Elis SA (ELIS) Earnings Call Transcript & Summary

Ladies and gentlemen, thank you for standing by. Welcome to today's Elis H1 2020 Results Conference Call. [Operator Instructions] I must advise you that this conference is being recorded today, Wednesday, the 29th of July 2020. I would now like to hand the conference over to your speaker today, Xavier Martiré, CEO. Please go ahead, sir.

Thank you. Good afternoon, everyone. Welcome to our 2020 half year results conference call, which is also webcasted. The speakers on this call will be Louis Guyot, CFO, and myself. So after an overview of the H1 business highlights, I will hand over to Louis. He will detail the first half financial performance. And I will then come back to provide you with some comments on our full year outlook. Let me start with the highlights of our H1 performance, which is, in many ways, remarkable given the unprecedented crisis we are going through. As you will see, the group has quickly adapted to preserve its margin and cash generation and proved once again its great resilience. The COVID-19 crisis obviously had a very material impact on our top line, with organic revenue down minus 15% in H1 and minus 27% in Q2. This sharp top line decrease resulted in a significant drop in headline net results of minus 51%. EBITDA margin, however, remained solid and was up plus 20 basis points in H1, and free cash flow was EUR 56 million, up plus EUR 75 million versus last year. This good performance on both the cost side and the cash side is a consequence of all the efforts made by the management teams, all group countries, and of the very good reactivity we showed when the first containment measures were implemented. Around 100 plants have been closed and furlough measures have been put in place at all levels of the group to reflect the lower activity. We have also started to implement more sustainable cost-saving measures in our 28 countries to prepare the group for the sluggish activity levels. It is fair to expect in the coming 18 months. We have also reviewed and significantly cut our CapEx plan for 2020 and 2021, especially all the industrial projects to increase capacity. This will come on top of the mechanical decrease of linen CapEx as activity slowed down. Aside from the savings measures, I would also like to underscore the very good operational progress and productivity improvements that have been achieved in several countries like Germany, where performance was especially satisfactory in the first half. Finally, our available liquidity today stands at around EUR 1.1 billion with our net debt to EBITDA leverage ratio at the end of June at 3.5x, stable compared to last year. As a reminder, the crisis impacts every one of our countries, with a marked deterioration starting mid-March. What you see here are the dates where -- when the lockdown measures were implemented in our different European countries. These measures were put in place between March 10 starting with Italy and March 21 in the U.K. They concern all our significant countries, except for Sweden, where no real lockdown occurred. Looking at this time line, I would like to highlight again out great responsiveness as we were able to communicate to the market as early as March 17 regarding the first set of cost adjustment measures to cope with this exceptional environment. We usually comment our numbers by region, and this is what I will still do today. But given the very different trends between our 4 end markets, the performance of each geography comes down to the contribution of every end market to its revenue. In industry, first, we saw a contrasting situation. On the one hand, some of our clients have simply stopped their activity and closed their plants. That is the case, for example, of many of our clients in the automotive or aeronautics industry. But on the other hand, some other clients haven't seen any material impact such as in pharmaceutical, energy, food processing, among others. At the end of the day, we observed a low point at circa minus 15% revenue decrease in H1. After the implementation of the lockdown measure, the decrease of the first half is around minus 5%. Hospitality was, without any doubt, our most impacted end markets with France, Southern Europe and U.K. and Ireland, our 3 regions with the highest proportion of hospitality in the mix, the hardest hit geographies. Our hospitality business fell to close to 0 from mid-March onwards with all restaurants, all hotels either closed by law, like in Spain, or with negligible occupancy rates when they remain open. In H1, activity was down minus 55%. In health care, we have been able to sign some new business. For example, with some small hospitals that switched to the outsourcing business model during the crisis. But this was offset by the fact that most countries decided to postpone all non-urgent medical treatments to make more room for COVID-19 patients, open resulting in some hospitals being partially empty, which means lower activity for us. At the end of the day, health care was down around minus 10% at the peak of the crisis and around minus 5% in H1. Finally, trade and services, just like industry, show a mixed picture with some businesses very impacted such as all non-food stores and collective catering but with other businesses like small and large food retailers maintaining normal activity. All in, the low point hit in H1 was at circa minus 25% top line decrease with minus 10% in the first half. Looking at H1 revenue by