Accor SA

PAR:AC

Earnings Call Transcript 2021-Q3

[Operator]

Hello, and welcome to the Accor 2021 Q3 analyst call. My name is Josh, and I will be your coordinator for today's event. Please note that this conference is being recorded [Operator Instructions] I now hand you over to your host, Jean-Jacques Morin, to begin today's conference. Thank you.

[Jean-Jacques Morin]

Thank you. Good evening, ladies and gentlemen. Happy to be with you today for this Q3 '21 revenue presentation. As we did back in April and July this year and for the sake of clarity, we will continue to provide the RevPAR variation versus 2019 for the rest of the year. And this is to ease the understanding of performance and notably because, as you all know, of base effects. As for revenue figures, we will provide both variation versus 2020 and 2019 in the document when it's relevant.So without further ado, let's catch out the accelerating momentum of the business by going directly to Slide 3. So in July, we mentioned that we had seen a RevPAR improvement of about 5 points every month since April over summer and confirmed in September and in October, we've seen an accelerating momentum.So what's going on? Globally, there is a relaxation of the lockdowns. And you've got some illustration in the last part of the slide in front of you for Australia, the U.K., Singapore, and more importantly to come is the U.S., which will open up in November to foreigners, and that will typically set the tone for the rest of the world.So when you look at the trends for our hospitality business, which is the right part of the table, we had a strong leisure guest recovery in July and August that, by the way, extended past August and gave us over the period some pricing power. We also saw confirmation that the business traveler on MICE are back in October and also -- in September, sorry, and also in October. I mean, 2 consequences that you are all aware of, but I'll quote them is the decline of governmental subsidies as business is recovery, the subsidies are going down and some labor shortage, which are notably stronger in the U.S. and in the U.K.Moving to the next slide, Slide 4. You've got here the classical Q3 highlights. So on the left part, what you see here is that the Q3 2021 reflect the trading improvement that I was describing. After 3 quarter of Accor hovering around minus 60% of RevPAR, we now have an inflection point and RevPAR is only down, if I can say it this way, 37% in Q3 2021 versus Q3 2019.The second point is the last 12 months net unit growth on the network continued to improve, and we are now, for the last 12 months, at 2.5% and we plan to finish the year around 3%. And all of that translated into a group revenue of EUR 589 million, i.e., a decrease of 40% versus Q3 2019 or an increase of 79% versus Q3 2020 on a like-for-like basis.Now in all of that, we did sustain our strong operational control on the business and the costs. So we had mentioned back in the call in -- of the H1 results that we had a few moving parts. And we are now happy to upgrade for 2021 both the EBITDA sensitivity to RevPAR, but also the monthly cash burn guidance that we have provided. To be precise, the EBITDA sensitivity is now below EUR 17 million per point of RevPAR versus the slightly below EUR 18 million that we had disclosed at the beginning of the year. And the average monthly cash burn is now below EUR 35 million when we had talked before of EUR 40 million. In all of that, the RESET savings are per plan, on target, and we will have, as discussed before, more than EUR 70 million of EBITDA in the P&L of financial year 2021.If I move to the next page, which covers where we stand in terms of network at the end of September. As for gross opening, we opened 10,000 rooms in Q3, which is, in fact, a very good quarterly number versus our history. In Europe, we did maintain a consistent level of opening as position was no longer under lockdown and asset owners who are basically accelerating openings to catch the business. Huazhu was around 2,300 rooms in Q3, which is a level which is quite consistent with the 2019 level.Maybe just for registration, a few key openings. There was very nice Raffles in Dubai, a Mama Shelter in Roma, our first Novotel in Bolivia. And last but not least, a very nice property, the Century Plaza in Los Angeles that will just open up in the month of October.Overall, conversions did ramp up, and they now make up about 45% of the year-to-date openings as a result of lifted COVID restrictions since May and June for Europe, but also some conversion in Australia. As you well know, this level is above the industry standards.As for churn, it remained tightly monitored. It's always a point of focus when there is a crisis. And we have been sticking to a level which is around the historical level of 2% plus. So overall, when you add up the openings and the churn, our net unit growth was the 2.5% over the last 12 months that I mentioned.And the rationale that we've been applying is that with the current environment in Southeast Asia, which is very much affected by international travel, as you will see later on in the presentation, we plan to finish at around 3%. Just as a final comment on this slide, our pipeline is stable versus March of 2021 at 211,000 rooms -- 211,000 rooms, sorry.If you move to the next page, where we talk about the RevPAR by geography this time. You can see that overall, the group reported a 20 point sequential improvement from Q2 to Q3 2021. The strong demand translated in pricing power, and we have had prices over Q3 that were higher than the prices in like-for-like of Q3 2019 in the France and the British provinces, in the UAE and in the U.S. So when the demand came back, it came back very strongly.South Europe was, in the end, at minus 25% versus Q3 '19, as you can see from the graph, and that's close to a 40% sequential improvement. The French provinces were very close to the level of 2019, 6% below, but Paris was, in fact, the portion that was subdued as it had been since the beginning of the year because the main cities, the main capitals are much more affected by the international restriction that people have on travel.The good news was that in September, MICE took over from leisure, and that was a good confirmation that the business was coming back. When you talk of North Europe, you've got comparable patterns. The RevPAR improved sequentially by about 35 points. The U.K. was very similar to France, in fact, and the Provence in the U.K. was close to 2019 level with London