David Stevens - CEO Geraint Jones - CFO Alistair Hargreaves - Head of Service Cristina Nestares - Head of Product Noel Summerfield - Head, Household Martin Coriat - CEO, Confused.com Henry Engelhardt - Founder

Dhruv Gahlaut - HSBC Ravi Tanna - Goldman Sachs David Bracewell - Redburn Olivia Brindle - Bank of America Cameron Hussein - RBC Jame Shuck - UBS Thomas Seidl - Bernstein Greig Paterson - KBW

Good morning and welcome to our first half 2016 results. I’m David Stevens, CEO. This is my first set of results. I took over from Henry in May this year, probably 30 to be precise. And it’s lovely to be able to stand up in front of you and talking about such a good set of results and the Company’s such a robust state of health. My collogues and myself are going to talk about that in more detail. Geraint will start with an overview, then Alistair and Cristina two lucky managers from the UK operation will talk about the UK. Martin and myself will talk about price comparison in our international operations before I do a wrap up, over to you Geraint.

Thanks, David. Good morning everyone. I’m going to come with the results, then we want to talk about solvency and the dividend and finish up briefly on investment. So we’re very pleased with the first half’s results, you’ll see strong top-line growth with UK and at sight a record first half profit, a record interim dividend and strong solvency ratio will be impacted by big moves in interest rates and good progress in our operations outside the UK. This slide shows some of the highlights. At the top very strong growth in customers and turnover, we had 15% more customers at 30 June, 2016, 4.8 million in total. We’re going to move past 5 million customers around the group in the next couple of weeks, that growth came from all over the group, notably in the UK, where we added 0.5 million customers over the past 12 months. Turnover was also up strongly, maybe 20% higher at £1.3 billion pounds, double digit growth in all segments. Both of those figures are new records for first half. Pre-tax profits and EPS both up to record levels and return on equity still pretty amiable 49% which is in line with 2015 as a whole. And finally, as mentioned earlier an interim dividend of 62.9 pence per share, that’s a record for interim and the big increase compared to 2015 results from a second installment additional capital return of £$33 million, which we’ll cover shortly. Let’s take a look at that top-line growth in a bit more detail. So the left hand side of this chart shows turnover by segment and the right hand side shows customer numbers by segments, very nice to see lots of upward arrows on the slide. UK car insurance firstly, Cristina and Alistair are going to talk about this in detail, so I’ll just say that favorable market conditions meant, we were able to grow in the UK to nearly £1 billion of turnover in the first half and at 340,000 customers over the year to move past 3.5 million customers, so 11% year-on-year growth was the fastest rate of growth since 2011. Outside the UK, our insurance business grew their turnover by 44% to 159 million, it was a big step up on the 2015 growth rate. They also added about 125,000 customers over the past year to move past three quarters of a million that’s 15% of the group’s total customers. Household insurance and van also growing strongly, adding 160,000 customers, household approaching 400,000 customers at the end of June continued its strong growth. Last but not least, good growth in price comparison turnover 16%, including a very good first half result from Confused.com which Martin will talk about shortly. So moving on to profit, this slide analysis how the different business areas contribute to the overall result of £193 million for this half year. Moves to price see that the dark blue which is UK car insurance result, continues to dominate with a slightly higher profit in H1, £221 million more of which later. The proportion of the total is down slightly as you can see and that’s due to a better result from price comparison and other charges. International insurance is in the green, combined loss there was at the touch end of £13 million, which is a bit higher than the £11 million in the first half of last year. That result includes a further profit from ConTe in Italy that’s offset by continued investments particularly in the U.S. and France. Positive combined ratio trajectory within those operations continues especially in Europe. Price comparison in total made a loss of about £1 million, which is better than the £4 million of H1 last year. That’s largely made up of continued investment in compare.com in the U.S. where our share lost were steady at about £10 million in H1 and a good from Confused.com which improved to £8.3 million from £4.8 million last year. Finally in the yellow that’s the other segment, which includes household which made another profit while its growing very strongly and also includes other costs such as share scheme charges, debt interest and there were also some gains on forward foreign exchange contracts in the first half amounting to £5 million, those gains will reverse over the next year or two. Turning now to our capital position and then on to dividends, this slide sets over the current solvency position. Just as a reminder our capital requirement is currently based on the solvency II standard formula plus capital add on that we agreed with the PLA towards the end of last year. The position remains strong as you can see with post dividend solvency coverage of 180%. The dividend you see in the slight blue very shortly. Now, after the Brexit vote, the solvency ratio would have remained basically flat compared to the end of 2015, capital generation minus the interim dividend minus the effects of the pre-Brexit moves in interest rates at basically net off to pretty much zero. The fail of significant moves there in the yield curve since Brexit have negatively impacted the ratio. Just as a quick reminder the yield curve is used to discount technical provisions notably actual and set and potential PPO claims and the risk margin on our solvency II balance sheet, so lower interest rate means, higher long-term liability evaluation and bigger risk margin and increased capital requirements with some offsetting moves in fixed income asset values. So the combined impact of the post Brexit moves in yield curve was a reduction of around 20 percentage points in the solvency ratio. We cannot use the standard yield curve in our calculations, but we have applications with our European regulators, they use the volatility adjusted curve, which allows a more appropriate discount which you can see on the slide the adjusted solvency ratios and even stronger 196%. So we expect those applications to be reviewed shortly and hopefully approved during H2.In the back of the pack we include solvency sensitivities including further yield curve movements. No change in our plans for the partial internal model, the program progresses well and we’re on track to make our application during 2017. Once the model is approved, we’ll confirm our targets for solvency coverage and as you see on the slide, no change currently in our expectations to where that targets might sit. I’m sure you’ll remember with our 2015, we reported that we moved into solvency II with surplus capital and that we intended to return that surplus to our shareholders in the form of additional special dividends over the first two to three years of solvency II being in effect. That’s broadly after the point where our partial internal model is approved and in use. So just to update things six months on, we still have surplus capital as you saw in the previous slide with the adjusted solvency ratio of 196% or 180% from the standard yield curve. And our commitment to return surplus capital also remains, though the market conditions in H1, notably interest rates moves means, that we now project that the total of additional special dividends will be in the range of £100 million to £150 million, which is down from the £150 million to £200 million that we talked about six months earlier based on locking additions at that point in time. Those figures include the £33 million we’re declaring today and the £32 million that we paid earlier this year with the final 2015 dividend. But based on what we see today, we envision the final 2016 dividend in six months’ time will include about £20 million of additional capital return based on that new range. And just to reiterate, this is something we’ll continuously reassess as we move to the next 12 to 24 months and according to market conditions have notably interest rates take change or business plans change or other circumstance change and these numbers could change. Also important to repeat our expectations for the foreseeable future is to carry on paying underlying dividends in the order of 90% to 95% of the earnings, that doesn’t change. Turning next to the interim dividend, so like six months ago we split our interim dividend into three parts; first up, the normal dividend which was 65% of post-tax profits, then the first part of the special, which combined with a normal gets us to the familiar 90% to 95% level and then the additional capital return in the light blue that sits on top. The actual amounts as you see it on the slide is 35.9 pence per share for the normal part, 15.1 pence for the first part of the special, add those two together you get the 51 pence this year, which is the H1 payout ratio of 91%. Then the additional return that’s the second installment of the £33 million at 11.9 pence per share. Grand total of 62.9 pence which gets paid on October 7 and the dates you see there on the slide along with our dividend policy. Just quickly this slide shows our dividend history, as a you can see a pretty impressive record of growth slightly moves in line with earnings overtime, first two installments of the additional capital return you see in the yellow on the top of the final two bars and in the box we have pated the totals that we’ve paid out since flotation back in 2004, was £2.2 billion of the dividends in total, split broadly evenly between the normal and the special parts. Final slide for me is on investment, so on the left hand side of the slide we show credit quality and the allocation between cash tree of our assets at the end of last year and the middle half of this year and as you can see there’s basically no significant change in either since the end of 2015. On the right hand side, you see how investment income has moved over the past few half years and as we flagged over the past year or so, 2015 number was particularly was sorted by negative adjustments relating to withheld quote of share reinsurance funds and that accrual is now being fully released and the negative 5 million in H1 ‘15 being replaced by a positive 9 million in H1 this year. The current period also benefits from the £5 million or so of foreign exchange gains I mentioned earlier. Happy to discuss that with anyone later on if you’d like more detail. That’s it from me. Next up we’re going to hear about the UK insurance business from Alistair and Cristina. Thank you.

