AIA Group Limited (1299) Earnings Call Transcript & Summary

Good evening from AIA Central in Hong Kong, and welcome to AIA's IFRS 17 update question-and-answer session. I'm Lance Burbidge, Chief Investor Relations Officer. Together with me today we have Garth Jones, our group CFO; Sam Morgan, our Head of Integrated Reporting and other members of our finance team. AIA adopted IFRS 7 and IFRS 9 and IFRS 17 from the 1st of January 2023. And our upcoming interim results will be published on the new accounting standards. For comparative purposes, today, we have published our consolidated financial information on both IFRS 9 and IFRS 17 for the first half and the full year of 2022. We also released a slide pack and a video presentation to help understand the new disclosures, and I hope you found that useful. Also available on the Investor Relations website are a transcript of the presentation, a financial supplement spreadsheet and a reference guide that maps the key figures shown in the presentation to the detailed line items in the consolidated financial statements. Before we start the Q&A, Garth will make some opening remarks.

Thank you, Lance, and good evening. Firstly, it's important to note that IFRS 17 is an accounting change and does not impact the underlying economics of our business nor the way which we operate. There's no change to our capital management framework or our dividend policy. As we indicated at the full year in March, the adoption of IFRS 17 had a positive impact on AIA's 2022 results compared with those reported under IFRS 4. OPAT was up 1% to $6.4 billion, while net profit was $3 billion higher and will be significantly less volatile going forward as the new accounting treatment is better aligned with the underlying economics of the business. Similarly, shareholders' allocated equity and shareholders' equity are higher under the new basis at the end of 2022. Our leverage ratio improved to 11.4% based on an updated definition under IFRS 17, further reinforcing AIA's strong credit and financial strength ratings. OPAT remains a key measure of our underlying earnings as it continues to remove the short-term volatility from market movements. There are new enhanced disclosures breaking out the different components of OPAT providing greater visibility on the sources of our earnings. Our earnings remain at very high quality and arise predominantly from insurance contract services, which in aggregate contributed 74% of the total in 2022. Profitable new business is the core driver of our OPAT growth. As we add successive layers of new business, this delivers sustained growth in OPAT and shareholders' allocated equity over time. While OPAT represents the accounting view of earnings from our business, our embedded value is based on a prudent valuation of distributable cash flows. The comparison and the equivalent IFRS metrics with our EV equity and VONB demonstrates the prudence of our embedded value reporting. Therefore, VONB, EV and free surplus remain the key measures of shareholder value creation. -- with the adoption of IFRS 17, there's no change to our proven growth strategy as we continue to capture the immense opportunities for life and health insurance in Asia and deliver sustainable returns for our shareholders.

Thank you, Garth. We can now begin the Q&A session. So Kelly, over to you.

Ladies and gentlemen, we will now begin our Q&A session. [Operator Instructions] Now let's proceed. And our first question comes from Charles Zhou, Credit Suisse.

I have two questions. First, when I read some of the previously recent materials, it also talk about the CSM, contractual service margin, and also value of new business margin. So can you maybe just give us more like a teach-in about the CSS margin? And also my question are the trends of the CSS margin and also value of new business margin positively correlated? And what other factors that should we consider when forecasting the CSS margin? So this is my first question. The second one is also related to the CSM. So is the pace of the CSM and also -- or the risk adjustment release generally stable across periods for AIA? And what factors may affect the sales -- sorry, the pace of the release?

Yes. Thanks, Charles. I think the key thing to look at is the slide we showed, which shows the proportion of CSM, the new business CSM to VONB, Slide 17. You'll see that's 2x net of reinsurance. And that is broadly stable. It does vary by product and geography, but it's broadly stable overall. And it reflects a different basis of calculation effectively, but they are -- that ratio is broadly stable in the near term. In terms of the pace of CSM release, that's also very stable. We've shown that, that's 9.3%. And again, we wouldn't expect that to change very much from period to period.

The next question comes from Thomas Wang of Goldman Sachs.

I just want to focus on this EV and the CSM sort of comparison. On the 17, you highlighted the CSM to VONB looks about 2x. Can I just check how should we think about this 2x ratio? What's going to drive that movement? Because is it NBV margin? Because I'm thinking is that NBV margin is higher. It's also means some higher. So that ratio is going to be stable? Or is it something product mix or positive ratio that's going to drive this 2x multiple? The other -- so the second part of that question would be, if I look at VIF versus the overall CSM balance, which is I think was pretax, the growth number is $53 billion or so. VIF, last year, it was about $40 billion. So that multiple is more like 1.3x. How should we think about that 1.3x versus this 2x multiple for the new business comparison?

