Admiral Group plc (AMIGF) CEO Milena Mondini on Q4 2020 Results - Earnings Call Transcript

Admiral Group plc (OTCPK:AMIGF) Q4 2020 Earnings Conference Call March 4, 2021 5:30 AM ET

Company Participants

Milena Mondini - Group CEO

Cristina Nestares - CEO, UK Insurance

Costantino Moretti - Head, International Insurance

Scott Cargill - CEO, AFSL

Marisja Kocznur - Head, IR

Geraint Jones - CFO & Director

Conference Call Participants

Jonathan Denham - Morgan Stanley

Freya Kong - Bank of America

James Shuck - Citi

Ben Cohen - Investec

Thomas Bateman - Berenberg

Andreas van Embden - Peel Hunt

Alexander Evans - Crédit Suisse

Phil Ross - MedioBanca

Ivan Bokhmat - Barclays

Milena Mondini

Good morning, everybody, and welcome to this second, and hopefully, last year's 12 edition of our results presentation.

A positive set of results, as Geraint, our CFO, will detail later, with continuing growth and strong performance across our business, as Cristina, Costantino and Scott will explain, respectively, for U.K. Insurance, International Insurance and Loans. And these results were achieved while focusing on doing the right thing for all our shareholder. And I will come back to this at the end of the presentation.

So many things have changed this year. And to start with, the way in which we communicate, as you will see, in some of the cover slides of this presentation, we show pictures of some of the most typical Admiral event, pre- and post-COVID. For me, personally, there was also a major change in role. I am truly excited to work in the footstep of Henry and David. And I'm particularly excited had I meet everything, an incredibly strong team and a fantastic company and platform for continued future success.

Let me remind why I think this is the case by taking a look back over the last decade, before talking about 2020 and the decade that is to come.

So what are the drivers of Admiral's success?

First of all, there is a track record of delivering superior margin, with a combined ratio that is substantially better than our competitors. And this is a result of a strong focus on central insurance, claims and pricing, with an obsession for continuous improvement, combined with effective distribution, a cost-conscious culture and a lean operating model.

Second, these competitive advantages and the resulting growth in profit gave us the confidence to almost triple our customer base over the last 10 years, having 5 million of new customers, without even accounting for price comparison platform. Out of which, more than half were in new geography or handling new product. And the investment to develop this new venture was less than 5% of overall group profit. The secret sauce in this mix is our culture, possibly the most difficult advantage to replicateable. In every country and throughout the whole organization, we have great teams, focused on providing great service to our customer, but also on being great colleague, one for each other. We have been the only business that has been named in the Sunday Time and Great Place To Work list in U.K. since 2001. Why that's so important? Because people who like what they do, do it better.

Finally, an efficient capital model that we can leverage on very strong reinsurance agreement, that are highly possible because of the positive and long track record of underwriting results. This translates into higher return for our shareholder, with an average that is consistently over 50%, more than double than comparison.

Now let's take a look at what happened in 2020.

Last year, we delivered positive financial results and growth in U.K. and overseas, as Cristina and Constantino will explain in more depth later.

But what made me most proud are 2 things. First, despite all the operational challenges and the many changes caused by COVID disruption, we also managed to make substantial progress on our longer-term strategy, as I will explain soon. And second, in this difficult context, we remain true to our values, which helps us to navigate the unexpected and influence a lot of our choices. And we looked after all our stakeholders.

Starting with customers. First of all, we provide continuity of service by moving the majority of our staff to work remotely in just a few weeks. Second, we gave a refund of £25 per vehicle to 4.4 million customers in U.K. during the first lockdown in recognition of less driving. And this was followed by substantial discounts thereafter.

We supported more than 9,000 key workers by waiving, for example, the excess fees or providing free courtesy vehicles. And we allowed loan payment holidays for those who needed them. And the feedback for this has been overwhelming.

Staff safety was our top concern through the year. We provided flexible working arrangements and implemented many initiatives to support well-being at home of our staff, such as additional holiday stays. It has been incredible to see even stronger employment engagement scores despite the context. And we were named 14th Great Place To Work in the world.

But in true, we are aware that we work in a sector that has not been as negatively impacted by COVID as many others. And we tried to help the less fortunate through the Admiral COVID fund and many other choice initiatives, such as providing school equipment, coordinating tablet device to care homes. We also support our partners such as repair network in U.K. or outsourcer in Italy, without forgetting our commitment to the environment. And I'm pleased to report that we are now carbon neutral through offsetting, while also continuing to use 100% of renewable energy in our U.K. premises.

Last but not least, we delivered good outcomes for our shareholders, with profit exceeding £600 million, driven by prior year reserve releases and decrease in motor claims frequencies, and achieved almost 10% of customer growth, mainly enabled by stronger retention rate across all the businesses. And finally, I also believe that we are in a strong position to continue to deliver in the long term.

If we look ahead, we can envisage more demanding customer, the availability of more data and also material change in the way people will move around. In this context, our aim is to further diversify our product offering while delivering on customer expectations faster.

To assure, that we will focus on 3 areas. First, accelerate the evolution of all our core business towards what we call Admiral 2.0. And Admiral debt remains true to its historical strength, but it is even more agile, is digital first, embrace smart working. And above all, we aim at keeping improving customer experience, leveraging on data and advanced analytics even more than before. Second, we need to continue our product diversification journey to develop stronger proposition for our customers, but also to increase our engagement and enforce our relationship with them. How? First, by continuing to support successful or promising businesses, such as household or loans in the U.K., and then by continuing to explore new opportunities that are adjacent to our core, and while we think we can deploy our core strength. Finally, in the longer terms, change in mobility will come with risks and opportunities at this very early stage now, but we tend to observe, learn and prepare.

So why 2020 was such a strong year in that respect? It's fair to say that COVID played an important role here, putting initial pressure on our digital agenda, forcing us to move remote working almost overnight, and forcing customers to rethink their mobility and their living habits as well as also accelerating other pre-existing trends.

In 2020, we doubled the number of machine learnings models pushed to production, moved a large part of the business to scale agile, transition the majority of our customer data to the cloud and materially improve our NPS score in every country. And there is so much more to do.

On the product side, we'll not only improve household results substantially and strengthen our foundation loans, as Cristina and Scott will explain later, but we also planted more seed overseas with the launch of pet insurance in Italy and household insurance and the Olivier brand in France.

We also set up Admiral Pioneer, a team for the exploration of new opportunity in the U.K. To clarify, Pioneer is not an investment vehicle or a traditional incubator, our aim is to continue to do what we have always done in the past, pass new ideas, support the promising ones, and kill the others. But we will do so with a more intense approach, which will give us speed from one side and limit the distraction to the core, which remain our priority, on the other side.