Good morning everyone. My name is Alistair Hargreaves, I’ve been with Admiral for around eight years working in a number of different areas of the UK business. I’m currently the Head of Service, responsible for operations claims and marketing. I’m going to talk about the UK motor market and our performance and then Cristina will talk about some exciting aspects of motor and also household. So let’s start with the results for the market as a whole, here we see the market net combined ratio for 2015 at 102%. Compared to the last 30 years, it’s one of the most profitable, but it’s driven by 12% reserve release. By adding back the reserve release, we get a gross combined ratio for 2015 of 114% and that’s used to see the overall profitability of business written in 2015. If we start with the gross combined ratio of 114% and take away an estimated ancillary profit of 10% of premium and investment income of 3%, we get an all-in combined ratio of 101%. Basically without reserve releases the market is only just breaking even. Now, history suggests that in 2016 reserve releases will be lower, so market profitability will dependent on current year business or in other words, there’s plenty of motivation for increases in premium. So let’s take a look at what’s happening with premiums in the market. The ABI index shows increase in premiums written up 6% year-on-year. Earned premium, which is what flows through to market profitability is also up 6%. Now earned premium is stable 11% below the first quarter of 2012, but 2016 is the first year in five years that we’ll see an increase in earned premium. We expect to see more premium increases in the second half, the volume of increases and the 2016 market results is dependent on three factors, first, the amount of reserve releases still in the tank; second, large BI inflation and third, the rationality of the market as a whole. As shown on the last slide, historically the motor market cycle has increased well beyond the gross combined ratio of 114%. This time we’re seeing premiums increase already earlier than was the case of previous turns. Is this an indication of a slightly less volatile cycle going forward? Now, the price increases are leading to more people shopping for that car insurance and gross an increase in new business volumes for the market as a whole. Here you can see changes in our price competitiveness as measured by times talk on price comparison indexed to 100 in December 2013. Our competitiveness dropped in 2014, then we regained it and we’ve maintained it for the last 12 months. To maintain our competitiveness we’ve been increasing prices in line with the market. This combined with people shopping helped us grow new business volumes. At the same time I’m pleased to say, we’ve maintained our customer attention and renewed volumes. The end result is our 11% vehicle growth. So let’s look at loss ratios. Here we’re comparing the reported results for the market on an accident year basis with Admiral’s ultimate loss ratios. You can see some very positive movements in ultimate loss ratios on our back years. The 2015 year is at a level to 2014 at the same point. It’s a relatively conservative early pick reflecting the inherent uncertainty around large BI. But why is the gap between the two lines narrowing? Well, there are two potential answers to that. The first is that our competitors are becoming more disciplined and this has led them to improve faster than we have. The second is conservatives within our ultimate loss ratios. These have historically improved over time. Here you see that 18 months after the start of the accident year, our 2015 ultimate loss ratio is 6 points better than the market. This is comparable with previous experience. For example, at the same point 2013 was 7 points better than market. 2013 is now 11 points better. Only time will tell how 2014 and ‘15 mature. We don’t publish a loss ratio for 2016 this early, but we’ve been increasing premiums and seen moderate claims inflation, so a positive start. Now, how do those loss ratio improvements translate into reserve releases? On the left hand side, we’re looking at releases on our retained share. The positive developments on our back years have contributed to substantial reserve releases. In addition, the buffer we hold against best estimates also remain strong and at the top end of our range. What are the prospects for releases in the second half? Well, they should remain significant partly because the 2015 ultimate loss ratio is relatively conservative. The chart on the right hand side shows retained or commuted reserve releases. We typically choose to commute our reinsurance agreements two years after the start of the underwriting year when we’re confident in final outcome. In the first half we commuted the 2014 underwriting year, this was booked to the combined ratio of over 100%. Upon commutation we wrote off reinsurance recoveries relating to 2014, which reduced commuted reserve releases. However, on an ultimate basis, 2014 is profitable or in other words, we expect to future commuted reserve releases from 2014. Our key piece of the puzzle is our expense ratio. We’ve increased our new business volumes resulting in increased payments and price comparison sites and an increase in expense ratio from 16.2% to 17%. Overall this graph shows, we’ve continued to maintain our expense ratio advantage, which is very important to us. So how does all this come together? On this slide, you can see the increase in customer numbers, it’s flowing through its increases in premiums and other revenue and resulting in 16% increase in turnover. We have a strong combined ratio due to improvements in our loss ratios, which more than offset the slight increase in expense ratio. As Geraint explained, we’ve also benefited from a positive investment return. These positive impacts have reduced by lower commuted reserve releases due to the 2014 commutation, but overall profits are up 2%. That’s all from me. I’ll now hand over to Cristina.