Maybe I'll start on the CSM, the new business CSM to the VONB. As I said earlier, yes, we expect that certainly in the near term to stay fairly stable. One thing, I guess, which I'll add to that is that certainly our -- the China -- Mainland China business and the Hong Kong business are broadly around that average in terms of the 2x multiple. So geographic mix and product mix will change potentially that ratio, but I think they broadly move in tandem with each other. Certainly, the same factors behind the new business CSM and the VOM applied because it's calculated that inception.

Yes. And Sam, maybe you could cover the 1.3x point.

Yes. Thanks, Garth. For the comparison between the IFRS and the EV, it's actually better rather than considering just the VIF and the CSM. You're better off comparing the total, so the total IFRS 17 comprehensive equity to EV equity. The split of the comprehensive equity between the shareholders' equity and the net CSM is really about the timing of the total value. So really looking at those two independently can be quite difficult to compare, whereas if you compare the two of them together, so the comprehensive equity and the EV equity that is a ratio which is more sensible.

The next question comes from Kailesh Mistry of HSBC.

I've got a few questions here. First one is on discount rates and illiquidity premium. Just -- Will these be updated every interim and full year results? And secondly, related to that, is the risk free rate simply the government bond yield at this specific maturity stated? Or is it something else? And on the illiquidity premium, is there a basis point cap on the illiquidity premium? And as a result of that, does it move formulaically from one period to the next? Or is it purely based on management discretion? And what are main factors that may explain the differences in the scale of this illiquidity premium that you've received? And then I've got some more, but maybe if we could start with that one.

Okay. Thanks, Kailesh. Yes, the base rates, we'll update these on an ongoing basis. And obviously, we'll publish them on an ongoing basis. And so they are based on the government bond rates. The risk-free rate is based on the government bond rates and the curve. And then we have an illiquidity premium that's based on observable spreads netted off of our expected defaults and migration and is based on a reference bond portfolio, which is aligned with the standard. So there's no cap.

Okay. So there's no cap on that. That's fine. And so will that illiquidity premium change every half year like the bond yield as well, so they'll be updated concurrently?

That's right. It will move with the observable spreads and so on.

Yes. Okay. And then do you take a proportion of the observers spread? Or is it -- is that's done on a case-by-case basis based on the reference portfolio?

Yes. It's based on the reference portfolio, the expected defaults in the migration in the default portfolio.

Sure. Okay. And then just going on to the sensitivity disclosure. Again, just a few points to clarify really. On the CSM sensitivity for interest rate changes, am I correct in thinking this is a parallel shift and no change in the illiquidity premium effectively for the shareholder share of asset values and the discount rate? And then on the PBT sensitivity, will this basically come through the short-term fluctuations line? Or is some of it through OPAT as well? And I'm not sure I could find a sensitivity for changing interest rates on the OCI. If I'm missing something, grateful if you could just point out where that is.

Yes. All the sensitivities in the CSM, that is a shift across. It's a step-up as it were, parallel shift. So that's clear. I think one thing to remember with the sensitivities is this is the sensitivity of the CSM. And then that will obviously be released at a rate, as you said, it was 9.3%. So that means the sensitivities are very small. And there's no change in the illiquidity premium. That would be back to the reference portfolio.

I guess just add at that point as well, if you look at the variances in the roll forward of the CSM in 2022, you'll see that they correlate pretty closely in terms of the sensitivities. So the big driver of the negative variance in 2022 was the increase in interest rates. What was the question on the profit before tax again, Kailesh?

Yes. Yes. Just on that sensitivity will -- when we model that sensitivity, should we assume it all comes through short-term fluctuations? Or does some of it go through OPAT as well?

It doesn't go through OPAT.

No, no. Most of the -- the vast majority will go through short-term fluctuations, correct.

Okay. Fine. And then the other one was, is there any sensitivity on the impact on OCI, so both the discount rate and the asset value because it wasn't on the slides, and I may have missed it when I scrolled through the 150-odd pages of other disclosure.

You've had so many hours, Kailesh.

If I had -- 20 minutes, it would have been okay.

Yes, there isn't an interest rate sensitivity of the OCI. Maybe in part because there's an OCI on the liabilities as well on the assets, right, interest rate movements and...

Okay. So it's in the disclosure somewhere.

No.

Yes. You don't really need one. That's the point.

Because it's so small. Okay. Fine. And then on the Thailand negative spread write-off, this is obviously the key driver of the OPAT. But mechanically, how does it come through? Should we just think about effectively the cost of liabilities in the earnings has fallen because you've written off that negative spread. Is that the way conceptually just to think about it?

Yes. I think the way to think about it is that previously, that would have come through every year, that negative spread and be a reduction to the OPAT effectively. And now that's been capitalized into the equity effectively. So the equity has been reduced accordingly as we provided for it. And then every year, instead of that coming through OPAT. What will happen is that the that liability that provision will unwind and be released and that will meet the negative spread every -- each year.

Okay. Perfect. And yes, I think that's it. There are more, but don't worry. I've taken...

