AXA S.A.

Andrew Wallace Barnett - IR Thomas Buberl - Group CEO Gerald Harlin - Group CFO Gaëlle Olivier - CEO, AXA Global P&C Paul Evans - Group Chief Executive, AXA Global Life, Savings and Health and CEO, AXA Global Asset Management Bertrand Poupart-Lafarge - Deputy CFO, AXA U.K. Amanda Blanc - CEO, AXA U.K.

and Ireland

Farooq Hanif - Credit Suisse Jon Hocking - Morgan Stanley Nick Holmes - Society General Peter Eliot - Kepler Cheuvreux James Shuck - Citi Andy Hughes - Macquarie Johnny Vo - Goldman Sachs Andrew Crean - Autonomous Niccolo Dalla-Palma - Exane BNP Paribas Paul De'Ath - RBC Thomas Fossard - HSBC

Well. Good afternoon, everybody.

Welcome to AXA's 2017 Half Year Earnings Presentation. Welcome to those of you here in the room with us, welcome to those of you on the telephone, and welcome to those of you on the webcast.

[Operator Instructions] And of course, we will give preference as per usual to questions from here in the room. Here in the room we have with us Thomas Buberl, Gerald Harlin, Gaëlle Olivier, and Paul Evans.

We also have Amanda Blanc, the CEO of the U.K., in the room with us as well; who'll be happy to answer any U.K. related questions.

And without any further ado, it is my pleasure to hand over to Thomas for an introduction.

Thank you, Andrew, and afternoon everybody. It's a great pleasure to be here with you and to present the half year results 2017 to you and I'm looking even more forward to your questions.

If we go and start with the overall, we can look back on a very strong first half year, with a underlying earnings per share growth of 5%. This is in the middle of the range that we have communicated last year for the Ambition 2020 between 3% and 7%.

The adjusted earnings per share are at 4%. This is being carried by a strong improvement in all lines of business.

The Life & Savings business has improved by 3% in underlying earnings and I'm particularly proud that the P&C business has improved by 6% in the first half year. What is also very significant is the asset management improvement by 10%.

In Ambition 2020, we are not focusing on the overall volume growth. We are focusing on growth in specific segments.

Those segments are segments where we are closer to our end customer, where we have a natural end customer contact, which is the Commercial P&C business and the Health & Protection business. We have also followed that logic in the first half of 2017, with a plus 2% in Commercial Lines business and a focus of that in the non-motor business because we believe that the future is there for us and that is also where we have the most end customer contact.

The same is true for the Health business, where we have increased our growth from last year of 4% to this year of 6%. We further said that we want to improve the business mix on the Life side.

Today, we're one of the very few companies who have drastically changed their business mix in Life to protection, to Unit-Linked, and mutual funds, and I'm very happy to see that the APE in Unit-Linked has increased by 11%, which is driven by our two strongest markets in that area, France and the U.S. This, at the same time, where the NBV margin has increased again to 40% and also this is in comparison to our competitors, one of the highest levels.

We also look very much onto Asia because we believe that Asia is a very important region for us. We've got a great presence in Asia.

About 25% of our business is in Asia today. And I'm very happy to see that we have also been able to increase our NBV in Asia by 9%, with a very strong focus in the first half year on making sure that the business in China, which didn't use to be as profitable as we wanted it to be, is now at an NBV margin of 25% versus last year of 10%.

When we go into the technical margin, which was the second pillar of Ambition 2020. As I said earlier, I'm very, very proud of the fact that we have made such progress in the Property & Casualty.

And if you look at this 1.5% reduction in combined ratio this is a great achievement. Yes, you can say that we had less nat cat, but if you look into the numbers, the natural events have been higher than last year.

So we are talking about a 1.2% true reduction in the combined ratio, which is linked to the fundamentals of really improving the loss ratio and the expense ratio. And with that combined ratio on the current year basis, we are, from what I've seen, extremely well-positioned looking at the competition, but also extremely well-positioned when I look at the track record of continuously reducing debt.

This 96.1% all-year combined ratio is well in line with the ambition 2020 of 94% to 95%. We slowly want to move ourselves there, being very prudent on our reserves.

You have seen that we have reduced the prior years, this year relative to last year and have actually a comparatively, relatively low level to our competitors, because we do believe this business has to be steered of the fundamentals of the current year and not the reserves. If we go into the efficiency part, which is the third biggest bucket of Ambition 2020, we also see that we have made progress there.

Our total aim of cost reduction and avoidance of cost increase is EUR2.1 billion in the Ambition 2020. We have achieved EUR300 million last year.

We are at half year at EUR200 million. All entities are now in the process of really reducing their costs, and everybody has got their fair share of how this needs to happen.

We have seen -- you have seen that in all major countries, where a social dialogue is required to reduce the cost, we have also started that social dialogue. And in some countries, we are already in implementation after a successful close of that social dialogue, if you look at France, if you look at Belgium, if you look at Germany.

So, despite the fact that in Europe, reducing cost does take a little bit time at the beginning because we want to continue the active social dialogue, the things are in rolling or, as the French are saying, [Foreign Language]. I'm therefore very confident that continuing these efforts and with a disciplined approach, really working on all our three levers that we want to address, the administrative expenses, the claims expenses, but also the acquisition expenses, that we will reach our Ambition 2020 target.

If we look at the capital management, we can clearly say that all of these results have been you realized based on a very strong balance sheet. 201% Solvency II ratio is a great result.

It's even an increase compared to last year. And we have clearly signaled to the market that with the intention to lift a minority stake of our U.S.

business on the Life side and the Asset Management side, that we really want to accelerate the portfolio shift more towards a P&C and more towards a Health & Protection profile. With having said this, we are obviously continuously screening the market for M&A opportunities, where can we really both grow inorganically in the Health & Protection and P&C Commercial business, but also how can we deploy investments internally to really make that growth.

And you have seen earlier that the organic way of growing is quite a successful way. We have managed to grow in the Commercial and P&C business both in terms of putting through the pricing power that we have and growing in terms of number of customers and the same is true for the Health business with 6%.

AXA is also a very large insurance company worldwide and as we not only work in the payment of claims, but we have a very clear societal role both from an insurer perspective but also from an investor perspective. We've been continuing our acting really in the sense of a good corporate citizen and being socially responsible.

We have worked on two areas, where we have taken significant investment or divestment decision in the last two years. One was on the core front, where we have decided to divest from companies that are mostly exposed to the core sector.

We've gone a step further and have joined the so-called RE 00, initiative where we basically with other large companies, commit to only use sustainable electricity. We've also continued on the divestment part on tobacco after having announced of getting out of the tobacco industry.

We've also gone a step further and have cosponsored a very large declaration, where many, many investors have followed our initial goal of divesting from the coal sector. When you look at these results in the light of Ambition 2020, I'm very happy to tell you that we are on track when it comes to all the key KPIs.

We are determined to have an underlying earnings per share growth between 3% and 7% over the period from 2016 to 2020. Last year we had 4%.

This year we have 5%. The same is true for the free cash flow's on accumulative basis, EUR28 billion to EUR32 billion.

We are today at a similar level as we were last year when we look forward towards the end of the year. So, we can also -- so we were also stand by this commitment.

When we look at the adjusted return on equity, we said we want to be between 12% and 14%. And I hope you don't mind that we have surpassed the upper boundary of 14% in the half year with 14.7% return on equity.

And as I said earlier, Solvency II ratio, we want to be in a range between 170% and 230%. With 201%, we are almost in the pure middle of this range.

A solid set of numbers, a good performance, which shows that AXA is clearly on track to deliver Ambition 2020. And I would now like to hand over to Gerald, who goes into more detail.

Good afternoon. Okay.

So, let's go to more detail on the underlying earnings by segment. We could say that in our main segments across the Board, we are significantly up, with Life & Saving, plus 3%; Property & Casualty, plus 6%; Asset Management, plus 10%.

The Banking is very small, minus 13%. It's due to lower margin and deposit in Belgium.

And as far as the Holdings are concerned, two reasons for the minus 15% decline. The first one is higher fine charges due to the issuance of Tier 2 debts as a refinancing of our Tier 1 debt.

As you know, Tier 1 debt doesn't flow to the P&L account under the IFRS principles. That's the reason why.

The second reason is the non-repeat of tax -- of tax one-off in 2016. On the adjusted earnings, so we have -- we are plus 3%.

You can notice that we have realized capital gains, that's EUR307 million. It's exclusively equity and real estate and you can notice as well that net impairments are much lower.

