Aegon Ltd (AEG) 2014 Q3 法說會逐字稿

Good day and welcome to the Aegon Q3 results conference call. Today's conference is being recorded. At this time I would like to turn the conference over to Willem van den Berg. Please go ahead sir.

Thank you operator. Good morning and thank you for joining us for this conference call on Aegon's third-quarter 2014 results. As always, we will keep today's presentation short, leaving plenty of time to address your questions. We would appreciate it if you take a moment to review our disclaimer on forward-looking statements, which is at the back of this presentation. Our CEO, Alex Wynaendts, will provide an overview of this quarter's performance and will then be joined by our CFO, Darryl Button, to answer your questions. I'll now hand it over to Alex.

Good morning everyone. Let me begin by saying that this was clearly a quarter of mixed results. Earnings this quarter have been impacted significantly by the changes we have made to our assumptions and the enhancements to our models, something I will address in more detail very shortly.

Excluding these one-time items, underlying earnings were EUR512m, which is in line with what we would normally expect. And at the same time we are pleased with the strong profitable sales performance over the quarter and the continued execution of our strategy, including the recently announced sale of our business in Canada. Cash flows were robust, taking into account a higher new business strain related to the exceptionally large pension contract win in the Netherlands. There was also a one-time impact on a capital position in the Netherlands following a valuation methodology change for a Dutch mortgage portfolio. I will also return to this in a moment.

Let's turn to slide 3, where we have provided you with a comprehensive overview of the outcome of our assumption review and model updates. I am clearly disappointed by the outcome of our annual assumption review, but on that note let me say that this is a core part of what we do. We make assumptions. We validate them through our experience and adjust them when necessary. And that is exactly what we have done this quarter.

As we flagged earlier, we have updated our mortality assumptions and strengthened reserves in the US. This was the result of emerging experience supplemented by industry studies and relates mostly to the older age segment, ages of 85 and above. And this is a group for which history has limited experience. On the contrary, our assumptions related to policy behavior for variable annuity book have been too conservative and these were also updated to better reflect actual experience. The combined impact of assumption changes on underlying earnings was EUR138m, as you can see in the table.

Also included this quarter is the result of several model enhancements. Over the past year we have intensified our efforts to review and enhance our models. We've made many different products and therefore models so this is a very significant undertaking. More than 60 high-priority models have now been validated and, where necessary, adjusted. The impact here was EUR82m on underlying earnings and EUR151m on income before tax. We have also improved our processes and governance related to the ongoing maintenance of our models.

In addition to the one-time impacts shown in the table, the assumption changes and model updates will have a recurring negative impact on underlying earnings before tax in our US life and protection business of approximately EUR20m per quarter. This is significant, but I can assure you that we will be looking at all options available to mitigate the impact as much as possible.

Let me now turn to capital on slide 4. We have adjusted the level 3 fair valuation methodology for our mortgage portfolio in the Netherlands to use more market-observable data. As a result, our IGD ratio in the Netherlands stands at 220% and the Group IGD ratio now stands at 202%. This is a valuation methodology change only and does not change our view of this asset class or of our Dutch mortgage business. Therefore we expect the immediate capital impact to be recovered through cash flows over the lifetime of the mortgage portfolio.

The UK pillar 1 ratio was stable and remains at the target level. Here though I would like to point out that uncertainties will remain until the recently announced DWP regulatory changes are fully implemented in the coming years.

And finally, in the US our capital position is healthy at $1.1b over the S&P AA level.

Total sales were up 38% this quarter, which is a reflection of the continued strength of our franchise across our markets. Slide 5 shows the strong growth of our life and accident and health sales. Life sales were higher across all markets, and the highlights this quarter was securing the largest ever pension buyout deal in the Netherlands. In the US we continued to see higher demand for our universal life products, in particular in next universal life. Growth in accident and health sales has been driven by the addition of new distribution partners, successful walk-side marketing and additional amount for our products created by the introduction of the Affordable Care Act in the US.

Now turning to slide 6, strong momentum in deposits growth continued this quarter, with record gross deposits of over EUR15b.

Aegon asset management had a very strong quarter, with new mandate wins in each of its markets. Third party inflows more than tripled to EUR7b, driven in part by growing demand for Dutch mortgage funds, which has now assets of over EUR2b.