And finally, just let me bring an example of the exploration within the new insurance model. Veygo is the U.K. brand for short-term flexible insurance. It was launched in 2017 and counting 150,000 customer last year. It's more with a turnover of 14 million, but it's growing, and it is an interesting vehicle for us for further test and learn of new mobility trends.

So to conclude, despite the challenging cases of a terrible pandemic, we had a good year, with good results. We made strategic progress. And we remain true to our value and our culture.

And now to Geraint to share our results more in detail.

Geraint Jones

Thanks, Milena. Hello, everyone.

I'll cover some of the main features of a positive set of results, looking at the top line, and of course, profit. I'll give a quick update on a few pieces of financial news. And I'll finish by looking at our strong solvency position and increased dividend.

First up, these are some of the highlights. And in a year dominated by COVID, the story is positive pretty much across the board. Considering the backdrop, customer growth was especially pleasing at 10% and turnover was also higher. The profit increase, as you saw earlier, was significant. And our return on capital was in line with 2019 at over 50%. Solvency ratio remained very strong, just under 190%, after taking into account a healthy 12% increase in the full year dividend.

Let's get into some of the details, starting with turnover in customers. As I mentioned, given the backdrop in 2020, which included some sharp decreases in demand for car insurance at times, I think this is a positive set of numbers. Increased competitiveness and very strong retention led to a very decent 9% growth in the U.K. Motor business.

Turnover was only marginal yet though due to the £100 million impact of the stay-at-home refund and material discounting in the second half. The household business continued its positive momentum, closing it on £200 million of turnover and 1.2 million customers, both up double-digits again. Our markets outside the U.K. were more challenging in volume terms, and so growth was more muted in aggregate compared to recent years. But progress continues, 13% higher customer numbers, 20% of the group's insurance customers are now outside the U.K. French growth remained very strong. Positive top line results from the comparison businesses. And as previously flagged, we saw a fall in the net loans balance due to the prudent approach of Scott and team that we'll hear about later.

And now to look at some of the detail behind the big increase in profit. And that tends to be the case with our results. Loss ratios are a key driver of profit movements, and that's especially the case with these 2020 results. Group profit was £112 million or just over 20% higher than 2019 at £638 million. Remember that the 2019 was negatively impacted by a long-term discount rate change, which hit the result by nearly £35 million. And so I think of the increase more in the order of £80 million or 15%, clearly still a positive outcome. The U.K. insurance result nearly hit £700 million, with the increase predominantly down to 2 factors: firstly, higher investment income; and secondly, despite the stay-at-home refund and the significant discounting, there was a notably lower current year loss ratio, which offset reduced, though still high levels of prior year reserve releases.

The household business doubled its profit to around £15 million, also seeing a better current year loss ratio, and associated profit commission income, pleasing results. A lower loss ratio also helped the International Insurance segment reported its first overall profit and a £10 million improvement from '19 to '20. All the European insurers reported positive results, and there was a lower loss in the U.S., too. We signaled at the half year result that we expected a smaller loss from Admiral Loans in the second half, and that's proven to be the case. And we report a loss of just under £14 million for the year.

Scott will share the detail shortly. But we end the year with what we believe to be an appropriate provision, and we're well set for 2021 and beyond.

Not too much to say on the other costs. There are some nonrecurring items in there such as costs related to the comparison disposal as well as some of the big projects like IMAP and IFRS 17. Share scheme costs, still a big majority, though with a slight increase year-on-year, notably due to the higher share price at the end of the year.

Moving now to look at the strong solvency position and higher final dividend. This top chart shows the capital position in terms of the requirement of surplus and the solvency ratio. As you see, the ratio is broadly consistent and strong, running at just under 190%. In 2020 versus 2019, higher profitability has led to notably higher capital, but also a larger increase in capital requirement than we usually see, reflecting higher profit commission risk from those higher profits. In absolute terms, the surplus capital level is around £100 million higher year-on-year, even after deducting that final dividend.

Speaking of which, as you can see on the bottom here, the proposed final dividend is 86p per share, so a touch under 90% of the H2 profit. And that's 12% higher than the final dividend from last year, with the deferred part added back. That's consistent with the H2 profit increase. A 12% increase in the 2020 full year dividend follows 11% increases for the past 3 years.

A few quick update now on some other topics. Firstly, hopefully, it didn't pass you by, but we announced back in December that RVU will acquire most of our comparison businesses, with the exception of compare.com. It remains subject to approvals, and we expect it to complete in the next couple of months. The reasons are hopefully well understood, and we firmly believe it's a very good outcome for all parties. We will confirm the plans for the proceeds after completion, most likely with our interim results announcement in August.

On reinsurance, I expect we'll conclude extensions for the U.K. business by the end of the first half. No big changes expected there. And just a reminder that we'd expect investment income to come under a bit of pressure in 2021. No major changes to the mix of assets over the last year or to be expected in 2021.

And finally, a quick update on our internal model progress. Following discussions with the regulator in the latter part of 2020 and earlier this year, we're taking some time to review the design of the model. And that will result in a further delay to application, which we now hope to be next year and not in 2021. More news later in the year.

And to wrap up, key messages from my section. In a year totally dominated by COVID, it was pleasing to see continued growth in almost all parts of the group, along with some reductions where we'd want to see them. Profit was significantly higher, with various factors contributing, including lower current year loss ratios in several businesses and a familiar positive story on solvency and dividend.

Over now to Cristina to update us on the U.K. business. Cristina?

Cristina Nestares

Good morning. I'm going to talk in more detail about the results of the U.K. insurance operation.

Starting with the highlights. We had good growth in our motor book of 9% as we decreased prices more than the market. We also had good profits due to reserve releases and the impact of COVID on frequency. But also COVID had a negative impact in our additional income, and we decreased its income due to less claims and a shift to digital. Beyond motor, in our household book, we had good growth, both in terms of units and profits. And we experienced a limited loss in our small travel book.

For 2021, we expect a higher increase in frequency, which will translate in towards loss ratio.

Let's start looking in more detail at prices in the market. It's been dominated by COVID. In the first quarter, we saw some price increases. But then since COVID, we started to see prices in the market decrease. In Admiral, we have done bigger decreases in the market. As you can see in the graph on the top, this is the -- a comparison of the Admiral time stop indexed to 2019. As you can see in the graph, in the last part of 2020, we became much more competitive. And actually, we were 20% more competitive than the previous year.