We benefit from a better financial market, with EUR3.5 billion for adjusted earnings. Net income now, you can -- we have an increase of 1% on net income at EUR3.268 billion.

You can notice that the gains on economic hedges are minus EUR281 million. This corresponds to free-standing derivatives not benefited on the hedge accounting and the EUR127 million, it corresponds to the favorable impact of markets on IFRS P&L funds, which are mark-to-market, so plus 1%.

Let's move to the Life & Savings and let's start with the topline. And let's -- I propose you to start with the mature markets.

In the mature markets, APE is at 2% and I would say there are different elements that are necessary to explain just because 2% is an average. First, we have two countries where we have an excellent performance, starting with France, 7%; U.S.

13%. And besides, we have other countries like Switzerland, which suffered from; I would say one-off effect in 2016, because in 2016 in the first quarter, we had a very large contract and as this explains in itself, 1%.

That means that Switzerland is at minus 12% and we would be positive if we would restate for this one-off. So, this Switzerland -- this contract explains 1% instead of 2%.

We would have been at plus 3%. At the same time, keep in mind that we have AXA MPS; AXA MPS is at minus 15%.

The situation is significantly improving, but we started the year at MPS with topline growth at 50% of what it was one year ago. It's fairly better now.

We are -- in July, we are at minus 10% only, so an improvement. And so that's mostly it, so a contrasted situation.

As far as the emerging countries are concerned, minus 2% is due to Hong Kong. Hong Kong, which is at minus 5% due to the closure of a G/A contract in the third quarter 2016.

In line with our intention and in line with what we presented in our plan one year ago, we want a high profitable business and that's the reason why we decided to close this contract. In Southeast Asia, we are at minus 2%, and in the same spirit, it's a decline in Southeast Asia of 2% due to China because in China, we hold a smaller business.

That means we hold a smaller amount of new business. But we see profitability of 25% in term of NBV margin, where one year ago, we were at 10.

It explains the 9% NBV improvement that you can see here. So, as a whole, we have an APE at plus 1%, NBV plus 2%, which is a very strong NBV margin, one of the strongest -- the strongest among our peers at 40%.

Moving to the segments now, I would like to start first with Unit-Linked. You can see that we have an APE growth of 11% with an NBV margin of 32%.

Again, France which is at 25% -- 25% growth in Unit-Linked in the first half of the year, which is a strong performance. Just to give you -- to give more color around this performance.

Expressed in term of revenues, the savings products we write 43% in Unit-Linked today when selling savings product and the rest is in general account. The average of the market [Indiscernible] at 30% only.

So, we are far above our competitors and it's something which is still improving. This is not -- this is due to the fact that we can propose to our clients very attractive products including real estate products, so plus 11% in Unit-Linked.

Next is Protection & Health, so Protection & Health, plus 1%, and of which France, 7%. If I would exclude the contract I mentioned before in Switzerland, we would have been at 2.7% growth in Protection & Health, quite in line with what we shared with you one year ago.

On G/A Savings, minus 16%. This is all the reasons I explained, the combination of Hong Kong with the closure of G/A contract, with Italy as well.

So, as a whole, minus 16%, but as you can see as well, it's the lowest NBV margin. So, in the nutshell, I would say that it's a new business.

It has been extremely selective, with high quality and a very disciplined approach fully in line with the one we described one year ago. Let's move to underlying earnings by business.

This is on a pretax basis on the right-hand side, post-tax on the left-hand side. Let's start first with Unit-Linked, plus 35%.

With higher management fees, I should remind that the average reserves increased by 7%. We benefited from the favorable equity markets.

And at the same time, we have higher GMxB margin, which is in line with the market improvement. In the middle, G/A Savings, minus 4%.

I would say that it's due to the lower investment margin. Pretty in line with our own expectations and in line with the guidelines that we shared with you when we presented our plan.

Protection & Health, minus 10% and this is due to mortality model changes in the U.S., of which correction of a [Indiscernible] of policies after attained age of 100. In other words, we stop modeling after 100 years old and we had to correct this because indeed, it doesn't correspond to the reality and that's the reason why we were at minus 10%.

I should add that taking into account this correction; we have a combined ratio of 96.1%. You can see it in the appendices, as usual and we still maintain the target as presented in our plan, at the end of the period of the plan between 93% and 94%.

So, as a whole, I would say on the post-tax basis, total Life & Savings is at plus 3%. Property & Casualty now, and the total revenues are at 1.4%, and you can see in the bubble, plus 2.3% excluding Turkey.

As you may know, Turkey has been hit by a 30% price decrease of third-party liability following regulation change since April 2017. It's a price gap and this price gap translates into the topline and that's the reason why we are at 1.4% and excluding this, we would have been at 2.3%.

We are quite proud also on the right-hand side to show you that non-motor for personal lines as well as for commercial Lines at plus 3%, absolutely in line with our -- this correspond to our priority segments. Last, you can see that the price effect are quite still very strong, 2.7% and 1.7% for Commercial Lines.

Let's move to underlying earnings. Underlying earnings are at plus 6%, which is an excellent performance and this is explained by two elements.

First, the combined ratio, as explained by Thomas, we have decreased the current year combined ratio by 1.5% from 98.4% to 96.9%. We could add that we had natural catastrophes at the low level, 0.1% instead of 0.7% a year before, but the smaller natural events have been stronger than they were one year ago.

If we combine lower nat cat and higher smaller natural events, we could say that the improvement is 0.3 points only. So, we should have global view of the natural events like this.

So, it confirms at the same time that the attritional losses improved by 1.2 points. Prior reserve developments, as explained by Thomas, we have a strong decrease of 1.2 points.

We should say that we had a cautious approach. And I can tell you that we are still expecting for the full year to be between 1% and 2%.

So, as a whole, 96.1% versus 96.4% one year ago, minus 0.3 points and I still confirm that we're confident in our 2020 targets between 94% and 95%. Asset Management now, on Asset Management, strong performance, plus 10% in underlying earnings with I would say average AUM at plus 7% for AXA IM, with almost the same, plus 6% in revenues, and for AB, plus 2% in average AUM, plus 6% in revenues because the net inflows have been concentrated on the retail segments, with higher fees.

As a whole, 6% of revenues versus underlying earnings at 10%, confirming our capacity to benefit from operational leverage. Balance sheet now, I would say that we're not seeing very different from what we explained and commented over the past sessions.

82% of assets are invested in fixed income. Long asset duration, we didn't change.

I remind you that we are managing duration gap close to one year. And we have a slow yield dilution.

We moved in Life & Savings from 3.3% to 3.1%. For Property & Casualty we are flat, but taking to -- you should take into account this compares to the full year 2016.

And you know that in the first half, we have always the benefit of the dividends. So, we can expect in line with what we told you before we can expect to be slightly lower for the full year.

Let's move to the ratings and as far as the ratings are concerned, we have -- we're still investing, we have still an average rating, which is maintained at AA for the sovereign bonds and A for the corporate bonds. As far as new cash flows are concerned, so we invested EUR31 billion in the first half in fixed income.

You can notice here that it's been done at 2% on average versus 1.8% for the full year 2016. So, a slight improvement, mostly coming from the U.S.

dollar because we have slightly higher rates on the Eurozone, but spreads tightened. We are at 2%, which is a slight improvement.

And you can notice on the left-hand side that we have been investing in the first half 94% in investment grade credit and in sovereign bonds. Let's move now to the ALM, to the Asset Liability Management, you can notice that we still have a significant buffer between yields on assets at 3.1% on the inforce.

The average guaranteed rate at 1.9%, so we have still 120 basis points, even more relevant for new business because we have only average guaranteed rates at 0.3%. So, on average, we're at 170.

But I should add in line with my comments on the French business that across the Board, the new business is sold in combination with higher margin Unit-Linked business. As far as our shareholders' equity, we have a decline from EUR70.6 billion to EUR68 billion, explained mostly by the dividend of EUR1.16 that we paid in May, EUR2.8 billion.

Second, the net income for the period is EUR3.3 billion and the ForEx and others but mostly the ForEx, minus 2.3%. The euro versus U.S.

dollar moved from 1.05 to 1.14. Let's have a look at the debt ratio.

So, the debt is going down from 18.3% to 16.9%. We reimbursed EUR1.8 billion of Tier 2 debt and we issued in January $1 billion of Tier 2 debt.

So, a slight decline. Debt gearing is stable at 26%; interest cover at 15.6 times and the rating causable is AA- stable.

Solvency II ratio, we increased from 197% to 201%. Not a big change, you can see on the numerator and the denominator.

Key sensitivities are not very different from what we shared with you in December. My only comment is that the operating return is quite strong, nine points.

