© 2024 Alpha Spread Limited. All Rights Reserved.
www.alphaspread.com is not operated by a broker, a dealer, or a registered investment adviser. Under no circumstances does any information posted on www.alphaspread.com represent a recommendation to buy or sell a security. The information on this site, and in its related newsletters, is not intended to be, nor does it constitute, investment advice or recommendations. The information on this site is in no way guaranteed for completeness, accuracy or in any other way. In no event shall Alpha Spread Limited be liable to any member, guest or third party for any damages of any kind arising out of the use of any content or other material published or available on www.alphaspread.com, or relating to the use of, or inability to use, www.alphaspread.com or any content, including, without limitation, any investment losses, lost profits, lost opportunity, special, incidental, indirect, consequential or punitive damages. Past performance is a poor indicator of future performance.
You don't have any saved screeners yet
Intrinsic Value is all-important and is the only logical way to evaluate the relative attractiveness of investments and businesses.
Ladies and gentlemen, welcome to today's Legrand 2020 Nine-Months Results Conference Call. [Operator Instructions] For your information, this conference is being recorded. At this time, I would like to hand the call over to CEO, Mr. Benoît Coquart; and CFO, Mr. Franck Lemery. Sir, please go ahead.
Thank you. So good morning to everybody. Franck Lemery, Ronan Marc and myself are happy to welcome you to the Legrand 2020 Nine-Months Results Conference Call and Webcast. First of all, I hope that all of you, your close ones and colleagues are doing well. Let me remind you that we have published today our press release, our financial statements and a slide show to which we will refer. The documents are available on the Legrand website. Please also note that this conference call is recorded and webcasted on our website. So I will start with a few opening remarks. Following which, Franck and I will comment into more details our results. Then we will conclude with the anticipated trends for the fourth quarter of 2020. I'll begin on Page 4 of the deck with the 2 key takeaways of today's release. First, Legrand recorded good showings in Q3 2020 and overall good performance since the beginning of the year, which overall demonstrates once again how solid Legrand fundamentals are. Second takeaway, since the beginning of the year, Legrand has actively pursued the deployment of its model through -- many initiatives, sorry. So let's now move to Page 6 and start with an overview of sales. So on the backdrop of an unprecedented crisis, sales retreated in total by minus 8%. This trend resulted from an organic decline of minus 10% with more particularly a stable plus 0.1% in Q3 alone. This was completed by the increase in the scope of consolidation with a positive impact of plus 3.7% while the impact of exchange rate was negative by minus 1.5%. This is, of course, for the 9 months. Looking now for the end of the year and taking into account acquisitions already completed and their likely date of consolidation, the scope of consolidation should come to around plus 3.5% in fiscal year 2020. Also, if we apply the exchange rate of the months of October 2020 to Q4 2020, the theoretical impact of FX fluctuations should come to about minus 2.5% for 2020 sales as a whole. This was for the total group. Let me now go into more details regarding the like-for-like evolution of sales by geographical zone. Please refer to Pages 7 to 9 of the slide show. So in Europe, organic sales were down minus 10.5% in the first 9 months of 2020. In Europe's major countries, sales declined by minus 13.1%, while strict lockdown measures took a toll in Q2. Sales then rose plus 2.2% in Q3 alone. But, in particular, by the resumption of projects suspended and the success of many commercial initiatives, this was for Europe's mature countries. As far as Europe's new economies are concerned, the 9 months sales were up plus 5.1% and plus 10.8% in Q3 alone. Sales to September showed sustained growth in Turkey and were up slightly in Eastern Europe. I will add that figures of Q4 2019 will be a demanding basis of comparison for Q4 2020 in this area, i.e., in Europe. Let's now move to North and Central America on Page 8. Organic sales retreated by minus 8% in the first 9 months of 2020. In the U.S., sales declined minus 6.8%, including minus 1.5% in the third quarter alone. Compared with September 30, 2019, the steep rising sales of products for data centers, including busways and PDUs and the strong performance of residential business, in particular user interfaces and AV infrastructure solutions were not enough to offset retreats in other areas. Let me now move to the last zone on Page 9 with rest of the world. Sales were down minus 13.1% organically in the first 9 months of 2020. In Asia Pacific, sales retreated minus 10.4%. This resulted from marked declines in many countries including in India. On the other hand, there was limited fall in China and a rise in Australia. In Q3 alone for Asia Pacific, sales rose plus 1.9%, driven, in particular, by good showings in China and Australia, which offset declines elsewhere including in India. In South America, net sales were down in many countries, declining minus 19.8%, all 9 months, and then rose slightly by plus 0.8% in the third quarter alone. In Africa and the Middle East, sale to September were down minus 14.8% and fell minus 6.3% in Q3 alone. These were the main comments on sales. Let me now pass the mic to Franck for more colors on our solid financial performance.
Thank you, Benoît, and good morning to all of you. I'll start with profitability on Page 10. On this slide, you can see the adjusted operating margin before acquisitions, meaning at 2019 scope of consolidation for both the third quarter and the first 9 months here of 2020. In the third quarter alone, the margin rebounded to 21.6%, up plus 1.4 points on the back of stable organic sales which highlights the good showings of the quarter. This figure was 18.8% in the first 9 months, down only minus 1.6 points, while organic sales retreated at minus 10%, which demonstrates the good resistance of our margin since the beginning of the year. Moving now to Page 11 to focus on margin trends recorded in the first 9 months of 2020. Against the backdrop of a steep decline in sales volume, the adjusted operating margin before acquisition shows good resistance in profitability, reflecting the effectiveness of measures taken in response to the crisis. This was achieved more particularly, thanks to: first, a balanced management of sales and purchase prices; second, a marked reduction, in part temporary, in production cost and in admin and selling expenses; and third, structural adaptations of the organization with, in particular, EUR 55 million in restructuring costs, excluding net gain in building disposal. The dilution on adjusted operating margin coming from the acquisition was modest at the end of September with a slight negative impact of minus 0.1 points. Moving now to net profit on Page 12. It was down minus 21% from 1 year earlier, reaching EUR 493 million. This decline comes from the decrease in operating profit together with an unfavorable trend in net financial results. A positive impact came from the decrease in absolute value of corporate income tax while the income tax rate rose slightly to 29%. Let's share now the last part of the financial performance on Page 13, regarding free cash flow generation and the balance sheet structure. On the left-hand side, 4 points to be stressed. First, you can see that cash flow from generation -- from operations, sorry, came to EUR 781 million or 17.4% of sales, down minus 0.6 points for the first 9 months of 2019. Then working capital requirement stood at 9%, and the ratio of free cash flow to sales was stable at 13.8%. Moreover, in absolute value and in Q3 alone, free cash flow increased by plus 22% from Q3 of last year. Last, the balance sheet remains solid, with notably a net-debt-to-EBITDA ratio of 1.9, equivalent to September 2019. Coming to the right-hand side, normalized free cash flow was up plus 2.2% from 9 months 2019, and it stood at EUR 773 million. These were the main elements I wanted to share with you, good showings of the quarter and good performance shown in our financial key indicators at the end of September. This underlines the efficiency of the measures that we have taken and also the resilience of our model.