that was lagging just like Paris was lagging with about the same numbers. I mean, Provence about 5%, 6% below 2019, the cities like London or Paris about 50% below the level of Q3 2019.Germany, which as you all know is more business than leisure will see a recovery which is more backend loaded, but the sequential improvement was also extremely impressive over Q3.Asia Pac is kind of the exception, and we had started the discussion around that back in July with the deterioration of the situation in Australia. What you see in Asia-Pacific is a sequential slowdown of 20 points to reach a level of minus 57% versus Q3 2019. Pacific itself was at minus 56%, and it was really impacted by the large cities lockdown, Sydney, Melbourne and so on.Good news is that since then, Sydney has been releasing its lockdown. And in fact, it's substantiated or rationalized by a very high level of vaccination. The progress made by Australia have been stunning. They are now close to 60% double vaccinated, i.e., around what you see in Europe or in the United States. So they've been very, very much compensating because they were nowhere at the beginning of fiscal year 2021 in terms of vaccination.Greater China saw some COVID outbreaks that was back in August. And so in 1 month, they lost 50 points of RevPAR. And they ended up the year -- the quarter, sorry, at minus 34%. Southeast Asia was flat with a RevPAR at minus 72%, what I alluded to a little bit before. The vaccination is gaining traction in the region, but it's still behind. And so you will see a delayed recovery of Southeast Asia.When you move to IMEAT, which is Middle East, Africa and India and Turkey, you see here a sequential improvement of 20 points with RevPAR at the end at minus 23%. The UAE was very strong, has been very strong for months and is now back to the 2019 level. The virus position which just started at the beginning of October will fuel this healthy trend for the coming months. The person that was less rosy was Saudi because there was still a restriction on pilgrimage. And so the kingdom basically remained closed for countries like Egypt, Indonesia, Pakistan and that affected the numbers for Saudi.In the Americas, 25% sequential improvement, and it's basically in South America, a vaccination that has been swiftly increasing. I mean, as an example, Brazil was around 13% double vaccinated at the beginning of July and is now 50%. So obviously that helps. And same thing -- and on Canada, sorry, what helped here is that the U.S. did reopen with Canada on August 9. And so Canada did benefited from that. And again, as I said earlier, the U.S. will reopen to other countries like Europe notably and some countries in Asia, starting November, and that should help.If you move now to analysis by segment and not by geography, and I am on Page 7. You see that the Accor revenue reached EUR 589 million in Q3 2021, which is 79% up versus last year and is 40% down versus the same quarter 2 years ago, 2019, and this is very much in line with the RevPAR variation I have commented before of 37%.For HotelServices itself, the growth is 94% versus 2020, minus 42% versus 2019. Again, here you see a decrease, which is a little bit sharper than the RevPAR as we would typically have. M&F was 45% down, and I'll detail that by geography on the next page. And Service to Owner was down 40%. Again, here it's the mix phenomenon of having in the U.S. [indiscernible] cost and with U.S., which has RevPAR decrease which is way better than the one of the group average. So that's a mix phenomenon, if you will.As for Hotel Assets, the revenue growth was a 57% increase versus 2020 and a decrease of 38% versus 2019. You may recall in H1 Australia was really strong. And with the lockdowns, we saw a strong decrease from Q2 to Q3. But what was good in the end is that Egypt and Turkey did to cover and notably because Russia tourists were able to travel to those 2 places. And they really fueled the properties that we've got in Egypt and that we've got in Turkey.And so that's why the variation is the one that you see. That was an offset. As a reminder, this segment includes new businesses and here, nothing new, the digital services are behaving better than the travel-related activity and the travel-related activities are very much behaving like the rest of the travel activities of Accor.So that's on the analysis by segment. Just drilling down, going a little bit deeper on M&F. And you have on Page 8, the M&F management and franchise revenue by geography. You see that versus Q3 '20, M&F revenue doubled by region. The variation that you see here reflect the variation that we have been discussing through each of the geography.The important point here is that now we have occupancy which is above the 40% threshold. And so you see recovery of incentive continuing to improve. We have already seen that back in Q2, but it further improved in Q3. And again, to basically answer the question that I would have for sure, you may recall that M&F revenue portion made of incentive was about 1/3. So 1/3 of the M&F revenue is incentive in a normalized year.In last year, we were at 8%. And this year, we are at 20%. So we are not exactly where we used to be and no surprise, but we are moving along the line of recovery also on the incentive, which is, again, no surprise.If you look at Q3 '19, the same analysis, you've got RevPAR which was 37% down, annual revenue which is decreasing by 45%. The delta between the 2 is, as usual, the incentive in the management contract.So just to conclude the discussion and moving to the last slide, you've got the priorities going forward. The accelerating momentum of your Q3 and the preliminary data for October confirmed recovery continuation over Q4. And now we are focusing the company on the rebound of 2022. The finalization of RESET is, I've been saying that for one year, but I'll say it again, is totally in line with the plan that we [indiscernible] ourselves back in July 2020, and it remains obviously an important element of the recovery plan.Balance sheet, liquidity, obviously we've been looking at that, we continue to look at that, and we've got a strong position here and we'll continue to optimize whatever may be optimized. And as part of the recovery, we'll capitalize on the pent-up demand, which is local and notably through what we had launched before the crisis, but that has been accelerated by the crisis like the lifestyle and the augmented hospitality strategy.Thank you very much for your attention and now the floor is yours.

