‘for mid-tier and challenger banks and lending institutions and consumer credit businesses, the challenge is equally demanding to that of the large global banks and in many ways requires the development of a more difficult bespoke solution that is proportionate, fit for purpose and not over engineered for ‘business as usual’ going forward.’


Much has been written following the arrival of IFRS 9 Financial Instruments back in 2014, as to the technical requirements and practical complexities of implementing the new rules.

As a recap, IFRS 9 is applicable for accounting periods starting on or after 1 January 2018 – so for those with a calendar year end, the 2018 financial year will be the first year end reporting under the new rules.  However keep in mind that comparatives and opening balance sheets will also be required.  The data, modelling, policy, methodologies and IT changes required are significant and we have seen major 2 or 3 year projects developed to meet the required changes in time.

From experience, the implementation and ‘go live’ challenge comes down to data, data, data …

IFRS 9 has three elements: ‘Classification and Measurement’, ‘Hedging’ and ‘Impairments’ – each of these areas have their own complexity and implementation challenges, however by far the new impairment rules are the most complex and troublesome and is the focus of the rest of this article.

The new impairment rules move the IAS 39 ‘Incurred loss model’ for financial instruments such as loans, receivables and certain investments (i.e. you recognise a provision only when a loss indicator/ event has arisen), to a new ‘Expected loss model’, where you have to now provide for future expected losses (possibly for the life time of the instrument).

It goes without saying, that this is a fundamental game changer, resulting in more provisions being recognised and those being recognised sooner.

This will fundamentally impact income statements, balance sheet, liquidity and key ratios with provisions increasing from 50% to up to 300/400% or more depending on the credit quality of the underlying portfolio.

Overall challenge:

Moving from the old ‘incurred loss’ model to the new forward looking ‘expected loss’ model, will require a fundamental change in accounting policies, credit policies and methodologies, systems, data, risk and governance to name but a few areas.

In practice, reporters will require (or some equivalent of) the following, to be able to implement IFRS 9 Impairments and report under these rules going forward:

  • Definitions and polices as to what’s a significant increase/ decrease in credit risk
  • 12 month and lifetime Probabilities of Default (PD’s) per exposure
  • 12 month and lifetime Exposures at Default (EADs) for each exposure
  • Loss Given Defaults (LGD) for each exposure, Cure rates and haircuts
  • Macroeconomic inputs

As IFRS 9 Impairments is forward looking, the ability to estimate, report and govern future estimates of expectations is inherently complex which will result in very material amounts and will be a key challenge.

Identifying the data needs, what data already exists’ what data needs to be cleaned or generated, the system changes and raft of accounting and credit policy choices and methodology decisions (minimum of 40 for a starter!) are not to be taken lightly.

Additionally although the rules have been out for a while, market interpretation is developing as others adopt.  The audit and accountancy profession are also in the throws of developing their own interpretations, opinions and views of what is needed.

This constant state of flux, is evidenced by the ITG (the IFRS Transition Resource Group for Impairment of Financial Instruments), who continue to debate and provide guidance on the required application of IFRS 9.

At one end of the scale, the large global banks are employing hundreds of staff globally to develop the solution and implement and are gearing up to make literally billions of calculations each period.

A proportionate challenge:

The large global bank’s challenge is in itself an unbelievable and monumental task.  However for those mid-tier and smaller banking and lending institutions and consumer credit businesses, the challenge is equally demanding and in many ways requires the development of a more difficult bespoke solution that is proportionate, fit for purpose and not over engineered for ‘business as usual’ going forward.

The challenges for such lenders are many and obvious, but worth outlining from experience:

  • Many won’t have a team of dedicated credit modelling specialists, or if they do, they may have one person who already has a full ‘day job’
  • Life time PD curves, EADs and LGD’s may not exist – for example lenders often have a limited 6 month spreadsheet of historic losses as the only information
  • The raft of data needed may not be available – do you create new data or use assumptions?
  • Depending on the maturity of the lenders, historic trend data to build loss expectations may not be adequate – this is quite common in emerging challenger banks or lenders with new market activity
  • Current credit modelling engines are often limited and somewhat of a ‘black box’ environment – how do you ‘open’ this box and determine what changes are needed and make those necessary changes?
  • Management may not have the technical skills and/ or capacity to develop solutions
  • Some lenders have outsourced their current IAS 39 engines to external providers – many of these providers do not as yet have a tried and tested IFRS 9 solution
  • The overall change is significant and needs to be carefully governed – a formal ‘Change Management Office’ or at least a well-rehearsed and controlled change approach may not exist
  • Software change capacity and/ or expertise to write new credit engines may not exist – even ‘replumbing’ data conduits to an ‘off the shelf’ solution (once available) may prove troublesome, time consuming and expensive
  • The required macroeconomic forward looking information may not exist – or if it does, is it bought in, or is it one person who generates this – if so, how do you control and govern its adequacy, consistency and reliability?

