Technology Perspectives from the IFRS 17 Journey

Kokleong Choy, FIA, Director, Risk Software, Aon Center for Innovation and Analytics (ACIA), Singapore, explains some of the technology challenges facing insurers as they implement IFRS 17

As the implementation deadline for IFRS 17 in 2022 draws ever nearer, the insurance industry is in the midst of preparing for the standard. However companies are at various stages of preparedness from having already started implementation to just preparing the gap analysis or selecting technology vendors.

Many of our clients that license ReMetrica as a capital modeling and reinsurance pricing platform are now dealing with the challenges of implementation and in several cases, use our technology to provide actuarial inputs to IFRS 17. Our experience working with our clients has given us some perspective on the challenges and opportunities of IFRS 17 implementation.

The primacy of data to the implementation process

Without question, data is the most important and challenging component of IFRS 17 implementation. The calculations behind the standard have been judged to be complex and demanding in both the quantity and granularity of the data required.

"Data is the most challenging component of IFRS 17"

The most immediate impact of the standard on some companies is that they find themselves lacking the requisite amount and quality of data needed to implement IFRS 17. As a result, these companies would need to make simplifying assumptions in their calculations. While transitional arrangements may suffice for now, going forward, these companies may need to create new processes to capture the data needed for future calculations.

For companies that have the requisite data, the challenge then becomes one of putting in place the required processes to marshal the data for calculations.

With IFRS 17, while the calculations needed are highly granular, it is essentially the same calculations that need to be performed across different groups of contracts. This naturally suggests the use of technology to automate calculations.

Importantly, we have identified that automation used for IFRS 17 must:

  1. Be performed accurately at high throughput;
  2. Have strong data archival, retrieval and audit capabilities. It is important to note that just as IFRS 17 implementation consumes large amounts of data, it will also generate large amounts of data that need to stored and referenced at future periods, e.g. the contractual service margin of groups of contracts at inception. Archival and retrieval is hence important. For the same reason, audit and data governance and control need to be key aspects of the automation system.
  3. Be interoperable with other software platforms, since data in each company is likely to come from a variety of sources with an equally rich diversity of formats.
  4. Where possible, be capable of interfacing or being integrated with the general ledger system for reconciliation and posting purposes.

Even with the use of automation technology, data in its current state in most companies usually requires substantial processing before being fit for IFRS 17 implementation. At a minimum, data would need to be broken down to the appropriate granularity for calculations while at the same time retaining the ability to be aggregated for reporting purposes.

Having worked with clients on internal capital model development, we understand the importance and difficulties with data. We introduced the automation workflows feature in ReMetrica to allow our users to process data in situ in ReMetrica, enabling seamless data handling and modeling. This feature has been pivotal in allowing our clients to leverage the use of internal model builds for facilitating IFRS 17 implementation.

Because of the central importance of data to IFRS 17 implementation and the considerable difficulties most companies experience, we recommend that all companies set aside a liberal budget in both time and resources to handle data issues. The challenges of getting internal company data to specification are significant and should be respected.

Emergence of new key performance indicators to track

The implementation of a new accounting standard also brings with it the emphasis on new measures of performance. For one, the contractual service margin represents the future profit embedded in insurance contracts and is almost certainly going to be one of the key performance indicators (KPIs) that will be tracked over time, among other indicators.

Alongside the implementation of IFRS 17, companies should consider how management processes should evolve to track the new KPIs that arise. The introduction of new tracking processes is arguably best done alongside the considerable work that needs to be done to satisfy the disclosure requirements mandated by IFRS 17.

Figure 1: Example of ReMetrica IFRS 17 components

Source: IFRS 17 Essentials AON - InsuranceERM – Winter 2019

ReMetrica is most commonly used by our clients for capital modeling and projections. Just as with the Solvency II Reporting component that we introduced prior to its implementation, we now provide the IFRS 17 Reporting component for companies that want to stochastically project their future income statements and balance sheets on an IFRS 17 basis.

The IFRS 17 Reporting component in ReMetrica calculates all major components of IFRS 17 calculations, including the contractual service margin at inception, release of risk adjustment and contractual service margin during contract term, best estimate liabilities and liability for remaining coverage, as well as considering the impact of reinsurance.

"Having projections made on an IFRS 17 basis provides business insight "

Having projections made on an IFRS 17 basis provides insight into how the business may perform in the future and hence facilitates business planning and strategic decision making. Just as importantly, new performance measures represent one of the most visible outputs from the entire process of implementation to management. This high visibility is a reminder of how IFRS 17 will impact the way the insurance industry does business in the future. When tracking of performance measures is introduced alongside implementation, this will help ensure that teams within the company remain committed and on target for implementation.

Given that companies in the insurance industry will be judged and valued in the future based on IFRS 17, it would be prudent to devote some of the bandwidth set aside for implementation on devising processes related to performance measure tracking as well.

IFRS 17 as business transformation

IFRS 17 implementation will be costly for the insurance industry as a whole. However, an implementation need not be viewed purely as an expense, but can also be construed as an investment in the future.

Some companies in the insurance industry, for various reasons, have chosen a minimalist approach as necessary for implementation of the standard. Common reasons for doing so are to reduce costs and project implementation risks.

Other companies may take a more future-oriented view, taking the perspective that IFRS 17 represents a forcing function that prompts a re-engineering of the company to one that is more dynamic and data-driven. These companies may consciously decide to invest more in the implementation process so as to derive greater value from the work that goes into implementing the standard.

Both approaches can be valid depending on the individual company's situation.