And for the six months and dividend based on the last year, our paid dividends is at minus five points. That mostly explains the increase from 197% to 201%.

Now, I'll hand over to Thomas for the conclusion.

Thank you very much, Gerald. To conclude, three key statements.

Number one, these half year earnings show a very strong earnings -- underlying earnings per share growth of 5%, which is supported in an improvement of the performance in each of the business lines. We have clearly seen that the focused topline growth in Health & Protection and in the P&C Commercial has been fully followed through.

And that certainly, on the Life business, where we would have expected more, we have very clear remediation plans be it in Italy, be it in Japan, be it in Hong Kong. The strong technical improvement is absolutely key for that improved performance.

And I will we particularly highlight again the P&C business, where the minus 1.5% improvement in the combined ratio is a great achievement. All of this, based on our continued strong balance sheet.

Thank you very much for your attention. And we would now like to come to your questions and our answers.

And for that, Gerard and I will be joined by Gaëlle Olivier, who is responsible for the P&C business globally; and Paul Evans, who is responsible for the Life, Savings and Health business.

Farooq?

Firstly, -- does it work? Hi, can you hear me?

Testing, 1, 2, 3, 4.

Yes.

Perfect. So, I don't believe I just said that.

So, on the P&C business. So, I noticed that AXA Corporate Solutions has been one of the strongest growth areas.

And that's also slightly different from your message that you're looking at more of the small and medium-sized type of commercial business. But I'm just kind of wondering what's going on there and what you see as the growth rate of that business because it seems fairly profitable.

So, that's question one. Question two is on the cost-cutting.

So, call it a lot left to deliver. You seem to be talking recently about your reduction in headcount.

Just wondering what are the levers sort of going forward over 2018? What are we -- which are the areas you now need to work on?

Is it just geographical or are there acquisition costs and various other levers? Thanks very much.

So, I suggest that Gaëlle, you take the first question on AXA Corporate Solutions. I think it's important that we explain again what is in the commercial business bucket.

And I would take the second one on the cost. Gaëlle?

Absolutely. So, thank you, Farooq.

And indeed one of you or yourself, you had asked a similar question at the full year 2016 when we presented the progress of our 2020 Ambition. We have disclosed in our 2020 plan that we want to book our overall Commercial Lines book at 3% to 5% CAGR by 2020 -- between 2016 and 2020 and at the same time, deliver a 94% to 95% combined ratio by 2020.

And if you remember, what we had said in 2016 was that this includes as well the international segment, meaning including AXA Corporate Solutions. And at that time you had challenged us, saying looking at where the profitability is today on AXA Corporate Solutions, are you going to meet your target?

And my answer that time was twofold. One is yes; we are reinforcing our target of 94%, 95%.

And two, part of our Commercial Lines development will come from the development of what we have started to do in the past years, including the larger costs, but also a strong acceleration on notably the SME part of our business, which will be supported by the stronghold we have on our distribution -- property distribution, notably in the European market. And what we see today is we see the combination of two things.

One, we see AXA Corporate Solutions growing extremely strongly in the half year, supported both by European markets with those new territories, Brazil being a very strong accelerator of our growth in Commercial Lines in the AXA Corporate Solutions business, together with a strong improvement of our profitability in AXA Corporate Solutions. And if you get our SME business, we have a growth of our SME business which is at roughly 3%, with an overall combined ratio which is below 95%.

So, we are actually solid and disciplined on where we want to grow business. Our first objective is really to get back to a stronger underlying profitability base.

And once we have this, accelerate our business development and that's exactly what you'll start to see in the numbers in 2017.

So, let's come to the cost savings. Again, the EUR2.1 billion is a combination of cost avoidance.

So, if my salary is at 100 and there's inflation every year, if I avoid the inflation, if it stays at 100, it counts for the EUR2.1 billion and it's a reduction in extra cost. When you look at the numbers and make a math, we are at EUR200 million this half year.

If we assume that the first half is equal to the second half and you make the math, you will almost get until 2020 to the EUR2.1 billion. As I said to you earlier, it always takes a bit of time to get the machine rolling.

But these cost reductions and cost avoidance are on the one hand fully implemented in the local plans and if you wanted to, you could do a field test now and ask our tour representatives of the U.K. market, Amanda Blanc, who is the CFO -- CEO; and Bertrand Poupart-Lafarge, who is the CFO, what it means for their country.

Every country has their package, so the EUR2.1 billion is divided amongst the countries. And we can obviously not only expect that the countries are saving all of the EUR2.1 billion, we also as a Group need to contribute to it and show a positive example.

And if you were an employee of AXA, you would have seen that a couple of weeks ago, we have launched an internal project to look at our overall setup when it comes to Group, the regions and the countries and to see how can we simplify the model and get to a model where you have a clearer base of accountability. So, I'm personally very convinced that we will get to those EUR2.1 billion.

And again, if you look in the newspaper and see the announcements that have been done in Belgium, reduction of 600 people that have been done in Germany, agreement of reducing the staff by 800 till 2020. Look in the French newspaper, where you see that AXA has gone from five insurers in France to one by putting it altogether.

You have very clear in tangible proofs that this is moving ahead and that I'm very confident that we will make it. Jon?

Make it easy, we go through there and there.

Thank you. Jon Hocking from Morgan Stanley.

I've got three questions, please. Firstly, just to come back on the cost base.

If I understand correctly, what you're saying is that you've already basically achieved sufficient initiatives that if you seem you can maintain that run rate over the next period, you're likely to achieve the EUR2.1 billion. Does that not suggest that there's further upside of cost over the plan period if you identify additional initiatives over that period?

That's the first question. Second question, just to come back on what you mentioned in terms of P&C combined ratio.

You said you wanted to focus on current year profitability. I'm just wondering, what sort of reserve margin you're building in and presuming there's an expectation of some, albeit maybe small, positive runoff over time.

And then just finally, I'm wondering if you could just comment on what you think the impact of Macron will be on the business, particularly in terms of tax, life insurance and anything else you might think is relevant for the Group. Thank you.

Good. So, I suggest that I'll answer one and three, and Gerald would do question two.

We start with Gerald.

Okay.

Time to think that.

So, Jon, on the P&C current year. Yes, we have an improvement and we are focusing with fixed objectives.

The way we manage it is that we fix some objective in some of current year combined ratio towards the entities. And the prior year is something which is managed in collaboration with the different entities.

If you ask me, do we have -- that was -- do you some positive runoff? Yes, we have some positive runoff and you can easily -- if you go to the SFCR, you can see that's there is some positive runoff and that's the reason why I said we are at a relatively low level at 0.7% first half.

And I was quite relaxed to tell you that we'll be between 1% and 2%, in line with the guidance that we gave before for the full year. Yes.

So, no worry on that side.

So, on your first question, Jon. I mean, the EUR2.1 billion need to be positioned in the line of history, they don't come as the first-ever cost savings of AXA.

If you look back at the plan 2010 to 2015, we have already achieved EUR1.9 billion cost reductions. So, -- and as you know yourself, the first savings are always much more easy than the second and the third.

Getting to those EUR2.1 billion is -- requires a lot of rolling up your sleeves. So, I do believe that there is hardly any further potential.

If there was further potential, I would most likely come to the conclusion with my management team to reinvest it in the business and to transform the business even further because if you look at this EUR2.1 billion, it's a net number after investment. We need to make sure that we keep the steam engine running and this requires, on the one hand, being more efficient, but also to invest into the future.

Your second question on Macron -- or your third question is a very interesting one in the current French environment. Obviously, as you can imagine, after the election of Macron, there's a lot of enthusiasm in France.

And Macron has very much kept very tight to his program. One of the points in his program was the simplification of the taxation of life insurance, which, in general, we all welcome.

Where the details now need to be worked out is the question. How do you simplify and how do you make sure that another goal, which is very important for him and his government, which is the achieving growth and investing into the real economy is being achieved.

And I believe there is a very positive way of combining it and saying, look, we need to, as an industry, move more towards capital light and towards products where you have a better upside for -- a better combination for the customer within upside and guarantee, which also means that you have to go in more productive assets, be it private equity, be it infrastructure, or be it equities. And you have seen that he himself in one of the last -- his last action as Minister of Economy at the time was to launch a new law with a life insurance and private equity after real estate has become a subject.

So, the next logical step for me would be to extend the same to infrastructure and I do believe that the simplification could go quite well with our own view of how the industry needs to move in a better balance between upside and guarantee.

Nick Holmes of SocGen.

It's very good to see you Nick, because the light is -- okay, now I see you.

