Sodexo S.A. (SW) Earnings Call Transcript & Summary

Ladies and gentlemen, thank you for standing by, and welcome to the sell-side analyst meeting. [Operator Instructions] I must advise you that this conference is being recorded today, Thursday, 16th of April 2020. I would now like to hand the conference over to your first speaker today. Thank you, please go ahead.

Yes. Hello, everyone. This is Denis. I'm with Marc and Virginia. I hope you are all well. We are in Paris, confined in our home, but for the road show, we came together to the office. We are the only ones in the office. We are in a good shape and ready to answer your questions. So we'll let you start. And thanks for being with us.

[Operator Instructions] And your question comes from the line of Jamie Rollo from Morgan Stanley.

So I guess I'll kick it off. So maybe I could just start with the guidance for the second half, the sort of EUR 2.4 billion to EUR 2.8 billion number. Can you maybe start by giving some recent figures just so we can get a feeling for how you got to that figure? And also, what are your assumptions on the lockdown ending? Because there's quite a sharp improvement between the third quarter and the fourth quarter. And finally, is there any sort of differential there between the FM business and the catering business? You've given us the regional splits and so on, but maybe a point on FM, please.

Thank you, Jamie, for your question. So from 18th of March when we published our first announcements about the COVID impact, we have done -- it was our first version of the model, so we were at early stage, and we ran actually -- between 18th of March and 9th of April, we ran 2 or 3 more models to get to that point. And with a level of granularity, so that the teams on the ground could really look at -- it's really at the intersection of the segment and the regional segment in the country that it did the job, so it's been done country-by-country and within each country by segment. And it was also addressing all the support we can get from the government and also all the news of our start date of confinement and potential date of exit of confinement. So this is where we are today. When I look at Corporate Services, and I've heard that I think some of you found the numbers for Corporate Services too low, what you must remember is that we are primarily impacted in food and, to a lesser extent, in FM or sometimes not at all in FM, depending on the timing. So we do a large proportion of FM at Sodexo, 30% overall, 70% food, so that explains part of it. And then after, you really have to get down to geography by geography to really understand what's happening. So for instance, if I take the U.K., the U.K. is a country where we do a lot of global strategic account and a lot of FM and, to a certain extent, much less food. Therefore, the U.K. is -- is not impacted that much compared to other countries because of the large FM and global strategic comp. What you need to understand is most of our global strategic accounts do work today. We are present at their sites, and they are paying their bills. And yes, we've made some adjustments to services and invoicing and so forth, but the volume drop in global strategic account is minimal. In FM, in many sites in March, we were functioning almost normally, and some sites we had to drop, but -- so the U.K. will be less impacted in Corporate Services than the other regions because of its large components of global strategic account than FM. If I go to the U.S., the U.S. is a lot more food. I think it's about 75% or 70% -- 70% to 75% food. But in the U.S., some of our food contracts have particularities, which was not so good in the past, but which we like now is that they are cost-plus. And 50% of our food contracts in the U.S. are cost-plus, which means that contractually, we are entitled to invoice the client for whatever fixed costs we have on site. Therefore, we are actually quite -- invoicing quite a bit to clients on those cost-plus contract. When it's a P&L contract with a lot of retail elements, obviously here, we have a higher loss in revenue, and we suffer much more from a margin point of view, too. And then in FM in the U.S., again FM and global strategic accounts are quite present in the U.S., so we're not suffering so much in the U.S. on the FM and global strategic. Where we suffer a lot is in France because in France, a lot of our activity is food and a lot of our activity is in -- around Paris, and in those big towers at Île-de-France and so forth. And being a lot of food there, most of the sites are closed and empty. So here in France, we do experience a much higher drop than the average 30% in revenue. And -- but the combination of all of this, and if I were to take Brazil as an example, in Brazil, we usually invoice a lump sum and a variable part, the lump sum covering our fixed costs. So what we are expecting going forward is that in Brazil, we should be able, at least, to invoice a lump sum. Now also, the lockdown did not start at the same dates everywhere. And you see we had good results in Asia, we had good results in Lat Am and in -- and in Brazil in the month of March, and March is the first month of our quarter. So all this put together, we get to 30% in Corporate. Where we may be a little bit optimistic is the 10% for Q4. What we are doing here is that we are reexpecting a start sometime in May, with picking up in June and so forth. So we are expecting that, yes, the Corporate Services activity will restart gradually in May, which, for instance if I look at what we've heard in France from our President recently, it's not inconsistent. And I hear that we are talking about end of lockdown in Switzerland, in Spain and that Trump is also talking about it in the U.S. So God knows when will be the start date, but we are expecting a ramp-up starting, let's say, from mid-May gradually. If I look at Sports & Leisure, Sports & Leisure, there is hardly any activity left in Sports & Leisure. I think anecdotally, we have the Lenôtre boutique open in Paris over the weekend, over the Easter weekend, but that's about it. I think everything else is shut. And this clearly, for Q3, we only had operations for 10 to 15 days at the beginning of the month, beginning of the quarter, and then almost nothing. And it will continue for Q4. We are expecting slightly better in Q4 but not a lot better in Q4. Now in education, we are assuming that the schools and universities will stay almost closed at the level we are experiencing them today until the end of the year. What we are expecting is that they will reopen in the new school year and some reopening in the U.S. is in August, some reopening in Europe is more early September. So we do expect some reopening turnover in Q4 because the schools in the U.S. and universities in the U.S. reopened. Now we also have some FM contracts in education, and we also have some minimum invoicing in some universities and schools. For instance, in schools, we are producing more and more meals for children in need financed by the local authorities. And today, I think, Denis, we are producing almost 300,000 meals a day.