Thank you, Franck, for this overview of the financial performance. Let me now move to the second main point of this presentation with the continued deployment of the Legrand model. So Legrand is actively pursuing initiatives to preserve and sustain its development model. The first example of those initiatives with structural moves as depicted on Page 15. These include the continued docking of recently acquired companies as well as the adaptation of cost and organization, for example, through the adjustment of the group's cost base to trends in business, the streamlining of the industrial and supply chain footprint as well as the digitalization of the back office and the front office. I will now give more details on the ongoing innovation momentum in the following pages. Starting with Page 16, Legrand maintained a robust drive for innovating, developing and deploying new products. In the first 9 months of the year, 5% of group sales were dedicated to R&D. On Pages 17 to 19, you will find some examples illustrating how we sustained the stream of new product launches in the first 9 months of 2020. This was the case, in particular, for offers underpinned by structural and long-term favorable trends, for example, linked to the environment, the rise in data flows, the need for safety, the new ways of working or the search for more comfort. Page 17 shows on the left-hand side some of the products launched this year for building energy efficiency, such as Drivia with Netatmo, the industry-leading connected electrical panel, but also Smarther with Netatmo, a connected smart thermostat for energy meters to help measure energy consumption. On the right-hand side, we present new solutions for data centers that contribute to securing the housing of active devices and bring scalability and efficiency with a new structured cabling solutions of the LCS3 program. On the next page, you can see example of safety products such as the Uraone connected emergency lighting as well as lighting fixtures, providing high integrity for critical spaces, such as clean rooms. Also, we launched new residential solutions that facilitated working from home, such as the new WiFi routers and Power over Ethernet switches in North America. On Page 19, Legrand added offerings to its catalog dedicated to comfort in residential spaces with new user interface solutions and sound diffusion systems. We also launched the new Edge Acoustic Architectural lighting solutions for commercial spaces that provide ambient noise absorption. I will conclude with the right-hand side of the page, stressing that Eliot's deployment keeps going on with now smart wiring devices rolled out in 41 countries, meaning 8 more than at the end of 2019. So we covered the 2 main elements of this release. Let's move now to Page 21 to share the anticipated trends for the fourth quarter of 2020. Taking into account a persistently difficult and very uncertain environment due in particular to new health measures in a number of markets, and given the demanding base for comparison recorded in the fourth quarter of 2019, Legrand anticipates an organic decrease in sales in the fourth quarter of 2020. The group is confident in its ability to keep developing its market share and will continue to actively protect its adjusted operating margin. Finally, Legrand is resolutely deploying its CSR roadmap. Well, this concludes our presentation. Franck, Ronan and I are now ready to answer to your questions.
[Operator Instructions] We have first question from Gael de-Bray from Deutsche Bank.
Can I have 2 questions, please? The first one is about the prospects of -- well, it's not even prospects anymore, but the tightening of restrictions in Europe again. Is there any lesson from the spring lockdown in terms of how you're getting prepared for the coming months? That's question number one. The second question is I'd like to get your views on M&A and product innovation going forward. I mean in the current context with the development of work from home, how could you try and skew the business more towards data centers, more towards residential and possibly less nonresidential?
So on the first question, a couple of comments. Number one, we have to recognize that, of course, the tightening of sanitary measures and the implementation of lockdown, curfew and other measures in Europe is not a good news, obviously. And this is one of the reasons why -- and there are a couple of others, this will impact the Q4 going forward. This being said, a couple of things. Number one, we have to recognize also that the restrictions are not yet to the level they were in April -- I mean in March, April and May in Europe. I think there is a better balance, which is currently being implemented between controlling the epidemic and the impact it could have on the economy. So for example, France, it has been officially stated by the authorities that the building industry was to remain open. And actually, most of our customers being either professional distributors or DIY distributors as well as contractors have said that they would continue to operate. Schools are open. So as a result, people have less constraints to stay at home to keep their children. Factories are open. So it's not as tight as it was in April so far. Now, of course, in the weeks to come, it will probably depend on the state of the epidemic. As far as Legrand is concerned, we are also in a different -- we are obviously more prepared. Since April has been implemented in all of our factories, logistics centers, offices, a set of very strict measures, of course, the masks, [ hydroelectric gels ], organization of spaces, organization of remote or smart working. So we now have the processes, the tools to face this epidemic, not to mention the fact that our people do also have -- are acting differently and have a different attitude with the epidemic. And everybody got, unfortunately, accustomed to the virus and there's less fear, for example, to come to work than there used to be in April. So I think, number one, it's not a good news, and it will, anyway, impact the confidence, the economy and many things. Number two, with this being said, lockdown are a bit different than the ones which we implemented in April, and the reaction of the company and of the people, we are more prepared than we were. As far as M&A and product innovations are concerned, it has always been our approach to anticipate a number of trends and to shift part of our R&D expenses or CapEx expenses to products which would be supported by interesting trends and to acquire complementary companies that would help us to become more sizable in the trends which we believe would continue to grow. So we will continue to do that. Take, for example, the data center that you mentioned. We acquired the past years 11 companies and data centers in the U.S., of course, but also in the Netherlands, but also in India, but also in China and in a couple of other countries. Of course, we will continue to do that. Now, I would not like you to think that in the interesting trends in the so-called the next normal, new normal, you only find data centers and working from home. Security, for example, is a trend that will remain present for a very long time. Assisted living and all what we can do to help our old people to stay safely at home is also a very important trend. Energy efficiency and everything which helps regulating the temperature and reducing CO2 emission is a very important trend also. So yes, we are looking carefully at all the trends and making sure that whether organically or inorganically, we do whatever it takes to answer those trends. And being more specific on R&D, as I said during the presentation, you will have noticed that even though we have done a number of cost-cutting measures to cope with the crisis, we have maintained a very significant R&D effort. So R&D, let me give you a number of -- the total number of people or staff for Legrand went down by 9%, like-for-like. So significant reduction in the number of people in order to cope with the crisis. As far as R&D is concerned, our R&D teams were almost stable in the first 9 months of the year compared to last year. So we have really kept our R&D capabilities untouched. I also mentioned during my presentation that our R&D ratio to sales was 5% in the first 9 months of the year. So it's still a very significant ratio of our R&D to sales. As far as M&A is concerned, it is true indeed that we haven't done any deal since Focal Point in the U.S. in February. But I can tell you that we maintain a very close relationship with a number of owners, and we are still keeping alive our pipeline. So to make long story short, yes, we are tracking all the trends, which goes much beyond working from home and data centers. And we remain very active in R&D and prepared in M&A.