Thank you. First question is on Unit-Linked sustainability.

A lot of people are very skeptical about Unit-Linked. It's a very volatile product.

I wonder if you could reassure us that maybe things are changing and Unit-Linked is more here to stay in terms of interest rates, guarantees not being existent and more bonds and et cetera. I wondered if you could share your thoughts on Unit-Linked sustainability.

The second area is Asia. And I wondered -- two aspects.

I wondered if you could talk about what's happening in Hong Kong. Clearly, there is a real product focus here on Protection & Health as opposed to savings, and this is not what many of your competitors are doing.

I wondered if you could talk us through that. And then health insurance in the context of Asia, you are conspicuous and being underweight, I think, in health insurance in Asia versus Europe.

What are your plans for health insurance in Asia? Thank you.

Thank you, Nick. The best person to answer all these three questions in a brilliant way is called Paul Evans.

Thank you, Nick. You were very kind.

Let me take those in the order that you asked. So Unit-Linked sustainability.

So, it is true that Unit-Linked can be more volatile in terms of consumer sentiment so that when markets are volatile, it's natural for investors to be more hesitant even if, frankly, the best time to invest is often after a crash. The natural behavior is more volatile.

However, if you counter that with a view on general account, so in today's environment, the returns that we can offer on general account are now very low. And we're not prepared to offer an economic return to generate topline because they're very capital-consumptive.

So, the reality is more and more competitors are now seeing the reality of the G/A return-on-equity equation. So, more and more are offering Unit-Linked.

And therefore, customers are being driven towards that as a long-term investment solution. If you're saving for a pension, don't save in G/A; save in Unit-Linked over the longer term.

So, I would say, it is sustainable. There will be some volatility from year-to-year.

You've seen a very strong period, during which we've seems relatively stable markets. Asian markets and Hong Kong, in particular, so yes, we are focused indeed on Protection & Health and, indeed, in Protection Hong Kong.

AP was up 25% in the first half of the year. That isn't the same, however, that we aren't prepared to also offer saving solutions.

So, we did withdraw a product in the third quarter of last year because interest rates have moved to a point where it's no longer generating another good return on capital. So, we have seen the impact of that loss in volume in the first half of this year.

And that's why revenues are down 5%. However, with interest rates now moving back up, that product can be redesigned and made profitable again.

And we do intend to relaunch into the G/A Savings market in the second half of this year because we can make money again and because fundamentally, there is no Unit-Linked market in Hong Kong because the regulation GN15 and 16 that lead then to the impossible to compete in Unit-linked. So, it isn't to say that we don't like it; we're just very determined to only distribute profitable products.

But we will be seeking to move back in the sales market in the second half. Health in Asia, yes, health in Asia is relatively small.

Obviously, it is stronger in Hong Kong, and we do have businesses in another Asian markets, and it is an area of focus. Indeed the team are currently seeking to lift of the management team for health in Asia and we do expect to see growth in that market.

Nothing to say right now, but it is an area of focus.

So, we do one more question here, and then we move over here.

Peter Eliot from Kepler Cheuvreux. Can I just come back on the prior reserve -- prior developments?

Because I was quite interested, Gerald, by your comfort that you can get 1% to 2% for the full year. If I just look at the sort of the moving parts, I mean, okay, we had Turkey in the first half, but that was only 0.3%.

Otherwise, the deterioration seems to be quite uniform. And when I look historically, most of your reserve releases are always coming in the first half of the year.

So, I'm just wondering, what's giving you the confidence that we can sort of buck that trend if it were in the second half? The second question was on the Life's running yield which -- again, I'm a little bit confused of the moving parts because it's come down 50 basis points over the last 18 months from 3.6% to 3.1%.

And given the duration of that portfolio and the reinvestment rates that you've been doing of sort of around 2%, I'm sort of struggling to see how it's come down so quickly and to spread falling 20 basis points. But then when I looked at the individual countries, it doesn't seem to have moved very much.

So, I'm struggling to understand those different moving parts. And perhaps finally, if I could just ask on the Solvency sensitivities.

As you said, they haven't moved very much, but you have made the interest rate, one, more symmetric. So, the upside has increased without the downside increasing.

And just wondering what's happened there. Thank you.

Gerald?

I propose you to move to page B38 to answer your first question. And just have a look and we can notice that the level of the reserving ratio remains pretty high, and we are 2%, 3% at the end of June 2017, so that's it.

And that confirms what I told you that we are not at all in a situation where we could forecast any decrease in this capacity to release some reserves. On your second point, you should keep in mind that, first of all, there are some situations depending on the type of investments, so it's relatively difficult first half to first half to make a very reliable comparison in terms of investment income.

But when I told you that what is relevant is just to look at the 71 basis points, which is the investment margin. And the investment margin is absolutely in line with what I told you.

And you remember that on June 21st, 2016, we said again that we would be between 65 and 75 too. So, to be at 71 basis points in the soft semester of this period gives me a lot of comfort.

So, that means that I confirm this objective and I don't have any fears. There could be some situation.

And the second and the other argument which at least partially, we'll answer your questions -- your last question, is the fact that we have more and more flexibility vis-à-vis our policyholders to decrease the remuneration when rates are at such a low level. We have flexibility to look at what the buffers that I showed on the enforced business.

It's quite strong. So, no, it's not at all, for us, a matter of uncertainty I would say.

Let's move forward to this side.

It's James Shuck from Citi. I had three questions, please.

First thing on the underlying growth, you showed that at -- 5% H1, great result. But I think at the Investor Day, you kind of suggested that it takes some time for some of these initiatives to actually stop biting and the profit pro forma, a little bit more backend-loaded that we might expect.

So, my question is really, what's going better than you expected at H1? You have talked about those digital initiatives that you're investing in, that it's time for some of those to start paying back.

Are you able to give any specific examples about those that actually contributed towards profitability? That's my first question.

Secondly, around the central liquidity. You stopped disclosing the cash level of the Holdco.

I'm just interested to know, what is the level of cash liquidity? And how do you actually want to manage that balance?

Do you actually have a set target that you want to manage around? I kind of connected to that.

Are there -- is there any surplus linked to the SFCRs? Germany seems to have quite a high level of Solvency II ratio.

Is the surplus an end of the operating units that you would want to upstream to the Holdco? And then finally, around the debt optimization plans, the Ambition 2020 points to 10 points of debt optimization coming, I think, from basically not rolling over -- or rolling over the Tier 1 -- the maturing hybrids into senior debt as opposed into qualifying capital.

It doesn't seem like you actually rolled that over into senior debt. It seems like it's going into Tier 2.

So, are you still intending on setting aside 10 points to Solvency II capital by -- as part of Ambition 2020? Thank you.

Thank you. I suggest, Gerald, you answer all the questions, except for the digital ones, which I would take.

Okay. So, let's -- your first question was about what is going better.

Unit-Linked. Remember, when we presented our plan, we said that it was assuming that the equity market will be flat, and we've been -- we are benefiting from higher equity markets.

So, this obviously, it's a matter of improvement. The cash level at Holdco, it's not because we don't want to present the cash level in the holding company.

It's only because before, maybe you remember, we were netting the net debt on the cash that we had at the holding company, which means that we were not consistent with some of our peers. And many of you asked us to be more consistent with our peers, that's why.

Yes, we have some cash. We have a comfortable level of cash, I would say, at the holding company.

You remember, we said that we had capacity to invest EUR1 billion in M&A per year. So, that means that we have -- we had EUR1 billion available that we got last year.

This year, we should get one. We did some share buybacks 36 -- 37 million shares.

It's EUR1 billion less. So, we have at the whole something like sufficient in order to face quite a significant acquisition.

But at the same time, I would say that the surplus, which was -- it's a soft part of your question, the surplus in these subsidiaries, you remember that we have a remittance ratio of -- we have a remittance ratio commitment between 75 and 85. We have effectively some surpluses that we intend to upstream this year and -- but you noticed also that on the 10th of May, we said we would recapitalize by EUR1 billion the U.S.

So, on a net basis, what I can tell you is that we will remain -- although we will have to do this recapitalization, we'll stay within these 75 to 85 points -- 75% to 85% of remittance ratio. So, we're comfortable on this side and we have some capacity to reduce the excess capital, as we explained two years ago.

And the debt is very simple. When you have such an opportunity and when the spreads are at such low level, you prefer this time to borrow and to issue some debt other than senior debt.

Because the difference between look at the spreads two year ago compared with today on sub debt and senior debt from an opportunistic point of view, I can tell you that I prefer to issue some debts. That's what we did.