Yes. Yes, exactly. 270,000 in schools and about 30,000 for universities.

Okay. So this is coming us up. And at the end of March, we didn't have it, it started early April. In Healthcare & Seniors, we have 8% and 5%. The 8% comes mainly from a drop in retail. We were doing quite a bit of retail in health care, the famous -- the kiosks that you find in hospitals. And here currently, with the lack of visitors and the situation, the retail element has dropped significantly. What we also experience is that we have some units being closed, for instance, it happened in France with some private clinics because they were doing more comfort medical support than absolutely necessary medical support. And today, everybody and everything moves towards the hospital treating the COVID, and there is no other, actually, intervention or theoretical interventions and so some sites have been closed. But at the same time, in health care, we invoice more because we are asked more services in terms of deep cleaning and so forth. So it's a mix of things. But the main element is a drop in retail. In Benefits & Rewards, it is actually a rather complicated model. We are assuming that in Q3, there will be a drop in issue volume and drop in reimbursement. The issue volume will drop a little further than the reimbursement. It will weigh on the revenue because most of the revenue comes from reimbursement. But we will have a bounce back in Q4 because we are very much linked to the reopening and when are people coming back to work. And we believe also that the volume should be maintained for people working from home. The key question we have and that we are not able to assess at the end of March looking at the numbers is what's happening with people being in technical unemployment or partial unemployment. Are they getting vouchers and benefits or are they not? So we would be a lot more knowledgeable at the end of April, but our models today tells us that there should be a drop of 20% in revenue in Q3 and 10% in Q4.

That's really helpful. Can I ask a follow-up? You gave a lot of information there. So you can't give us any recent data? You can't give us the March or the first sort of 5 or 6 weeks as opposed to 1?

I can give you some indications. But when we did the model, it was before we knew the March numbers, and the March numbers do not change the model. The March numbers are a confirmation that our early estimates and modeling was correct. So we -- from out of March numbers, I will not make change to the model.

Yes.

There were some interesting anecdotes, for instance, just for BRS, we shared this with some investors. In Italy, which was the first European country to go into lockdown, the issue volume for BRS was flat versus last year. There was no drop in March. So it is very difficult to assess what will be the impact for BRS, because we were expecting a drop in production. And most of it is in paper, in our case, in Italy, and we did not have a drop. So we will have to see with April what's going on. March is actually not telling us much for BRS Except that the numbers in March were pretty decent. They were not positive, but they were pretty decent for the first month of the lockdown. Obviously, on site, there was a drop, but it is in line and it's not actually contradictory with our hypothesis for the future.

And April will tell us a lot because April -- March was really a transition month, both for revenue but also for flow through. And I think in April, we have more of a, I wouldn't call it steady state, but more plateaued month. And I think in the end of April will give us a better refinement of our model.

Okay. And you mentioned FM doing better, could you give us a rough split of what FM might be then within that first quarter figure?

FM is 30% of the overall business. It is obviously more in Business & Administrations than in the Healthcare & Education. So the fact to have a high content of FM is actually a better top line protection in today's environment than if we were only food. So we experienced a further -- a stronger drop when there is a lot of food than when there is more FM. So the FM component is actually a component which helps slow down the drop in revenue.

Sorry, I actually meant what drop in FM are you factoring in? Clearly, it's better than catering, but what -- how much better?

Well, it really depends.

Yes.

But today, the behaviors of our clients has been to maintain a good level of FM. In March, the level of FM was pretty good in most geographies where we operate. It was the case in China in the past, too. There is a small drop, but I will say it's not huge. Compared to what we are seeing in food, it's much lower.

