David Stevens - CEO & Director Geraint Jones - CFO & Director Cristina Nestares - CEO, UK Insurance Alistair Hargreaves - Head, Service Scott Cargill - Head, Loans

John Irwin - UBS Wajahat Rizvi - Deutsche Bank Edward Morris - JPMorgan Chase & Co. Dominic O'Mahony - Exane BNP Paribas James Shuck - Citigroup Barrie Cornes - Panmure Gordon & Co. Thomas Seidl - Sanford C. Bernstein & Co. Andrew Crean - Autonomous Research Dhruv Gahlaut - HSBC Philip Ross - Berenberg

Good morning, everyone. I'm David Stevens, I'm the CEO. Welcome to our 2018 half year results, and thank you very much for coming in this August morning to hear what have to say. Brief introduction from myself, very happy to stand in front of you reporting at our positive set of results, characterized by growth pretty much across the board in the company, growth in customers, growth in turnover, growth in profits and growth in dividends. One particular milestone I'd like to flag up is the arrival at profitability of our EU Insurance companies, as we're calling them, somewhat preempting Brexit at Spain, Italy and France, collectively making a profit for the first time. Talking more about our numbers will be Geraint, the CFO; talking about the group numbers, Cristina, U.K. Insurance CEO; and Alistair, Head of Service for the U.K., talking about core U.K. business. And then I'll come back and talk about price comparison and then International Insurance. Thank you.

Thank you, David. Good morning, everyone. I've been trying to work out all week why Investor Relations have seen fit to include a picture of me with egg on my face at the results, but I haven't quite figured that out yet. I'm going to do my usual run through the results highlights. I'll talk about capital and the dividend, and I'll finish up with a brief update on Admiral Loans. To start off, this is our usual highlight slide. David has given a flavor already, so I'll pick up a few more out. The top line figures were at the top, continued really good growth across the group: Plus 14% on both customers and turnover measures in H1; £1.7 billion of turnover; 6.25 million customers; all parts of the group are growing, those are both record figures; similar reach of growth to those we saw in 2017; pretax profit was up to £212 million, plus 9%.; EPS was up similarly, both of those are record sales; another very positive return on equities, you'd probably agree, 54%, basically in line with last year, reflecting strong profitability, efficient balance sheet; and our interim dividend of 60p per share was largely up in line with earnings per share; and finally, our solvency ratio still is definitely qualifying as very strong, just under 200%. This slide looks at customer and turnover growth across the group. It looks a familiar picture from last time, all the arrows are green. Particularly impressive growth again in households, up around 40% period-on-period in customers and turnover. That means that we added nearly 120,000 household policies in the first six months. Our International Insurance businesses had a nearly 20% customers and turnover whilst improving the results nicely, as we'll see shortly. And a couple of nice individual milestones to report. Elephant hits 200,000 vehicles in the last day of the first half and Bolivia and France moved past 150,000 in the same month. Of course, not at all to forget UK Motor, which is cars and vans. They review very well in the first half, 14% in terms of turnover, 12% in customer numbers. We added over 200,000 cars in the first half to move past 4 million, and the total you see there includes over 200,000 insured vans. Our combined price comparison businesses grew a bit more modestly. Turnover was up 6% to nearly £80 million. Moving on now to look at profit. As you've seen already, we made £212 million pretax in the first half, £211.7 million to be precise, that's 9% up half year on half year. We'll go into detail on the various segments throughout the presentation, so I'll just make some high-level comments here. The main contributor clearly is still UK Insurance. Profit was up 9% to £247 million. Bad weather hit the household results, and there was a loss there of £2 million, and that was around £7 million lower than it would've been with normal weather. Motor result was strong, £250 million of profit in the first half, 11% up. There were a few offsetting factors as usual within that result, with higher other revenue as a result of the growth. It was an important driver. Just a reminder, UK Car Insurance profits are still on reserves at Ogden at minus 0.75%. It's particularly pleasing to see a big improvement in the International Insurance result, which moved from a loss of £10 million in the first half of '17 to basically breaking even in the first half of this year whilst growing the top line by nearly 20% period-on-period. That results made up of a further profit with ConTe in Italy and actually a combined profit, as David has already mentioned, from our European insurers for the first time. That was just under £3 million. That total figure was slightly offset by a loss, albeit improved period-on-period, around £3 million at Elephant in the U.S. So very pleasing international results. Price comparison, as you can see, was essentially flat with a good increase in profits with Confused.com in the U.K., offset by lower profits in Europe with Compare.com in the U.S. broadly flat period-on-period. Loans, I'll come on to very shortly, but the increased loss we saw there was entirely reflecting the significant growth of that business in the first half compared to the first half of last year. And Other, which is everything else, we've put the detail again at the back of the slide so you can see the full breakdown. But the main points, as you can see: firstly, 2017 benefiting from some realized gains on investments held in the parent company, not repeated in 2018; and then secondly, higher share scheme charges. And the increase there in the current period is mainly driven by changes to the vesting assumptions in the variable awards. Moving away from income statement, let's take a look at capital. This slide shows our solvency position at the half year in terms of the amount of capital, capital requirements and the resulting ratio. At the end of June, the half year point is 196%, standing from 205% at the end of '17 and 214% in the middle of last year. Main movements in the current periods, as you expect, economic profit for the periods, deducting the interim dividend and an increase in the capital requirement, which is due to growth. The fall in the ratio in the first six months is due to a combination of that small increase in the capital requirement and a very small reduction in own funds. The basis of the capital requirement is still the standard formula plus the capital add-on. And in terms of internal model, our team has made very substantial progress this year in preparing our application, and we're having good positive interactions with the PRA. We do expect, however, that those discussions will carry on a bit longer, the plan that we now expect to apply for model approval into 2019 rather than towards the end of this year. And in terms of what that means for the endpoint of model approval, in a good scenario, it doesn't really change. We'd be looking at the middle of 2019 for approval, or on a less good scenario, we might be looking at the end of 2019 for approval. When we report our full year results in early 2019, we'll be in a much better position to give clearer guidance on that and hopefully start to share some numbers with you as well. In terms of future guidance for solvency, another repeatable. I said six months ago, we think that 150% is a good indicator of the upper end of the target range once we get through model approval. And up until that point, we won't make further comments on surplus capital that we are currently holding. Moving on to the interim dividend. As you can see, we've declared 60p per share, 7% higher than 2017's interim and is 97% to the first half earnings. Not too much more to say on the dividend. It moves largely in line with profits, as we've guided and as you've seen before. The repeat of the formal policy on the right-hand side of the slides, and I'd reiterate it unless, or in so we say otherwise, the guidance for dividends is in the 90s in terms of the payout ratio, and we got the relevant dates on the slides as well. Now one of the pillars of the Admiral Group's strategy is diversification beyond insurance, and Admiral Loans is the most important part of that right now. And this slide gives a brief update on our progress in H1. As you can see, we've continued to grow quite significantly and over £200 million of loans were outstanding at the end of June. We are very satisfied with the way the business is going, including with the quality of the business we are writing. And we continue to be prudent in terms of risk profile. We finalized the next stage of the external funding for the business at the end of June, and now the majority of that balance is funded externally rather than from group resources. And the new funding will see us well into 2019, and plans are well underway for the future stages. The loss for the period, as you can see, was £6 million and the guidance for the full year slightly raised to £10 million to £12.5 million of loss, and that changes due to growth and the front-loading of costs in that business rather than anything adverse. I'd also say, at this point, we currently expect 2019's result will be better than 2018's, obviously subject to things developing as we currently expect them to. We plan to have Scott Cargill, who is the CEO of Admiral Financial Services. He's in the room today. He will present at the full year and give a more detailed look at Admiral Financial Services. That's it for me. I'd leave you a couple of highlights. Firstly, more growth, growth, growth, plenty of it. Secondly, a really pleasing improvement in the international results. Third, continued improvements in household and loans. The results for the first half were impacted by weather for household, strong growth for loans. And last but certainly not least, a very strong solvency position maintained after paying a record interim dividend. Over now to Cristina and Al to talk more about the U.K.