US variable annuities had record deposits, both on a gross and a net basis. Deposits remain strong in our US pensions business despite fewer takeovers as we continue to see higher recurring deposits.

Other parts of our business are also contributing to deposit growth. Assets on our UK platform increased 20% to GBP2.4b and the platform continues to be one of the fastest growing in the market.

In the Netherlands our online bank, Knab, is also growing. This is a great example of how we are using technology to get closer to our customers and provide the products and services they look for. It took some time to generate momentum. However, since we have repositioned the business earlier this year, the number of customers has more than doubled, with assets now stand at over EUR800m.

So, in summary, while I'm not pleased with the negative impacts on this quarter's earnings, I'm proud of the many ways in which we are successfully executing on our strategy. Sales are higher across almost every part of our business, underlining the successful expansion of our franchise. We announced the sale of our business in Canada and we continue to work hard towards repositioning our Company successfully in the digital world. And therefore I am confident that Aegon is well positioned to not only manage future challenges but also to take advantage of the many opportunities we see across our businesses.

Darryl and I are now happy to take your questions and I'm sure you will have a few. Thank you.

(Operator Instructions). David Andrich, Morgan Stanley.

Hi. Good morning. Thank you for taking my questions. My first question is just around the model assumption changes. And I was just wondering, going forward now, what kind of stricter controls have you put in place and how much more consistent are the assumptions across your different business units?

And then second of all, I was just wondering, in terms of the UK business, you state that the cash flow generation is challenged due to ongoing regulatory changes. I was just wondering, has something changed since Q2 or is this more just along the lines of the continued transition of the business in the UK? Thank you.

Hi David. It's Darryl. I'm going to take your first question just on the model updates and assumption changes. First of all, on the assumption changes, the biggest change we did have is the one we flagged back in the second quarter on the mortality. This is really an area where our industry data has been limited in this ultra-older-age category. Our own data as well. But we've seen enough combination of our data and industry data that suggested that we need to make this change, and so we've done that. I don't think that we're going to be needing to change the mortality assumptions again any time soon, so I think we've dealt with that.

Not all the assumption changes were negative. We had a significant positive on the variable annuity on the lapsation as well, so that's part of our annual update.

On the model updates, I would say that we have had an intensified effort over the last year to really go through and scrub all of our high-priority models, and Alex mentioned that earlier. So I certainly hope that this lends itself to fewer adjustments going forward. As part of that process I can also say that we've also strengthened our own internal governance for ongoing review and maintenance of these models going forward, so I hope that also contributes to a much lower likelihood of these assumption updates in the future.

David, on the UK, this is a continuation of the transition which we are in. It's a transition, as you know, from a business model. But it's also the changes -- the regulatory changes which are affecting our business and which still have to be implemented. I can just name a few of them.

The implementation of DWP. As you know, it's not only a price gap which has been included, but probably more importantly in fact could be the elimination of the difference between what we charge active members, members that contribute to the pension plan and members that don't. So all of that has to be worked out. That also requires we engage with each of our customers individually because these are large customers for which we have a contract with their participants. You're well aware of the annuity changes. There's been recently another change in taxes for individuals.

So all these changes all create uncertainty. And I just wanted to make sure we understand that we are working hard towards repositioning our business, that we are executing on what we committed to execute on. But we are in an environment that has a lot of changes, which also means that it requires a lot of attention of our people to address those changes and implement all these regulatory changes, of which some are still not even clear how they need to be implemented. So it's a continuation of this transition phase.

Okay. So no change in terms of the expected cash flow from the UK business going forward then?

Well, we have given you a range which we see as a target for the UK. I've been very clear that we are looking at the lower end of the range. And I believe that these changes will make it probably even more challenging than it was before these regulatory changes were in place to achieve that target.

Okay. Thank you very much.

Ashik Musaddi, JPMorgan.

Yes, hi. Good morning everyone. Three questions. First of all, can you give a bit more color on what this EUR20m per quarter is and how do you arrive at this number?

Secondly, it looks like all this assumption changes and model updates has no impact on your cash flows, i.e. is it fair to say that this is purely an IFRS accounting thing and has no impact on the capital position on a local basis, i.e. you don't have any concerns from the upstreaming dividend from US or Netherlands because of all these accounting assumption changes and model updates?