And within the next years when it will be much more normalized, I anticipate that it will be much more some senior debts. But for the time being, it's clearly opportunistic.

If I may just clarify. So are you still expecting to -- for the Solvency II ratio to four by 10 points by 2020 as you're all over that debt?

Yes, that means that nothing has changed on that side. Nothing has changed on that side.

But again, it's -- no, you can issue sub debt today with a spread of 160 basis points I'm seeing like this, which is really low.

Quickly on the -- sorry?

No, no go ahead.

On the digital investments, I think you have to differentiate it between three different types of investments and they all come into play in a different time frame. The first investment area is around the question, how can we improve the customer journey?

The customer -- improving the customer journey means how can we automate processes where you don't need people anymore, where you can automate and streamline. The second one is the question, how can you better play the interplay with direct and agent, with the aim as well as to reduce the commission.

And the third one is the question, how can you be more transparent towards the customer and also push self-service with the customer, which again, you then don't need people to do it. If we take a simple example, AXA France.

In AXA France, which is one of our biggest markets, we -- this is already in implementation. So, we're working and have achieved -- and this is one example of many significant increase in the automation rate of the business, simplifying the business, centralizing it.

We have achieved in AXA France a better interplay between direct and agents. If you want to buy a motor insurance in AXA France today, you will most likely be focused on a product called Clic & Go, which has a much lower commission ratio than the traditional product used to have in the agent channel, with the effect that the agent sales for motor has fallen to almost 30% from a very different level.

And then when you go further and look at how have we increased transparency on the question, where does your claims payments down, how can you also register claims online, this has also gone up significantly. Those investments have been done and have also been done in other countries and will pay off relatively quickly.

We have a second category which is around data. In data, we have invested significantly in the so-called data innovation labs in Paris and in Asia where we employ data scientists to be better on using the data we have.

Again, this is something that I would expect over the meantime to really kick-in, but we also have, obviously, focus area in the short term that kick-in. We have been working a lot on the question of reducing claims fraud.

Gaëlle has been working a lot with her team on cat modeling. Those things are already in place, but there is many, many more initiatives.

We have, for example, started and rolled out in AXA France something called 360°. It is basically like an Amazon approach where you get to understand what is the next best product to buy with the aim of cross-selling customers.

This has worked so far in the pilot very well and, when it's going to be rolled out, will also kick-in. And then you've got a third phase where I would say they are more the longer term ones where we invest in new business models, where we create new business models.

We have launched an incubator called Kamet where we have very different business models to launch already that are now in the phase of growing. Again, to take a very classic example and real example, we are in London here, one of the start-ups is called [Indiscernible].

It is a telemedicine service for foreigners in London to have local doctors connecting to them. This has been founded, but it's obviously now with very few customers already working, but needs to grow now.

And there, I would expect those things do come longer term.

Andy Hughes, Macquarie. Two questions.

First one is about kind of the growth in Life & Savings. I mean, the business seems to be growing the fastest at the moment within AXA.

It's actually the U.S. Life & Savings variable annuity business, which is kind of a little bit strange because that's the business you want to get rid and take capital at.

So, why the strategy to grow rapidly in the U.S. ahead of a disposal where the peers, if you look at, for example, Brighthouse, they're trading at very low PE ratios.

So have you, for example, looked what's happened to that one post IPO? And does that call you a picture as to whether you're going to proceed with the strategy?

And when you get the proceeds back, presumably, the EUR1 billion you're paying into the U.S., does that change the Solvency II recognition of your surplus in the U.S. and we have to write-down the chunk of the surplus capital in the U.S.?

So, will it still count with the Group Solvency, because you can't get that back. And is that to unwind the guarantee you're applying to AXA Arizona at the moment?

I'm sorry. The proceeds question.

The business seems to be lacking in Asia, particularly [Indiscernible] was -- you don't know which profit business are on call, which, I think, is probably a massive handicap for the moment compared to AAA in Peru. Is that something you can buy or build using the proceeds from the disposals?

Thank you.

So, I suggest question one and three is answered by Paul, which is one, being on the why and where is the U.S. growing so fast and what about the risk profits in Hong Kong.

And Gerard will take the middle question of your whole array of questions around the U.S.

So, looking at the contribution of growth of APE from the U.S., I think we need to also understand the factors which are depressing growth in APE elsewhere in the Group because many of those are step-change factors which are disguising through underlying growth. So, we've touched on some of that already.

So, in Switzerland, for example, where we're reporting minus 12%, actually that had been plus 1%. But for the large Group Life contract this time last year, Japan is down 10%, but that's entirely due to change in fiduciary regulations in Japan.

That Australian dollar bond product that we used to sell, which is now pretty much closed business, that's a step-change in volumes. Underlying business is still minus 1% in Japan, but that's due to the [Indiscernible] of our health product and we're launching a new product in the -- late in the third quarter.

Belgium is minus 5%. That minus 5% in the context was an enclosed retail business and focused on core Group Pension business, which is plus 27%.

Spain. We decided to move out of G/A Protection.

So, Spain is down 28%, but the underlying business is up and was 80%. And Hong Kong, we touched in the second part, is down 5%, and that's because we closed new business, the Fortune Goal Saver product.

If I need -- and I'm not going to say you should correct for those things. But the underlying business, including the U.S., is growing at 5%.

If I take all of the U.S. out, it's still growing at 3%.

So, come back then to the U.S. I think the story really is there is underlying growth in the rest of the Group that has been constrained by actions we have taken to hold back the business in G/A and, of course, very much in Italy by the impact of BMPS on Italian volumes.

That will start to unwind as the core businesses restore their growth in Q3, Q4, and into next year. A look at U.S.

U.S., clearly, we've seen very strong growth of the structured capital strategies, which is an alternative to the GMxB in the variable annuity product space. Very compelling product, I made the same comment to you at the full year results.

It was very strong last year also. And I think that is simply a very strong offer.

In the context of a market which is performing strongly post [Indiscernible], frankly, and pre-DOL. Now we don't yet know what the impact of the DOL will be.

So, DOL payment in place in part in the middle of June. So, we'll start to see in Q3 and Q4 what the outlook for volumes is.

We are expecting some contraction of VA in the U.S. in the context of DOL.

We haven't changed our estimate, which we sent to you as a potential impact in terms of APE in the U.S., but we'll have to wait and see what that impact might be. But nevertheless, for the first half of this year, the U.S.

performed very strongly in the context of strong underlying markets and a very strong product, which is very compelling to advisers and to customers. Do I think that, that business should constrain growth or adapt to the strategy in response to its IPO?

And looking at it frankly, no, I don't. I think it's performing exactly as I would expect it to in the best interest of shareholders today and, frankly, in the future.

Asia, does having no profit business constrain us? Well, I think what constrains us in Hong Kong in particular was, a, we were a market leaders in Unit-Linked.

And of course, GN15 regulation effectively stopped the unit-linked market. So, we had to rebuild general account presence.

Secondly, we are very disciplined in our general account profitability standards. So, therefore, we come in and out of the market when those, as you say, when profit funds have more off-balance sheet capital and can afford to stay in the market during the highs and lows of interest rates.

That is a disadvantage, but nevertheless, we believe that through product R&D work, which we're doing at the moment, we can introduce such satisfactory products for our clients in Hong Kong. And we intend to do that in the second -- the third and the fourth quarter of this year.

So, I think we can compete in Hong Kong without having profit business. So, I wouldn't be saying that we should acquire one with the proceeds of the IPO.

Thank you, Paul. Let's go the question that concerns Gerald.

Yes. With the U.S., I would like to remind you that we are in a quiet period.

So, nevertheless, what I can tell you is that you cannot say you will dispose. That means that it's an IPO and an IPO means that we'll remain, as you can imagine, shareholders and that we have a direct interest in this company performing well.

The second part of your question was about what will be the impact on our -- on the Solvency. So, long as -- and we clearly said when we announced in our press release on May 10th, that we would intend to reinvest it and to reinvest the proceed in acquisition.

And we said that whatever would happen, we were confirming our objective of 3% to 7% and underlying earnings per share growth of the plan period, which confirms that we would prefer to invest in our business. If not, it's quite clear that we would do share buybacks.

So, that means that in terms of Solvency, because it was your question, Solvency II should be neutral. So, you should expect that in the end it will be neutral.

Let's move down here, and then we'll go over to Andrew.

Hi, it's Johnny Vo from Goldman Sachs. Just a couple of questions.

So just first question, just on the prior year releases. Is there any relationship between nat cats -- low level of nat cats and prior year reserve releases?