origin. Now organic revenue was down minus 14.7% in H1 with all geographies impacted by the crisis but with some differences linked to the mix. Geographies in which hospitality is the biggest contributor showed the strongest revenue contraction in H1. So that is France at around minus 20%, Southern Europe at minus 32% and the U.K. at minus 27%. Conversely, Central Europe and Scandinavia showed good resilience in H1, driven by, first, the higher proportion of industry of trade and services in their revenue mix; and second, the softer lockdown measures in some countries like Sweden or Germany, leading to a lower number of business shutdowns compared to what we observed in the Southern Europe. Latin America, and especially Brazil, remains a good growth driver for the group with positive revenue growth in H1 driven by good commercial activity and a favorable business mix with health care representing around 2/3 of revenue in the region. I would now like to return to all the measures taken by the group in a very timely manner to adapt to this unique situation. First, as soon as the lockdown measures were implemented and revenue started to decline, we reorganized production to concentrate the remaining volumes in a smaller number of plants. This resulted in up to around 100 plants being either shut down or virtually stopped, including 5 that will not reopen. At the same time, we also adjusted headcount at all the levels of the company, blue collar, field agents, sales team, maintenance team, plant managers, in addition to central cost savings at 8 offices of our 28 countries. Now on the cash side. We have also been doing tremendous work to limit the impact on our numbers. The obvious gain was on CapEx, where we have extensively reviewed the 2020 plan and canceled all projects related to capacity increases. We will also see our linen CapEx investment decrease during the months of low activity, as the linen that is currently not used by our clients will not need to be replaced. Finally, we have been working on redefining new commercial offers to address the needs created by the sanitary crisis. First, our pest control business is offering disinfection services which, as you can imagine, was met with real success. On the linen side, we had many successes in workwear with an increasing demand for surgery gowns in replacement of single-use products. We also noted a significant rise of interest for our outsourcing model with many players that were previously asking their employees to watch their work clothes at home, deciding to sign contracts with us in the light of health care concerns. These adjustments were, in fact, a combination of what we could call emergency measures, leading to immediate savings and of midterm action plans implemented in all countries. In all our geographies, this led to an absorption of the top line decrease of at least 60% on EBITDA, with Central Europe doing even better with close to 120% on the back of the strong operational progress achieved in Germany. Latin America did very well, too, notably thanks to the signature of short-term, very profitable health care contracts that boosted margin in the first semester. All in, gross margin was up 20 basis points in H1 despite the very difficult environment, which is, again, a remarkable achievement. Let's now go through our different geographies in more detail, starting with France. So after a good beginning of the year, we registered a sharp activity slowdown in March caused mainly by the sudden interruption in hospitality, which normally represent more than 1/3 of the country's total revenue with very good margins. So on top of that, the lockdown implemented mid-March also impacted our industry and trade and services businesses. But we have been noting a pickup since the end of the containment measures mid-May. Health care remained stable with the same phenomenon I described before, some gains of new contracts, notably with some public hospitals, offset by the room made in some private hospitals for COVID-19 patients. In total, H1 organic revenue declined by minus 20%, with a low point in April at minus 45%. We put in place a lot of savings measures in the country to limit the impact on margin to 120 basis points in the first half. As mentioned before, performance was good in Central Europe in the first half with a 210 basis point margin improvement in H1 and a limited organic revenue decline of minus 6%. The end market mix in the region is more favorable with a strong contribution from health care and industry clients, which respectively represent 40% and 30% of total revenue. Industry showed good resilience in the Netherlands, Poland and Germany, partially helped by softer confinement measures. It's not a surprise to see Switzerland was the most impacted country in the region due to its higher exposure to hospitality. As far as the margin progression is concerned, I would like to stress the operational progress made in Germany, the second largest country of the group in terms of revenue. We have been working over the 2 last years on improving the quality of our local teams, and we are starting to see the first results. We are, today, more efficient when integrating acquisitions. Our average productivity in the country has materially improved, and we have reduced the churn rate in H1. Moving on to Scandinavia and Eastern Europe. We see a pretty similar performance with organic revenue down only minus 7% in H1 and margin up 130 basis points. In this region, hospitality