Safestore Holdings Plc (SAFE) Earnings Call Transcript & Summary

Hello and welcome to Safestore's Full Year Results Presentation. [Operator Instructions] I will now hand you over to Frederic Vecchioli, CEO of Safestore Holdings. Frederic, please go ahead.

Thank you very much and good morning, everybody. I'd like to start by putting a bit of perspective on our journey. I've been building this company for 25 years. I've seen phases of acceleration and phases of deceleration. But overall for anyone that look at it from a long-term perspective, it has been a story of fundamental value creation. And my view as an investor is that the fundamental of the equity story are as strong as ever when one look beyond short-term volatility, which we have highlighted in the graph that shows our annual storage revenue have evolved monthly over the last 20 years. It is pretty clear in the last 3 years we outperformed, I think we outperformed the sector and we also outperformed our own longer-term trend of revenue CAGR of 8%. So it should come as no surprise that, that is followed by a phase of consolidation, which is the way I would describe 2023. The drivers of these short-term fluctuations are simply the fluctuations of inquiries. Our conversion rates and pricing power are unchanged and they rarely move. The way customers behaved on the journey is very stable. The single and only driver of these variances is the fact that levels of demand fluctuate on a short-term basis. But fundamentally, it is a long-term growth of a relatively new asset class through various economic cycles. Coming back to our journey. Out of an historical investment net of debt of GBP 600 million, we have already paid back GBP 400 million of dividends and we have created a net asset value portfolio of GBP 2.3 billion without asking shareholders for more money, apart from a miniscule GBP 30 million raise 10 years ago, which I deeply regret. Since that, we delivered a total shareholder return of more than 600%. That is our business model: invest wisely in prime locations, deliver operational excellence as shown by our REVPAF, create values and cash flows which enable self-funded future growth. And there is still a long way to go to carry on with that strategy. We have already built a significant pipeline of 18% of our current MLA, which together with the non-like-for-like stores should deliver more than 30% EPS growth on top of the growth that the like-for-like stores will also provide. Our loan-to-value is only 25% and it is based on what we consider relatively conservative valuation when looking at recent transactions and there were close to EUR 1 billion of transaction last year and our loan-to-value is well below our guidance of 30% to 40%. The team has built a pan-European management platform that is ready for further growth. And coming back to 2023, I'd like to go straight to Slide 7 now to talk about inquiries. As you know, now it all starts online. Investing in our digital platform is something that we have done and we continue to do constantly. We have always had a state-of-the-art leading website in our industry and we think that is a differentiator compared to the vast majority of our smaller competitors that have nowhere near the capabilities of the major players. In our portfolio of keywords, we constantly achieve very high impression shares and absolute top rankings. The cost of our paid search operations is modestly up on last year as a proportion of revenue from 3.6% to 3.8%, still well down on close to 6% 6 years ago. If we look at how our inquiries have evolved over time in our 2 main markets, U.K. and Paris, that is on Slide 8. I think it illustrates how putting new supply over the year is what creates the market and creates the demand. It was harder to fill the initial 3, 4 stores I ran in Paris 25 years ago than it is now to fill any new store that we add because the awareness and the demand have increased over time and as soon as I add a store, customers move in. Demand structurally grow as more people use the product, which is in part driven by new supply. Supply is constrained in the best market by the barriers to entry and new supply in the good markets in which we operate comes at a slow pace every year. In the last few years, COVID was clearly an accelerator. I guess it created more projects and changes in people's lives clearly faster than normal, well above the secular trend. So a bit of pullback is in my view normal. Year-on-year inquiries in 2023 were down 7% in the U.K., but they were still up 41% on pre-COVID. The London versus non-London split of inquiries and move-in has been remarkably stable over the last 10 years. Both London and non-London growing consistently over the period. The share of London inquiries has been fluctuating over the last 10 years within a very tight range of plus or minus 1.5% versus non-London. It is in reality super stable. In Paris, inquiries are marginally up year-on-year around 1% and up 21% on pre-COVID. In the new territories on Page 9, the growth of inquiries was essentially driven by the implementation of our digital platform in the business that we purchase. If you look at Spain, 2023 growth was also fueled by the opening of Madrid stores. But before that, the acceleration in 2020 to 2022 is impact in Barcelona of our digital platform, which happened in 2020. Similarly, in Benelux where most of the stores are the same, you can see the benefits of the implementation in 2020 of our website. Moving to the next slide, I'll make a few comment on the full year trading. Group revenue is up 5.5% that follows 3 years of outperformance. Over the last 3 years, our revenue grew by 27.5% on a like-for-like basis. We've discussed U.K. and French inquiries. In Spain, we are making good progress. I'm very happy with how the new stores trade. They're on line with the business plan and same in Benelux. I'd like also to dispel a myth about Safestore, which is that we are a company that predominantly caters to business customers more than peer. That may have been true 10 years ago, but that is not true today. Our split at around 80% domestic and 20% business is similar to what our peers report and in occupied square feet, it is also similar. Our average unit size is around 60 square feet, which is again similar. Finally, on recent trading. In the 2 months since October year-end, we took some limited pricing actions that were efficient to reduce the year-on-year occupancy gap. The U.K. like-for-like occupancy improved from minus 3.8 percentage points year-on-year at Q4 to minus 1.4%. In France from minus 0.4% to plus 0.3%. The group revenue broadly was flat, like-for-like revenue down 0.6%. It could decline slightly further in the quarter, but the impact of pricing action will gradually dissipate. The actions we took were small. If you look at the U.K. achieved rate sequentially in the last few months, it was at end of October at the highest ever for the month of October. It decreased through November, December and was 1% lower in December compared to October. But in January, the achieved rate is already back to the last October level, up 1% from the prior month, which