Blog

Today, we have an interview with Sufyan Khan. Sufyan Khan is our member and the Regional Product Manager of IFRS 9 and Tax from AxiomSL. Sufyan has more than 14 years of experience in the area of sales, client service, business development and product management. Over the past 7 years, he has helped clients to make informed decisions in the area of trading and best practices across enterprise-wide risk management.

As of 1st January 2018, institutions will need to comply with the new IFRS 9 standard that will have a significant impact on how financial instruments are accounted for.

These changes may directly impact the P&L and corporate balance sheets. So today, I would like to ask a few questions to Sufyan on this subject.

You can watch the video recording of the interview here

Below is the transcription.

Boris: Sufyan, what are the main challenges associated with the implementation of IFRS 9?

Sufyan: So first and foremost, thank you very much Boris for this interview. If you actually look at the previous standard, which was based on IAS 39, it was more related to the finance function. However, IFRS 9 is not just a finance function, it’s more related to risk; it’s more of an amalgamation of finance and technology – the two working together. Also, the previous directives calculations were based more on the incurred loss approach, whereas IFRS 9 has more of a forward-looking approach. That’s where risk and finance have to work together and collaborate. I think that’s one of the key challenges because risk and finance has to be reconciled which means the generation of the models. And again, these models are generating PD/LGD results but they are not just looking at a historical approach, you are looking at a forward-looking approach in the model which means you are not looking at the previous regulation related to the incurred loss. In that scenario, you will have to generate the Probability of Default (PD) based on forward-looking macroeconomic factors. The same approach is to generate the Loss Given Default (LGD). Large volumes of data are generated in the process.

You are not just looking at one number. One of the big differences between IFRS 9 and IAS 39 is that you’re looking at the present value of future cash flows which have been generated, and that means generating huge volumes of contractual cash flows and behavioural cash flows. You’re forecasting it not for one year but looking at the entire lifetime value of the asset and that’s a huge amount of data, a big complexity in the calculations which have been generated. A lot of processing which is going on behind the data and calculations as well as running it across multiple scenarios and multiple simulations, if necessary.

I think that it becomes a huge exercise not just from a risk or finance perspective but it’s also very technology-heavy directive. Not just a quantitative element but a qualitative assessment needs to be taken into consideration.

There’s a model validation process which needs to be done. Banks and financial institutions want to see what are the impacts they will have on their balance sheet and on their provisions.

So , the challenges are not just from a calculation perspective, there are also challenges from a qualitative perspective, on data, data quality, reconciliation, risk management, as well as the alignment of two different teams - where both the CRO and CFO functions have to work together.

Boris: Thank you. Sufyan, could you please tell me in detail how is it possible to integrate risk and finance within IFRS 9?

Sufyan: One of the key aspects of IFRS 9 is the alignment of Risk and Finance. You need to have your risk and finance data integrated and reconciled. A lot of the banks face data, data quality and reconciliation issues and I think that’s where the key challenge is. If your risk and financial data are not reconciled or with bad data then this would have an impact on the provisions.

I think that’s where you need to have some sort of flexible infrastructure, your technology has to be flexible enough where you can reconcile data easily. Additionally, reconciliation needs to be not just on the ledger but sub-ledger account level too. Risk should be talking in the same language as finance. Finance should be talking in the same language as Risk. Both the CFO and the CRO should be aligned. Not just the data, it’s also the cleanliness of the data that matters and that’s where the data quality process comes into the picture. It’s a lot of pressure on IT to get the right data aligned and reconciled as well as deal with data gaps or data issues. IT departments have to work closely with Risk and Finance.

I think that’s one of the key challenges and that’s where it is extremely important for a bank whether small or large to have a user-friendly infrastructure in place. As the data increases, they should be able to add and reconcile it on the fly.

Boris: Alright. Tell me please, Sufyan. How can you source a new data attributes in this environment?

Sufyan: In banking technology, we generally see lack of flexibility. When IFRS 9 directive comes into the picture; a new set of data, new attributes, new cash flows, new risk parameters – the complexity will increase significantly. I think that’s where it becomes extremely important to have a flexible infrastructure. e.g., do I have the flexibility to create my own model based on my business requirements? Can I see the results of the model? Do I have the flexibility of the qualitative and the quantitative factors? What are the macroeconomic scenarios which have been used?

So that’s another new set of data that is required. If I’m looking at my asset side information, what are the factors which have been generated using the lifetime value of the asset. It’s a new amount of data, a new set of results which have been generated. What is my Loss Given Default (LGD)? What are the results which have been generated which I then need to map onto my FINREP reports if you’re looking from an EU perspective, or it has to be mapped onto my balance sheet? So I can use a set of attributes which have been generated and mapped. Also, from the shareholder’s point of view and from the board’s point of view, what is the impact between the previous directive and IFRS 9?

