David Stevens - CEO Geraint Jones - CFO Cristina Nestares - CEO UK Insurance Martin Coriat - CEO, Confused.com

David Bracewell - Redburn Partners Thomas Seidl - Bernstein Andreas Van Embden - Peel Hunt Andrew Crean - Autonomous Research Iain Pearce - Berenberg Ed Morris - J. P. Morgan Will Hardcastle - Twelve Capital

Good morning, everyone, and welcome to the 2016 Full Year Results Presentation, my first, full year presentation as CEO. And my predecessor, throughout 25 years of being a CEO, announced an unbroken string of record profits, and I stand before you to announce a fall of a quarter in our profits. I'll get my coat. But, in truth, very much the impact of the Ogden change, as we all know, and when you look through that, you see a business that is growing robustly across practically all the areas in which it operates, and which is demonstrating its resilience in the form of a maintained and substantial dividend payment in 2016. To talk to you through the results in more detail, Geraint and I are going to stand up and talk about numbers and the capital position; I will come back and talk a bit, a big picture about our own performance in car insurance, and about the market as a whole in car insurance; before Cristina Nestares, our newly appointed UK Insurance CEO, talks about car insurance and home insurance in more detail. And then, Martin, CEO of Confused, will talk about price comparison, both in the UK and across the world. I will then talk about the Direct insurance businesses beyond the UK before wrapping up. Thank you all for coming.

Thanks, David. Good morning, everyone. Thank you for making the early start. I'm going to cover some of the main features of our 2016 results, spend some time on Ogden, and I'll finish up on capital and dividends. So, let's begin. We said at the half year that we were very pleased with our 2016 performance at that point. And if we look through the Ogden impact, and look back at the full year at the underlying business, then much remains the same today. This slide shows some of the key points from the results. At the top, you see customer numbers and turnover, both strongly ahead: 16% and 22%, respectively. Those percentages were 15% and 19% at the half year, so moving even further ahead in the second half. Profits and EPS, as you can see on the left and right, both down about one-quarter. That is entirely down to Ogden. We'll see the exact impacts shortly. Return on equity, bottom left, also similarly hit by Ogden, 37%, though, still not exactly a terrible return, down from 49% last year. Bottom right, full-year dividends for 2016 were 114.4p per share, which is exactly in line with 2015. And that goes -- the same goes for the final 2016 dividend as well. And finally, right in the middle, the solvency ratio, after accounting for the Ogden impact and the final dividend it's still very strong, it's 212%. But let's get into Ogden. On this slide, I'm going to explain to you how the new discount rate affects the balance sheet, our profits, capital, and the business. I think we probably all know that the rate has changed from 2.5% to minus 0.75%. The best estimate loss ratios that David will come on to shortly, and a capital position, that I'll show you in the next few slides, are both prepared on the basis of that new rate. So it's a best estimate position of minus 0.75%, rather than a booked position. We estimate that the impact to Admiral, which we will see in higher claims costs on our net reserves and lower profit commissions, will be something like £150 million after tax, or £190 million before tax. As you can see, the impact before all co-insurance, reinsurance, and tax was over £300 million. As we've said many times before, we take a prudent approach to reserving in the financial statements. And reflecting the uncertainty that remains, we've held the margin in the book reserves above best estimates, consistent with the level of the past couple of reporting dates. That margin remains prudent and significant, and is the top end of our reserving policy range. The profit impact this year is just over £100 million before tax, so about £85 million before tax to come over the next few years. I'd estimate about three to five years, probably. Technical provisions on the Solvency II balance sheet also valid at minus 0.75%; so, as we'll see shortly, despite that change, the capital position remains very strong, 212% after the final dividend. Now, naturally, we would expect such a substantial shift in the discount rates that will have a knock-on impact on market pricing. We moved our rates materially back in December when the MoJ released its statement, and we expect to move a bit more in the next week or two. Two points to make on reinsurance. Firstly, there will, most likely, be material changes in excess of loss pricing, the next renewals, and David will cover this; and secondly, on quota share reinsurance, all our UK motor capacity is placed up to the end of 2018. We don't expect the new Ogden rate will change the results for the reinsurers, and we don't expect the new Ogden rate will impact the next negotiations for 2019 and beyond, which we'll come to, probably, later this year. Just to bring out what the exact impacts of Ogden were this slide shows the pre- and post-Ogden positions for the key figures that were most impacted. Profits and earnings per share would both have been up a couple of percent, instead of down one-quarter; and return on equity, in the middle, would have been in line with last year at 49%. Moving away from Ogden, let's now look at the top-line figures across the Group in a bit more detail. I think you'll agree, you see some very strong results for 2016. UK car insurance turnover just missed the £2 billion mark, which is pretty much what the Group delivered over the past two or three years. And we continue to add customers, 350,000 new customers in UK motor over the year as a whole, as market conditions remained favorable. But positive as those stats are, they start to look a bit paltry when you compared them to the rest of the Group. Household grew by one-half on both measures; that follows even more stellar growth than 2015. And international insurance was nearly 60% higher in turnover, and 30% higher in customers. And last, but certainly not least, price comparison, which grew its turnover by 20%. Pleasing performance in all markets. Moving now to profits, this is our usual slide, which shows the composition of the results for the year. As usual, in the dark blue, that's the UK insurance result, which covers car and household; the dominant part, this year with a post-Ogden profit of around £340 million. As I said earlier, that's just over £100 million lower than last year, 2015. As you can see, the change is entirely down to Ogden. Household reported another profit, whilst continuing to grow very fast. The green section is international insurance, where the results improved from a loss of £22 million in 2015 to £19 million in 2016, as the combined result in Europe continued to improve. As well as the strong growth we saw earlier, the overall combined ratio across all operations continues to track in the right direction. And ConTe, in Italy, was profitable again. In the light blue, just about visible, is price comparison, which, in aggregate, turned around a loss of £7 million in 2015 to a profit of £3 million in 2016. Strong improvement in Confused.com, which increased its profits by 30% to £16 million contributed, whilst there was also record-profit at Rastreator, in Spain, and continued encouraging progress at compare.com. Finally, in the yellow, that's the other segment, which mainly includes share scheme charges and the debt interest costs, and that was pretty similar in value to 2015. Turning to capital, a quick reminder before we start, our solvency capital requirements, the SCR, is based on the standard formula from Solvency II plus a capital add on. And despite the change in the Ogden basis that we've talked about, the solvency position is very strong at 212% after the proposed final dividend is accounted for. I'll cover the dividends on the next slide. 212% is a reasonably significant improvement from the position at the half year, which was 180%, and a number of things, some of which have offset each other, have contributed to that change, which you can see on the bottom right-hand side. And I could spend the whole presentation talking through the impacts on the bottom right-hand side, so I'm going to offer up to talk individually about those impacts later and move on. As usual, the slide pack includes solvency ratio sensitivities. And you'll note that the change in the PPO assumptions, referenced on that slide, have led to a significant reduction in sensitivity to interest rates compared to the position at the half year. Now, the final dividend, we've taken the decision, this year end, as you now know, to hold the final dividend at the level of last year's final, excluding the return of surplus capital, the super-special dividend that you can see there on the left-hand side, in the dotted red lines, above 2015's figures. And so, despite our second-half profits being £65 million after tax, the proposed final dividend for 2016 is £144 million, or 51.5p per share. And that brings the full-year dividend for 2016, as you can see, to 114.4p per share, which is also in line with 2015. On the right-hand side, you see a reminder of the dividend policy, which I hope is familiar; that hasn't changed. For the foreseeable future, we'd still expect the earnings-related dividend to represent a payout ratio in the 90s%. You might notice here, there's no update on the future returns of surplus capital. The solvency ratio, at over 200%, is higher than we would expect to operate that in the long term, once we're up and running with the internal model, hopefully, in 2018. Our philosophy as regards capital in the Company is the unchanged. We don't intend to keep surplus capital in the Company; we intend to give it back to shareholders. But we think, for the near term, it's sensible to run with that higher number, for a number of reasons; mainly, until we have certainty over the internal model capital position, which we expect to get as we move into 2018. And there's also residual uncertainty over how Ogden feeds through, what happens with the Ogden discount rate review, regulatory review of the current year capital add on, and so on. The dividend dates are shown on the slide, and the payment date is June 2nd. Finally from me, a reminder of key points from this section; firstly, very strong growth across the Group in turnover and customers. Profits were inevitably hit by Ogden, but there were improvements in the international results and the price comparison results. Capital position remains very strong, despite the Ogden change. And finally, the dividends for 2016, including the final, are in line with 2015, despite the profit hit. That's it from me. I'll hand you over to Cristina and David to talk about the UK.