Next question from Andre Kukhnin from Crédit Suisse.
Can I just start with a couple of technicalities first? Could you quantify any potential impact from less holidays being taken in Europe and I think in the U.S. as well during this quarter during kind of traditional holiday months of July and August, if that's possible at all? And also in the same vein, in terms of kind of stocking decisions by distribution, was there an impact in Q3? And if you could help us quantifying that, please.
Well, I don't believe that there was such a change in the number of holidays being taken in France and in the U.S. or elsewhere in July and August, which would have a very significant and material impact on our P&L. Our personnel cost is going down by 7% year-to-date. And it's a mix of many things. It's a mix of more days off being taken in France, for example, during the lockdown, and we have asked our people to take a number of days off in order to do savings, but it is also the result of the cut in headcount. I was telling Gael that our head count went down by 9% like-for-like in the first 9 months of the year. It could also be the result in view of that country of number of pay cuts, and a number of countries have decided, for example, to do voluntary pay cuts. So all that, a big part, if I may say, explain the fact that we have personnel costs being down by 7% in the first 9 months. But it's a bit difficult to identify each of the various impacts. Also, more as -- all those positive impacts were also sometimes compensated by the negative ones. If you take countries like India and China, for example, during the lockdown, so with sometimes no sales. I remind you that the sales in China went down by 50% in Q1, and sales in India were almost 0 in April. Still by the law, we had to continue to pay the people and, of course, the surcharges. But all in all, if I may say, taking positive and negative impact, the cost of personnel went down by 7% in the first 9 months of the year. As far as the distributors are concerned, there was very strong -- and I have to admit, they've probably never seen destocking in Q2, which is a very logical decision from our distributors, given the uncertainty and the low level of demand in many markets in April and May. There was probably somehow some sort of restocking in a number of countries in Q3. I don't believe this was a major impact. It's really difficult for us to quantify because we don't always have the sell-out measures everywhere. So there was probably a bit of restocking in a few markets. But again, I don't see restocking as the main reason for a good performance in Q3. Going forward and looking forward, it is always a big challenge. Nobody knows what the distributor is going to do in Q4. And again, purely their call, not our call. Well, it is not completely illogical to think that they could continue to optimize their inventory level in light of the lockdown and in light of the uncertainty. Now we'll see. We'll see what will happen from now to December. So strong destocking in Q2, somehow a bit of restocking in Q3. But again, not that material and difficult to quantify. For Q4, we'll see. But it is, again, one of those uncertainties that could weigh on the fourth quarter.
Thank you, Benoît. And I think I apologize I slightly misled you with the first question. What I was wondering about was the impact on revenues for you in Europe and North America in July and August. From what we hear is construction industry taking less holidays to catch up post the lockdowns. And less holidays in France...
Andre, sorry but I didn't get to your question earlier.
Yes. I apologize.
Well, it is a fact indeed that post-lockdown -- and it happened in Europe but also in the U.S., post-lockdown, there were 2 phenomenons, which positively impacted our top line. In other words, there was a sort of catch-up of all the projects that were started prior to the lockdown and that were only completed once the people could get back to the field and then finish the work. So it clearly positively impacted our top line starting from June, let's say. Second impact, there was a sort of a strong appetite of people to invest back some of the money into their home. And for example, consumer markets and DIYs. DIY shops did very well. After the end of the lockdown, people just wanted to buy switches and tech trackers and connected products because they have experienced during the lockdown how difficult it was to have one room, which was not renovated enough for them to remote work, for example. So this also had a little positive impact in the U.S. and in Europe and actually in many countries. Now quantifying all that and relating that to the fact that people took less days off is a bit difficult. I think it's more coming from the fact that some works were stopped and frozen because of the lockdown and the appetite from consumer to invest money into the home. But clearly, yes, it had a positive impact, and it supported our Q3 top line. Will it be sustainable enough to impact Q4, question mark? And it's also one of the reason why we will see this Q3 as a bit boosted by that, this catch-up effect, for example, will not last forever. So again, this was positive in Q3 and question mark for Q4.
Got it. And may I just ask lastly, just conceptually, in an environment where your top line is down somewhere around kind of low single-digit to mid-single digits compared to 2019 or kind of pre-COVID levels, would you be comfortable to say that you can sustain profitability at kind of those levels, at least given the track record of Q3 and given the structural cost measures you've taken?