was December and we had a slightly higher volume. Our price list in January is broadly similar to last year and the gap between move-in rate and vacate rate is as usual. As you would expect after a certain period of reduced move-in activity, vacates are down significantly down 13% in square feet year-on-year. The total occupancy is around 4.3 million square feet. The total occupancy variance is 70,000 square feet so that's about 1.6%. Out of this 70,000 square feet; 50,000 square feet is business, domestic occupancy is close to flat. Business shortfall is almost entirely on the larger units above 250 square feet, I think that's 90% of this 50,000 square feet. That suits us as we continue our program to repurpose unit size in the U.K. stores towards typical domestic unit size. The U.K. occupancy CAGR over 8 years has been 5.3% for domestic versus 1.3% for business. In the last 5 years we converted 100,000 square feet of large units into small units and we'll continue that program. Trading in the other countries: positive in Paris and strongly up in Spain, Netherlands and Belgium. Moving on to REVPAF on Slide 11. Our operational business model remains to focus on REVPAF and whatever combination of rate and occupancy that delivers the highest REVPAF. We believe that the strategy that focus on constant high occupancy requires operational strategy that will systematically deliver lower REVPAF and quite spectacularly so. We break down in the Slide 11 our REVPAF in the various region and against our main listed peers. I won't name names, but one of our peer follow a similar revenue optimization strategy as we do and they achieve almost identical REVPAF. The other one that focus on occupancy achieves spectacularly inferior REVPAF. There is absolutely nothing new with these numbers. They have been similar for 15 years. But as I keep being asked why we don't pursue a 90% occupancy strategy, I thought that this slide would provide an adequate answer. I will probably skip the next 2 slides that are usual information, but don't require any more comments and go to Slide 15 to discuss the portfolio. The key points are: we added 0.5 million square feet of MLA in 2023 and on top of that, we have a pipeline of a further 1.5 million square feet, which again represents together with a non-like-for-like store more than 30% EPS growth over the next few years on top of the like-for-like portfolio further growth. This pipeline allows us to build scale in our new geography. We'll soon have 22 stores opened in the Benelux and 16 in Spain. That pipeline is also concentrated in the best market as shown in Slide 16. In fact 96% of our pipeline is London, Paris, Amsterdam, Barcelona, Madrid. When we look at our portfolio, we see that 65% of the value of our portfolio is in the capital cities, 75% if you include Southeast U.K; 89% is located in metropolitan areas of more than 1 million inhabitants. And when we add the cities about 0.5 million people, we are more than 95%. And that is probably a second miss about Safestore, which is that we are the regional operator and more so than listed peers. That is again simply not borne by facts. None of our main peers has a greater concentration of assets in capital cities and none has a greater concentration on key metropolitan areas. And our intention is to continue to focus our development efforts predominantly in the largest key metropolitan areas and our current pipeline clearly shows it. It is described in Page 17. It is almost entirely in capital cities, but I'll go straight to Page 18. We added 0.5 million square feet, as I said, in 2023. We'll add 800,000 square feet in 2024. We have planned again 0.5 million square feet in 2025 and for the moment at least 200,000 square feet in 2026 and beyond. But the pipeline is continuing to fill and it is in fact already bigger than the one disclosed here. We are very happy with that pipeline because of the quality of the location and the cost effectiveness of the development at GBP 160 a foot, which compares very favorably to competitors. And we believe that our track record in making high returns out of our investment will support further earning growth in the long term. Our confidence in our pipeline is simply based on our track record, which you can find on Page 19. We have a conservative cash-on-cash target above 10% on all investment. Cash-on-cash being net earnings after tax before funding cost so free cash flow before interest. Our average return is 15% ranging from 21% for our pre-2016 investment to double digits for all new stores and acquisition done until 2019, on track to double digit for the most recent store development. Our 2020 openings are already above 8% cash-on-cash. In Slide 20, we show how the cash-on-cash build over time with the investment bucket by year of store opening. This graph exclude acquisition, it is only about organic development. Again there is a clear and consistent trajectory of delivering and exceeding the conservative cash-on-cash targets that we underwrite. The combination of investing sensibly, not overspending on a square foot basis and achieving high REVPAF thanks to our operational model is what allow us to deliver this high cash-on-cash. We also normally invest in properties that can be quickly put to production, which allows to recycle cash flow as quick as possible and invest for further growth in a compounding way. That is our business model. A quick mention on Slide 21 about the reliability of our leasehold portfolio with U.K. lease regularly regeared providing remarkable stability on our unexpired lease lens. And in France, leases being simply as always automatically renewed by law. The combination of our operational strategy and the way we invest and fund our growth has delivered industry-leading EPS growth. We think that strategy will continue to structurally underpin higher EPS growth trajectory over the long term, particularly as there are still lots of growth opportunities and low supply as you can see in Page 23. Finally, on ESG. I think our investors have now understood very well that we have a very low carbon footprint already. We reiterate in the next few slides how we plan to go to neutrality. We've gone through these slides already so I'll focus on Page 28 where we show the progress. Well, we are on track. The number that really matter is the market-based emission intensity of carbon per square meter of MLA, which is a number that you will find below the chart. That is a number that matters together with the EPC ratings and it is what is on the green elements of our FCF. There is a requirement to report the location based as well, which is why we do it. The difference being essentially that market-based takes into account the green electricity that we purchase from wind farms whereas the location-based is on what the U.K. grid delivers on average. Our intensity, which starts from a very low level, it was 1.36 in 2020, very low in the real estate sector, has been already reduced by 1/3 to 0.91 in the last 3 years and we continue our journey towards neutrality. We are ahead of our own targets. With that, I'll let Andy go through the financials.