Okay. That's helpful. Also, if I can just follow up on one more. You said cost-plus sales still invoicing in the U.S. So is that even when the site is closed because they're employing the [ start-of-self ]? Does that mean you can still invoice even at the self-starts?

When you are in pure cost-plus and the contract is well written, for instance, your depreciation, your unit manager and some ongoing fixed costs you have can be recovered through the cost-plus contract. Now some sites are closed, but some sites are partially closed, so there is a lot of discussion with the client right now. Even if we don't have a cost-plus or whatever, it's all a question about discussing with your clients as to what's fixed -- what fixed element you should keep and so forth. For instance, in France, even though we have mainly P&L accounts, we are discussing with every client because it is known by the clients that we've put some CapEx, that we have with unit managers and so forth. And so it's a question of negotiating with the client how much we should be invoicing them on a monthly basis to recover those fixed costs. And we have some successes. I mean, there are ongoing discussions, obviously, and not everybody is amicable to that discussion, but we have good results. So we can invoice. In many, many situations, we can invoice. Sometimes not a lot, but the key question is to recover, at least, your fixed cost.

And your next question comes from the line of Jaafar Mestari from Exane BNP Paribas.

Just a very quick one for me, if I may. What about the cash burn of the business in your scenario? If I say EUR 2.8 billion times 25%, it's about EUR 700 million. Your H2 EBITDA last year was about EUR 700 million. So is it fair to assume that you're breakeven operationally and then you're only burning a little bit of cash if you decide to put some CapEx in for finance costs, et cetera?

Yes. It's a question I've asked for the past 2 days. And my answer to that is that when you look at our cash generation for fiscal year '18 and fiscal year '19 during H2, if you look at the free cash flow for H2 the past 2 years, we were circa EUR 900 million. So H2 is clearly a semester where normally we generate quite a bit of cash. Here, because of the drop-through of 25% to, let's say, an average of EUR 2.6 billion, this drop-through at UOP level will be also a drop-through for most of it at cash level. So it will obviously come as a deduction to cash. We will reduce CapEx and the indication I gave is that we would like to reduce CapEx to half of the level of H1 to something around EUR 135 million. So we will reduce CapEx. And then the question is what impacts on the working capital. So our working capital is negative. So at the end of August every year, we have a significant working -- negative working capital, especially on the on-site and BRS, but on onsite. With the drop in volume of activity, we are expecting this working capital to contract. Our working capital will mostly be impacted by the Q4 activity, not so much the Q3 activity, because our payment terms from a client side and payable side are usually about 60 days, 70 days. So Q4 is really what's going to be making our level of working capital. So if we assume some contraction of the working capital, so we will be consuming cash by having a less negative working capital, altogether put -- if we put all things together, it looks like, and this is what our models are telling us today, that we will not be generating cash in H2. So we were generating, let's say, EUR 900 million on average, we should be around 0. Now this is around 0, plus or minus a couple of hundred million right now. I'm also a bit in the dark. But this is what the models are telling us. So our cash burn, on average, it's more a question of noncash generation rather than cash burn. And if there is a cash burn, I think it should be modest. But so what I'm thinking is that the net debt I have at the end of February is probably a good proxy for the net debt I will have at the end of the year, provided that there is some currency impact on the assets and liabilities. So they could also be -- the reais has been weak, the dollar is a little stronger, so we could also have some current rate impacts on the debt. But this is where I am today with the model.

Excluding time...

That's very clear for H2.

Excluding, yes.

Yes, constant rate.

It's constant rate.

And outside the seasonality, if, hypothetically, the situation remains the same for an entire year, are you at a steady state, no cash burn, no cash generation sort of level outside of the seasonality?

This is what -- when I wrote the model and when I ran the forecast for working capital, this is what I have today. And when I said plus or minus EUR 100 million or EUR 200 million, because this is not an exact science, and in terms of forecasting free cash flow with -- it's never been perfect in the past. But this is where I am, yes.

And your next question comes from the line of James Ainley from Citi.

Two questions, please, if I may. Just following on Jaafar's questions then about cash burn. I mean, I guess the danger is that these things continue for much longer, that you risk, I guess, a technical covenant breach, not because the debt's blown out, but just because there's very little EBITDA being generated. What conversations have you had with financing providers about potential covenant breaches? And what is their response? And what measures do you think might they take? And then secondly, thinking longer term, there's a lot of discussions around how we all are getting used to remote working, people thinking about different strategies towards remote working. What conversations are you having with clients about their longer-term plans on remote working strategy, please, and potential risks to lower volumes in your B&I business in particular?

