Sabre Insurance Group plc (SBRE) Earnings Call Transcript & Summary

Hi. Good morning, everyone. Thank you for joining us. What I know is a very hectic morning for some of you. It's almost exactly a year since we found ourselves rather unexpectedly during our results presentation by Zoom. Not really sure I've moved very far from this chair since we did the last results presentation. Certainly, the first one, I wore a normal shirt in the last year. We're very much live and in person, as you can tell. There'll be a fine for whoever speaks while on mute. As we go through these slides, we are very conscious that COVID is not just a claims frequency issue. There's been some real human impacts, some of which have impacted us here at Sabre as well. So I think it's important to keep our presentation this morning within that context. We're going to move through this fairly quickly. Charlie, plenty of time for questions at the end. Both similar format to normal, and we have at various points given forgoing to our internal insurance [ readiness ]. A few more of us speaking. The normal is we don't have to run up and down to a podium. Myself and Adam will do most of the presentation, Trevor will talk about claims inflation, and James will talk about how we've gone through pricing. As we go through this presentation, if there's any technical thoughts that you can't hear, we got here so the screen's up. If you could then please put that in the chat. When we get to questions at the end, if you can use the sort of hands-up facility, Hanro, who's facilitating, will open up your screen, will open up your camera and your mic to take questions. So today's agenda, the highlights, financial results. We will, as usual, give you a fairly full view on market context as we see it, a little bit on our investment case and then a summary and outlook for next year. So financial highlights. Continued absolute focus on profitability. The key thing for us has been to maintain very strong foundations during quite a complicated and turbulent year. COVID-19 has clearly impacted our premium. On the flip side, we've experienced reductions in claims -- in traffic and claims frequency. We look to pass the benefit on to customers by pricing appropriately as we've seen reduced traffic ahead of us, but we have had residual benefits, which gives us a loss ratio benefit in this year. Reserve position, we'll talk about, reflects quite a lot of moving parts, actually, in the year, which we'll talk about later in this presentation. We're very pleased to pay a full year dividend. Slightly complicated way of describing it this year. If you include the deferred special from last year, it's GBP 21.2 million. If you exclude that, 16p. SCR coverage, 203% post-dividend -- pre-dividend, 155% post-dividend, very comfortably within our range. You can see on the right-hand side, good net loss ratio. Expense ratio, slightly up, mainly because premium's down. And the combined ratio, very much in line with our long-term target. The operational highlights are really we supported customers and staff throughout this period. We've maintained pricing discipline as we've gone through this year. Our prices are up by in excess of 10% in a market where premium has probably been flat or slightly reduced. So we've maintained a strong foundation for future growth, which we'll come back to. We've covered long-run claims inflation. We've continued to optimize volume within our 70% to 80% combined operating range, and we launched about 5 or 6 new rating factors during the year to ensure our pricing sophistication remains at the forefront. We are continuing to expand our presence in there and testing new ways of rating and some footprint expansion now. The Saga opportunity has been inevitably delayed as we weren't able to get on site to test the new product, but we anticipate that going live in Q2. And as I said, we've maintained very strong foundations for growth. The other activities, we've had a minor operational restructure. That's really to allow us to accommodate the growth we see coming without increase in our headcount, which will benefit our expense ratio. We're in the process of test launching a new cloud-based rating and administration system that sit alongside our current setup. This allows us to test new products very quickly and very cheaply, and we'll talk more about that later. We think this gives us an ability to test just moving slightly sideways from our current footprint. We've maintained all employees on full pay, and we continue to pay rises and bonuses as we go into this year. All employees that work from home are virtually no additional cost. Very detailed plans in place to respond to the forthcoming regulatory changes, which we'll describe in much more detail later. And I'll say, I'm going to reiterate, a price increase in excess of 10% to ensure we are entering this year very fully funded. Finally, a roadmap carbon neutral is being developed, and we'll publish more on that as we go through this year. COVID has clearly impacted in a couple of ways. Premium, we've looked to pass back or to appropriately allow for the premium discounts as we've entered this lockdown period, and James will describe in more detail how we've gone through that. Premium is significantly down due to a lack of car sales and young drivers entering the market. There's normally about 1.6 million driving tests a year in the U.K. with about a 50% pass rate. So a very substantial number of young drivers haven't entered the market this year or last year and into this year. And clearly, car sales, especially secondhand car sales, have been very limited. So that could have an impact on the availability of premium in the market. How quickly volumes bounce back will really depend on how quickly and how well lockdown loosens. We think April 12 is quite a key date when, hopefully, car driving tests can recommence and car showrooms open up again. The flip side, claims factors, short-term benefits through reduced traffic, I think, is well understood. Perhaps less well spoken about is that theft is not down by the same proportion. Larger PI claims settlement have been significantly slower to settle. There have been delays of courts and delays in obtaining medical evidence for obvious reasons, and we don't yet know how consumer behavior reacts as lockdown fully ends. Really, what does the new normal look like? Is it 100%? Is it 105%? Is it 85%? Is there some sort of curve as we go through the next few months? It's probably one of the unknown factors at the moment. So for those highlights, I will now hand to Adam, who will talk through the financial highlights.