Good morning, everybody. Alistair and I will cover the results for the U.K. Insurance operation. I'm going to start talking about the results for motor and the pricing of the market in Admiral. And then Alistair will cover some aspects of our motor operation, including claims, expenses and customer services. And then I will finish talking about our household results. But first, these pictures. It shows innovate. It's an annual event that we hold with staff from all parts of the business, and they come together to work on innovative ideas. Some of the ideas coming out this year included applications - or more applications for robotics and also ideas in certain cover insurance. Now before we go into the details, this is a reminder of the growth of our motor and household business over the past few periods. In summary, we have seen very strong growth in household and good growth in motor. For the first half of the year, our motor book grew by 12%, and that includes an 8% growth in the number of cars on cover and also the growth of our back book, which is now around 210,000 customers. Household grew by 42% in this period and profit in motor grew by 11% in the first half. And as Geraint said, the household result was impacted by the weather. In terms of loss ratios, this slide shows the current base estimate loss ratios with the numbers in brackets showing the movement in the ratio in the first half of the year. Now the best estimates that we sold tend to be conservative, especially for the most recent years. So the changes that you see in the first half are normal, in line with what we normally saw, and it reflect positive back year development. The 2017 year currently at 74%, it looks like it's going to be a good year. However, we continued to experience some disruption in the pattern of settlement of large claims due to Ogden uncertainty, although we expect some changes to this in the next few months. And finally, we have included some additional information in the Appendix of the presentation, including best estimate ratios on an underwriting basis and also perfect sensitivities on a number of different loss ratio changes. Moving on to look at reserve releases. You see another large release of 26% of premium, above the long-term average of circa 15%. The large release is driven by positive prior year development as well as a slight reduction in the reserve margin. The margin remains prudent and significant, which is consistent with trends over the past few periods. As usual, we will expect to continue large reserve releases in the future if things develop as expected. In terms of commutation, we have fully committed all years up to and including 2016 in the first half of 2018, as we will normally put. There was no commutation in the first half of 2017 due to the Ogden situation. Therefore, there is around a £30 million adverse impact on the income statement, which impacts the comparison with various periods. Moving on to pricing in the market. It's clear that market prices have declined in the first half. We have put a graph on the left that shows two indexes. The ABI index, it's in green, and it reflects or it's based on business written. So it includes new business and renewals, and so it's a small decrease in Q2. The Confused index in blue, it focuses more on new business quotes. So it tends to solve much more radical changes and it overstates in both directions, but it saw at 11% decrease in Q2 year-on-year. On the graph on the right, you have Admiral's time top. The graph is indexed to 100 in June '17. If you focus your attention from January this year, you can see that we have been losing competitiveness, and that is because we have kept our prices flat. That is we have not decreased prices in the first half of this year. Now it's hard to predict the outlook of the market for the rest of the year, but we believe that the reductions that we're seeing in prices are unlikely to continue for much longer. So we think that we might not see premiums dropping as fast as in previous cycles, and this is due to the underlying claims inflation; the government reforms, which are likely to be pushing to the future, as Alistair will explain; and also more rational pledges in the market. In Admiral, we have always focused on the balance between growth and margin. And recently, during Q3, we have then a small price increase to reflect the underlying claims pressure. And now I'm going to pass it to Alistair to talk a lot more about claims.

