Ladies and gentlemen, welcome to the IFRS 17 at Zurich conference call and live webcast. I am Sandra, the Chorus Call operator. I would like to remind you that all participants will be in listen only mode and the conference is being recorded. The presentation will be followed by a Q&A session. You can register for questions at any time by pressing star and one on your telephone. For operator assistance, please press star and zero. The conference must not be recorded for publication or broadcast. At this time, it's my pleasure to hand over to Jon Hocking, Head of Investor Relations and Rating Agency Management. Please go ahead, sir.
Good afternoon and good morning, everybody. Welcome to Zurich Insurance Group's IFRS 17 education event. I'm Jon Hocking from Investor Relations, and I'm joined by our two presenters today. We've got George Quinn, our Group Chief Financial Officer, and Karthik Thilak, who's the Group Head of Financial Accounting and Reporting. George and Karthik aren't going to present every slide, but the full deck is available to download both from the webcast platform and also from our website. The presentation is going to last around about 45 minutes, and then we'd be very happy to take questions. When we get to that point, if everyone could keep it to two questions, that'd be much appreciated. With that, we'll hand over to George to kick off.
Thank you, Jon. Good morning or good afternoon, everyone. As Jon mentioned, Karthik is with us. Prior to his role as group head of financial accounting and reporting, Karthik was our IFRS 17 program leader, so he's particularly well-placed to guide us through the effects of IFRS 17. The aim today is to provide a view of how the transition to IFRS 17 will affect Zurich's financial results, disclosures, balance sheet, et cetera. We'll touch on that together with some introduction to how it will impact KPIs in the future. It's going to be largely qualitative, but the information we've given you is based on the best of our knowledge at this point in time.
We'll go through some of the decisions that we've made, some of the judgments that we've taken, and walk you through how we see these impacting balance sheet and income over the next few years. A couple of caveats before we start. This year is obviously not yet over, which means that I can't predict precisely what the outcome for this year is going to be under IFRS 17. We're still learning. We're still fine-tuning things. It will continue for some time to come, but we obviously have sufficient confidence to make this presentation today. Many of the comments are going to reference 1/1/2022 as the starting point, and you'll hear this especially as regards equity impacts.
The IFRS challenge this year is not that the IFRS 17 balance sheet is especially volatile, but the IFRS 4 balance sheet is. For example, the reduction in equity that you're going to hear referenced at the beginning of the year is much smaller by the end of the first half, and it's not at all inconceivable by the end of the year that IFRS 17 equity could be greater than IFRS 4 equity. I'm going to try and avoid slipping into jargon immediately, but as you're going to discover quite quickly, that's not easy. There is a new language to learn as part of this effort.
The event we're having today is also probably a bit earlier than we might ordinarily have planned, but given we're bringing IFRS 17 in the shape of targets to the Investor Day, we felt it was important to establish the right context for that presentation in November and hence why we're presenting today. As John said, there's going to be plenty of time afterwards for Q&A. Some key messages. The most important point that I want to make today is that we don't expect the adoption of IFRS 17 to lead to any significant changes at Zurich. It doesn't lead to significant changes in the group earnings. It doesn't lead to changes in the underlying economics of the business. It doesn't change cash or capital.
It does change equity, but as that's simply a different presentation of the same economic balance sheet, it doesn't really impact us. Consequently, the things that matter most to us, strategy, risk preference, and of course, most importantly, the dividend policy remain unchanged. Now, if that's all there was to say, this would be a pretty short presentation. Of course, while the big picture is not changing too much, the underlying detail is. Here you can see a bit of the journey we've been through. We've been running a dedicated group program focused on the joint implementation of IFRS 17 and 9 since 2017, even if it sometimes feels like it was 1917.
I think I've said to some of you that I had no children when the IASB added this topic to their agenda, and the children that I've acquired in the intervening period are not children anymore. To put it positively, we've had plenty of time to prepare, and we think we're well-positioned. There are some unexpected benefits. The transition has afforded us opportunities for transformation and simplification, including the development of some new platforms. For example, we've accelerated the adoption of cloud-based technology. We've completely overhauled the actuarial systems, and we have new workflow tools around data analytics and performance management. I'd also highlight the efforts of the CFO Forum on implementation and especially the work together with my peers at Allianz, AXA and Generali.
Together we've prioritized consistency and comparability because we recognize the value that this has for all of us. IFRS 17 is up and running. We're going through a quarterly parallel closes under both IFRS 4 and IFRS 17 standards. We're running the financial planning process for the next three years entirely under IFRS 17. We're looking forward to welcoming you back here soon to present the 2023-2025 financial targets at the Investor Day in November. In terms of disclosure, the full year 2022 results will continue to be based entirely on IFRS 4, but we'll include a detailed presentation on the IFRS 17 transition impact.
While the three-month update that we give in May next year will already be based on IFRS 17 and nine, the first full reporting for us under the new accounting standards will take place with the disclosure of a half-year 2023 results, which is next August. IFRS 17 introduces an approach to accounting for insurance that's generally described as the general model. Then there are variations or simplifications that reflect the different extent of policyholder participation in investment or insurance entity performance. Broadly, these are the non-participating or indirect participating model referred to as the building block approach or BBA. The direct participating model, which are here referred to as the variable fee approach or VFA.
For short duration contracts, IFRS 17 allows the application of a simplified so-called premium allocation approach or PAA, which can be applied to contracts that have a coverage period of up to a year. Or for multi-year contracts, where the results using PAA wouldn't differ significantly from those using the BBA. Zurich's general approach across the group is aimed at applying the standard while achieving as much consistency across the reported results across businesses and geographies. Therefore, the simplified or PAA approach is mandated to be used for all property and casualty insurance business. While the default accounting approach for our life business is the general model, so the BBA or the VFA. You can see this approach outlined on the slide together with, at the bottom, what is a very short list of exceptions to those general rules.
The standard is principles-based, and that means that there's a number of areas we have to make choices or judgments in its application. Where we've made those choices, the group's main criteria has been to limit volatility, as far as possible, ensure the stability of earnings, achieve operational simplification, as well as much as we possibly can to recognize the underlying economics of the business that we write. Let me highlight a number of key accounting choices that we've made in this respect. IFRS 17 introduces a current measurement for all insurance liabilities. It allows, and we've opted for the use of locked-in interest rates, to discount loss reserves, although not for VFA, and this reduces volatility in the P&L.
The standards requires the revaluation of the balance sheet reserve using current rates at each reporting date, with the difference in evaluation between the current and locked-in rates being a component of other comprehensive income. We've also chosen to adopt interest rate definitions that as much as possible are close to the local regulatory requirements, at least as far as is permitted under the standards. This introduces some diversity at the country level, but it means that the accounting basis is a bit more consistent with capital generation. We've adopted for zero OCI at transition. The P&C businesses will apply the relevant discount rates for the period from 2015 onwards, 2014 for the U.S. business, but 2015 for the rest. Transition date discount rates for all prior periods.
This reduces the expected positive impact of discounting on OCI at transition. It does create somewhat of an earnings headwind as the discounting effect builds up to run rate over time. Determining the discount rates, we're applying the default bottom-up approach, so therefore risk-free plus an illiquidity premium. For risk adjustments, we're applying the percentile or confidence level approach. The calibration of risk adjustment considers the probability distribution of future cash flows and the additional amount to be held above the best estimate, an estimated probability in order to meet that target confidence level. In doing so, we've defined different percentile ranges for life and non-life separately.