And to that point then, if we have another half year of a very benign period of nat cats, would we get the 1.3% minimum release from the prior year? Is that what you're saying, Gerald?

That's the first question. The second question just relates to the sensitivity to Solvency II.

Clearly, it looks like your Solvency II ratio is very stable, but you haven't given a sensitivity to sovereigns. So, can you provide the sovereign sensitivity?

And the second thing is can you provide the sensitivity to write increases post the 20-point where the UFR kicks in only? That's the second question.

And the third question just relates to the elephant in the room. And I think you partially answered this question.

The problem with the 2020 Ambition, I guess, is the use of proceeds for -- to meet the 3% to 7% target. So, can you give us more of a split potentially between acquisitions and buybacks or whatnot?

So, we can feel more comfortable that the group can actually reach the 3% to 7% with effectively a sixth of the earnings disappearing over the period or potentially moving down?

Gerald?

So, I would say that the prior year release has nothing to do with the current year level of nat cat. It's completely independent.

It's how it works. That means that we have an experience, and it's only after few years that we can say, okay, we have an excess of reserves or not.

So, honestly, it's completely independent. The second question, it was impact of sovereigns on Solvency II.

Look at the -- we can take it off-line, but look at the solvency sensitivity; you can make your own math sensitivity on the interest rates. So, that means if you have -- you can make pro ratas and so on.

But so long as you have sensitivity to a 50 basis points move, taking into account the sovereigns and the percentage of sovereigns that we have in our own balance sheet, it's a EUR575 billion of general account, you can make your own math, but we can do it together after, but it's very simple. Last is the -- yes, the UFR.

UFR, roughly speaking, we can expect next year to have an impact of around three basis points. So, that's not something huge, three, four something like this and the use of proceed for 2020.

Remember, we had different boxes when we presented our plan. And we said that EUR1 billion investment per year corresponded to 1% additional growth.

So, that means out of the 3% to 7% compounded growth of underlying earnings per share, 1% was due to the [Indiscernible]. We have absolutely no commitment, and we didn't give any percentage or any guidance on any kind of share buyback.

We've committed only, remember, on share buyback forms of investment eventually for the reinvestment of the proceed of the IPO, and secondly, for the dilutive impact -- for the dilutive instruments, which are much smaller and so I put this aside. So, we didn't give anything but -- any guidance, but it's 1%.

It's 1% which is at stake. And -- but you can imagine that many times during the session, we confirmed that we had an objective to stick to this 3% to 7%, so we'll do it.

We go down here and then move over.

On the Non-Life, so this is from memory and I may be wrong, I thought Turkey was actually a positive in terms of combined ratio when premiums were capped and I just wondered if you can explain what happened there and what's going to happen going forward? In the cost cuts, the 0.2% of claims handling cost reduction is that included in the EUR200 million cost cuts or does the come on top?

And then one question -- I think Gerald you've already answered, but maybe if I put it slightly differently. So, your debt is coming down EUR18 million to EUR16 billion, does this is mean that you've got more comfort now to raise the payout towards the 55% end of the scale?

Thank you.

So, Gaëlle, do you want to do the question on Turkey? I'll try again with the costs and Gerald will handle the last question.

So, on Turkey what you see this year is an impact negative on the growth on the volume development due to the market cap because the cap that the regulator has said is roughly 30% below the market average. So, we have, like most of our competitors, a strong drop of renewed driven by this cap.

These are the topline side. Now, what we have done, and we have started to do that for the past two years following the change of regulation on Motor TPL, back in 2015.

What we have done is taking strong action on BS [ph] business picks in Turkey, significantly decreasing our market share in Motor TPL. Back in 2015, if you remember, it was close to a 25% market share.

It is now closed to 6% market share. So, we have significant actions that we have taken like this one on our business picks, but also actions on our operational efficiency.

We have been able to significantly decrease the combined ratio, which is what you see today in half year 2017, where Turkey has a combined ratio which is decreasing more or less by three points versus last year. Then going forward, we will continue to have a strong action on business picks operational efficiency so that we are not too much impacted by regulation, whatever happens on Motor TPL in the coming years.

You should be reasonably comfortable about this.

So, going back to the costs, yes, the EUR2.1 billion includes three components, administration costs, claims handling costs, acquisition costs. And what you see in terms of numbers we are working on all of these levels.

And just to give you -- I mean administration cost is simpler, and it's probably the most straightforward. The claims handling costs, yes, I mean going back to the example that I said earlier on being smarter about frauds.

Frauds you need less people to go and inspect fraud and follow frauds, if you can do it more digitally and more data-driven. I'll take the other example that I mentioned on Clic & Go, the French motor product.

If I pay less commissions and if I have less people to follow-up on the acquisition administration part, I can reduce the costs. So, yes, the EUR2.1 billion covers all three areas.

Gerald, last question.

The last question, I would say that if I made of good try in order to discuss about the payout ratio, but we should not confused Solvency and liquidity and you know this by heart. In other words, it's not because -- first of all, we are not short of cash.

I had a question before connected to this and the dividend and the Payout ratio is an impact of -- on Solvency and it's not because we have a constraint in terms of cash that we don't go to the upside at 55%. So, that's -- these are completely, as you know, independent factors.

Andrew?

Good afternoon. It's Andrew Crean, Autonomous.

Could I ask three questions? Paul, you were discussing Unit-Linked.

I just want to ask a question, we're eight years into a bull market and you've got negative net flows in Unit-Linked. Last time we were in a bull market in 2007 you had 7% positive net flows.

Why is there such a structural and secular change there? Secondly, on the Non-Life side, I think you said in current year claims ratio has improved by 1.2 points if you adjusted for both nat cats and attritional weather.

Could you also talk about large losses because they can be a volatile element and extract those? And then thirdly, in terms of the improvement in the Solvency II coverage ratio, I think nine points came from the operating side.

Could you just tell me how much of that was management actions and how much was underlying?

Thanks Andrew. Paul, for the first question, Gaëlle for the second one, and Gerald for the third.

Could I take mine last because I don't recognize the question and the number. I've got positive EUR1 billion in my-

Okay. Then let's go in the other direction.

Gerald starts, Gaëlle, and then Paul.

I would say on the management action on the operating side. I would say not so much.

That means that anytime -- this is real easy, underlying performance of our business. Remember that we said that we would be between 15% and 20% per year and we are in the upper range.

We were roughly at the same level last year. So, that is real easy, underlying performance nothing in this 9% concern, for example, the additional debt, the debt, debt pre-financing, and so on, which are on the right hand side of the roll-forward.

So, that's what I can say. Paul?

If you can find A-20 Thomas. Perfect.

Thank you. So, that's what I have, Andrew I may have misunderstood your question, but I have positive EUR1 billion net flows Unit-Linked and that's despite negative EUR1.3 billion of the GMxB in the U.S., so that includes the EUR1.6 billion of SCS business I referred to earlier.

So, it's down very slightly on last year, but it's still EUR1 billion of net positive.

But I think if you look at the financial supplement it's negative. So there's a difference.

But the fundamentals of the question is that net flows were 7% of start year reserves in 2007. They're now either slightly minus or slightly plus.

There's a secular change there.

But if I may. The big difference in 2007 was that at time, the interest rates were at much higher levels.

Today the competition is either you invest in Unit-Linked with a much broader range of products that you can -- where you can invest or you get 2% today and maybe 1% tomorrow.

If interest rates were higher, I'd invest in general account product.

Yes, but you have today a very different competitive landscape. I think if you go back 10 years, the asset managers in the banks have not been as developed as they were today.

And I think the other factor might be, if you go back to that period, Andrew, of course, we owned the U.K. Life business in 2007.

And if you look at the main Unit-Linked markets, they are the U.S., France and the U.K. So, you've taken out a substantial portion of what was a Unit-Linked-only market in 2007.

But maybe I'll take your question offline, if I may, because there's clearly something that I haven't understood in your question, perhaps.

Good. But we still have a third question left for Gaëlle.

So, let me try to bring you towards the roll-forward. 1.5 points of current year combined ratio improvement.

We have said we're benefiting from a low level of nat cat events, 0.6. We are suffering from less favorable development of small nat events 0.3.

Net, it's minus 0.3, positive effect on the combined ratio, positive effect on the loss ratio as well, obviously. Then we have 0.6 improvement on frequency which means that the pricing actions that we have taken have resulted into better profile in overall of our portfolios.

We have large losses which are slightly improving versus last year. It's flattish to slightly improving.

And we have claims, of course, deteriorating, due mostly to spare parts inventories from car manufacturers. If I make a long story short.