And thirdly, basically can you give us a bit more color about these model changes? How much of that is driven by low interest rates? How much of that is driven by underwriting? So a bit more thought on that would be really appreciated. Thank you.

Hi Ashik. It's Darryl. I'm going to try and hit those questions, all three of them. On the EUR20m per quarter, how is that arrived at, basically when you're strengthening the assumptions, it's really strengthening the assumptions that goes to both the reserves and the DAC in combination. Effectively in this particular case it was strengthening reserves. What it does is it changes the trajectory of the reserve build over time. So there's a one-time catch-up on the reserves and that's what you've seen through the P&L period going forward.

There's also additional premium that has to now be set aside to increase the reserves further so the trajectory of the reserve build is increased. And that's what drives the EUR20m a quarter going into the future.

It's actually -- it is not an IFRS-only. There will also be a regulatory cash flow impact out of the US as well. I expect that to be in the neighborhood of maybe even slightly larger than this EUR20m on regulatory earnings in the US. So that will impact their operational free cash flow going forward.

The third question, I must admit I didn't --.

Some color on the model changes.

The model, yes, the model changes themselves. Yes, so you can see that there were a collection of different model changes, and you can see that on the (technical difficulty).

Generally it falls in the category of using -- really upgrading and making our models more sophisticated on the life side, increasing the model points that we now model for a lot of the life assumption changes was getting better premium persistency assumptions down to lower cohorts, which means basically getting more granular data points down into the models, which would have been difficult in the past due to technology restrictions. But as we continue to upgrade technology and we?re able to basically get that finer granularity down into the model point.

Same thing goes on the variable annuity side as well, an update to the withdrawal assumptions enabled to take that down into a finer cohort of individuals and data points. Those are really the nature of the changes. The one on the fair value side related to getting a better projection of our future hedging costs down into the model. So they all fall in that generic camp of more sophisticated, newer technology model enhancements.

That's really very clear. But just one follow-up on that, the EUR300m of the total impact you flagged in third quarter itself, how much of that is actually going to cash, i.e. how much of that will weaken your cash position?

And just a follow-up from that simple question again is how should we think about the trajectory of cash flows because if I remember correctly you mentioned that from US you should get roughly, give or take, 900m dividend to holding company from Netherlands. I think it's around 300m. So what changes today from the -- on these two numbers specifically?

Yes. Ashik, there's actually very limited impact to the capital position in the US from this change today. I caveat that slightly in that at the end of the year we'll do our annual cash flow testing work at the end of the year and there may be some impact on that. But for the most part, there's really no cash or regulatory impact from these IFRS changes today.

However, the lower mortality and the adverse part of the assumptions will obviously bear through regulatory earnings as that experience emerges into the future, which is why the regulatory earnings impact will be slightly higher than the IFRS earnings impact going forward.

Yes. Okay. That's very clear. Thanks a lot.

William Elderkin, Goldman Sachs.

Hi. Good morning everybody. I've just got two questions left, one following up from the previous one. Effectively if we're looking at US underlying capital generation, we should be lowering that, by my expectations, by around $100m a year, if I understood the answer to the previous question correctly.

And then secondly, just on long-term care, a number of your competitors had a few problems there. I was just wondering if you could give us an update as to your position and whether there's anything there we need to be worrying about from your side.

Yes, William. This is Darryl. I'm going to take those questions. Yes, your math is right on the US adjustment to the operating free cash flow. That's a pretty good estimate.

Long-term care, we had a small -- so as part of our assumption changes there was a small hit. It was about EUR16m, so that's a pretty small number in the grand scheme of things in terms of updating the assumptions. I will say as part of the actual quarter performance, we did flag that it was not a strong quarter from a mortality and morbidity perspective, and that was around EUR60m total. About half of that was the morbidity experience, EUR30m. That was related to long-term care. And we have seen some higher claims this quarter and we strengthened the, effectively we have strengthened the IBNR reserve, which is setting aside what we think we need to set aside for future claims.

So that is impacting the Q3 results by the EUR30m plus the additional EUR16m on the assumption change this quarter. Otherwise we're comfortable with our reserves and I don't foresee anything different than that.

I think we can take the next question, operator.

Albert Ploegh, ING Bank.