Given the much longer duration of the life business, we're applying a much higher percentile risk adjustment to the life business, which we expect to have a stabilizing impact on earnings over time. The calibration approach for the risk adjustment percentiles is in line with our target SST ratio of above 160%. IFRS 9 introduces an option to account for certain equity securities at fair value through other comprehensive income, but the group does not intend to make use of this option. This choice will increase the volatility of the investment result within the P&L and consequently NIAS, but recognizes that over time, we expect to generate income from these investments. We already have volatility to some degree with realized gains and losses, and impairments in the current P&L.
The group has also applied retrospective transition approach whenever practical, and expects that most groups of insurance contracts will follow either the full retrospective approach or the modified retrospective approach. The fair value approach, which we'll apply whether there's no reasonable or supporting information to enable us to apply the retrospective approach, we expect to be impacting less than 20% of the CSM. As you can see on this slide, and subject to my earlier comments about the trajectory of equity over the course of this year, shareholders' equity under IFRS 17 at transition in this context being 1/1/2022, is expected to reduce compared to the equivalent IFRS 4 position. That's largely the most notable change that results from the application of the new standard.
The transition equity impacts can be broadly grouped into accounting policy choices that were made under IFRS 4. For example, DAC amortization, real estate valuation for us. IFRS 17 accounting mechanics. The creation of CSM, the introduction of discounting, risk adjustment, and things like the level of aggregation for onerous contract testing, and then transition calibration decisions. For example, the things we've adopted with the intention of reducing volatility in future. For example, risk adjustment. The majority of the expected equity reduction will be driven by retained earnings, and that's mainly as a result of the somewhat more back-ended profit recognition that you get under IFRS 17. We're looking at the change from a segment perspective. The group is expecting limited to positive equity impacts from short-term P&C business and reductions from long-term life.
As I mentioned on the prior slides, earnings are not expected to change significantly compared to IFRS 4. Thus, essentially the changes to shareholders' equity that I've just described will lead to an expected mechanical increase in return on equity of up to about 300 basis points compared to the current underlying levels of around 15%. However, given that IFRS 17 is an accounting change only, it does not impact the underlying economics of life and P&C and therefore does not change our view of their relative attractiveness or in some cases, unfortunately, unattractiveness. Having said that, given the changes, we'll reflect this mechanical change and the starting point for our ROE ambition when we set the targets for the new strategic cycle at the Investor Day in November.
The introduction of IFRS 17 and 9 represents an accounting change only. Again, no underlying changes in the economics of the company. Therefore, we don't expect to see any change in cash or capital generation. There might be some second-order impacts, but these will be limited to areas where we see or we accept IFRS 17 as a starting point as a convenience. Leverage is expected to remain at a comparable level on a comparable basis post-introduction of IFRS 17. Our reported ratio will include CSM and risk adjustment. This is necessary and appropriate given the loss-absorbing characteristics and the fact that in the case of CSM, it can be monetized. As a consequence, our dividend policy is not impacted by the adoption of IFRS 17.
To conclude, for this part of the presentation, just let me reiterate the key messages. No significant change. There is a reduction in life equity and a mechanical increase in the ROE, but the underlying economics of the businesses are unchanged and earnings are not expected to be significantly affected and cash and capital are unchanged. I think once we all get used to it, IFRS 17 should lead to better comparability across lines of business in the reported results. Should also improve the visibility of profit emergence in the life business. Again, we don't expect this to lead to any change in our targeted business mix. Finally, neither our dividend payment capacity, the dividend policy, or our communicated payout ratio target will change following the transition to the new standards. Let me now hand over to Karthik.
Following Karthik's section, as Jon said earlier, we'll open for Q&A. Thank you.
Thanks, George, and good morning, good afternoon, everyone. It's a real pleasure for me to be here today because as someone who spent the last few years of my life working on IFRS 17, this represents an important milestone because it's a first in a series of communications which we have planned on IFRS 17, and it's all becoming real, so it's really exciting. You know, as George said, as an industry, we have come a long way in working with the IASB in defining these new standards and making them fit for purpose, you know, not just for management, but for you as the users of the financial statement. As you know, there were three main objectives of the new standard. It was about increasing consistency, especially with the economics of and the accounting.
It was about improving transparency in long-tail businesses, you know, like the life book. Most importantly, it was about improving comparability of the financial statements. You know, the aim was not even to improve comparability within the insurance industry, but even across industries. Now having been through the experience of implementing IFRS 17 and enabling the actual transition, I think we are on track for the improved consistency and transparency. Achieving a greater degree of comparability, it's going to take a few years. I expect that there needs to be more convergence of practice, which will form and new baselines will also be set because the presentation's going to be different in the new world.
Today's session is also aimed at really accelerating that because we would like to give you a closer look on the application choices which we've made, which we believe reflects our business model and also the underlying economics. With that, let's start with P&C. Given that we have applied the simplified approach or the premium allocation model for the whole of the P&C book, we expect to see only limited impact on the P&C financial statements. With the changes in IFRS 17, and the simplified model, there are two new concepts, right? It's discounting impact of the reserves, and it's also the recognition of risk adjustment.
We'll spend some time going through these new concepts because you will notice that they create a visible impact across the whole of the P&C financial statements. There are also a few other elements which will lead to slight differences in a combined ratio and in the reserves, but we expect these to have a limited impact compared to today's picture. IFRS 17 also requires us to test for contracts if they are loss-making at a much more granular level, but we don't expect the P&C onerous contracts or the loss component to be material at the group level. On the presentation side, one immediate benefit which we do see is that, to a certain extent, IFRS 17 helps comparability between the different lines of businesses as well.
Overall, you know, while these new mechanics, they can impact presentation, and in some cases, they can also affect the timing of profit recognition, we don't expect any of that to be substantial to have an impact on our target business mix. The P&C business mix should remain unchanged. On this slide, let's start with a comparison of the old versus new of the most important P&C KPIs. Other than insurance revenue, which now becomes the new top-line metric, the other metrics are what you're already used to. Insurance revenue, it will replace GWP, but please note that it is not the same in definition. Insurance revenue is much more comparable to gross earned premium from the current IFRS world.
The other main difference is that in our case, we will include premiums from single parent captives on a net basis within the insurance revenue. I expect insurance revenue will become a really important metric going forward because it is the main driver of the profit for the P&C business, and it will also replace net earned premium as the baseline for the calculation of the combined ratio. The combined ratio, that will also change, and where the insurance service result, which is on a fully loaded basis, including the reinsurance result, will form the new numerator, and gross insurance revenue becomes the new denominator. We call this the net gross combined ratio. We've provided you some illustrations in the deck, which would highlight how purely the presentational differences can also lead to slight changes in the ratio itself.
We expect that the P&C book and the P&C shareholders' equity will remain fairly stable. Though at transition, the P&C shareholder equity will increase, and that's mainly due to the effect of discounting of the P&C reserves. I'd also like to highlight two principles which we followed in definition of the new IFRS 17 KPIs, not just for P&C, but also for life. First, you know, as the KPI definitions themselves, they are not stipulated by the standard. To the greatest extent possible, we've considered peer alignment in those definitions. And it's in our interest to stay comparable with the peers. However, we'll continue to, you know, monitor how the market practice is developing, and if required, we will fine-tune our KPI framework. Second, we've always tried to keep the financial statements as the basis for the derivation of the KPIs.
That way we don't have to maintain any more shadow or offline metrics. For you, it should also help you trace the KPIs back to the actual financial statements. On this slide, you see the new phase of the income statement under IFRS 17. One key point to note is that while IFRS 4 always focused on revenues over expenses, you will notice that IFRS 17 brings a bundled margin view. In my head, I always think about them as concentric circles, which is centered around the insurance service result. Within the insurance service result, you notice that the reinsurance result is also presented separately. It is much more transparent on how is the direct business performing and what is the contribution from the reinsurance. Investment results, so this is mainly the result of the investing activities.