Good morning. My name is Alistair Hargreaves. I've worked in Admiral for around 10 years in a number of different roles in the U.K. business. I'm currently Head of Service, responsible for marketing, claims and operations. So I'm going to start talking about claims. On the left-hand side, we see the market claims frequency over the last five years. There's a long-term trend of gradually declining frequency and some fluctuations due to winter weather. Most recently, there were some very positive reductions in 2017, which reversed in the first quarter of 2018, largely due to particularly bad winter weather. Our own claims frequency trends are very similar. On the right-hand side is an exhibit to give a sense of the relative size of claims costs that drive our ultimate loss ratios. It shows the three key components: damage claims, large bodily injury claims and small bodily injury claims. Large bodily injury claims are large in value, but very small in volume, and hence, result in some volatility. This year, we continue to reserve at a discount rate of minus 0.75. And this, combined with expected changes in the discount rate, make it harder to comment on large bodily injury trends. Though I can say that after seeing a relatively low level of new large bodily injury claims in 2017, we have seen an increase somewhat in the first half of 2018. This is prudently reflected in the current year loss ratio, and these large claims take time to develop and will be influenced by any changes to the discount rate. On older years, we continued to see positive developments on large bodily injury claims over time. On the next slide, we see drivers of the small bodily injury and damage costs. The graph on the left-hand side shows the 12-month rolling total of small bodily injury claim notifications for the market. There is a declining trend through 2017, and this has leveled off with lower levels of notifications maintained in 2018. On the right-hand graph, we see market severity for damaged claims. This shows continuing inflation at a rate of 6% due to increasing technology within cars. Our experience on both BI and damage is broadly in line with that of the market. So there are the underlying trends, but to understand the claims outlook, we also have to consider the change to Ogden discount rate and whiplash reforms. It's been signaled that the Ogden discount rate will move from minus 0.75% to between 0% and 1%, reducing the average severity of large bodily injury claims. The whiplash reforms include the increase in small claims track, and new tariff system is intended to reduce the costs of small bodily injury claims. The key point in this slide is that, although both of these measures are within the civil liability bill, the expected timing is different. Following royal assent, the discount rate review is currently expected to be completed by September 19. The whiplash changes are also dependent on a new portal to allow customers to notify whiplash claims direct, and these changes are currently expected by April 2020 at the earliest. In summary, overall, Admiral's claims trends continued to be similar to those of the market, relatively low frequency and small bodily injury frequency with pressure on damage severity and an increase in new large bodily injury claims. The most significant development in 2019 is likely to be the increase in the Ogden discount rate. So moving on from claims costs, let's take a look at the expense ratio. In the first half of 2018, our expense ratio increased, largely as a result of increased acquisition costs due to faster growth. The graph also shows the comparison to the market, which is more difficult because of a change in data available, but we maintain a sizable advantage. On the right hand, we show that there are two drivers of our expense ratio advantage: our higher average premium and our lower cost per policy. Interestingly, the mathematical advantage from higher average premium only equates to 4 percentage points of our expense ratio advantage. We continue to have higher risk, higher premium policies, but our growth in recent years combined with others taking more high-risk business has somewhat reduced this difference. However, the most important component of our expense ratio advantage is the actual cost per policy. A key component is acquisition cost, which is over 1/3 of total expenses, excluding levies. I'll talk a little bit more about this. Our acquisition approach of disciplined marketing spend and embracing price comparison keeps costs down. But a key driver of acquisition costs is the number of customers who are newly acquired, or to flip it, the number of customers who stay with us. Persistency is a measure that reflects the proportion of customers who stay with us for a full term and then also stay to start a new policy at renewal. The graph shows we have better persistency than the market. However, we'd not expect this to be the case necessarily. A greater share of our customers come to us from price comparison, and we have a higher proportion of higher risk, higher premium customers. Both of these factors should result in lower persistency. So why do our customers stay? Well, it's always important to remain competitive on price, but another key factor is our service. On the left-hand side, we see indicators of customer satisfaction. Our customers stay because they receive great service. Underlying this is our culture, which Cristina will touch on in a moment, but the key point here is our people, including agents who serve the customers directly and others who work on products, projects or technology, really care about the service that we deliver to our customers. On the right-hand side is an example showing calls received per active customer. The types of calls we receive is something that we focus on. Ease of service is important. Can we make changes to save the customer having to call? This exhibit shows the reduction in calls of 15%. This reduction means we have less calls to answer, and it's also saving the customer effort, which is good for the customer and hence, persistency. The improvements are due to a number of initiatives, but include improvements to allow customers to make more changes to their policy online. Online service is an area of focus for us. We also offer live chat, online claims notifications, and we're trialing a total loss portal. That's all for me. Now I'll pass back to Cristina.