Yes. You're quite right. Your question is quite right. There is a net debt on one side and there is the EBITDA on the other and the covenant is a ratio of the both. So today, all the models we've been running tell us we are fine with the covenants, so we're not worried about this. We have a healthy conversation, regular conversation with our USPP holders. Some of them, the biggest one, gave us the call in March after our publication and after also S&P confirmed our rating, even though with a negative outlook on the 25th of March. So we have regular contact with them, and right now, there is nothing to worry about. We -- now if one day we need to do something, we'll do something. But right now, nothing tells me that I need to do something.

Right. And in terms of your second question, James, the -- yes, definitely, we -- what I hear from our clients, from my peers is that I hear 2 things. First, that companies have been able to operate remotely and that they've conference calls and everything. The systems have worked and they have developed pretty sustainable systems. However, I hear that CEOs want people back in the office, probably not in the same magnitude with and definitely having a little more home office will help optimize real estate. And so that's something that will definitely have an impact. What we see is probably more of a hybrid model coming into play with a bit more home office, but really companies wanting people to get together. And the ramp-up that we will live through now with progressive -- we see lots of companies, particularly in China, as Corporate Services sites reopen, we see very much the Red Team, Blue Team or Team A, Team B scheme working, operating like one team in the home and the other team in the office. And next week, the contrary with deep cleaning in the middle. So those things will remain for several months, I believe. And -- but we -- but I hear that people want -- CEOs want people back in the office. What it will change? Yes, probably there will still be an impact in terms of volume, but the good thing is the on-site business is very much a variable cost model, so we will adjust. And at the moment, by the way, we are -- our teams are engaging with our clients to really anticipate what they have in mind in terms of restart of the business, to properly size, particularly labor, because the big thing is on labor and the food, you provide food that you deliver, it's as simple as that. But labor is the more -- is the equation that we have to solve. And so less volumes, but also a way to deliver our services differently. We've seen, of course, in this past month, a lot of grab-and-go offers, to-go type lunches, et cetera, that we have delivered. We've also increased our directly B2C offers like typically offers for dinner that we provided for tough people in hospitals. And so we have been able to capture a share of wallet in our consumers that we hadn't necessarily anticipated and that we've been quite efficient at capturing. So different ways of delivering services. What we need for sure is to, moving forward, is to ensure that we really flex our operating models, be more multichannel. We already were in that momentum, but we have to accelerate this. We saw possibly less volumes, but as we flex our offers, we believe that we can also conquer segments of clients that are of a smaller size. We capture a lot of the big companies that really do outsourcing. But as we flex a lot, we think that we can be also attractive with more like convenience corners. If you remember well, we spoke about the Enjoy offer. I think it was at full year result last year. And typically, it's sort of a small on-site offer, but very light, Very convenient for smaller companies like mid-size. And so I think we can also -- the more we flex, the more we'll be able to capture different sizes of clients. So yes, changes in trend, but lower volumes, not necessarily.

Okay. Can I just have one quick follow-up question, just on what the business environment is like? I imagine there's a bit of a standstill on new business, but also, by the same token, I mean, you're not losing so much business in this environment. Is that fair?

Yes. Yes, I think we've seen a bit less activity on tenders, definitely. At least we haven't seen new tenders for the past month. I don't recall any, but maybe few here and there, but not massive. What I see in the next months is probably a better retention because less tenders coming up, but not massive development either, of course, as a consequence. We also concentrate a lot on retention. So in some segments, and I have Healthcare in mind, in some segments, Healthcare in the U.S., particularly some hospitals, have still kept their tender process. But some have stopped it and I have one in mind where the client stopped the process and say, finally, "Well, you know what, let's be simple. I stay with you," and we renewed the contract for 5 years. So that's good. One that decided to go self op and sent us a letter saying, "Oh, by the way, no, no, no, I won't go self op, I stay with you." So that's good. And some others that are still considering maybe a change. And again, back to -- sorry, a small parenthesis on this, but in Healthcare North America, still uncertain and I still have some concerns. We are all hands on deck, but I still have some concern on Healthcare North America on our side. But overall, yes, probably a better retention and a bit less development in the coming months.

And your next question comes from the line of Richard Clarke from Bernstein.