[Operator]

[Operator Instructions] Our first question comes from the name of Jamie Rollo from Morgan Stanley.

[Jamie David William Rollo, Managing Director]

Three questions, please. Jean, you normally give or you have been getting recently the monthly RevPAR data. It will be quite helpful to us to get the exit rate in what was a pretty unusual quarter and maybe discuss October? And also the statement I think says you're expecting further sequential improvement. So does that mean Q4 should be better than the minus 37%?Secondly, is there any flavor you can give us on corporate negotiations, particularly on pricing? Clearly, pricing has been better than expected with leisure, but is that continuing? And then finally, on the pipeline and the signings, it's hard to see exactly what the signings were without knowing the conversion brands and the attrition rate. But just a general comment on forward net unit growth for next year, should that sort of start to pick up a bit from here, do you think?

[Jean-Jacques Morin]

Yes. I'll take the last question first because it's an easy one. I mean, you've seen that when we look at that indicator, which is the last 12 months trailing net unit growth, if you look at the number at the end of last year, at the end of Q1, at the end of H1, you will see a progression on the number of the last 12 months trailing net unit growth, which I just mentioned that you will see a further improvement of the number at the end of the current year.So the 3% will be an improvement obviously from the 2.5%. So you should see that trend continuing because we are in a recovery mode versus the point in time where people were basically not opening hotels, whether they had no business for opening the hotel or the hotel could not be finished because people were not able to work or people had financing difficulties in order to progress on some of this. And it's all the things that you know so well, which are linked to the crisis are gradually disappearing. And so that's why the trend that you've seen in our numbers will continue either and inertia, if you will, of improvement.In terms of corporate negotiation, we are in the midst of it. In 2021, there was not much. There was basically a rollover. And what people are now doing is that they are willing to travel again. And so we are in a better position to get the discussion on stronger rates. And obviously, the clients are trying against those rates to secure volumes. And so the discussion and the typical game between rates and volume is the one which is occurring today. But as there is a demand, I think we have a good positioning to negotiate.As for the RevPAR, in fact, I never gave the monthly RevPAR except last time because I'm trying not to enter into details that in the end will not really help. And I gave the monthly RevPAR last time because there was clearly an acceleration from a situation that had been pretty gloomy, I would say, happening in the month of May and continuing in the month of June. And so that was to give you comfort that we would see improving numbers in Q3, which was not a natural thing when you were to look at the overall Q2 figure, which was in the end very close to the number of Q1, which was very close to the number of Q4.So I don't want to enter into too much detail. But to answer your question and give you an idea, yes, we should see further improvement. I think I've said that between the lines in the comments I made. And so you should see further improvement of Q4 versus Q3.