So let's take a look at the household results. A key theme of the household business since we launched five years ago has been the strong growth, and we have continued with this trend in the first half of the year. The factors, the drivers behind have remained the same. The growth of price comparison channel, which we believe will continue in the future, but also the growth of our direct and cross-sell activities. We use multi-cover to sell car and homes to our existing customers. In terms of the results, we have mentioned a few times that they have been impacted by two weather events, the freeze that affected much of the country in March and also the flash flooding that impacted mainly the East Midlands in May. As you can see on the graph on the left, these two weather events added 25 points to our loss ratio, and they accounted for £7.5 million. As a result, we have posted a loss of around £2 million. However, if we exclude the impact of these weather events, we will have somewhat a profit of £5.6 million. One key advantage that we have been able to replicate from our motor book to our household operation is our excellent expense ratio. As you can see on the graph, our expense ratio continues to improve and its [indiscernible] advantage versus the market, which is around 45%. This is particularly relevant when considering our strong rate of growth as expenses in insurance tend to be front-loaded. This advantage is based on a more efficient operation and on our acquisition strategy. For the remainder of the year, we expect to continue growing and to return to profitability, of course, weather permitting. Now before finishing, I would like to highlight one aspect of our business that we're particularly proud of, which is our culture. We believe in Admiral that people who like what they do, do it better. And also that behind happy staff, you have a lot of happy customers. So we came this year number three in the Best Workplaces survey, but also we are the only company that has been part of the Sunday Times 25 Best Big Companies to Work For since the inception of the survey back in 2001. Also this year, for the first time, they have published a study of best workplaces for women, and we were very pleased to come number three. But also, and I guess sometimes more important, I think our inclusive culture is also reflected on our pay structure, and we have the lowest pay gap among major U.K. insurance. So in summary, a good growth both in motor and especially in household. We have grown our car book by 8%, despite not putting prices down in the first half. We have improved our result for motor, and the household result has been impacted by the weather. And finally, we expect a more stable UK Motor pricing outlook, mainly due to growing claims inflation pressure. And now over to David to talk about our international results.