Thanks, Frederic. Good morning, everyone. I'll start with the financials on Page 33 of your slide presentation. So in the year, we achieved 2.2% like-for-like revenue growth. Once adjusted for constant exchange rates, that was 1.7% and that flowed nicely down to 2.8% growth in EBITDA as a result of strong cost control over the like-for-like cost base. We then overlay the non-like-for-like stores and that's anything we acquired or opened in 2022 or 2023 and the most significant one of those is clearly the Benelux acquisition, the 80% that we bought from our JV partners back in April 2022. And once those non-like-for-likes are overlaid, we delivered 4.8% revenue growth and 4.5% EBITDA. After leasehold costs, EBITDA grew by 4.2%. And not on this chart, but we'll come on to it later, but the finance cost did increase by around GBP 5 million so earnings per share grew by 0.8%. We'll come on to that a bit more later. Consistent with our dividend policy of growing dividend at least in line with earnings, the dividend for the year is up 1%. From a cash flow perspective, a oneoff working capital movement which I'll come on to in more detail and higher interest costs resulted in free cash flow declining to GBP 89 million. Moving it back on to the balance sheet. We saw a 9.3% growth in our investment property valuation, which we'll deal with in more detail shortly. And that band with around GBP 100 million higher debt resulted in a 1.8 percentage point increase in LTV to 25%, well within our internal targets. Interest cover was 6.7x and we'll deal with that a little bit later as well. I won't dwell too long on this chart. Frederic's dealt with some of the trading and the story really is pretty consistent across all of the geographies. In the year we've seen strong rate growth on a like-for-like basis, you can see the middle line there. We see 5% rate growth across the year and that's offset a pullback in the average occupancy, which is around 3.3%. Combined with solid ancillary revenue growth of 2.8%, we delivered that 1.7% like-for-like revenue growth overall; which is 1.2% in the U.K., 3.6% in Paris and flat in Spain. Again I won't go through all of this slide. This is the presentation of the P&L that drives our headline adjusted diluted EPRA earnings per share number. You can see at the top there's an overall GBP 10.2 million increase in revenue and that's driven a relatively flat EBITDA margins on a total basis at 63.4%. However, if you adjust for like-for-like, our EBITDA margins move forward by 70 basis points to 64.8%. As mentioned before, leasehold costs did increase by GBP 1.1 million. That's slightly more than normal. It's around 8% driven by some rent reviews in the U.K. and in Paris, but also the annualization of the lease costs on the Benelux business. We would anticipate that around 5% is the normal sort of level of leasehold cost increase over the medium term. As mentioned earlier, finance charges grew by GBP 5 million as a result of the slightly higher debt levels plus interest rates going up since the prior year particularly on SONIA and Euribor, which our RCF is impacted by. So as a result, we grew to GBP 47.9 earnings per share, up 0.8%. Just looking at some of the drivers of the EBITDA in a slightly different way. You can see the GBP 10.2 million of revenue growth in the first 2 green blocks; GBP 3.5 million of that was from the like-for-like stores, GBP 6.7 million from the non like-for-likes with by far the biggest chunk being the Benelux business. So partly driven by a couple of new stores, but also the fact we owned that business for 12 months this year [Audio Gap] in EBITDA after rent. Going back to the cash flow. We did see a oneoff working capital movement in the period and that combined with higher cash interest costs really drove that 12% reduction in the free cash flow. The working capital movement related to national insurance payments on the vesting of the oneoff 5-year 2017 LTIP, which was significant in its quantum and also was a 5-year scheme as opposed to a 3-year scheme. So we'd anticipate that the working capital movement relating to those sort of factors going forward reverts back to a more normalized level similar to 2022. So free cash flow GBP 89 million, down 12%. Investment wise, we invested around GBP 115 million this year outside of nonproperty CapEx or maintenance CapEx and that compared to about GBP 200 million last year. The biggest piece last year obviously being the Benelux acquisition. This year the GBP 21 million that's highlighted relates to the acquisition of the freeholds of 2 of our leasehold stores in Birmingham and Barcelona and the extensions that we're in the process of building out in Paris and London. GBP 91 million relates to the stores we've opened in the year and the pipeline that we'll open in the coming years. Dividends wise, the GBP 66 million relates to the 2022 final dividend and the '23 interim dividend. And you can see that we drew down just over GBP 100 million of debt in the period. Moving on to the balance sheet. The property valuation, as I mentioned earlier, grew by 9.3% largely driven by the improved rate performance. Exit cap rates were pretty much flat year-over-year. On a like-for-like basis, that valuation grew by 5.4% and you can see the geographical split of the growth there; 6.5% in the U.K., 8.1% in Paris, 43% in Spain driven by new stores and 27% in the Netherlands and Belgium. It's worth noting that the valuation implies a net initial yield on mature stores of 5.92%, which we think is very comfortably