It is imperative to have the right technology and the right infrastructure where you can source that level of detail, where there is flexibility in sourcing information, where you don’t have to go through a huge ETL or a transformation process. We at AxiomSL increase the user experience. The whole IFRS 9 data sourcing, calculation and reporting process should be more of a business as usual exercise to see the impact on provisions and should not just be treated as a regulatory exercise. AxiomSL’s strategic regulatory platform can extract the best out of this directive to see the impact not just on the group but on different lines of business.

Boris: Alright. Sufyan, please tell me. How can banks leverage their financial reporting data in this new situation as well?

Sufyan: I think it is a key element and a starting point where existing financial reporting data can be leveraged as long as the data is available at contract level and not just aggregate level for reporting.

Under IFRS 9, the directive clearly states that your expected Credit Loss needs to be computed at a contract-level. Which means the values computed under IFRS 9 that populate the financial reports would be an aggregation of the results computed at the contract level.

Certainly, calculations have to be carried out on a very granular level. Yes, you can leverage the data, but only if you have the right level of data for the contract and not just an aggregate value. Results based on the data used should be auditable. Auditing not just for an internal or external auditor but the bank should be able to see calculations at the contract level. Where is my exposure high? Which contracts are accounts which are actually falling in value or which assets are in Stage 1 vs Stage 2 and Stage 3? What is the impact on my provisions across different reporting dates? Why is my provision number increasing?

Boris: Thank you. Sufyan, how does the IFRS 9 solution of AxiomSL support scenario and impact analysis?

Sufyan: From AxiomSL’s point of view, we continuously enhance the user experience. We have made the solution end to end, which means we are not just taking the data on one side, also going through a data quality check process. Clients have the flexibility to add more data, which means down the line that if you want to get more data inside the platform, the client can do it with ease.

With our IFRS 9 solution, you can compare not just previous results but also compare results between IAS 39 and IFRS 9. We provide flexibility where the client can have not only standard PIT models but also bespoke models based on their business requirements. With AxiomSL’s strategic platform, you can generate and execute PD/LGD models inside the system, you can compute contractual and behaviour cash flows, all the way to calculations, to report generation and GL postings. As part of our standard package, we are already generating IFRS 7.35H for our clients which provides impact analysis. The report actually tells you what is the impact between the previous result versus the new result.

The platform has a flexible approach whereby the entire end to end process can be executed as a one-click workflow. We at AxiomSL make the entire process automated so that clients can run the workflows and see the impact results in an automated manner. The process can be automated across multiple scenarios. If the client wants to do an impact analysis based on their own scenarios and from their own bank’s internal assessment perspective, they can run that within AxiomSL.

For example, our clients who are currently implementing IFRS 9 are running the process in an automated manner not only from a regulatory point of view but also from impact analysis perspective with the ability to run impact analysis on their provisions if certain global macroeconomic scenarios took place.

Also, they are able to run this both at group and entity level as well as a line of business across different jurisdictions. Since macroeconomic factors would be different for every region, our clients are able to run different scenarios for different regions which then aggregate up to the group level. This gives them the flexibility while automating the process with minimum manual intervention.

I think one of the key aspects is the ability for a client to audit the results not just to the calculation but all the way from where the data originated. The same approach we have adopted for the models to make it completely transparent. AxiomSL does not have a ‘black box’ mentality, so we ensure the calculations are open and transparent which means end business users can drill down from the results all the way back to where the data actually originated from. So it’s not just one or two steps but all the way to the point of origination. This way we continuously enhance the user experience.

Boris: Okay, thank you. I think maybe the last question I would like to ask you today is what mistakes do you see the organizations trying to implement the IFRS 9?

Sufyan: I think that right now everybody wants to push towards the implementation deadline. Going forward, an organization should look at a strategic approach instead of a tactical approach. It is inefficient when organisations work with a patchwork of multiple systems and excel and flat files. What they really should be doing is taking a step back and thinking, how can we ensure to not only get accurate results but also establishing this as a business as usual activity moving forward? Which means, if we run a process, how easily and quickly can we generate the results? How quickly can we respond to an internal or external audit?

I believe a strategic approach should be adopted by banks and financial institutions with a long term vision in mind. What about the next regulatory reporting deadlines? Is it going to be yet another struggle? Will you continue to deal with costly and time consuming patchwork of systems or excel and flat files or are you going to be looking at a long term strategic approach where you will have the ability to run the entire process in an automated manner?

I think institutions are going to change down the line. For example, a bank that is pretty strong in their lending activities in one sector might start expanding into another sector. So how flexible would it be to expand and what is the impact going to be on my provisions for the new set of asset classes?

I think that’s really what banks and institutions should be looking at. They should be looking at a strategic approach. They should not be thinking tactical which will turn out to be inefficient in the long run. Sometimes you just have to take a step back and think how this process will fit into your entire infrastructure.