Yes. Good morning. Thanks for taking my questions. The first one is on the Dutch mortgage impact, the change in basically in the fair value accounting thereof. What are the implications for that also on the cash flows going forward? It seems actually it probably will move up a little as a result. And will it also impact maybe your strategy to invest in Dutch mortgages going forward and whether there's potentially any impact or any difference in the impact between Solvency I and Solvency II rules on that?

Then the second question is on the business review of the lower performance -- performing assets, and in specific France. Is there anything to mention on the progress on that specific file?

(Technical difficulty) pointing towards [the real]. Is there anything you can say on that? Thank you.

Albert, I must admit the phone cut out after your second question. Could you repeat your third question?

Sorry. The third question is on interest in the Dutch consultation to participate. Your local competitor made clear statements last week, so any color from your end would be welcomed as well.

Okay. I'll jump in with the first one. On the Dutch mortgages, yes, so we have updated our valuation. You're correct actually. It will have a higher cash flow going forward, so it's really a point-in-time valuation change. We really feel very strongly about the ultimate performance of this asset class and the underlying asset cash flows. So any valuation change that we make today will come back into capital and cash flows into the future, so you're correct on that.

In terms of impacting our investment strategy into Dutch mortgages, we actually have gotten to the point where we are really, from a concentration risk perspective, filling up on our capacity on our own balance sheet for Dutch mortgages. So starting this quarter in fact we've started diverting more of our Dutch mortgage production into our fund -- our fee business, so the asset management fund that we've opened up to sell third party. So that's where I predict more of our Dutch mortgage production to go into the future. Obviously that's a little more lower-margin for us going forward, but it comes back to the concentration that we have to manage on our own balance sheet.

Yes. Darryl, I'll take the two other questions. In terms of the business reviews, I was obviously -- we were very pleased to announce the sale of Transamerica and Canada. As you have probably guessed, this has been quite a long process, a process where we have really looked at all the different options and tried to ensure that we would get the best possible outcome for shareholders.

And the same applies to France. Again here we are reviewing options. We've committed to give you clarity by the end of the year. We're looking at the various options we have, all of it with the same objective of optimizing the returns for our shareholders. And, as you can imagine, if you're in a hurry, that's usually not the best way of optimizing returns. So we will do what is right here.

Your last question was about the consolidation. I will repeat what I've said all along. Aegon is focused on organic growth. In the Netherlands we have a very strong position, and we've made that very clear. We've not participated in the process around [real] and we are not intending to do anything different there.

When you look at our organic growth in the Netherlands, you can see we're well positioned. We're particularly well positioned in the pension business, where we're able to secure the largest ever deal. It was not only won on the fact that we have a good proposition, but also because here we were able to provide the participants of the plan the kind of products and services they were looking for. And that has made a difference. So that is really at the heart of our strategy, growing organically, and that applies to our businesses across the world.

Thank you very much for your answers.

Nick Holmes, Societe Generale.

Hi there. Thank you very much. Two more questions on the mortality review, please. The first one is can you give us more color on the level of confidence that you have on your new assumptions? And I'm thinking things like how do you compare with the industry? And how much scope have you built into the assumptions for further deterioration in mortality?

And then secondly, you mentioned options to mitigate the EUR20m recurring impact. I wondered what those might be and, in particular, I wondered is longevity something which might be a natural hedge to what you've got on the mortality? Thank you.

Hi Nick. It's Darryl. Yes, so on the mortality assumptions, I think now we compare very well and very conservatively, I think, on the older age mortality issue. This is something I've mentioned in the past. The industry data has been sparse but is now starting to emerge. I've seen enough benchmarking data to show me that this is not just a Transamerica issue, that this is an industry issue. So I think you're going to hear more on that as we go forward. But I feel very comfortable, now that we've made the change, we've made a significant change and to a point where we're very comfortable with our assumptions.

On your second part related to longevity, the only think I would maybe mention there is that we have been I think very upfront that we have, with our hedging programs here in the Netherlands, the risk that really we are left with what dominates the balance sheet becomes longevity risk. And so we are very actively looking to pursue longevity-hedged transactions and move those into the capital markets, and we've executed on a couple of those deals already. And we continue to work on that and you'll see more from us on that front.

Great. Thank you very much. I just wanted to come back to the options to mitigate the EUR20m. I think, Alex, you mentioned that. I just wondered what you were thinking of.