underpinned by external transactional evidence. Net debt grew by GBP 106 million and that combined with the valuation movement resulted in the LTV moving to 25.4%, up 1.8 percentage points compared to last year. Interest cover ratio did decline to 6.7x as a result of the higher interest charges, but that's well within our internal targets and our banking -- our lender targets as well. And you can see there the level of headroom we have within that metric. Effective interest rate after capitalized interest was just under 3% and our debt capacity at the end of the year was just under GBP 200 million. The weighted average debt maturity was 4.5 years. And just after the year-end in November, we did extend. We took advantage of 1 of the 2 options we have to extend our RCF by a further year. So in reality just after this point, that number will have gone up. It's worth mentioning that going forward we only have a EUR 150 million refinancing of a private placement note to do in 2024. Otherwise, our debt maturities spread between 2026 and 2033. And at the year-end, 73% of our drawn debt was at fixed rates. Moving on to this slide. We've shown this a few years now. I think it's just worth showing. This is the performance of the 112 store portfolio that we had in 2013 and we still have today. And the purpose of putting this in is really to show based on true factual evidence what is achieved over a sort of longer-term period of time or what could be expected to be achieved in stores. So you can see here over 10 years those 112 stores, the rate has grown 34% over that period of time. The square foot occupancy-wise has grown by 31%, which on a closing occupancy percentage of MLA basis is 17.6 percentage points from 63% to 81%. And that's driven 77% of revenue growth in that portfolio over that 10-year period. So just a bit of an illustration particularly given the pipeline that we have coming forward of what can be expected to be achieved over a period of time. I'll just touch briefly on this. You'll recall at the half year we started to talk. Given the pipeline that we had over the last 2 years has grown very significantly, we started to sort of try to give more disclosure around what we think the impact of that pipeline is going to be. And we just wanted to look at the earnings per share growth split like-for-like and non-like-for-like from the 2023 perspective as the sort of base for looking forward. So you can see in the bottom there, we've got 0.8% of total EPS growth in the year. We estimate that on a like-for-like basis, that was actually around 3.4%. And you can see that we had about 0.9p per share dilution from the non-like-for-likes in the pipeline. So that's really the basis for the next slide. And what this slide is trying to do is show -- you will have seen this for the first time in the half year results and it's changed slightly obviously as the pipeline moves and as interest rates have moved a little bit. But what this really shows is the extremely strong accretion that this pipeline will deliver for us over the medium term. Frederic referenced the number as well. But if you look at 2028, we think this pipeline will give us 15% of growth on our 2022 earnings per share number and go 5 years further out, 32%. So that's the pipeline and the non-like-for-like stores. It does include the Benelux business in there because we're looking at this from a 2023 definition of like-for-like. It is worth noting as well, you can see on the left hand side the 0.9p of dilution we saw in 2023. We will see more dilution in 2024 as that pipeline is built out and we expect that to be at this point in time around 3.8p, which will mean that earnings for 2024, as we've previously communicated, will probably go down slightly depending on what like-for-like growth we deliver on the revenue line. Finally, on to the usual guidance slide. I won't go through all of this. I'll just focus on a couple of lines. On the interest charges, we're expecting around GBP 28 million to GBP 30 million of gross interest, but we expect to be able to capitalize between GBP 5 million and GBP 7 million of that. So overall, I expect net interest to be around GBP 23 million to GBP 25 million in the 2024 financial year. Post capitalized interest, we expect that to be around 3.1% effective interest rate. Clearly despite the news today, if we do get any interest rate reductions over the coming months, a 25 basis point interest reduction would be worth about GBP 600,000 on the gross interest line. You then got to take into account what's already been capitalized as well and that's on an annualized basis. Finally, on the CapEx side, the total CapEx related to the pipeline was GBP 232 million. We've spent GBP 104 million of that with GBP 128 million therefore still to be spent and we show the phasing of that; just below GBP 65 million in 2024, GBP 26 million in '25 and GBP 37 million beyond. We'll have maintenance CapEx of around GBP 8 million per year. And over and above all of that, anything we do in terms of buying out leaseholds or such things would be on top of that. As Frederic mentioned, the pipeline results in us having a total spend per square foot of GBP 160, which does compare very favorably to what we think has been disclosed by our listed competitors. And bear in mind that of the 30 stores in the pipeline, only 2 are leasehold stores and all but 2 I think are in capital cities. So that concludes the presentation. I think we're going to go to questions in the room first and after that, we'll go to online questions.