Well, you have to -- I have 2 answers to that. Short-term answer, which is -- you can look back at what we did. Q2 drop in sales of 22-point-something percent. And our EBIT margin was slightly higher, 15%. Q3, significant leverage from top line, and our margin expanded nicely. Now the key question mark for Q4 is, of course, the top line. And we clearly stated in the press release was that we expected the organic sales to go down in Q4. And there are, of course, several reasons for that. The fact that number one, the lockdown measures and curfew and also sanitary measures, even if not as tight as in Q2, will have an impact on the market. It will have an impact on the level of confidence, on the ability of people to work on the field and so on. And actually, looking at the IMF measures -- IMF, let's say, survey, they expect Q4 to be more difficult than Q3 in terms of economy. Number two, you have a number of one-offs, if I may say, which positively impacted Q3 and might not repeat in Q4. Take, for example, this catch-up effect from all the work that were started before the lockdown. It can last a few months but not a few quarters. And number three, on top of that, we have a basis for comparison, which is demanding for Europe. I remind you that Q4 2019 was about 2 points higher in terms of top line growth than the full year 2019 in Europe. So we have -- so to give you numbers, we were growing in Q4 2019 in Europe by 5.1% like-for-like, and the full year 2019 was 3.3%. And in the U.S., the base of comparison is tough compared to our peers because most of the companies went down in sales in Q4, and we were up. So for all those good reasons, big uncertainty on the top line for Q4. And as I said, we expect organic growth to -- or organic sales to decline. In terms of profitability, you have with Q1, Q2, Q3, a good sort of example of what we can do. Midterm, this is a different story. Midterm, we haven't canceled our mid-term model even though we have suspended our 2020 guidance in March, but we haven't suspended our midterm model, which, as you know, stayed for, on average and excluding major economic downturns, an EBIT margin of about 20%. So we are still convinced that this is a model, which is -- which we can run and which is -- which is working.
Next question from Lucie Carrier from Morgan Stanley.
I have 3 questions. I will go one at a time. The first one is, it's a bit of a follow-up on the current discussion you've had with Andre. Can you maybe help us quantify how much cost savings you think you have annualized from your structural measures since the beginning of the year? And I think in your previous answer, you were also mentioning that some of the countries have taken voluntary pay cut, which I'm assuming are not going to continue. So I guess what I'm trying to know is how much savings have helped maybe the third quarter? And how much of that is structural versus what is a bit more maybe temporary considering the current condition?
Well, Lucie, it's really difficult to quantify. I can take an example. The fact is that since February 2020, our travel expenses has been going down very significantly. And because of the lockdown measures, it's very difficult to travel from one country to another, and we have replaced a number of travels by, as you do, I guess, also by online trainings, online calls with customers and so on and so forth. How much of that is sustainable? So in the so-called next normal, how much will we be able to convert from physical to digital? It is extremely difficult to say. So I cannot draw any lines between what is sort of one-off or which will not last forever and savings, which are a bit more structural. What I can tell you is that 3 things. Number one, our total cost in the first 9 months of the year, so including production expenses and SG&A, our total cost went down by 8%, for a top line going down by 10%. So we have had the ability to decrease our cost quite nicely. Number two, I believe that there are a number of habits that will last. We will travel probably less than before. We will do more customer calls using the digital tools. We will do more customer visit using digital tools. We will switch more advertising from offline to online. So a number of those habits will remain. While there will also be a number of additional costs that will also remain, I don't believe that we will be able to get rid of the mask, for example, or sanitary measures, plexiglass and so on before quite a long time. Third element, as you could see in the press release, we have booked in the first 9 months of 2020 restructuring expenses of EUR 55 million. You remember that an average year for Legrand as far as restructuring is about EUR 20 million to EUR 25 million. So in 9 months' time, we have booked, let's say, between 2 to 3x what we usually book on a yearly basis. The payback of those restructuring, it differs from measure to measure. In the U.S., you have, for example, a very happy payback. And it partially explains a very good performance in terms of margin of Legrand North and South America. In Europe, it is true that the payback could be longer, it could be 2, 2 or 3, even sometimes 4 years. But this is structural. So even though I'm not able to quantify, to draw a line between structural and less structural, I can tell you that we are working on both fronts. And at the end, expense is down 8% and restructuring expenses amounting EUR 55 million.
My second question, I was hoping you could comment maybe across your portfolio of how much growth or decline you have seen in your residential business versus your non-residential business in the quarter because you were mentioning, obviously, that the DIY business has been performing quite well.
I think it really depends on the geography. And there are some countries in Q3 which rebounded pretty nicely, and this rebound was not only in data centers, in resi, but also in the number of the non-resi activities. So the main driver especially in many European and rest of the world countries was more, let's say, the country situation even more than the state of this or that specific vertical. For this being said, there are a couple of verticals that did pretty nicely all across the world. Connected product is better than the rest of the group. We will comment on that more in our full year numbers. But in the first 9 months, our Eliot program is better than the rest. Residential and especially DIY and retail is better than the rest. Data center did better than the rest. Even assisted living, better than the rest. And of course, by definition, the other verticals or product family did slightly worse than the rest. But again, I think the important driver was even more the state of the country and the economy and the state of the -- stages of the works being performed even more than verticals. I also have to mention that the Q3 performance, I don't know if it was clear enough reading from the press release, is also coming from a number of initiatives that we have done. It is a fact, indeed, that in Q2, a number of product launches were frozen because you don't launch a product when you have your customers back at home. But after the lockdown period, we have launched a number of new products. So we insisted a lot on that in the press release and the presentation because it was, I believe, material to our performance and a good way to reconnect to our customer. Take, for example, Drivia with Netatmo, which was mentioned in the press release. It is a very, very interesting connected cabinet. We started to launch it in France and Italy in September. We have good results, and we are progressively rolling that out to a number of countries. We pursue the launch of connected wiring devices, connected thermostat and a few other products. So to make long story short, number one, don't forget the country situation, which is as important as vertical. Number two, this being said, it is true indeed, this number of verticals which did better than the others. And number three, we supported all that by the number of product launches. We didn't give up on launching new products, even though the economic situation is not as good as we would like it to be.
And my last question, if I may, is around the EU green deal on the recovery plan for building renovation. I just wanted to have your view in terms of now that we have a bit more details on that plan that was published last month. How long do you think this is going to be kind of hitting the different countries and ultimately hitting your P&L in terms of potential benefits? And what do you think about the target that were provided? Because 1.2% of renovation of the building fleet in Europe for 2023 to 2025 is not a big uplift versus the 1% that were renovated in 2019. So what's your view on that for Legrand?