The next question comes from Leon Qi of Daiwa Securities.

This is Leon Qi from Daiwa. My first question is on the IFRS net profit under IFRS 17. On your mark-to-market of liabilities, I understand that now the market-to-market movements for our business is absorbed directly within CSM. But for non-par, it's still -- the non-par part still hits the net profit. We do notice that for some of your peers in China, they actually choose the so-called, the OCI option for the mark-to-market part of their liabilities, basically both par and non-par, the mark-to-market goes through OCI rather than hitting the net profit. I'm just wondering what factors are we considering when we only basically only use the OCI option or the par business rather than the non-par business? That's my first question. And secondly, if I may follow up a little bit on your CSM release rate. Is my understanding correct that the CSM release rate will be a little bit higher for Mainland China business because Mainland China's OPAT's contribution is a bit higher under IFRS 17? Basically, the IFRS 17 OPAT is revised up a bit more proportionately than your other regions. Just wondering if that implication is correct. And thirdly, well, I just wanted to make sure that actually our product strategy does not change after all these accounting changes because we keep saying the underlying economics doesn't change despite the fact that actually the profitable release of our different products are a little bit different under IFRS 17.

Yes, three different questions. Ain't that perhaps, Sam.

On the first?

First one, yes.

Okay. On the first one. So on the first one, the question you asked was, where does the interest rate movements on the non-par go? I think you said that it went through the net profit. In fact, it does go through OCI. So we are aligned to what you were saying the Chinese insurers do. And -- yes. And yes, we've done that effectively to remove the volatility in our net profit and to give something which is smoother over time.

Okay. Okay, good to know. Thanks for the clarity.

But obviously, the change in equities and real estate, that does go through net profit and we strip that out in the short-term fluctuations, as you'll see in our OPAT to net profit reconciliation.

Of course, I understand. yes. Okay. Yes.

I'll do the CSM release on China. I mean I think the CSM release just generally varies by product type and protection products are generally -- have a slightly faster release than savings products. I mean that's a kind of general theme that we see across our business. I think you've kind of conflated that increased OPAT in the China business, is actually -- that's the move from IFRS 4 to IFRS 17. So it's not really there. It's the release. The reason it's increased is that the shift in China, we've adopted a group standard in the way that we're recognizing profits from different products. And actually, we were, I guess, under reporting China earnings. And so it's become a bit faster. China -- AIA China relative to the 9.3% is a little bit higher, but not far away. And then the last question was on product strategy.

Yes. And the product strategy doesn't change. We're very clear. It's protection long-term savings. Yes.

That okay? Do we get all of those, Leon?

Yes. That's very helpful, very clear.

The next question comes from Dexter Hsu of Macquarie.

My first question is regarding the short contract boundaries imposed on the retail -- insurance business in Australia. Can you explain more about this? And the second question is actually, the next year, I mean the comprehensive equity. Would this be countercyclical in the future? Or how should we look at these things?

Yes. Thanks, Dexter. Just on the Australian business, under IFRS 17, the reinsurance on the renewable term business, it's long-term renewable term business. The reinsurance on that is treated as a long-term business. And therefore, the cost of that comes through into equity and so on whereas the profit side of the equation is treated as a short-term contract. And so we only have 1 year's profit that comes through. And so there's a mismatch between the two. And the impact of that is what we see in the numbers there on the equity and the OPAT. In IFRS 4, those were both treated as long-term contracts. So obviously, you didn't see that. In terms of the comprehensive equity, we've shown that at 100%. What we understand that the rating agencies are likely to take some or all of the comprehensive equity into account. That is 100% or 0 of the CSM. Some of the CSM they will take into account. We're not sure exactly how much. Some have said informally that they're trending towards, including all of it. But the final basis hasn't been decided upon. And so we've shown it at 100%. We've been very transparent and very clear. And then we'll see how the rating agencies finally come out.

The next question comes from Edwin Liu of CLSA.

My first question is to follow up with Kailesh's question on illiquidity premium. So thank you for sharing all the statistics. Just very quickly, could you also share some color on ultimate forward rate if available by region? Second question is on risk adjustment. I noted risk adjustment seems to be quite small for you. I'm not sure it's a good ring or bad thing. Your confidence level is 75%, which is slightly lower but should we still interpret your small [indiscernible] as a good thing because that indicate your risk profile is better than peer? My last question is on change of measurement KPI. Would there be any change after the IFRS 17?

We haven't disclosed the ultimate forward rates. But -- so what we've given is what we are disclosing.

And the best estimate liability is based on, obviously, it's 50% probability. I think it's a distribution, so implicitly allows for TVOGs and then 75% is chosen to add some prudence in that. So -- and it's comparable to with others in the market. The KPIs won't change. We see as the KPIs are still OPAT, VONB and so on.