Three questions if I may. First one is just on -- in U.S. universities. The meal plan revenue, I know some of your peers have sort of suggested that some of the revenue kind of gets maintained because it was paid a long time ago. Maybe if you can just update us on what's happening on that, if you had to refund that. And then just on work -- second question on working capital dynamics in Benefits & Rewards. You mentioned in the release something about a cutoff date issue. Maybe you could just give us some details on that. And then how are you thinking about working capital in the second half if people are earning their vouchers but they're not necessarily able to redeem it because all the merchants are closed, is that positive for you? And then maybe just also comment on currency in B&R as well. Just to confirm that your guidance is not including the drop in the reals? And then third question, just feeding on a little bit from the point on work from home. Any other comments on the other structural changes. What will you have to implement in terms of social distancing within your canteens? Is virtual queuing going to be a necessity? Is cashless payment going to be a necessity? And what does this mean for your kind of CapEx requirements going forward?

So on the U.S. universities, well, we have a mixed situation. Some universities have asked to refund, some have not. And when we refund, it's just a cash situation. We haven't incurred the cost, so we can refund the viable part. Of course, the fixed part is the fixed part. It's the cost that we have for operating. So we've had here and there some requests and we negotiate with the clients. But with -- in other cases, there hasn't been any. I wouldn't call it massive. The big question moving forward is what is going to be the meal plan structure in the future? Of course, that's part of the unknowns. And I can tell you that at the moment, it's very hard to understand what -- first, we don't know if universities will reopen, to a much earlier point. At the moment, we believe that it will reopen in mid-August, but would that be the case, still unknown. And how many will reopen? And what will happen to the meal plan? So there are some question marks here. I'm not immensely worried because that scheme will remain. But what parameters will evolve is still a question mark. And that will link to the second -- your last question, if I may answer before Marc talks about working capital. Yes, of course, social distancing, some of the disruption linked to the COVID-19 will impact the way we operate our services. We also believe that there is a big opportunity, and that's what we're working on at the moment. There's a big opportunity for us to be the thought leader and really as a front-runner in terms of how we accompany our clients in the recovery phase and the restart phase. And we are truly unique with both food and Hard FM and Soft FM, a lot of experience in deep cleaning and disinfection in -- particularly in Healthcare. And at this moment, we are very vocal. We go to webinar-to-webinar to talk with our clients on what would be our recommendations and how we can help them both with FM services in the workplace, arranging the workplace, managing social distancing in the workplace and all the cleaning activities that goes with it, but also social distancing in restaurants, yes. We already have done that at the beginning of the crisis, when there was not yet any lockdown, and it's been very efficient. We have deployed very fast some very strong protocols, rearranging tables so that there is less -- so there's more distance, longer opening hours of our restaurants, pushing more for like grab-and-go offers like these people don't stay in restaurants. Cashless payment, yes, but the move was already engaged, so we'll just deploy faster. I think this will help us. Sometimes in the past, we have the technology, but sometimes clients they're still -- they're slow. They don't want it. I think this will help us go faster on this. So will there be a bit of CapEx? Possibly, but it's not going to be massive. But yes, going into contactless or virtual will -- I think that COVID-19 will accelerate the digital transformation of the way we deliver our services. On working cap, Marc?

Yes, Richard, the H1 situation, the working capital variance in H1 was not good because of the Rugby, because of the prompt payment code and because of cutoff issues, because of benefits, I should say, last year from BI. BRS, the model is prepaid. We give credit to some clients. But in Europe, we've had some cases where we issued a lot during the last few weeks of Q4 without actually reimbursing that much. And so when you have this with a couple of service at the same time and you issue a lot and you don't reimburse, the problem is that it inflates your cash flow at the end of August and then you reimburse the following year. It's a bit like what we had with the Rugby. We invoiced in July and August and we incurred the cost in September, October. Similar situation we had with BRS and with the Rugby. The second half issue volume versus reimbursement. And this is a tricky part actually, and this is why the modeling at BRS is complicated, and we need some data to see how people behave in their confinement under lockdown. Because obviously, I mean, you issue, you get the cash very quickly because we don't -- we give very little credit terms. We get the cash and we are reimbursed. In some cases, at 30 days, some cases, 45, but I would say, maximum 60 days. So obviously, I mean, we are sitting with the cash between those 2 days. When issue volumes go down, if your reimbursement volume goes up or remains high or higher than the issue volume, then it squeezes your cash. But when the reverse happens, when your issue volume goes up and your reimbursement fall behind, because normally, when your issue volume goes up, you reimburse what you issue a couple of months later. So when the issue are on the rise, you generate cash and it goes into your working capital. What we've assumed today is that we should have a drop in issue volume and reimbursement in Q3, but the drop in reimbursement will be lower than the drop in issue volume. So we will normally be burning a bit of BRS working capital during Q3, but because Q4, we see the exit of the confinement and so forth, we see the issue volume repicking up to past levels, while the reimbursement will slow a little behind. So we will be remaking the working capital by Q4. The BRS team sees a situation which balances itself quite well. But because the reimbursement in Q4 will be delayed and Q3 will be lower, there will be an impact on revenue. The beauty with BRS is whatever you don't reimburse in one quarter, you will be reimbursing in the following quarters. So whatever we do not make based on reimbursement, if it's just a delay, we'll make it later. With regards to currency impact, I do a similar exercise every half year. I projected the current March exchange rates till the end of the year to calculate our average rate for fiscal year '20 to become our new current rate when it will be at the end of August '20. And it's true that currently, the reais being at BRL 5.60 or BRL 5.70, and we were at nearly BRL 5 at the end of February, means that BRS and on-site combined will have a drop in UOP at current rates. And so this drop, I estimate that the drop at the group level will be between 10, 12 basis points at group level on current rate margins if the reais at BRL 5.7 remains as such till the end of the year.