Thank you, Geraint. I'm going to do a couple of pages on big picture, our own performance in the UK car insurance market, before talking about the market as a whole. So a year of substantial growth in the number of vehicles we cover, up 11%; that takes us up by over 20% now since we started growing again, in the middle of 2014. Turnover up 16%, profits down 24%, largely reflecting the impact of Ogden. Without Ogden, the profits would have been flat. There was still some impact there of the commutation of 2014 on the 2015 to 2016 comparables, which we talked about at length at the first half. Most of you will be familiar with these graphs. On the left-hand side, you have a comparison of the market's claims ratio and our own claims ratio, or loss ratio. The red line is the market. It's the market as at 2015 reported loss ratio with the reserve releases redistributed back to the years to which they relate. And our own number is the number at December 2016 within the brackets, the movements between June and December 2016. Our own numbers fully reflect the impact of the move to minus 0.75%. What, historically, you've often seen with our numbers is a more substantial number in the bracket-bracket that move down, particularly on recent years. You're not seeing that this time, because of the impact of Ogden. It's very early to call to 2016, but our first projection is 82% at this point, which is consistent with the first projection for 2015, in December 2015, which was also 82%. But, of course, the 2015 number at that time was pre-Ogden, the 2016 number is post-Ogden. On the right-hand side, you have our reserve releases, always a feature of our results, as we wait to be 100% sure that underwriting profits are true profits before we actually allow them to feed through into the reported profit. The releases would have been 23%, without Ogden, they are actually 14%. And, as Geraint mentioned, we remain at the top end of our reserving range. And there has been no reduction in the level of conservatism in our buffer at the end of 2016 versus the middle or end of 2015. Now, moving on to the market, I think what we're all interested in knowing is what's the impact of Ogden likely to be on claims costs, and how will that feed through to price changes, and when, and so we've attempted to do some work to try and help with that. And part of that work is to understand what the profitability of the market in 2016 was pre the impact of Ogden, which is not an easy thing to do, because reported results, with a heavy roll of reserve releases, don't necessarily give a fully accurate reflection of current profitability. So the way we've done it is we've gone back to a year which we feel is largely developed, namely, 2012, and then we're rolled forward the premium changes and the claims changes. Now, we're showing here a market combined ratio for 2012 of 104% as at December 2015. Now, let me just caveat markets here, and I should have done this on the previous page in relation to the red line. Market is those companies that report through PRA returns, and so it excludes Gibraltar; it also excludes ourselves, Hastings, Saga, it excludes Dublin, Zurich, excludes Lloyds, but Lloyds are no longer very material. So you're looking at something like, actually, two-thirds of the market here. They were at 104% in December 2015. Roll it on through 2016, and on to where it ultimately might have finished without Ogden, and it might have finished around 100%, which we think equates to a return of capital of around 28%. The reason we say 28% is we're assuming 11% add on in investment income return, and a capital requirement of 40%. What's happened to premiums between 2012 and 2016? What's happened to claims? For premiums, we've taken the ABI data and we're seeing a mid-2012 to mid-2016 fall of premiums of 8%. The actual falls in written premiums that took place in the second half of 2012 and through 2013 and early 2014 still outweigh the increases that took place thereafter, if you look at it on an earned basis as at the middle of 2012. If you look at it on a written basis, it would be about minus 3% or minus 4%. Then, you look at the claims situation, and you're looking at a situation where we've actually had, with the benefit of hindsight, and using sources like the third party, working party, using some ABI sources, but also having management estimates, particularly in relation to 2016, so all these numbers are really not points, they're really in the middle of ranges, looking at those, we think the cost-per-vehicle year change in that period was something like 6%. Unusually for the market, mainly driven by damage, which was down a bit in frequency, but up in severity, small BI shows the benefit of the LASPO reforms over the period, and large BI was unusually benign for those four years, pre-Ogden, with frequency down, offsetting any increase in severity. So you roll those through and you get a combined ratio for that part of the market that we're looking at. The 100% turns into 112%, and, essentially, the market is around breakeven. Then, look at the impact of Ogden. It's not as easy as you might think to know what the price of the cost per vehicle year impact of Ogden is. Our estimate is that's it around 10%. But I would treat that, again, as a midpoint within a range. And our estimate is, for the market as a whole that around 60% of that accrues to the reinsurers and about 40% to the primary insurers. There are a number of insurers who will have a materially smaller share of the initial impact who have low retentions. There's ourselves that have a higher retention than the likes of esure and Hastings. And they're in the middle of the road. And you can tell from the pack, we're roughly 50-50 split between reinsurer and ourselves. And then, there are other players, particularly the big multinationals, who have materially higher retentions, again, and will bear more of the immediate cost. You take that 10%, you add it to the claims cost, and we're sort of saying that the market, if it was not to respond to the Ogden change, is substantially loss making; or at least that proportion of the market that reports through the PRA. That is the material qualification. I mentioned that the market as a whole, at the beginning of this analysis in 2012, was 104%. Our combined ratio, our equivalent number for 2012, was 74%. So this is not an analysis of our own profitability, which is very different, this is an analysis of the market profitability. And it's, obviously, subject to a number of adjusting factors. You could argue it's too pessimistic, because price increases have gone in the second half of 2016 that aren't fully reflected in this analysis, but they haven't been that dramatic. And you could argue it's too optimistic, because, actually, there's pressure on material damage inflation, caused by the fall of sterling. When the announcement was made in the middle of December that the Ogden rate was being reviewed in an urgent fashion, we took a conservative view, both on the expected outcome of that review, and the inevitability of some change, and we made an immediate price increase in December to reflect that conservative view. Our view, looking around the market, is we may have been alone in that. Certainly, we were largely alone, if we were not entirely alone. And what we have seen, though, which has encouraged us, is ever since the announcement of the actual rate prices have begun to change, and our percentage times top which fell materially in the middle of December, is beginning to rise back up. What that early and we think appropriate move does mean is that the volume of business that we've written in the first quarter is lower than it otherwise would be, and the business has not grown during that first quarter, which it probably otherwise would have done. I said we took a conservative view on the ultimate outcome. We didn't actually assume minus 0.75%, and so we anticipate some further price rises from ourselves in the near future. So, over to Cristina to talk in more detail.