Yes, Nick obviously we always look at our business, ways of optimizing. And what we are saying here is we are going to have to look again as to where are the options to further improve efficiency. I'm looking at our big back book in terms of expenses, margins. So it's not something specific that I can share with you at this point in time. This is at the end of the year. It's a process around the budget. And you can -- I just want to make sure you understand that both Darryl and I are going to make sure that we do everything here to mitigate as much as possible the impact.

So, sorry to labor this point, but just, very finally, does that mean there might be scope to offload some of this business, that kind of thing that you're looking at?

Again, Nick, at this point in time I can say we will be looking at all options to optimize our position here and clearly -- and I have said that I'm disappointed by the outcome. I'm disappointed by the one-time hit but obviously, equally disappointed by the fact that it has an impact going forward. So we'll be looking at all ways of optimizing our position here.

Okay. Thank you very much indeed.

Farooq Hanif, Citi.

Hi everybody. I hope you can hear me. Thanks a lot. I want to just go over some topics again, if you don't mind. Firstly, on the RMBS assumption change, I would have thought that the reserve, with the additional valuation hit that you've taken for that, could come back quite quickly into capital, given the duration of the book. So could you give us an idea quantitatively of how much of the, I think the EUR500m or so, is going to come back into cash every year as a result of that? That's question one.

Question two is on the upstream. I know it's early to comment on Q4 and next year, but it seems to me that you've got a very good surplus generation in the US and obviously a hit to IGD in the Netherlands. So I'm just wondering whether this is going to really change your policy on payout or level of upstream from these two businesses. Could it have an impact?

And lastly, you talked about there being no capital impact from the assumption changes, but theoretically-speaking, if we had a Solvency II number, which I know we don't have, but if we had one, would there be any impact on Solvency II from any of the changes you've made? Thank you.

Hi Farooq. It's Darryl. Let me try and address those questions. On the first one, on your first one, yes, just basically divide by seven gives you a rough feel for how much that cash flow will come back in. So that gets you to about EUR70m a year going forward on a EUR500m adjustment. So I think that's a reasonable number there.

In terms of upstreaming, yes, so we've talked about the US and obviously the excess capital position in the US remains very robust. And the US, we are in the fortunate position that the US cash flow is not impacted in any way by new regulatory standards, i.e. a la Solvency II. So I expect the US cash flow will, as we've suggested through the year, will come in the second half of the year -- sorry, in the fourth quarter.

And the Netherlands, I think it's very much a fourth-quarter decision. And we're having conversations obviously and will be having conversations with the Regulator in the fourth quarter. Obviously the Solvency I IGD ratios are very strong, but we continue to prepare ourselves and implement -- and get ready for implementation on Solvency II. So that's very much going to be a fourth-quarter discussion and conversation in the Netherlands.

On the third question --.

It was the Solvency II impact of the assumption changes, yes.

Yes, it's a very hypothetical question I guess, but I think it's fair to say if you look at the upfront impact that we've had on the assumption changes this quarter and you combine it with the quarterly impact that we've talked about going forward, I think that gets maybe a little bit of a feel of what might be on more of an upfront basis within a Solvency II context. That's probably about the best I can do. We're obviously not running Solvency II numbers on our US business.

Yes, of course, yes. So it basically has no impact, does it? Of course. I wasn't thinking.

No, it has no mechanical impact.

Mechanical impact, yes.

Again, our working assumptions for Solvency II on our US business remains that we will get equivalency and use deduction and aggregation.

And just one, just to go back very quickly. So on the US business you very carefully didn't use any adjective when you were describing the fourth quarter's cash flow. You just said it was going to come. But you weren't saying if it was going to be better or worse because of the higher surplus.

Yes, just because we've been running above our capital targets in the US and we have -- obviously the difference in the US is we have much better clarity on the regulatory framework. And so I think the excess capital we have in the US combined with the clarity that we have on the US statutory and regulatory framework, gives us a lot more sightlines and visibility on the US upstreaming. We continue to prepare ourselves for Solvency II here in Holland.

Okay. Thanks very much.

Gordon Aitken, RBC.

Yes, morning. Just on, back on to this US mortality assumption change, understand it's 85 to 95 year-old male smokers are not living as long as you expected. I know you flagged this in Q2, but it is a surprising development given that the general increases in life expectancy, especially we've seen those at older ages. So firstly, why do you think you are seeing this in the US and not, say, in your UK business or your Dutch business?