So a couple of comments. Number one, even before talking about the green deal, the good news is that in a number of countries, you have had some measures that have been taken locally, specifically on energy efficiency in order to support the activity. Take, for example, a country like France, you had an interesting initiative called [ Matrim Renove ], which is providing some subsidies to a number of consumer regardless of their wealth in order for them to invest into energy saving-related products. So it is true indeed that a lot of those subsidies are granted for insulation or from heating system. But part of -- they're also available for what we call active energy efficiency products, such as thermostat, for example. This was for local initiatives. As far as the green deal is concerned, the intent is obviously very good, and we believe that the European Union is committed to that. Well, our estimate is that if you really want to reach carbon neutrality by 2050 and be consistent with the 1.5% deal reduction, you would have to move from 1% of the buildings renovated to 3% of the buildings renovated. So it's a big, big uplift. This being said, the objectives of European Union remains ambitious and interesting. The key question mark, and I have to admit that we don't have the answer, is how long it would take to flow into the economy. It is, of course, a good signal sent to market operators. But there are a lot of question marks that remain to be answered. And actually, it is not yet fully voted. The way it will be applied is not yet completely decided. So how long will it take to flow into local regulations? How much of the incentives will be granted to active solutions, so the one in which we operate against the passive solution? Will the incentive be good enough for, really, people to implement them? What will be the reaction of public operators? And we know that state and local governments in many countries and not only in France are real estate owners and have the ability to do significant innovation, and they want to do so with maybe a shorter decision-making process than private owners. So all those questions remain still unclear. We have to admit it. And it's sometimes difficult to get full clarity on all those topics. My assumption is that it will take some time before it flow into the local regulations and into business. So it's not a matter of months or quarters, it's a matter of years. The good thing, if I may say that whether for residential buildings or for commercial buildings, we have the appropriate offer, and we are working hard, especially in R&D, to make it even more relevant to tackle the topics. In residential buildings, for example, we have solutions that allow to do significant 30%, 40%, 50% reduction in energy consumption. Smart cabinet, the smart wiring devices, smart thermostat, so on and so forth. And in commercial buildings, we are targeting more, let's say, the mid-sized commercial building, not the big ones. Still, the market is extremely interesting because I remind you that buildings below 1,500 square meters in Europe represent 80% of the buildings. So it's typically small shops, small office buildings, museum, showrooms and a number of other buildings. And for those buildings, we do have new lighting control solutions, temperature control solutions, a high-efficiency UPS, which are efficient solutions. So not a clear answer to your question, Lucie, I'm sorry for that. But it's still unclear, I think, to many operators. The good thing is that whatever happens and -- we are ready in terms of product offerings.
Next question from William Mackie from Kepler Cheuvreux.
I'd like to ask some questions about the regional trends. You've mentioned the rebound in certain verticals in Q3. Can we spend a little time talking about the importance of how the countries have behaved? I mean, I think specifically, could you give us an idea of the evolution that you've seen through this year, Q1, 2, 3 in France or Italy specifically within Europe, which have been or appear to have been big swings? And then when we come down to the more general question across your portfolio, can you update us on the classic commentary regarding pricing on your product catalog? But then also when we think about the challenges you've had with trade wars and supply chains over the last 18 months, can you perhaps talk a little about what you've observed in the evolution of the direct costs associated with your purchase costs, the inflationary costs in the cost base, excluding the employment and indirect costs you mentioned?
Okay. Look, as far as regional trends are concerned, let me first start with the group numbers, and I will then focus more specifically on the countries you have mentioned. So the month of April was down. So it was really the worst both in terms of the pandemic and the lockdown measures. We were down 41%. And then May and June, we were down, I think, 14%, 1-4. And then Q3, as we released this morning, we are stable, plus 0.1%. So obviously, there is a clear change in trends, even if -- and I have to remind that clearly to everybody, even if we expect Q4 to be down again in sales. By how much? We don't know. It will depend on many things, including the tightness of the sanitary measures being put in place, the behavior of our distributors and many other things. So we don't believe that the Q3 performance is repeatable in Q4. But looking at the April, I mean Q2 and Q3, this trend was clear. It was the same in France, Italy and more generally in Southern Europe, even though it was -- the situation was even worse in April. You have to remind that France, Italy and Spain were at the center of epidemic in Q2, and we have in some of those countries, sales going down by 60%, 70% or more percent in April. So drop in sales, which I have to admit, I've never seen myself in my 23 years at Legrand. Then we have had a similar pattern then for the group, less drop in sales, if I may say, in May and June and a recovery in Q3 , which was not up 10%. It was -- in those countries, low single-digit growth but still appreciable growth. At the end of the period, France, Italy and Spain remain down by more than the group and Northern Europe. If you take those 3 countries, year-to-date, they are down like-for-like, minus 15.3%, whereas the total Europe is down by 10.5%. The total group is down by 10%. So even though Q3 was -- did show some sort of catch-up, at the end, this catch-up was not enough to offset the very strong drop in sales which we experienced in, let's say, end of March, April and beginning of May, so during the lockdown period. Well, going forward, it's always the same as the same topic. We will see, even though you have understood that because of the lockdown measures and the number of other things, we believe that Q4 is going to be very difficult. As far as your pricing/raw material and component questions are concerned, so in the first 9 months of the year, our price -- prices were up 0.9%, so 0.9%, and inflation of raw materials and components was down minus 2.2%. So maybe I have to explain what happened in Q3 because for those of you who will do the math, you will see that the pricing was in Q3 was pretty consistent with the one we had in H1. But the inflation of raw material components was down more than you could have expected. So to give you the numbers, in Q3 alone, price was up 0.8%, and inflation of raw materials and components was about minus 3.1%. We were a bit helped in Q3 by the U.S. situation, where we benefited from refund and exemption on the U.S. tariff imposed on some of the products imported from China. Without those refunds and exemptions, the minus 3.1%, which we had in Q3, would have been minus 1%. So still we would have had a benefit from the difference between selling price and purchase price, but the benefit would have been lower than the one we recorded. And for the 9 months of the year, without the refund and exemptions, the minus 2.2%, which I mentioned, would have been minus 1.6 points. Needless to say, those refund and exemptions are temporary measures. They have a temporary effect on Q3. In fact, of course, the whole of the year -- I mean the first 9 months, but don't expect them to be repeated in the quarters to come. Then last, have we had any significant issue in our supply chain because of the COVID-19 epidemic? No, it was the case indeed in the very beginning of the year because of the closing of the Chinese factories. Now the Chinese factory, not only ours, but the one of our suppliers and subcontractors are fully up and running. So we do not have any supply issues for products or complaints coming from China. And I also have to mention that despite the various, let's say, lockdown 2.0 measures or curfews being implemented, 100% of our factories are working, 100% of our offices are working even though a number of people are remotely working. So no, we don't have any supply chain issues. The question mark in Q4 is more a demand topic than a supply chain topic.