Good morning, everyone. David and Geraint have talked to us about the impact that the change in the Ogden rate has on the market, on Admiral, and I'm going to focus on the fundamentals of the UK motor insurance, because it remains broadly unaffected; and also, will talk about the progress of our household book. First of all, let's start with expense ratio. This graph has been, long time, a feature of the Admiral result's presentation. It shows that we have constantly achieved a significant expense ratio advantage over the market. Last year, the expense ratio remained flat at 16%, despite the growth; and also, despite the restructuring cost by the change in the IT system. One important contributor to this is our efficiency in acquiring new business, and also our ability to retain customers, which brings me to the next slide. Our retention rate has been stronger than the market for the past few years. And during 2016, retention in the market decreased, while we managed to increase our retention. And we did this through a combination of factors, great customer service; a number of improvements in the renewals department; and by allowing the base of renewal price increases to lag behind the base of the new business price increases. We're very focused on giving great service to our customers, and this is particularly important at claim stage. You can see on the graph that the customer feedback, following a claim, has always been quite strong, and it has improved in the past few years, which also encourages customers to renew with us. Let's now move to claims. Most of our loss ratio advantage comes from superior risk selection, and also from the handling of large BI claims. However, a significant part also comes from better handling of volume claims. And this is due to a combination of factors. First, our investment in detailed first notification of loss and our embedded commitment to capture and handle third-party claims ourselves. And this translates into faster settlement and more efficient settlement, too. Secondly, our approach to our supplier network, where we have long-term relationships. And thirdly, we take a firm approach to claims fraud. And we're not afraid of going to court to help reduce leakage. And then, underpinning all this, it's our continued ability to select and train the right people, who truly care about their roles, their achievements, and their business. Let's move to our household book. Last year, we grew by more than 50%, while managing to increase our profit. We had claims frequency reduction, and we also enjoyed a benign year in terms of weather. We also had made some operational improvements, and we benefited from an increase in scale, which translated into an expense ratio improvement from the previous year. This expense ratio that I'm talking about is in written basis. And this is even despite the impact of the flood re-levy, which added 3 points to the expense ratio. Now, growth in the household book last year was driven both by a growing price comparison market, and by more customers being drawn directly to our household products. The growth of the household aggregator market accelerated during 2016, reaching just over 50% of the new business market. This is longer than the motor aggregator market. So we expect this to continue to develop over the next few years. Our direct sales also increased in 2016. And, as you can see on the graph on the right, we have started moving the direct percentage up on the second half of the year. It accounted for 17% of our new business sales. Included in this percentage is the cross-sell volume. With the rollout of our new policy system, Guidewire, we have been able to increase our ability to cross-sell household policies to our existing motor customers. In summary, both our motor and household products have performed well. On motor, we believe that our strong expense ratio advantage, great customer service, and efficient claims handling leave us well placed to continue performing the market. Also, the progress of our household business provides an exciting area of growth for the future. And finally, we were very pleased to have won second-best large company to work for in the UK. We are the only company which has been on this list since it started, more than 15 years ago. So, that's it from me. And now over to Martin.