Thanks, Geoff. Hi, everyone. Thanks for joining us remotely this morning. I'm going to run through the financial highlights for 2020. So I'll talk through our results focusing on premiums, claims, expenses and dividends. I'll try to highlight the impact of the unique market conditions during 2020 we've had on H. Our monoline model really allows us to present a straightforward set of results. So test the slide? Great. So the headline premium at GBP 173.2 million is down on 2019. It's largely due to our maintaining strict pricing discipline during challenging conditions in 2020, which is some of the item picked in the next few slides. Our combined operating ratio was in line with our expectation and the guidance we issued for the year at around 75%. Looking through that, we'll see that's a result of a very strong loss ratio and an uptick in our expense ratio. On investment return, we've removed some of the volatility here by classifying the assets in our buy-and-hold portfolio at fair value through other comprehensive income. So our investment return from 2020 onwards reflect the effective interest rate on assets hold, removing the temporary impact of market value movements. The return of GBP 1.4 million is, therefore, effectively the long-term yield across our portfolio. Profit before tax of GBP 49.1 million remains primarily a function of our premium and income and combined ratio. With a decrease in earned premium and a slight uptick in combined ratio follows that our profit is down in 2020, which consequently feeds through to earnings per share. We've made the decision to pay out effectively 100% of our earnings as a dividend for 2020. The 2020 dividend quoted on this slide firstly includes the 5.2p deferred special dividend in respect to 2019, which was paid at the interim stage. After the payment of the year-end dividend, our solvency coverage ratio remains very comfortable at 155%. So I'm going to dig into a little bit of what's been driving our top line during 2020. Underlying all of this remains our basic philosophy. We'll chase profit, not volume, making sure that we price our policies to cover all associated costs. Because of this, despite the temporary impacts of COVID, we continue to price in our long-term view of claims inflation. That's against the market backdrop of static or shrinking average premiums over the last few years. We still believe that a significant correction to market pricing is required. As you can see from this graph, which shows our price changes versus the market, a point during 2020, we did reduce our rates to reflect our assumptions for the temporary reductions in claims frequency during various lockdowns. James will give you a bit more detail on how we thought about pricing as the pandemic progressed. These discounts were always based on our best view of the cost attaching to policies. We were never going to chase the market down. We believe that these discounts on new and renewing business were fair and proportionate without undermining overall profitability. We suspect that some in the market were discounting far more heavily, particularly during the most severe periods of lockdown, which, of course, had an impact on our competitiveness on overall premium. But the relative levels of discounting weren't the only issue. This slide shows the relative levels of quotes for new policies received through price comparison websites versus the same point in 2019. At Sabre, we've always been skewed more heavily towards new than renewing business. So during those periods of low shop out during 2020, we think the less premium was available to us. All of which means our premium during the year look like this. You can see that we started to build a little momentum during Q1, pre-COVID, where rates appeared to be happening. Then during lockdown, we saw a significant dip in our premium income. This recovered strongly immediately post-lockdown but then shrunk when restrictions were reintroduced. So we expect our dip in top line income to be explained through, primarily, a loss of competitiveness relative to the market as competitors' policies were over discounted; and a significant reduction in the events, which would normally drive new business towards us, such as new drivers entering the market and car sales. It's perhaps a testament to our competitiveness within the more specialist areas in the market that despite this, we retained our premium at 88%, 2019 levels across the year. Now for the other significant impact of the pandemic on our business, which was in our claims experience. The headlines, our loss ratio is down year-on-year to 48.6%, while our expense ratio is up to 26.7%. I'll talk you through our loss ratio first. We can split it into the costs recorded in respect of accidents in 2020, which is our current year loss ratio; and the movements on our estimated cost of set being claims recorded in previous years, which is our prior year loss ratio. On the current year loss ratio, this came in at 51.2%, which is 11.6% below last year, indicating more favorable claim experience. It's likely that this improvement was primarily driven by the reduction in claims frequency due to fewer miles traveled by policyholders in 2020. While we did discount new and renewing policies in order to reflect these savings, the improved loss ratios came through largely to on business written pre-lockdown or where we had underestimated the benefits. Our reserve position continues to reflect our consistent long-term approach and allows for increased volatility on the latest claims. We have noted a marked slowdown on the settlement of personal injury claims, which is, of course, also reflected in our reserves. The prior year loss ratio benefit is relatively low compared to previous years at 2.6%. We've always said that the benefit from prior years will reduce as our model normalizes and exceptional reserve releases drop away. I previously said that this would lead to a normal prior year reserve benefit of around 4% to 5%, which would largely reflect the runoff of risk margins on open claims. The prior year benefit is lower than that, in part because we have modestly increased our provision to cover things like increasing cost of providing care, particularly in respect to potential periodic payment orders, but also in respect of other large claims. Overall, our reserving methodology has been completely consistent with our views on the ultimate cost of claims representing the best data available at the right time. On to our expense ratio, which has increased year-on-year to 26.7%. This next slide shows a simple bridge between the 2019 and 2020 position. You may recall that 2019 saw a one-off accrual release of a little over than GBP 3 million, which obviously didn't recur in 2020. Otherwise, we've faced some cost pressures from increasing industry levies, increases in audit costs and the cost attached to recruiting new members to our Board, although we've largely offset these costs by keeping a tight control over our core expense base. The main rate driver of our increase in expense ratio is the reduction in our income. While we have a significant element of variable costs, the dip in premium will always have an adverse expect in our expense ratio. We've maintained our staffing levels on full pay throughout the pandemic and rewarded our staff with bonuses during the year despite the decrease in volumes. We're forward-looking so we're careful to manage the risk of overwhelming our operations should growth surge back as it has done in the past. Moving to fully remote working has been broadly cost-neutral with a small cost attached in running additional service negated by lower cost of growing the business during the year. I should also add that we haven't taken advantage of any government assistance during lockdown. I've added this slide to help observe us when trying to compare expense ratios across the market. Our expense ratio is all in. As a relatively small monoline business, there's no way to hide our costs. All of our costs, including head office expenditure, share scheme expenses, marketing costs and commission, are wrapped up in our expense ratio. This next slide shows our allocation of invested assets at year-end. Our new asset management relationship with Goldman Sachs was timely given the economic turmoil that persisted throughout 2020. During the year, we stepped carefully into a more diversified but still very low-risk portfolio of assets. The highly rated corporate and government-backed bonds performed well during the year with no significant downgrades or losses. Our strategy centers on capital preservation in our investment portfolio while attempting to a reasonable yield on the low-risk assets. The way we record income on investments, that's a little different this year, as I mentioned earlier, because we generally hold all our assets to maturity. The amount recognized in profit is the effect of interest on those assets, which is reflected for the yield to maturity. Short-term market value fluctuations in the bonds are now taken below the line. Whereas in 2019, all fair value movements were taken through profit. The dividend. Our approach continues to be to return excess capital to shareholders, with that excess being defined with reference to our current circumstances while keeping the post-dividend period end capital in 140% to 160% range. This year, we've chosen to pay out about 100% of our earnings, which takes us a little away into our capital range but still leaves us plenty of headroom. Our full year dividend, excluding the deferred 5.2p in respect to 2019, is 16p per share. My last slide is one I've shown in the past, our capital position over time. As you can see, we remain strongly capitalized. As we generate excess capital very quickly, our capital position has tended to exceed our preferred range, mainly because of the time between calculating the capital available to fund the dividend and paying that dividend. We continue to aim to hold post-dividend period-end capital at 140% to 160% range. And with the post-dividend capital position of 155% at year-end, we're very comfortable to be inside it. Thanks. And now back to Geoff.

Thank you. Adam. I'd just grab back control of the screen. Okay. We are, as usual, going to give you our views on where the market's at. Please, first few slides, I will drive the screen and James would talk. So we're going down in 3. Next slide, please. Approach. James, do you want to pick up?

