19 disruptions.

The annexure of the above letter provided guidance for the institutions to take into account when using forecasts to estimate the ECL:

• The collective assessment of the significant increase in credit risk (SICR);

The ECB perceived deterioration to the GDP outlook could lead to manifold increase in probability of default (PD), in turn leading to transfer of the financial instrument from stage 1 to stage 2.  In times of economic disruptions, such as the COVID-19, even factors usually assessed individually become very sensitive to macroeconomic scenarios. For tackling such a situation, the ECB advised the institutions to :

• Consider whether a top-down approach to stage transfer can be taken and in the context of that approach, recognise lifetime expected credit losses on a portion of the financial assets for which credit risk is deemed to have increased significantly without the need to identify which individual financial instruments have suffered a SICR.

As per para B5.5.6 of IFRS 9 (Ind AS 109) A top-down approach to ECL is such, where the identification of a particular sub-portfolio on the basis of shared risk characteristics that represent customers whose credit risk is expected to have increased significantly, is not possible. In such a case, an assessment can be made of a proportion of the overall portfolio that has SICR since initial recognition.

Example of Top-down approach will be as follows:

Where there is high chance of a risk of default occurring and thus an increase in credit risk, results from an expected increase in interest rates during the expected life of the portfolio held by a financial institution. In such a case, assessing an increase in interest rates of 200 basis points will cause a significant increase in credit risk on 20% of the variable interest-rate portfolio and accordingly, recognition of lifetime ECL on 20% of the variable rate mortgage portfolio, and a loss allowance at an amount equal to 12-month ECL for the remainder of the portfolio.

• The use of long term macroeconomic forecasts ;and

The IFRS 9 provisions lead to the conclusion that where there is no reliable evidence for specific forecasts, long-term macroeconomic outlooks will provide the most relevant basis for estimation. The ECB provided guidance for financial institutions to:

• use available historical information, but only as far as this information is representative for the long-term horizon and free of bias;
• Where historical information depends on macroeconomic variables, consider information covering at least one or more full economic cycles or that is otherwise adjusted to avoid biases, e.g. a recency bias.

• The use of macroeconomic forecast for specific years

To include forward-looking information in ECL models, it is quite a common practice among significant institutions to establish a baseline macroeconomic scenario. In doing so, ECB guided the financial institutions to consider the following:

• Consider macroeconomic forecasts for specific years and long-term forecasts, as previously described;
• Factor these two types of forecasts into the baseline scenario with weights based on their respective relevance;
• Assign more weight to the specific-period macroeconomic forecast for the short-term outlook and systematically reduce that weight as the forecast loses relevance for time horizons in the more distant future;
• Use the long-term forecast (e.g. the long-term GDP growth rate) whenever the specific forecast has lost relevance.

Institutions that use other scenarios in addition to the baseline scenario are expected to:

• Estimate the probabilities of these scenarios and their deviation from the baseline according to the above principles;
• Ensure that the probabilities assigned to scenarios and the deviations of such scenarios beyond the reliable forecasting horizon reflect experience collected over a sufficiently long-term horizon and not be based on specific year projections.

The Basel Committee on Banking Supervision (BCBS)[3] had set out additional measures to reduce the impact of COVID-19 on the global banking system. In its press release, the BCBS reiterated the importance of ECL frameworks, referring to both IFRS 9 and Current Expected Credit Loss Model (CECL) by the Financial Accounting Standards Board (FASB), urging banks to:

“Use the flexibility inherent in these frameworks to take account of the mitigating effect of the extraordinary support measures related to Covid-19.”

BCBS further stated that:

“Extraordinary support measures should be taken into account by banks when they calculate their ECLs.”

“Regarding the SICR assessment, relief measures to respond to the adverse economic impact of Covid-19 such as public guarantees or payment moratoriums, granted either by public authorities, or by banks on a voluntary basis, should not automatically result in exposures moving from a 12-month ECL to a lifetime ECL measurement.”

The approach towards the assessment has to be holistic, considering the effect on all the components considered in the computation of ECL.