[Operator]

Our next question comes from the line of Bilal Aziz from UBS.

[Bilal Aziz, Associate Director and Equity Research Analyst]

Two for me, please. I think you already talked about the mechanics of the incentive theme. I guess your RevPAR like-for-like difference was broadly consistent with the first half. Just to clarify, did you suggest that will now narrow into the fourth quarter and going into every quarter into next year as obviously, the recovery continues? That's the first question. And secondly, you suggested 45% of openings were conversions year-to-date. And can you have 3Q numbers roughly, please?

[Jean-Jacques Morin]

Yes. The incentive will continue, and that's what I was suggesting when I gave you the various figures of the Q3 of 2019, the Q3 of 2020 and the Q3 of 2021. And why are they going to continue? Because as you know, there are some step functions in the computation. So it's not exactly a linear relationship to RevPAR. But as RevPAR is improving and in fact, as occupation rate is improving, what you see, you see in fact a profitability at the property level which is improving.And so because the profitability level is improving, what you see is you see our incentives kicking in, if you will. And so if you go with what I just answered to Jamie, which is that RevPAR will continue to be better, then our profit will continue to be better at property level. And as our profit will continue to be better at a property level, then you will have an increased level of incentive. So that's I think the -- the answer, sorry, on the incentives.As for the conversion, the 45% is a number which is a year-to-date number. If you look at what we did in history, right, the 45% was probably more in average something like 40% in 2018 and 40% in 2020 and a little bit more in 2019, where we had a lot of conversion. So what I'd say is that the number has been improving in terms of conversion because some deals were cut over the quarter, which helped in fact the year-to-date number to improve versus what it had been up to the end of H1 because you may recall that I had told you that we were not seeing significant amount of additional conversion versus a historical trend.So Q3 was very good on conversion. And again, the number will fluctuate, right, i.e., depending on the deal you cut, we cut a deal in Australia that is helping the number of the quarter. But depending on the deal that you cut, you will get fluctuation the conversion. I think the part which is important here is that the way to recover our net unit growth, that is, to some extent natural and helps the fact that new built properties who are not going at the same speed during crisis as what they used to be going is through conversion. So I think as we've been saying for one year, you should see a good level of conversion going forward. That kind of makes sense, if you will, from a physical perspective, and we saw that in the Q3 number and in the year-to-date figures.

[Operator]

Our next question comes from the line of Vicki Stern from Barclays.

[Victoria Jane Lee Stern, MD & Equity Analyst]

Just firstly on -- circling back on the signings point. I think you obviously referenced again tricky in Southeast Asia, just given the sort of border closures. But just a bit more color, please, across the different regions in terms of anything getting better in the other regions as things are starting to open up and just your sort of expectation on that signings pace as we look out into next year?And sort of circling back on the management incentive fee point, and thinking about it in the context of drop-through, obviously, helpful that you've kind of given us the color now for drop-through for this year being a bit better than the EUR 17 million. For next year, obviously, there's a massive element of the cost savings increase coming into that better drop-through. But just stripping that aside for a second, could you help us understand a bit better now as we get more of the management incentive fees back, how we should think about that drop-through improving alongside that RevPAR increase?And then just finally, to sort of try and push you a little bit more on the Q4 RevPAR piece. I think the consensus out there implies something like a jump to kind of minus 18% RevPAR versus 19% in Q4 from the 37% or so you've just done. Obviously, you've referenced that you do expect further sequential improvement, but just sort of to that extent? Or any sort of comments on either side of that number.