Good morning everybody. I’ve been here presenting the results of our European operations a number of times. But this time I’m going to focus on a few aspects of our UK operation telematics, our new IT system guide work and the household book. I moved to Cardiff a year ago and I’m Head of Product, which I think that includes pricing and analytics underwriting, but also ancillaries, telematics and household. Now, we’ve been asked a number of time, what are we doing on telematics, how big our book is? And I’m very pleased to say that Admiral has the largest telamatic book in the market. We have about 190,000 policies followed by insured books static line and hasting’s [ph]. Telmatics is still quite a niche product, it accounts for about 5% of total market share because we have to take into account that the telematics policies have a very high expense ratio due to the fact that the cost of device is also very significant. We have to pay for the books for installation and for data cost. But there is also another important cost which is the discount we give to the customer to sell these policies. So at the end of the day, we have this higher expense ratio which is compensated by other loss ratio. And Admiral has a strong interest in understanding telematics because we think this is a very important technology that is going to play a bigger role in the future. So having the largest data set in the market allow us to understand it better, allow us to price it better and understand to serve this product in the future. So let’s move to guidewire. We’ve been quite busy doing 2016, basically continuing our roll outs of our new IT system and today almost all of our policies are solved using guidewire. One of the main advantages that this system brings is that we can add new products more easily to the system. It also makes cross selling, the process of selling different products to the customer much easier and faster. Guidewire also brings operational efficiencies, today many processes are faster than with our old system I90, particularly mid-term changes and renewals. And the system also allows the faster development of changes in the future and it makes it easier to train staff. Let’s take a look now at the household market. The market continued to be profitable in 2015, but the combined ratio deteriorated due to the bad weather events during the last part of the year. 2016 also started with bad weather events which had an impact on the loss ratio and there is also pressure in the market on the expense ratio due to the Flood Re levy. The market remains quite competitive, but after a few periods suffering and decline, we’re starting to see some signs market stabilizing. And also we’re getting very good news coming from distribution, advocators have gotten 50% year-on-year and now they account for about 50% of total new sales. So in this context our results have been quite good. We’ve continued growing our book and we now have about 380,000 policies, which represent the growth of 23% since the end of last year. Also our combined ratio continues to improve. We finished 2015 just over 100% and during the first half of 2016, we’ve continued improving our combined ratio. Our loss ratio remains stable despite the bad weather events and our expense ratio continues to improve. We’re getting more efficiencies and we’re also getting a larger proportion of renewals. You don’t have it in the graph, but in the first half of 2016, our expense ratio finished at 30%, very similar level to 29% for the year full year 2015, but when you take into account that this 30% includes 5 points for Flood Re, our underlying expense ratio 25, which is a significant improvement from the last year. So we’re very pleased with the progress with our household book, we have our combined ratio very close to breakeven and we have our expense ratio result very much better than the market. And our loss ratio is higher than the market, but this is partly explained from the fact that we acquire most of our business from advocators. So for the future, we expect for the market and our turning prices, most of it [ph] and more growth for advocators and for the Admiral household book more growth come. So this is all from me and now I’ll pass you to Martin.