Good morning. For those who do not know, or remember me, my name is Martin Coriat, and I'm the Confused.com CEO. My task, in the next few minutes, is to present the very good results of our price comparison businesses around the globe. As you can see on this slide, Admiral Group operates four price comparison businesses in four different markets, and this slide gives you a snapshot of each operation. And you can note that we have operations at different stages of maturity and also, we operate in very different markets, so we have to adapt to local customer needs. It doesn't come as a surprise then that they all have to follow their different growth strategy on their respective markets. But despite being a very different operation, you can see some common themes in the 2016 achievements. Growth, every operation has grown in the last 12 months, new product development as several have launched new products and services to feed their strategy, and improvement, as key metrics have improved in every operation, making them operationally more efficient. So, overall, 2016 has been a good year for the price comparison operation of the Admiral Group, and that's the main message for today. So, let's see in more detail how each operation performed in 2016. Let's start with Confused.com results. The UK is a rather mature market, as far as price comparison is concerned, and has remained a growing market in 2016, as you can see on the top left of this slide. As for Confused.com, 2016 has been a busy and productive year, and the results are strong, both in terms of traffic and profit growth. Quotes are up by 7%, while profit before tax has grown by 29% to reach £16.1 million for 2016. More importantly, Confused.com has changed its positioning to become a one-stop car savings shop and to focus on motor-related products with clear benefits, and the results are encouraging at this early stage. The new positioning has differentiated Confused.com in a more-and-more undifferentiated market. Several new products have been added to make drivers win and save on Confused.com. And we've seen rapid growth of the business since we launched the new campaign as Confused.com has been the fastest-growing aggregator for car insurance in the second half of the year. All this can be summarized with Confused.com claiming to be number one for car savings websites, with more opportunities to save than any other website. But beyond the results, I also wanted to take this opportunity to detail how we see our strategy expanding beyond car insurance. Being driver-and-saving centric gives us room to grow beyond car insurance. You have here several categories of car-saving opportunities that we offer at Confused.com. Some offer opportunities to interact more often with drivers, and there is a car-saving tool banner, like MOT or petrol comparison, while others have significant potential to offer a driver's win and create significant business in the future, such as car finance comparison or car buying. So let me give you more details on a driver's win with our car finance product that we have recently launched. It presents unique features, such as comparison across several car finance products with a true and personalized APR, and not a representative APR. And we see potential in this product to make British drivers make an informed decision while choosing car finance, especially when you know that more than 3 million cars a year are bought on finance. This is how Confused.com help drivers and reinforce the number one for car savings ambition. Moving to Continental Europe, where we operate, as you know, an aggregator in Spain, Rastreator.com, and one in France, LeLynx.fr, both businesses had a good year in 2016. As we indicated, though, aggregator market remain smaller than in the UK, so it's good news to see growth in terms of quotes and revenue. They are both very well-known brands. To give you an example, Rastreator has a prompted awareness of 96% in the Spanish market, which is fantastic for eight-year old brand. And they both have key market positions for car insurance and are now growing far beyond this product. But they also have to deal with very different market dynamics. LeLynx is focusing on developing the market and educating customers and insurers on the value of aggregation. The message is well received, but it needs time to impact customer dynamics in a massive way. In that context, marketing investments remains crucial. And it's good news to see the market becoming more competitive with new entrants helping to further educate the French consumers. As for Rastreator, they are now a clear leader in several verticals, and they are showing strong growth in new territories, such as mobile phone comparison, or other financial services. They are clearly on an upward trend and have ambitious growth plan for 2016 to leverage their unique assets, such as their brand, their customer loyalty, and also their technical expertise. Moving to compare.com, and our progress in the US, I'm happy to report that we continued to make good progress with compare.com in the last 12 months. First, we're seeing important key metrics heading in the right direction, indeed. First, we see good progress on adoption of our model by insurers. compare.com offer more and more prices to customers because of the expansion of our panel of insurers on one end, we displayed 50% more brand than in December 2015, for example; and expansion of the footprint into more states, on the other hand. The second source of encouragement comes from improvements in terms of customer acquisition, following a change in our approach in 2016. We have, indeed, decided to cut back on marketing investment in 2016 in order to focus our activities on key states, to be more visible in some chosen geographies. In those states, numbers show dramatic acquisition cost improvement, on the back of better focus and better targeting, bringing, overall, a better response rate to our marketing activities. So we see that there is good progress, and we are overall pleased with this improving metrics. Is it enough? Clearly, no. Do we need more improvements to start generating profit on a larger scale of volume? Yes. But we are encouraged by these numbers. And we continue, with our partners, to invest on this operation, to seize the opportunity of the US market. But we expect further investment is needed for compare.com in 2017 with losses in the region of $15 million to $25 million. We intend to have Andrew Rose, compare.com's CEO, to come here for the half-year result announcements to go in more detail on compare.com progress, so watch this space. As you have seen in the last few minutes, 2016 was a good year for price comparison operations across the globe. There's a lot going on in this fast-paced operation, but key messages are consistent: first is a strategy that fits the customers, grow across different products and verticals, and prepare the business to capture new opportunities. I will now leave the stage to David Stevens, who will give you more details on our international insurance operations.