Can I call out one last question with respect to...
Yes, sure, sure, please.
With respect to the rest of the world region, the profitability there was considerably better than I expected. Were there one or 2 countries or one or 2 specific reasons that seem to enable you to report a very strong level of profitability in rest of the world?
Well, looking at the numbers, it is true indeed that the rest -- maybe, I give you the numbers to make sure that we have our numbers clear. So in the rest of the world zone, the margin -- EBIT margin before acquisitions was up in the first 9 months of the year by 240 bps. Excluding acquisitions, this would have been plus 230 bps. So negligible impact of the acquisition, if I may say. Well, it's coming from, let's say, 3 things. Number one, don't forget that we have very different profitabilities between countries. As we've always said, profitability depends very much on the market share we hold on a given country. And in this area, you have countries with 5% EBIT and you have countries with 25% or 30% EBIT. So the total, let's say, profitability of the zone depends very much on the mix of countries and which ones are growing faster or decreasing less than the others. Number two, we have implemented a number of cost-saving measures, and this is an area where those cost-saving measures can be implemented faster than elsewhere because of the local regulations. And on top of that, we have had a good management of our pricing. Third, don't forget that we had a significant onetime gain of asset disposal recorded in Q1 because you remember that we sold a building in Chile in Q1, which was about EUR 15 million of capital gain. And of course, on a zone, which is about 20% of group sales, it has a very significant impact. So you have to take this one-off into account. It probably explain, let's say, half of the increase in margin that we -- that we recorded in the rest of the world area. Last, maybe just to make sure there's no misunderstanding, the profitability of this area should really be looked at the EBIT level, not at the gross margin level because the gross margin is influenced by inter-company sales, which are very significant from this zone to other zones.
Thank you very much. Great results. Congratulations.
Next question from Alasdair Leslie from Societe Generale.
So it was just on the Capital Markets Day last year, you discussed the scope to kind of optimize the back-office operations. For the data centers, you were talking about rationalizing the product platforms, 8 to 4, localizing the supply chains. I was just wondering if that was one area where you've maybe had a jump-start on structural cost adaptation, whether you can maybe give us an update on how that's progressed. And are you launching similar programs just in respect to your comments around structural takeout? Are you kind of launching similar programs for areas like lighting and audiovisual as well given those are the kind of areas where you've been particularly active from an M&A perspective as well in recent years?
Well, we have many programs, as you can imagine, going on to optimize our cost structure. It's not much about platforming because platforming, of course, it's a -- those are projects which are going on, and those are mid-term projects. You don't, let's say, create a significant platform, which you could leverage within a few months. Amongst the programs we've been launching to optimize our cost base, as for example, the optimization of our logistics and industrial footprint, we have -- to give you a number, we have closed or announced the closing of 14 sites since the beginning of the year. The usual pace, because you know that we are constantly doing, of course, optimization -- the usual pace for Legrand is between 3 to 4 sites a year, and we have closed or announced 14 sites. In many countries, including actually in France, in Germany, in China, in Russia, in the U.S., in Brazil, in Turkey, so this is more material, I think, to our performance, and it explains also the reason why we have such an important restructuring charges more than platforming. And on top of that, we are doing a number of things. This kind of crisis is already opportunity to reshuffle a bit the organization to make sure that we have the best person for the best position, to look again at our remuneration strategy, to look again at our process and our organization, to make sure that we are not becoming too heavy in some places, to look again at our R&D program. So it's always a good opportunity to have a fresh view of many, many different topics. Now again, I wouldn't like you to conclude that midterm, we'll go from 20% EBIT margin to 22%. A number of those initiatives do have a long payback. Some other will be -- some of them will be compensated by additional costs. And again, the sort of midterm model that we have to go remain in this horizon of about 20% on average EBIT margin.
Okay. And maybe just a follow-up. Is it fair to assume that maybe the majority of the benefits from the EUR 55 million of restructuring will accrue in 2021? Is that kind of how you look at it?
No, no, no. Some of these benefits is already in 2020. When you launch a restructuring plan in the U.S. and when you have 40 -- 400 people leaving your organization, you booked the charge in April, and you have -- you start the benefit from the next month. It's very, very quick and fast payback just because this is the way the U.S. labor law is structured. And typically, for the U.S. and you may have noticed that excluding other expenses, the profitability in the U.S., so excluding acquisitions and excluding other expenses, we have maintained the same EBIT level in the U.S. in the first 9 months of the year compared to last year. So no degradation in profitability. Well, this is clearly coming from the fact that not only, of course, we have very capable teams in the U.S. But on top of that, you have the ability in the U.S. to do restructuring a lot faster than in Europe, and you have the benefits of that, to restructure in Europe. So no, typically for Europe or for the U.S., for example, a lot of the benefit already started to impact on P&L, or P&L late Q2. Some other will, of course, have some impact in '21, '22, '23, all other things being equal.
Great. next question from Andreas Willi from JPMorgan.
I have a follow-up question to an earlier discussion on the component costs and the question on the U.S. lighting business. If I start with the U.S. lighting business, maybe you could talk a bit what you see in that market in terms of demand? And also, whether you have looked at whether UVC lighting for disinfection and so on would be an interesting opportunity, whether you're offering something in that area or considering to do that? And the question on the component and raw materials, maybe you could give us an updated number, what percent of sales that cost bracket accounts for, so we can kind of calculate the temporary benefit you had in Q3 from the tariff reduction you gave earlier.
Yes. So as far as the question on lighting. So I didn't get which company you were mentioning actually. I can comment, of course, on the -- our U.S. lighting, but you asked whether we could be interested by which product or which company?
No. To use ultraviolet lights for disinfection, upper airflow disinfection in offices. Other lighting companies have introduced products, solutions based on that, whether you're in that field as well and whether you think this is an attractive opportunity or not?