So, to your specific question on large loss, we're not -- where we're at this stage, we have the same development of large loss as what we have had last year, slightly better with margin and EBITDA I would say.

Next to [Indiscernible].

Niccolo Dalla-Palma from Exane BNP. My first question is on health insurance.

You kindly provided us with more disclosure now pulling together the bits from P&C and Health, -- and from Life, sorry. My question is the following, so we can see there's 6% growth.

So, first question, how much of this is coming from Group business versus individual business? I guess the individual one is more valuable, longer term, to your Ambition.

And the second question is taking aside potential M&A opportunities, what can you actually do organically to continue to drive this. The second question -- I take the opportunity as we've the -- as we're in the U.K.

with management in the room. There's little note in your report on the fact that Ogden has no impact because you have -- your reserve level is fine.

By memory, no one so else had had that, so just if you could give some color on that. Thank you.

Good. So, Paul I guess you are starting with the first question on health.

And then Amanda, do you want to take the second one? Or Bertrand will take the second one.

Good.

So, the health revenue growth, which is 6% overall, is actually split, 6% growth Group, 6% growth individual. So, actually both parts of the portfolio are growing equally.

You are right that it is no more for the individual or more profitable in the Group account. On an inorganic, yes, it's one of our strategic growth areas and it's something that we would have an eye to.

Perhaps, Thomas do you want to comment on inorganic, but it's certainly in our horizon to do something in that space.

Yes. I mean the inorganic, as Paul said; we're looking at this very carefully.

Fortunately, it is an area that not so many competitors are interested in. Because if you look at AXA globally, you see that obviously, the very large U.S.

players are the biggest, but they are very focused on their domestic market. If you take it off, you see that AXA is with it's almost EUR13 billion of premium, hopefully, by the end of the year, the number one player internationally, very strong in six markets, so we have a very high interest and we are looking very carefully at where can we reinforce our sales beyond those six markets where we're already quite large.

If you take one local example, the U.K., with PPP we are number two in the market, we're gaining market share, but it would be very difficult to acquire further. So, we need to go beyond those six markets, and there we have very clear plan, which are these markets and which targets we are looking at.

Ogden, Bertrand?

On Ogden, it's very difficult of course to comment on our competitor's results. But as far as we're concerned, we had a very prudent reserve policy and we have been able to absorb the change of Ogden rates without any P&A impact.

Further questions in the room? Difficult to see.

Thanks. Paul De'Ath from RBC.

Just a couple of questions if I can on China. So, you mentioned, obviously, the decline in the topline on the Life side in China due to a change in the product type.

Going forward, is this again a step change and then you see kind of growth going forward? And how does that compare to what your peers are also doing in the markets.

Presumably, now your peers are the local players rather than the other foreign-owned insurers. And then the second point was just on the Alibaba announcement that came last year.

I think we talked about it last year. We haven't really heard anything more on that as far as I'm aware, is there any update on what's happening there and where that can go in the future?

Thanks.

Paul, do you want to take the first one? Should I take it?

What would you like to do?

50-50?

50-50, it's a good deal. You do the first 50.

Look, our priority in China has been to substantially reduce the loss-making business -- or the very low margin business that we were selling in China. In the first part of last year, which we have seen, we saw the MBB margin improve to 25%.

And that came with a cost of 3% lower volumes. So, we have changed the mix.

So, in that sense yes, it's a step change. Going forward, yes, we are a foreign player, therefore we are partnering with a bank and therefore, we are working with our bank partner to understand their growth strategies for the coming years and where the Life & Savings propositions fit in.

But we would expect to deliver strong growth from this point from China. Thomas your 50%.

So, my 50%. I can give you a travel report of my last travel to China which is about two weeks ago where we discussed about the Life strategy which fits perfectly into what Paul has said.

Both partners have a very clear understanding that the single premium business that is so popular in China around the new year, and therefore, you always see it in the first-half figures, is something that we want to de-prioritize going forward and we want to focus more on the regular premium products where you have already seen a first step shown in this margin improvement. So, there's very clear understanding on both sides that this is the way of going forward.

The second big piece that I would like to mention to you going forward is that we have relooked at what are actually the two strengths of the two companies. And with ICBC, we've got the world's largest bank as a partner.

Just to give you a sheer idea of the size, their Shanghai branch makes as much profit as AXA globally. The second thing is, we have AXA as big as the Shanghai branch.

And that knows very well of how to manage Asian networks. And if you look in China, at our composition, the have a strong Bancassurance channel with the 10,000 bank outlets of ICBC out of 14,000.

And the second one is we have an agent channel of roughly 3,000 agents. The idea is, going forward, to bring those two closer together which is a value proposition that hardly anybody else has.

And then coming also to your question of competition, the competitors have very different models. You have agent-only models focused on regular premiums-only and you have at the other end of the spectrum which is mainly the local players very, very heavily single premium-driven models.

When you look going forward, our positioning today is very unique in a way that nobody else has such a partner in the local market. So, we have unit positioning, going to use it is extremely important.

And the Chinese regulator is extremely vigilant that the practices in the Chinese market are focusing more and more on more sustainable business and more profitable business. So, if you look at a very simple action, in the last year, the capital charge has been significantly increased force in the premium business which will drive the market as a whole, more into regular premium and more into regular premium-plus riders.

And if you look at our composition of having a bank channel with a large amount of customers, and today we are only penetrating 2% of the ICBC customers and a strong agency salesforce and a good partnership with a mutual understanding, I think we're a quite well-equipped for moving forward. You're second question on Alibaba.

The Alibaba Corporation was focused on two areas. One was the question of ensuring the exports of Chinese companies outside of China, and second one was to insure traveling Chinese outside of China.

We have started the corporation. As you can imagine this is never easy to first get to the corporation and then get it started.

But we have already signed the first insurance contracts recently and we do hope that with a certain starting time that this will continue to grow.

Thank you so much. The first question is a bit of a cheeky one.

So, the impression I have is--

Another tricky one?

Not, cheeky, cheeky, sir.

Cheeky.

Never tricky. So, my impression is you deliberately were more prudent than usual in the first half results, and that meant you aimed and landed at that 5% spot.

So, you had the EUR100 million in Life in the U.S. And if I think about the reserve release, that's probably about another figure of the same magnitude.

If equity markets remain strong in the second half, you'll have the same problem in the second half. Where are you going to park the money if that continues?

So, that would be my first question. And then the second question, on the way in, and I'm sorry I was late, I met some guys who had come out of your meeting and I think they were little bit less enthusiastic than I am.

And -- so what are you going to tell us to make us more interested at your Investor Day? Because you can't talk about, what the other guy said is, the elephant in the room, or whatever.

And what is there which can get people excited, I think the story is great, but we need to bit more sizzle if you think, see.

If I tell you now what we're going to tell at the Investor Day, I have a risk that you won't come. Gerald, on the first question, I know you are satisfied that we don't have 4% prior year like other companies have, but Gerald might tell you how much we have parked or not parked.

No, but I've -- honestly I have no problem to park the money, so I don't know how to answer your question. It means that, yes, if things go well; it means that we're quite optimistic on this full year, so that's it.

And we go back to your previous question about liquidity. Earnings is not liquidity.

Liquidity, yes, it can be managed. My first priority is to have earnings which are in line with our plan and to -- if possible if the can have -- to be in the, I would say, to go to be on the way and not on the low part of our commitment, which was 3%, of course, I'd be very happy.

So, just quickly coming back to Investor Day. I'm not going to show you the agenda.

But if you look at the questions today, there are probably some questions that need a bit of further work and further detailing. I'll give you one question, there was about three or four questions about cost saving.

I think it would be good to show you a little bit more in detail with concrete cases, what's going to happen, where it's going to happen. We spoke a little bit here in the front about new business models.

I'm sure you want to know what is that telemedicine service? Or what are the other business models that you are doing?

You want to see these guys that are doing it. How many customers, what's the revenue?

So, there's quite a few things that have not been answered yet, which I do believe you should be interested in. But I'm stopping myself now.

Hi, Andy Hughes from Macquarie. Very quick follow-up question if I could on the prior-year profits.

So, I noticed the prior year profits have fallen most of all for the Direct business. Is there a reason for that as to why, because I assume given the things you mentioned about the technology, about settling claims online, et cetera, that would trigger more prior year releases from the Direct business as put through lower claims handling costs.

So, obviously, the ongoing Commercial business seems to be okay. And the second thing was on the Direct business in the U.K.

there was the -- at the year-end we talked about growth areas. You actually highlighted the success of the U.K.

Direct and your trading business. Yet there appears to be weakness of the revenues in the Direct business now.