In terms of retention, similar story. You can see in the graph on the bottom that the Admiral retention has been traditionally higher than the market, but in the past few months, this gap has increased. 2 reasons for this. First, the decrease in our rates, which we have done towards savings to the customers, but also an increase in the trust in our brands following the refund.

So overall, strong increase in units of 9%. However, turnover only increased by 1%. Why this? Several reasons. First, the rate decreases, the rebates, but also a change in the mix of business that we had in the market. We have less young drivers, less new cars. And finally, a decrease in additional income.

Moving now to talk about loss ratio and profits. During 2020, we had a strong profit. Like in previous years, reserve releases are an important part of the profit. And we made a release of 23%, which is in line with our average. Also, we have kept a similar level of conservatism in our reserves.

We also experienced a very good loss ratio in 2020. In the graph on the bottom, you have the difference between the loss ratio in 2019 and the loss ratio in 2020. As you can see, very strong reduction in frequency, which has been partially offset by an increase in severity and a decrease in premium.

Now let me explain you in more detail what has happened to claims in the market. We've been talking about the reduction in frequency, which was especially strong during the first lockdown, but we have also had an increase in severity. Some of these have been caused by pressure on our guard sheets, which has made the length of the repair longer, increasing the cost of rate higher. Also, in the case of Admiral, we have provided ongoing financial support to parties and also to NHS and emergency workers.

On the bottom, you have 2 graphs that compare Admiral performance to the rest of the market. In the first graph, you can see the average of our claims costs, which are significantly better than the market. And also, on the graph on the right, you can see a comparison in terms of complaints per claim. I think it's very important to highlight that we managed to have very good claims costs, while also delivering very good service.

Moving on to additional income. As you can see in the graph, we can see a reduction of £5 when we compare the second half of 2020 to the year before. 2 key reasons that explain this. First, part of our additional income is made of income at nonfull claims. So a reduction in frequency has impacted this. Secondly, there has been an acceleration to digital, which has also impacted additional income. For the future, we expect this revenue to stay flat and not increase to previous levels. So even though we expect an increase in frequency, which will increase this revenue, we also feel that the weakness reform will increase the cost of providing motor legal protection by a few pounds. And also, we expect to see a continuation in the shift to digital.

In terms of our household results, we have seen good -- or 2020 has been a good year for profit and growth. COVID has had an impact on the claims. We have seen an increase in accidental damage, a decrease in the spend on escape of water and a reduction in such cases. In terms of growth, we have seen a good rate than to premium reductions that we have made to pass savings to our customers and also to the continuation of the cross-sell that we do with our multicover proposition. For the future, we expect to continue growing in size and profit, but we also think we will return to more normal levels of growth in profits.

Now before finishing, I would like to talk about our views for the future.

For 2021, we expect frequency to increase to more normal levels, which will impact the loss ratio, making it higher than it was in 2020. We think that the market needs to unwind some of the COVID-driven discounts that were made in the past 12 months. And even though the whiplash reform will impact prices, we think it's going to be a small impact. That is because the market has already started taking this into account.

In the case of Admiral, we have already started unwinding some of the discounts, and we have recently done increases in our rates.

Beyond this, what to expect from the FCA pricing change? Well, it is still early to know exactly what is going to be the content and the timing of this reform, but in any case, we think that Admiral is in a good place, because, first, we have a customer base that is very used to comparing prices every year. Secondly, we think that after the reform, being a good underwriter, it's going to be more important than ever. And that is basically handling claims well and being able to price based on risk, and this plays to Admiral's skills. And finally, we think that in Admiral, we have the capacity to adapt well to changes in the market. So overall, even though there is lots of uncertainty, we hope that we will be in a good position.

So in summary, 2020 has been a good year for the U.K. insurance business, good growth and a strong profit.

And now I'm going to pass it to Costi, who will talk more about international insurance.

Costantino Moretti

Thank you, Cristina. Good morning, everyone, and welcome to the International Insurance.

If you had asked me a year ago, do you think it is possible for International Insurance to achieve a strong result during the global pandemic, with everyone working from home and no face-to-face meeting for over a year and guaranteeing an excellent customer service? I probably would have say no, and I would have been wrong.

I am pleased with what the team delivered in 2020, an overall strong performance, proving that our business is resilient. We made good progress in the U.S. and delivered another year of sustainable growth in Europe.

Let me start with our U.S. business. In the U.S., we consciously prioritized bottom line improvement rather than aiming for top line growth. 2020 has been a year of good progress in the U.S., in particular, in 2 areas, risk selection and loss ratio improvements and operational efficiencies.

I know you might be asking, how can you say that Admiral has improved the loss ratio net of the positive currency effect? Well, if we take a look at the graph on the bottom left of the slide, it represents the changes in the loss ratio indexed at 100 in Q3 2018. You can see that Elephant, the blue line, shows a larger improvement than the market, the red ones. Hence, taking the market trend as a proxy of the positive COVID effect on the loss ratio, this implies that Elephant has also improved independently from the lower frequency related to the pandemic.

The improvement of the combined ratio over the last few years is a result of a combination of factors. We have focused on improving technical fundamentals as well as invested in digital enhancements to strengthen the business. We expect this to continue in the future, although not necessary at the same pace. In conclusion, good progress in establishing a sustainable digital insurer in the U.S.

Let's now move to Europe. For the first time, all the 3 European businesses are profitable, and we achieved a record profit on a combined basis, again, driven by our largest operation in Italy ConTe. The progress of the European insurance is evident also net of COVID, although less strong. Indeed, whilst the pandemic has had an obvious positive impact in lowering the claims frequency, it has at the same time significantly pressured our volumes and periods. We estimate that removing the COVID effect, this would still have been a year of growth and profitability, with Admiral Seguros and L'olivier nearly breakeven and Condé largely profitable.

Digging deeper into the top line figures, there are a couple of additional points I would like to make. L'olivier in France delivered a very strong growth for another year in a row. And we continued to experience our premium conditions in Italy and Spain, which resulted in a modest growth in [indiscernible], whilst we achieved a notable growth in their customer base.

Finally, in Europe, we are progressing well in delivering more digital services to our customers through agile operating models that help us in making our customer happier and our business thus more efficient.

In conclusion, International overall delivered a strong performance, proving that our business is resilient. We made good progress in the U.S. and deliver another year of sustainable growth in Europe.

Thanks. Now over to Scott and our Loans business.

Scott Cargill

Thank you, Costi. Good morning, everyone.