Okay. Yes. So as far as the U.S. lighting is concerned, so let me remind you -- maybe how our performance did break down in the U.S. for the first 9 months of the year. Data center up at high single digit. Residential business, so not only wiring devices, but the residential piece of Milestone, typically, so mounts for the residential buildings, was up single digit. All other businesses were down about double digit, especially those related to commercial buildings being -- not being data center. So this is the case for the commercial part of Milestone and the AV piece. This is a case for Cablofil and Wiremold, so wire mesh and cable management solutions. This was the case for our shading business, for example, and this was also the case for U.S. lighting businesses. So we haven't seen significant difference between, if I may say, our U.S. lighting business, which I have to remind you, is mostly geared at commercial buildings. We have almost no lighting for residential compared to the other non-residential businesses, excluding, of course, those related to data centers. As far as products, allowing disinfections are concerned, we do have with our Kenall, Kenall is a specialty lighting company we bought 2 years ago. So specialists of lighting fixtures for demanding environments such as hospitals, tunnels, the prisons and stuff like that. We have within our Kenall offering dedicated lighting fixtures with some treatments that is purifying or clearing the air. Well, it's still a small business. I don't expect this very well business to -- you don't have such an impact that it will suddenly, a lower lighting business to grow significantly in the U.S. So yes, it's an interesting piece. Still a very small piece of our lighting business. Now as far as all products, which could be potentially used to, for example, disinfect or kill the virus, we are, of course, looking at the solutions very carefully. Well, we also have to be careful because some of those solutions could do more harm than good. If you take, for example, UVC technology, not sure if it's a good treatment for the skin nor for the good within the building. So we have to do that carefully, of course. But again, to win that share, we do have an interesting offer, but which will not, by itself, compensate the market impact on our lighting sales. As far as the COGS are concerned, well, raw materials represent, as you know, 33%, so 1/3 of our sales. So -- and maybe to remind the numbers which I've mentioned, purchasing price all in, so including the tariff exemptions and acquisitions was minus 3.1% in Q3. And if we exclude the positive impact from tariff, it was minus 1% and not minus 3.1%. Does it answer your question, or do you want more granularity on our costs?
No, that was very clear. Maybe on the pricing, obviously, you have a good backdrop at the moment with basically falling costs and a good environment for pricing, given also there were component shortages and people were restocking and so on. How long this -- do you think this will last, this environment? Are inventories now back to more or less normal of your products at distributors? Or do you still expect a little bit more of kind of a restock demand in Q4?
Well, so the pricing -- so 2 questions, pricing and stocking. Well, the pricing we had experienced in the first 9 months or so, so plus 0.9% in positive price is very standout in the Legrand history. We did record a lot more last year in 2019. It was plus 1.8%, but this was influenced by the tariff of the goods imported from China. I recall you that we had more than EUR 40 million of additional costs to finance because of the tariff. But having a slight increase in sales here -- on pricing, sorry, year-on-year, it has been part of the Legrand story forever. So we don't intend -- we don't think there will be a significant change in trend neither in Q4 nor next year. I mean this is part of the story, and this is, I think, the know-how that we are able to implement. As far as the stocking or destocking is concerned, it's extremely difficult to have a complete view at our distributors for many reasons. Number one, it's their call, not ours. Number two, we have many distributors. We always mentioned Rexel, which, of course, is a leading company in Australia. But if you look at the global level, you have many, many distributors. Number three, we have many catalog numbers, many SKUs. In a country like France alone, we will have 20,000 SKUs, which are distributed. So having a complete, let's say, view at what our distributors could do, it's extremely difficult. Well, we don't believe that our -- so it's -- let's say, it's a feeling. I don't believe that our distributors do have a very, very low level of inventory. And I would not be surprised, but it's not information that I have, but I wouldn't be surprised if they were to do a bit more destocking in Q4, for example. It could very much happen depending on their views about the start of 2021 and their willingness to optimize their working capital. I don't know. It's a big question mark. What I'm pretty sure about is that they don't have an obvious lack of products given the, let's say, state of the demand.
Next question from Martin Wilkie from Citi.
It's Martin from Citi. Just a question coming back to end markets. You've obviously seen some strength in residential and areas like that. But on commercial, I appreciate a lot of projects that were perhaps paused during the second quarter kicked back off again in the third quarter. I know you obviously don't have your own order backlog. But in terms of new commercial projects starting, there obviously are some fears that office, retail i.e. other parts of commercial could see some weakness next year given the sort of lockdown trends, work-from-home, these kind of things. What are you looking at in terms of how you look at commercial going into 2021? Do you get involved in tendering and sort of pricing that gives you some sort of sense? Or is your visibility really just sort of limited to perhaps a quarter or so? Just understand how you look at commercial into next year?
There's absolutely no leading indicator that will give us a good feel about what the market is going to be in 2021, neither actually for commercial nor for residential. So we are completely blind, and this is not new to the Board. This is the nature of our business, not to have any of those leading indicators. Now it's true that you have -- look, things that could positively weigh on the non-residential market, and you've named one of them, remote working. And the fact that if people start to work 1 or 2 days from home, a bit less space might be needed. By the way, it's not as much as that, except if people are working 5 days per week at home. But except a few companies, this is not the trend most specialists are expecting. They expect remote working to be 1 or 2 days a week, which, by the way, will not reduce that much the need for square feet. But in front of those, let's say, trends or megatrends that could possibly weight a bit of negative impact the non-commercial -- the non-residential space, you have many trends that could also be quite positive. Well, we discussed a bit about the green new deal and the need for energy efficiency. This will lead a number of spaces to be modified and going to be more energy efficient. If you have people working remotely from home, you will need more common spaces to meet, which will have to be equipped with a good AV product, good connectivity products, bandwidth and so on. The -- not to mention, of course, the fact that working from home or buying or selling more online, all this stuff will also lead the data center market to probably grow and more need for bandwidth and capacity. So I wouldn't take for granted that the next normal would have such a negative trend on non-residential. Even though, of course, you can assume that some of those spaces talking, for example, about high-end hotel rooms could be negatively impacted. Maybe I can remind you so that you have the right orders of magnitude, the right -- the breakdown of our sales. So again, with a limit that a lot of the products that we sell are sold in [ housing-type ] of verticals, so new residential represent about 20% of our sales. Renovation for residential represents another 20% of our sales. Data center, it's about 10% of our sales. New commercial building, excluding residential -- excluding data center, it's about 20%. Refurbishment of commercial building, again, excluding data center, it's about 20%. And all other type of buildings, so industrial building, infrastructure and others, represent 10% of our sales. So this should help you, according to your own assumptions, about the market trends to do your own computations for 2021.