Thank you, Cristina. So my name is Martin Coriat, as you probably noticed, I’m French. I joined Admiral Group eight years ago and spent time on price comprising business lines both in France and UK and since 2013 I’ve been the CEO of Confused.com based in Cardiff. I’m here to present the result of our different price comparison operation. As you can see we’ve business up and running in several countries, in the UK, Spain, France and the U.S. and we’re working towards the launch of a new operation in China. We also partner with big name for this internationalization and that’s the reason why Admiral does not always own 100% of these businesses. As you can see we have now a nice portfolio of price comparison operation in a variety of countries in term of maturity and size. UK is probably the most mature market as far as price comparison and penetration is concerned, we still see a big potential in some under developed markets where price comparison is still a very new concept such as the U.S. or China. Let’s have a look in more detail of the performance of these businesses. So after of years of slow and limited growth in the UK market, the market is growing. This growth as you can appreciate at the bottom left of the slide is mainly driven by two factors; first, the turn of the insurance cycle as indicated by several indexes including our own Confused. com index and that’s stimulated the market. But on top of the insurance cycle per say is the increase of the insurance premium tax at the back end of last year given us a push to the aggregative market. Second, car sales hit an whole time peak in the first half of 2016 for both new cars and renewed cars [ph]. As you can imagine car buying is a natural trigger for insurance trigger and does definitely fuel the aggregative market. So overall it’s very pleasing for me to see this market growing again. However, you should not jump to the wrong conclusion the aggregative markets remain a very competitive place. If you’re watching TV you know that the main aggregator that’s spending a lot to grab customer attention, to grasp customer attention. In the first half of the year we estimate that the industry has spent just short of £70 million in media, that’s the second biggest amount in history. So media spend is one of many drivers of the level of competition in the market. Another evidence of the toughness of this market is the difficulty for new entrants to succeed in the market whether they name Google or Tesco. Overall the industry remains very competitive and challenging, but at least this competition occurs in a growing market. So how did Confuse.com perform in that environment? It was a good first half of year for the confused.com and we have pleasing numbers to report in term of customers and profit. The number of customer reported on the compare.com website in the first six month of the year has grown in line with markets with an 8% growth compare to last year. More importantly profit as risen significantly to reach £8.3 million for the first half of the year. So we are talking about a very pleasing 73% increase in profit compared to last year. Looking at drivers of the confused.com profits, there are two main factors to highlight on top of the customer approach. First is the better acquisition cost. As we experienced lower than usual inflation and we used smarter online acquisition strategy. Second is the better conversion, on the back of several and different promotion run on confused.com. Looking at what future has in store for confused.com. We have decided to focus on marketing efforts on the motor related product as it remains by far as a biggest market for aggregator. It is also in a real way confused.com has more legitimacy and expertise. Current driver-centric positioning is supported by a new advertisement campaign featuring James Gordon and it positioned confused.com as a car specialist. Moving to operations in continental Europe, we have seen some good performances from operation in both France and Spain. We have indeed seen some healthy numbers in term of volume with quote numbers up by 7% to reach 4.4 million quotes combined. Turnover has grown by 10% explained by the launch of new product as well as continued improvement in our customer experience. In Spain, Rastreator has experienced strong growth in the first half of the year across several product lines. Insurance, telephony and finance. Pursue of clear multi-product strategy and expand of product range. In France, LeLynx remains market leader for car insurance and continue to educate the consumer to benefit of comparing insurance prices. So overall it was a very positive start of the year for both Rastreator and LeLynx. As we continue to successfully build strong, profitable market leading businesses and by capitalizing on their expertise and solid brand in their respective countries. If we now go across the pomp to look at our U.S. price comparison operations that we launched back in 2013 this huge but under develop market in term of price comparison and option. As again see in graphs the operational metrics continue to improve dramatically. Let that three key metrics versus business, volume, acquisition cost and panel development as they all –and see how it is developed. The volume measures in quotes rose again and we reached a symbolic milestone earlier this year with more than 2 million quotes since inception. If you compare it to last year 2016 quotes volume have already exceeded 2015 totals. Acquisition cost measured by the cost per click for insurance has been reduced dramatically in the year. A panel development as we are committed to grow a panel of insurers is also going to the right direction. We have now partner with more than 80 car insurance rents including 13 of the top 25 carriers. That includes well-known brand name such as MetLife has just joined our panel, so overall essential and promising improvement to the metrics in our US operation. It is also interesting to note that in States where we operate for longer, we have better than this average metrics and economics numbers. That gives us confidence into success of operation but still requires time, patience and long-term investment and as a result we expect compare.com to incur losses in 2016 in a range of $30 million to $35 million. So this concludes very rapid two of our price comparison operation, overall a very positive first half of the year with encouraging development for all of price comparison businesses. That’s all for me and I’m happy to leave the stage to David who will present the latest results of international insurance division.