Sure. Morning, everybody. Okay. So we thought we'd use the -- this morning as we have an opportunity to kind of describe how we went about pricing during 2020 with respect to COVID-19. So this is COVID-19 specific as opposed to underlying kind of claims inflation. But I guess it's important to reiterate what Adam said, which was that we kind of -- we stuck to our knitting in terms of our pricing discipline. And when the lockdown struck in March 2020, what we've got here is a chart which shows our volumes. Premium volume, the yellow bars, and then we've taken the government traffic volumes as a starting point. And if I can have the next slide. We were looking to predict what would happen based on what we can [indiscernible] so far. So I think it's fair to say the lockdown started, and we entered the first couple of -- the first week in April. The first thing we didn't do was have a knee jerk reaction. We stayed firm, collected some data model. So take a view based on the information that was out there as to how we think we're going to price going forward because, clearly, we still have a lot of uncertainty around what was going to happen, and what we could see was the traffic volume's down. So I guess this first yellow -- the first yellow line shows what we thought would happen to relative traffic, and that was one of the inputs we used to get into our premium calculation, appreciating, again, as the guy said, the you've got things like fraud, you've got things like theft not reducing during lockdowns and then you've got this kind of knock-on effect on claim severity, increased severity costs potentially owing to COVID. So this is really just a starting point as to how we were thinking about how frequency might turn out as of April 20. We can have the next slide, please. So you can see this is the government data and how things start to evolve. So actually, we didn't get it too bad, actually. I mean we undercooked it a little bit, and you can see the volumes really came up. But on average, over the period, we were kind of in the right ballpark. And then I think as more information -- we unlocked to go over as some of more information became available. And at that point, I mean things were kind of going back to normal. I think if we go on another slide, thank you. We then kind of saw scores went back. Traffic grew as near as to normal as it had previously been. But then obviously, we had the announcement that we'd be going into a lockdown during November and December -- half way through November and December. And we could see -- we took an estimate at that point as to what would happen, and you can see our estimate wasn't too aligned with the government, actually what happened with hindsight. So you've kind of got November and December kind of up and down, with December up because you've got a Christmas effect, where we knew everyone was going to be allowed out for Christmas, so to speak, or so we thought at the time. It turns out they won't have allowed out for as long. And we took a view at that point that as at December, that we took the view that we didn't know we were going to go into a full lockdown for the first part of 2021, but we had a feeling that we weren't going to return to normal. So we took a view on where the traffic was going to be and where that would impact claims numbers, and therefore, reflected that in the premiums. So if we can have the next slide, please. Actually, what happened was then as we got the new information that we would be in a lockdown, we redid the pricing. So each of these kind of iterations is a summary point of different iterations of pricing. And as soon as we found out we were going back into a full lockdown from the beginning of January, effectively, we changed our assumptions around the pricing. And I guess this really reflects one of our most recent assumptions around pricing, which is we think we'll be back towards what I would call normal, whatever normal looks like around the June, July mark this year. But -- and so far, I guess it's showing that our modeling hasn't been 1 million miles out in terms of the shape with respect to the actual observed traffic volumes. I think what we're kind of assuming is the unwinding of lockdown measures we saw over last summer is going to be very similar in terms of behavior, in terms of traffic volumes. But like I said, we -- our expectation, based on the information we have from the government and everywhere else, is that we'll be back to normal by around about July in terms of traffic anyway, sorry.

Thank you, James.

Thank you.

So James has 2 or 3 master degrees and I don't. So he did those slides. Mine is a simple summary. We've applied very scientifically valid discounts in a cautious way for a limited period. That price is reflect to the claims spend over the policy life, not current profitability. Some proportion has benefited this year's loss ratio, and our view over discount on our optimistic assumptions and have very severe long-lasting but quite slow to emerge impacts, and we've been very keen to avoid that risk. I guess for those of you who've known us for a few years know we expect to shrunken growth in various market cycles. On here, you can see our year-end policy numbers in the GWP. I think our view is that COVID feels like it's extended the down part of the -- or the weak part of the market cycle by around the year. We see no reason why that market isn't going to turn. We saw very strong evidence at the start of this year of market rate increases tried to come through. So if we look back on with just over a year, we were putting through quite a lot of price discounts and our volume was holding very steady. So very clear indicator at the start of this year that market price increases coming through. Clearly, that all got stamped on when lockdown came in as market prices shrunk again. We've seen some fits and starts attempts to reincrease rates, but fundamentally, it doesn't seem to happen in the market. We think there is a risk that claims frequency benefits in this year's P&L line competitors to some of the underlying claims cost increases or our rates to be held lower for longer. To reiterate, traffic reduction is not directly proportionate to claims cost because of things like theft claims. So we firmly believe that for many competitors, rates need to adjust pretty sharply upwards in the near future. We have sort of actually dusted down our sort of scales of inflation versus deflation. For the eagle eye amongst all spot, this looks almost exactly the same as last year, really. Really, COVID has just pressed changes happening. We're going to talk through these in a little bit of detail. And for the first one, Trevor, you are going to talk through the claims impacts if that's okay.

Thanks, Geoff, and good morning, everybody. So overall, our view of inflation remains pretty consistent in that bucket of 7.5% to 8%. I guess what I would say is missing from this slide is the impact of theft in terms of severity. But as we've outlined, both from Geoff and from James, theft remains an issue in terms of frequency. But the main drivers are around first and third-party property damage inflation of 10% to 12%, and I'll unpack that in a couple of the slides that I'm going to go through. And we've also got personal injury severity inflation of between 5% and 6%. And I guess if you think that in terms of general damages, the bulk to the valuation of the judicial coverage guidelines came out of in 2019, and that had a 7% impact in terms of inflation there. So I think overall, these things tie together. We'd expect to see our competitors in a range around our numbers. Thanks, Geoff. Okay. So just sort of looking at what's happening around bent metal. We're seeing increased cost of parts. I'm just going to take you through some examples in the next couple of slides. We also believe that there are still to come potential post-Brexit impacts on there. Obviously, at the moment, we've got significantly reduced volume. So the full story isn't yet unfolding. There is greater use of credit repair models by our competitors. So over the last couple of years, we've probably seen something like a 25% increase in the proportion of claims that go through a credit repair model versus a normal insured subrogated model. And of course, costs under a repair or under a credit repair model are higher than the traditional costs, and we're seeing higher levels of inflation there. And increase on repair costs is driving an increase in total loss valuations. Now we look to combat these inflation factors by managing our own damaged clients in our network alongside our partners, Innovation Group. And of course, we're looking to control third-party costs where we can really to mitigate the industry-driven increases. Thanks, Geoff. So with Innovation Group who manage both our First Notification of Loss Facility and our repair network, we enjoy a material increase in terms of the buying power that we would have stood alone. So I'd like to thank them for some of the analysis that they've provided to support this presentation really by providing some detail in relation to repair costs across a basket of vehicles. So what we've done is we've taken the most popular car makes and the list is there, really just to look at how the cost of commonly damaged parts has been moving over time. So if we look at the first one, please, Geoff. So this is the average cost of a car part for those top 10 manufacturers. And you'll see between 2020 and 2019, we had an 8% increase significant increase across 2017 to 2018. So this is year-on-year increases that we've seen for a number of years now, and these are parts that need to be fitted to these common vehicles. So we've broken this down in the next slide between. If we take a right headlamp, so this is typically a component that will be damaged in front ax in a focal accident, so where a vehicle may have gone into the rear of another. This headlamp technology has moved on in the increased costs. Last year, we saw an increase in excess of 17% in terms of the average cost of that component. Then if we look at a full accident, so a rear bumper where clearly, we've gone into the back of somebody, again, inflation on that component of almost 10% across that average basket of vehicles. Thank you, Geoff. So we're facing into some other premium inflation factors. We've finally got the rules over 100 pages of rules and practice directions in relation to the Civil Liability Act and the changes to whiplash. So we now know that it will be implemented from the 31st of May this year on an accident current basis. So accidents on or after the 31st of May will be subject to the new rules and the new tariff. The original estimate was that there would be savings of circa GBP 30 per policy in relation to cost savings from the introduction of the new tariff reforms. Our house view is that currently, we believe that, that benefit will be materially less. That's partly due to the fact that there has been improved personal injury claims experience in terms of frequency across the last few years, and that's prior to the COVID impact. So we were seeing progression on reduction generally in terms of PI frequency. I think a word of caution, however, that it's going to be sometime before the impacts can be properly confirmed and assessed. These claims do take some time to settle. And therefore, it's going to be a year or 2 before we're getting a good idea in terms of how severity is looking and frequency is looking in relation to these whiplash reforms. So we're continuing to see significant ongoing claims inflation. As I said previously, our current view is between 7.5% and 8% driven by increased costs due to technology in the newer vehicles and more of a push in terms of credit repair models. If current dynamics continue, and we see no reason why they shouldn't, then we do anticipate that there will be a competitor margin squeeze. It may now be a significant market turn. Whilst we don't know exactly when as claims frequencies return to a normal position and claims inflation, which has always been there and through, we think that, that will be putting pressure on the market rates. Thanks, Geoff.