You have the fee result. Fee result will include fees from any unbundled services which we provide to our customers, so like risk engineering services. In our case, most importantly, it also includes the result of the Farmers Management Services company, which is unaffected by IFRS 17. Other result, this will mainly be non-qualifying expenses and other. Note that these bundles, they don't change whether it's a P&C business or a life business, and it also doesn't change based on what measurement model you apply. That's one common view of presenting the IFRS 17 results. Also note that the bottom line metrics themselves, they don't change. What changes with the simplified model? I mentioned it's mainly discounting and risk adjustment. Let's talk about discounting in more detail.
The concept of discounting itself is not new to you, but the difference now is that it applies to the whole of the P&C reserve. How do we discount? We apply the bottom-up discount rates, and that's based on risk-free, and we will include the illiquidity premium. The base rates for risk-free are very similar to EIOPA curves, so we should be consistent with our major European peers as well. Illiquidity premium is determined using an approach which is very similar also to the EIOPA volatility adjustment. To limit the volatility in the P&L, we have chosen also the locked-in option. We lock in the current accident year rates, which are used for discounting and any financial impact from the difference between the current rates and locked-in rates that is recognized in the other comprehensive income in the balance sheet.
How does the mechanics work? All the claims reserves, they get discounted on day one, and then the effect of discount is a benefit to the current year, in the P&L because you create lower, liabilities. The unwind of the discount, however, in the subsequent periods, is a charge to the P&L. Note that there is an important presentational dynamic here because the standard requires us to recognize the unwind of the discount, you know, which is a charge in the subsequent period as insurance finance expense, and that is included in the investment result and not in the insurance service result. This presentational difference can also lead to slight changes in the ratios itself. In the current interest rate environment, it can have significant effects.
The last component of the transition impact is the discount. As part of transition, as it was mentioned by George, we will go back until 2015 with the historical yield curves for discounting the historical reserves. For any periods prior to that, the historical reserves are bundled together, and then we apply the December 2021 forward curve, and this is the zero OCI method. Overall, discounting will have a positive effect on the P&C shareholders' equity at transition, but with the zero OCI option, that effect is lower. Now that we've covered discounting, let's talk about risk adjustment, which is an additional allowance for uncertainty. While discounting reduces the claims liabilities, risk adjustment is an additional risk margin, so it increases the claims liabilities in the current accident year.
We will use the percentile method for calculating the risk adjustment, which in our case is based on SST, and we will also calibrate it for the cost of capital. As George mentioned earlier, we are going to be using different percentiles for life and P&C, and that's mainly due to the difference in the duration of the underlying liabilities. Mechanically, the process of booking risk adjustment is very similar to discounting, but it just works in the opposite direction. Risk adjustment, which is applied on day one, it's a charge. Then over time, when it is released back to the P&L, it is a benefit. While in the picture it shows that risk adjustment and discounting are offsetting each other, it is true. Directionally, they will offset each other. The net impact per accident year depends on various circumstances.
For example, the business mix or the interest rates which we use for the discounting, or in certain cases it could also be changes in the capital framework which affects the risk adjustment calculation. Directionally offsetting, but quantum can be different. An important point to note on risk adjustment is the release of risk adjustment back to the P&L in the subsequent periods will be included as part of PYD, prior year development. We expect PYD to also remain in a favorable level in the future. To some of the pictures on liabilities, you know, the implementation of IFRS 17, we also don't expect it to have a significant impact on our nominal reserves, and this is mainly because we already apply best estimate principles to reserving in the P&C space today.
IFRS 17, though it doesn't recognize the concept of reserve strength, so we will make an allowance for limited historical experience within our best estimate liabilities. Think of this as an additional prudence within our P&C reserves to cover for any reserve deficiency, for events which are not sufficiently reflected in the historical experience. I believe some in the market are also calling this as resilience reserves. To summarize the application choices, for P&C under IFRS 17, the nominal reserves, they remain similar, including the LHE, while recognizing the risk adjustment adds, you know, additional prudence and it increases the liabilities. Then discounting is more a financing component and that reduces the liabilities. Now let's move on to ratios. I had mentioned earlier that IFRS 17 combined ratio is more comparable across the different lines of businesses.
We have selected two examples here for an accident year combined ratio walk, which shows you the difference between IFRS 4 and IFRS 17. Adding the discounting and risk adjustment, you know, impacts to the overall picture, it provides for a better portfolio compatibility, especially between the short tail and then the long tail lines. Discounting has a bigger impact, giving a larger benefit to the longer tail lines, because you anticipate that there's adequate investment income to absorb the higher reserve unwind. Risk adjustment is slightly larger for property, and that's because there are more volatile factors like cats or large losses in that line of business.
The other factors which impact combined ratio, it is the use of reinsurance because now the reinsurance result is entirely presented together, and then also the fact that insurance revenue now becomes the new denominator for the calculation of combined ratio. Again, remember that other than the effect of discounting and risk adjustment, which can have an impact on timing of profit recognition, the rest are purely presentational differences, and we don't expect any of this to change the relative risk appetite between the different lines for us. Last point on P&C, investment result, right? Investment result in the future, it will also include the unwind of the discount, which is now booked as discount of the insurance liability. Now what this effectively does is it turns the net investment result into an investment margin view.
You know, this is the second bundle of the bundled margin view. The levels of the P&C investment income itself, as well as the capital gains and losses from hedge fund portfolios, which we today include in the BOP, that will not change because it's not affected by IFRS 17. In summarizing the P&C section, there are new elements of risk adjustment and discounting, which will result in both changes in timing of profit recognition in certain cases and also presentational differences. The ultimate outcome, it never changes in the new standard. Insurance revenue will become a key metric, and combined ratio definition will also be reset. On the actual financial impacts, the strength of our reserves doesn't materially change.
We don't expect any material change in the P&C bottom line, and the P&C shareholders' equity will increase on transition date mainly due to transition effects. We don't anticipate any of this to be substantial to have a change in our target business mix. Now we can move on to life. I keep saying life is more complicated and life accounting is also more complex. I'm going to do my best in trying to break it down into three blocks. We'll cover first the measurement models, then we'll talk about the mechanics and dynamics of these measurement models. Lastly, we'll also cover presentational differences under IFRS 17. The life business, it sees the most significant changes with the move to IFRS 17.
While we don't expect our earnings level to materially change compared to IFRS 4, the profit signature of the life results will be different. Remember also the transparency objective of IFRS 17. IFRS 17 does bring the margin view also to the life book, and you will see that in two places, you know, where I really like the new transparency. One is the presentation will increase the visibility of the profit emergence. Two, there will also be more transparency on how new business translates into the IFRS results, and I'll demonstrate that to you. From a balance sheet view, the equity at transition will be lower, and there are multiple drivers here, but the most significant one is the recognition of the future stock of profits, the contractual service margin or CSM, as a liability on the balance sheet.
While a lower, you know, equity, it will result in a higher accounting ROE, we don't expect this to change the relative attractiveness of the life business or any of the underlying life products. To be clear, our focus on protection and on unit link, it will continue. Lastly, unlike P&C, there are different measurement models which are applied to the life contracts, and this is based on the nature of the business, and we'll see about that in the next slide. You probably noticed that you don't see GMM or general measurement model on the slide, and let me just clarify that. Purely for convenience, we've chosen to use VFA for all direct participating contracts, and for the rest, we use BBA or building block approach just to differentiate it.