Thank you, Martin. And if you look at the international insurance operations on regional basis, Europe which is Italy, Spain and France and there is U.S. It has been a year of accelerating growth, direct insurance operations, the European operations account for 8% of the policy holders, about 60% of the premium, in U.S. 10% of the policy holders and 40% of the premium. European operations have tantalizingly closed to profitability, making a loss of $2 million in the first half of 2016. US operation, to an increased loss from 8 million to just under 11 million and I will expand on each region separately. So first looking at Europe, we saw here, a growth in customers of 19% and growth of premium of 30% and this country by country break down shows you that Italy dominates that growth, it has benefited from a 20% growth in the price comparison market in Italy and it has managed that growth of 21% in customers while also delivering it’s third successive profit, third successive reporting period profit. The other striking feature of this exhibit I think is the growth in the French operation, 74% in customers. You may recall that we initially outsourced our operations in claims in France and over the last 18 months, we have been in sourcing. We held back on growth until that’s been completed. That has been completed and we understand taking the shackles off in France and are looking to grow the business to its potential, I want to focus on one priority feature regions, not the only priority but a key priority. In the case of Europe, it is to expand our shares sales coming from price comparison as it takes a bigger share of all price comparison sales happen and there are three things we are doing to achieve that. We are widening our foot print where that’s possible and that’s the biggest example, that is France, we are improving our operation effectiveness to make sure that when someone does through from our website from price comparison side or when they phone as a result of quote on price comparison site, we are maximizing our effectiveness of turning that into sale, that is a key skill in these market. And lastly we are investing behind TV marketing and sponsorship to build some brand awareness. We have spent 2% of GWP in the first half of 2015 and 6% in the first half of 2016 and sponsorship. And the goal there is not so much to drive direct traffic, it is to persuade the customer that sees our price in top of price comparison list of prices, to persuade them to except that pricing to drop their insurer of traditional insurer or alternatively to persuade them if they see our price second or third to buy us rather than to buy the first price and part of that is about brand awareness. So moving on to the U.S., also a period of rapid growth, customers up 23%, premium very substantial 4% to 6%. The half had one notable event which was unusually dramatic hail season in Texas. Texas accounts to roughly half our business in U.S. and there is a hail season, typically in the second quarter of any individual year, which had it been a normal level of hail it would have cost us roughly $2 million and in the event it cost substantially more cost around $8 million, because there were a lot more of them and severity of events and location of the events unfortunately is rather more urban. What are we focusing on in this states, but one of the things we focusing on is improving our loss ratio, we have seen some rapid price increases in our U.S. operation, 15% in Texas, 8% in Virginia, we have also decided to reduce generosity of our renewal capping what we have seen in the U.S. market this year is a very good outcome in terms of acquisition economics. And those improved acquisition economics has given us bit more space to be less generous in terms of trying to maximize retention so we’ve concentrated more on increasing the way loss ratio and we have invested in the size of the team and in our rating software to accelerate the rates of pricing refinement in the U.S. market as we get more and more experienced as business growth. Other notable feature of the U.S. market is our expansion out in Tennessee and Indiana, two new states; we are going from four states to six states. Two advantage to this, this is a route to faster scale and the other advantage is it gives us a blank huge paper where we can trail different marketing approaches, both of those consider value to the long term health of the company. So that’s the direct insurance operations internationally, rapid growth, good progress. It seems to me appropriate given this is my first set of results to attempt a bit of big picture overview and this is the big picture. The value of our shares predominantly built around the value of our core business, our UK personal lines business and our strategy has two legs to it. One is to ensure that we maximize the value of that core business in the long term realize its full potential. And the other is that we create future sources of additional value. And I hope from the presentation today you gain confidence that we are doing what is necessary to maximize the value of our core business. I think I’d point to our telematics leadership, our better understanding of key technology, I’d point to our renewal of our core systems laying the foundation for the next 10 or 15 years of growth and I’d point to the fact that the franchise still has some room for growth. Half a million new customers puts on over the last 12 months, 35 million from the car insurance business and 150,000 from the house insurance business, so signs of health and progress on the core business. Homes are creating feature value, the opportunities we are familiar with, the direct insurance operations overseas and our price comparison businesses, notably our price comparison businesses overseas which are on early stage of development. All these businesses we feel have a very manageable downside and a very substantial potential upside. Lastly and perhaps, a word of point, you would necessarily have seen in the past, all the rest you had seen in the past is UK new product. We do believe that the skills we have and the assets we have, may allow us to develop a competitive advantage in some products beyond insurance. One of our objectives to next two or three years is to test whether there are such products particularly in the UK, products that play to our strengths, fulfill certain characteristics and likely to be business to consumer internet products, they are likely to be products that has an element of risk selection involved. They are likely to be a market which are experiencing or likely to experience disruption. And ideally they are not necessarily would also like them to be products that we can readily sell to our 4 million Admiral customers in UK. The first of these products and one I think that very much fix those criteria is personal lending so from quarter one of next year we will be trialing an Admiral loans product, primarily our car finance products, initially focused on our - primarily on our own customer base with potential to expand beyond that. We are going for a process Admiral way of testing and learning and understanding if there is an opportunity to create something interesting to create competitive advantage to create some differentiation in a world which we think may will be changing quite rapidly over the next 10 to 50 years and these are indeed long term plays. There might be another one coming around behind it, maybe two and those will be part of our focus to next two or three years. So all in all, very glad to be reporting very rapid growth, bigger profits, a sustained phenomenal return on equity and continued evidence of amazing cash generative powers of this business. And lastly I think I have to say with some pride, as a sort of parent looking on a child, Italy has achieved second best work place to work in 2016 eclipsing UK performance which is still very predictable of sixth best work place and I think also demonstrating how those cultural aspects of Admiral have been key to our success are being diffused throughout our operation. Thank you very much for your attention and we will open it up to questions. Q - Dhruv Gahlaut: Good morning guys. Dhruv Gahlaut from HSBC, a couple question firstly, could you tell us what is the duration of your active liability and how’s that - I mean what’s the mismatch there and any changes we should expect to the investment portfolio in terms of actions you take to reduce gap there. Secondly, with could you also update us, if you want to get the volatility adjustment approved what would the new sensitivity you will present given what you’ve given in the past the 26 percentage point. Thirdly in terms of US, is there any operations which you could think that’s now close to breakeven - as it actually related [ph] to California about two or three years back now, are we getting any close to breakeven levels the total number, we must be getting distorted by the new reasons you’re going through. And lastly, you mentioned that the reserve buffer is still fairly high as compared to the reserving policy, so what is the range as we should look at it and where is that number from that [ph]

The duration of our assets, all liabilities over the past six months. The liability side is the number of claims dominated by relatively short-tail stuff but a growing proportion of the reserve over the past five years is now coming from PPO claims, which obviously got much longer life. If you just look at peer straight average of asset and liability durations, not that great difference particularly we don’t have any part of our asset portfolio that’s helped into the 30, 40, 50 duration territory, can see impact on solvency ratios. So the asset side is two or three years longer stuff, we thought as the UK goes as you know. So it is not particularly big mismatch, we are looking whether it is make sense rest of move some of our assets, I think longer the stuff over the coming months and years without rushing into anything clearly, we’ll update you should we do anything. Next one is volatility, adjustment sensitivity. So the sensitivity I think remains basically the same just starts from high level.

On the reserve release I don’t think we disclose the actual range but the stock there, there is slightly down on the full year but it still towards the top end of the range. I think we have said the reserve release will average out about 14% over the time and that has not changed.

We are getting competitors coming [ph] as a result of U.S. It is true that U.S. is a big country that far as insurance concerned, it is more just the position of states, whatsoever countries. So we see by it, allow difference from one state to next performance, but we still consider the U.S. as one country and one operation and we are to grow and succeed in other country level not state level. So we only report the figures at the country level.

Thank you. It is Ravi Tanna, Goldman Sachs. I’ve three questions please. The first one is on pricing and claims and you made a reference to you alluded to the 2016 outlook and prices tracking a little bit ahead of claims inflation, I was just wondering on an average premium basis it looks like premiums are up about 3% presumably because of de-risking of the book and I am just curious, how far or longer are you in that kind of broader de-risking journey and be what suggested for that, what is the pricing claims, kind of dynamic look alike in terms of 2016 on that basis. Second question was on reserving buffer and again the comment around it being the top end of the range. In the last six months year it has obviously changed to prudency reserving policy on kind of new business on account of the fact that the reserve by force at the top end of the range and was wondering whether that we should assume that be implications of how you book, book policies going forward given the buffer still at the top end of the range and the third question just on the Solvency II ratio, when the Solvency II dividend but when 150 to 200 million guidance was originally said, I am just wondering what are the kind of underlying interest rates assumptions, that was predicated on please.