Thank you, Martin. Particularly, thanks to Martin. Following the deferment of the results, he had to fly back from the French Alps today to join us, very briefly; much appreciated. I'm going to talk about the international insurance results. As you can see, the rate of growth in both Europe and the US has been substantial. In Europe, we're up one-third, both in value and in number of vehicles covered, while seeing the profits, sorry, seeing the losses fall substantially and going to single-digits at 5 million. That's our share of the losses. If you actually look, on the right-hand side, at the whole account picture, Europe, that's Italy, Spain, and France, very close to breakeven in 2016. Looking at the US, also a picture of very rapid growth, combined with falling losses. The volume of number of cars on cover up 20%, the premium written up 49%. With that increase in average premium, a combination of rapid price increases and some mix effects. Losses, down from $23 million to $21 million; and down, in percentage terms, from 27% to 22%, but still materially further away from breakeven than the European operations, which I think are on the cusp of delivering material value to the Group. Let's look at Europe in more detail across the three countries. Italy, which, yet again, reports a profit, remains our biggest market and is growing as fast as Europe as a whole, up one-third. And the other highlight, I think, is the emergence of France with 60% growth; 92,000 vehicles on cover at the end of the year, and last week we went through 100,000. I said at the half year that we were investing in brand advertising across Europe. And our objective for doing that was to grow our share of price comparison sales so that our businesses can grow not just on the back of price comparison growth, but on the back of taking an increased share of price comparison. We've seen that increased investment feed through into increased brand awareness, where you have here 2016 brand awareness versus 2015, indexed, at 2015, at 100. And this isn't just an ego thing. We're doing it because it has a direct and measurable impact on outcomes. One of the really important outcomes that we measure is people's willingness to go below the cheapest price on price comparison to buy from one of our brands. The example you see here is for Qualitas Auto, which is the biggest of our two brands in Spain, which has seen a 25% increase in the number of people clicking through from second or third place to buy from our brand. In the U.S., the focus is on reducing losses. And the principle focus in that context is reducing the loss ratio. We're doing a lot in terms of claims and risk-selection initiatives. But one of the most important things we're doing is simply increasing prices, very materially. On the left-hand side you can see the price increases in blue, for ourselves in our three key states, which between them account for well over 90% of our premium. In red, you see the average increase for 2016 for the top 10 insurers. And you can see that we are substantially outpacing the market in terms of price increases. On the right-hand side, you see the Elephant loss ratio in 2016, a 3% reduction, from 83.5% to 80.5%. You might say that's a bit disappointing in the context of these price increases on the left. Well, actually, the price increases on the left have yet to fully feed through in terms of earned premium. Some of them were -- many of them were back ended in 2016. And there is also the influence of a particularly expensive hail season in Texas, which is perhaps masking some of the improvements, which we would expect to feed through in 2017, provided that hail reverts to a more normal level. Across international, substantial growth in all countries. Investing in Europe as it moves towards being a valuable and exciting part of the business, and an example of that is in stronger brand awareness; and working on better margins in the U.S. Cristina mentioned earlier that the UK came second in the UK competition for best places to work, while Italy, chip off the old block, for the second year in a row, came second in the Italian competition. That makes us very proud and very convinced of the sustainability of its model, going forward. There we have a set of results influenced by Ogden. But looking through, I think, reassuring in terms of the long-term health of the business. The impact of Ogden should be one-off, provided that the market responds to the increased claims costs. And the analysis that we shared with you on market profitability gives us confidence that the market will feed the cost of Ogden through in premium increases, potentially in two bites, some of it now, to reflect the cost of the primary insurers; and some of it later in the year, particularly at the back end of the year, to reflect increased reinsurance premiums, which are inevitable. That process will start, perhaps, with esure's renewal at the end of the first half, with some other small renewals in the autumn. But most of the reinsurance renewals won't happen until December. I can't complete this presentation without referring to our outgoing Chairman, Alastair Lyons. There's no individual, I think, or very few individuals, who deserve as much credit for Admiral's success as Alastair. The Board, as a whole, and Henry and myself, in particular, have benefited from his wisdom, and diligence, and worldliness and, in moments of crisis, his composure. We're very grateful to have had him as Chairman. He joined back in 2000, when we were a second-tier UK-only car insurance only operator. Over the 17 years in which he's been Chairman, he's transformed, or overseen the transformation of, the business to being a substantially larger and more interesting business in almost every respect 10 times as big in turnover, 12 times as big in profit, even allowing for the Ogden impact. In fact, the only thing that really hasn't changed substantially in my view over those 17 years is Alastair's youthful good looks. It's sad that Alastair isn't here. He would have been here a week ago. Had he been here, I might not have done that good looks joke. But I'm very glad to say that we do have on hand, literally on hand, Annette Court, Alastair's successor, who joined the Board five years ago and immediately made a strong and positive impact, and brings to Admiral a wealth of insurance expertise and experience, and experience and expertise beyond insurance. We're all confident that over the next 17 years she will preside over a further 10-fold increase in turnover and 12-fold increase in profit. Thank you very much. Any questions. Q - Unidentified Analyst: [Multiple speakers] Three questions. Firstly, starting with reinsurance, given the commentary around the excess of loss, would you be looking at taking that further up, as in lowering your retention even further? Also, on reinsurance, your comments that you don't think there's going to be a material change from a negotiation perspective, if you could illustrate a bit more there? Second, on UK home, could you quantify what the expense ratio is now you've got a market share of maybe slightly above (inaudible). And I'm assuming as scale develops that should improve further. Thirdly, on the US aggregator business, it seems, in the Chairman's commentary, there was a talk around the partnerships. Could you, one, clarify what you guys are thinking? Secondly, are there states where you're close to breakeven? Thanks.

I'll do XOL, Geraint's going to do proportional, and Cristina will do home. Martin, are you able to do compare? Excess of loss, we would anticipate, depending on how prices change, in our excess of loss program we'll stay pretty much unchanged in 2018 and beyond, because we would expect the prices to adjust to reflect the underlying costs.

On quota share, the live quota share that we have in place, which covers 2015 and 2016, we expect both of those underwriting years will be profitable underwriting years, and so the quota share reinsurers will get their margins and make their return. The negotiations for future reinsurers will depend on the reinsurers' confidence in our ability to continue to deliver the outcome for them. We demonstrate that we put prices through and the impact has largely been neutralized, or fully neutralized, and I suspect it won't have a strong bearing on those negotiations. Cristina?

The reported expense ratio in 2015 was 33%, and in 2016 it was 34%. You need to take into account that in that 34% there are 3 points of the truck related, so it's around 31% without it, on a comparable basis. And, as a reference, the expense ratio in the market is around 50%, so it's a significant gap.

For compare.com in the US, so you were referring to the partnership, so I assume it's like the partnership we have with insurers. And we've seen more and more the panel growing quite substantially in 2016, and we see that. Still more important than number of partners we have in panel is the number of prices with these state customers. And we see the numbers going up quite dramatically on the basis of new insurers joining the panel, but also current insurers on the panel expanding their footprint to other states or other type of risk. And we see that as a good sign of insurers' adoption of the model in the US. And that's what we continue to work on in the next few months and years.