Today, I would like to share with you how our Admiral Loans business has also proved to resilient in 2020. Yes, our growth was impacted by COVID. As we took the prudent decision to post new owner regulation for part of the year. Having said that, our customer think is, and we made large provisions on the projected future unemployment. We finished the year optimistic of the near-term future, and we will return to growth in 2021. One of the things that I'm proud of and that gives me confidence for the future is that through Admiral way, we deliver the right solutions for customers, and they like what we deliver. You see two factors here. The first feel that in early 2020, is a picture of us serving our 100,001 customer in the office. The second is from just last month, when Admiral Loans won best lender in both comparison of loans and car finance cost as the money factor awards. These awards are explicitly voted for by customers, a nice way to start the year.

As Geraint evident, we have reached £13.8 million loans for 2020. What you can see is asset provision charge of £25.8 million. This is £11.5 million higher in 2019 and includes £15 million for loans, which remains fully up to date, reflecting an expected increase in U.K. unemployment. We have a track record of consistently improving actual loss performance. As you see in the chart on the bottom left hand side. This shows 12-month customer payment default outcomes for each monthly cohort since our launch in 2017. And it's pleasing to see from industry benchmark data we're among the best performers over that period. You can see the actual loss performance has consistently improved during that period. And then 2020 has, in fact, been in line or better than prior periods. Also, we're highlighting is that we finished the year with less than 1% of the book on payment holidays. What this looks like in coverage ratios is an increase of 4.9% at the end of 2019 to 10.4% at end of 2020. And importantly, a 3.8x increase on the provision for performing loans from 1.5% to 5.8%. This reflects uncertainty in the macro environment.

We started 2020 proving the growth capability of the business by writing record volumes during early Q1. As you can see from the chart on the bottom left-hand side, when the COVID first emerge, we responded quickly, and for this new business entirely in order to focus our attention on existing customers. We reopened to new business in July, where we chose to take a cautious approach for the rest of H2, having now evolved our models under the right approach, have returned to cautious growth in 2021.

Looking forward, we believe the evolution of the market plays to Admiral's strengths, where we believe key differentiators will be risk selection, product innovation and a low-cost to serve. Consumer behavior is also changing to be ever more digital and a wider range of behavioral data becoming more important to protect that the loans will perform in the future. We also welcome the regulatory direction, the recent role review of the unsecured market, which does suggest in time and with more transparent pricing and products, which aligns with Admiral Loans' aim to provide affordable lending products for net range risks via real rate pre-approved offers.

In terms of guidance, we expect gross loan balances in the range of £500 million to £550 million by end of 2021.

In conclusion, Admiral Loans has also proved to be resilient in 2020, validating the strength of what we have built over the past 3 years and further convincing us of the opportunity to build a scalable business, making a significant contribution to the Admiral Loan. We had a rapid response to COVID and paused new loan originations for part of the year, resulting into fair growth. And we are optimistic for the near-term future of return to growth in the early 2021.

With that, I'll pass back to Milena to wrap up.

Milena Mondini

Thanks, Scott. One of the collateral effects of a crisis, it is to test what people and business done for, are there priorities and the balance between the year now and the long-term change went under pressure? So what does Admiral stands for? This came out even more clearly for us this year. We want to ensure that our products deliver good value and are fairly priced, and therefore affordable and inclusive for more people, even when this meant giving back £110 million. We want to help and provide peace of mind for the future, particularly when the future is more uncertain, as it was the case this year. We want to look after more people, particularly when they need it the most. And we're trying to find new ways to do things better, every day, little by little, by using more data, testing and learning, always trying for excellence.

And we do so together as a team that is deeply ingrained in our culture.

In summary, help more people to look after their future, striving for better together. This is the revised purpose statement of Admiral. And you can read through -- -- you can read these through different lens. First, the customer, as we protect what's important to them. Second, our staff, as we empower them to achieve the full potential; and third, the larger community. Not only has we provide great employments opportunity in a company that is a great place to work, but also, as we contribute to address another challenge as social mobility or climate change. And we want to make a contribution because we care. Is this alignment of different stakeholders has always been a distinctive picture for Admiral and is very ingrained in our model. We do have long-lasting relationship with our partners, the distribution, our insurers, our customers who stand stay longer with us, and our staff who has impressive average tenure in the business. And we managed to do this, not only because we really care, but also, we think because we take a long-term perspective in our decision-making. And something that underpin this is our reward system that is based on Admiral shares rather than short-term incentives. All Admiral employee are shareholders.

In conclusion, we continue to strengthen the company year after year. Our strong foundation has supported positive results that we've delivered and outcomes for all shareholders. We've been progressing product during our core business and U.K. And our unique focus on creating long-term sustainable business is preferring us for tomorrow and beyond.

Now before we close, I would like to take a moment to say a big thank you to David. Thank you, David, on behalf of all Admiral team for your invaluable contribution to the business and for being such a great reference and fantastic role model for all of us.

Thank you very much for your attention, and we are now happy to take questions.

Question-and-Answer Session

Operator

[Operator Instructions] And the first telephone question is from the line of Jonathan Denham from Morgan Stanley.

Jonathan Denham

Firstly, U.K. insurer is going to be impacted by an increase in the corporate tax rate. I was just wondering how much you thought would be passed on to consumers via higher prices? And then second, today's release had you confirm your intentions for use of the Penguin's portal proceeds after completion. Just wondering if you still expect to return the majority of the net proceeds to shareholders?

Milena Mondini

Thank you for your question, Jonathan. And Geraint, do you want to take these 2?

Geraint Jones

Can do. Jonathan, on the first one, on the tax increase, I wouldn't expect us to pass that to customers. I think our approach is to pay up most of our net profits to shareholders. And so I'd imagine that the impacts might be felt there rather than in pricing. And I didn't catch the second part of the question. So could you repeat?

Jonathan Denham

Sure. I was just saying, today, I think you said you'd confirm the intentions for the use of the Penguin portal proceeds. On the initial release, you said you'd expect to return a majority of the net proceeds to shareholders. I was just checking if there was a change there or it's still expected to return majority.

Geraint Jones

Yes. No change in that. We'll confirm the plans after completion with the results in August.

Operator

The next question is from Freya Kong of Bank of America.

Freya Kong

I've got 2 questions on U.K. Motor. So first question is, your reserves are typically set quite prudently because of bodily injury, and they tend to unwind in outer years. Could you help us understand the composition of the 2020 accident year reserves and whether we should expect to see a lower level of unwind down 2 years? Second question is, we've seen a lot of positive development on the 2019 book loss ratio, from 92% down to 76%. What's driven this? And will you can sit -- will you continue to set loss rates this conservatively in the future?