I think the best way to read this slide is if you go vertically by the different lines of businesses which are shown on the slide, and I'll cover the most important portfolios. Most of our continental Europe savings product, they will qualify for application of the variable fee approach. Given the long tail nature of this book, these are also the most significant contributor to the CSM stock at transition. Under IFRS 17, unit link products, they may be considered either as insurance products, in which case we apply the variable fee approach, or in certain cases, they may even be considered as an investment contracts, in which case they are not under IFRS 17, but they are accounted for under IFRS 9 basis.
The classification between insurance and investment, it really depends on the materiality of the risk which is transferred as part of the contract. On the protection portfolio, the bulk of it is accounted for using the Building Block Approach. Then the truly short-term protection book, which we specifically sell in Latin America and through the Zurich, Santander joint venture, we will apply the PAA or the simplified approach also for that portfolio. Already today, we effectively manage it as a P&C line of business. Now, before we look into the mechanics of these measurement models in the following slides, on the right side, you see how material are these different measurement models, and that's both based on the reserves and the investment liabilities, but as well as the contribution to BOP.
Keep in mind that this picture is from first January 2022, so it still includes our German back book. Once that disposal is completed, the share of VFA, it will be materially lower. In summary, you should think of a life business as three blocks. You know about 75% of the business is accounted for using the VFA or the BBA approach or the general model. Then you have the PAA block, which you already know how it operates. It's very similar to P&C. Then the IFRS 9 block, which is purely investment contracts, and they are managed as assets under management, cost-income ratio, fee revenue. Those would be the KPIs for that. To the interesting part. Let's talk about the mechanics and dynamics of these different models. I'm going to explain this more from a P&L point of view.
Just assume that by this point we've already calculated the CSM, and that's recorded. The CSM stock is recorded. The BBA it is the default measurement model for insurance contracts under IFRS 17. This model will apply for all of our protection business other than Latam and also certain indirect participating contracts, like the ones we sell in Spain. Under the BBA model, the key profit driver is the earn through of the profits through the CSM amortization and the release of the risk adjustment. If you follow through with the chart on the right side. There is one dynamic to note here that the risk adjustment release under P&C, it typically tends to follow the same pattern as the release of the reserves. However, on the life side, the risk adjustment is released based on the underlying risk driver.
It's very much possible that the risk adjustment release has a different pattern than the CSM release pattern. You book the loss component for anything related to onerous contracts. We don't expect loss components to be material at a group level, but it may be visible at a regional level. There is a volatility reduction choice here, so all economic variances such as change in discount rates, they will impact the shareholders' equity directly through other comprehensive income without hitting the P&L. If operating variances occur, there is a check, right? It is first validated if it is pertaining to the past or the current periods, in which case that's recorded against BOP. If it pertains to future profitability, then the operating variances are also absorbed by the CSM, and then they affect the P&L only through the CSM amortization.
Together, all those components, they give you the insurance service result. Once again, the first bundle. Then beyond that, for the BBA business, we also expect that there is a positive BOP contribution from the investment result, and the other result will mainly be non-qualifying expenses. Now, if we move to VFA, I think it's a bit more easier to explain VFA through BBA because it's a modified version of BBA. The variable fee approach of VFA, it can be applied to all direct participating contracts. In our case, this applies to our Swiss business, our German business, the U.K with-profits products that we sell. It will also apply to the unit-linked portfolios, which are treated as insurance, and they will also fall under the VFA model.
Similar to BBA, the key BOP drivers for the VFA business is also the CSM amortization and the release of the risk adjustment. The key difference to BBA is that any impact of economic variances, they do not impact the P&L directly. They are first absorbed by the CSM, and then they, you know, they affect the P&L only through CSM amortization in future periods. Operating variances and assumption updates. Here in cases where we have policyholder participation, like we have in Switzerland and Germany, also the actual versus expected experience on operating variances is absorbed by the CSM, and it impacts the P&L only through the CSM amortization. Also note that investment result under VFA model is also reflected as part of the CSM, and it flows back to the P&L through the CSM amortization.
This can be quite counterintuitive at first because, you know, you expect to see an investment result. A simple way to think about it is, under VFA, where it's all about direct participating contracts, the investment performance is also part of the core insurance service, and hence there is no separate investment result. This mechanism, it's also a driver for the equity impact of transition for the VFA portfolio. To give you an example, any capital gains from real estate portfolio, which we have in Switzerland and the IFRS 4 basis today, it's already part of the IFRS 4 equity. As part of transition to IFRS 17, a portion of it moves from equity into the transition CSM, and it's amortized then over time.
Between VFA and BBA, now you understand the P&L dynamics of more than 75% of the business. You know, remember the rest is PAA and IFRS 9. We've talked about the mechanics and dynamics. Now I'll just cover two slides on presentation. To the transparency features of IFRS 17, this one is something I really like, because it brings much more discipline and rigor on how new business contributes to both the value of the in-force portfolio as well as to ultimately to the profitability. Now, those of you who cover life insurance businesses, you would know that, you know, new business metrics doesn't really have much of a direct connection to the IFRS results.
That will change with IFRS 17, where it becomes clearly visible as to how much of the new business CSM, you know, or in other words, how much of the unearned profits, is added to the book in any given period. You can also see how much of it accretes back to the P&L. The chart you see on the right side, that's the CSM walk, and I think it will become a powerful tool for you. It shows whether the business is growing or is it shrinking, and that's by comparing CSM accretion versus dilution. It also provides you an insight on the year-on-year operating variances. It gives a flavor for the reliability of the underlying best estimate liability assumptions. The CSM walk is also a balance sheet view.
Now let's move to the BOP walk, which is the P&L view. This is the second feature in terms of better visibility of profit emergence, that you get under IFRS 17. The key driver of the profit is the margin release, and that's made up of CSM amortization, risk adjustment release, and that's for both the BBA and VFA portfolios. We will also add the technical result of the life short-term book, which is accounted for on the PAA basis into the margin release, view. The investment result, which is then relevant only for BBA and PAA. Remember, for VFA there is no separate investment result. The fee result will cover the performance of the pure investment contracts, which are accounted for on an IFRS 9 basis. The other result is mainly non-qualifying expenses.
For completeness, this illustrative graph, you know, it also gives you the actual versus expected experience and that can move in both directions. It also shows the impact of onerous contracts. The view, it's not directly available in the disclosures, but we will provide it to you as part of our supplements going forward. Now we've covered a lot of measurement and dynamics and presentation. A question which may now be on your mind is what does all this mean to the volatility of the results under IFRS 17? At this stage, it's quite difficult to give a one clear answer because there are elements of the new standard which point in both directions. Let me take a few examples.
If we take market impacts, they will become much more visible under the IFRS 17 insurance result, especially for the VFA portfolio, because they are reflected in the CSM and in the effect of the P&L through the CSM amortization. At the same time, you know, under IFRS 4, we used to have locked-in assumptions, so that was limiting volatility from the market impacts on the IFRS results. When you have operating variances or assumptions updates, they will continue to impact the P&L also in the future, but we expect that this impact can be much more subdued. That's because of the CSM mechanics, where it's absorbed by CSM first and then amortized to the P&L. Today, a lot of volatility in the life result, it also is a result of one-offs. You know, these can be positive one-offs or adverse one-offs.
We also expect in the future that will become less pronounced, because these one-offs gets absorbed by the CSM, and then it gets amortized over time. At this stage, based on dry runs which we have performed, we don't anticipate very significant changes to the volatility of the life results. Of course, we'll continue to manage the business and prioritize growth in areas which contributes to more stable earnings. That brings me to the end of my presentation, and let me conclude on life. We welcome the shift to IFRS 17 because it addresses a number of limitations in the current accounting rules. The new presentation, it will improve visibility, especially around new business contribution and profit emergence.