Second, to what extent is this a new development over the last 12 months or for several years have you seen more people dying versus your expectations?

And finally, in Q2 you mentioned that you were seeing lower than expected reinsurance recoveries on this. Why aren't the reinsurers paying out?

Yes, so generally what I would say on the US mortality, there is a general trend to increasing longevity. That's still there. But what we've had to adjust back is that that increase in longevity is not as much as what we had predicted it would be.

You have to go back into, if I go back a little bit into why the data is just really starting to emerge, you really get back into older-age state planning, second-to-die spouse products and things like that. These are the products that really extend out into these older ages. And they were popular in the 80s and 90s, are when those products started being sold to people in their 50s and 60s. So that's the cohort of people that are now in the 85 to 95 area that are coming online now that's giving us the data points that we're seeing.

So we've had to pull back our assumption. And, again, I will repeat what I said before, I've seen pricing benchmark data from the US that suggests this is not ours alone. But we see a trend and it's important. We still sell these products and we are still very much committed to this part of the market. So it's important that we acknowledge and address the trends when we see them so we can get our pricing assumptions updated and make sure that we don't grow into this problem into the future. So that's I think the importance for us to deal with this now.

I think you're going to see a good, continued emerging experience as the data comes online. There was a recent, going back a year or so, an old-age mortality study performed in the industry that really did start to correlate with our own data, which is giving us the confidence to make this assumption change. I'm not sure I can add a whole lot more. I really don't want to comment on competitors or other experience, or how you've been hearing that from others. I don't think that's my place to do here.

Reinsurance?

Yes, so reinsurance, they are in a position -- they do have -- they are in a position to see more cumulated data than any one of the individual companies. I think actually the experience there has been mixed, and that's a question you're going to have to ask them in terms of what they've seen in this part of the market in terms of this data trend.

Just at Q2 you seemed to imply that they weren't -- the reinsurance recoveries were down.

Sorry. Sorry, yes, you had asked specifically. Yes, well, that's part of, to be honest with you, that's actually part of our assumption updates as well in terms of so the largest number that I've been talking about is the older-age mortality adjustment. We also had an impact for adjusting our own reinsurance recovery assumptions in the model and that's also impacting the life and protection numbers in the US this quarter.

Thank you.

So basically we've strengthened those, yes.

Francois Boissin, Exane BNP Paribas.

Yes, good morning everybody. I have two questions remaining, please. The first one, on Dutch mortgages, can you give a bit more detail on what drove the depreciation in value? And does this have an impact on your nominal value of mortgages or does this lead to a lower market value of mortgages only? That's my first question.

The second question is on the market value -- sorry, market-consistent value of new business, you had quite a decline in Q3, mainly in the US and in the Netherlands. I wondered whether you could provide a bit more details on why that was and how you saw the outlook for that. Thank you.

Sure, I'll take the first one. It's Darryl. On the Dutch mortgages, what drove the change, it was a change in our discounting methodology to come up with the fair value of the mortgages as we report under IFRS 13. It's a level 3 valuation, which means we can't go into the market and get an actual fair value on mortgages from transactions directly so we are obliged to come up with a fair-value calculation.

We basically have gone back into the market and we have adjusted our methodology to incorporate more market-observable information. We've taken a look at some of the RMBS transactions that have been done over the last year as well as updated our methodology to include more of the junior tranches on these securitizations in the calculation. Effectively what that means is that we increased the discount rate, which lowered the fair value of the mortgages. And that's, the fair value of those mortgages are included in IGD calculations and will be included in the Solvency II calculations as well.

So it's strictly a market-value adjustment and it does not impact the nominal value. It also does not impact the IFRS earnings, only the fair values, which do not come through the IFRS P&L statement.

In terms of your question of the value of new business, what we are seeing in this quarter is very strong sales. We see sales up 38% compared to last quarter. Sales are up in our key markets and that has had a clearly positive impact on MCVNB. However, the positive impact was more than offset by two items.

First of all, we've had lower interest rates, which means that the value of new business with lower interest rate is lower. And what we see also is, Darryl mentioned it earlier, we are generating a significant amount of mortgages now for our third-party asset management business. There's a lot of demand for it. And that part of the mortgage obviously does not come into the market-consistent value of new business because it does not come through our general accounts. And that explains why you have, despite very strong sales and you would have expected high MCVNB, slightly lower MCVNB this quarter compared to last year.