Sorry, can we leave that to Andrew's presentation at the half year. As Martin mentioned, we, in the second half, focused on some key states, and that has proved to be a very positive move. But beyond that, I'd like to wait until Andrew's available to answer further questions.

David Bracewell, Redburn. Two questions. On the US pricing movements that you put through, is that because you got your pricing wrong, historically? Or this is in anticipation of what you see in terms of higher claims costs and you're moving ahead of the market on those price rises? And on the second question, on the UK car insurance, you put through significant policy growth, I think it's about 11% year on year, yet profits still seem to be broadly flat, perhaps slightly down. I appreciate there's an earnings drag, and there's a reserving, etc., so perhaps you could talk about why you've got top-line growth and no bottom-line growth yet on the UK?

I'll do the second one, first. The biggest single impact is the impact on the comparable of the computation of 2014, which is somewhere between 35 million and 40 million, which makes 2015 looks better than 2016, or makes 2016 look worse versus 2015. And we can talk more about that offline, if you'd like to do that. In the US, there are two things going on. Yes, there is an unusual level of claim inflation in the US, and we are very keen to make sure we move ahead of any claim inflation, rather than wait for it. But also, I think there is a strong belief that in order to prove the model in the US we need to have a loss ratio which is a sustainable long-term loss ratio. And a loss ratio in the mid-80s%, like 2014, simply isn't really going to give us the outcome we need. So we're driving hard for a sustainable loss ratio for the long term which is somewhere in the 70s% as an important step towards proving the model.

Thomas Seidl, Bernstein. First question on, also, the US price comparison. You showed $10 million lower loss for 2016, is this mainly due to the cut in marketing, or is it also due to substantial growth in quotes and conversion? Secondly, on the reserve strength, I think in the press release you said that you were not willing to absorb the Ogden intake in reserves; now you said the reserve spread is unchanged. Can you just confirm that there is no change in the reserve level, and this is really fully flowing through the P&L? Thirdly, on the home business in the UK, can you talk us through the market share on the price comparison? Are you mainly gaining from the growth in the price comparison penetration, or you're also taking share within this?

Martin, Geraint, and Cristina?

Start with compare.com and the better results we had in the second half of the year, which is mainly marketing efficiencies. We spend less, as said, because we choose, in some geographies, where we wanted to be more targeting and focus, and that brought much better acquisition cost in those states. So, yes, we had a very much better second half, and limited losses in the second half of the year.

Reserve strength, the buffer in the booked reserve is above the best estimates that we show you earlier is unchanged, relatively speaking, year-on-year. And so, we haven't reduced the size of the buffer in respect of Ogden. The P&L impact this year is just over £100 million pre-tax, and there's about £85 million pre-tax left to come. Some of that will come through in profit commission, so it's not necessarily a reserving thing. So it's profit commission that we will -- we do expect to earn on the most recent underwriting years, but aren't yet recognizing in the accounting. And there'll ultimately be less of that than there was previous exchange in the Ogden discoveries. So some of it is profit commission, as well, which will come through in the next few years.

And on household, there are three sources of growth last year. The first one is our market price comparison, which is the majority of the sales, has grown. Second one is we have grown the number of sales that we make in that market, so we're growing our share of price comparison. And the third one is the direct sales. So still, even though price comparison accounts for the majority, 83% of our sales in the second half of the year, we are increasing the number of people that come directly to us because of our direct proposition. But also, we increase the number of cross-sell, so we manage to sell household policies to our existing motor customers. But of all of these three items, the biggest growth has come from us growing our percentage, our share in the price comparison business.

Andreas Van Embden, Peel Hunt. Two questions, please. Coming back to solvency, your nice table on slide number 8, I know you've -- I don't want to go into too much detail, but I see you've given some quantification in terms of plus-plus and pluses in terms of the overall positive contribution to your solvency ratio from economic profit, PPOs, etc. Could you maybe attach a percentage contribution to those pluses as well, please? And on the PPO propensity, what assumptions have you made there? One third of your reserves are large BI, part of that would be PPO, part of that would be Ogden, do you know more or less the split, top of your head? Second question, on price comparison, I see that your -- the quotes you get through the system have gone up again. I just wonder whether you could comment on your conversion rates; what the level is, and whether they've moved up in the past year. And also, looking strategically at all your price comparison businesses, in Europe, in particular, what are the synergies between them? Do you manage them as an integrated business, or are they four standalone businesses with no internal synergies? Or are the synergies in technology, pricing techniques, and particularly data? Do you aggregate data? And do you use that across your four businesses? Or is it just four separate silos?

Let me stop at three, partly because we'll forget. Geraint, do you want to do the first one?

Solvency, we're not going to give percentages at this point. We've given qualitative indicators of how these things contribute to the movements. And there is so much going on we could spend a lot of time talking about, so at this point we're not going split the debt. One thing I would say is that the profit and the dividend basically offset each other. So the other movements are what are driving the change half year to full year. PPO propensity, two things to say, I think if the new discount rate stays at minus -- if the discount rate stays at minus 0.75% we largely expect PPOs to die away. I think if you talk to a claimant solicitor, they'll recommend no one takes a PPO with a discount rate of minus 0.75%. We've made two changes in PPO propensity. The first was pre-Ogden, where, in line with what we see in the markets and the third-party working party studies, we reduced the level of propensity in our reserves to be in line with that market benchmark. And that's above what we see in our data. So based on the small samples, actually, I think the right thing to do is to go on with that market benchmark. And then, post-Ogden, we've effectively something less than halve the number of cases we expect to settle as PPO. So we've still got a reasonable portion of cases in our balance sheet reserve, and they will settle as PPO, despite the fact, you might well argue, that's a very cautious assumption in this new environment. The split in the reserves, I'll do it off the top of my head and what we share in exactly, what is PPO, what is a lump sum Ogden related, and so on.