Next question from Wasi Rizvi with from RBC.
Just a couple left from me. Firstly, on data centers, could you give us more color on that business? And how much of the U.S. right now -- is now because you've made quite a few acquisitions, as you said, and it's been growing faster? So could you help us quantify how big that is within your North America region and also kind of trends you're seeing and expecting? And if you could maybe split that between hyperscale and other, that would also be helpful if you have that visibility. And then the second one was on restructuring costs. I mean, you highlighted that they've been elevated this year. I mean we don't know what next year looks like, but if we were to assume a return to business as usual, to some extent, should we expect restructuring to be lower than your usual range because of what you've done this year? Or does it go back to the normal level?
Yes. So as far as data center exposure in the U.S., it's slightly more than -- it's about 20% of our U.S. sales. Sorry, it's about 20% of our North and Central America sales. So it's a significant part of our business. It's mostly geared at white room. So we are not much -- as you know, in data center you have white room, which is where you have the servers. And you have, well, what we call a gray space, which is a building infrastructure, where you have the protection, transformers, UPS, circuit breaker and so on. Our position in the U.S. will be mostly geared at white room. And we do have a number of leadership positions in some very critical product lines, PDUs, which are sort of emergency outlet, smart sockets where you connect the servers and which help you monitoring the servers, including heating and so on; busway, where you connect the racks and cabinets, the racks and cabinets themselves, connectivity and a bit of fiber. In terms of type of data centers, we are selling to all type of data centers, even though it is true that a large part of the growth, which we do have in 2020, it's probably coming from what we call the hyperscale data center where the market is literally booming, including with very interesting and very added value customers. So this is for the U.S. Just a word on outside the U.S. Outside of the U.S., our position is slightly different. We have lower exposure to data center because we are more of a newcomer in Europe and in the rest of the world. We have a more balanced presence between a white room and the gray space because we are more significant and bigger player in Europe in the rest of the world, in the circuit breaker, transformer and UPS and so on, than we are in the U.S. And we are currently in the process of rolling out our U.S. offering through our local organizations in Europe and in the rest of the world, leveraging local organization and the nice portfolio of products we have acquired in the U.S. So this is for data centers. Well, as far as restructuring is concerned, it clearly depends on the state of the economy and how much actual restructuring do we need to do in 2020 and what are the prospects we have for 2021 -- sorry, for 2022. It's a bit too early to -- it's a bit too early to comment. That's as far as the restructuring is concerned, focus on Q4 before getting into 2021. So no answer at this stage, no guidance for restructuring. It's a bit too early. We'll probably discuss that in February. It will partially depend on what we can see as far as the market, the economy -- the market and the economy are concerned. It will never come down to 0, by the way, because we always have a number of interesting ideas to the site, do a site consolidation. We are acquiring regularly some companies. So we -- in a reach, if I may say, of sites that we can consolidate and then transform. So definitely not 0, but to give you a number for 2021 is a bit too early.
We have a new question from William Mackie from Kepler Cheuvreux.
Okay. I just wanted to come back to M&A. I think you're flagging that the marketplace for transactions has improved. And can you just help reconcile between your key focus areas for growth with the Legrand Way and where you see opportunities for M&A? I mean do you want to flag up any particular areas? And any sort of size that we should be thinking about with respect to your M&A policy? And how quickly you can develop that M&A policy?
Well, we can resume quickly if we want to because we have retail, the relationship with the contact, and we have enough targets, even though most of them are of modest size. But this is the Legrand model. We have enough targets to be able to pick up the right ones, if you wanted to resume a bit more actively on the strategy. So it will not take quarters or years before we can start again. As for the sectors are concerned, I wouldn't want you to think that we are only focusing on the 3 or 4 sectors, which are resisting more than others to the crisis, data center, residential or assisted living. If there was an interesting leader in -- even in a geography, which suffered a lot from the crisis, take, for example, Southern Europe or Latin America or India, we would definitely be interested. So our focus will not only be on looking at data centers and stuff like that, but it will also be the traditional Legrand strategy of looking at nice leaders in good categories. We discussed lighting, for example, lighting fixtures. We discussed a bit AV. Do I regret the acquisition of Milestone and a couple of lighting acquisitions we've done because the year 2020 is a bit difficult because of COVID-19? No, I don't. I think that long term, those are very interesting acquisitions. They have reinforced a lot the Legrand grip on the U.S. market. They have brought us a leadership position, very nice margin. Very nice cash flow. Those were very, very good acquisitions to make. So we will not only focus on, let's say, faster-growing field of activities or geographies which are resistible to the crisis, but we'll remain very, very focused on looking at the nice leaders. May I also add that the day we will be willing to resume more actively acquisition strategy, we do have, of course, the balance sheet to do so. You may have noticed that our financial leverage, so the ratio between the net debt and EBITDA, is 1.9 at the end of September, which is exactly the same that at the end of September 2019. So it's, let's say, reasonable leverage, which allow us to do deals if we wanted to do so. We have significant cash on our balance sheet, EUR 2.7 billion of cash on our balance sheet. So now, we have no obstacle, if I may say, that would prevent Legrand from resuming its acquisition strategy. And the day we will do so, it will be the tied to Legrand vision. So targeted at small to medium-sized targets, so good leaders in interesting -- good leaders complementary to Legrand.
Next question from Jonathan Mounsey from Exane.
On the cost savings, I imagine when you enacted that EUR 55 million program, you probably thought -- were doubting that the world would recover quite as fast as it has in Q3. While we appreciate some of that may be temporary, pent-up demand from Q2, et cetera, it is interesting, the headcount has come down by 9% this year and now revenue is back to flat. Do you feel that maybe some of that -- those cuts need to be reversed? Or actually, was this an opportunity to restructure the business more aggressively? And in fact, you can still supply the level of revenue you bounced back to in Q3 with this renewed cost base? