Good day, ladies and gentlemen, and welcome to the ASR conference call on the 2016 annual results. Today's conference is being recorded. At this time, I would like to turn the conference over to Michel Hülters, Head of Investor Relations at ASR. Please go ahead, sir.
Thank you, operator. Good afternoon, good morning to those of you listening in from the U.S. Thank you for joining this conference call on ASR full year 2016 results. Presenting today are Jos Baeten, CEO, and Chris Figee, CFO. Jos will start today's call with a summary of the full year results. He will also discuss some strategic highlights and business progress. Chris will then provide further detail on the financials. He will talk you through solvency and capital as well. Following these presentations, we will have ample opportunity for Q&A. Please also have a look at the disclaimer in the back of the presentation for your perusal. Having said that, Jos, you are on.
Thanks, Michel. Ladies and gentlemen, 2016 was, to our opinion, without doubt, a successful year for ASR. Our strategy of value over volume delivered on its promises. We are proud to report a very strong set of financial results for 2016. Throughout the year, we have been able to keep our business momentum at a high level. Our full year performance is in line, or sometimes even better than our medium-term targets. Make no mistake, these targets are challenging. It is hard work to get these results. We will continue to work hard, and that is a promise. We will discuss our financial performance and strategic developments in more detail. Let me start off with an overview of some of our key metrics. Those are on slide two.
The highlights on this slide clearly show that our performance in 2016 has been strong on almost every key metric. Our Solvency II ratio is robust at 189. This is based on the standard formula, as you may know, after deduction of the proposed dividends already. Before dividend, the ratio, by the way, was 194. The quality of our capital remains high as well, with Tier 1 capital alone representing almost 160 of the SCR. There is still plenty headroom to maneuver in both terms of Tier 1, over EUR 1 billion, and Tier 2, almost EUR 700 million. Our operating result was up 11.5% to almost EUR 600 million. This yielded an operating return of more than 14%, compared to our targets of up to 12%.
The strong capital position, combined with the business operating results and the return on equity above target, show the strength of our franchise. This performance triangle is key in assessing how well we are doing. We are particularly proud on the combination of a solid return on equity, robust solvency, and low leverage. The operating expenses. Those went down by 1%, and this already includes the absorbed regular cost base of EUR 13 million from the business that we acquired, as well as IPO-related cost. The focus on continuous expense reduction delivers results. In our non-life segment, we've been able to keep the combined ratio in the 95% range ahead of target of 97%. Please bear in mind that this combined ratio already includes the significant claims from hail and water damages in the first half year of 2016.
We notice recent comments in the industry on exposure to bodily injury claims. We have only limited exposure to this, with reserves amounting up to almost EUR 400 million, i.e., less than 2/3 from our annual operating profit. We do not expect any adverse development from this. Our business generated more than EUR 300 million organic capital in 2016. This is in line with the guidance we gave at our IPO and is still based on our original investment return assumptions. For your reference, in 2015, we generated EUR 264 million of organic capital. When we also take into account the additional capital generated by excess investment returns and operational efficiencies, the total capital accretion amounted to EUR 475 million. Chris will come back to this later. This last number is, of course, before dividends. Our strong solvency position enables us to remain entrepreneurial.
As we have said, everything above 160% makes us to be entrepreneurial, to pursue profitable growth and pay an attractive dividend to our shareholders. Talking about dividend, this is significantly up too. Driven by the strong financial performance and confidence that we have in our business, we have decided to raise the dividend to EUR 187 million. This is up from the EUR 170 million last year and also exceeds the guidance at IPO of a discretionary dividend for 2016 of EUR 175 million. Our strong solvency enabled us to also participate in the recent sell-down from the Dutch state. We purchased 3 million shares in this transaction, roughly EUR 66 million. As already said several times, at the upcoming annual general meeting, we will request a new market-consistent mandate to buy back our shares.
This may provide us the flexibility to participate in further sell-downs by the state. Let's now turn to our business portfolio and show some strategic developments during the past year. This is slide three. In 2016, we have also made considerable progress in executing our strategy and optimizing our business portfolio. I'm sure you're familiar with this matrix in which we plot our businesses, and this slide highlights some important developments and achievements. In the top left of this slide, in Box A, are our businesses that provide stable cash flows, and here we focus on organic growth. Disability is a key product line in this segment, and a proposition that combines disability and health, the so-called doorgang proposition, is gaining traction. An advantage from this combined offering is the increased retention levels of the profitable disability product.
Both in disability and P&C, we were able to grow our premium levels. Market share data is not available yet, but we believe that we have been able to grow our market share significantly, whilst at the same time maintaining a very healthy combined ratio. Our funeral business has successfully completed the integration of AXENT. This was executed very well ahead of schedule, and we've been able to absorb the business with minimal additional headcount. When we acquired the business, headcount was 62 FTE, and we now run the same portfolio with only 29 FTE. Now it is integrated, we actually can achieve the cost benefits from the migration to our low-cost platform. Our funeral business will now turn into integrating NIVO, which was acquired at the beginning of last year. This should be done at the latest in the first quarter of 2018.
Further on, we remain interested in funeral books. However, it may take some time before a book becomes available. In the capital light space, that's in Box C at Slide 3, we have made further progress as well. We've acquired BNG Vermogensbeheer, and the integration and acquisition, in the meantime, has been completed. So the business today is fully integrated in our own business. Another example is the launch of the ASR Dutch Mobility Office Fund. In December last year, we bought the office portfolio from NS, the Dutch Railways. This portfolio comprised 15 offices, and nine offices were included in a newly founded ASR Dutch Mobility Office Fund. In February, we sold the other six offices. Clearly, we have strengthened our position in asset management and fee income business as promised at our IPO.
Thirdly, in this segment C, we are also pleased to see organic growth in DC business. This has been accelerated over the last year. Assets under management in DC more than tripled and sales doubled. And finally, the acquisition of SuperGarant and Corins have been completed, and together with the existing distribution activities, we expect this to show further traction in 2017. In Box B, the left angle down, are the large service books that we manage. Maintaining a low cost base is crucial. With expected declines, in particular in the individual life portfolio, roughly 50% in the next 10 years, we are verbalizing the cost base so that our costs keep pace with the declines of the book. We're on track in realizing the medium-term cost decrease. Finally, Box D.
As you already know, we also dare to take tough decisions in divesting businesses, and last year we divested SOS International and stopped our real estate development business and divested some of the real estate development projects. So finalizing this slide, the heart of our equity story is about capital generation, reflecting in an accretive solvency number, which we can invest in our business and to pay attractive dividends. We're very disciplined in deploying our capital in areas where our skills and expertise allows us to offer customers good value in products and services while achieving attractive returns for our shareholders. We are not capital hoarders. Let's now turn to the next slide of the operating results. The operating results increased EUR 62 million to EUR 599 million. Lower earnings in the non-life segment was more than offset by an increase of EUR 110 million in our life segment.
While the combined ratio remains strong at 95.6 in 2016, exceeding the target of 97, the operating result in non-life was mainly impacted by lower direct investment income, the hailstorms in June, and lower contribution from the equalization system in health. The lower contribution was, by the way, EUR 16 million. The P&C business performed well, including the absorption of the hail and water damage claims, which impacted us by EUR 25 million earlier in the year. The increase of EUR 110 million earnings in Life is primarily related to the positive contribution of the acquired companies, EUR 22 million, and a higher investment-related result on swaptions. The increased release of realized gains reserve compensated the lower direct investment income, as you may notice. The operating result of non-insurance activities showed a decline of EUR 15 million, mainly due to higher interest expense in the holding of EUR 17 million.
This is related to the issuance of the Tier 2 subordinated debt of EUR 500 million in September 2015. Acquisitions contributed to an increase of the operating result by EUR 8 million in the distribution and services segment. Let's now turn to slide five. Not only the full year went well, also the quarter-by-quarter developments show the strengths of ASR. I'm especially proud of the quarter-on-quarter combined ratio. This is already the 12th consecutive quarter that our combined is below 100. This, in combination with the organic growth of our non-life business, shows that structural underwriting loss-making non-life business is not necessary to grow our top line. On slide six, on cost. One of the key drivers of solid operating earnings and long-term value creation is our ongoing focus on cost, a discipline which has become part of our culture and daily operations.
In non-life, the expense ratio improved from 8.9 to 8.3. In Life, the expense ratio was also better, 11.7 instead of the 12.3 of 2015. All in all, operating expenses decreased from 575 to 569, a decline of 1%. This picture, by the way, is actually somewhat distorted by the acquisitions we've done in 2015 and 2016. On a like-for-like basis, that means including the full annual cost base of all of the acquisitions, the 2015 comparative cost level would have been EUR 604 million, and this would then result in a decline of EUR 35 million. We have been able to absorb the full cost base of the acquired businesses. Measures taken to reduce our cost base are fully on track and on target. Let's now turn to the non-life segment, which is on slide seven, and let's have a closer look on this segment.
In the non-life segment, our underwriting expertise is market leading. All non-life product lines showed combined ratios below 100, and we're proud of that. Even including the impact of hail and water damage claims, we've been able to keep the combined ratio in the 95-ish range, better than the target of 97. Also noteworthy is the favorable development in our expense ratio, as I already mentioned. Market developments towards more rational prices allowed us to both grow our top line with an overall growth of 6% in the P&C and disability business. Operating result in the non-life segment continues to be strong. The exceptional hail and water damage led to a specific claim cost of EUR 25 million after reinsurance.
We also, by the way, experienced an increase in the number of large claims, roughly EUR 7 million, relative to multiyear historic averages, which have been covered in part by reinsurance contracts, by the way. Even after those, our P&C combined ratio has remained strong without unduly relying on reserve releases. The underwriting results of the disability business improved. This is driven by growing business volumes, reflecting the recovery of the economy in combination with the expertise in claims handling, prevention, and reintegration. Our health insurance businesses reported lower earnings due to lower benefits in combination with higher claims estimation from the National Health Care Institute. In addition, we also experienced higher dentist claims for supplementary health insurance. The total effect amounts to a decrease of EUR 25 million, but still delivering at the IPO target of 99% combined ratio. Now let's turn to our Life segment. That's on slide eight.
As you may know, our Life segment comprises three major product lines, individual Life, which is 40%, pensions, which is 50% in terms of reserves, and finally, funeral insurance, which is 10%. Although by the very nature of the product, we would expect funeral to increase gradually in the future. Gross written premiums of the total life segments rose by 10% to more than EUR 2 billion. The decrease in the individual life portfolio was more than offset by the growth in the funeral business, including the acquisition of AXENT and NIVO. Our pension business due to the acquisition of the Eendragt. The DC pension also contributed to the growth, including customer switching as a result of the commercial integration of Eendragt. Single premiums in the life segment increased by EUR 162 million to EUR 734 million.
The increase includes the transfer from the funeral portfolio of NIVO and the pension contract for AstraZeneca. New business went up by EUR 60 million to EUR 152 million in 2016. Excluding NIVO, by the way, the underlying growth of the life segment was EUR 8 million. In the pension business, the shift from capital-intensive defined benefits products to capital light products is making progress. We, as already mentioned, noticed a doubling of new business. From an earnings perspective, and I'm on slide nine, the life segment is a major contributor to the overall group earnings. The operating result rose by EUR 110 million, driven by higher income from, first of all, the realized gains reserve, shadow accounting, from a higher contribution from acquired businesses as well, and from a higher investment-related result on swaptions whose gains also feed through our capital gains reserve.
We also benefited from some portfolio management decisions, Chris will discuss those later. Important to note, as the bar chart on the top shows, the lower direct investment income is offset by higher regular contribution from the realized gains reserve under shadow accounting. This shows the stabilizing effect from the shadow accounting method under our interest rate hedging program. Operating expenses in the life business, including the additional costs of acquisitions, which by the way, were eight in the life segment, decreased by EUR 2 million to EUR 303 million. Due to the successful integration of the acquired businesses into ASR ICT platform, we were able to capture scale benefits. As a result, the cost premium ratio improved 6 percentage points to 11.7%. During 2016, further steps were taken to achieve cost savings ambitions. As discussed, this includes the migration of several product and system combinations into a new single platform.
Overall, the life segment delivers very good returns. The operating return on equity increased to 11.9%, while the life insurance margin rose to 3.7%, being 3.4% in 2015. Turning to slide five, to the various activities in non-insurance. These are performing broadly in line with expectations. In the distribution and services segment, we have acquired SuperGarant and Corins, and we expect them, together with the existing distribution entities of Van Kampen Groep and Dutch ID, to gain further traction this year. In the banking and asset management segment, the acquisition and integration of BNG Vermogensbeheer has been completed and showed early success in winning an asset management mandate of EUR 1.7 billion. The operating result of ASR Bank was lower than expected, reflecting actions to further improve the organization. Now let's move to the comparison with our IPO targets.
I believe in the past year, we have delivered the proof that we are executing our strategy diligently and consistently with our equity story. One quote in the reports we've seen this morning summarized it even better. The quote was, "ASR continued undisturbed on its path of over-delivering on its IPO promises." We couldn't have said it in a better way. Our financial results are strong and profitable, and our balance sheet is robust. As I mentioned in the beginning of the call, make no mistake, it is hard work to get these results, and these targets are challenging in the Dutch insurance environment. I'm confident that our ambition stands high in any fair comparison in the Dutch market. On slide 12, to finalize, year-on-year, we achieved better financial results, driven by strong business performance.
The steady increase of operating results has enabled us to also increase the returns to our shareholders. Over the past years, we have built a solid track record of paying dividends. The proposed cash dividend of EUR 1.27 per share is an increase of 12% compared to last year. The return of cash to shareholders is also underpinned by our recent share buyback of 3 million shares in the sell-down of the Dutch state. This year, in 2017, a new dividend policy has become effective. The annual dividend will be based on a payout ratio of 45%-55% of the net operating result attributable to shareholders, i.e., net of hybrid costs. We apply, by the way, as you know, a boundary condition based on our Solvency II position, where we would not consider to pay a cash dividend should the Solvency II ratio fall below 140%.
The proposed dividend of EUR 187 million is fully in line with the new dividend policy. Now for more financial detail and further information on our Solvency II position and capital generation, I will hand over to Chris. He will continue to build momentum towards slides 22 and 23.
Very good. Thank you, Jos. Over to the financial update and continuing on the momentum that characterizes our fund, or our stock actually, for our company. We will move from page 13 to page 25. I will take a few pages that I will talk shortly about and a few pages that I will elaborate more. Starting with the financial update, that is page number 14. As you can see, the increase in operating results, whether we look at the IFRS results, which is up 4%, or the operating result, which is up 11.5%, we are seeing an increase in results. Details are in the appendices A to E. Just a reminder on this page, the difference between operating result and IFRS result are the capital gains and incidentals.
On the investment side, we this year had a more normal year in terms of capital gains of around EUR 170 million, which is kind of where we were in the long run in terms of capital gains. You might argue it is slightly lower than normal because we traded a bit less in our land portfolio. Last year we had exceptionally high capital gains as we rebalanced our equities portfolio. In terms of incidentals, last year we had a negative EUR 93 million incidental relating to the provision for real estate development. This year, we had EUR 100 million-plus positive incidental from the finalization of the modernization of our pension scheme, where we bought off the previous unconditional inflation commitment. This elimination of inflation exposure led to EUR 100 million reduction of the defined benefit obligation, so a plus. Overall, IFRS result up about 4%, operating result up about 11.5%.
Operating ROE this year, 14.1%, similar to the 14.5% operating ROE we had last year. Actually, the decline in the ROE was solely due to the higher base. In the appendix C, you can see the ROE calculation. The denominator in the ROE calculation moves up from EUR 2.5 billion to EUR 2.9 billion. The small decline in ROE is simply because the denominator went up. Had we had the same denominator as last year, our ROE would have been 16.16%. Moving on to the investment portfolio, page number 15. The portfolio increased in value to about EUR 57 billion. Details are provided for in the appendix H and I. There you can find breakdowns by asset class and by sub-asset classes. Couple of points to make.
During the year, we made a set of tweaks to the portfolio to further optimize our return on capital, especially in a Solvency II context. We basically continued to rotate out of equities and shifted to credits to mortgages and real estate. In this move, the direct investment income of our book went actually up 3%, despite low interest rates. The direct yield, the direct IFRS yield so to speak, is about 2% to 3%, 2.5%. If we include release from shadow accounting in the capital gains reserve, the direct yield is still safely above 3% and appears to stay there for the plan periods. Highlighting mortgages and real estate, our mortgage book is now 18%, 18.6% of the total asset base. Gross mortgage production was EUR 1.3 billion. The net increase, EUR 700 million in mortgages. Our book is now 50/50 split between government guaranteed and non-government guaranteed.
We'd like to point at about 75, three-quarters of our mortgage book either has a government guarantee or loan to value of below 75%. Given this high-quality nature, you understand that the performance is good, that the book is developing very healthily. The arrears numbers, mortgages in arrears, 90 days arrears are less than 1.5% of the mortgage portfolio, and the actual default or foreclosure costs are less than one and a half basis points. A very healthy, solid mortgage book. There's actually quite a lot of client demand from institutional investors wanting or desiring to invest in our mortgages, we're turning it into a mortgage product. In terms of real estate, Jos already alluded to the fact that we acquired the office portfolio of the Dutch Railways, EUR 275 million. Effectively, we warehoused these assets over the year-end.
We bought them in December, they're on our balance sheet. In January, we sold EUR 60 million of non-core assets. We placed EUR 20 million already in funds for third-party clients. The remainder, about EUR 200 million, will be part on our own book and part managed for third parties in a new fund. Please know that EUR 275 million office acquisition in real estate is actually a step towards another asset management, real estate asset management solution. In real estate, please note about 40% of our real estate portfolio is in land. As people say, they don't make that stuff anymore. We're very pleased with our land business. The yield after vacancies in our real estate portfolio is 4.3%.
For those of you with a more black perspective on the world and on our risks The exposure to Italian banks is EUR 130 million, all in fixed income, all in what we see as high-quality institutions. Exposure to Monte dei Paschi is EUR 7 million, only in a senior bond. The amount of direct risks from some of the remainders of the crisis is very limited. Finally, we made a number of changes in our liquidity portfolio to deal with swap spread exposure and to lock in our swap spreads, we'll talk more about that when we get to page 25. Let's first turn to page number 17, when we kick off the discussion on capital. Key developments in cash and capital. This is one of the pages that I'll talk about very shortly because Jos has gone through the numbers.
Solvency II standard model 189%, post the foreseeable dividend. It was to 194% pre-dividend. Compare that to 180% in our day one report, means a total accretion of capital since the beginning of the year of 14 percentage points. Organic capital accretion came out at 9%, in line with the IPO guidance. During the year, we had numerous comments and suggestions to harmonize and update our definition. We'll talk about it later. Important to mention that the 9% is pretty existing, the old methodology. We met our targets not by changing the model, but by delivering on our goals. Dividend at EUR 127 per share.
If you add back the cash dividend, or if you add up the cash dividends of EUR 187 plus the EUR 66 million shares we bought back, or value of shares we bought back in January, the total cash return since IPO is EUR 253 million or 8.7% of the IPO valuation. We hope and trust that those who had confidence in our stock at IPO were duly rewarded. Moving on to page 18, continuing to build momentum on SCR development. Some people call me old-fashioned, but I sometimes like to look at book values. In this chart, you can see the IFRS equity book values and the solvency eligible own funds. IFRS equity moved up from EUR 4.2 to EUR 4.4 billion. If you exclude the hybrids from EUR 3.6 to EUR 3.8. An increase in book value ex hybrid to about 6% over the year.
If you look at the equity base that we use for ROE calculation, which you can find in the appendix, moved up by 5% during the year. Eligible own funds up to EUR 6.3 billion, excluding hybrids moved up by 4%. If you take different book value lenses, whether it's solvency own funds, whether it's IFRS, taking or excluding hybrids or with and without realized capital gains, by all means the book value of the group went up. We think in the long run, despite volatility, in the long run, book values are a good guidance for the development of any company. We're pleased with the continuous accretion of funds of book value. Page 19. Group solvency figures. Page 19 depicts the own funds under required capital. Eligible own funds of EUR 6.3 billion, required capital of EUR 3.3. Divide one by the other, you get 189% post-dividend solvency.
Couple of points mentioning worth on stock. Tier 1, as Jos said, 84% of total funds. Tier 1 ratio alone, 158%. A pretty solid construction of solvency. Significant headroom available. We've noticed some discussion in the market around Tier 3 and Tier 3 capacity and tiering risks. Our DTA of the group is EUR 11 million, which compares to the total gross headroom of EUR 501 for Tier 3 capital. The net headroom in Tier 3 capital is EUR 490. EUR 501 minus EUR 11. That means even if our DTA goes up, even if interest rates move to Tier 3 capacity or Tier 3 tiering is not at risk. Of course, if you were to use the Tier 2/Tier 3 space, you need to think about your ability to absorb any changes. At this point in time, no tiering risk to the solvency of ASR.
In terms of eligible own funds, in the year, we absorbed a decline in the volatility adjuster, which moved from 21 basis points in the beginning of the year to 18 for half year and 13 at the end of the year. Effectively a drag or a headwind of around nine points in solvency. When you compare the 180 day one to 194 pre-dividend, please note that is after absorption of a nine-point VA drag, so to speak. Just give a bit of color around the underlying accretive capacity of the group. Double-clicking or zooming in on the required capital. In the appendix, Appendix G, we've got a little bit more intelligence or data on the sources of the change in capital. If you go on market risk, our market risk is still 49% of the pre-diversification required capital.
Where we want it to be, remember, we're an insurance company, not an investment fund. We believe that market risk should hover around the 50%. Could be a bit above, a bit below, but not too far off. 50% is a number we feel very comfortable with. Inside the market risk buckets, during the year, we lowered allocation or capital allocation to equities, we lowered capital allocation to currency risks, and we increased capital allocation to spread risks and to real estate risk. That actually is a reflection of the portfolio choices we made. There was a slight increase in allocation to the long-term interest rate shock, which is a technical phenomenon. As the curve changed during the year, the curve steepened during the year. The interest shock, the capital charge for a shock in Solvency II went up.
By and large Out of equities, out of currencies, into corporates, credits, and into real estate. That's the delta behind the market risk number. Other capital components, the life risk charge went down for the year. In the life bucket, we had an increased charges for longevity, mainly, which is a second-order effect from the change in interest rates, offset by reduced lapse risk. Remember our mass lapse insurance. Due to lower cost risk, which is a benefit from the integration of AXENT. In life, an increase in longevity more than offset by a decrease in lapse and cost charges. Counterparty risk went up due to the allocation of mortgages and LAC-DT, a small support to capital from delta and LAC-DT. We'll come to talk more about it later and no doubt in your questions.
In summary, when you look at the capital requirements, we had reductions in required capital or an increase in availability of capital, if you wish, through reduced charges for lapse risk, cost, equity, currency risk and LAC-DT. We allocated more required capital to spread risk, real estate risk, counterparty risk, and in our business P&C, to some extent, longevity risk. Please note, real estate reflected the EUR 275 million effect of warehousing we did on real estate for the office fund. In summary, looking at our numbers, we believe we've got a rock-solid solvency number, strong tiering, no tiering risks. We are pleased with the level and the quality of our solvency.
Again, from our perspective, well-controlled, measured developments in the underlying solvency components, where we continue to assess the sources and uses of capital to optimize our balance sheet and to provide good returns to all our shareholders. Page 20, organic capital creation. Let's go into the delta of our solvency development. As per the hub here, we'd like to break down the delta solvency and underlying components. One word of caution here. Any breakdown in the delta solvency is judgmental by its nature. Right? The insurance industry is still trying to find stable ground here. We aim to run at the forefront of capital disclosure and share how we think, any bucketing of delta capital has an element of judgment to it.
We'll follow on in the approach we took last year by defining organic capital generation, organic capital creation in three buckets: operational, net release of capital, and technical movements. The remainder, the category other, is called market and operational developments. Let's start with the technical movements, work our way from right to left. The box technical movements contains the UFR unwind and equity transitionals. The UFR unwind for the year was EUR 110 million or about 3.3% of SCR. It includes this metric, the equity transitional, the amortization of the transitional rule for equities. After that specification is about 0.7 SCR points. The total technical movement, the technical drag is 4%. Please note, again, this has nothing to do with management skills or whatsoever. This purely is a technical shift between stock and between flow.
It's almost like running up on a downward moving stairway on a running escalator. You'd have to run faster than the stairway to make progress. The annual drag from this point was about four percentage points in the last year. Second bucket, net release of capital. This consumes or contains the release of SCR, the release of risk margin, and investment in the new business. The resulting number here is 5.7%. Think about SCR minus new business and risk margin of equal size. The 5.7 you can divide into two. Half of it is the release of the risk margin, and half of it is release of SCR minus new business investments. The SCR release was tilted up a bit because the lapses on our nominal life book have moved up a bit during the year.
In our country, people redeem or have redeemed their mortgages more than they used to do. We used to run at unnatural, unexpected lapses, about 50 basis points a quarter. That has moved up to 60 basis point a quarter and has been stable throughout the year. Some acceleration of release of capital through the redemption of mortgages, but again, pretty well sustainable. 5.7% in terms of net base of capital, 50/50 between risk margin release and SCR minus new business. We've got the operational capital generation that is comprised of excess returns, technical results, the fees, and then holding costs and hybrid costs. In total, 7.2% of day one capital. With excess returns the largest component and the technical results exceeding the holding cost.
This gives an approach where our business generates about 7% of capital plus a release of capital of 6%, eaten up partially by a 4% technical drag between stock and flow. Measured in EUR, it gives EUR 301 million of organic capital creation or 9% of day one solvency, which is in line with our guidance and expectation. The EUR 301 you can break down into own funds and SCR charges. On average, EUR 230 million in increase in own funds and EUR 38 million in lower SCR charges. The bucket other, market and operational developments, it's for the second year in a row, it's now a plus, so that we outperform, we add, or we have added for the last two years over and above the organic capital creation. In this bucket, you've got a number of pluses. You have excess returns over and above the assumption in the OCC.
The cost benefits lapse insurance, LACDT positives are a plus. Negatives would be the decline in the volatility adjuster. Negatives would be increase, for example, allocation to interest rate and to real estate. Finally, some modeling changes, but all modeling changes together have are basically cancel out. Where does this leave us? Compared to the model we choose at the beginning of the year, we have delivered on our guidance, 9%, EUR 300 million capital creation up from EUR 264 last year, in what was not an easy environment. Second important point to note, the operational capital generation exceeds the release from a book. The business generates more than the release from the book. This is the way we manage our company. We are a book, a business about capital generation, not just capital release.
In the way we manage our company, the factor other was again, a plus over and above the OCC. To give you a little more color, we have moved to page 21. You can see the movements in numerator and denominator, the delta in own funds, and the delta in required capital. I will not spend too much time talking about it's more for your perusal. Again, you can see about EUR 230 million net increase in own funds and a EUR 38 million benefits in required capital. If you multiply the EUR 38 by the average solvency of 1.84 during the year, you get to the EUR 301 OCC. Again, here it shows that operational capital generation EUR 242 is the largest component of what we deliver in terms of organic capital increase.
Finally, please note market and operational developments. The required capital element of that is only EUR 2.
EUR 2 million in required capital for market and operational developments. That is kind of actuarial speak for there were no major net modeling changes to speak of. In terms of modeling changes, anything at all canceled out and did not lead to a massive increase of the required capital. Page 21 elaborates further on the organic capital generation. Cap gen going forward, I would ask you to move to page 22, the definition going forward. Seth, in our interaction with investors and analysts, we've got lots of feedback on the way we calculate the OCC. We have been challenged if we were not too conservative, especially in the assessment in the long-term spreads. We have conducted a very thorough analysis, also used some external support and looked at UFR and VAs, traditional spread assumptions and what have you. With that, we've updated our models.
Again, not with the purpose to meet our goals, but with the purpose to be market consistent in our assessments of organic capital generation. After the assessment, we've concluded better to move the transitional rule for equities like others do to the bucket market and operational, and we've adapted our spreads. Most notably, we moved up the spreads on mortgages, equities, and real estate. We've stratified our bond spreads. We are different, differentiate between core govies and non-core governments. Also, let's be true and responsible, liquidity does come at a price. We observe some industry participants plot a zero spread for government bonds. Well, in this day and age, that is not realistic. Government bonds do provide a drag compared to the solvency curves. We estimate for the medium term, spreads on core sovereigns are -20 basis points.
Using this, we get a refined number of EUR 348 million for capital generation in 2016. So on a market consistent methodology, market consistent way of measurement, we get to EUR 348 million of capital. The delta, think of it like this. Take the starting OCC of EUR 301 million, add about EUR 40 million from excess spreads, deduct EUR 15 million from the negative drag from government bonds, and add EUR 24 million on a full year basis from the reclassification of transitionals. That will give you about EUR 348 million to EUR 350 million for the year. Going forward, we will work with this definition. The long-term investment margins are what they are, long-term investment margins. So we tend to keep them stable from now on. The one thing we will continue to assess is, of course, the government bond spread as interest rates and swap curves move.
At this point in time, it is fair to assume that there is a drag on anyone, any insurance company holding government bonds. That is roughly the price for holding liquidity. Page 23. An alternative view on capital accretion. As we said before, OCC, organic capital creation, is just one way to slice and dice your delta and solvency. In reality, of course, this is a number that hinges on an ultimate assumption, on assumptions you make on spreads. We have shown you how we define it. We want to be fully transparent, but the definitions make the number. The number that cannot be changed is the number at the beginning of the year and number of the end of the year. Those are hard numbers, are audited numbers.
In order to give full disclosure, and we would like to lead the pack here, we have also provided you with an alternative view on solvency developments, namely through sources and the uses of SCR. Again, this is the number based on the audit figures starting of the year, ending of the year solvency. And if we include all the relevant elements, the sources of capital were EUR 674 million and the uses of capital, EUR 386 million, out of which EUR 241 million was returned to capital providers, namely dividends EUR 187 million and hybrids EUR 54 million. So EUR 674 million minus EUR 386 million gives an accretion after dividend of EUR 288 million, and the total returns to capital providers included in this is EUR 241 million. Why do we believe this model is important? Because it reflects the way we run our business. We strive to outperform the long-term investment margin.
Excess returns do not show up in OCC, they show up in the bucket of other. And the bucket of other may therefore sometimes have a structural component to it. So we strive to outperform the LTIM. That is what we are mandated to do, and it shows up, of course, in sources of capital. Secondly, we run a life book that effectively closed grows through M&A. Post M&A, we restructure the acquired business. We take out costs. When costs are out of an acquired business and taken out, that adds solvency, not through OCC, but in the bucket other, namely to lower costs. For example, the integration of AXENT, and we expect the integration of NIVO, will lead to cost savings that will show up in the component of other. And actually, it is a source of capital.
We believe assumptions, changes of business developments do contain some things which are the heart of our business model, but it cannot be captured in the OCC number. That means that the capital accretion of the group bucketing source of use of funds is a good way of looking at how this business develops, how we run our shop, and it was EUR 288 million or EUR 475 million pre-dividend. As you remember, in January, we used another chunk of this, buying back EUR 66 million worth of shares. We think it's only fair to complement or accompany traditional capital generation numbers with capital accretion sources and users of funds. I understand you all are getting tired, it's a few more slides to go. Hang in there. It's almost like a game of cricket. Once you understand the rules, you're ready for tea.
Two more pages to go. Interest rates. As you can see on this page, impact of interest rates on stock and flow. Interest rate sensitivity is limited, stable, not so much because we changed our hedging policy, but because the increase in interest rates in the last half year, last month of the year, reduced the convexity of our business and the convexity change reduced interest rate sensitivity. More importantly is actually the sensitivity of SCR to a lower UFR. On this page, you can see the solvency ratio, but different levels of UFR. 189% is where we are today. 178%, it used to be or 178% at 3.7, the number that used to be contemplated by EIOPA and a number of other figures. Please note the bottom end of 2.2.
In our industry, everybody, participants, regulators, are all struggling to define what is the long-term across the cycle UFR. What's the right rate to use? There's a long debate whether 4.2 is actually a relevant number to plug in as a UFR. Internally, we started to take another view, say, what if we plugged in the long-term investment yield that we're making today? The basic idea is you should not discount your long-time liabilities at a rate that's higher than you make today, because then you eat up your own solvency, or at the rate that's lower than you make today, because then you understate your solvency, i.e., what if we plugged in a UFR that's related to your investment yields? Would you then still have a solvency that's safely above 100%? The IFRS yield on our book is around 2.3%-2.5%.
In today's market, the long-term direct yield is probably somewhere between 2%-2.5% ex capital gains. Of course, if rates move, this moves up. What we did was we plugged in a 2.2 UFR and asked ourself the question: would at that level, which is somewhere close to where the long-term direct yield is, would we still be safely above 100%? Because that means we could freely distribute cash or invest in future ventures. Again, we found a solvency level after SCR shock, after tiering, of 142%. That means with this level, we can have responsible, thoughtful financial management and feel confident on future distributions. We have developed a fairly advanced set of modeling technology to analyze and test this and play around with different numbers.
We will adjust it if rates move, for us, a UFR that is linked to your investment yield should give you a more economic view in the standard model. Although we understand that may be a contradictio in terminis, the economic view in the standard model is what we strive for. We believe this is the way forward for the industry. This is not our formal policy, not an internal model, but a way to think about economic solvency. Also on this page for your perusal, and by popular demand, we have plotted the impact of different UFR levels on capital generation. As you can see, lowering the UFR to the EIOPA ambition level would reduce our solvency a bit, but also make us still stay above the hurdles and also increase the annual flow, the annual capital generation.
Here you can see the move between stock and flow at various solvency levels. Needless to say, strategically, we wouldn't mind if the UFR would be lowered a bit. Again, you can also see the impact of government bond spreads at a EUR 7.7 billion core government bond book, a spread in the OCC of 20 basis points. The cost of interest rates is a EUR 50 million drag a year. That is the cost of liquidity. Page 25, some final observations on LAC-DT and on swap spreads. A lot has been said on LAC-DT. I guess a lot will be said on LAC-DT. A few points by us. Our regulator has put out new guidance in February about how to think about LAC-DT and how to model it.
Remember last year, this time around, we as a group marked down our LAC-DT considerably out of prudence, out of anticipation, and we wanted to limit the dependence on future fiscal profits. We further developed a model during the year. We received feedback on our day one model and moved on. To the best of our abilities, the DNB guidance that is supposed to be implemented by June has been reflected in our models. Some points may require some further clarification, but our models appear to be fully in line. We do not expect any major negatives. Actually, some points could be a small positive, depending on how we interpret some of the more complicated elements of the guidance.
For full year 2016, we've got our LAC-DT on the life business at 60%, non-life at 75%, basic health at 0%, supplementary health at 25%, which is in line with the model that we used, and the model, as far as we can see, is consistent with DNB guidance. It has very limited use on component four on future fiscal profits and is robust against fiscal year 2015, historical fiscal years falling out of the equation. We believe when it comes to the complex world of Solvency II in combination with taxes, it's better to be lucky than to be smart. We believe when you plan well, when you anticipate well, you increase the odds of being lucky. This development shows that in the industry, in its first full year of Solvency II, things are still in discovery mode.
From our perspective, LAC-DT has been implemented and very limited to no material downside. On swap spread hedging, please note in Solvency II, your liabilities are discounted on the base of swaps, and the assets comprise a large of investment that are not swap related or at least priced on the base of a government curve, i.e., an almost EUR 8 billion of government bond portfolio or liquidity book. In this space, the swap spread widening, which as we have seen in the last years, has supported solvency across the industry. Also, we have benefited from this. Given market developments, we have decided to lock in some of these benefits while maintaining the liquidity thresholds in our portfolio. On a relative basis, you sell government bonds and increase swap exposure. In total, we trade about EUR 4 billion in transactions.
Effectively, we sold long-dated core government bonds, buying back short-dated credits, short and medium-term government bonds from France, Belgium, Spain, and Ireland, plus some of the receiver swaps. With this, with the intent to lock in the swap spread benefit. At this point, and after a few last weeks in January, we believe the swap spread exposure of the group has been halved. That means at least half of the swap spread benefit has been locked in, and the swap spread exposure has been halved significantly. We think it is a way to be ahead of any changes in interest rates, ahead of any changes in swaps developments. In summary, Solvency II is and will be a complicated world. It has market value effects, tax effects, second order effects all playing a role. In our risk management, we aim to identify opportunities and threats early and anticipate.
This means that our group was well prepared for any changes in LAC-DT and is well prepared for any changes in the world of swap spreads. Finally, page 26, the numbers. I will not repeat them to you anymore. We hope this presentation has shown you that our performance in 2016 has been strong on nearly every metric. We delivered on our promises, and especially we're proud of the combination of operating performance, a solid ROE or market consistent capital generation or capital accretion. Very pleased with solvency levels up to 189% standard model with no tiering risk. Not just capital gains, underpinned by strong technical results from our business, the combined ratio in the 95% range, and low financial and double leverage, which you'll find in the appendix.
Again, the increased dividend to EUR 1.27 per share shows the confidence we have in ASR's operations and our willingness and ability to share the good fortunes of our group with our shareholders. That concludes our presentation, giving back the floor for questions.
Ladies and gentlemen, we will start the question and answer session now. To be registered for the question and answer queue, please press star one. Star one for your question or remark. Go ahead, please. The first question is coming from Mr. Cor Kluis, ABN AMRO. Go ahead, please.
Good afternoon, Cor Kluis, ABN. A few questions. First of all, on disability insurance, the legislation, can you already give a kind of idea what the impact might be on the premiums for 2017 and 2018? Because last year, your disability premium went up 4%, but this could be somewhat more material positive. Second question is about the internal model. Could you give an idea what the impact would be if you put your own internal model? How much higher would the Solvency II ratio be then? A third question is about the EGM. I thought you asked for a share buyback request of a maximum of 10%. Given the share overhang and your very strong capital position and cash flow, why are you not asking for a larger share buyback possibility? Last question is about, it's more a technical thing, the size of the realized capital gain reserve.
What's the size at the end of 2016 of that figure? That were the questions.
Thanks, Cor. Let me start with the question on disability
It is too early to give final guidance on how the season went. A first view on it is that a lot of smaller companies decided to return to the public system. In terms of number of customers that returned to the public system, we've seen more going to that system than expected. In terms of premium, in terms of new business, we see a neutral effect until today. Probably we will not fully meet the expectation we had for the medium term. This was only the first year, and we think it may require a little bit more time to meet the top-line growth target. Those are only the first views on it, because numbers are not final yet, and new business is still coming in.
The main reason for that is that we have kept to be disciplined in our underwriting and premium. We could have done more if we had wanted, as we have done before, said to the disability management, "You are allowed to do as much business as you want, as long as you stick to the underwriting principles." We have seen a fairly disciplined market, not everybody was as disciplined as the market in general. Some participants have been fairly aggressive, we've decided not to take part in the aggressive pricing. That's on disability. The question on why don't we ask for more than 10%. In our view, we have every year an AGM, we can every year ask for a new 10%.
In our view, there should be a balance between the year-on-year generated capital and the capital return. From our point of view, given the developments in solvency, the uncertainty where EIOPA ends up with the UFR, we have said, the total return of capital should be balanced with the capital generation. To our opinion, 10% should be sufficient on a year-on-year base.
Okay.
On your question, Cor, on the internal model, where do we stand? We do not have an internal model. An internal model is a pretty complex beast. The closest thing we have is an ECAP model. The ratio of the ECAP model is 226%. Please note, this is not a model that's gone through the same rigorous validation process as the Solvency II standard model has. On your question, will we move to an internal model? At this point, we have no plan. Reason is, we believe the regulator will always look at both models when it comes to distributing cash and capital to shareholders. If I look at the banking sector, I'm not sure the banking sector will provide guidance, there we can see like a harmonization or internal movements and obviously internal model and standard models move towards each other.
Internal models with a floor, I find an internal model with a floor just becomes a very expensive version of a standard model. We believe at this point in time, it is not sure whether it is the best way to spend shareholder money to go through all the lengths in validating and approving the ECAP model if the solvency is what it is, because we are already at pretty safe level. You had a fourth question, I cannot even read my own handwriting. What was your point again?
The realized capital gain reserve.
Realized capital gain reserve at the end of the year is about EUR 3.6 billion. Actually, it was still a net addition to that reserve. To give you some color, we believe that the release from that realized capital gains reserve in the plan period, assuming rates stay relatively where they are, will be similar in the next three years as it was last year. If I look at the amortization schedule, amortization pattern of that EUR 3.6 billion, if rates stay roughly where they are today, the contribution of that will be same when actually the plan period as it was last year.
Okay. Very clear. Thank you very much. The next questions come from Mr. Albert Ploegh, ING. Go ahead, please.
Yes. Thank you for taking my questions. I've got basically a few on the capital generation. First one to be clear on the new methodology on slide 22. You mentioned you've moved the transitional equity rule to the operational variance and market bucket, so to speak. First of all, I thought it was something like EUR 45 million per annum. Is it then correct that in the EUR 348 that's printed on the slide, on the new definition, that it is then not including that drag, while the EUR 301 million did include it, so to have that at least clear. The second question I have on the capital generation is a little bit on the non-core sovereign bonds at the minus 20 basis points. I think on slide 24, you mentioned that basically has a drag of EUR 15 million or so on capital generation.
How to square that with the actions taken to basically lock in the spread, because that you also mentioned in your opening remarks. Is basically the starting point for 2017 already meaning that maybe that the EUR 15 million drag is already reduced by 50%? I'll leave it for now for there.