Aaron Guy from Citi. Frederic, just a quick question. You've had 25 years plus experience in the business. Referring back to Slide 3 where you looked at the revenue long term, acknowledge the flattening of the income this year. But if you step back a little bit and look out 2 to 5 years, would you expect a significant deviation in that trajectory going forward? And I refer to some of the discounting that you've been doing. Should that be just considered a short-term sort of yield management phenomenon or something a little bit more concerning?

No. It's absolutely a short term and very limited yield management and I would expect the rate to gradually pick up. In terms of what next 5 years are, I mean if you can tell me your forecast on GDP growth and general activity in Europe and in the U.K., then probably it would be easier for me to answer your question. There is no reason other than anything exceptional happening for the trajectory to change because the fundamentals are still there. There are a small amount of supply, strong barriers to entry if you look at where our stores are located and where our next stores will be located because we won't deviate from that investment policy. So I mean we start from a very low base in terms of intensity of use of the product and the more supply there is, the more people use it. The more people come across storage, the more there will be a propensity to use it when there is a storage need. So I fundamentally believe that there is still a very compelling investment story and I intend to continue to invest in it even personally. And so no, I don't think. What the rate will be? Of course it's absolutely impossible to forecast and we will be as always exposed to short-term volatility, which drives short-term volume of inquiries. But I think the structural story is there and given the strength and the importance of the barriers to entry, I think that story is going to be still there for many, many years.

Just switching to the property valuation up 9.3%, most other property sectors are sort of down last year. Can you just give us a bit more color on the buyers, the sellers in the market? What freed up and created the extra transaction evidence last year? And whether, with interest rates possibly having turned the corner, should we be more or less confident about the growth rate in 2024?