only represents 15% of revenue. And some countries like Sweden, Norway and Finland, which were not put under heavy containment, continued their activity almost normally. Denmark was a bit more impacted at the heart of the crisis, but the trend has been improving since then. As far as the margin concerns, the improvement is a consequence of the cost-saving efforts made in the first half and of a favorable mix effect as hospitality, the most impacted end market, has a below-average margin in Scandinavia and Eastern Europe. Finally, please note that the sharp improvement you see in June on the chart reflects the benefit of a material calendar effect. U.K. and Ireland now is, without any doubt, our most difficult geography with a very adverse economic situation. Our mix is not favorable there, with 1/3 of our business in hospitality, which has suffered significantly since March. But we also noted for industry and trade and services that a higher number of collective catering clients are very impacted by the crisis. In health care, we noted a gradual decrease in volumes throughout H1, as NHS has been rolling out a massive plan to postpone non-urgent treatments to make more room available for COVID-19 patients. And to make things work, the nursing home sector, which end up quite well during the crisis, is generally not outsourced in the U.K. Therefore, we did not benefit from such buffer as we did in some other geographies. As you can see on the chart, situation remains very worrying in June with a revenue decline of more than 40%, way more than in the other countries. In terms of costs, support plans from government came a bit later than the other countries, and we have adjusted our headcount from April onwards. On top of that, the more structural cost-cutting plan is currently being implemented in the U.K. This should bear fruit over the next 18 months. All in, H1 organic revenue decline is of minus 27%, and margin is under pressure at 25.6%, down 2 -- minus 250 bps year-on-year. So all in, the U.K. today, among the countries, is where the crisis has the biggest impact on our business so far. Moving on to Southern Europe, the most impacted region in the first half from a revenue standpoint. The beginning of the year was very good with plus 9% organic growth in January and 6% in February before the crisis massively impacted March at minus 40%. The mix is very unfavorable in this region, too. Around 60% of our revenue normally comes from hospitality clients. And all the Spanish hotels have been ordered to close by the government between March and May. However, Italy was slightly up in H1, with many of our clients operating in health care or in resilient industrial activities and no hospitality. We made the same efforts on the cost base than in the other geographies, but the top line decrease was so strong that we could not avoid a sharp margin decrease of 440 basis points. To conclude this walk-through, let's touch base on Latin America, where performance was very good in H1 with nearly plus 4% organic growth, driven by favorable end market mix as circa 90% of our revenue comes from health care or resilient industry clients. Again, we have seen many cancellation of non-urgent medical treatment leading to material volume decrease in health care in March and in April. Nevertheless, our Brazilian commercial team has been very good in signing high-volume, very profitable, short-term contracts with health care clients regarding protective overcuts for medical staff. This explains the strong activity pickup in June. At the same time, we have implemented headcount reduction measures in our 3 countries of the region: Brazil, Colombia, Chile. This, combined with material productivity gains, led to very strong margin improvement of plus 530 basis points. Profitably -- profitability in Latin America is now very close to what we deliver in our best-in-class countries, highlighting once again the value creation Elis can deliver, thanks to M&A. As I would like to remind you, we built our presence in the whole region through successive acquisition. Let's now have a look at our M&A activity in the first half, which was obviously subdued given the crisis. What we did is essentially closed deals we had previously announced or where negotiations were well advanced. We closed 4 deals in H1, the largest one being Kings Laundry in Ireland. It is pretty much the same rationale as what we did in Spain with Indusal or in Brazil with Lavebras. We were already #1 in Ireland, and we had the opportunity to take over the #2 and further consolidating our leadership in the country. In the U.K., we acquired Central Laundry, small player in the health care market with below EUR 5 million of revenue, where we believe we will be able to generate good synergies in the coming months. We also bought Haber in Germany, a EUR 20 million revenue health care player, further consolidating the health care market in the country, where we are already #1. And finally, we acquired TWC, a small player in Czech Republic, to increase our production capacity near Prague. As far as future M&A is concerned, we announced in March that we would not pursue any new deals given the situation. As of today, we want to provide an update on this. Even if we intend to maintain a very -- our very careful approach, we also want to have the freedom for potential M&A should good opportunities arise in the coming months. I will now hand over to Louis to comment further on the financial results.