The next question comes from Jian Li of Huatai Research.

I just have one question about the CSM between the VFA approach based reserve and the CSM based reserves. I noticed that VFA-based reserve is around 40% of the total CSM. But in the earnings sources, I see the number from par business is only 16%, while the profit from non-par and the fee-based profit is 58%. So it's kind of -- the difference is big, why the par business contributed 40% and the CSM only contributed 16% of the earnings?

Yes, I think I'll take that one. So obviously, the -- as you said, the VFA represents about 40% of the CSM balance. But I think as Lance mentioned previously, the release rate on the par is parent savings is generally a little bit slower than the release rate on the protection. And so what you'll find is that 40% because of the slower release rate ends up being a smaller percentage of the total of the source of earnings pay. So that's why par is smaller than the rest.

The next question comes from Charles Zhou of Credit Suisse.

Okay. Great. So I have two questions -- two more questions. One is on the leverage ratio on page -- I think, Page 16. We noticed that your leverage ratio just declined from 22% to 11%. I think that's mainly because of the different formula here. So you also mentioned the inclusion by the rating agency for leverage ratio calculation. So I may ask you, is this something that you use when issue with the debt? And also for the new formula, so is this only for international or Asian insurers or the Chinese insurer? Because I just want to get more colors like for -- to understand this 11.4%, is this something that will be used by everyone or not? So that's my first question, whether this will be recognized by all the rating agencies or can you maybe give me some color how this leverage ratio will be used? And my second question is also on Page 17, 1-7. On the title, you mentioned reinforced the prudence in AIA's embedded value. I also recall that you talked about a comprehensive equity versus the EV equity and EV equity is like smaller than the comprehensive equity. You also mentioned 2x of the new business CSM compares with VONB. So can we maybe say if this ratio is higher, say, comprehensive equity, divided by EV and so that means more prudent? Or in other words, for other companies, if the ratio is lower, it means maybe less prudent. Can we -- so can we come to this conclusion?

Yes, thanks. On the leverage ratio, we expect that as IFRS 17 will be published globally that the rating agencies will start to look at comprehensive equity. They may well look at the composition of the comprehensive equity between the [indiscernible] and the CSM. But we believe that they will include at least some of the CSM in the calculation of leverage ratio going forward. Some of them did that in the past when they looked at the expected future profits. But I think it's more explicit now that we have the CSM. And exactly how that's going to be done, the rating agencies are still assessing. But we would expect that this will become a standard way of looking at things going forward. And then the rating agencies will come up with their models and there will be an input into their overall credit rating.

And I think on the comprehensive equity to EV equity, I mean, I won't speak for any other company. But certainly, the ratio that we show, shows us that ours is certainly prudent and we've given the reasons as to why they're different as well and why we believe and continue to believe the EV, VONB and free surplus are the best measures of shareholder value rather than comprehensive equity, but we understand that people will look at this number.

The next question comes from Shengbo Tang of Nomura.

Can you hear me, sorry?

Yes, we can, Shengbo.

So on Page 11, regarding the 16% par in the earnings resource breakdown, is that sensitive to investment return for the par? And what's the difference from the 11% spread. My understanding is that the spread is sort of the gap between the investment return and the policy dividend. Can you give me more color on the 16% par.

Yes. So for our power business, our power business all goes through the VFA measurement model. And as we, I think, have previously mentioned, VFA, it essentially will all come through the insurance service results. So all of the economic movements, both interest rates and equity movements will flow through into the CSM, essentially, the asset values will change. We will then offset that by the liability values, and it's the difference between those two, which will then flow into the CSM and then that change in the CSM will then amortize through over time into our profitability. So yes, it is sensitive. But if you look at the sensitivities at the back, the other sensitivities, you'll see on Page 34, they show that we -- although those sensitivities are of a certain size, they will amortize. And so we're not particularly sensitive from an OPAT perspective on the par business. When you look at the 11% spread, the 11% spread, as we mentioned on Slide 10, that 11% spread is fully associated with the GMM business. So this is our -- essentially our non-par business, and it's really the excess of the expected return and our surplus assets. So it's the expected return which the returns on our non-par and surplus assets over the unwind of the liabilities of our GMM liabilities.

I think I might just add on to the first question as well. So if you go back to the waterfall of the CSM movement in 2022 on Slide 9, where you see the 3.9% negative billion of variances, the majority of that is coming through on the CSM on the par business essentially. So as Sam says, although you have -- the investment return is coming through there, then it will come through OPAT effectively at the CSM release rate. So it gets amortized over a long time. So that may be an easier way to -- or another way to think about it as well. I think that's exhausted all the questions that we have. As Kerry said, there's quite a lot of information to look at. So please, if you do have further questions, obviously, come through to the Investor Relations team who are available whenever you need. So thank you very much, everybody, and good evening.

Thank you.