Thank you, Cristina. So before going into price comparison International Insurance, we have a picture here of the inaugural European Insurance MasterChef competition where employees from Spain, France and Italy gathered to compete, and our CEOs in the front row volunteered to judge. Big success from the Welsh management team is very keen to get involved next year. We're confident we can teach the Spanish, Italians and French a thing or two about gourmet cuisine. Price comparison first. So confused, a good six months, 6% growth in turnover, 29% growth in profits up to just under £6 million. Small increase in market share, but most of that increase in profitability is around operational efficiency, particularly reallocation of digital spend towards the more efficient digital media. The European insurer price comparisons, Rastreator and LeLynx, had a tough year, slow growth and some shrinkage in margin. A few factors involved there, but the major one was the launch in France of a third television advertising price comparison site. So now you have ourselves, LeLynx, LesFurets from competitive market and Meyer, Esurance as TV advertisers. Great news for insurers that sell by price comparison and ultimately potentially great news for the sector as a whole as we collectively educate the French motorists on the most efficient way of buying insurance, but something of a headwind in terms of the efficiency of marketing for LeLynx in the first half. Going over to the States and compare. The biggest single influence on Compare's results in the first half was a very strong change in sentiment in the U.S. auto insurance market. This shows the U.S. auto insurance combined ratio, and 108%, although it might not sound too frightening for those of you familiar with the UK Insurance market was something of a shock to the U.S. auto insurance market. And the focus in '16 and '17 was very much on repairing margin, increasing prices. As you can see, the Combine is coming down on earned ratio, and a number of major insurers are now writing business profitably and are keen to grow the top line as well as repair margins. That's led to those insurers, particularly competing more aggressively for the attention of the 25% to 30% of U.S. motorists who shop every year, and that's caused some inflation in the marketing arena, an arena in which Compare competes. As a result of that, the operational improvements we made in the first half have been offset by a more challenging marketing environment, and the loss has stayed roughly flat year-on-year. Conversely though, because more insurers are interested in growing their business, more insurers are interested in participating in Compare. And we've seen 2 or 3 quite substantial insurers coming to us during the first half of 2018 and asking to get involved. Such that by the end of the year, we are expecting to be in a situation where 15 of the top 25 insurers in the U.S. are represented on Compare in some way or another. Now I have to caveat that because it might be that they're only in one state, it might be they're only with one of their brands, but 15 will be there by the end of the year, and that's a very positive progress and good for the long-term prospects of the business. Having cracked the top 4, I mentioned last time that Encompass, the third biggest brand from all states, had got involved last year, and we've made further progress with insurance, their direct brand coming on board in a test mode just early in the second half of 2018. So that's progress, but still a way to go in terms of cracking the Top 4. U.S. Insurance, Elephant, a very strong first half, a return to substantial growth, 10% in six months, up to 200,000 customers. And that's the persistency strategy beginning to pay off in terms of customer growth as well as the attractiveness of the underlying business. The business is still loss-making. You see on the right-hand side that losses on a whole account basis, essentially turnover, were at 11%, down from 16% a year ago. Roughly 1.5% of that is a more benign hail season in Texas than the previous year, but still a substantial progress and it feeds through into a $4 million loss for us on the business as a whole. And I'm very proud to see that in our biggest states, we're now outperforming the market as a whole on loss ratio. EU Insurance has come up a couple of times already, but very happy to see the business reporting a profit of €3 million, a continuation of a trend. And over that period, we've also been increasing the reserve strength across Spain and France as building on the conservatism that we've already seen historically in the U.K. and Italy. So very encouraged by that, and particularly encouraged by the fact that we haven't had to buy profitability by stopping growing. 18% growth alongside the emergence of profitability. I won't go into more detail because there is a European Investor Day on September 17. Just to set expectations, we're not planning a dramatic announcement in Rome on that day. We've had a number of investors over the last few years who have asked for more color, more detail on our international businesses, and we'll give you that, and we'll give you that about the markets in which they operate as well. Also a chance to meet some of the management team there, important part of the Admiral talent pool, and hopefully, give you - do attend, the opportunity to get a feel for the valuation relevance of the EU business over the medium to long term. So in summary, increased profit for Confused, a return to growth, and lower losses in Elephant and European Insurance profitable. Now I do like to wrap up with a longer-term perspective. Investor Relations have put this picture in first which is me thanking staff in Wales for 10 years of service. I notice my colleague is staring glassy eyed into the distance, which may be a subliminal suggestion from Investor Relations that I don't go on too long. So I'll keep it short, and focus first on the core U.K. business and the long-term trends. One of the long-term trends is the growing importance of focused, listed U.K. personal lines players in the market or perhaps, to be truly up-to-date, focus listed in price and private equity owned U.K. personal lines players, who now represents - or at the end of '17, represented 43% of the market. It's higher now. And by the end of the year, we'd expect it to be well over 45%, approaching 50%. And that's a function of obviously, companies floating. But it's also a function of those companies having above average growth, noticeably ourselves, Hastings and esure. So what? Well, I think the so what is that when you have the market increasingly dominated by players that have that personal lines' dominance in the U.K. and the discipline of the market, you get a more rational, collectively more competent market, and that is good news for everyone. It leads to increased stability in our view, good news for shareholders who don't like the variability that the market historically has delivered, good news for customers who don't like the roller-coaster pricing of extreme cyclicality that has been a feature of the U.K. market over the last few decades. So I think a better context for us going forward and for our peers. This is the business that accounts for the majority of the value now. Going forward, I'm convinced that our diversifications will deliver incremental value to our shareholders. Taking the same sort of time perspective and looking at our business where the bubble size is defined by the number of policyholders or, more recently, loan recipients. I can't do price comparison on that model, so it's in the numbers at the bottom, but it's not in the bubble sizes. What you see in 2010 is an irrelevant diversification in terms of size, but not irrelevant in terms of losses at minus - at £30 million. That's losses beyond UK Car and Confused. 2014, it's still looking pretty irrelevant in terms of size, but it's looking even more relevant in terms of losses at minus £30 million- plus. And now 2018, I think it's a visual way of saying that these diversifications are relevant, and importantly, are also less expensive. So we're spending less on this in 2018 than we were in 2010. So I think a healthier environment for the core business and good prospects for the diversifying businesses. Thank you for your attention. I'm happy to take questions. Q - Unidentified Analyst: I'll hit you with three questions. One is I wonder if you can just talk about the ABI initiative around discounting of new business? Because traditionally, you have the biggest gap between renewal and new business pricing. Second, just a point of clarity. I'm not sure if I heard right. When you were looking at the second half of this year, in terms of motor, did you say the rate of increase would accelerate or deaccelerate? I just didn't quite hear that. And then in terms of the reserve strengthening in Spain and France, was that - was there some strengthening that went on in the first half of this year, or are you talking over the last few years? I'm just trying to understand where there's a headwind...

I'll do that last one first. We're always taking a conservative approach to reserving. That's been most relevant in the U.K. and Italian markets because they're exposed to bigger bodily injury claims. We have extended that into France and Spain as they've become - particularly as France has become more material, I was mainly making the point to say that we haven't delivered those improved results by accelerated release of reserves and pulling forward good news, if anything, the opposite.

There wasn't a change in your prudence?

There's been a change over time in favor of prudence over a number of years. Cristina, are you happy to take the first two points?

Yes. Second question is decelerate. So basically, we have seen in Q2 strong decreases in prices, and we think it's unlikely that these will continue for much longer given the claims inflation. In terms of the AVI initiative, we're fully supportive. We think it's a market feature that there are changes between new business and renewals. And there are things that we can now collectively do to make it easier, the journey for our customers, especially focusing on building more customers. I also believe that due to price comparison, motor insurance is one of the most transparent industries in the market when it comes to prices.

Because your discount is usually 5%, 10%? Is it still 5%, 10%, the discounting?

We don't give that range.

Other questions? The gentleman over there. Thank you.

John Irwin from UBS. Two quick ones from me. So on the motor pricing outlook, and I know that you're highlighting that it's still uncertain and that you're expecting stable pricing from here. I wondered why stable if we are seeing signs of high claims inflation across the market, Admiral and peers? Admiral is now increasing prices a little bit in Q3. And Q2, there was a sharp deterioration as you mentioned. It feels like perhaps pricing should be rising a little bit at the market level. And secondly, on the Admiral loan portfolio, when would you expect this to return to profitability - well, to make a profit?

Do you want to do the pricing, and maybe Scott's in the audience - might talk briefly about the prospects for loans. Cristina?