The Institute of Chartered Accountants of India (ICAI), as a joint initiative of Accounting standards Board (ASB) and Auditing and Assurance Standards Board (AASB) has come out with a publication ICAI ACCOUNTING & AUDITING ADVISORY Impact of Coronavirus on Financial Reporting and the Auditors Consideration[4] which has provided guidance on various areas of financial reporting and Audit of financial statements, which require particular attention in respect of financial statements for the year 2019-20.

The guidance has provided the following points in relation to impairment losses as per Ind AS 109:

• Measurement of ECL: the impact on business of borrowers or debtors to impact credit risk parameters in the following ways:
• Increase in PD due to reduced business activity.
• Adverse impact on value of collaterals and business cash flows in turn affecting the loss given default (LGD)
• Full utilisation of undrawn limits and loan commitments in times of disruptions to impact exposure at default (EAD).
• Entities to develop more than one scenarios considering the potential impact of COVID-19.
• As a guidance from Appendix A of Ind AS 109: Borrower specific concession(s) given by lenders, on account of economic or contractual reasons relating to the borrower’s financial difficulty, which the lenders would not have otherwise considered. Such a condition to be considered as an evidence that a financial assets is credit-impaired.
• Entities to consider impact of Prudential Regulatory actions to sustain the economy such as loan repayment holidays, reduction in interest rates etc.
• Disclosures of the impact of COVID-19 on various credit related aspects such as methods, assumptions and information used in estimating ECL, policies and procedures for valuing collaterals etc.
• Non-Banking Financial Companies (NBFCs) and Asset Reconstruction Companies (ARCs) should also carefully consider the recent guidance provided by Reserve Bank of India (RBI) on implementation of Ind AS[5]

The major component in ECL computation is the probability of default (PD). To arrive at the PD, we use historical PD by assessing the entities internal credit rating data as well as forward looking-macroeconomic factors in determining PD term structures. While assessing the forward looking PD, entities will have to consider the disruption in the business of the borrowers. Such disruption would have resulted in reduced economic activity, which in turn would have significantly increased the likelihood of default.

The effect of a pandemic on the ECL parameters have been discussed at length below.

The effect on ECL parameters in times of disruptions

The parameters used in reaching to the ECL are:

• Probability of default (PD)
• Loss given default (LGD)
• Exposure at default (EAD)
• Discount factor (df)

Further the ECL formula can be defined as “ECL= PD*LGD*EAD*df”.

To arrive at the PD, we use historical PD by assessing the entities internal credit rating data as well as forward looking-macroeconomic factors in determining PD term structures. While assessing the forward looking PD, entities will have to consider the disruption in the business of the borrowers. Such disruption would have resulted in reduced economic activity, which in turn would have significantly increased the likelihood of default.

The LGD is the indicator of quantum of loss from a transaction, given that a default occurs.  The LGD takes into consideration the exposure at default and post default classifications, valuation of any collateral as well as allocation of the same. The resultant decline in the economic activities due to lockdowns can have severe effects on the value of collaterals as well as cash flows of the business.

EAD is basically the estimation of the extent to which the financial entity may be exposed to a counterparty in the event of a default and at time of the counterparty’s default. EAD also takes into consideration, the expected payments in future scenarios. In estimation of EAD, the strength of the financial entity, in terms of assets (i.e loan given) and liabilities (i.e. loan taken) are kept in tandem to for assessment of contractual cash flows. However, in such times of disruption, the trend may be for the borrowers to fully exhaust the credit drawing limits and loan commitments, creating stress on the exposures, more so in case of revolving lines of credit facilities.

The discounting of the above components have to be done using the original effective interest rate (EIR). The EIR is arrived at after considering the current interest rate for the facility and adjusting with it, any fee or income charged. In times of such global disruptions, there is always a likelihood that the prevailing market interest rates takes a hit. As already we have seen by the RBI purchasing Rs 10,000 crore in government bonds under the open market operations programme as a measure against the disruption, the same would result in change in the risk free rate as well.