Okay. That's quite clear. In terms of outlook, should I understand that basically now your ability to write new mortgages for your own accounts has been limited? So should we expect lower margins or, let's say, should we not expect margins to recovery in the Netherlands going forward on the back of that?

Well, these mortgages have been put on the books so these margins are there. These margins will be there for the maturity of the mortgages. And what we are doing right now effectively is using our franchise in the market, with now 13% market share mortgages based on very strong origination capabilities, to leverage that on behalf of third-party business.

This is attractive business. It needs no capital. It's business where we get a fee. Fees are attractive, between 40 and 50 basis points. And they stay -- they are very sticky because once you buy a mortgage, an institutional investor buys a mortgage, it will stay -- it will have to hold on that mortgage until the customer, the one that bought -- that issued the mortgage, will redeem it or repay it.

Yes, I understand this, but this is more a fee business than third party, but how should I think of your general account business in the Netherlands?

As I was saying, these mortgages that are on the book, they will hold on these margins. The margins will stay there.

Right. Thank you very much.

Thank you.

Jan Willem Knoll, ABN.

Yes, good morning gentlemen. Thanks for taking my questions. Back on the model assumption review, you mentioned you reviewed 60 high-priority models. What percentage of your reserves has been covered by this review and what is the euro amount of reserves that will be covered by the ongoing review going forward?

And then on Dutch non-life, you mentioned a negative impact of a number of large claims in general insurance. Maybe you can give a bit more color on the size and nature of these claims.

And then lastly, on the Dutch pension buyout market, maybe you can comment a bit on the pipeline you are seeing there, any impact, let's say, of the low-interest-rate environment on the appetite of the pension funds to move to a structural solution.

And also the fact that your Dutch IGD ratio has been declining in the quarter quite significantly, would have any impact on the short-term or the short-to-medium-term on your appetite to grow in the Dutch pension buyout market. Thanks.

Yes, I'll try and cover the first one. I can't give you an exact reserve coverage number, but what I can tell you is, as I mentioned before, we really intensified our efforts over the last, really over the last year in terms of a comprehensive model review and update across the organization. We've focused our efforts and segmented them into the high-priority models, which were a function basically of the bigger reserve coverage and the bigger numbers. And we've really gotten through all of those high-priority models. And I mentioned earlier on the call that we've strengthened our governance program in terms of ongoing model validations and governance and control reviews.

That now carries on going forward obviously on these high-priority models, but also extending to all of the lower models. So while I can't give you an actual reserve number, what I can tell you is that we've covered all of the material and larger-ticket models with this update.

In terms of general insurance in the Netherlands, as you know, there are two parts. One is I would call it property and casualty. That's where we've seen a number of exceptional large claims. They relate to a number of fires which we've had. You don't need many actually to move the number from a profit to a loss here in this quarter. However, what we see is a further improvement in our disability segment and this is the result of actions we've been taking in the last year to improve the quality of our portfolio, improve underwriting, but also to increase prices. So we are here on a good trend but have had a couple of exceptional items.

Your question on the Dutch pension market, just to put it in perspective, the large buyout of the miners' pension had an impact of EUR70m on our new-business strain in the Netherlands and that obviously is exceptionally high and it was the largest ever contract. In general what we've seen in the past quarters new-business strain has been closer to the level of EUR15m to EUR20m and that is what we expect also to be going forward.

This business is lumpy, so you'll have big contracts then you might have a couple of quarters of much lower ones. We are still very interested in this market. We believe we are well-positioned. We want to make sure that any deal we do is a deal that also meets our requirements in terms of pricing.

This deal, the mineworkers deal, the pensions for the mineworkers, clearly is within our pricing discipline. What's important is that it fits very much in our strategy because we're not only looking at the pension plan in itself, but we are looking at the participants. And I've talked about that more often. These are customers, the participants in the pension plan. They need more products, more services. You know governments are retiring here in the Netherlands, as they do in many other parts of the world. The responsibility shift goes to the individual and this is at the heart of our strategy. But we will do that only if we are able to achieve our pricing requirements.

Thank you.

Okay. Thanks.

Steven Haywood, HSBC.