Okay. Albert, on the transitional rule, in the movement from EUR 301 to EUR 348 with a plus of about EUR 23 million from the transitional rule, that was moved out. There was a negative in the EUR 301 that no longer occurred in the EUR 348. It was reduced during the year. Two effects. One is diversification kicked in. Due to the portfolio developments, the impact of the transitional rule post-diversification was a bit less. Secondly, we sold some equities. We reduced equities. Part of the equities that we actually divested were subject to the transitional rule. There's less to amortize, you've got less equities. Partly it's technical. It's a diversification effect. Secondly, the equity base that was subject to transitional was lower. In terms of non-core spreads, the EUR 15 million drag is one going forward.
It's actually a bit less than last year, I agree with that, because the government bond portfolio will decline. Remotely, EUR 7.7 billion is the portfolio that we have roughly where we are today. That's a small decline. In terms of what does the swap spread hedge do, the swap spread hedge basically is not so much to lock in the spread, but lock in, I would say, the delta in these spreads. I mean, the spread widened in the last years, the swap spread. That supported our solvency levels. I mean, you discount your liabilities at a higher rate than you discount your assets. We wanted to lock in that benefit. That means if swap spreads reverse and the spread declines, you don't lose that benefit.
The swap spread change itself does not do too much on organic capital creation, but it aims to protect the stock of solvency that we have.
Very clear.
The next question is coming from Mr. Steven Haywood, HSBC. Go ahead, please.
Thank you very much for the presentation. Just a couple of questions. Can you go back to your core and non-core sovereigns? I know obviously one is now minus 20 basis points excess spread, and the other is plus 50 basis points. In reference to slide 36, can you define which bonds are core and which ones are non-core? Secondly, when you talk about the UFR changes on slide 24, I just want to know your opinion about what you think is most appropriate to use, whether you should use the direct yield or the total yield, including gains. Whether you'd use the 2.2% or whether you'd use over 3% as your assumed UFR. Thank you.
Core, we define Dutch and German government bonds as core. All the remaining in Europe is actually non-core. Dutch and Germans are our core, the rest is non-core. In terms of what is the right level to use, the direct yields, which is coupons, dividends, rents, et cetera, it's between 2%-2.5% ex-shadow accounting release. The actual yield that we make on a portfolio is larger. We made last year, EUR 170 million capital gains. If you look at the average of the last three to four years, it has always been in the EUR 170 million-EUR 200 million range of capital gains. The actual yield one makes is over 3%. If you think about a fully economic UFR, probably a number over 3% is justifiable. However, capital gains fluctuate over time. You may have a bad year in which there are capital losses.
We felt from a long-term prudent perspective, we'd like to work with direct yields. We could understand that it's arguable that you could move it over to over 3%, including a fair amount of capital gains. That's a judgment call that one can make. We believe in terms of being prudent and being fair to our policyholders, the direct yields is something we can observe, and barring defaults, that will happen year on year on year. Actually, you can bank on that. Your point is valid. You could argue that the yield you make, the return you make, is probably above 3%.
Yeah. That's great. Thank you very much.
The next question comes from Mr. Farquhar Murray, Autonomous. Go ahead, please.
Hi. I think you may have given this on the call, but when was the broad timing of when you did the swap spread lock-in, i.e. the EUR 4 billion transaction? More generally, has that increased your spread sensitivity to Belgian and French sovereigns? I presume it does. Actually, are you able to give the kind of sovereign spread sensitivity overall for the group? Then finally, can you give any indication on how Solvency II has developed so far this year, given we've seen some quite significant sovereign moves overall? Thanks.
Farquhar, thanks. In terms of when did we execute the swap spread change, in steps since September. It was actually done in Q3, Q4, and the remainder actually in January. It was a series of trades. Honestly, we started doing this by shorting 30-year bond futures. It's the most efficient and liquid way to doing it. We found liquidity in the market was too limited. At some point, we owned too big a share of that market. We moved to peripherals cross swaps. It did increase exposure to those non-core bonds. Page 36, you can see the holdings in French govies went up from EUR 800 million to EUR 1.4 billion, Belgium from EUR 600 million to EUR 1.2 billion. There's an increase of almost EUR 800 million-EUR 900 million in government bonds from Belgium and France.
We still feel very comfortable with holding those government bonds, especially if they have a midterm maturity. We do not yet disclose sovereign spreads sensitivities, something to pick up and to think about going forward. There's no issue around it. We just haven't disclosed it. We don't really track it that much. You can see on page 36 the changes in the portfolio.
On your last question, the development of solvency during the first six to seven weeks of the year, without giving any numbers, we have seen a fairly stable development until now.
Okay, brilliant. Thanks very much indeed.
The next question is coming from Mr. Matthias De Wit, KBC Securities. Go ahead, please.
Yes. Hi, good afternoon. I would like to start with a small follow-up question on the equity transitional. Could you comment what the remaining benefit is to the Solvency II ratio at this point in time, and how that benefit amortizes over time? Secondly, I had a question on the organic capital generation of EUR 348 for 2016. I guess this is based on the start-of-the-year balance sheet or averages, whereas there are some changes in mix and rates in the meanwhile. Just eager to get your comments on how that number could develop into 2017, how we should think about capital generation in 2017. My last question is on LAC-DT. I noticed that the benefit to SCR increased to EUR 586 from around EUR 500 at the end of H1. What is exactly the key driver behind that increase?
Is it in the life business where you moved from 50% to 60%? Is there any conservatism left in your current approach, or do you think it's currently a fair, taking into account some regulatory risk that might remain? Thanks.
Matthias, on the equity transitional, let me look it up. I don't have that number on top of my head. We'll look it up. We'll feed it back
Yeah
through IR to you. In terms of the OCC, that was defined on the beginning of year asset mix last year, 2016. Used the beginning of the year asset mix. For 2017, we'll also use the beginning of the year asset mix. In terms of where are we on that number, one thing we learned throughout 2016, it is a number, especially interest rate component, is pretty sensitive to interest rate movements. In 2016, we saw a V-shaped long-term yield development, and especially the UFR unwind, of course, is pretty volatile and sensitive to rate moves. What I can say today is the business performance that underpins, that is behind or underneath this OCC, we feel very comfortable with the performance of our business.
I mean, the year 2017 is only a couple of weeks old, but it's kicked off in the same notion as we ended last year. In terms of business-wise, this company is still performing at the same level as we ended last year. Rates have stayed stable, moved up a bit in the first half of the year, but we need to see how they develop, and OCC is a number that is very vulnerable to interest rates. Not so much the total solvency numbers, the delta solvency is relatively stable. In the slicing and dicing, the OCC has a rate volatility, which could be a headwind, could be a tailwind. We find it hard to give, at this point, major guidance on this number. Safe to say that the business trading has gone off to a good start.
The investment portfolio hasn't changed much during the year. That gives you some clue going forward, and we just need to see how rates develop. Just give you a feel for sensitivity, for example, one point of combined ratio, better or worse, is about EUR 5 million-EUR 6 million in OCC. When you go and want to model this, that's kind of where the sensitivity is, and the rest is really all about rate movements. In terms of LAC DT. LAC DT on P&C business remains stable at 75%. There are no use yet of the famous component 4, no use of future fiscal profits. On the 60% in life, has very limited, I think it's 58% of the 60% is DTLs and historical profits, no future profits. It's a pretty stable, solid number.
The use of component four of future fiscal profits has been limited substantially by the DNB regulation. Obviously, you have to make pretty strong assumptions to substantiate significant use of component four. That's something that we will look into. It will require a bit of work and dialogue in the industry to understand exactly how to interpret some of the rules. Some points we may have interpreted conservatively. We believe the 60% is well supported. Is there conservatism left in the number? There's realism left in the number, that's one thing for sure. I still think it's a responsible number. Downside risks to that extent are limited from our point of view.
Okay. That's very clear. If I could just follow up on the organic capital generation, I also have a bit of difficulties in analyzing how or in getting a sense of a sustainable recurring SCR release because there were a lot of changes in the organic numbers. Is there anything you could say in that respect?
I think on the SCR release, the number we provided you for the full year was about 6%. If you look at the first half year, it was about 3.5%. The total release of capital the first half year was 3.5% and 5.7% for the full year. The difference between the two is an increase in new business. You may see our numbers, our P&C volumes have grown by EUR 80 million, disability has grown by EUR 20 odd billion million. The growth in our non-life business took place in the second half of the year. We believe something like a 6% release of capital is not a strange number. The main driver actually is how much new business we write. Again, from H1 to H2, you can show an increase in volumes in what we see as profitable non-life business.