Thanks, Trevor. We've discussed in previous sessions that claims inflation is only really one part of the overall cost inflation. There are some other things going on. The first is the MIB levy. The MIB levy is certain to increase due to the Ogden rate. The recent Supreme Court judgment on the new suggests that we may be not picking up those liabilities to the MIB, but there's some work to go on how that works and indeed when that cuts in. The MIB is also picking up the operational cost of the new MOJ signatures. The FSCS levy delivery doesn't get spoken about so much, but it's quite an important factor. There's 100% increase in general insurance levies this year, and it's not a particularly small number. There have been several insurance company failures, and I suspect there may be other local insurer failures to come. Clearly, impacts different insurers sorts in different ways. I guess a high volume/low margin insurers may take more pain through this. For us, this is just part of our overall cost inflation assumptions. A big thing is the FCA pricing review, which I know has been spoken about in many other sessions. I think everybody is aware that this effectively bans price walking between new business and renewals. I'm not hearing much industry pushback on the fact these reforms should come through. I am hearing quite a lot of pushback on the timescale. I guess our view is they shouldn't be that long needed. What's been proposed to has been known for quite some time as a potential solution. We believe there is a real risk of some insurance operating. So think of hail mary strategy to just shut the ball forward a long way to build the portfolio as much as possible before the FCA reforms come into place. It's a last hoorah to try and build a book before the new business renewal prices come through. We suspect that may come in to the market. Some people might execute that well. Others may be forced into top in may not execute it. So well, we don't intend to indulge ourselves in that game. On the reinsurance cost, reinsurance cost increased at 31/12 in. As you know, our reinsurance renewal is midyear. Same to center at about 5% increase. Besides, price depended on your own claims experience. I think we would agree with some previous comments that the value of some of the lower levels is looking a bit shaky. So we will think about how renewal strategy as we go into this half year, but we won't make any fundamental changes to our growth. This, I think, is perhaps one of the key slides actually in the presentation. We've decided to steer our own path. I think it's fair to say as we've come through this COVID-impacted period. We've decided not to follow people down the rabbit hole of chasing reduced volume by overdiscounting premiums. We believe that leaves us in a very fully funded position as we sit here today. If we think about the tailwinds that we announced in over the next 12 to 18 months, we've reflected on here what we think will happen in March and how that may impact us. This is not necessarily in day order. We don't know where these tailwinds kick off, but some of the things that we think will have to happen here. Some competitors will already have whiplash MOJ discounts in their prices. We have not yet fully put that in place, so we have some amendments to wait from that. Many competitors may still have some pretty big COVID discounts to unwind. As James has described, we haven't really done that. We have a very minimal amount of discount to unwind from this position. As quote volumes increase, and we hope that starts coming from April 12, we should benefit from that increased volume coming through in the market from car sales and new drivers starting to take driving test again. We do believe there's a significant market pricing correction needed. We don't need to do that. We need to cover ongoing claims inflation in a disciplined way. We don't need to correct. And I think probably most people on this call know, we don't differentiate price between them and renewal business. So as the FCA pricing impacts come in, if that does involve market new business price increases, we don't need to do that. We can hold our prices. So generally, across this line, we would expect to see our competitive position improve as the next period goes by because we've been very disciplined in maintaining proper, effective, fully funded pricing through the last year or so. I'm going to spend just a few minutes talking about the investment case. I'm really conscious many of you have seen this before. This first slide continues to be our key principle: Maintain a wide underwriting footprint. At the moment, we're quoting for, I think, around 99% of risks in the U.K. So anyone coming to a cost comparison website can expect to see a price from Sabre or on a consistent margin approach. Market-leading underwriting footprint. So we're continuing to target in mid-70s combined. Strong cash returns and control growth across the market. Average quoted premium last year was over GBP 2,000, and average written premium was over GBP 1,000. So we maintain our slightly nonstandard positioning. But the largest premium last year, interest, and it was over GBP 30,000 for one risk, and we had several premiums over GBP 20,000. On the other side decline, our lowest premium was GBP 150, and we had a lot of those as well. So we're very happy to write premiums at a very high or relatively low at a rather consistent margin. I'd hope we managed to demonstrate our deals to 3 of these principles. The ones we're looking forward to demonstrating now is the controlled growth across the market, which we do believe is coming. Our pricing approach, we put this in last year, the bullseye for us in stable market conditions is 75% combined. I think we were quite clear last year in a soft market, we're right towards the right-hand side of this graph, and we are prepared to write much near 80% combined. We only really expect to be below 75%, where we've been able to give us some good news has come to, unexpected good news, although we've increased prices because growth is outstripping our ability to operationally handle that. Absolute focus on profitable business, but the key thing here is our combined ratio has been very consistently outperformed in the market for a very long time, and we don't intend to upset the apple cart on this because of a slightly complicated year that we've just been through. I think important on our profitability. It goes up and down in the range, but we avoid boom and bust. So if you look at our profit here at the bottom line, you can see its grooves tend to be smooth. It goes up and down a bit, but we're looking to maintain a regular flow of dividend and a sustainable profit going forward. Adam has mentioned our efficient use of capital. As we've described before, we really focus on underwriting profit. We have a prudent user insurance, a low-risk investment portfolio and consistent reserving where we see bad news coming through claims cost, but we'll reflect that very quickly reserving. We won't shy away from it or kick the can down the road on that. Reliable dividend flow, as Adam said, up and around 100% of earnings. We've been very clear that we're prepared to use our capital ratio to support the dividend in weaker parts of the cycle. Clearly, as we go through '21 and '22, we'll be earning through lower premiums, and we're absolutely prepared to look at using our capital range to support the dividend as we go through that period. These are our competitive advantages. High-quality underwriting is always going to be at the center of our business. We have a data lake full of the data that we have collected in a consistent way over the last 18 or 20 years. We use cutting-edge data enrichment, sophisticated pricing models and the scientific taking the human impact and subjectivity out of our pricing models. Cost advantage, as Adam has described, we outsourced where it's beneficial, but we are very happy to invest in things that make a difference. We invest a lot of money in price and enrichment. And we've maintained and intend to maintain our headcount in claims ahead of that anticipated growth, despite the fact we have excess capacity at the moment. Distribution, split between brokers and direct. We get the best of both worlds. We are channel-agnostic when it comes on it. We make effectively the same profit either way. This is a bit of a personal bug there for me, if I'm being honest. This is some stuff that I've carbon pasted from some recent sales pitch documents for Insurtech. So automated underwriting, state-of-the-art pricing, data lakes technology, automation, lightning fast speed, API links. It also has very good and very modern. I do wonder whether people think we set about with scrolls and quilts when we do our underwriting. The point here is we do all of these things as well, and I guess we had the additional element that we have got a proven ability over many years to make a profit as well. Growth. We really thought that 2020 will be the start of a growth phase for us. And the first month or 2 reconfirm that with the price increases we saw in the market. As we described, COVID has held back that phase. I guess everything we described as we are pretty confident growth is on the horizon. The timing is slightly uncertain, but it is coming. We see growth being driven by 2 main pillars and potentially supported by 2 others. And important within this, any growth initiatives have to fit within our dividend-focused strategy. We won't undermine our ability to pay dividends to attract growth. So this is our sort of pillars, and our focus pretty much works from left to right across these pillars. The key thing for us is to optimize and maximize what we already do in our existing product set. The second column is product expansion. We've described that we're looking or we are rolling out some brand changes this year. We have other thoughts of some product enhancements, some adjacent product development later this year. Going forward, we probably expect to research and test and potentially launch 2 new product adjacent initiatives a year. To the third column, if we identify opportunities that we don't have the skill set, we're prepared to think about recruiting people to help us with that, but the DNA of those people and their culture must be completely aligned to what we do. And last, and in this case, definitely least is M&A. We tend to look at M&A, but it must be good value, and it must be complementary and supportive of our growth strategy. So our growth strategy very much from left to right across these pillars. So to wrap up on the summary, very focused on our long-term strategy. We haven't amended it, and we don't intend to amend it. We believe growth is coming in the foreseeable future. We described the tailwind. Exactly when they start to play was a little uncertain, but we're pretty confident it will come this year. In the meantime, we're confident we can maintain our profitability. Net actual variation in the absolute level of profit as volumes flex, and our pricing discipline over the last 12 months has given us a very strong foundation to continue to pay dividends and take advantage of growth. So by some fluke, that was exactly 45 minutes, and we now have time for Q&A. If you can use the raise hand function, Hanro will open your mic and your camera if you're happy to be seen, and we're happy to take any questions. Hanro, to you.