Thanks, Xavier, and good afternoon, everyone. Let's start by looking at the full P&L. Revenue decreased by 15.7% in the first half and 14.7% on an organic basis. As Xavier already mentioned, EBITDA margin is up by 20 bps. Below EBITDA, D&A were up plus 5%. The reason for that is that linen depreciation represents 60% of the total depreciation. Linen have depreciated over a 3-year period on average. So they will follow the fast shrinking of the linen CapEx, but with the lag in the first half, we started to see a decrease in linen depreciation year-on-year only in the month of June. The rest of the depreciation, so 40% of the total, is linked to other assets, mainly industrial, which are depreciated over time ranging from 5 to 50 years and creating a strong inertia to the total D&A. On those assets, D&A are up circa 10% for H1 year-on-year. Now new current operating income and expenses. They are made up of restructuring cost, first, for EUR 12 million, which correspond to permanent shutdown of a few plants. In many cases, we had the project before COVID crisis from we decided to accelerate. The cost of -- also cost of provision for relevancy programs in Brazil, U.K., Chile and France. We have also EUR 22 million of COVID-related incremental cost that has exceptional bonuses to reward workers in premises during lockdowns. The fee we paid to waive our covenant test until the end of '21 and protection equipment for the employees. Moving further down. Financial results improved by nearly EUR 30 million. The lower financial charge is due to the lower cost of debt following the '19 refinancings onto the high H1 '19 base, which included some exceptional costs related to the refinancing. Headline net results remained in positive territory, that is half the level of last year, reflecting the very significant top line drop. We show further the classical reconciliation between net result and headline net results. Keep in mind that this method is always the same. Basically, we reset IFRS 2 and 3 expenses and the major noncurrent costs, net of tax, when appropriate. Let's have a look now at the cash flow statement. The good news is that we have not burned any cash in H1 with free cash flow at EUR 56 million in absolute value and above last year by EUR 75 million. Net cash flow is positive, too. Our net financial debt is, therefore, slightly lower than end of '19. Into more detail now, starting at the top of the chart. We see the cash effect of the exceptional items we previously discussed. CapEx were significantly down in the first half because the premium investment significantly reduced with the crisis. All the capacity-related industrial CapEx was also canceled for the year, but all the industrial CapEx that has maintenance CapEx were, of course, performed as our main priority has always been to preserve the quality of our plants and machinery. Working capital retirement is slightly positive in H1 with very different effects in the balance sheet. Inventories on the balance sheet are up because all -- because of all the offshore linen orders, which continued to arrive by ship. We don't need the linen yet, so we stock them in central warehouse. Receivables in the balance sheet are significantly down in H1 as a consequence of 2 effects. First, a technical one. When the revenues decrease, we collect bigger amounts of billing than the current one, creating a positive flow. But secondly, it reflects on the strong focus we have put on cash collection. And third, for the payables, the technical effect is exactly the opposite of the one of the receivables. Like for the P&L, the lower amount of net interest paid reflects the 2 refinancings done in '19 with a lower cost of debt. Income tax paid does not exactly match the ratio we usually get for the full year. At some countries, same as last year regardless of their forecast. And other countries have negative earnings before tax. The EUR 56 million of free cash flow is, therefore, a very good performance for our first half, even if we benefited for these technical effects on working cap. Nevertheless, it highlights the very strong focus we have put on cost control, CapEx control and cash collection since day 1 when the crisis began. Below free cash flow, the EUR 34 million paid for acquisitions correspond mainly to Haber, TWC and Central Laundry. Indeed, Kings Laundry was closed in July only and will therefore be accounted for in H2. Bottom line, net financial debt slightly decreased in the first half. So again, very satisfied not to have burned any cash in H1. Next slide. You find the bridge between what we actually paid to lenders on the amounts [ fingering ] in P&L and cash flow statement for financial. Two comments. First, the interest paid reflects the good refinancing done in '19. We are now around 1.5% of average cost of debt, all of it being fixed. Second, please note that we pay more in H1 than in H2 because we have a big annual coupon that we pay in February. You remember that we are targeting, for the full year, around EUR 70 million for the cash flow statement. Finally, let's have a look at our debt structure. Again, you remember we took advantage in '19 of the excellent conditions of the market to refinance some of our debt so that we have now a debt, which is long and well spread from '23 to '29. We have reduced the cost, which was above 2% to circa 1.5%, all of it being fixed. So actually, we don't have any major maturity before '23, which gives the market plenty of time to recover because -- before we need to start thinking of any refinancing. As far as liquidity is concerned, we have been working hard in H1, both from the cost side and on the cash side, to maintain and even further improve our financial flexibility. In total, our liquidity is fully intact at EUR 1.1 billion, out of which EUR 900 million of revolver undrawn and nearly EUR 200 million of cash. Regarding the covenants, you remember that we have obtained waiver regarding the bank test as of 30th June '20 -- December 31 '20 and 30 June '21. Please note that the discussion were fast and friendly as only the bank revolver and the borrowings, USPP are concerned. They are partners with whom we keep the line open in a very transparent way, which I believe is highly appreciated. Please note, by the way, that with the leverage at 3.5x end of June, we expect the initial covenant of 3.75. But of course, we didn't want to take any chance in this matter. Better safe than sorry. As a conclusion for these section, the key messages. Organic growth is down 15%, impacted by the crisis but with the split, which demonstrates the interest of geographical and segment diversification. Workwear and health care did quite well, leading to moderate impact on Eastern and Northern Europe as well as Brazil. EBITDA margin is slightly up on the back of the speedy implementation of a broad range of cost-cutting measures across all countries. Headline net results divided by 2 at EUR 50 million, impacted by the lag effect of the depreciation. Of course, we will benefit one day on the other side of the curve. Free cash flow is excellent at EUR 56 million, reflecting the control on cash, on cost, CapEx and cash collection. I will now hand back to Xavier, who will provide you with some comments on current trading and '20 outlook.