Yes. Two aspects that we have to consider. The first one is 2017 is probably going to be the best year in the cycle. So we start with a very healthy margin, and that's why we might not see increases in prices yet.

Scott?

Let me start by saying that the ambition for this business is that we are looking to create something that's meaningful and sustainable and certainly over time become a very significant contributor to the group profit. We - as Geraint said before, we are playing in the markets that are personal loans and car finance, and that's our focus. Over the last few years, we've recruited our team which is experienced in these markets. We're seeing distribution moving in our favor, particularly for a direct in price comparison. Distribution lines are familiar for Admiral Group generally. And over the last two years, we have been building and testing our underwriting capabilities, and we're happy with the early progress there. Over time, I'd also flag that we would expect and hope that when we get to a certain scale, we'll be able to create an expense advantage, and something that we are starting to strongly believe in. So a positive start, and I look forward to seeing more at the full year.

Thank you, Scott.

Waj Rizvi, Deutsche Bank. A couple of questions from me, please. One, on the UK Car Insurance top line growth. So part of that, I guess, would be the reinsurance renewal and then your competitiveness being restored, post the opt-in. But is there anything else, any other initiatives or anything which are driving that growth because the growth is pretty sharp, if I compare it to last year? And the second one would be on price comparison business. So you've talked about how diversification is a pillar of your business model. There, your strategy seems to be in stark contrast to your peers who are looking to diversify in other verticals of energy and others - stuff where you're focusing on the car market. So just wondering if you can comment on that, in terms of diversification.

Okay. So I'll do price comparison after - maybe, Alistair, do you want to do growth of the UK Car base and what's driving it?

Shall I do that first?

Yes. Okay.

So I think in terms of that 12% customer growth, Cristina showed our Times Top which is - was higher than last year, so that's partly our competitiveness in the first half. But the other thing that's maybe a bit different is the van book is included in those numbers. So as Cristina mentioned, we've now got just over 200,000 vans on cover.

Thank you. Price comparison, I think you have to look at the world Confused and the others primarily. So in the Confused context where you have four established players in the mature markets, what we've chosen to do there is - as a source of differentiation, is to focus particularly on an offering to drivers, which is car insurance, car finance and other car-related products, where I think it's important to differentiate in a world where you have maturity and established players. Internationally, we're investing in diversification. And one of the bullets, in terms of the lower margin in the European Insurance entities are - price comparison entities, was around investing in diversification, and energy would be a case in point in France, for example, where last month we launched an energy vertical. In Spain, we've launched recently a mortgage vertical and a car buying and selling vertical. So we are broadening the product offer in those markets where price comparison is still very much an emerging phenomenon and where we're typically in leadership positions. Okay. And then we'll go one forward.

Edward Morris, JPMorgan. First question is just on the impact of weather. You told us how much it cost you in the home business, but I wonder if you could indicate whether there was any impact in motor? Was your profitability affected at all in the first half from weather? And the second question is on the new financing for the loans business, the warehouse arrangement that you have. Can you just explain how the economics of that work? Is there a certain limit that it runs up to? And ultimately, is it something that you would look to securitize? Or what is the - just explain that to me, please.

On the loan part of the question, I think what would be valuable for our investors is a deeper dive into loans in general, and that's what Scott will deliver at the full year results. So - and one of the important elements of that will be understanding the funding we put it in place and the options going forward. So forgive me if I defer the answer for six months on that one. The first bit of the question was around household?

It was the effect of weather on household.

Sorry, on household, yes?

So on Car I think the impact of weather is best seen in the chart on H1 frequency, despite that we see on frequency in Q1 2018, that is showing a weather impact.

And going forward one? Thank you. Dominic O'Mahony: Dom O'Mahony, Exane BNP Paribas. So two questions from me. The first is on the impact of commutations. So was a negative impact clearly in this period. I'm just thinking about how to think about that going forward. Clearly, '17 was better than '16, but '17 also seems to have been quite a good year across the market. If - so - two-pronged answer I guess to this. If '17 improves by a point or two, are you expecting to see a negative impact on the commutation of that year? And then, more broadly, going forwards, should we get used to the idea of specific small negatives when you commute which then unwind into positive developments in future years? And then a second question is just on PYD for UK Car. 26 points is well above your 15% long-term figure. Could you just help us understand what that would have been without the slight relaxation in prudence? So I think this time six months ago, you told us it would have been more in line with that long-term figure. Does that still hold?

Commutation is - the reason we commute to reinsurance when we have the option is when it makes economic sense to do so. So reinsurance is cheapest to us if we commute it as early as we can. If we think there's enough headroom in the profitability of an underwriting year, then we'll choose to do that as soon as possible. What tends to happen is when we book an underwriting year at a loss-making point too early on it's life, it means that we recognize an accounting loss at that point. I think it's probably fair to assume that we will carry on doing that as long as our underwriting years continue to look profitable. And they will have this sort of £30-ish million negative impact usually in H1 which then reverses over time as an underwriting year will release. So I think that probably is a trend. On - do you want me to do reserve releases? And on reserve releases, I think 26%, you're right, is quite a bit above the long-term average. We reduced the size of the margin, modestly I would say in the first half, and so the release would have been closer to the long-term average, but still above. And I wouldn't go any further on than that, I think.

I think there's a question coming in on the phone. I don't quite know how to access it, but...