Okay. So price comparison in Europe, also, we run these businesses quite independently to fit a local market and to pursue the independent strategy. There are clearly some synergies in term of IT, in terms of techniques, but all in terms of experience sharing. We run price comparison in the UK for about 15 years, we run price comparison in Spain for about eight years. There's a lot of knowledge and experience that we've gained and that we share along this kind of different operation. But mainly, that's the three aspects of the synergies: on IT techniques, but mainly on sharing knowledge and experience. You had another question on conversion rates, but I'm not sure in which country you're, in the US or in the UK. In the UK, we have seen conversion rate, because they've quite a mature market, it's overall quite stable. There is some movement, but it's quite stable. It's going slightly up, but more because the market is quite dynamic, especially with all the price changes that insurers have put through in the last few months, almost year now. We've seen more switching behavior in the UK market, which improves conversion rates, but it's slight.

Andrew Crean, Autonomous. Can I ask in two areas? Firstly, on the excess capital and special dividends relative to excess capital, you obviously suspended those until you got a clearer view of your internal model and Ogden. From my understanding, you're fairly clear that the internal model will be similar to your current position and Ogden could well improve in the autumn. So I don't quite understand why you're so conservative and I'd like to have a view from you as to what you think may happen to Ogden minus 0.75%? Secondly, on the investments overseas, I think, in Europe you said that you're on the cusp of creating substantial value, after 10 years. Could you give us some sense of that, and what your strategy might be to those different operations between hold or sell? And then, finally, just as the European businesses are coming into profit the US is making substantial losses. You give us a figure every year for how much you're going to spend on compare.com. Are you prepared to say whether you'll be breaking even beyond this year, or whether we're going to have to suffer further losses?

If I do the speculation on potential future changes on Ogden, before sending it back to Geraint for his comments on conservatism in the suspension context, and then I'll probably pick up all the international ones. So the government announced the move from 2.5% to minus 0.75%, making the point that they had no choice other than to do that, in their view, and then promptly announced that they were going to review the whole process and rationale. We are very committed to the appropriateness of compensation at the right level for claimants often with very serious injuries. But minus 0.75% is an irrational, crazy, as the ABI called it, outcome and we are convinced that any review should lead to a much more rationale method of setting it and rate it's set at. Having said that, rationality doesn't always triumph on these occasions, as has been evidenced over the last few weeks, and the government is very congested in terms of its timetable. This will require legislative change. And whether, in fact, enough space is made in the timetable to address this issue, which is of vital importance to us, it's actually very important to employees across the land, who have got public and employer liability, it's important to the government in terms of NHS and local authority liabilities. But I think it's very evenly balanced as to whether, actually, a rational outcome is arrived at, and in any way will take 9 months to 12 months to seek clarification on that, in our view. Any further comment, Geraint?

No, not on Ogden. On special dividends, I think our philosophy is unchanged: we won't keep surplus capital in the Company. But to make that call we have to have certainty over the amount of capital that is surplus, and so we will wait until we're into internal model environment before making a call on future returns of surplus capital. The position in the internal model environment is not fully in our grasp. So we have a view that it shouldn't change materially from what we see today. It's already on the standard form of a plus and add on that we calculate internally and, hopefully, agree with the PRA. But, ultimately, the PRA will decide on whether the internal model capital position is what we see it as, or what they see it as. And we don't have that clarity yet, and won't until into next year.

On international, I would say that it is my belief that the European operations are on the cusp of being of material value. To combine 30% growth with getting very close to break-even is an unusual and difficult combination in the world of insurance. That's not to say it will make money next year, because it may well choose another year of substantial growth and marginal losses. Would we look to sell them? Henry used to say everything's for sale, apart from his wife. In the interest of marital harmony at home, I think I ought to make the same point. At the right price, everything's for sale. But, in truth, it's not our intention to sell them. It's not why we created them. We think that the value that's potentially in these businesses is a long way from being recognized by our shareholders and probably, by extension, by potential buyers, but we don't discount the possibility. In the US, in the case of Elephant, our direct insurance operation, we are accelerating moves to make sure the economics are better than they've been historically. Minus 22%, which is what the whole account result was in 2016, it isn't good enough in the long term and so you see that in the form of price increases and a move to particularly improve the loss ratio. And I think that opens the door on a business which is clearly delivering lower losses and more ultimate potential value. On compare, it's been an encouraging six months. We have, it's suggested that the losses next year will be between $15 million and $25 million, and that is a substantial amount of money. But we feel that the size of the prize here is potentially very, very substantial and it would be a mistake, given the progress we are making, not to continue further to bring to the US something which is quite unusual in the US context in terms of distribution of car insurance. The amount of money that is spent on acquiring new car insurance customers in the US is so huge that our ability to take even a small slice of that pie in the medium term and create a business like Confused, like Rastreator, is such an attractive prospect that we will persevere for a while yet. And that does not mean, that means we cannot guarantee there won't be losses into 2018.

Iain Pearce, Berenberg. One of the slides you had showed part of your loss ratio advantage was from handling of large bodily injury claims. Could you just outline what you do differently from some of your peers, and whether PPO propensity will affect that competitive advantage, perhaps? And then, on Confused.com, quite a few of your peers have shown good growth in lines outside of car insurance, home insurance, credit cards, this sort of thing. Are there any plans to move into those areas as well, or is it mainly to focus on car insurance there?

Cristina, do you want to do the claims one?

Yes. On large BI's, we do a combination of things. The first one is that we stay very close to the claim during its whole life. So our lawyers tell us, because we don't have any market comparable data, that in other cases people outsource the claim to these law firms, whereas we are very involved at every stage of the claim. And I think that is quite important. Secondly, we have a commitment to try to settle claims as quickly as possible, as long as it fits with the claim, in that case, and with the claimant. But it's something that we care about, is striving to settlement quickly. And the third one, I think, is our staff. We're based in South Wales. All offices -- we have three offices, and they're very close to each other. So we have a very close team of people, who also have an interest because they are shareholders in the Company, and they feel very proud to feel there. So they care about the result in terms of customer satisfaction, but also in terms of handling claim efficiently. And it's a combination of both those three give us a superior advantage.

For Confused.com, clearly, our strategy is to be driver-centric and not car insurance-centric. And driver-centric means a lot more than car insurance. And that's why I tried to demonstrate earlier with car finance, for example, having a vertical that we see great potential, car buying, everything that is car related. That we see more growth potential in this kind of verticals than in [indiscernible] or credit card for Confused.com, because what matters for us, really, to differentiate ourselves from our peers.