It should help to get better understanding of our life business and its performance drivers and increase the confidence when you're valuing the business. From a financial point of view, life equity at transition will be lower. It's mainly due to the recognition of future profits, the CSM as a liability, and earnings level of the life book will be largely unchanged. We do not expect any of this to have an impact on our strategy and our focus on protection and unit-linked will continue. With that, thank you, and I look forward to taking your questions. Over to you, George.
Thanks, Karthik. I hope that you found the introduction today useful. This is obviously the beginning of the transition rather than the end of it. Karthik, Jon, me, and the entire IR team are here to support you as we go through the transition together. From today's presentation, I hope you've got a better sense of how IFRS 17 changes some of the reported information. We'll continue to provide a detailed supplement, and if you've got a particular request about what you would like to see in there, please let us know. IFRS 17 might change the way we report performance, but it doesn't change the way we run the business. It doesn't change the way we generate value or returns for shareholders. With that, I think we're about ready to start the Q&A. Jon?
Thank you, George. Just as a reminder, if everyone could keep to two questions, that'd be much appreciated. If we have time at the end, there'll be an opportunity to ask follow-ups. Can we ask the first question, please?
We will now begin the question and answer session. Anyone who wishes to ask a question may press star and one on the touch-tone telephone. You will hear a tone to confirm that you have entered a queue. If you wish to remove yourself from the question queue, you may press star and two. Questioners on the phone are requested to only unmute while asking a question. The first question comes from Peter Eliot from Kepler Cheuvreux. Please go ahead, sir.
Thank you very much. Thank you for a very useful presentation indeed. My two questions. The first one, very simple, just wondering what information you will provide us with next year to enable us to compare 2023 with 2022 and prior. Second question, on the accounting choices you've made, George, you mentioned that you'd sort of spoken closely with Allianz, AXA, and Generali. I'm just wondering if you can share with us, you know, how comparable you expect these to be across, you know, your close peers and the industry, and maybe where you expect the main differences to be. Thank you very much.
Thank you, Peter. Maybe I'll do the second part, then I'll ask Karthik to say a few things about kind of what the disclosures might look like when we publish next year. I think it's been clear to us from a relatively early stage in the process that I mean, you can optimize these things individually to make your own results look the way you want them to look. Actually to improve the credibility across the industry, the aim has been to try and be as consistent as we possibly can. With the CFO Forum, and in particular with the companies that you mentioned, we've tried, I mean, as far as we can, to work together to increase the comparability. Now, what are the limits around that topic?
I mean, we've discussed the key KPIs, if that's what you can call them, to try and make sure that we present them the same way. We've tried to bring some of the definitions of things that aren't really IFRS 17 driven, but impact the way that the companies are viewed. Things like operating profit. You'll have seen in the presentation earlier, that we've changed our definition very slightly to make it more consistent with the peer group. I think the limitation on this thing really lies around some of those key judgments. I mean, Karthik was describing how we approach things like risk adjustment, how we approach best estimate reserving. I mean, those are things that we obviously can't share in advance.
I mean, we might have a sense of what people intend to do, but we just don't know. I think that's why both Karthik and I think that, I mean, there's bound to be some transition beyond the transition. I mean, hopefully that's not a significant component of what we do. I think there'll still be a period where businesses improve the alignment, improve the consistency. We have made a significant effort already to try and derive that. Karthik, on 2023 comparatives or the transition?
Hi, Peter. Just to give you a flavor of what's happening from a process-wise. We are currently running the quarterly closes under IFRS 17, but as you would expect, they always run with a lag, and it's going to be the same in 2023. We'll still be closing the year-end 2022 under the IFRS 4 basis, and publish results in February. Subsequent to that, there will also be year-end of IFRS 17, which happens internally. In the year-end 2022 IFRS 4 disclosures, you would see transitional disclosures, which will mainly cover the transition impacts or the opening balance sheet of IFRS 17 on a 1/1/2022 basis. The full year 2022, we currently plan to provide you certain supplements, which is more in April-May timeframe.
As George mentioned, Q1 statement will already be based on IFRS 17. The full comparatives of the half year results for 2023, that will be published in August, obviously next year, on an IFRS 17 basis with comparatives for 2022.
All right. Thank you very much.
The next question comes from Andrew Sinclair from Bank of America. Please go ahead.
Thank you, everyone. I agree, it was a really useful presentation. Thank you very much for that. Two from me, please. Firstly, it was just looking at slide 22, and just trying to understand some of the impact, some of the ability for some of the elements to kind of smooth profits from year to year. It looks from that slide like the net impact of discounting plus PYD looks slightly larger in IFRS 17 than PYD was in an IFRS 4 world. Is that the right interpretation, or otherwise? Or will you be separating out kind of PYD and discount impacts going forward? That's my first question. Second question was just on life. Really interesting just to understand the different calculation methods.
Again, it looks to me, if I look at slide 34, there's quite a lot being amalgamated together, in the margin release column. Are we gonna be able to see the different sources of earnings between protection, unit link, guaranteed savings, and the like, or is that still gonna be all amalgamated together? Thanks.
Andrew, maybe I'll start with a general comment on PYD, and then we'll try and touch on as much of the question. I think at this stage, I'm not sure if we can give you a complete picture of everything we intend to do on the breakdowns. Again, the aim would be that we're gonna try and make this as transparent as possible. Again, if you've got particular requests, please let us know. I can't promise that we can build an enormous supplement, but certainly if there's a number of topics that are of common interest, rather than answer questions frequently, it just makes sense for us to include it in the supplement. Generally on PYD, the concept remains the same.
There's probably two things that can drive it in directions that may be different from what we've had before. We've got risk adjustment. That adds an additional element to PYD that hadn't previously existed, and you would expect that to be beneficial to PYD. There'll be a discounting effect. The type of PYD is gonna impact, I mean, what you'll see in terms of the ultimate effects in the P&L overall from the P&C business. I mean, we've run various simulations, both looking at some of the historical information and based on what we would expect to see in the future. I think at this stage, we would still conclude that we would expect to see PYD in the range that you're familiar with from us.
That leaves some margin for uncertainties through the risk adjustment. Overall, we think that PYD continues to be in the range that you've become familiar with from us. Let me ask Karthik if he can help a bit on those two topics. Karthik?
Sure. On the granularity question, you know, I think the best way to think about it is the components which are parts of the disclosures. Like I mentioned, the income statement itself, you know, it doesn't differentiate between the accounting models or, you know, whether it's life or P&C. The more, the supporting tables, you know, which are notes to the financial statements, they would be much more based on the measurement model. You would see everything which has a CSM component, so that's pretty much life, BBA and VFA. They would have real roll forward tables. You would see how the risk adjustment and discounting and all of that is flowing through. In the supplements, currently, we are planning to provide you a breakdown, which is also based on the line of business.
You would see the split between protection, savings and unit-linked. That information should be available for you, going forward as well. That's a supplementary requirement rather than from the disclosure itself.
Super helpful. Thank you very much.
The next question comes from Andrew Ritchie from Autonomous. Please go ahead.
Oh, hi there. Yeah, echoing my previous colleagues, you definitely are leading the pack so far on all the IFRS 17 presentations I've sat through. First question is on the topic of onerousness-ness, if that's a word. I thought I'd see it mentioned a bit more than it was. I saw it in the slides. In particular, I would have expected on some long tail P&C business where you are trying to be ultra-conservative as far as you can be on the risk margin, that might have become onerous post the risk margin consideration and then be released over time. Maybe just clarify where are there onerous contracts and why, in particular on.