Any questions on the phone?

Is there any questions on the phone?

The first question from the telephone is from the line of James Shuck of Citi.

James Shuck from Citi. So I had three questions, please. Firstly, just on the change to the Munich Re agreement. That car insurance has been in place for kind of 20 years-plus as far as I can remember. I'm just wondering why now has been the time to change the allocation on the installment income so that you actually keep that, because that's obviously driven quite a high proportion of the increase in the other revenue per vehicle. A second question, around claims inflation. Just returning to the large bodily inury side, so you're saying there's a kind of uptick in claims inflation from large BI. I've been hearing that the propensity to PPO had actually fallen very sharply. And I would have thought that would have been an offsetting impact. So could you just explain a little bit about what's driving that negative development on large BI? Whether that's a frequency thing or a severity side of things? And then thirdly, just around your systems and capabilities. I guess we haven't really have any update from where you are, particularly in UK Motor I'm referring to, or UK Insurance, I should say, in terms of your kind of capabilities. So are you moving to kind of more flexible products? Do you have things in the pipeline that can allow a much more tailored approach towards customers? We know that there's a number of things in the pipeline from many of your competitors, so I'm just keen to know how far advanced you are on that side of things, please.

Geraint, do you want to take - thank you, James. Do you want to take Munich Re, Geraint; and Cristina, systems; and Alistair, PPO?

The Munich Re car insurance contract has, you're quite right, been in place for quite a long time since the early part of the 2000s. We've made a couple of reasonably significant changes to the arrangement over those years, which generally moved more of the profit to Admiral and away from Munich Re, which is our objective, because they're not in the room. And most recent change we've made is to keep all of the installment income where previously they kept their 40%. The - on the face of it, that sounds like a very dramatic change, but actually it means that we will earn less profit commission back from Munich Re because they'll earn less profit. And so the net overall impact of that is actually quite small. It's a positive to Admiral, clearly. But does affect which line in the income statement that change flows through, i.e. more installment income, less profit commission. But the overall change is not that material.

Alistair, do you want to do large BI and PPO?

Yes. In terms of large BI, I was talking about the trends for new large BI, so essentially, frequency rather than severity. In terms of the PPO, I think we haven't seen large numbers of PPOs anyway. They tend to be used in specific circumstances, but logically, a lower discount rate makes the lump-sum option more favorable.

Systems?

In terms of systems, we completed our migration to the Guidewire system, and it went well, so we don't talk more about it because it's finalizing. As we said at the time, it allow us to launch new products, but also to improve overall operations. And just to give you a few examples. After the launch or the migration to Guideware, we have been able to launch MultiCover, which allow us to sell several products under the same policy and to price them together. We have also done a move to online, stronger than in the past. And Alistair was talking about the reductions in calls, and that's partly because we have a better system that allow us to do these changes. So overall, pleased with our system, but it's an area that we continue investing heavily and are always keen to improve.

Back to the room? The back row and then one in the front. Yes. That one, and then...

Barrie Cornes at Panmure Gordon. I've got a couple of questions, if I may. First of all, just wondered with the relatively good weather, whether or not you anticipate any substenance claims on the horizon on the household book? And secondly, with the announcements from the last couple of days, just wondered what your views were on M&A in the sector and whether you're very happy to continue organic growth?

Substenance?

Yes. We're, mindful of the risk, given the very unusual weather, I mean we had a wetter spring, followed by a dry and hot summer, so it's a - we're going to keep a close eye with it. We have seen smaller spike in July, but this is too early. Normally, we will tend to see this in September and October, so it's a bit early to comment, but we're keeping an eye.

And obviously, interesting to see esure out, led as you can tell to a hurried change in one of our exhibits. I think it perhaps supports the view that the sector as a whole is looking more and more interesting to sophisticated financial investors. In terms of our own growth plans, we've always grown organically. We don't dismiss the possibility that we could buy something at some point, particularly perhaps to accelerate the growth of some of our secondary - common secondary products, but we haven't come across anything as yet that we feel is an appropriate purchase. One forward and then we'll go back to the very back.

Thomas Seidl, Bernstein. First, on Compare.com, you used to have Andrew Rose here so I may - I have to ask you, David. So can you explain how the competitiveness is increasing there and how it influences your outlook for Compare.com, given it seems to stagnate right now? Secondly, in home, you strip out the 7 million events, but clearly there must be a normal weather ratio for this business. So what would be a normal level of weather losses given your much more diversified book right now? And then on large bodily injury, can you just give us by how much in the current year reported losses large bodily injury claim have gone up, in percentage?

So household weather, do you want to take that one, Cristina?

Yes. So the 7.5 million that I was mentioning, it only this - it's only the impact of these two weather events, so it doesn't include normal weather. And we normally don't give result splits by the different perils.

On BI, it's ridiculously early to try and quantify something that is very volatile in terms of numbers and is very unpredictable in terms of ultimate cost. So it would be misleading to try and put a specific number on the pattern we've seen in the first half of '18 versus the first half of '17. What was the other one?

Compare.