A proportionate solution:

As noted above, the challenges for mid-tier and smaller lenders to successfully implement the complexities of IFRS 9 Impairments are somewhat unique and different for each lender – the key to managing this change efficiently and effectively (and ultimately ensuring compliance with the new standard is achieved in time for a clean audit report), is your approach and the support you get in making this happen.

Whether this support is ‘in house’, contractors or through external advisors, you will need someone to help you not only advise, but to help ‘hold the pen’ in how you implement.

The challenge for advisors is that this is both a technically complex issue and one that requires a high degree of practical insight and experience in how best to implement.  As a mid-tier or smaller lender, you cannot have this ‘done’ to you by external advisors or modelling firms and risk ending up with an ‘off the shelf’ implementation project or solution forced upon you.  You will have unique needs and challenges and your approach must cater for this to ensure that you have a proportionate, not over engineered and ultimately workable solution.

Some are looking to engage external advisors or modelling firms to build or enhance their own credit risk engines (whether financial or regulatory) or buy in a software model: the key for them will be that they know upfront precisely what they need and ensure what is built or commissioned exactly meets those requirements.

In order to do this, they will need to have made the key technical decisions already, have performed a detailed gap analysis of where they are now, where they need to be and only then can they write an RFP for assistance, build specifications or detail the needs of an ‘off the shelf’ model solution.

There are a number of ‘smart’ approaches that advisors should bring to you for consideration in developing a proportional solution.

For example one area that should not be over looked is the use of assumptions – you will have areas of data missing and be faced with the choice of what to do.  We have seen from in-flight implementation projects, a number of areas where lenders can be ‘sensible’ in how they address this issue.  There is the potential to avoid expensive data creation work streams, by adopting assumptions in certain areas.

If you can prove that the assumptions are valid, stress test them now and going forward, to show even if you are out by +/- 100%, then the impact is not material, then why would you want to incur the cost and effort of generating new data and reproducing this on a regular basis?

When do you need to start?

In short, already…

If you have already commenced a project to understand and implement IFRS 9 Impairments, that’s a healthy position to be in, but momentum and progress needs to be ensured at all times going forward.

If you have yet to start, then time is an issue for you; right now is the latest that you can feasibly leave to start the process – there is no quick and easy way to approach this issue.  Understanding what you need, how best to do it and who you need to advise and support you, is not a quick and easy process – realistically we have seen even what would look like the simplest non-core IT change take 9 -1 5 months to complete and UAT.

If you leave this any later, then what will happen is that the end testing (parallel running and UAT etc) will be condensed and rather than the minimum three cycles of testing (one to build, one to correct and one to confirm it’s ready to go live), will all be condensed into an unworkable and high risk time period.  This can be avoided by starting the process now.

Next steps:

If you have already commenced with an understanding exercise of the requirements and maybe moved into a gap analysis – then taking stock of progress so far, reflecting on the thoughts above and building these into your continued implementation plan would be beneficial.

If you have yet to start, you need to get things moving quickly, but in a sensible pragmatic way – a lot of the stress of implementing IFRS 9 Impairments can be reduced by having a well thought out and achievable way forward planned – by doing this, you will remove the uncertainty and have comfort that you can meet this demanding and technically complex new reporting standard.

Such an understanding and having a plan, is vital for you when your various stakeholders ask for an update on progress and potential impact (such as the C-Suite, Audit Committee, NEDs, Auditors, Investors and your Regulator).

The people you have leading this project will be critical to your success, whether they are internal or you engage external consultants, you need to ensure those you work with have both the technical competence, but more importantly the practical experience of implementing IFRS 9 Impairments – such people are in great demand right now and also few in number.  Ideally your advisor will be able to leverage off a range of in-flight implementation projects, a varied range of clients and also preferably have early adopted it already (the latter is somewhat unique given only one major group has adopted early).