Thank you, Louis. So we like to regularly mention one of the key characteristics of the group, which is the resilience of the business. So looking at this graph, you see the evolution of the top line and the margin performance over the last 2 decades. The backbone of this resilience is twofold: first, the diversified geographical footprint with France representing less than 1/3 of our business; and second, the diversified portfolio of clients in terms of size and end markets. It is worth noting that this resilience profile was significantly improved with the acquisition of Berendsen and the addition of new countries in Central Europe and Scandinavia. Consequently, you can see on the graph that margins have constantly been evolving at high and stable levels within 200 bps range regardless of the external events. The first half of 2020 demonstrates once again the veracity of this resilient pattern. On top of that, one very interesting characteristic of our business is that linen investments come on and with top line growth. That means that conversely, they mechanically go down during bad top line years with a favorable impact on cash generation. I will now give you some first trends regarding the first 3 weeks of activity in July. For industry and trading services, we continue to register good progress with circa minus 5% and minus 10% year-on-year expected in July, respectively. On top of that, collective catering and facility management clients should hopefully further pick up in the next month with a lot of offices reopening in September. Health care is now almost back to normal, and we expect July to be flat on the back of the contracts we recently signed with hospitals and clinics to bring back linen garments to operating rooms instead of disposable products. Finally, hospitality remains under pressure with a very contracting picture. On the one hand, we see a pretty good level of activity from domestic tourism, especially for mid-range hotels on the French Atlantic Coast and French Riviera as well as on the Spanish Costa Blanca and Andalusia or on the west part of Denmark. But on the other hand, activity remains very weak in big cities such as Paris, London or Geneva, where most of the 5-star or palace hotels are still closed and will not reopen before September. All in, we expect hospitality revenue to still be down around minus 55% in July. As you can see, the environment remains very complex, which makes us -- that difficult when it comes to providing a full year guidance. Also, we do see continuing revenue improvement in July. It is still too early to provide a firm top line outlook for the year given the many uncertainties that remain when looking at September and the autumn period. Uncertainties concern the sanitary situation in our countries. Will there be a second wave? If there is, what will be its magnitude? Would there be full lockdown measures again or partial containment measures only? Today, nobody can answer this. Furthermore, we know that H2 should be impacted one way or another by the economic crisis that is coming. One should expect the unemployment rate to increase in most, if not all, countries as well as a rise in business bankruptcies across the globe. With that in mind, we feel it would not be reasonable to provide the revenue guidance for the full year. However, we can reaffirm our capacity to protect margins and cash generation on the back of our ability to react quickly and to adjust our cost base to the activity as we clearly demonstrated in the first half. So finally, the midterm cost-saving plans we initiated in H1 should also start to bear fruit in H2. For all these reasons, we feel comfortable when looking at the next 18 months and we think that in full year 2020, we will be able to deliver EBITDA margin and free cash flow after lease payments quite close to what we posted last year. So I thank you all for your attention, and we can now move on the Q&A session.