[Jean-Jacques Morin]

Yes. The RevPAR is as usual a tricky question, right? There is a good reason for why nobody is giving RevPAR guidances because I think everybody got bound in history. So I will not really answer precisely your question. I've just said that it's going to be better, which is already some kind of an indication, but I won't give you the velocity. I mean, Vicki, look, in the months from July to August -- or from August to September, sorry -- no, from July to August let's take this rate goes the opposite way anyway, I mean, China lost 60 points of RevPAR.Between June and August, we probably got 40 points of RevPAR coming down in Australia. Was I predicting that? The answer is absolutely no. So that's why I'm a little bit cautious this year again. But with time, you get more and more comfort that when -- even if we -- we are getting from time to time backwards, in the end, we do 2 steps forward and one step backward. So we are moving in the right direction there.And even if you look at the consensus that you're referring to, I mean, in the consensus, I mean, it's an average between people being at 55% drop year-on-year and people that are at 35% drop year-on-year. I mean I think these are the 2 extremes from the top of my head that you've got in the current consensus or whatever it's worth. So when I do a mathematical average of that, I'm not quite sure that it is fully precise. So with all of that, just take that it's going to get better, and I'll stop here.When you talk about the net unit growth, Southeast Asia was a provider of strong growth in history. 50% of what we used to do, and I'm sure you recall that number in term of openings, was Asia. And Southeast Asia was probably 15% to 20% of the growth of the group. So today, Southeast Asia is basically altered. I hope this situation will change. Thailand, for example, is talking of opening its border in November. Thailand is a very important region in that geography. You've got Singapore, which has been extremely cautious on any opening just like they're always very strict on everything. But at one point in time, they will reopen to go and do business. So I'm confident that this situation is not forever. But today, we're really suffering from that because that vector of growth is not there.As for the other geography, I think we have got good results in fact in Europe. And we've got good result because we've been able to save a few opportunities, notably in the U.K. but also in Germany and some in South Europe. So the numbers, in fact, in Europe in general are, in fact, better than what we used to have in 2019. So that's an area into which we've been able to grasp a few good things.Pacific because of the situation that is not easy, so Pacific, I mean, with the variance one way or the other has not been doing good numbers this year. And Middle East and Africa, which was the second vector of growth for the sector for many, many years, is, in fact, doing good. I mean, they are not doing as good as you could see them do in history, but they are way above the other regions in terms of absolute net unit growth.And that shouldn't change because I was describing to you the situation in Dubai. I think Saudi is talking about revising what they've been doing for now many months in terms of the opening of the country. So trying to give you some color on the fact that this situation and the dynamic will continue to improve on net unit growth.As far as drop-through, I mean, you will continue to see some improvement on the sensitivity that we provide. I mean, the cash burn will turn into no cash burn at some point in time. And same thing with the RevPAR sensitivity, it will continue to improve next year. It is really a function of the business volume that you can forecast. So the exercise of forecasting the level of RevPAR for next year will be very critical at what is the number. On top of what you described, i.e., the fact that you benefit from RESET being at full speed at the end of next year.Last but not least, I did mention that as part of my first slide, we used to have significant subsidies, right? We did benefit as all of us of subsequent subsidy. So that plays the other way. Again, I had quoted that number before. So I'll quote it again. We used to have around EUR 100 million of subsidies in 2020 overall when I add up everything, the following the things that Germany did as one-off. And so that's a significant number.We are to the point today where there is, so to speak, not much more subsidies in Q3, but we did benefit of subsidy in H1. So that will go in the other direction in the computation. And I was talking about those moving parts when I was -- when you guys were pushing me, I think Jamie was asking me very stringent questions in the H1 result, trying to understand why I was not willing to give more.And he was rightfully asking the question, and I was rightfully not giving him an answer because of all these things that you don't fully control. I mean, Australia being one, the level of subsidy being another one, the capability to see the business coming back in September being another one. I never had any doubt on leisure over summer because I knew it, I could see it in the July [ behaviorals ], and I am super happy that we've been able to see in September, in October some MICE, in September and October some business travel. And that's why I am happy to release more to you in terms of guidance and directions. That's what I would say.

[Operator]

Our next question comes from the line of Richard Clarke from Bernstein.