In terms of transactions?

Property valuation.

I think proper valuation, I mean I don't see clearly cap rate at the moment and I think the talk of the day is more at some point a reduction of interest. So I don't see. I think that the demand on this asset class of investment particularly from all stores of private equity players is enormous and there is clearly much more demand and much more capital willing to be deployed in that asset class and there is anything for sale even if the close to EUR 1 billion of transaction last year is a record year I think. And from what we can see is a cap rate at which transactions took place are lower than our valuation. And you have to consider where these assets are located because there were not a single transaction in London, Paris, Amsterdam, Madrid, Barcelona. They were all into a combination -- I mean in locations which were much more regional. So I think that is a very strong underpin on our cap rate and I can't see and at least as of today, I don't see the amount of interest for investment in this asset class reducing. So I think that is a very strong underpin on our valuation. Why did we see more transactions? I think probably there is a generational element to it, quite a few people started maybe 25 years ago and some are probably selling and just a normal transition. And as there is more stuff around available, it would be normal to expect, but it's just a speculation to see more transactions happening over the next few years as the asset class grows. But it only happens slowly because it's bloody hard to find good stores in good locations at sensible price.

John Cahill from Stifel. Your occupancy has come down not much, but a little bit over the last year. But your yield management systems have meant that your overall revenues are up because rate growth has been so strong. Is that implying that actually the proportion of your tenant base that's priced inelastic is actually probably quite a lot higher than maybe you'd first expected? Can we imply that? It almost looks like there's quite a lot of customers where irrespective of what you do with the price, they seem to stay. Even those ones that are there for the life events, et cetera.

Well, I think that our customers are not stupid and so when they accept to pay the price that we charge, it's because they find real value and they understand all the quality of the service that we provide. I don't think that we have any pricing issue. When I look at the rate at which customers move in versus the rate at which they vacate, the gap has not changed. All the structure is very stable. I think that this yield management and this rate action that we took, some of which are front-loaded, are really on the margins. And I don't think that structurally there is anything that has changed in the way we price or in the way we sell and in the way we manage rate versus occupancy. And as you can see, that delivers the highest REVPAF that you get in the industry, but there is no change of paradigm.

Markus from Bank of America. Can you go a bit into occupancy trends in the first months? You talked about pricing, but how is occupancy in January comparing to October and for year-on-year also?

Well, I think we have that if you look on the outlook as of end of December. So apologies, I don't have the number as of this morning, but just as 2 weeks ago. And you saw that the year-on-year occupancy in the U.K. was up -- sorry, the difference was reduced. So obviously we are on a declining season so you are -- I mean it's always a weak season so you are comparing 2 curves, which always decline every year. But clearly we were much closer to last year as of end of December in the U.K. than we were at the end of Q4 and we were up in Paris versus last year as [Audio Gap] would you say 0.04?

Yes, I think October 0.3% down in Paris and we were 0.4% up in the end of December. In the U.K., we were 3.8 percentage points down in October and 1.4% down in December. So it's a slightly improving trend.

And because the pricing started to get negative in the U.K. with closing occupancy so not like-for-like below 80% more or less if I understand well. Is this a trigger where you have to go into big discounts or...?

No, there's no such thing. I mean we really don't have a sort of a general approach to it for us. Each store is a different business and even within the store, the 25 square feet units and the 200 square feet are different markets as well. So we really look at it very granularly and we try. Our goal is to get the highest possible REVPAF because in the end that's what we want. I don't pay dividends with certificates of occupancy. I pay them with cash and that's what I'm trying to get. And as you can see, the results are there because we do have the highest REVPAF even having a slightly lower occupancy. So we will continue to do that. This is what we have always been doing. But there's no such thing as a magical number at which you can increase the rates because it's a business with churn. So if you need a discount to fill our stores, it's not because your store is full that you're going to remove your discount because then the people will start churning and you will need the discount. So I think it's based on something else. And so we don't look at -- I mean so there is no such threshold in our approach to revenue management.

Just on the transaction you have seen last year, what cap rate on average we have original transaction and was it mainly U.K. or was it across Europe?

It varied. I mean we don't see -- the statistic we quoted was Fedessa's, the European Self-Storage Association's, estimate of transactions for the whole industry. So you don't get necessary disclosure on every single transaction that happens. So there has been quite a variety. I think there was a business in Ireland went for something that we think was between in the mid 4s from memory.