Just maybe to add the BRL effect is not taken in this hypothesis.

No. And in the hypothesis, we are at constant rate. Because we are compared versus last year, so we are at last year's rate. But the currency impact in H1 was mildly positive, positive 1.6. And at H1, we were compensating the reais going down by actually a relatively strong dollar and a steady sterling. So today, from a currency point of view, we are slightly up in H1. I think this could be disappearing by the end of the year and be close to 0.

And your next question comes from the line of Joe Thomas from HSBC.

Three for me, please. Firstly, as we think about potentially the margin, a little bit more working from home and also the impact of social distancing and so on, as you renegotiate contracts, can you give us a sense of what you might do in order to protect returns, thinking around CapEx, depreciation and covering those sort of fixed costs that you have in the business? The second thing is I wondered whether, as this rolls on and hopefully it doesn't, but let's say that COVID does roll on for a period and goes beyond the year-end, do you see the scope for taking large amounts of cost out of the business? I'm talking about costs that might otherwise have been thought of as fixed. And then finally, you talked about some remaining concerns on -- in the North American Healthcare business. I was hoping you might elaborate on what they are and what you thought states of that was going into this period. And any sort of high-level thoughts about the other businesses that have perhaps been under pressure in the past?

So of course, our goal is to keep the level of margins or even improve the level of gross margins that we have on our operations on site. So when we renegotiate, typically with new services or when we incur more costs, particularly, we know that we will incur more costs with personal protection equipment. We -- it's part of the discussion that we have, typically, like when we have the inflation pass-through and whether we have to -- that's what we -- it's a business as usual, right? I don't anticipate, at the moment, that we would have to throw more CapEx into our operations with -- the COVID-19 doesn't change massively the way we operate, hasn't changed massively the needs for the clients. And so there are some -- there are no fundamentals that are changing. So what we need to do is to remain competitive. That means I believe that some of our clients, which will be probably under economic pressure with the economic crisis that we will live through, yes, some clients will ask us for selling, so we need to be prepared. We need even to sometimes proactively propose ways to operate on a more frugal way, on a more simple way so that they get the savings and we keep them. But that's -- this is -- and again, also on the ramping-up phase, I'm not too worried. We'll have to, of course, be very, very careful in the way we ramp up, but we ramp up on sites that we know, it's a remobilization, again, on sites that we know, so I'm not extremely worried about this. In terms of fixed costs, yes, we have this program that we call Fit for the Future that has delivered results. We had already some results in H1 linked to Fit for the Future, some savings. But we will have to go 1 step further. I think what this crisis has taught us is the fact that we were able to be very agile. As soon as the crisis started, I really put the decision level at the lowest in the organization to be very agile and adapt to the very granular situation on the field. And we've demonstrated a lot of agility and creativity and responsiveness. So I think we've learned some good things. And I think we've discovered also some of our structures or some of our things are probably not that useful. So that will help us take some cost out of the business. So we will streamline more the way we operate our business, that's for sure. We're starting this now. We're starting the design and have nominated a subgroup of the Executive Committee to work on that. So yes, yes, we can -- you can expect some -- a bit of restructuring moving forward. Now on North America, I said before, talking about Healthcare, I'd say that we have a good dynamic in North America in many segments. Corporate Services is having a good organic growth, good dynamic. Energy & Resources, also quite, quite strong. Government agencies with -- we still have some improvement to do in the U.S. Marine Corp., but it's really improving month after month. So although there are some stronger -- strong fundamentals, schools is doing well as well in North America, for sure. And so I have some element that tells me -- that tell me that we are stronger than we were before. We have a strong management team now, maybe 1 or 2 moves still to adjust. But I think very pleased with the new Region Chair in North America. He has a lot of experience. He's been with us for 10 years. He's been with Aramark and Compass before and knows very well the business and gives a great momentum to what we are doing. And they have -- that team has been amazingly reactive in terms of adjusting the workforce when the crisis hit in North America. So I think we're good there. Healthcare, 2 things. We see that the operations are more solid. And the best example is that our margins, our gross margins are improving. However, they are impacted a little bit by the medical cost. But the medical costs are beyond the, I would say, the control of the operations. So -- and when you have gross margins improving, it demonstrates that you are doing a better job. You don't increase your margins if you don't do a good job on the field. So that's a -- that's -- for me, that's an element that tells me that our operations are much more the control than they were 3 years back, where we had all these problems that turn into clients being dissatisfied and then thinking about changing providers, et cetera. The concern that I have, as I said, at the end of the fiscal year last year and in Q1 was that we still have some, let's say, sizable contracts that are due to tender in process. And I still have -- still not fully secured on some of them. And of course, it's -- it doesn't -- I'm not more concerned than I was 3 months back, but I'm not more optimistic. It's still this balanced situation which still has a lot of unknown. Team is all hands on deck, but we're still -- we've seen -- in the aftermath of that turmoil that we lived 2, 3 years back.