[Richard J. Clarke, Research Analyst]

Three questions, if I may. When you talked about the normal gap between HotelServices, like-for-like and RevPAR, it looks like that's expanded a bit. It's about 2% at the half year. It's 5% at Q3. That's not down to incentive fees because you said those are getting better. So I presume that's down to services to owners, which you've not really broken out. So is that not really keeping pace with the RevPAR recovery and anything to call out there?The second question is, you talked about kind of expecting a RevPAR improvement and you've upgraded that guidance on drop-through. Is that guidance on drop-through kind of based on your assumption that the RevPAR will get better? And maybe particularly in Australia, are you assuming that things get better because we know that sort of feeds into that? Or is it -- you're going to hit that kind of come what may?And then the third question, so Huazhu reported earlier today I believe, and they've got 125 Accor brand hotels in their pipeline. At the end of 2019, that was 214 Accor branded hotels in their pipeline. So that's almost half over just less than 2 years. Are they beginning to wean themselves off Accor hotels? And does that create you any opportunity? We've Hilton today talking about launching Hilton Garden Inn themselves directly in China. So any opportunities you can sort of move away from that agreement?

[Jean-Jacques Morin]

Yes. I'll take the various questions in order. There is really nothing to read in STO, Service to Owners. So there is nothing that I can think of that would explain things except some timings of billings and some one-off phenomenon, but there is nothing really to derive from the quick computation that you did. And you should take it offline with Pierre-Loup, but I can't think of anything.On the drop-through, you're absolutely right. And I said it myself. I mean, Australia was one of the reasons for why I was cautious in the way we talked to you about improving the drop-through. In reality, and I kind of alluded to that, I think Australia will strongly recover between now and December. That's what we get from our operations. The first -- the rule that Australia has been dictating is that as soon as they reach a given level of vaccination is when the borders reopen.When I talk of borders, I talk of borders between states. And as you know, we're entering into the summer season in Australia. And so the capability from people from Melbourne, Sydney, Perth to go and get a vacation on the Golden Coast, for example, where we got all these [indiscernible] properties plays a role into our numbers. So what we've assumed is that October would be still difficult. November would improve, and December would be fundamentally better than the 2 previous months assuming that people would reach a level of vaccination between 70% to 80%, somewhere at the end of October or mid-November. That's kind of the, I would say, business assumptions that we took on Australia. So that does have an effect.And on Huazhu, yes, I have not seen the detail of that announcement as I was working on mine. But what you described to me is something that I am aware of, i.e., that Huazhu has been launching, pushing some of their other brands. Does that mean that we're going to quit anything with Huazhu? The answer is absolutely not. I mean it's relationship that works well. I mean, and they have been basically making big strides for us in China at getting our hotels developed, which we were never able to do.You may recall, we did more in 3, 4 years than we've done in 40 years in term of openings. So I think it's probably also a sign of some of the difficulties that may have risen in China in term of signing some properties. And you're right also on the fact that Hilton is trying to find a way to also enter into those territories. So China is a huge market. There is a lot of potential in China. And so you've got to figure out what is the right partner to go business with. And today with Huazhu, we've got a very good entry point and a very good partner over the last 5 years. So there is nothing really that changes at this juncture.

[Operator]

[Operator Instructions] Our next question comes from the line of Andre Juillard from Deutsche Bank.

[Andre Juillard, Research Analyst]

It was more a qualitative question. When you look at what is going on in Q4 in October and your pipe, can you give us some more color about the mix of clientele you are seeing? And my question, there is -- do you see a real improvement in the seminar in the MICE segment and in eventually new clientele coming in your hotels? Do you see a real tracking from people working from your hotel on this kind of thing?Second question was about M&A and the [ Ledo ] discussion. Could you give us a little bit more color about that? And do you see some opportunities coming on the market also which could make sense for you?

[Jean-Jacques Morin]

Yes. On the qualitative comments, I'll be quantitative, then that helps. Yes, we are seeing more MICE, and I said it in the comments. But I think if you want to just use a rule of thumb, we probably are today at MICE level, which is to the tune of 40% to 50% of what we used to be doing in 2019 on a full run rate. So it's, in fact, a different nature of MICE, Andre. It's smaller meetings. It's meetings which are held in different locations versus where they used to be held before.They used to be held in very large convention and in central, big, large cities. But we've got a lot of those MICE events which are now also hybrid, but the bottom line of it is that the numbers have been good. And by the way, there has not been a lot of cancellation in Q4. There used to be more in Q3. And like in 2022, things like Salon de [indiscernible], Bauma, all the big, big, large events that makes up those convention revenue are confirmed, or as far as I know, are being held and there were series that also were held, notably in Germany in fact in the month of October. So again, I am cautious here. But if you want to get a sense on the quantitative element of why do I think that business is coming back, that's to me, a good one.Then your question on Ledo when we are discussing there is nothing here which is done. It's just one of the discussion. Why do we talk to them because we see that as a potential way to continue at minimum investment this augmented hospitality strategy. And so that's why we are entering into discussions with the Ledo.Are there more than that? There are always things that we study. None of them is of any significance because I've told you that we are hyper-focused on the rebound, the marketing effort, the capability to make sure that we deliver the new way of working optimization and benefits. And so that's really what we've been focusing on. And we start to get some win in ourselves, as you would say when you are a seller. And so we just want to make sure that we keep the momentum and not defocus everybody, and it has not been an easy task to ride over the last months. So I think everybody is really willing to get finally that 2022 rebound. So that's really where we are.