All right, Geoff. So first question will be from Ming Zhu.

Just 3 questions for me, please. Since you've already increased your price more than 10%, how come you're still guiding the combined ratio in the sort of mid-70s target for this year? And I thought that is really a long-term target, and given where we are now with the lockdown, I would expect things to be better than that. And the second is I think previously, you commented that your customers are not that price sensitive. So given your top line and challenge last year, I just want to get a color in terms of the nonstandard risk target market. Has that part shrink a little bit? Is that what the challenge is? And also, my third question is, what is really a normal level of reserve release going forward?

Okay. I'll take the second. Adam, will you take COR? And maybe James, you can comment on the reserve releases. So I think on the price sensitivity, Ming, we said at IPO, we thought around 3/4 of our portfolio at that time was nonstandard. If you look at what's happened to premium, we pretty much have ended up at around that sort of level. So we don't believe we've really lost any competitiveness in the nonstandard area at all. As we expected, we think we have a competitive moat around the nonstandard stuff, and we tend to lose business in the more mass market area, and I think that's pretty much what's happened over the last year. Adam, do you want to talk about the COR?

Yes, certainly. So if you think back to the end of 2019, we came into 2020 rising towards the top end of our combined operating ratio range in a relatively soft market. And as Geoff said, that's where we'll replace our combined ratio target at that point. We have increased prices over the year with the goal that we can meet the inflationary costs that apply to that. We haven't necessarily stepped back down the combined ratio range. Overall, notwithstanding some of the benefits that come through from claims frequency from COVID, for example. So -- and of course, what we have done, to Geoff's point, during the year is make sure we've discounted in what we expect the claims frequency benefits might be. So on that basis, we wouldn't necessarily expect a significantly better combined ratio than one which you might have in a normal year, but nor do we expect our combined ratio to be at the top of the 70% to 80% range as it might have been where we not to be in the situation that we currently are.

Thank you, Adam. James, do you want say anything on reserve releases?

Yes. I mean, I think as Adam has been saying for quite some time, we have been expecting exceptional releases to come towards an end at some point, they can't carry on indefinitely. I think it's very -- I think if I had the exact answer of how big the relations are going to be going forward, I probably wouldn't have a job. But the -- I think it's fair to say that we said that the reserve releases would start to reduce or may have produced slightly more than expectation, but not outside of the realms of acceptability in terms of how volatile they can be.

The next question is from Nick Johnson.

Two questions, I think. Firstly, some of the peers are talking about increased price sophistication. Just wondering how confident you are the loss of competitiveness that you've seen in 2020 is sort of general pricing adjustments in the market or -- and not increased sophistication by some of the peers? That's the first question. The second question on capital ratio. I think you've said in the past that the capital requirement is forward-looking. So just wondering what level of growth is baked into the capital requirement that you've got at the moment. Are you assuming a return to pre-COVID premium income in your capital requirement number? Or are you assuming some additional growth as well?