Is none of the P&C business falling within that after application of a conservative risk margin? Second question. You talked about leverage on slide 11, I think it is, where you're talking about using a leverage ratio, including CSM and risk adjustment. Is that, I mean, is that your view or is that broadly what your peers have agreed also, and or I'd I guess rating agencies have not sort of opined on it yet. I mean, in principle, over time, I mean, that could probably mean you could run with more debt than you used to under IFRS 4 because you're taking into account in principle more unearned profits. I mean, do you think in principle that those would support a higher level of of debt than what you used to to roll out? Thanks.
Thanks, Andrew. I'm gonna have a go at both of these. I think on the onerousness topic, we've tried to avoid turning business that isn't onerous into onerous business by virtue of the assumptions that we set that are the judgmental elements that we can apply as management. I think if you look at where we've done much more on risk adjustment, put probably more on the life side. I think when we did the initial calibrations, you could see some businesses that we knew were not onerous that popped up as onerous. To me, onerous is the granularity of risk adjustment. We made it more granular in some cases and actually avoid that outcome. Lost some point, which does drive volatile.
We have to try to move in by pushing. I think overall, I mean, we feel comfortable with where we are across. I have that much of the book finishing. I think on the leverage point. I mean, I think you're right. I mean, the whole thing is, I mean, I wouldn't say in flux. I would expect that most people would look at CSM and reach the same conclusion that we have around what it is, why it exists, and why it would be part of this calculation. However, I come to a different conclusion from you on the last point. Would this...
I mean, I guess theoretically, if you manage the business based on that leverage threshold based on the old basis, and you treated the new one as being identical, you would now have more room. In general, we look at a wide range of metrics. I mean, the more traditional measures of leverage haven't been things that we've placed too much weight upon. We look more at balance sheet structure. We certainly look at the rating agency requirements, and we would continue to make sure that we respect those. I think in theory, you're right about the mathematics, but I don't think it changes in any significant way, debt capacity.
Okay. Thank you.
The next question comes from Dominic O'Mahony from BNP Paribas. Please go ahead.
Oh, hi, folks. Really helpful presentation. Two, I think relatively detailed clarification questions. First is can you just help us understand the significance of using a locked-in discount rate in P&C? I get. As I understand it, you still recognize the impact of movements in interest rates. It's just that it's gonna go through OCI. Is the difference just that it goes through OCI rather than through the main face of the income statement? And then secondly, in terms of the impact of rates moving, as I understand it, but please correct me if I've got this wrong. Essentially, investment income won't really change if the risk-free rate moves.
What will actually happen is, for the same business, you'll actually see when interest rates go up, I think the combined ratio comes down, but future investment income doesn't, because of just the discounting effect, on the liabilities at inception. Is that right? Is that the right way to think about it? Thank you.
I'm gonna try and do the second one. I'm gonna get Karthik to help on the first one. I think on the second topic, investment income moves. Yep, they impact the new way the calculation of the combined ratio works because of the impact of discounting. I think the point, though, about investment income not changing. I mean, that's only true for the in-force. Of course, that does dominate the expected investment income, so you won't see a change. But when Karthik answers the first question, you'll see why it doesn't actually have any impact on the overall result. But for the new business that's coming on, if you write new business in the quarter that attracts a higher discount rate, that will be because interest rates have moved.
Given the application of ALM, you'll have an expectation that for that part of the book, there will be higher expectation of yield. Karthik, on the first one.
On the first one, as new business is written, you know, the locked-in interest rates. Interest rates are locked in in the current accident year. Of course, you know, it works on a quarterly basis. We normally take a year-to-date accident year for in the individual quarter and then true it up for the end of the year. What happens is that amount of locked-in discount is then, it's just locked in. Until the reserve runs off, we don't change the effects within the P&L of the accident year rate. IFRS 17 requires us always to compare that with the current rates at each reporting period. That's the difference which goes into the OCI.
Anything which is the current rate versus the original locked-in rate, and given that, you know, in certain P&C portfolios you could still have multi-year periods where the reserves still exist in the book. That difference is what goes through OCI.
Great. Thank you.
Thanks, Dom.
The next question comes from William Hawkins from KBW. Please go ahead.
Hi, gentlemen. As with Piers, thank you for a really helpful presentation. What percentage of the total change in shareholders' funds is likely to go through OCI rather than the P&L in what you would consider a normal year? George, you made the throwaway remark that maybe you're adding three points to your ROE, so we're now at 18%. But presumably there's gonna be a different change in the book value overall. I mean, is it positive or negative? Is it very positive or very negative? And that question then leads to the other one, which is, I mean, how are you gonna want us to be thinking about the level of volatility in equity? What is the level of volatility in equity likely to be?
'Cause I mean, one of my concerns is that you've, you know, quite optimized for stability in your earnings, but there's still potentially a huge amount of volatility that's gonna go through the OCI. If anything, from our point of view, it's gonna make life harder because it's volatility on both the asset side and the liability side. It's volatility that these days we may have to call economic volatility, even though you've hidden it from the P&L. Again, you know, given that we care about dividend adjusted book value growth, how should we be thinking about the level of volatility in the equity, even if it's not in the P&L?
I mean, I think the three points is a gauge to the impact on book value. I think you can. Even though we haven't given all the numbers today, I mean, that's a comparison to the ROE under IFRS 17 has been compared to an ROE under IFRS 4, where we already eliminated the impact of changes in gains. That three points is a good gauge to what was happening at underlying book value at the point of transition, which in this case is 1/1/2022. As I mentioned earlier, the actual reported book value for IFRS 4 has been moving much more through the course of the year because of interest rates, but that three points is unrelated to interest rates.
Level of volatility and equity. So I don't think the I mean, IFRS 17 clearly changes some of the volatility dynamics, especially if you've got a very large life book. If you've got most of that life book in VFA, then I mean, it does do quite a bit of smoothing. I think if you look at the types of risks that we prefer, I think the rationale that drives that risk preference holds through IFRS 17 because the things that we have less preference for, I think are the things that are still likely to generate the most significant volatility in equity. It's obviously closer to the more economic models, it's not quite the same.
The things we're doing around some parts of, for example, the back books, I think having less of those elements within the overall portfolio actually improves the transparency, reduces volatility over time and just makes the outcome more predictable. I mean, there will continue to be volatility in equity. I think for us, I mean, if you look at that, I mean, the traditional drivers that exist under IFRS 4, under IFRS 17, I mean, key issues are gonna be, I mean, as Karthik described in his presentation, especially around CSM, where we're making changes to assumptions that affect past claim topics rather than future claim topics.
I mean, those could be much more significant impacts than we've previously seen under the existing kind of locked in general approach to life accounting. I think our approach to volatility generally works for the various capital models. It doesn't work to quite the same degree for IFRS 17, but it still works for IFRS 17.
Thank you.
The next question comes from Ashik Musaddi from Morgan Stanley. Please go ahead.
Thank you. Perfect. Thanks, George. A very helpful presentation. So just a couple of questions on classification again. I guess this was the second last question from here. The discounting benefit that will go into the investment bucket, investment result bucket. If I understood correctly, you're saying that it will not go there, it will just go directly into OCI, and therefore the combined ratio that you'll report will just be on the first bucket of the margins, right? Is that understanding correct, or am I confused here? That's one question. And secondly, Karthik, I guess you mentioned that because insurance revenue will be used as a denominator of combined ratio, it will be higher than the current basis.