Milena Mondini

Sorry, would you need. Thanks a lot for your question. Cristina, do you want to take the second one, and Geraint, take the first one?

Cristina Nestares

Yes. Thank you. So on the reduction of the loss ratio in 2019, several reasons. The first one is the normal releases that we expect given our conservative approach. But secondly is that, when we look at 2019 on an underwriting year basis, there was a reduction in frequency that affected some of the policies underwriting that year.

Geraint Jones

On the first part, the 2020 claims cost does have lower levels of large injury claims than in the previous years. The mix is actually slightly different, and there's actually a slightly bigger a large new claims in the total, but the absolute number is clearly smaller with the lower loss ratio. So in absolute terms, it's probably fair to say that there is less positive development that came into future years on 2020 accident year. But the approach remains the same, with the numbers that you see in the slide will be cautious. And the numbers that we book in the accounts will continue to be cautious. And the relative size of the buffer in the claims reserves is flat year-on-year.

Operator

The next question is from the line of James Shuck of Citi.

James Shuck

My first question is around the new level of conservatism on reserves. And so within that, if I just look at the 2020 margin as an assessment, which I would look at on an underwriting year basis, so I look at be ultimate versus the booked. And it looks like you're booking the loss ratio then 9 points above the ultimate for the 2020. In normal years, in 2018 and 2019, that margin has an assessment of that underwriting year. It looks like it's normally around 5 points as opposed to 9. So if you could just comment on that. And linked to that, if I look at the book loss ratio in the accounts, which obviously is a slightly different basis. But it improved by 20 points to 72%. The accident year bridge that you show, which was on a different basis, but that's improved 11 points. So if you could just help me square that movement as well, that would be helpful.

And my second question is around the FDA reform. So obviously, there's going to be changes to the pricing of the business. And despite new business pricing will go up in order to justify the same customer lifetime value, as those products are introduced into the market, which should look very much likely that's going to happen in 2022, would you expect a one-off earnings impact as the new business prices through? And how do you think about the EPS impact as that happens?

Milena Mondini

Thank you very much for your question. Geraint, do you want to take the first one? And Cristina, take over the FDA reform.

Geraint Jones

Yes. Thank you. James, the -- I'll take the second part of the question first, which is about the movement in the book to loss ratio in the accounts versus the ultimate in the slides. What you saw in the slide is the '19 accident year improved a few points. So if you look at the original 2019 estimate on accident year basis, that was 79, actually, low 80s. And that improved something in the order of 15 points to the first projection of 2020. And I think the movement in the book to calendar year loss ratio in the accounts was 14 points. So we consider that to be basically in line. And first part was about the amount of conservatism is booked in the current underwriting year. As you say, it is 9 points on 2020 booked versus ultimate underwriting. And there's no particular reason for it being slightly bigger than in the previous years. And it's been in 9 points or actually higher before. We think of the margin, as I think we've said before, in aggregate, across all the underwriting years, you usually see a pattern where the level of conservatism, i.e., the gap in book versus ultimate loss ratio will increase as the year becomes less developed, and that's typically the case. But no particular change, the level of conservatism, i.e., the book loss ratio overall divided by the best estimate claim reserve overall is flat year-on-year.

James Shuck

Before we move on because it's quite a complex topic. So just very quickly, I mean it looks to me like the 2020 book to loss ratio in the accounts actually improved by 20 points, from 92 down to 72. But obviously, some of that is going to be frequency-driven. But again, I understand that that the opening has moved down and the expectation you've just given, but it still seems a large movement because -- maybe we can take it off-line. And if you can add anything else.

Geraint Jones

Yes. If you look at page, I mean if you go to the accounts release in front that we -- if you look at Page 23, we've got a reconciliation there, shows us move from 87 down to 72, which is 15 points, and one point is explained by the impact of the Ogden discount rate change last year, so it's net down to 14.

James Shuck

Okay. I was looking at the loss changes on Page 71 of the results, but I'll start the back on that. That's fine.

Milena Mondini

The second question?

Cristina Nestares

Sorry, apologies. Just to answer the second question he asked about the impact of the FCA reform on market prices. So I would agree with what you mention, that we expect both for car and for household and increasing new business prices basically right after the FCA reform is implemented. And then a decrease in renewals. However, it's going to take some time until the market adapts all these changes, and we fully understand the dynamics in the market. Just to point out that new business in general tends to be more elastic and renewals more inelastic. And therefore, it will take some time and it will take some changes until a profitability in the market is the same.

James Shuck

So you wouldn't expect any earnings impact in 2022, 2023 if the proposals are enacted?

Cristina Nestares

There could be -- yes, there could be an impact, especially when the reforms becomes live, and we could see an erosion in margin. But I think that it should be temporary and companies will adjust prices accordingly. In the case of Admiral, we feel that because our customers are very used to shopping and comparing prices every year. And because we tend to be very strong in underwriting, both claims handling and pricing, we should be better equipped to take advantage of this reform.

Operator

The next question is from Ben Cohen of Investec.

Ben Cohen

I'd just like to ask on Slide 18, where you have the bridge on the accident year, and you discussed the improvement in frequency and the worsening and severity that you saw '20 or '19. I just wonder if you'd make any sort of forward-looking comments in terms of how you see frequency and severity developing this year. And maybe a little bit more detail around what the swing factors could be. And also related to that, if you'd be prepared to say a little bit more about what you're seeing on motor pricing now in the market, please?

Milena Mondini

Cristina, I think that will go also for you?

Cristina Nestares

Yes. Well, first, in terms of severity, we think that the impact or increasing severity of just basically repairing costs is related to COVID, and therefore, it's temporary. So ex-COVID, we don't expect a significant change in the severity trends that we have seen in the past. However, with last reform will start impacting claims from -- or after May 31. And therefore, that would start impacting severity and frequency of small BI claims. In terms of frequency, it's a bit harder to predict. It really will depend on what is the evolution of COVID and how the lockdown measures evolve. If we take the current estimations from the English government, we're talking about frequency towards the end of -- sorry, towards the end of June at the time where things could start going back to normal. However, the new normal will possibly have lower frequency that we have seen in the past.

You also asked about what we're seeing in prices in the market. As you saw on the graph, we were mentioning that we have seen prices decrease even more in the last quarter of the year. Also, the confused index for the month of January said that prices during that month had decreased 5 points versus December, which is a very strong decrease for a month. In this environment, in Admiral, we have started putting prices up. We think it's the right thing to do because we expect frequency of the policy that we're selling today to be higher than the policies we were selling before, basically because of the impact of the measures of lockdown easing out. Hard to tell how the rest of the year will evolve, but we think the market will need to unwind some of the discounts that were given in the past 12 months as frequency continues increasing.