And your next question comes from the line of Daria Fomina from Goldman Sachs.

I have 3 short questions. The first one on CapEx. Can we talk a little bit about 2021? And sorry for going that far in these uncertain times, but anything from the current environment that can help us leap into the higher or lower coverage in '21, the lower tendering activity, but on the other hand, some delayed CapEx? The second question on labor intensity. Can you help us understand from the clients that are opening up in Asia how the labor intensity of the business is affected, basically probably less, birthday lunches, more portioned food, more stuff for that? Anything changing on that front that could affect the profitability for a little bit longer? And my last question, just definitely to check, did I understand you correctly that you expect operating -- sorry, that you expect net 0 cash for the second half, basically, both CapEx and interest? Meaning that you expect actually generate cash on the operating level? And that's it.

I mean CapEx 2021 will depend of -- we are a client-centric organization. And if we have good business in front of us, what we said a few years back is that we are happy to put some CapEx. Now it's true that when I look at the Sports & Leisure segment, I will not be expecting them requesting a lot of CapEx given the slow start of the year they may have next year. And universities today are very prudent, so we will see about the reopening. We passed the message so that we are controlling tightly the more in-house or the internal CapEx, the development we do in our systems and so forth, so that we keep only the strictly necessary ones, but we need also to get our IT infrastructure to evolve and we embarked into a revamping of our IT structure. So we will want to do some IT CapEx. So 2021 is a bit early, but I will not be expecting massive CapEx in the Sports & Leisure segment. Universities will be very fluid. The fact is that also in the past, we took some commitments to make some CapEx at milestone and some milestones will be coming even this year, but some of them will be coming next year. So we need to reassess all of this. Labor intensity in China. What's happening is that when you reopen the site, you've been there before, you know the tariffs, you know what you have to do and you discuss with your clients at what level they want you to reopen. So maybe you were at 100 before the crisis, now they wanted to reopen at 80. The game is for you to start, at least, you're reopening at 70 so that you rebuild your margin and then you progress from there. So the advantage you have when you restart is that it's a restart something you already know and you have a client asking you to restart and then you have a dialogue before you restart, so that's what Denis was mentioning earlier today. It's all those insights and relationship we have with our guys when we are accompanying them to restart. We are in a stronger position to restart correctly. And what we see that China is a very tiny country for us. It's important, but tiny. What we see in China is they've done well in the restart.

Yes, China has demonstrated fantastic agility, both in the closing of the site but also in the restarting. That's for sure.

We've learned a lot. Now can we expand what we've learned in China globally and do every country self-success done in China, that's the question. That's a big question. Zero cash in H2, I'm talking what we call internally LGO free cash flow. So it's auto finance mode minus CapEx minus working capital volume. This is what we have today on our page, I forgot the number of the page, but maybe Virginia, if you can help me. It's on Page 13 as free cash flow, which is negative, minus EUR 243 million this first half. What I'm saying and what I was commenting earlier is that number for H2, currently in my model, runs around 0.

And again, it's a model one.

Plus or minus a few mistakes here and a few opportunities there, but -- while in the past, and if you look at the past 2 years, H2 was more in the EUR 900 million arena. But then you still have to pay for acquisitions, but there will not be much acquisition, share buyback. There will not be any dividend in H2. The other cushion we have is other changes, and this is where the currency comes into play because we have asset -- cash assets in reais and they will depreciate. We have dollar debt. I don't think they should be moving much. So we could have a potentially small negative number in other tangent.