[Operator Instructions] Your first question comes from the line of Chirag Vadhia from HSBC.

My first question is -- you mentioned that you've shut 5 plants. How do you see plant evolution going forward? Will we use this time to make more strategic decisions related to where you want to keep plants open or not? And my second question is on permanent layoffs and furloughing. Could you give some color on the cash flow support you've received from furloughing in France and the U.K.? And could you elaborate on the scale of layoffs that you've done? And finally, last quarter, you mentioned that you had flagged around EUR 10 million potential bad debt. Do you see any of this changing in the second half, as you mentioned, that you might see businesses going bankrupt across your geographies?

Okay. So first question around the industrial strategy. So the 5 plants where we have decided to close definitely, it's 5 very small plants, not with a good level of productivity. And this is the reason why we benefited, if I may, on the crisis to take the final decision. And at this stage, we don't have in mind to close some other plant definitively. We have probably 1, 2 or 3 situations where we will not reopen too early the plants, but not with some definitive closure in mind. The second question was linked to restructuring program and cost of layoff and so on. I will give the opportunity to Louis to answer. And with the size of this country by country, it will be different, of course. In Latin America, for instance, where we didn't have any help from the government for partial unemployment, we had to put in place some definitive layoff program. And we are talking about several hundred people, of course. In France, we have launched something at the headquarter level. It's around 70 positions. It was more massive in U.K. where we have made some efforts not only in the operation but also with 150 positions less at the headquarter. And the third question was bad debt risk. Of course, we know that in the second semester, it's -- cash collection will still be a very challenging subject for us. What we can just highlight is the fact that cash collection beginning of July is very good. So it's a very good signal to see that we are able, in this context, to collect the cash at a very good pace, a little above what we expected, to be honest. And so it gives us more comfort for the second semester on this topic. And then, Louis, you can perhaps answer for the question of cost of layoff.