Ben Cohen

Sorry. Could I just ask a follow-up on that? Have you seen in putting up prices with the market reducing prices? Have you seen a material impact to your retention and your new business, please?

Cristina Nestares

It's hard to mention. It's hard to see it because the price increases that we put are very recent.

Operator

And the next question, is from Thomas Bateman of Berenberg.

Thomas Bateman

Just the market is clearly a little bit concerned about the discounting that you put through in H2. I guess what can you say that gives you comfort on these actions? I mean, obviously, we're still quite a large amount of claims frequency benefit to come through in the first half of this year. But what else would you point to sort of justify those price decreases? And secondly, [indiscernible] quite well in government and social aspects of ESG. But thinking bigger picture, what are you doing to grow faster transition to electric vehicles? Can you talk about your EV book at all? And maybe and slightly a radical thought here, but could you envisage a market where you subsidize EV somewhat from the combustion engine vehicles that you write?

Obviously, this has to be a kind of market-wide reform. But is that something that is being talked about at all?

Milena Mondini

Thanks a lot, Tom. Cristina, do you want to answer the first one? And I go with the second.

Cristina Nestares

Yes. In regards to discounts in the second half of last year, we feel comfortable that, that was the right thing to do based on the frequency packages that we will see at the moment, but also the frequency we'll see now. So the policies we saw during the second half of last year benefited from the second and the third lockdown. And they are seen -- or basically, these policies are experiencing less frequency than before.

The second point is severity trends. As I mentioned, there was an increase in claims cost inflation, but that was mainly related to lockdown. And we're seeing those trends unwinding. But the thing that should really give you comfort is that Admiral has been independent, very rational. And the fact that we have already started increasing prices is the best signal that we will continue to be rational. And we could continue to price policies adapting to what we see in the market, especially in terms of frequency and severity.

Milena Mondini

On your second question, Tom, I think ESG is very important for us. We welcome more attention of the market on this. And we are investing more resource and tuning our plans. On the climate change, more specifically, and the environment, the way we look at this is threefold. The way we operate, so as I mentioned, very pleased to say that we are now offsetting our carbon emission. Our carbon emission has been verified by the trust. And we're also looking at waste management and using 100% renewable energies and many more things. Also looking at investment and how we invest in a more sustainable portfolio over time and growing in that regard. Probably the biggest impact is, as you pointed out, is supporting this change in these trends of electrification, that is going to be quite massive, expected to be 50% of new vehicles in U.K. being electric by -- in the next 10 years. As we have done for telematic, we intend to embrace these trends and invest resource to make sure that we become a great underwriter for electric vehicles. And I think that's going to be the biggest. And understanding the specificity in terms of claims, understand the -- how price electric vehicles better and try to help the customers that deserve to have a cheaper price, and in this way, support the electrification. There are also other little initiatives that may indirectly support the world with less carbon, like some initiative around a product that provide more flexible insurance terms. So some people maybe decides to not own a car, but rather use it on a temporary base, and I mentioned vehicle, sorry, as an example, earlier on. So just to mention that, also, we do already underwrite a lot of electric vehicles. We have a good share of the market on electric vehicles. And as I said, looking forward to increase this over time. We will have spoke more and speak more about our target and our commitment at half year.

Operator

The next question is from [indiscernible] of Deutsche Bank.

Unidentified Analyst

Just 2 from me. So first, you talk about Admiral 2.0 and also product diversification. Can you provide a bit more color around both of those? And any associated investment spend for them? And second, you're keeping compare.com in the U.S. What's the long-term plan for this business? And are there any targets that you can give to us?

Milena Mondini

So on your first question, both Admiral 2.0 and product diversification, it's not something new for us. The main message for me today was that we are accelerating on those areas, particularly in Admiral 2.0, 2020 was a very strong year, partially because of some external factor that were accelerated by COVID. It is a trend in the market, and we are investing more. If you look at our expenses, our expenses in tech, in general, has been growing in the last few years. I think now we're at a higher level.

We don't expect to necessarily additional grow in the future or substantial grow in the future. But what's normally happened is that you increased a bit of the spend in tech in digital and you may have some other lower expense that are driven by more efficiency as well. So not necessarily major expectation in terms of the overall spend as a whole.

Our priority in terms of Admiral 2.0 is not only the digital, but it's also the data and analytics. There is so much more data available and also tools in terms of advanced analytics and methodologies. And we do believe there is the potential to do more. And this is going to be a continuing trend in the next future years. There is also an element of how we operate. So we move to agile and smart working. Remote working are also important features of this evolution. Product diversification is also something that David announced. When he took over as the group CEO, we spoke about an ambition of diversify outside motor insurance and outside insurance as well. If you look at the next 5 years, probably the existing business like loans and households are the one that's going to have an impact. And we believe they're going to contribute positively to our bottom line as well. In the longer term, we hope to be able to prove our competence also side those business. And Admiral Pioneer, as I mentioned, is just a vehicle that will help us to accelerate this investment and this exploration. I don't necessarily want you to believe or to take away that we're going to massively increase our investment because we will continue to test and learn and use a relatively cautious way and approach, as we've always done in the past.

Cristina Nestares

Your second question was on compare the compare is -- we always look at our strategic option, and we evaluate them every year and regularly for all the business. Compare is -- it was at early stage compared to the other price comparison platform, but also different market, yes, that requires more investment. And it's a longer learning curve that acquisition costs are higher, so it takes a bit more time to mature.

We are -- we reduced a lot of our investment in compares and brought it to a level that we feel comfortable and improve a lot of underwriting metrics. So we're pleased with the progress this year. We remain committed to continue to make a success of it. Having said that, of course, we always revalue our options.

Operator

The next question is from Andreas van Embden of Peel Hunt.

Andreas van Embden

Yes. I've got 2 questions, one on reserves and one on capital. Starting with reserves. There's still significant reserve releases coming through from 2015 and prior years. And it's also impacting profit commissions from that period, which is still quite elevated. How much of is left here and sort of in the tank on these older sort of vintage years? And what is driving these releases?

And my second question is on capital. I'm just curious. What are the key final hurdles you have to sort of take in rolling out your internal model? What is causing the delay? And could you put your finger on the key issue?

Milena Mondini

Geraint, do you want to jump on both?