And your next question comes from the line of Vicki Stern from Barclays.

I've got 2 questions. Just as you're coming back on China, could you help us understand a little bit about what you're seeing? I appreciate it's a small country for you, but just what is the activity level now back to -- in B&I and some of the other segments? It's just quite helpful to know as we understand the shape of the recovery. And then just circling back on H1. Your margins in BRS sort of impressively good, particularly given the mix of the revenue decline, i.e., Lat Am weaker. Can you just shed a bit of light on what's been going on there and what's driven that margin growth? How we think about that?

Thanks, Vicki. So regarding China, roughly speaking, so we had our Healthcare, as always been, of course, at full speed for the -- since the beginning. B&I is -- all our sites are reopened, except one. And we are seeing good activity levels. Of course, with the shape, the split of services in China is more skewed towards FM. We have a lot of FM. And as Marc was saying earlier, the impact on FM is, of course, less. But we see -- what we see restarting is we see lots of sites restarting, Team A, Team B or Team Blue -- red blue -- Red Team, Blue Team, sorry. And so a bit less volumes in food, but still some, I would say, some reasonably good volume. So we have been quite satisfied with March numbers. They're, of course, not as before, but the decrease is not massive. So we're pretty satisfied. Schools have not reopened. We heard that they would reopen in the coming weeks, but there is still a question mark, given that there are still some restrictions. China is scared about this COVID-19 coming back. So they're very, very careful about this. Borders in Hong Kong are still closed. Our lounges in Hong Kong are still closed. So it's still a big situation, but we see good level of activity in B&I, definitely, yes. Yes. And for BRS, Marc, H1 margins?

Yes, they were good. Yes. The margin for BRS was good because a couple of years ago, we embarked into a transformation of -- digital transformation and investment in BRS, especially on the employee benefit side. We also went through diversification with Rydoo and incentive and recognition. But the results, we -- are starting to come through on the employee benefit side. And actually, we had a reduction in cost also because the transformation on some topics is now done. And we had an increase in productivity and better gross profit. So it's a mix of both, but we are seeing the results -- the beginning of the results of the transformation in BRS.

And sort of were we not living in the COVID-19 world, I suppose, how much more of that is the flow-through over the next 6, 12 months or so?

The BRS business, the trajectory we have with them is that they are -- they should be improving their -- everything remaining the same pre-COVID, they should be improving their UOP over the next 2 years, given the investments we've made in the past. So now with the COVID ball in the air, but I think they'll be pretty quickly back on their feet.

Yes.

It's an agile business. We went through crisis with high unemployment before. It lasts a few quarters, but then you reconsolidate your volumes. I think it also gives us -- the crisis has given us some opportunities because some of the clients who said, "I want to get delivery of my vouchers," and says, "Now, we call." Now if you move to cards, we can. So we also see some governments coming towards, asking us some new services, helps how to manage benefits to governments and so forth, and we have the agility to do it. So I mean all these digital investments we've been making over the past years will help us be more agile in the future.

And your next question comes from the line of Felix Schlueter from Goldman Sachs.

Perfect. It's a very quick one. I just wanted to confirm if I understand correctly that you said Healthcare volumes were up. And if you could just give us a picture of what's happening in Healthcare? Because I presume that occupancy might be down given that the normal operations albeit put on hold. But is there anything else going on that is supporting volumes?

Actually, Healthcare volumes are not up. We see, in hospitals where there is a COVID-19 activity, yes, we see volumes up, definitely, because we need to do more cleaning, and they're, of course, more active. There is more staff, so we feed more people. But on the other side, we see that some hospitals have less activity, some private clinics have less activities because they've postponed some nonurgent surgeries or things like this to be able to accept COVID-19 patients. So we see some even have closed. And -- but we also see a small drop in revenues in healthcare because of the retail that is being closed. There is no more visitors in the hospital. So the only services that remain are -- the immense majority of hospital is just patient feeding, staff feeding and FM. So in retail, depending upon the country, it can be 10%, 20% of our activity. So -- and then this has an impact. Yes. All right. I think we need to get moving. So I'd like to thank -- thanks a lot for your presence. And hopefully, even though I think we've proven efficient in these virtual meetings, I would hope that the next meeting will be a physical one. So thanks for being with us. Take care of yourselves, and have a good evening.

Bye.

Thank you very much.

Bye.

Thank you. Bye-bye.

Thank you. And that will conclude our conference for today. Thank you all for participating. You may all disconnect.