I’ll just make one comment and then hand it over to the Alistair and Geraint. I don’t think we have changed our initial reserve prudence on new claims that are emerging, I didn’t think. That’s a misunderstanding or something we met at. The reserving prudence start when we put on our initial claims and what the levels we book our initial years, very much in line with previous experience. Alistair?

So in terms of pricing, yet average being gone up by that 3%. And I alluded to the fact that our price increase is similar to ABI and so the difference between the two is a change in mix. It is the change in mix isn’t a long term plan we write business according to what is most profitable as we always have done. In terms of claims inflation, the other side of that, I think if you look at the industry data the ABI data on frequency looks fairly benign, the MOJ portal information suggests that claims registered in the first half of 2016 is only about 5% from the first the half of 2015. So frequency looks recently positive and we are fairly consistent with that. In terms of sedulity there is some inflation that I think others have commented on the repair side, which we also seeing some of us as well. But the big thing is bodily injury, a churn about the market I think we probably seeing is very early days but for 2015 and first half of 2016, we have seen some positive trends in terms of what we are seeing on large which hopefully links back to that slight shift in mix we have seen, but it is very early days so we will have to wait and see how that comes forth.

The final part that was yield curve, so six months ago we said we expected to pay 150 to 200 million on special dividends, that was based on what we saw at that time, it was based on business plans and our projections and the yield curve and the data that we got effectively at the end of January.

No that’s all right, can I just make the follow up on the first one around, there well maybe misunderstanding but I understood that you were booking your loss ratios closer to ultimate going forward than you have historically on the basis that the reserve buffer was at the top end of the range but I may have misunderstood.

Do you want to comment about that?

One way, I have certainly we move our range, our position in the range down from top end to the middle end. Couple of ways you can do that, you can also reduce that on the conservatives on the back book or you can book a little ratio on the current year and trust me, we think about the margin as a whole across the whole years, particularly important, which part it comes from. Naturally we want to hold a bigger buff on the most and underdeveloped which tends to be the most recent years, which is quite likely the initial margin[indiscernible] pretty consistent in the most recent years.

Okay thank you.

Hi, it is David Bracewell here, Redburn, few questions. One, you reiterated the range for Solvency of 125 to 150 when you gave you internal model approve. I’m just wondering given the sensitivities that you have published today were quite large or some of the scenarios, I am just wondering how you could, if you could confirm why you are happy with that kind of low buffer if not that sensitivity will change in your internal model basis. Second question is on your opinion operations, growth is fantastic but I was expecting to see larger growth from Spain rather than Italian operations and equally on the price comparison I was hoping to see the Spanish price comparison website doing the better that given the changes of going on in that market. I wonder if you can comment there about your fort [ph] and what is happening in Spain. And the third question just reading on cross selling ability on the home insurance which is curious, you have mentioned the few times now the new system cross sell will be a lot easier going forward and you also mentioned the price comparison for whole insurance significantly, I just wondering if you are going to give us an idea of where you expect policy growth to come from will it be from cross-sell or will it be from the growth of the price comparison and equally just on the new products you mentioned, I’m just wondering what your ability really is to kind of cross sell a car finance loan to your customers and what your expectations are there for kind of marked in developing the product.

I’ll do Spain and I’ll do new products and then hand it over to Cristina for household. Spanish operation has been held back over the last few years because of situations back in the premiums falling and frequency falling and the number of cars on falling [ph]. We are looking forward to a time when that changes, it has partially changed in a sense that, economic activities picking up, car sales are picking up but the actual premium changes, the relatively modest so far, maybe 1% or 2% and that hasn’t yet fade through substantial increase in short. So although we have grown the business somewhat, it is not growing very rapidly in the stage simply because the price comparison and shopping isn’t growing very rapidly at this stage. Terms and new products and is not an easy to cross sell household, it is not easy to cross sell loans to motor. There is an interesting interaction that we have with customers around change of vehicle which means we are very aware when customer is considering or making a change and we also have awful lot of information about the car they apparently have, how long they had it for and things like that. One of the big things we will be testing is part of the loans operation in 2017, is how we leverage back in knowledge in to a cost effective cross selling operation. Household shifting?

Yes, similar to what you were telling, the new system allows us to cross sell much easier. First because we cannot - product we can change but also because it makes the customer journey faster. So instead of having plus or the questions we can start to give ambiguity place after just asking five questions. And if you imagine that you have been on the phone for a while talking about motor insurance it makes it very easy we can say five more questions and we will give you this. So with now almost all the new sales on guidewire that allow us to start doing a lot of test in cross selling but cross selling is despite the difficult technique to achieve a lot have valid and then break it. So for us it is beginning of a journey, we are going to start to do NP for ways, we start to offer different promotions, but I wouldn’t want to have a promise that is going to be significant growth in the next few months, it will just be a journey.

Final one was the guidance on the solvency coverage, I think what we are saying is that we haven’t framed operations but once we get the internal model approved in a year and half’s time or whatever that will come. I think your challenge is fair with the start of the year with a 1.25% solvency ratio and then also 1.2% moving yield curve balance that would be in a rent comfortable [ph]. We won’t comment on how things might look into the positive model, it is too early to do that and we will continue to look at ways to jump it of it, sensitivity ratio, commencing.

We will go to the right and we will come back to the front and over to left back.

Hi, there it is Olivia Brindle from Bank of America, two questions. One again on the target solvency range, just wondering when you originally gave that guidance what was your thinking around this whole volatility adjust ratio because obviously you just said the best starting point would then be higher. So it is unchanged how do we think about the whole volatility just received within that. And secondly just in terms of congestion in UK, obviously expectations are for a recession, certainly a downturn in macro environment which we haven’t seen yet. Just wondering how you think about what market does in that scenario, you talked about maybe come backs just being a bit more rational, maybe responding to trend bit more quickly and how do you think pricing and competition looks if you do go into much worse environment.

The volatility adjustment is my first. The volatility adjustment applications [indiscernible] them. It is accelerated, clearly as a result of interest rate moves we’ve seen. And what that does, it does basically gives a more appropriate discount rates, purely risk free rates which is purely inappropriate. I view targets almost doing is really changed that much, the volatility improves the solvency position by about 15 percentage points and so the start point from where we go from here will hopefully be 16 points or so higher. That’s really changed the end game I don’t think.