We were happy to spend some capital release in organic growth in our business. I think the number we've produced so far has been relatively stable, and you could apply it going forward. The key driver here is the amount of P&C and disability volume we're able to attract.
All right. Very good. Thanks, Chris.
The next question is coming from Mr. Bart Horsten, Kempen & Co. Go ahead, please.
Yes, good afternoon. I have a bit of a bad line, so hopefully you can hear me well enough. Also on capital generation, if I may. You gave an indication, I think, in your guidance on net operating life results of 75%-85% translating into capital generated within life. Is that bandwidth still valid or do you think that would move upward as well? I was wondering, too, will you be revising assumptions on exit turn? Is that permanent? Is that a periodic schedule? That's on that topic. Second one, you said that your life insurance margin went up from 3.4% to 3.7%. I recall that during the IPO, you already stated you expected the life insurance margin to go up. It went up higher than I had anticipated. Is that a level which you expect to assume going forward, or do you see further improvements?
My final question relates to the right to buy back shares. I think the lock-up of the NLFI will end at April 17, and your AGM, it will be in May. Suppose the NLFI will sell down before your AGM. Do you have an opportunity to participate or will you can only do that after the AGM? Thank you.
Bart, thank you. On the life insurance business, the capital conversion ratio from life to profits or to cap generation, it is a bit lower than we thought, simply because the share of capital gains reserve, shadow accounting contributions to the life business was higher than last year. Going forward, it's probably easier to model of the OCC number than to model of the life profit conversion figure. The chunk or the share of capital gains release in life was higher than last year. The conversion ratio this year is in the lower end of the bandwidth that we provided, simply because there was larger capital gains. I think going forward, it's better to model off the absolute number of OCC. To the life insurance margin, it had moved up to 3.7%. It moved up faster and more than we guided to anticipate at IPO.
This appears to be a reasonably stable number. If I look at the life insurance business, I have no reason to doubt that this thing will change materially. The plateau at where we are appears to be sustainable. In terms of assumptions, we tend to have Q3 as the assumptions quarter. Normally we have once a year when we update all our non-economic assumptions. Cost assumptions, lapse assumptions tend to take place in Q3 with some overflow in Q4. Unless there is during the course of the year, a really striking phenomenon that you have to take into account. Normally, as for most insurance companies, Q3 is assumption season in our actuarial family. That's when we tend to happen.
Okay.
On your last question, do we still have room to maneuver if and when NLFI would decide to further sell down before the next AGM and after the lock-up period has ended? The answer is as simple as clear. We've used our full capacity with the first buyback opportunity because we wanted to give a strong signal to the market. If and when NLFI would decide to do a further sell down before the next AGM, we will not have room to maneuver in terms of buying back shares. The limitation is not our capital position or the unwillingness to do so, but just we're just not allowed to do so.
Okay.
It is probably fair to say that we do have the intention to participate in placings during the year, as we've done in the past. The magnitude of timing is out of our control, something for our shareholder to decide. The magnitude depends on the time at hand, but it's our intention to support the sell-down by the State as we've done in the past.
Okay. Thank you. If I may, I just found one other question I would like to ask, that's on the development of your DC business. It's moving quite okay. Tripled in assets and doubled. Could you tell us what the recent dynamics are? Is there also some capital release already shown in 2016 from this move from DB to DC, if I assume that these were mainly existing clients which you have? Thank you.
The increase in our DC portfolio were mainly new customers. We were happy with welcoming new customers. That didn't lead to significant releases in the DB book. The main driver behind that is our improved product. I think some of the other market participants decided not to be as active in this market as they were before. Today, we see three to four active pension insurance companies in the Netherlands actively involved in new business. I think the market becomes pretty small in terms of number of providers. That's helpful in acquiring new business.
Okay. Thank you very much.
The next question is coming from Mr. Kunal Sawhney, JP Morgan. Go ahead, please.
Kunal, you are asking question. I am not on your line.
Hello? Hello. Can you hear me?
Yeah.
We are. Sorry. There is a bit of confusion. This is Ashik Musaddi from JP Morgan. Just a few questions. First of all, can you give us a bit of color about UFR drag? If I understand correctly, and if I remember correctly, at first half, it was 2.2 points, and at full year, it is three-point something, 3.5 points or something. Your capital UFR drag in second half went down compared to first half, whereas given what interest rates have done, it should have gone up materially, UFR drag. What is going on there? That is number one. Secondly, going back to slide number 22, you are using some spread of core sovereign bonds of -20 basis points and non-core of 50 basis points. What is your thought process behind it?
If I look on Bloomberg at the moment, year-end spreads for, say, Germany was 60 basis points minus, for France, it was 40 basis points minus. Your core sovereign spread should be like -50 range, as well as your non-core sovereign. If I look at, say France, Belgium, Austria, supranational, everything was negative, and you are assuming 50 basis points positive. What is the rationale behind using this? These are market consistent data which we can track every day on Bloomberg. That is the second one. Third, can you give us some color about your life earnings, which there is on slide number nine, there is something called additional investment results. Looking at the slide, it looks like it is roughly EUR 60 million, which includes M&A as well. In the previous slide, you mentioned M&A is EUR 20 million. That means additional investment income is EUR 40 million.
That is a big jump from EUR 440 million to EUR 480 million. What is driving that? Majority of the asset allocation shift you have done is in second half, any thoughts on that would be great. Thank you.
Very good. In terms of the UFR drag, we do calculate in every period from the beginning of the period interest rate and the ending of the period interest rate. We looked at the UFR contribution as per January 1st and the UFR contribution as for 31st of December. That is the number we use in our analysis. Rates during the year had a V-shaped development. Beginning of the year, end of the year rates we put into our model. We have seen other people doing a funny averaging method. We have used a constant zero model using beginning of the year and end of the year solvency. That reflects where we believe the right way the UFR drag develops.
It does front load some of the UFR drag instead of averaging it over time, of using beginning of the year, end of the year numbers. In terms of the long-term investment spreads, as I said, we slice and dice. We slice the delta solvency over time. The number that is hardest is beginning of the year solvency, end of the year solvency. Then you have got the solvency accretion, which is a number that you cannot argue with. It is just the delta in solvency that you achieved. OCC is a way of bucketing into what is a sustainable, replicable level of capital generation. Here we have assumed long-term investment margins are reasonably stable across the cycle. Again, some of these numbers are a bit more optimistic. In real estate, the direct investment yield is still 4.3%.
I think if you look at the numbers, it comes out 3.7% applying our spreads. On equities, we use 3%. The actual return you make on equities is larger. The spreads we make is actually a blend of direct income plus capital gains across the cycle, where we believe in terms of Core government bonds. There is a clear drag today on holding those. Otherwise, they may yield better. Again, in other categories, for example, real estate and equities, we still have a fair degree of conservatism. Across all categories, we believe in the long run, this is a bankable set of indicators, especially if you take into account the capital appreciation of some of the other investments. Our mortgages are at 110. The actual spread is higher. The default cost is virtually nil. We have also absorbed those, not completely in the numbers.
The 110 is also reasonably conservative. Across the numbers, we feel this is a sustainable, defendable, bankable number across the cycle. In terms of the live earnings, page number nine, there is a set of shaded bars. Let me just give you the numbers that are in those bars. In 2015, the top number is EUR 441. If you go down, you add EUR 15 for swaptions, EUR 146 for shadow accounting release, EUR 173 for investments, and EUR 107 for other. That fills that chart. If you go to 2016, the EUR 551 breaks down in EUR 57, which is the shadow account release from swaptions valuation, EUR 65 from M&A and additional investment result, M&A acquired businesses, EUR 212 from regular shadow accounting release, and EUR 105 from investment results. If you look at those numbers, investment result plus the regular release was EUR 319 last year, EUR 317 this year. A pretty stable number.
On top of that is additional results of M&A of EUR 65 and a EUR 57 contribution of earnings from capital amortization of the capital gains on the swaptions portfolio.
That's very clear. What I was trying to get some clue about is this EUR 65 million, the second bucket in 2016. If I look at slide number eight, for example, it mentioned that operating results from your acquisition was EUR 22 million. Actually, by the way, this was same at first half as well. I don't understand how this happened as well. A number at first half unchanged at second half as well. Anyways, even if we say EUR 22 million of acquisition benefit, that still means that additional investment result is plus EUR 43 million. That's quite a lot on a base of EUR 440 million. What's driving that? Is it asset risk or is it some sort of capital gains which may or may not disappear? Any thoughts on that?