[Operator]

Our next question comes from the line of Alex Brignall from Redburn.

[Alex Brignall, Research Analyst]

Just 2 questions, please, on signings and growth rates. So on the conversions, obviously as the denominator has fallen as conversions have gone up, but where are they on the sort of an absolute basis against pre-crisis levels? And then just in terms of that unit growth, I'm not sure it's quite the best way to ask this question, but if we look at say 2022, when would those hotels have been signed? And I guess, really, what I'm trying to get to is if we are seeing lower signings as we are seeing at the moment, when would that actually manifest in hotel room growth or system size growth? I suspect the hotels being opened next year will started sometime before COVID. So just trying to get an idea of that.

[Jean-Jacques Morin]

Yes. If you talk about conversion, just to give you absolute numbers, in year-to-date, what we are discussing is 10,000 rooms, right, which compares in history in 2019 for the full year to 17,000. And for the full year of 2018 to 11,000. So in 2021, we've done year-to-date as much as what we've done for the full year of 2018, but less than 2019. I told you that 2019 was very good. And obviously, way more than whatever was 2020, which was for the full year less than 10,000. So again, trying to be very specific on your question, these are the numbers. That's on conversion and you had -- yes. You had another question, sorry.

[Alex Brignall, Research Analyst]

On -- just on the signings rate, and when those signings actually manifest into room being opened, I guess it relates to your room night guidance for next year, but how that then develops in the outer years?

[Jean-Jacques Morin]

Yes. Again, I'll make you the lawyer's answer, it depends. That being said, when I say it depends and you know that, it depends on what signing you are discussing. If you're talking about a luxury hotel, then it takes at a minimum, I'd say, 3 years. That's what it takes between the time that you sign and the time that you get something in place. And by the way, just take the example of the Century Plaza that we were discussing today.I mean, it took more than 3 years between the time of signing to the time that the hotel really opened. So because of the complexity and the nature of what you go do. On conversion, it can be super quick. So the conversion is a question of months, weeks, in some cases, because you've got to just do whatever it takes to put the standard at the right level in the hotel. So depending on the brand that you use, if you use some of the conversion brand that we've got, like MGallery or Mercure or Grape, these kind of brands are -- you can go really fast in the conversion process, which obviously is not the case if you go to something which is way more normalized like a Novotel, for example.So again, here, it can go from a couple of weeks, some weeks, to months if you are more in the midscale and economy, and I'm talking conversion here. If you talk about the new builds, even if it is and it can on economy, you talk more than 1 year for sure. And so trying to make it simple for you. When we do the modeling, we use something between 2 to 3 years for the time for which, from the time that you sign, in average, you get the fees kicking in.And you are really at full run rate first 3 years in term of your fee capability because not only do you have to open the hotel, but you also have to ramp up the services and the occupancy rate of the hotel, as you know. And again, depending on where you are in the world, can be relatively quick if you are in a very affluent place, and it can be not as quick if you did something which is in a place where you need to create the business. So did I answer your question?

[Alex Brignall, Research Analyst]

That's fantastic. I'm just trying to, if I look at signings year-to-date, sort of a little below 50% of what they were a couple of years ago. So I'm just trying to work out when those signings would result in a slightly slower or changing rate of system growth. But that perfectly explains it.

[Operator]

We have no further questions in the queue at the moment. [Operator Instructions]

[Jean-Jacques Morin]

Okay. So listen, thank you all for the time. Thank you for the questions. And we really look forward to the next earnings call, which will be the full year sometime in February. So thank you very much, and stay safe. Bye-bye.

[Operator]

Thank you very much for joining today's call. You may now disconnect your handsets. Hosts, please stay on the line. Thank you.