Geraint Jones

Yes. We'll do. Andreas, on the latter point, on the internal model, it is a bit of a frustration that we are seeing another delay. It's a very tough process, I know. As I'm sure you know, everyone has gone through, it will have, I'm sure, experienced a different time line to the one they originally set out in their plan. We're keen, very keen to get the application right in a very good standard when it goes in rather than trying to rush it in. And we think that will probably be counterproductive. There aren't huge major problems. And there are certainly no concerns, certainly from our side, at least on the numbers that are coming out of it. And the regulator is not indicating they've got any problems with the numbers per se. I think it's particular to look at the way it's designed, the -- particularly the approach to modeling large claims. It's a very important risk for us. Reserve risk is our biggest risk. And large claims are an important part of that. So we're taking some time to think about that. Nothing to be concerned from a numbers point of view, like I say. We'd expect to get back into the pre-application process this year and then application next year.

On releases, I think that is a fair point, that it is slightly higher than usual, I think, on 15 and 16 at this particular point. We would expect book loss ratios to develop over that kind of time. 4 or 5 years is not unheard of, some of the loss ratios to get down towards the ultimate level. What you see in the ultimate chart in the slides is that they stop moving at around about that time. So there are a couple of years on the far left-hand side of that chart, where there's been no development in the ultimate. And so once that starts happening, we'd be expecting to approach the ultimate point and be surprised to see much development on 15 and 16, so 50 and certainly on a booked basis after 2020, if that helps.

Operator

The next question is from Alexander Evans of Crédit Suisse.

Alexander Evans

First one, I'm just thinking about the risk for higher inflation in the market going forward. Does that change your potential view of claims inflation? Or do you think current pricing actions are sort of capturing this and you've seen a lot of the movement already? And then secondly, just on the outlook of loans. I was wondering what your view on the U.K. unemployment was after yesterday's announcement and the revised forecast there. Is there a risk that you're too conservative with this year's provisioning?

Milena Mondini

Thank you. And Cristina, do you want to take the first one, then Scott, the second?

Cristina Nestares

Yes. In terms of claims inflation, in general, the market tends to be very quick to adapt. And as you said, they translate this into pricing actions. The more worrying trend is when you have a very unexpected increase in inflation, especially for the older claims, because you might end up having to pay a claim, like a large BI claim in a few years, and then a high level of inflation could affect that. But in general, and in terms of the more normal inflation, the market prices.

Scott Cargill

And on -- Alexander, the main assumption that we took in our provision was that the weighted unemployment would be 9.2. I think that's a key figure. If you look into the notes, I think we showed some sensitivities. And the latest forecast given after yesterday's budget would indicate that it's nearer to our up term scenario. It depends on the timing. I think the previous forecast was indicating a peak in the summer, and a peak of around 7.8 from the Bank of England. Latest forecast indicating a peak of around 6.5 in Q4. So there may be implications, but we'll have to wait and see how it plays out.

Operator

The next question is from [indiscernible] of Field Gibson Media.

Unidentified Analyst

2 questions from me, please. Firstly, just on general basis, how the experiences you had in 2020? How does that shape your kind of outlook for the future? And secondly, obviously, it was mentioned during the presentation concerning the FCS proposals about general insurance pricing, on price working, et cetera. I know that was mentioned in the presentation. But can I just get your thoughts on that and what your kind of general feeling is about those proposals, please?

Milena Mondini

Thank you very much, Paul. Is your question on the outlook of the future of motor insurance or more in general? Can you repeat the first question? Sorry. I think I didn't get it.

Unidentified Analyst

Sure. Yes. It's largely in general, just about your sort of past experiences, how they've shaped your outlook and risk modeling, et cetera?

Milena Mondini

Okay. So 2020 was definitely not like an ordinary year in terms of its influence, like COVID impact on results and COVID impact on the motor frequency. I think in general, we expect a bit of a return to normal in the future. 2021 still will have a lot of impacts from COVID, and so we will see a bit more of what's happened in 2020. But the big difference is that the premium that were collected towards the end of 2020 and beginning of 2021 will already have embedded a lot of assumption for a reduction in frequency. So if you think about 2021, probably the underlying underwriting year loss ratio is going to be higher, and therefore, the margin lowers. Of course, part of our model is also prior year reserve release. That also need to be taken into account. In terms of the market, we expect that the other big impact outside COVID is going to be the FCA pricing reform that Cristina spoke about before. And that's, again, that is something that may be material, but it's going to take time. It's going to be a revolution after revolution because a lot of customer have these habits to shop around.

So we'll see some increase of new business price and renewals price, but signs and extend is still there to be seen. And after that, in the longer term, I would expect that those 2 things to normalize a bit more and so probably to come back to what we were experiencing before 2020.

In general, we also see like digital trend to keep extending, and data analytics and a lot of those things are also to continue. But I think, in general, we'll see a bit of a back-to-normal. But COVID and FCA are definitely 2 things to take into account.

I think we are in a strong position, as I mentioned. We invested -- we are not a lot investing, navigating the year now, but a lot in the future. I'm quite excited about the prospects for us also outside U.K. Motor, in International Insurance, in Household and Loans in particular. It was a bit of a high-level view of the future. Happy to follow-up if you have more questions.

And the second one was on -- gosh, remind me. I see. Yes, Cristina, do you want to take this one?

Cristina Nestares

Yes. So I understood the question to be focused about our views on the reform. So I want to start by saying that we welcome this reform. That we understand this is something that wasn't working very well in the market and we think it would lead to better practices in the future. It's hard to comment exactly because we don't know yet that the content of the reform fully and when it will be implemented. But overall, we think it's going to cause important increases in new business, which certainly will have as a consequence that some customers that are very used to shopping and tend to benefit year after year from reduced prices are going to see a significant increase in the prices. So that will be our biggest concern.

In terms of Admiral, we have a very large renewal book. So of course, this could have a significant impact. We think our customers are very used to shopping around, compare prices on a regular basis and are happy with our level of prices. So we hope to be in a stronger position than some of other players.

Also, as the FCA has mentioned, the household market is going to be particularly impacted because that's when you see a much bigger difference between the older renewal books and the new business. So we expect to see much more changes there with very high increases in new business and also high decreases at renewal.

Overall, wait and see, but it could be an interesting time for Admiral.

Operator

The next question is from Phil Ross of MedioBanca.

Phil Ross

Just one question from me on Solvency, please, on the FCR increase, specifically. If you could just expand a little bit on the moving parts. I guess this is mainly driven by the capital add-on. And obviously, you mentioned the increased expected impact from profit commissions. But perhaps within that, we might have expected a lower charge for PPOs if they remain less attractive given where rates are. So yes, just a little bit of extra detail on that would be helpful, please, if you can.