Ed Morris, JPMorgan. Two questions, please. First is just on the growth outlook in the UK. I appreciate your comments about Q1 being a little bit slower because you adjusted prices quicker than everyone else. Is your anticipation, if the market reacts, that your customer numbers can continue the trajectory you've seen for the last couple of years? Just some thoughts there. And the second question is on some of the new ventures that you talked about at the half year stage, Admiral loans, for example. Can you give us a quick update on what you've learned over the last six months, and what future you see there?

I think, had it not been for Ogden, we would have been guiding towards growth in the UK, but not at the 11% rate. In terms of timing, that would have been what we would have expected to be optimal. In the context of Ogden, obviously, it's been a slow start to the year because we moved first. The balance of the year, it'll depend a bit on whether -- how quickly people move; how much of the reinsurance element they incorporate in their prices immediately; to what extent some of the ones with lower retention decide to try and take some more volume during the next nine months. So, probably it will be a lower year of growth for us than it would have been pre-Ogden. But then, in 2018 those competitors with a low retention will have pressures themselves and so some of that will unwind in 2018. And the other point was loans. Yes, we had been piloting loans during the course of 2016. We made a decision to up our investment in terms of the size of the team and the IT support for it, and we're going to launch a proper loan product, on a free standing IT system, within the next few weeks. Because the pilot gave us enough belief that the brand works, it works for our customer base and there's an opportunity, not in the short term; this is a very long term play where we're planting an acorn here that we expect to be an oak tree. But that takes a while, as you know, and so it won't have a material impact on 2017, apart from driving a small loss as we set up the business.

Thank you. Jonas [indiscernible] from [indiscernible] Capital. I have two questions. The first one is you have a very high payout ratio. Why are you not using part of those funds in fueling the growth of the growing businesses? The second question is which are your toughest competitors, the ones that are doing maybe something better than you? What is that thing that they are doing better than you? And what are you doing to keep at the state-of-the-art? And the third question is what is your view on the self-driving cars movement? And how is it going to affect the car insurance market? And how can Admiral adapt, or take that opportunity? Thank you.

Okay, Geraint, will you do the capital one? And then, maybe, Martin and Cristina come up with one competitor from your areas, and I'll come up with one from international car insurance.

On the payout ratios and funding capital for growth, and so on, you're right to say that over the past 10, 12 years we've paid out something around 95% of our profits. And during that time, we've doubled or trebled the size of our UK car insurance business; we've grown the household business from nothing to nearly 0.5 million customers; we've set up five, we now have four international insurance operations, which have 0.75 million (inaudible) customers, and we've set up three or four price comparison operations. And so, I wouldn't exactly say we're starving new operations of capital. They present budgets, and we discuss, and we allow our businesses to grow at the rate they think is appropriate, depending on their individual circumstances, the market they operate in. And so, we are quite comfortable that they get the amount of capital they need to grow in the way that we think is most appropriate, and they think is most appropriate. And having the combination of paying out nearly all the profits and growing all these operations from scratch to where they are today is quite a nice combination. Christina?

Yes. In the UK car, if you look at the trend in the past few years you'll see that they are the specialized motor players, the ones who are growing faster. If I have to choose one of them, I think I'll choose Hastings, that they're growing quite healthy in the past couple of years. I think that they are moving to a new system, and they are focused very much on online, so very interesting things that they are doing. On home, I would choose Legal & General. They have been growing a lot in the past, and they are doing interesting things. So, that would be my two.

It's a very different question because we the only group that operates in four different markets. So in each market different comparisons. But in UK, for example, I think that the main competitors remain comparethemarket, on the back of their very successful, but unique, meerkat campaign that really had a grasp people attention over the last six or seven years now.

Internationally, in some of the markets there are some really good low-cost players, GEICO in America, Mutua Madrilena in Spain, which represent challenges and are players we can learn from. In terms of self-driving cars, in the short run, we think the car insurance market, globally, is set to grow. The US market alone has grown $20 billion in the last two years. But that's going to probably continue for the next 5 or 10 years. It's going to plateau as semi-autonomous, and then self-driving, cars emerge. And we expect the car insurance market to start to shrink 20, 25 years' time. And that's partly why we're doing the home insurance investments; why we're doing the lending investments, to have other lines of business that are in place, and have the potential to be substantial, which can grow into replace any shrinkage of car insurance. But that is quite a way off. Our average car age is 8.5 years. It takes a long while for these changes to feed through to the car pool.

I'll say this very quickly. The first question, I think how you laid out the Ogden impact on claims is really, really helpful. I was just wondering whether you might be able to give us a similar idea on what the, potentially, positive effect of whiplash reform, later this year, might be, so trying to think about the two.

Whiplash reforms can be very substantial in their impact. We're not expecting them to actually emerge until later next year. So we think October 2018 is the current view on when they might get implemented. And we would expect their materiality to be similar, but in the opposite direction, as the Ogden reform.

Will Hardcastle, Twelve Capital. Can I have one last stab at page 8, on the [roll forwards] of Solvency II, and it might be something that I've done wrong? Thinking about if economic profit and dividend knock each other out, effectively, do Ogden and PPO adjustment broadly knock each other out as well and, therefore, your Solvency II is increased because of the volatility in yield curve?

It depends on which order you do the math in, I think. You might say that the top two cancel each other out, the profit and the dividend, the volatility adjustment and the yield curve cancel out the Ogden basis, and what's left is the PPO adjustment. So it depends on the order you do them. We're not giving the quantum.

Okay. So, broadly, they'd be knocking, that would be about right. Because am I looking at it right in terms of eligible own funds was £850 million at half year, is that a like-for-like number, and has since gone up to £1.07 billion?

That's pre-dividend. I think it was £850 million after dividend. So the £0.93 billion versus £0.85 billion at the half-year number.

Perfect.

Thank you very much for coming on this early results. And enjoy the rest of the day. Good luck with the Budget. Let's hope IPT doesn't move.