Is that because reinsurance result is not included or once the reinsurance result is included, probably will land up at the same level? These are two questions. Sorry, just for clarity purpose. Nothing else.
Yeah. No problem. I'll take the first one and ask Karthik to help on the second one. On the first one, if you're confused, I think at some point in this process, everyone gets a bit confused. I think the way to think of this is that the discounting component impacts the initial recognition of the amount of the claim reserve. It's definitely a benefit within the overall insurance part of the different groups of margin that Karthik referred to earlier. We put the discounting benefit into that combined ratio at the very beginning that we now calculate on an economic basis.
The unwind of that discount, so the build back up from the discounted level to the actual payment that we'll make to our policyholder at some point in the future, that's deemed to be a financing expense. That goes to offset the actual yield that we earn on the portfolio, and that's why Karthik referred to the investment part as now being a margin-based view. The discounting benefit impacts the insurance performance. On combined ratio, Karthik.
The combined ratio will be on a discounted risk-adjusted basis going forward. The point is the discount used for that is the locked-in discount rate. To answer your question on why is the insurance revenue higher, you're right. You see in this block where it's the reinsurance result, which is, you know, your ceded premiums net of recoveries, and the entire reinsurance result is now part of the numerator of the calculation of the combined ratio. If you think about it from a pure, you know, what's the baseline, the denominator, the insurance revenue, it is before the netting off of the reinsurance premiums or the ceded premiums.
That's why the insurance revenue is higher compared to current basis, because on the IFRS 4 combined ratio, we use the net earned premium, which already adjusts for reinsurance. Yeah, long way of saying you were right. Reinsurance results moving into the numerator is why insurance revenue is larger and on the denominator.
That's very clear. Thank you.
The next question comes from Will Hardcastle from UBS, please go ahead.
Hey. Afternoon. Thanks, everyone. Quick one on onerous contracts. Just trying to understand the level of granularity. You gave a good example there on life that you've added a bit more. I guess on the P&C side, are we talking sort of big picture line of business? Are we talking region or are we going line of business per region? If you see what I mean, to try and understand how big, if one moves into onerous contracts, how big a delta it could be. Something from a presentation this morning, which we were treated to, on liability reconciliations, we should get a bit more granularity on that apparently. Just trying to understand if that's a sort of a regional benefit that you'll be providing that from a disclosure-wise. Thanks.
Will, can I ask you to explain a bit more what the second point was, 'cause I didn't really understand what that was.
Yeah, sure. What we were told was we should be getting a sort of a claim reserve development, a liability, I guess, development including how the onerous contracts move year on year. Start year to end year, key moving parts behind that, and it's not just at sort of a group level, liability total. It should be more disclosure development, apparently.
Yeah. I'm gonna ask my expert to help discuss.
I can also start with the first question. Onerous contracts, they are tested at a unit of account level. Now what is unit of account? In our case, it would be the line of business, but at an entity level, in certain cases, or on an individual underwriting year basis, right? The onerous contracts are tested for every period at a cohort level, so you can see what is related to the current period and to the prior period. It's much lower than the region, but it's LOB at an underwriting year level.
On the liability reconciliation, I mean, if I understand your question correctly, for the P&C business where we apply the PAA model, there will be disclosures detailing out the liability for incurred claims, which in the future is your nominal reserves discounting this risk adjustment. It will also be presented both on a gross and on a ceded basis, probably that's the additional granularity which comes with IFRS 17. In case of onerous contracts, you know, we do provide separately a loss component, you know, where it is relevant, but that loss component is not even part of the liability for incurred claims. It will flow through the loss ratio and the combined ratio, but you would also get much more granularity on the loss component itself.
You think about liability for incurred claims where a roll forward is available, and then the loss component will be visible. Now, given we don't publish results on a lower level, so you would see that on a regional P&C level, of course.
That's great. Thank you.
The next question comes from Vinit Malhotra from Mediobanca. Please go ahead, sir.
Yes, good afternoon. Thank you, and again, excellent presentation. Thank you for the effort. Two questions from me. The first one is, you know, at the moment, at least I attach or some of us attach a lot of importance to underlying loss ratios or underlying combined ratio trends. Just, you know, on the slide 22, you mentioned the importance of risk adjustment to the current year or the current accident year numbers. Obviously, the breakout doesn't seem to show much. I mean, should we anticipate not a major change to these very important metrics as far as I look at the stock? So just any comments on underlying loss ratio, please. That's the first question.
Second question is just, George, we've talked about how the business is not changing from this quite an expensive project. What's the benefit of this exercise you see? I mean, one benefit could be that more granularity leads to more discipline in the group. Do you foresee that happening? Do you expect or hope that to happen? Any comment on the positive benefits of IFRS 17 for you would be helpful. Thank you.
Yeah. I'm gonna try to avoid the temptation to be cynical on the second question. I mean, I said in the earlier remarks that and Karthik and the team certainly feel this more than I do. We've been working on this for a long time, and if you remember, we've had a number of planned implementation dates that have slipped along the way. Not so long ago, the intention would have been that we'd be doing IFRS 17 already. Having said that, I think I mean, what benefits do we expect from this? I think I mean, Karthik laid out some of those in his presentation.
I think within the detail, it gives you more of a view of the business in the way that we would typically view it from an economic perspective. Of course it's that we use to determine what our preferences are. I think it does help align the external communication with some of the internal management thinking. I think in terms of the industry, I mean, you know as well as I do, I can't imagine that we have any significant peer that's not doing something that is like this in some way, shape, or form, and has actually been doing this for a very long time. I don't expect that to be a big driver of a change in behavior. It may be different in smaller parts of the markets.
Again, I think there is an element of discipline that it can help, but I certainly wouldn't overstate it. It has been an expensive project. We are obviously gonna be compliant, which is very important to us. I think, and probably the most important thing, I was telling a few heads, the CFO of one of our major business units, especially one of the European ones. We've used this to try and integrate more the different things we have to do, whether it's regulatory, internal economic capital models, and now the accounting. It just helps avoid too many conflicting signals. I think there is a benefit to this thing that it will feel quite intangible for a while.
I think just helps improve consistency across all the key metrics that we use. On the first part of your question, I think consistent with what I said to Andrew earlier, for the P&C business, we haven't put so much risk adjustment on, but we would transform something that we believe to be a good piece of business into something that is recognized at a very conservative level to produce profits in the future rather than profits now. I mean, it's important, but it doesn't change any of the decision-making, the risk adjustments. I think you need to see it in that context.
Okay. Thanks, George.
As a reminder, if you wish to register for a question, please press star and one. We have a follow-up question from Thomas Fossard from HSBC, please go ahead.
Oh, yes. Good afternoon, everyone. Two questions on my side. The first one will be related to your IFRS 9, 17 shareholders' equity. I'm still struggling to understand why it should be lower than the IFRS 4, since all the comments of your European peers have been highlighting that actually their shareholders' equity under IFRS 17 should be higher and actually should be not so dissimilar from Solvency II own funds. So I'm still struggling to understand why your shareholders' equity is down. Maybe you can bring a bit more granularity around that. The second question would be related to the reserve strengths, and because actually IFRS 17 is based on best estimates.
We know that actually you're holding reserves at the local entity level, and then you've got also on top, you've got group bulk reserves under the current regime. I want you to better understand where all these buffers went or how you have been able to recycle part of them and where this is going to show up in the balance sheet. Is it just capturing the risk adjustments, or does it show up somewhere else? Thank you.