Milena Mondini

Thanks a lot for the question. Geraint?

Geraint Jones

You're quite right to say that there's very little change in the element of that capital requirement, which is PPO-related. The big increases, it's profit commission risk. So that's the amount of profit commission risk that would be derecognized from capital in the 1 in 200 scenario. In the past couple of years, profitability has increased quite a lot, as we've talked about in 2020 and 2019 underwriting years, particularly. And when loss ratios improve to that extent, there's a lot of profit commission income that we would get. And clearly, early on in those years' development, that's still at risk. And so that's what's driven the increase we see in the 2020 year-end position versus '19. So very limited change in PPOs. Big increase in profit commission risk at this particular point. We'd expect that to subside over the next couple of years.

Operator

The next question is from Ivan Bokhmat of Barclays.

Ivan Bokhmat

I have a few questions. The first one would be regarding the reserve releases. Just wanted to make sure we're still talking about this midterm 15% to 20% release guidance, I think, given some of the changes at least within that. And maybe also, if you could talk a little bit of how should we think about the commutation revenues given 2020 was so unusual. And how will the commutation be booked in the future?

The second question, I wanted to ask you about the reinsurance renewals that you mentioned, would be completed over the second quarter. I'm just wondering if you're looking for any structural changes there that could affect your levels of retention or the profit commissions you'll get.

And maybe the final question, a rather small one, if I may. We did hear about the success of your refund that has boosted the retention levels. I'm just wondering if you could give a little more color on that. At what level is your retention at the moment in the motor market and whether that trend has persisted since May?

Milena Mondini

Geraint, I guess the first 2 are still for you.

Geraint Jones

Yes. I think I'll actually get 3 actually. I'll try under 3 as you have. Ivan, on reserve release outlook, I think it's a fair point. We've historically said that 15% was the average. That's quite a long-term average. Over the past 4 or 5 years, it's been in the low mid-20s. And in the immediate future, I'd expect that sort of level to repeat. We've still got a conservative reserve, which hasn't really changed year-on-year. So there's -- if things develop as we expect them to, there's quite a lot of release to come out. So no real change in the outlook there.

On commutations, the impact this year has been less visible, less material than in previous years because of where we are booking the most recent underwriting use, i.e., at a profitable level. So the impact of computation on the accounting will really depend on the profitability and the booking of individual years. I'm happy to spend a bit more time on this. That may be a bit too detailed for the call.

And reinsurance renewals. We aren't, at this point, expecting any structural changes or anything fundamental, I think. And we're hoping to get that done. I'd probably roll forward in a pretty similar fashion. We'll expect to get that done by the half year.

Milena Mondini

And Cristina, do you want to comment on the impacts on retention of the COVID refund and where we stand versus market?

Cristina Nestares

Yes. Well, it's hard to give concrete figures about what was the impact of the refund on retention. What we have is 2 things. First, clear improved results on our brand tracker. So we -- on a regular basis, we survey customers in the U.K. and ask about several brands, and also their level of trust, whether they are -- they will consider them for the future, whether they will choose them. And there is a clear, clear impact of the refund. Secondly, in terms of the impact on retention. What we have is a lot of anecdotal feedback on the website comments and also on the phone, on customer saying, maybe the price is higher than they would like it, but they are comfortable to stay with us because of the refund. I'm not sure how long this will impact our results, but it's clear that it will stay there for some time as we continue to hear some of this feedback.

In terms of actual data on the market, we don't disclose this. But I'm happy to share that, in general, and despite our channel being price comparison and our average premium being slightly higher than the market, we tend to have retention that is slightly higher than the overall market. And in the past 6 months, it's been almost mid-single-digit gap.

Operator

And we will now go to the webcast questions. Please go ahead.

Marisja Kocznur

There are 2 questions from the webcast. The first one is related to customer growth. The question is, the volume of new customer growth is substantial. Are you comfortable on the infrastructure in place to manage this new volume growth? Would you be comfortable repeating this growth again in the near-term if opportunity arose?

Or would you be looking to consolidate this level of growth and ensure margin is sufficient? That's the first question. And then the second question, related to comments on more insights into household pricing, claims frequency and claim severity patterns that we've seen and expect to see in 2021.

Milena Mondini

Cristina, I'll start with the first and then happy for you to chip in and add on top and move to the second.

So in general, we were happy with the growth. We tend to be very rational in flexing our growth depending on the market condition, and also, in this particular case, on our expectation of how frequency is going to change depending on COVID, what's going to be the impact of the reform to come whiplash and FCA. So we'll take a rational approach to the growth in the future, as we've always done. We tend to react relatively early to change in the market and in the external factor, and we'll try to continue to do so. So we'll be seeing -- we'll see what competitors will do and what the market will move. We tend to optimize our own decision based on our own data point. Our infrastructure is solid for growth. We have very solid infrastructure and operation, and that's true for U.K. and overseas. So we will take the growth as it comes. It's what is best for the business.

Cristina, I don't know if you want to add anything on top or move to the second part of the question.

Cristina Nestares

Well, just to say, Milena, as you mentioned, that we are comfortable to take growth because we understand we operate in a very cyclical business. And if you look at our track record, there have been years where we have grown at a similar level as in 2020. And there have also been years where we have been happy not to grow. So yes, we operate in that sense. Also, in the next couple of years could be particularly interesting. We don't know yet, but there might be an opportunity once the FCA reform is implemented and prices in the market go up. So we might need to be ready to take volume very quickly. And also, I think the fact that we have done a very strong move to digital and we have a very strong system and team makes it easier for us in the future to take more growth.

Moving on to the trends in household in terms of claims. Well, in terms of frequency, we saw a reduction in claims frequency in the early lockdown. But after the lockdown, it was recovered, and it went back to much more normal levels. In terms of severity, what we have seen is basically a change in claims mix. So more accidental damage, although small, reduce set of claims, clearly, as people stay more at home, and then severity around escape of water hassles have been reduced. And we think that might continue in the future. So it's clear that as people are allowed to go out more and to travel, we might see claims going back to previous patterns. But we think some of these changes that COVID has brought in terms of spending more time at home will continue in the future. So our expectation for 2021 is similar to 2020, but less strong COVID impact.

In terms of whether 2020 has a bit of a bad weather, especially at the beginning of the year, we had a few storms costing Admiral actually £5 million. But this year, so far, the weather has been quite benign. So that's also important to consider.