Yes, indeed. So we had to accelerate in U.K., Brazil and Chile because the furlough programs were very limited. Job retention scheme in the U.K.; law 936 in Brazil; Chile, nothing. So we have to accelerate for the full country and for France, as Xavier mentioned, only for the headquarter to reduce the fixed cost base. And we are speaking around EUR 8 million of total cost or provision at this stage.

Sure. Sorry. And that was the -- and the color on the cash flow support, sorry, that you're getting from furloughing.

Part of the EUR 8 million, a part has been paid in H1. And the second part has been just provisioned to be paid further in H2 or in '21.

And your next question comes from the line of Simona Sarli, Bank of America.

Just a couple of questions from my side. So the absorption of the top line decline of 69% in H1, so clearly was better than the 50% that you initially guided for. And you also mentioned some medium-term cost-cutting initiatives. Does that mean that going into H2, we should expect a similar absorption of the cost? And also, the second question is related to the free cash flow improvement year-over-year of EUR 75 million. So how much was related also to exceptional government measures? And how much is effectively underlying? And if that is sustainable also going into the second half of the year. And lastly, I was wondering if it is possible to have the organic growth decline in Q2 by end market.

Okay. So first question, we will have different events to manage in H2 with different impacts on absorption of the big decline of the top line. All in all, we are quite comfortable when we say that normally, we should be quite close to the margin of last year in the full year. So what will happen, we will have more cost in some area like prospection, so cost of sales because during the first semester, we have cut quite everything. It was quite impossible to sign contracts. So all the sales force was in partial employment in every country where it was possible. And in H2, we will start again to pay some -- our sales force to visit a customer and to sign new contracts. So it will add some cost. But on the same time, as you said, we have put in place some more structural action plans that will give some additional gains to the company in the second semester. So everything should mitigate the extra cost of new sales, and it will mitigate also the fact that in some countries, we could have the end of the governmental support. It will be the case in the U.K. and in Denmark, for instance. But with the structural action plan we have put in place, we will mitigate this impact. So it is the reason why all in, we consider that we should be able to keep this level of cost-saving in H2.

You had a question regarding public -- governmental package. Well, we have a couple of figures of what employees at home received by the government. I would like to underline that it's not exactly a fair question because we, of course, use governmental package. But if these governmental package was not there, it will not be just to pay people doing nothing at home. It will be like in Brazil, direct layoff. So I would like to rephrase the question around out of the good performance of the free cash flow how much is public money. That being said, I will give you some color. In France, technical unemployment was in the region of EUR 15 million. Spain, ERTE was around EUR 80 million. Great Britain, job retention scheme was in the region of EUR 10 million. Those were the main contributors from governmental scheme. Now the last question around organic growth by end segment. Well, roughly, we can say that with the pickup of June, hospitality was down minus 85%; industry, minus 10%; trade service, minus 15%; and health care, minus 5% for the Q2.

If I may, just -- it's a follow-up on the free cash flow question. So it was more to understand if the EUR 75 million year-over-year improvement, you think that it might also be sustainable in the second half of the year because, of course, it's sure that you might lose some of the government support schemes, but probably you will be having some other cost-cutting initiatives that will be helping in compensating that.

Okay. Sorry. I see now. Well, the good way to look at the model is not probably looking line by line. So take your assumption and turn over, which clearly we don't provide because nobody knows. Look at what Xavier said regarding the EBITDA margin. You can kind of conclude around what is the CapEx because of what's happening. But the big change is working capital. With regard with what I said around inventories are up in -- during the lockdown, that will decrease when we reopen because we will use the linen stock. On the other end, if the billing month after month regain, then we will have a negative effect on receivables on the small contrary on the provider of the side. So that is the impact that we may have. Also positive impact, we paid more tax and more interest than 50% of the full year. That's also a concern to take into consideration. So that's how, at the end, you go to the conclusion Xavier mentioned in terms of guidance for the full year.

Your next question comes from the line of Lucas Ferhani, Deutsche Bank.