So, could you just update where those things are going? Thank you.

So, Gaëlle, would you like to take the first question the second question is going to Amanda.

On the first one, I don't see any specific dis-balance on prior year reserves between Direct retail or commercial. We have a very well diversified book of business.

It's developing well. We have an underlying profitability improvement which is very strong in most of our entities.

If I take Europe, for instance, in Commercial Lines at half year, it's growing by 3% in Commercial Lines decreasing its combined ratio by 3%. So, all this is a very good sign, which makes both Gerald and myself, reasonably confident that we should have a good full year lending of all P&C business, including some positive news on the prior year reserves, which we expect to be within the guidance which we have had in the past years, between 1% and 2% of our premium development.

So, we will stick to that, not specifically to one business or another. On the U.K.

Direct piece, I think my colleague here; Amanda probably is better suited than me to answer this, where we have specific events in the U.K. which can explain the trends that we see.

So, I think from a U.K. Direct perspective, obviously, the business on motor is almost 100% digital, so ultra-efficient, but there are challenges in the market.

We know the discount rate challenge and the challenge around accidental damage inflation. And we've been pricing for that and therefore, we're not prepared to sacrifice profitability for volume.

So, we have gone for profitable business and actually that business is very profitable.

Any other questions in the room? Amanda do you want this on a personal mic?

I'd just like to ask about the future direction of the capital requirements, so the SCR, particularly, if we look over any multiple years. So, the Life business you're writing is capital generative in day one.

If you think about what you're capital requirements might be to fund growth, whether it's in Commercial Lines or the Protection Act outside of kind of inorganic things. It is difficult to see any particular increases in that SCR, particularly given diversification benefits.

So, if we're looking forward to the end of 2020, are we looking at an SCR that is broadly stable to where we are now? Or are you thinking -- well there's more you can do on the balance sheet side of things to increase your risk appetite and manage your overall solvency ratio in that basis?

Gerald?

No, I don't expect it to be fundamentally different. There will be a -- and I may be enlarged a bit your question, but I believe that, as you know, you have companies using different models than the others.

As far as a stronger formula is concerned, there will be some adjustments that will be discussed next year. But what will be much more important will be the discussion around the internal model and I believe that what gives with the volatility adjustor will be really key because, for the time being, it's a bit detrimental to the diversification, to the long-term investment, and so on and so forth.

So, I believe that we could imagine an improvement on that side. As far as the rest of the SCR is concerned, as far as we are concerned, it's based on our internal model and the internal models are not negotiated.

It corresponds to a reality and, of course, it's -- there are some exchanges with the regulator, but it won't change -- I don't expect it to change a lot within the next years. So, that's what I can say for the time being.

So, risk appetite, which is the second part of your question. I don't think so.

We have some adjustments of our risk appetite, but if your question is on the asset side, on the asset side, it's still again -- and many times we have been discussing it. Today, it's more than 50% of our SCR which is substantial.

We don't want to increase it because in the end, it would bring nothing. You are not investing in an insurance company.

You know that to invest in a -- in order to realize some technical profit and to invest in our hedge fund. So, that's the way I would answer.

There could be some adjustment but nothing major really.

Thank you.

Any more questions in the room? That doesn't seem to be the case.

Then let's go to the web and see if there is any questions there.

Hello, there's three questions that have come in from Thomas Fossard at HSBC. The first one is on Life & Savings.

He says, in spite of the effort of the Group to direct new flows -- news flows, sorry, to the most attractive products in the line, the NBV margin is flat in the U.S. and minus one point in France.

At 40% NBV margin for the Group, are we starting to reach a ceiling although still staying on a high level? And the second question is, the P&C Personal net new contracts on slide B33 had a rather negative momentum in the first half of 2017.

Leaving aside the specific issue in Turkey, there is still a loss of 100,000 contracts in France, 100,000 in the U.K. and 100,000 in direct.

Should we be faring this negative dynamic and should we expect this trend to continue? And the third question is on slide B18.

Can you comment on why capital light product inflows were only EUR0.7 billion in H1, versus EUR2.3 billion in the first half of 2016.

I suggest Paul you take questions one and three, and Gaëlle number two.

So, first question was has the new business margin now leveled out? And I would say that's going to be a function more of geography and mix than anything else going forward.

So, if you look at Protection & Health, our margins there are 60% and that's a strategic area of focus. If you compare that with Unit-Linked at 32%, G/A at 18%, mutual funds at 8%, then you can see how the margin could still develop by further improvements of business mix, away from general account towards Unit-Linked and Protection & Health.

You can also see from the detail in the pattern, the margins by countries, do vary, In particular the growth that we saw on the first half which came from the U.S. The U.S.

is one of our lower new business margin geographies compared to others. So, I don't think you should see the new business margins having leveled out.

Your third question on capital light inflows. Yes, they have reduced.

They reduced notably in Japan which relates to the change of Fiduciary rules in Japan which led to the -- what was called the Australian dollar product which was a capital light general account saving solution effectively coming off the market and there we saw a 47% reduction in capital light APE. We similarly saw reductions in Italy because Italy was still selling general account, but it was capital light.

And of course, the reductions we saw following the BMPS situation, has flowed through to the capital light area. And Spain, whilst the products there were capital light, we nevertheless were concerned with their profitability and we reduced volumes there too.

So, in each area, the capital light volume reductions are linked to individual country instances where, either we have chosen to, or have found ourselves in a situation where those ones were reduced. That doesn't signify in any way, a reduction in the focus of shifting away from traditional G/A towards capital light G/A, or even better, to Unit-Linked.

And in fact you are seeing a substantial reduction in traditional G/A. You can see that the cash flows in the first half were minus EUR2.9 billion in traditional G/A.

So, you're seeing a substantial reduction in our focus on traditional and instead you're seeing the flows go to UL and to capital light, but in this first half, less to capital light due to those three particular countries.

Thank you, Paul. Gaëlle?

Net new contracts. So, net new contracts is the retail contracts that we win or lose over the period.

So, it's a focus on the retail Personal Lines development. It does not include the commercial lines development.

So, if you remember what I said earlier on Commercial Lines development, we have started to experience a strong -- or an accelerated business development. On the Personal Lines side, what we see if you put Turkey aside, is indeed a drop of the number of retail customers, notably in Europe, mostly coming from the U.K.

And it's due to the U.K. property actions we have taken to turn around our portfolio and to increase our prices to improve our profitability.

And that's two actions which Amanda and the team have taken in the U.K. which were necessary to increase the profitability of the U.K.

household portfolio, and to a less extent in France. Now, if you add altogether in Europe the number of customers that we have lost on the Personal Lines front, it's 88,000 customers in Europe that represent 0.3% of our customer base in Europe on the Personal Line side in P&C.

So, you see it's a marginal number. What we want to do and continue to do in the coming years is to continue to grow our business in a profitable manner.

We see an opportunity to diversify our business beyond motor and that's where if you relate it back to what Thomas and Gerald have said earlier, we see a gross non-motor which is at roughly 3%, which is higher than the growth we are experiencing in motor. And that's part of the business mix actions that we are taking because we still see a strong opportunity to cross-sell and up-sell on our existing customer base in Europe.

So, yes, net new customers going down in Europe, marginal given the size of our customer base in Europe, 0.3%, and all in all, driven by actions which we have taken to improve our profitability in some specific portfolios. So, nothing which is worrying us at this stage.

Are there any more questions in the web?

Just one. [Indiscernible] who asks, what should be expected in terms of M&A?

Alliance is in the Credit Insurance business through Euler Hermes and why is AXA not in the Credit Insurance business?

So, these are two very different questions. What should we expected in the M&A?

I repeat what I have always said. You should expect us to look very carefully at all the M&A files that are in our focus areas, Health and Commercial, P&C.

You should expect us to not look at very large deals because we have global size today. I repeat again, AXA is most probably one of the most diversified global insurers which -- where eight to 10 markets make 80% of the profit.

And we will also not look at very small deals. We look at deals in the area of EUR1 billion to EUR3 billion but very selectively.

On the Credit Insurance, Gaëlle, would you like to answer that question?

I can take it. Credit Insurance is a commercial line business opportunity that we look at in terms of new territories.

But we look at it in a very disciplined and exhaustive manner because it is quite a capital heavy line of business. And as you remember, we are extremely attentive at the level of financial risk that we have on our balance sheet.

So, we look at it in a very selective manner I would say.

That's it in terms of questions from the web. What about the telephone?

No question, anymore questions in the room? Then I would like to thank you for your attendance and particularly for your questions.

And wish you a great afternoon and hope to see you soon. Thank you.