Effect of the moratorium grant on loan repayments on ECL by the RBI[6]

To address the stress in the financial sector caused by COVID-19, several measures have been taken by the RBI to mitigate the burden on debt-servicing caused due to the disruption. These measures include moratorium on term loans, deferring interest payments on working capital and easing of working capital financing. The lending institutions have been permitted to allow a moratorium of three months on payment of all instalments falling due between March 1, 2020 and May 31, 2020. The below explanation specifies effect of the moratorium.

Effect on credit risk and stage shifting

Since the moratorium is to be considered as a repayment holiday where the borrower is granted an option to not pay during the moratorium period, the same cannot be considered as a factor in determining change in the credit risk complexion of the borrower. The provisions of para 5.5.12 of Ind AS 109 are self-explanatory on the point that if there has been a modification of the contractual terms of a loan, then, in order to see whether there has been a SICR, the entity shall compare the credit risk before the modification, and the credit risk after the modification. Sure enough, the restructuring under the disruption scenario is not indicative of any increase in the probability of default.

Accordingly, the same should ideally not be considered as a factor for considering SICR and in turn, should not result in shifting of the financial instruments from one stage to another.

Effect on rebuttable presumptions about credit deterioration

The moratorium granted by the RBI seeks to amend the payment schedule without resulting in a restructuring. There is a rebuttable presumption that the credit risk on a financial asset has increased significantly since initial recognition when contractual payments are more than 30 days past due. However, the rebuttal may be offered in case the payments are more than 30 days past due. The very meaning of “past due” is something which is not paid when due. The moratorium amends the payment schedule. What is not due cannot be past due.

Effect on Effective interest rate (EIR) for the loan and income during the moratorium

The whole idea of the modification is to compute the interest for the deferment of EMIs due to moratorium, and to compensate the lender fully for the same. The IRR for the loan after restructuring should, in principle, be the same as that before restructuring. Hence, there should be no impact on the EIR.

As the EIR remains constant, there will be recognition of income for the entire Holiday period. For example, for the month of March, 2020, interest will be accrued. The carrying value of the asset (POS) will stand increased to the extent of such interest recognised. In essence, the P/L will not be impacted

Also, there will not arise any modification gain or loss as per para 5.4.3 of Ind AS 109 since the EIR remains constant.

Requirement of Impairment-testing of financial asset

The revision in the payment schedule does not result in a modification of the financial asset, which could have resulted in an impairment testing of the financial asset. Since the contractual modification in case of the moratorium is not a result of a credit event, the question of any impairment for this reason does not arise.

Conclusion

The concept of ECL being a fallout of the Global Financial Crisis, it will be interesting to see how fairly the model lives up in providing transparency to the users of financial statements at this time of global disruption. It is worth mentioning that ECL, being a reserve built up in good times which could be used in bad ones, the current situation is difficult, creating high levels of uncertainty but certain measures may be adopted by entities in curbing the situation to some extent, such as:

• Developing more than one scenarios for the potential impact of the COVID-19 disruptions treated as macro-economic information as per para B5.5.4 of Ind AS 109
• Effect of measures taken by the government and the regulators in the true spirit with which the same is implemented.

[1] https://cdn.ifrs.org/-/media/feature/supporting-implementation/ifrs-9/ifrs-9-ecl-and-coronavirus.pdf?la=en

[2]https://www.bankingsupervision.europa.eu/press/letterstobanks/shared/pdf/2020/ssm.2020_letter_IFRS_9_in_the_context_of_the_coronavirus_COVID-19_pandemic.en.pdf?b543f9408a8480e04748a3b0185d8cf3

[3] https:/www.bis.org/press/p200403.htm and  https://www.bis.org/bcbs/publ/d498.htm

[4] https://resource.cdn.icai.org/58829icai47941.pdf

[5] https://rbidocs.rbi.org.in/rdocs/Notification/PDFs/NOTI170F341D8DE49C04D4B8382D855A9858583.PDF

[6] https://www.rbi.org.in/Scripts/NotificationUser.aspx?Id=11835&Mode=0