Alastair, do you want to talk about the impact of the recession and?

I think we went through recession through this sort of 2000 latter after the crunch and beyond and the current rate cycle seem to have been fairly independent of those wider macroeconomics and I think there will be impacts until maybe people drive less against frequency but last time I think what we saw was those things were offset by other dynamics within car insurance and so I think we expect to going forward. But we certainly did see also is that even depth of recession people didn’t give up on the cars, they largely kept them and just drove them less and that is not the end of the world from a current shows point of view.

Good morning, it is Cameron Hussein from RBC. I’ve got three questions, first one just coming back to Alastair’s comments about conversations. So obviously the first half came down far away from the level we saw last year. I guess from the second half of this year and let think about the - I guess the lost ratio pick for next year [ph] looks about in line with where you are or where you are at 2014, so should we expect commutations next year to be equally like. The second question is on U.S. price comparison. Do you think you need to invest more in that market and I guess what’s stopping the 60% of the top 10 on your platform joining and the third question just on the loan - well, the potential loan book, just a quick question, is there going to be on your balance sheet or is that with the model be you find the loan taken for you and pass on someone else’s balance sheet. So that is my three. Thank you.

In terms of commutations there aren’t any commutations in the second half and next commutation is going to be 2015 which we are likely being in 2017. The 2015 underwriting year is booked at lower loss ratio in 2014 at the same point in time and for that reason we expect if that plays through as we anticipated we expected it to be lower the impact of 2015 commutation in the first half of 2017.

Answering in regard to [indiscernible] in U.S. I think it requires more investment because it is a huge market of cross marketing. Price comparison is still relatively new concept for insurance in this country. And I think the results we achieved so far support ongoing investments because this is dynamic and we will continue to do best in this operations have a huge potential given the size of the market.

I think it is very rational also for at this point of development of price comparison for some of the top players to hold off as they did in a number of markets. In a sense what we - hypothesis going into compare is that there is a large of number of U.S. players that find completely against [indiscernible] progressive in all state and state front increasingly challenge in U.S. market is concentrating towards those players to come in supportive of compare [ph] that’s the growing evidence or that brands that people migrate to, there’s big TV spends and so in that context it make sense for the other players to come in supportive of compare and that is the growing evidence, that’s is the way the world is going but which will have to be a much bigger player and much bigger part of market before some of the biggest brands decided to makes it even with the sense to go on board. In terms of lending, when it is very smaller, it is will be on our balance sheet, it is not worth the effect to try and go very complicated in terms of funding as it gets bigger in the same way that the for example insurance capital to help fund our insurance operations we certainly will look at number of different ways of building a business capital efficient.

Thanks it is Jame Shuck from UBS. I have three questions please. I just want to return to the Solvency II issue again and just looking at the sensitivity so slide you gave, the full year I think kind of indicated about six points of sensitivity based on January based on experience and I think January experience would it be the around 25 basis points. So in the first half we probably had closer to 100 basis points of reduction in the wake of Brexit, certainly looking at the 30 a year which would have been consistent with that about 25 point that you have indicated. The slide you gave at the back of the pack now is suggesting the 50 basis point reduction leads to 25 points of solvency to ratio which indicates, sensitivity about twice as much as previously indicated attributed, just that would be helpful and if I update the Solvency II ratio for now, ignoring the VA application for the time being but 180% would go to updating for what happens since half year I mean about 155 now, how would you view that in the context of your 125 to 150 which I appreciate is on a partial internal model basis but really want to comment on that partial internal model is going to end up versus the standard formula, are you able to give some guidance on, at least directionally which way it might go and second you have, you actually viewed that volatility in general. I mean is there just a quark of interest rates and risk margins and discussions with PRA etcetera or is it something you are actually going to take into account and to reduce that volatility and perhaps to bring up some of the insurance excess loss for example and two extra ones, just the accident, if I looked at the combined ratio X, the prior year development in the UK, actually kind of improved by about 60 basis points. So just giving the comment, this is a H1, just giving comments around pricing and claims inflation, I am just interested to know, why wouldn’t see the accident combined ratio improve more than that, I appreciate there is a number of written parts but if you understand that, why that didn’t get better. And then finally just on claims and experience at H1 just a little bit more you gave kind of indicative kind of directional things but you guys actually give a number about what your claim experience was and how you think it compare versus the market on large BI experience would be helpful, thank you.

The sensitivity point is the fair observation I think, since we brought at the end of year was December to January yield curve, it was uneven route across the yield curve so it was 20 basis points, some points lesser than the point, for which I mean here was flat 50 basis point across the curve particularly to the later point of it to the until sensitivity is as we move bigger probably slow post Brexit it is about 40 to 50 basis points move from the 20 something for June to the end of July and that’s consistent with this 25 basis point, upside 20 move in Solvency. We understand 50 basis points flat across curve and is consistent with what we saw post Brexit pretty much. The second part of partial model guidance and trajectory to 150, 125 to150, we are not giving guidance of what we expect the partial interim model to deliver. I think we should expect game changing insolvency kind of changes for the partial model approval and in the absence of other guidance we are not really hard to objection to be seen in the rest of year, the growth in the capital kind of have capital generation move or it has some dividends and I think you should take the 150s as guidance based on what we see today in terms of additional specials to get us there into that range in two or three years’ time or two years’ time though. So in sense of general, we will look at it obviously where we can damp ourselves between [indiscernible] in general at the end of the day where we are trying to have PPO gains in our books, where insolvency two moving acquiring very low, discount rates to the value of [indiscernible].

In terms of the accident loss ratio, I think you were asking about the movement from 2014 to 2015

H1 actually.