If I take the first one. And then James, you can chip in and then Adam on the second. I think on price sophistication, we don't stand still. I think it's fair to say. I think I'll start to introduce [ Matt Wright ], who is our Deputy Chief Factory, who with James spends really the majority of their time looking at how we can stay ahead of the market on price and sophistication. We test a lot of data on a regular basis. We test new techniques. I'm very happy that sophistication is not what we're losing. I think it's pretty mindset, Nick, is the difference, where our mindset is profit first, volume second, and that's how we set ourselves up. So James, can you talk about that?

I'd kind of echo Geoff's comments. I think that -- I mean if we -- what you guys don't see in the background is us wearing away on these kind of R&D, trying to find new things. So our data scientist is forever trying to beat the current approach, and we are implementing little things on the side that you guys don't necessarily see and we don't necessarily share. Similarly, as we alluded to in the presentation, I mean we -- last year, we deployed our single biggest week scheme to change in terms of we think it's very exciting, and it's a big piece rating factor we've implemented for probably 4 or 5 years. So I think there's no question in my mind that we continue to develop and challenge ourselves and find new ways of differentiating risk, and that can go away from the traditional prorating. I think what we're seeing is, as we have seen in previous cycles, is quite a lot of irrationality in the market. And as we've tried to stay, we've -- as in my slides show, we try not too far ahead because otherwise, we'd just be chasing the market down. We're basing everything on the data and the information we have at the time and being influenced by things like what the government are telling us is going to happen rather than kind of -- if last April, we'd have taken a pun and reduced our rates significantly, yes, we'd had written more volume. But we wouldn't have known that. There was no way of knowing. It would have been a pun, and that's not really the same.

Yes. Thanks, James.

Yes. I think we've been fairly open that we don't know exactly when the market will turn, but we expect it at some point. I think the key thing here is we've still got a lot of capital headroom, and we will be very comfortable moving around within the range that we've currently got in order to fund that. And the capital strain generated through growth is there is some, but it's not as substantial as it might be of, say, our combined ratio was a lot higher because the business rising is profitable. It's sort of self-funding from that perspective. So we've left ourselves enough wiggle room that we won't necessarily need to constrain growth because of where we are in the capital range, but I suppose the development of our premium levels will depend a lot on what happens throughout the rest of this year.

Thanks, Adam. Thank you, Nick.

The next question is from Ivan Bokhmat.

Can you hear me now? Can you?

We can.

Perfect. Well, the question is -- it's a couple actually. The first one would be on the expense ratio and on the degree of fixed versus floating costs. Obviously, your quoting volumes have been much lower, but it didn't seem to have brought a lot of benefit to the expense ratio. And secondly, it's just on the capital efficiency. I've noticed that if we look very simply and divide your premiums by your -- or divide your SCR by your premium volumes, that capital intensity has increased quite a bit throughout 2020. Could you maybe give us a bit of an outlook on how your capital generation going forward should look like? I mean is it going to be further increase in capital intensity? Or are there any one-offs that you would want to flag?

Sure. Thanks, Ivan. Adam, I'd love to give you time. I don't think you're taking those 2 straight away.

Yes. No, no problem, Geoff. Thanks for the question. Expense ratio, yes, we still have a pretty high portion of variable costs. If you look at the slide I presented earlier, it separates out the commission element, which is almost entirely about report, I think that was around 8.6% of the total expense ratio. Beyond that, we do have staff costs, we do have property costs and other overheads of running the business. It's sort of back of the envelope calculation to get to how much of that remaining pod of costs is fully variable. I think it's fair to say where we're not in the situation we are where a big proportion of our cost is variable. Our expense ratio might have looked somewhat worse with the level of premium decline that we've seen over the past few years. So while we've seen the expense ratio tick up by a few percent, I think that has been relatively resilient given the fluctuations in volume and is entirely within the realms of what we would have expected. There have been some structural cost increases due to being a listed company, which in growth times, that means things have been slowed up by premium increases, but which happened in the current situation. On capital efficiency, it's a really good question. We have seen an increase in our capital requirement throughout 2020. Largely, that's been driven by the revisions to the investment portfolio. So bonds coming in presenting spread risk and a slight increase in the term risk on interest rates, all of which has been designed to increase our yield to an acceptable but still to a relatively low level, but at the same time, not generate any actual increase in our view of what the downside risk might be of holding those investments. So that is a genuine one-off, I suppose, that will move around a bit depending on whether we're in or out of certain bonds. But generally, we've taken the hit to our capital to putting that new investment portfolio in place. Other than that, we might have expected the capital requirement to shrink a little as we shrunk in size.

Thank you, Adam. Thank you, Ivan. Anything else, Ivan? Was that okay?

That's it.

Thank you much. There's a question that's come through on the chat facility from Charlie Beeching of KBW. One is about what do we expect to happen to claims frequency pre and post COVID? Could there be an impact from customers driving more than previously as lockdown eases? I think this is exactly the big question, what does the new normal look like? At the moment, for one or too many better data, whereas even the new normal all looks like the old normal, clearly, we'll be tracking that very closely as we see how lockdown loosens. There are theories that people might work from home 1 or 2 days a week. That could have a decrease on frequency. They may, therefore, be more driving at weekends. They could be driving at different times of day. What James and Matt will be doing is tracking that data very closely and adjusting our pricing as we see reasons to move. The second question here is the market impact. Do we expect any significant market consolidation as margins deteriorate? I think we know that some insurance and some distributors are having a tough time. We've seen some M&A already in the recent past. There may be some more to come. I think it's interesting if you look at the public companies like Admiral and Direct Line who've presented very strong, very good companies. There are other companies out there that are not so well capitalized and don't have the same sophistication. So I think it's wrong to judge the market just perhaps on Admiral and Direct Line than our slightly nonstandard approach to life. So there could be impacts that come through.

Our next question, Geoff, is from Ben Cohen.

Can you hear me.

Yes.

I just wanted to go back to Slide 12, please, on sort of the quotability that you have. I just wondered, to the extent that the sort of the further deterioration at the start of this year implies that the market is even more competitive than it was through last year, whether there's a sort of year-on-year effect. And I suppose just more generally then, as you start the year, how much are your premiums actually down? I think there are some later slides that imply that premiums are also down year-over-year. And is this simply capturing the lockdown impact?

Yes, sure. So I think we're very happy to talk about the start of this year. In the full RNS statement, we've spoken about the fact we've had to run hard to maintain about a 20% drop year-on-year. An important clarification is that's against a non-COVID period. So the COVID impact didn't really start to kick in until just around now last year. So we're comparing a COVID period this year with a non-COVID period last year. So we're not really in a normal comparison period there. I think there is a risk of a really intense bit of almost a blood bath has been kicked off by people chasing volume. So some people have tried to hold out as they've lost more and more volume. I think other people are now having to try and complete to get some of that volume back. So their own portfolio is entering too fast. We, in some ways, might benefit from that if some of the larger brokers need to maintain their own portfolio sizes. What I can see there being an intense period of competition through the very early part of this year as people start to play their price positions as COVID discounts online, and they need to try and maintain their portfolio.