See, the only thing I'm trying to understand is this EUR 40 million number sounds a bit large given the base of EUR 400 million.
EUR 40 million may sound large on the base of EUR 400 million. On the basis of a EUR 30 odd billion investment portfolio, you're actually talking about something like 10, 13 basis points of additional returns. The EUR 40 odd million is the right number. EUR 65 minus EUR 22 gives EUR 43. The EUR 43 is a reflection of a few things. Some increase in shadow accounting release. Secondly, you may recall in our half year result that post-Brexit, we re-risked our portfolio a bit. We took advantage of widened spreads at that point in time to take a bit more risk. In the second half of the year, we added more risk to our investment portfolio, partially by buying then what we thought were underpriced U.K. bonds. We allocated a bit more to equities and to mortgages.
During the year, the re-risking paid off, and indeed, EUR 40 million is a lot against EUR 400 million. Against a EUR 30 odd billion investment portfolio, you're looking at a 10 to 15 basis points additional return. That puts things in perspective from our point of view.
This is recurring for next, say, whatever is the life cycle of the business, basically. It's not like a one-off, something like that.
No, we believe this, if I look at our mid-year plans, at least for the plan period we can foresee, this is a fairly sustainable number.
Sorry, just going back to UFR. How should we think about UFR? Because if you take a starting and ending and in between interest rate remains zero, then is that the right way of reflecting the UFR drag or Yeah. Anyways, I don't know. I don't know the answer as well, but just any. Because the thing is that interest rates went down in second half. Second half interest rates were definitely much lower than first half, and your drag was actually lower. I just got a bit confused here.
Right. Kunal, this is Michel. Can we take this offline.
Yeah. Sure. That's good. Thank you. Thanks a lot for your answers.
Ladies and gentlemen, for any additional question or remark, please press star one. Star one for your question or remark. The next question is coming from Mr. Arjan van Veen, UBS. Go ahead, please.
Thank you. Just a couple of follow-up questions. Firstly, it doesn't sound like the LAC-DT change you made to be in line with the new guidance is particularly material. If you can give a bit of color on that. Secondly, you gave us a lot of very clear guidance or commentary around the buyback. Can you also maybe give a bit of color around, you did a couple of bolt-ons during 2016. What's the outlook for potentially more bolt-ons going forward? Thank you.
Indeed, Arjan, I think we're actually very pleased that the LAC-DT guidance of DNB was not material to us. There was certainly no material negative. There was a small positive, a couple of points that it added to our capital, but that was, if you look at the way our capital developed, kind of offset by the warehousing of the real estate portfolio. Net impact on our capital was limited. Again, I guess our key messages here is LAC-DT did not materially negative impact our business. Actually, it's a small positive, more to come, who knows? That depends on how the market, and we will interpret the number, but it's certainly no negative on LAC-DT to us. On the bolt-ons, we are open for business. Our solvency position is strong. We continue to look for options.
On the other hand, we have very strict investment criteria, and if we look at businesses to acquire, they should at least meet our financial criteria. We more often have said no to options over the last two years than we've said yes. If you take a look at slide three, where we show our portfolio, we would be willing to look at options in the non-life area. If there would be a non-life book for sale, then we definitely would have a look at it, as long as it comes at the right price. We're very interested in funeral business because it hedges with our longevity. In the area of business enhancement opportunities, the fee business. In terms of distribution, I think for the time being, we are done. We are particularly looking at business that can strengthen our assets under management, our fee business.
In sector B, we're in the middle of converting our own portfolios to a software-as-a-service book. We're halfway. If that's done, then we also would be in consolidating the Dutch life insurance business market, especially the smaller insurance companies. We're open for business. We have a very strict valuation in terms of finance metrics. We never comment on particular deals where we are looking in at the moment.
Okay. That's very clear. Thank you for your additional solvency disclosure. It's market leading, and particularly slide 24 is very helpful. Thank you.
The next question is coming from Mr. Matthias De Wit, KBC Securities. Go ahead, please.
Yes. Thank you. I just some small follow-ups. On Solvency II, you stated that part of the increase in the ratio in 2016 was linked to cost savings. Just wondering if that's positive variances or is this more linked to changes to your cost assumptions? Secondly, on the risk margin release, just wonder if you could provide somewhat more color on the amortization pattern of that release? I think it could be quite long in nature. Lastly, on bank and asset management, the operating results dropped quite significantly. I guess it's mainly linked to startup costs and integration costs. Could you confirm whether that's the case? Also going forward, what could we expect from this business line in 2017 and 2018? Thanks.
Okay. On the cost savings, there really was the integration of the AXENT funeral business at Ardanta. We believe, and we're pretty convinced in our belief, that we're the lowest cost operator in the funeral business. Remember in our Q3 call, we said there were a number of FTEs that came in, a number of FTEs that were left. I think we do this with much less people from 60, 70 people to less than 30 people for the same portfolio. Now, that is reflected in the lower cost charges in your best estimate liabilities. That provides a capital uplift. In the Solvency model, that shows up actually in the bucket variances in other. Markets and other, that's where it shows up. No, sorry. It shows actually up in the capital charge in the life business.
In the delta capital, it shows up in the bucket variances in other. In terms of stock, it's lower life capital charge. In terms of flow, it shows up in the bucket of other. Where in terms of the run-up of the risk margin, let me look it up for you. Risk margin release is something that happens over a significant period of time. Over the coming years, we believe that there is still risk margin release to come. It is somewhat, I think, front-loaded. It's not an equal number over the entire period. The first years will be a bit higher than the latter years as the individual life book runs off. The exact pattern is not something I have at hand here, but please.
I guess for the UFR benefit, it's the other way around, I guess. It starts high and gets lower over time or?
Yeah, the UFR unwinds. The UFR drag is higher in the first period and moves down over time.
Yeah. Okay.
The offsetting risk margin release is higher and lowers over time. There's a plus and a minus that are both higher in the early years. In terms of banking and asset management, that segment, two things at play. In our bank, we have a relatively small bank. The profit of the bank this year was more a financial profit than operating profit. They were more in the shape of capital gains on a fixed income book. You can see the IFRS profit in the segment actually keeping up reasonably well, but the operating profit lower. In terms of the business, we believe that asset management is a growth business, but we made costs in terms of launching the Offices Fund. We made costs in terms of hiring people. We had integration costs of BNG without the full-year results kicking in.
There's really more costs preceding returns. It's the investment rather than underlying performance issue. We believe that if our funds kick off and if our goals materialize, this will be a solid profit contributor going forward.
Okay. Very clear. Thanks, All. Thanks a lot.
There's an additional question of Mr. Robin van den Broek, Mediobanca. Go ahead, please.
Yes. Good afternoon. Just one question to clarify, probably an answer given before, but the EUR 348 of capital generation, you've indicated that you're locking in some core spreads during H2 and then early 2017. Does that effectively mean that half of that portfolio should be assigned a 50 basis points excess return rather than the -20 basis points excess return, which is basically inflating that EUR 348 further? Or not?
I am not sure I understand your question, Robin. Could you please elaborate?
Well, on slide 25, you indicated EUR 3.8 billion has moved from basically core to short-dated non-core sovereigns. I assume the EUR 348 you reported on the new framework takes into account the average mix of the portfolio in 2016. The fact that the mix now is more towards non-core sovereign, although it is very short-term paper, but it is still non-core sovereign, should we assume that that is an incremental excess return compared to the EUR 348?
Well, the EUR 348 actually is based on the beginning of the year portfolios, the beginning of the year 2016. Compared to the portfolio at the beginning of the year, you will see a relative decline of core versus non-core. It is actually the swap spread trade should actually support the operating capital generation rather than dilute it, because the EUR 348 was based on the January first portfolio, now we have a portfolio that has more yieldy assets.
Okay. Your answer is yes, basically?
Yeah.
Okay. That's very clear. Thank you.
Mr. Chairman, there are no further questions. Please proceed.
Well, thanks for the time you took to listen to our story. Hopefully we were able to reflect on all your questions in a proper way. To close this call, again, we were very happy with the results we could present today. We delivered upon our promises. I said in my introduction, it was hard work. We intend to keep on doing so, to deliver at least in line with our promises also in 2017. I hope to see you all in person somewhere over the next period. Thanks. Have a good day.
Ladies and gentlemen, this will conclude the ASR conference call on the 2016 annual results. You may now disconnect your line. Thank you.