In H1, okay. So we haven’t disclosed our excellent year for H1. Our accident loss ratio, are you looking at the accounts? Okay. So H1 what we have written in the first half is very early days, so what we will be putting in there would be driven by very much by our conservatives and the first pick for because of uncertainty of large BI and I think that feeds through into your follow up question on claims experience, I think I talk about frequency on large BI, I said that we’ve seen improvements and it is hard for me to say what that is relative to the market, because we don’t really have insights to that and we are being conscious in terms of quantifying it because we won’t know what the large year experience would be for 15 and 16 for couple of years yet.

If I just follow up on that, if I look at the peers recent reported, some of those reported quite strong improvements in the accident year loss ratio, the calendar year, the accounting metric ex reserve releases at H1, you haven’t seen that coming through in your numbers, is that because I mean is the implication of the ultimate loss ratio big of H1 16 is actually flat to slightly up on 2015 number.

Can I comment on that by saying that the price increases we put through in the last 18 - 24 months, would you anticipate the loss ratio but we don’t take credit for those improvements until we got strong evidence of them and reliable evidence. In case our move to de-risk as Alastair mentioned previously, a lot of what we are expecting is an outcome is a lower big bodily injury claims, we have seen that on ‘15, in the first half of ‘16 but we are reluctant to take credit for it yet, because we have seen in the past the timing and size of the claim as a high degree of randomness. So if you want to take credit for them, make sure it is true and that tends to lead us to be very conservative on current years.

Thanks, Thomas Seidl from Bernstein, first on compare.com outcomes are you sure, this is linear improvement in votes [ph] my question here is number one, assuming this continuous, are you confident that you reach breakeven by end of 2017, second question if you don’t reach will you come close to breakeven 2017, are you continuous this assessment of some 30 million and third question this mix actually to lower risk business, does this allow you to earn the same underwriting margin or is it lower opportunity and lower risk business to run the same margins in the higher risk business.

I’ll do the first one and maybe Alas will do the last one. On compare, I think a key issue on comparison, you cannot say that evolution is going to be linear and therefore it's impossible for us to say exactly when the revenue and the expense line might cross. We are making good progress on both those lines, but they haven't crossed yet and we do anticipate continuing losses in the second half of 2016 and into 17. Do we give up if it's not making money again in 17, which I think was your question is I think that's too simplistic a question someone else asked about state experience. If we were seeing a number of states where we were seeing positive outcomes, particularly if we were seeing profits in a number of states that would certainly give us a lot of confidence to say ultimately this model can work and we should persevere and make the investment to try and realize the opportunity yet. Last year the U.S. car insurance market grew from $188 billion GWP to $198 billion GWP a $10 billion of growth in one year. So it is a very big prize if we can actually capture a material proportion of the shopping and we would only give up if we set forth the numbers were telling us that this operation does not have the potential to meet me to be very valuable to our shareholders.

And are you sort of - how do you do with this competitors, so do you see progress versus competitors taking market share. Well we're actually pioneering a different approach to price comparison in the U.S. And so although there are operations, which are similar, are trying to generate quotes in a similar space and sometimes even can run the risk of looking a bit like compared to the consumer, the actual delivery of a number of truly accurate immediately acceptable prices is only really currently being offered by compare in a price comparison model. There are a few internet brokers, look who do something similar, but it hasn't got the same breath of offer.

Can I go back and then we will move across the other side who probably feeling very neglected.

So in terms of mix the reason why where we're shifting that is because where we're driven by profitability and so we're looking at the margin where we're shifting that average because they're the attractiveness of it and I think it's worth saying that the reason why we've come down is because in 2011, 2012 we were very high premium business and that was the exception rather than the rule. I think David showed a graph at the full year and if you look back further than that, the differential between our average premium and the market now is similar to what it has been at most points in our history.

Yes, but no chance of that. Greig Paterson, KBW. One of clarification you made some comment about your average rate in the first off we knew that ABI could just explicitly state what was your base rate increase year-on-year to June and so there's no ambiguity, but what you're saying in terms of the rate increases. The second question, in terms of the 2014 underwriting year, which you have said has experienced elevated bargepole injury claims. If you listen to what Hastings said and they they've recently increased the loss ratios on the 16, they said it was not notifications I received in the first of this year. I was wondering if your increase in 14, over 13 BI was to do with increased [indiscernible] or increased actual notifications that you've had and then the third question is China, should we pencil in the same amount of marketing spend as we're seeing in the U.S. for the next few years on China - you can give us some kind of guidance was obviously significant losses in U.S. if we need to factor few in going forward for China.

Going backwards and I’ll start with China. China is a new price comparison site that we're looking to launch under our Preminen joint venture with Mapfre. Looking to launch it with a number of local Chinese partners, the dilution of ownership that results from the joint venture and the involvement of Chinese partners means that our share of the Chinese operation is in the 20% and it's a very early exploratory stage in terms of how much the appropriate investment is with initially as a sort of provincial rather than national launch. So we will use that opportunity of working alongside local partners and a fellow investor in the form of Mapfre to understand the Chinese market before we make any decisions, which would allow me to answer that question I think accurately. In terms of 2014 before I hand it back to Alastair for premium inflation. I just refer you back to previous presentation where we showed you the market data that said 2014 was just an unusual year with like of 10% higher level of big claims and a much bigger average value today because and so I think 2014 was just it was a surprisingly unusual market and very happy to refer back to that paper and that presentation of comment.

[indiscernible] you actually had the notifications or you speculating because of industry data that he notifications were higher.

I mean, we know about big claims relation to 2014 by now we don't necessarily know all their ultimate values, but we certainly know about them. Alastair?

I think I said base rate increase across new business and renewals were in line with the increases in the API and that was up 6%.

Just a kind of - so that’s average over the period 6% including new business and renewal business.

That's across, that's the increase over the last year-on-year. This is the 6% and that's across new business and renewal.

Is that as at June or the average over1H,’ 16, it makes a difference.

It is year-on-year.

So that’s in June.

As of June, yeah.

So 6% in June including new business renewal, business effects.

6% in June versus a year ago, yeah.

Apologies and sorry to cut you in the middle and to be unable to answer with a couple of questions. We’ll catch up later. Thank you very much for your attendance. See you next time.