Hi. Good morning. Thank you for taking the questions. I just wanted to clarify something. The model update for hedging costs on your GMWB VA book in the fair-value items, this EUR46m, what exactly does this relate to on this -- on the change in the model update for the hedging cost? I thought Aegon had implemented its dynamic hedging structure here fairly recently.

And second question, I see there's an adjustment for lower yields of about minus EUR8m in the Americas line. Can you just clarify whether this is in addition to the EUR10m per quarter dollar charge that you disclosed or is it -- is this related to that?

Yes, hi Steven. It's Darryl. On the -- I'll take both those. On the first question, specifically what we've done -- you are right, we've made no change to our dynamic hedge program on the GMWB; that's fully intact.

What this involves is projecting that program into the future and where that's important is actually on the DAC calculation. So it is a fair-value product and a fair-value hedge program, but we have to project those hedge costs forward into the end of the life of the product and then adjust our DAC models. So this change that we made was actually a more-enhanced way of projecting those future (inaudible) and hedge program through the life of the product and that had a one-time impact on DAC and that's what you see in the -- on the fair-value line.

On the drop in yields, it actually is just a one-time adjustment as we reflected and updated the curve that we use at the Q3. So it's, I would say, it's unrelated to the standing guidance, if you will. This is very much a one-time impact and I wouldn't expect this to recur next quarter. The standing guidance on the impact in rates is just the ongoing investment yield drag that comes from persisting in the low interest rate environment.

Thank you. Can I just follow up on the last one? The standing guidance, that's not changed either?

No, it has not.

Okay. Thank you.

(Operator Instructions). Archie van Riemsdijk, Dow Jones.

Yes, good morning. Thanks for taking my question. It may be a reiteration of an earlier question, but could you maybe once more give some color on the change in valuation of the Dutch mortgages?

And to what extent can this be described as accepting that the losses on the mortgages are higher than the model previously expected?

Yes Archie. It's Darryl. First, let me deal with the last part of your question first. It actually has nothing to do with the expected higher losses on the mortgage. I want to be perfectly clear on that. We still feel as good today as we did yesterday on the quality of the Dutch mortgage cash flows so that has nothing to with that.

It really is about a level 3 valuation under IFRS, which means we have to do a fair valuation. It requires expert judgment to do that and there's some subjectivity in that because we cannot point directly into the market and find the fair value of the Dutch mortgage, which means we have to come up with a discount rate to discount the cash flows. And there we've used more recent market-observable information to adjust our calculation.

And that leads to a valuation change day one. Obviously whatever -- because we still feel about, strongly about the cash-flow development on the future, whatever we strengthen day one will come back to us over the next coming seven or so years as the mortgages mature.

Okay. Thank you. And as an additional question could you give the number of how much capital -- well, it has increased, the capital requirement for these mortgages, by EUR0.5b, but what was the previous requirement and how much has it increased?

Well, that goes into our actual overall IGD ratio. So what you have to understand is that the fair value of the mortgage book is inside of the IGD calculation. So we've reduced the fair value of the mortgages by EUR500m. And that's EUR500m on a EUR26b, EUR27b mortgage base so to put that into some context, if you will.

Okay. So you have no number for the total capital requirements of this mortgage book?

I think that's something you can follow up with our Investor Relations or Robin actually would probably be your best contact on that.

Okay. Great. Thank you.

Okay.

Maud van Gaal, Bloomberg.

Yes, hi. I'm also looking for one further clarification and sorry if it's already been discussed, but it's regarding the assumption review. In slide 3 it says that a review of the lower-impact models will continue throughout 2015. Will that then lead to further adjustments to quarterly earnings? Will that have another impact?

Maud. It's Darryl. I can't promise that it won't, but I don't expect it to, is probably the best thing that I can say. And we've taken into this quarter everything that we can see or would expect. The other thing I would mention is that we've really dealt with the more material ones up through this quarter. So it's part of our ongoing -- and I really tried to describe that we really have strengthened the governance in terms of our ongoing model reviews going forward and this will be part of that. So I'm not expecting -- I think if there was something it would have at this point an equal chance of being positive and negative.

Thank you.

Yes.

As there are no further questions in the queue, I would like to turn the call back to our hosts for any additional or closing remarks.

Yes, I just would like to thank you for participating in this call and thank you for your continued interest in Aegon, and have a good day. Bye-bye.

Thank you. That will conclude today's conference call. Thank you for your participation ladies and gentlemen. You may now disconnect.