I think I'd throw some of that, Geoff. I think it's a really good point, Ben, that if there are certain segments of the market where we're not seeing newcomers, for example, when new people driving tests are taking place, so we're not seeing new drivers to the market as just one example, that means that the part is smaller. And therefore, if even just a small part of the market is down, everyone chases the rest of it much harder, and that's something that from our disciplined pricing approach, we've not been prepared to do. So whilst in the short term, we stick to our netting, and the volume is lower. In the longer -- I mean it's not sustainable to continually chase volume at the cost of profit.

Sorry, can I ask as a follow-up? Geoff, you mentioned the risk that there would be competition to sort of position ahead of the FCA changes coming in. Could you explain that in the context of if companies sort of lock in unprofitable business, then is it not sort of unprofitable forever? So how realistic is that as a scenario, would you say?

Yes. I think the way I've tried to rationalize this in my own mind, if you believe that market prices, I'm making these numbers up slightly, are going to increase by 10% when the FCA pricing review comes in. If you can buy a policy this year and you have a lifetime value model and think you can hold it into future years, that may be cheaper than buy it back for us comparison website next year. I think also some competitors have a profit model that's very dependent on non-underwriting profit. So for example, claims income. So if you go into next year too small, you only -- they had to buy back the policies, you haven't got the claims income coming through at the same time. So I can understand the strategy. So something we execute it well, I think someone. For us, it's irrelevant because that's not how we earn our money, so we will stay disciplined as that potential turmoil goes on. And as I think I mentioned in the presentation last week, there will be some jock into position as we will work out where the appropriate new business and when all price points are.

We still have a couple of hands raised. Next is Tom Bateman.

A few questions from me. A couple on top line, I'm sorry to come back to this topic. But if I look back last year, policy count is down about 24%. How much can you catch up in 2021 as new drivers come back on, as new car sales pick up? What's -- how much of that can you realistically make up? And I guess following on from that, you talk about maybe you haven't lost too much in the nonstandard risk market. But who are your competitors here? Who are writing these risks when you aren't essentially? I'm going to Slide 1 here on that topic as well. Just on electric vehicles, obviously, new car sales are a part of your book, given the increase in penetration in electric vehicles. How prepared are you to write more of that business? And just one final question on the SCR and investment yield again. The SCR is increasing because the market risk essentially. How would you look at the payoff between the additional yield you get on the portfolio and the capital requirement there? And any numbers on return on capital will be really helpful.

Yes. First off, I think maybe once you go back to 2015, actually, which is the last sort of really hard part of the market and it's been to the year I joined. I am joined and we couldn't contemplate fast enough to maintain the volume, to keep the volumes down at a level where we thought we could cope with operation. So we said, you know what, the growth was in that year. I think it was about 20...

I mean we would have heard something like 20 up -- we would have put in the region of 20 points on. I recall in March 16, we put over 6 points on a single month. So that we felt that the growth was so much that we would rather take the volume. So that operationally, Trevor's guys could handle the claims volumes with the quality that they -- and attention that they need because we didn't want to compromise that kind of our ability to set some claims. And I think it's an important point that someone might be, let say, taking the business, but are they handling the claims in the same way? That would -- I would challenge that as well. So if that business may well come back to us as things turn it out.

I think -- so the way I think about it is we unquote -- well, I think I'll make this up very slightly, around GBP 0.5 million business a week. And we quote for all of them, and we'll take any of them that generate our required margin. So there's no shortage of opportunity for us to write business in the market at all when the price points hit the right level. I think your next question was on electric vehicles. We're very happy with our electric vehicles. We're right at the moment. The claims are more complicated. Parts can be more expensive, but we're very happy about electric vehicles. We do them at exactly the same margin as any other vehicle type that we are. I think you asked who's writing the risks. I think, look, it just goes out to a combination of other insurers. We don't really feel like we've lost a lot of the nonstandard business. Some of it's not been there. So I'm personally very grateful to Kent Police for making -- for just sending me a stern letter from my speeding mishap rather than making me speed awareness costs. So I think you haven't had convictions coming on driving licenses. You haven't had new drivers. There's been less house sales. All the things that drive you back into the market as a nonstandard risk just have happened. So I don't think we've really lost anything in the nonstandard area that I can see. And the last question, Adam, was around SCR and the sort of payoffs that we think.

Yes. Thanks, Geoff. I mean I'm going to give you a fairly loose answer now, but by all means, we can dig into a little off-line. In very broad strokes, we have taken a capital hit because of the investment portfolio we're now in. There will be a payback period on that capital. The payback period will be longer in the current circumstances than it might have been if bond yields overall were higher. Nonetheless, I guess, spread's a factor into that as well. I suppose because we've got a fairly open range of capital in which we can operate. We've effectively just moved down in the capital range at no loss to the dividend available to shareholders. If we were sort of staunch the 140% mark, then maybe we would be easing up on capital we would otherwise pay out as a dividend. So I think the question for us more as a team is whether at the revised level of capital after we've taken that hit, we're still comfortable that we've got enough, and we are. That -- we keep it under review. So if we're not getting good value, we think, for the capital we've used in moving into the investment portfolio that we've got, we'll roll it back at relatively little expense. So yes, I'm not going to go into more of the math than that at this stage. But by all means, I think we should sort of keep it as a rolling dialogue with the market.

Thanks, Tom.

Our next question is from Robert Murphy.

Just coming back to the one-off prior year reserving increase that you mentioned. Can you actually quantify what that was? I mean you're sort of implying a couple of points. And also, is that mainly related to 1 year, sort of like, I don't know, 2019 probably, looks like one of the worst developing years historically, I don't know, if that makes sense.

If I get the headline, James, you can put in the detail if that's okay. So really, there's a few moving parts here. The first one is the care home cost increase, which was a -- is published on a periodic basis. The last one came out in November, which is the ASHE index. I think it was talking about more like a 5.9% increase in care worker wages. That's impacted across all largest claims. PPI capitalization funding is sort of a thing that's particularly related to that in how we can -- how the reinsurance capitalize some of those payments. And I guess just, James, reflecting all that data through as it's come to life. Do you want to sort of further any more detail around it?

Yes. I think you've kind of covered it off, Geoff. I think we did see in the second half and then to with a comment on the relevance of COVID in that sense, but we certainly have seen an increase in the expected in the cost of care by the ASHE index, not only impacts all of the triangles on a net back basis because it impacts some of the larger losses. It also impacts, as Geoff said, the view on PPO and potential capitalization claims. So that's essentially, in short, I think Geoff summarized it quite nicely.

And just a quick follow-up on the bond portfolio as well. Is that just credit risk you've increased or also duration risk on that?

Adam?