Compare, yes, sorry. Okay, well, I think what you have is you have a number of ways of driving a business in the U.S. and particularly the digital ways like Google and buying leads, it's just simply a supply and demand, and so the cost of being number one on Google or buying digital leads has increased in the U.S. market. So that's the short-term negative. Long term, the whole thesis behind Compare is actually the Top 4 eating up this market, and they've just gone over 50% market share. And the other - you might say, the other 600 because there's a large number of insurers, but actually, the other 30 - the next 30 are collectively losing share, and that's because they can't compete in the marketing context. Logically, Compare is a huge opportunity for them in the sense to pull marketing dollars and function in a way that lets them actually grow their businesses efficiently. That will happen if they give us their prices across a number of states and if they pay us a revenue per sale that is a true reflection of the alternative cost of driving business. So their appetite for doing that, I think, has materially increased by the term in the cycle and by the level of marketing aggression of some of the biggest players.

One follow-up.

One follow up.

Yes, yes, just on this - isn't there an issue that in the United States, you have with GEICO insurer which is likely to undercut any price Compare.com can offer, so as a customer, if I look for the cheapest price, you don't have GEICO on the platform, so why would I go on the platform?

Well, I think that exaggerates GEICO's overall competitiveness. I think off the top of my head, their market share is low teens. Yes, they would be a great addition to the platform. But when you take 20 insurers and put them up against GEICO, you beat GEICO a material percentage at the time. If you have 1 or 2 insurers against GEICO, you're much less likely to. So panel breadth is important.

It's Andrew Crean at Autonomous. Three questions, please. Can I clarify, on this £30 million shortfall in first half '18 profits versus first half '17 on the commutation, re you saying that, that will be made up in the second half? I'm not sure I entirely understand the situation there. Secondly, the slide you had on the expense ratio advantage, I think 8 points was per policy. How is that impacted if you throw the share scheme costs in? And why shouldn't you throw the share scheme costs in? So they seem to be a fairly permanent part of the P&L taking up 10% of the profits. And then thirdly, on PCW, could you say whether you get any insights in your underwriting from the PCW businesses that you have because they don't look as though they're substantial or will be substantial contributors to the P&L and diversification strategy and why wouldn't one sell them?

Do you want do commutation?

Commutation. So hopefully didn't promise that £30 million would reverse fully in the second half. And what it does mean is that we've have committed all years, up to and including '16, and so the reserve releases that will come from those years that have recently been commuted will be bigger than they would have been otherwise. And so you'll see a release in H2 from '16 almost inevitably that will now be bigger because we've done that commutation. Whether all that £30 million comes back in the second half will depend on how much we bring the '16 loss ratio down. The timing impact was the fact that we didn't do a commutation in the first half of '17, so there is a - if you compare like-for-like, there's a £30 million adverse in first half of '18. What we did do in '17 was the commutations were actually in the second half of the year. So the second half of '17 was impacted adversely. So when we come to look at full year versus full year, you probably see a commutation in '17 or a commutation in '18, and it will be more like-for-like. If that makes sense.

On the profit share, profit shares run at around 1.5% to 2% of turnover. We're not sure how some of those payments are accounted for equivalent payments in other insurers and whether they are included in the expense ratio. Clearly, there's transparency, and if people want to take that cost and spread it across pro rata, they've got the information to do that. In terms of price comparison, price comparison has been an important adjunct to our core insurance business over the years, helping us grow the opportunity in the U.K. initially and subsequently abroad, and also helping us have a deeper insight into the way the consumer was behaving. And we valued that over and above the P&L impact. Having said that, if it isn't as core to the business as insurance and some of the diversifications into other insurance lines and lending, and if people are - we - as Henry used to say, everything's for sale apart from his wife. And I suppose I ought to add my wife to that. Dhruv?

Dhruv Guhlaut at HSBC. Three questions. First, on the investment portfolio, as in, should we expect any changes from here in the next 12 months? Secondly, you've grown stronger in the motor segment. Has the risk changed materially? And how should we think of that for the second half? And lastly, in terms of Compare.com, losses have been broadly flat, but could you say as in how the revenues have done? As in have you seen - how the ticks have done, revenues have done, could you just throw some light there?

Investment assets?

We've got the breakdown in the back of the pack. It looks basically the same as it did six months ago. I think we always look for options to try and not radically alter the risk mix, but increase the returns, and we continue to do that. I wouldn't expect radical change.

Business mix in the motor book, no significant change.

And in Compare, marginally up on the top line because we're not chasing the volume buying loss-making businesses. We're saving our powder a bit until we've got what we expect to be an improved panel later in the year or early next year, depending on how quick IT takes. Yes, sorry. That didn't help, did it? What time do we need to stop, Marisja? We need to stop now. Okay. One question.

Phil Ross from Berenberg. On U.K. household insurance, you mentioned the expense ratio had come down around 5 percentage points, partly due to efficiencies. I wonder if there are any specifics on that, or whether it's driven partly by growth, partly by distribution as well?

No, it's a continuation of improvements. The bigger we get, the easier it is to be more efficient.

And so there was one more. I promised one, and then we're going to stop, I'm sorry about that.

[Indiscernible] from Mediobanca. Just one point of clarification. You made it clear that you're not going to look at excess capital return until you get the results from the PRA back. So that means even if we have a change in the Ogden rating, you have some visibility that for potentially at H1 2019, there still will not be any excess capital return until you get visibility from the PRA.

In the hypothetical situation you get an Ogden rate to change, which results in more capital, then it's likely we wouldn't hold on to that capital. But I wouldn't say that we would move radically below where we currently are until we get into model clarity. Does that make sense?

Yes, absolutely.

Lovely. Thank you for coming, and see you in six months.