Yeah. Thanks, Thomas. I mean, I think the first part of the question is quite difficult to answer because I think as I mentioned in my part of the presentation, you get very different answers this year, depending on which date you select. If you pick the start of the year, there's an impact on equity. If you pick the middle of the year, there's almost no impact on equity. If you pick the end of the year, it's entirely possible the IFRS 17 equity will be higher. I think that's why we've tried to, as we went through the presentation, to identify, I mean, what the drivers have been. For example, the P&C business, as you'd expect, there's more equity. And on the life side, because of what we've done on risk adjustment and on CSM, there's less equity.
It's hard to make a comparison to the peers because I haven't seen the peers yet. I don't know what dates they make the comments of. I mean, the presentation we've made tries to give you a sense of where we see it come from. On the reserve strength topic, IFRS 17 obviously requires a best estimate approach. I mean, the process doesn't quite work like this, but I'll use it as a simplification. I mean, essentially under the existing approach, we have things that I know you might classify as more prudent or more buffer like. We remove them, and we replace them with different components.
You see here, and as Karthik mentioned during his presentation, we have a new item, which is the allowance for limited historical experience, which is designed to reflect some of the things that were incorporated in some of the same elements that we had under IFRS 4. Definition is slightly different. Requirements are slightly different. You end up in the same place, maybe for slightly different reasons. But I mean, it's been our aim throughout, where possible and where consistent with the IFRS 17 requirements, to maintain a similar level of reserves. But even though it's not completely apart from the side, there has been some things were removed on IFRS 4 and they've been replaced by things under IFRS 17 that might be similar in quantum, but they're slightly different in definition.
The next question comes from Francisco Sebastian from Wellington Management. Please go ahead.
Thank you very much for this very useful presentation. My question is related to the interactions between IFRS 9 and 17 related to what others have asked before. In particular, as with IFRS 9, there is all these changes to the treatment of stocks, some mutual funds, expected credit losses that might introduce possibly some income or P&L volatility. Do you expect to absorb that or, you know, to just show that income statement volatility or maybe change your asset allocation? Or maybe is there something in IFRS 17 that would allow absorbing that income statement volatility coming from, again, primarily stocks, mutual funds, unexpected credit losses?
I mean, if you look at how we approach the business today, I mean, we already have volatility in results. We have impairments. We have realized gains and losses, which of course are determined by things other than market movements. So there's an unpredictable element to it already. But when it comes to making decisions about what we do with what we see as the underlying performance, what we do with the actual performance outcome, we tend to look through that. I don't see that changing in future. I mean, hence the commentary around the lack of change to a number of the key policies that we apply to the outcomes.
I don't think we're overly concerned about this. It will be transparent, so you'll be able to see it. You'll have an external market reference point, so it should make sense when it happens. It can't happen because suddenly we happen to realize a gain or a loss on something that would not have been entirely visible to you. It definitely introduces more volatility into the P&L overall. But I think given just a general approach that we're looking for what we believe to be the underlying performance and using the strength of the balance sheet to deal with the remainder, there's nothing in this change that would alter that philosophy.
The next question is a follow-up from Andrew Sinclair from Bank of America. Please go ahead.
Thank you. A couple more from me. First one was just on tax. Can you give us any color on what tax impacts you expect as book values get marked down, historic accounting profits backed out? Will you get a deferred tax asset and how can that be utilized? Secondly, just on kind of the loss of net written premium as a metric, how do you propose we track changes in retention and reinsurance programs evolving over time? That's me. Thanks.
Yeah. Thanks, Andy. The first one's an interesting one, because not every jurisdiction has entirely laid out the rules at this point. Based on what we've seen, you know the predominant way of where our earnings comes from. Of course, the taxation rules there are not gonna change, and the accounting is not gonna change there. Again, we're in the fortunate position that for a very large part of our book, none of this changes things, whether it's on the underlying economics or even on the tax topic. Now, having said that though, we do have other jurisdictions that haven't concluded and haven't made clear what the transition impacts are gonna be. We don't expect to be significantly disadvantaged or significantly advantaged in that process.
At this point, we still expect to see the taxation levels come out consistent with our current guidance, based on what we know at the moment. There's still some risk because not every jurisdiction has completely confirmed how it's gonna work. As we learn more on that, we'll update you. For the time being, no change to gains around the tax charge. On the net premium topic, and as someone who has personally struggled with the new combined ratio for what seems like years already, I still look at the thing and it. I mean, I get, I guess the part of the problem of experience, I'm very experienced in one particular measure, and my brain finds it hard to find the plasticity to cope with a new one.
Having said that, I think we'll adapt over time. I think on the business driver metrics, I think actually the growth topic is actually a better way to look at the issue that you just mentioned. I mean, Karthik, in his presentation, described two changes that we're gonna have. I mean, our premium numbers will move to be net of single parent captives. That reduces a distortion that's been a bit of a challenge for us in the past anyway. Then I think when you separate the reinsurance impact out from what's happening on the primary side with the business that's coming into view, I mean, that's how we do it already, to be honest. Actually, I think once.
Certainly I get used to this, we all get used to this. The separation is actually quite helpful. It creates another bit of volatility in the components of the result. But actually looking at gross incoming movements from clients is a better way to judge retention and other business drivers than maybe the net is at times.
Gotcha. Thank you very much.
Once again, to ask a question, please press star and one. Also, the next one is a follow-up from Peter Eliot from Kepler Cheuvreux. Please go ahead.
Thank you very much. It's just a follow-up on Nitin's question actually on the sort of cost benefit, because you discussed the benefits. I'm just wondering as well how the costs will develop from here. I mean, obviously there's been a huge cost to date, and I'm just wondering what the sort of annual cost of maintaining it, what you might expect them to be going forward. Secondly, given the comments that you feel very well prepared for this, I'm just wondering, could it possibly have been introduced earlier or more efficiently or, you know, do you think it's appropriate to have taken the time that we have? Thank you.
Thanks, Peter. I think on the first one, the annual cost of maintenance, I'm not gonna let Karthik answer that question. I mean, as I mentioned in my part of the presentation, we've tried to do a number of things to improve the efficiencies. It's certainly a more complex basis. I think there are elements if you look at things in the longer term, and you think about the skills that are required, whether you're an actuary or in the FAR financial accounting world or treasury or in our planning and performance teams, all of these skills start to align more maybe than they have in the past.
I think it creates interesting possibilities for the future that we haven't yet been able to think about because we've had to get the thing finished. I think, at the margin, I don't expect it to be cheaper to run this than it was in the past. Given the targets that my boss has given me, I don't expect it to be significantly more expensive either. On the second question, I think you might have to ask the IASB. I think they would have, they'll have views, I'm sure, on the process.
I mean, I guess whether you ask the industry, the preparers, the standard setters, the users, I'm sure it could have been done quicker, but it would have required all of us to have reached a conclusion just quicker in the end and more rapidly. We didn't. For various reasons, one or more of the stakeholder groups wanted it to take more time in favor of an outcome that they preferred. Therefore we are where we are. We could have saved some money if it had been quicker, that's for sure. I mean, I'm not big on kind of could have, should have, would have. We got it done.
I think we're ready for what comes next year, whereas, as you said, I know you're repeating what we said, but we feel we're well prepared. We've invested in this thing to make sure we do it properly. Hopefully that will show in what we do next year.
All right. Thank you very much. Sorry, unfair question, the second one.
Ladies and gentlemen, that was the last question. I would now like to turn the conference back over to Jon Hocking for any closing remarks.
Thank you very much for everybody for participating and your questions. If anyone's got any follow-up questions, myself or the rest of the IR team will be available. Thank you very much.
Ladies and gentlemen, the conference is now over. Thank you for choosing Chorus Call, and thank you for participating in the conference. You may now disconnect your lines. Goodbye.