For the latter group, we have seen some of our clients express the desire to invest in technology hitherto not used in their organization. Common points of interest among companies are in the use of business intelligence, data visualization and predictive modeling tools.

The rationale for investing in these tools is clear once we consider that:

1. As indicated earlier, companies will need to develop new processes to track the new KPIs that emerge as a result of IFRS 17 implementation. It then becomes a natural course of action for companies to consider the newer business intelligence tools available on the market that make business intelligence gathering and dashboarding faster and easier.

Equally, this reinforces the importance of interoperability between software platforms when choosing a technology platform for automating IFRS 17 calculations. A highly interoperable system will easily afford the generation of business intelligence through specialist third party software and hence the creation of actionable insights. The benefits of such a system arrangement will drive broader acceptance of the necessary process and organizational changes required by IFRS 17.

2. The highly granular data requirements of IFRS 17 dovetail with the needs of predictive modeling projects.

With data science and artificial intelligence at the forefront of many business conversations, more and more companies are trying to find business value in their repositories of customer data. Given that insurance companies will need to go to considerable expense to get their data up to specification for the implementation of the standard, it hence makes perfect sense to find more ways to use the data to justify the large upfront investment.

In many ways, insurance companies have always used data for the traditional concerns of pricing and marketing insurance contracts. Some have also used data for automating claims and underwriting processes. The difference now is that the granularity of data required for IFRS 17 may drive more targeted uses of the data and uncover insights previously unknown.

"The granularity of data required for IFRS 17 may uncover insights previously unknown "

With our own clients that use ReMetrica for capital modeling, we have recognized the same drivers behind adopting other technology as we are now seeing with IFRS 17, namely the desire to get more value out of modeling results and the desire to communicate said results more effectively.

Because of this, ReMetrica has the capability to interface with the two most commonly used open source tools by our clients, Python and R. Starting from the next release of ReMetrica, our users will be able to call ReMetrica externally from Python and work with model results in their preferred environment.

We recognize that companies frequently use different technology 'stacks' and that flexibility is important in any software platform. As we can see with IFRS 17, the implementation of a new accounting standard is now driving the desire to invest more in business intelligence and predictive analytics.

Integration of existing processes

In our interactions with clients, we have seen in some companies a general reluctance to overhaul or otherwise abandon existing processes. For example, spreadsheets (with VBA code) still play a dominant role in actuarial processes such as pricing in many insurance companies. This is despite the commonly stated desire to remove spreadsheets from business operations where possible, whether to reduce the incidence of errors or to enforce more robust version control. Competing with these reasons is the desire of many actuarial and finance teams to retain review and oversight over calculations, which is often felt to be best done in a 'transparent' medium such as spreadsheets.

Figure 2: Example of IFRS17 Workflow automation

Source: IFRS 17 Essentials AON - InsuranceERM – Winter 2019

It is almost the certainly the case that considerable time and effort had gone into establishing these existing processes such that they are part of the company's status quo.
While IFRS 17 is certainly transformative, realistically, not all existing processes need or should be abandoned in the course of implementation of the standard. The work that has already gone in developing existing processes and their associated artifacts can represent a starting point for implementation.

It is debatable whether retaining and integrating existing processes into a new IFRS 17 system is riskier than overhauling the entire system. However, the advantage of doing so is that it will likely reduce the time and work required for implementation, and it cannot be overstated, increase the comfort level for those involved in implementing what is widely seen as a challenging and complex project. This is particularly so for companies who have not embarked on a similarly complex project such as Solvency II.

We believe that companies should adopt a phased approach where they integrate existing processes into their IFRS 17 implementation, reducing time and work required. We have consulted with clients who have wanted to use ReMetrica in conjunction with their new and existing processes for IFRS 17 implementation and this can usually be done with a reasonable amount of customization work.

One direct insurer required an actuarial system that could process claims data from over 400 different product codes across several product families. In addition, these product codes were split across numerous categorizations that needed separate granular processing in multiple spreadsheets. We designed a workflow that permitted the loading of data automatically into these spreadsheets, which would then be automatically incorporated into a ReMetrica model. As a bonus, the workflow was designed to test the fidelity of the data prior to loading to minimize errors and omissions further down the workstream, which would have been much more costly to resolve later on.

"Companies should adopt a phased approach to integrate processes"

Another reinsurer stated that they would consider only actuarial systems that could demonstrably replicate the calculations from their own custom designed spreadsheets. While it seemed that this client was open to exploring the use of a system to replicate their calculations, it quickly became clear after several discussions that their intention was for the new system to aggregate the calculations from their existing approach rather than a wholesale replacement of it.

Based on our interactions with clients, we believe that the integration of existing processes is the more common approach instead of a wholesale replacement approach for IFRS 17 implementation. While perhaps conservative, for many of our clients, this is the usually the most desired approach and the most straightforward and workable one.

As mentioned before, a flexible technology platform for IFRS 17 should be interoperable and be able to accommodate existing processes.

Once new processes have had time to bed down, companies can then explore retiring older technology and processes in favor of newer approaches. This commitment is important at the outset as it will help to mitigate the risk of obsolescence in the future.

Conclusion

As a technology provider and partner to companies in the insurance industry, we have been privileged to be able to observe firsthand many of the challenges our clients have faced and be party to many conversations and discussions revolving around how to surmount these challenges.

Importantly, as outlined in this article, many of these challenges also represent transformative opportunities that rarely arise in the industry.

In implementing IFRS 17, we believe that a thorough assessment of the considerations we have highlighted here will allow companies to make a more informed choice of the technology to meet each insurer's strategic objectives.

To find out more information, contact Dean Roberts.