Some IFRS 9 Implementation Considerations

If what you are looking for is an IFRS9 Calculator, then don’t pretend that you want to know about IFRS 9!

It was not a coincident nor mistakenly missed points that IFRS 9 did not define default, nor did it define what constitute a “significant deterioration in credit risk circumstances”. This is only to send a strong message to each IFRS 9 reportable institution that it starts and ends with your institutions! No IT solution can have enough magic to spare you the pains of “Do It Yourself”. IFRS 9 is an exercise in asset quality!

 Here are some (not all) important considerations in the IFRS 9 implementation exercise:

1. Understand Your Own Data: not less than three years and five years is enough, and account for every loan in the portfolio: performing, under-performing, and non-performing.
2. Process your loan data (quarterly observations for commercial loans, and monthly observations for retails loans) to capture loss rates for each category: commercial and retail.
3. Look for shared risk characteristics between individual loans, and group loans based on these common risk characteristics.
4. So now we have grouped loans based on risk characteristics, and we have historical loss rates at the level of each facility and each loan group.
5. Expected Credit Loss (ECL) is EQUAL to Historical Loss PLUS a Forward-Looking Component. It is best if it is done for each loan group separately. You may end up with a different set of macroeconomic variables for each family (group) of loans depending on what may clearly have an impact on its loss rate.

1. The correlation between the loss rate and a given macroeconomic variable can be easily captured through historical observations – i.e., examine how the loss rate for the group of loans has historically changed as this macroeconomic variable changed. This is the easiest way to quantify ‘the future’!
2. It is a good time to ask yourself this question: is the [my] past a good predictor of the [my] future as far as the [my] loan portfolio is concerned? Your answer to this question dictates your choices of variables to incorporate in the forward-looking component of the exercise.
3. This is a good place to stress test for least likely to occur events with high impact.
4. The forward-looking component has two dimensions to it: the 12-month, and the lifetime such that the forward-looking of your stage 1 is only 12-month; and for stage 2, it would be best if you limit it to the life of the asset.
5. One more step to intimately understand your loan portfolio is to apply IFRS 9 retroactively. Stage the individual loans in your portfolio going three years back, and continue with the exercise till the present, or until the facility is completely settled (or otherwise worked out). Check the circumstances under which a facility migrates from stage 1 to stage 2 (or 3), and what is the average stay of a facility in each stage? This exercise helps you check, in retrospect, the quality of your past credit decisions. Accordingly, you decide to build on them or change them.
6. As far as your non-performing loans (i.e., Stage 3), and when you are factoring a collateral into your recoveries, make sure you apply a haircut; estimate the approximate time frame it will take to liquidate the asset; and discount that cash flow to the present so your calculations of the loss given default on this particular facility will be sufficiently (not exactly) accurate. 

Other considerations which are equally important, but have stronger strategic implications:

 Implications For Portfolio Strategies.

1.    Should we revise our credit portfolio allocation and lending policies?

2.    Should we reduce lending to volatile sectors with a poorer outlook? How do we reflect this in our lending policies?

3.    Should we weigh the financial duration of portfolios more heavily in our lending decisions and reduce lending on long-term transactions?

4.    Should we focus on collateralized lending portfolios to mitigate Loss Given Default and reduce lending to unsecured exposures? 

Implications For Portfolio Strategies.

1. Should we revise our credit portfolio allocation and lending policies?
2. Should we reduce lending to volatile sectors with a poorer outlook? How do we reflect this in our lending policies?
3. Should we weigh the financial duration of portfolios more heavily in our lending decisions and reduce lending on long-term transactions?
4. Should we focus on collateralized lending portfolios to mitigate Loss Given Default and reduce lending to unsecured exposures? 
5. Should we treat higher-risk clients differently in our lending decisions? Should scrutinize lending to performing high-risk clients more thoroughly? How should we reflect this in our risk appetite?

Impact On Commercial Policies.

1.    Should we rethink our product offering?

2.    Should we adjust our pricing to sustain profitability?

3.    Should we adjust maturity and amortization to shorten product lifetimes?

4.    How can we encourage Relationship Managers and Clients to shift to products with shorter terms or early redemption options?

5.    Should we raise prices for longer-term and less collateralized products and for higher-risk clients? Would that damage our competitive position?

Changes To Credit Risk Management.

1.    Should we strengthen our monitoring of counterparty and data quality to prevent increases in ECL?

2.    Should we improve our early-warning mechanism to detect any deterioration in client’s lifetime credit risk?

3.    Should we increase our monitoring of collateral data?

4.    How should we flag warning signs to our relationship managers to trigger remedial actions?

Evolution Of Deal Origination.

1. Should we adjust our credit strategy and policies to change the course of new business development?
2. Should we introduce new risk limits for clients, sectors, or products most affected by IFRS 9?
3. Should we change our origination process – for example, by adopting a delegated-authority system or improving the link between our risk-appetite framework and our under-writers?

Impact On People Management.

1.    Should we revise our incentive and compensation schemes for relationship managers (RMs)?

2.    Should we change RMs’ accountability?

3.    Should we change our performance metrics to reflect IFRS9-adjusted profitability?

4.    Should we provide training for our RMs on the consequences of IFRS9 and appropriate remedial actions?

Impact On People Management.

1.    Should we revise our incentive and compensation schemes for relationship managers?

2.    Should we change RMs’ accountability?

3.    Should we change our performance metrics to reflect IFRS9-adjusted profitability?

4.    Should we provide training for our RMs on the consequences of IFRS9 and appropriate remedial actions?