I'll take that one, Geoff. There's a slight uptick in duration risk, although we're still pretty well unmatched against our insurance liabilities, but it's largely spread.

So second last question is from [ Faisan Lacani ]. The next question will be Alex Evans.

Can you hear me?

We can. Now we can.

So the first one is just on the rationality of the motor market that you're talking about. It seems to be the case that this year at the start of this year, at least that's continuing. I just wonder what's the catalyst for these insurers rationalizing some of the frequency benefits? Is it the road traffic going back to normal levels? Or do you think the lower quote volume and more aggressiveness of some insurers means that, that's going to be beyond that? And then just secondly as well, maybe just on the FCA review, it seems like quite a significant data request to the insurers, and I appreciate it won't affect you in the same way as others. But how long do you envisage it to take to implement the infrastructure and everything in place and looking at the time line there?

Sure. I mean in terms of irrationality in motor, I think I've worked -- I've been looking at motion insurance for 30 years, and it's been pretty much consistently bonkers all that time. I would say, at some point, there's normally someone doing something irrational. I guess at the moment, it's probably being driven by a need to maintain top line. So we do have a reasonably flexible and variable cost base. So we are -- our mindset is absolutely focused on profit. I think a lot of insurers have a trade-off to make between -- an understandable trade-off between top line and profit. We don't do that. We were only focused on the bottom line. So I think probably the drivers are more to do with that. And with the FCA being such a radical change, I do understand why some insurers might want to maintain portfolio size going into that area. On the FCA data request, I've seen some amazing quotes on how much it's going to cost to make this change. I'm pretty certain Matt wrote a data request in a couple of days. Last time, we have all the data at hand. So we see, Matt, I think, no real cost of effort in complying with that.

We think it should be really straightforward. So we can take it on.

Thank you, Matt.

Okay. So let's see if we can get [ Faisan ] back for question.

Can you hear me now?

We can loud and clear.

I had some mic issues. I just wanted to sort of kick the ties on the pricing thesis. I mean if I may take the opposite view for a second, it appears the markets have over earned last year with code frequency benefit. Some of your peers have posted lower reserve releases than last year, which may suggest that the market is quite cautious. Now once that unwinds, there may be a bit of gunpowder to kind of help any sort of weaker current year margins. And even though the whiplash forma is probably said to be slightly lower benefit, there are sort of Q3 sort of big things pushing against a harder or a firmer pricing market. I mean what's your sort of view on that? And secondly, in terms of claims inflation, your guidance has stayed pretty steady. Just want to compare it to one of your larger peers, Direct Line, where they seem to be operating at sort of 5% to 6% claims inflation. Is that mainly due to business mix? Or is there an element of different views in terms of the drivers of the market or your capabilities?

Yes, sure. I guess on the pricing thesis, it wouldn't be a market if everyone had the same view at all times. I suppose our view is, and we've tried to outline this in some of the slides, when you're writing a policy, you could look at your claims experience for the next 12 months, not the current month you're writing that policy in. And so I think you're absolutely right, P&Ls could look quite good for the first quarter. Our concern is what will the claims cost look like for that policy across the life. James, do you want to add anything to that?

Yes. I think it's a good question, a fair challenge. I think what we observed last year was if you look at -- back to the presentation slides, if you look at our volumes over time and you overlay that with the traffic volumes, it would appear that there is a significant correlation between our relative competitiveness and the traffic volumes. And as we said, that doesn't perfectly correlate to pricing. We've seen that last year, I think we went from being something like the month of April was 40% down year-on-year and the month of June was 10% up year-on-year, now we took a strategic view on what we thought the pricing should be and went with that view. The traffic changed reasonably significantly, and we saw that reasonably significant change in volume. That would imply that there is a lot of irrational -- top line-driven irrational pricing in the market, I would suggest, and the fact that we have not been, what's the word, lured into chasing that. I think it's a set as a differentiator. And make no mistake, as Geoff has said, that if we write a policy today, we have to take a view on what the claims, what the exposure and what, therefore, the claims experience is going to be like, not just for the month of March and April. Truly, it's going to be for the next 12 months, and you really have to be confident. If you've got a crystal ball with the future in it, then great. But if you haven't, you've got to take a sense to that approach. And I would argue that, that is feeding potential rational pricing, along with the fact that currently, the P&Ls are probably looking quite healthy. But I guess what I'm saying is once that business be right today, if you've mispriced it, when it starts to earn through, it really starts to buy it, and we haven't been subjected to that because of our philosophy and our approach.

Do you have a view on which layers have been overly aggressive or irrational? I mean you've seen an, for example, a lot of business, but they seem comfortable with it. I mean do you have a view in terms of where the pressure will come?

I don't think it's right to comment on competitors on this stuff. I think it's important, and we've mentioned a few times, it's important not to base a market if you want to add more Direct Line than us. I think that's probably why we're tiny. But I think at more , 1/3 of the market, that's a lot of the market who don't report numbers publicly. So I think you shouldn't assume when we say competitors, we're not talking about just Direct Line and Admiral to any stretch of the imagination. I think an important point to make here is that if we have been too cautious, so if our assumptions are right, others have to catch up with our view. If we've been overly cautious and we're wrong, and we're not having enough to say that we're right, that means we've overfunded against our expectation, and that gives us the ability to reduce prices to get growth. So our assumption, our scenario A is that we are right, prices need to catch up. Scenario B is we'll be overly prudent, which case we can reduce prices if we've over assumed...

And I think to add to that, Geoff, then the balancing item would be in the simplest terms. If our margin is 20% and we decide we want that margin to be 19%, maybe we get 5% a bit away, we need to write at least 5% more business. And we do see, at certain price points, we do see a greater level of elasticity, and that's really where we're balancing what we expect to happen going forward and the actual elasticity in the market and making sure that we are kind of -- we are optimizing the absolute profit for our shareholders.

Yes. I think your second question, [ Faisan ] was claims inflation. So we are talking around 7% to 8%. I think others have talked both in been somewhat over 5%. The main difference could be around theft. I think it's fair to say, Trevor. We are prepared to write cars that are more prone to theft. We think that could have a slight uptick on our claims inflation compared to others probably in the region of a point or so. Trevor, anything you want to add?

Yes, possibly. Of course, we look at inflation across a long period. We're not just looking at it on a 12-month basis. And I suspect that if you ask a bunch of claims directors or a bunch of actuaries, you'll come up with different views in terms of actually how they measure inflation. But it seems to us quite clear that this isn't a platform that you can keep the cork in forever.

It doesn't seem like we've got any further questions. So no more raised hands or open questions.

Okay. If anyone does have anything, we are slightly here and happy to take any other questions. If not, thank you very much for your time today. Very much appreciated. I hope we've unpacked the result and our views on the future. And I know some of you now have to run off to another presentation, so I hope we've given you time to do that. Thanks so much for your time.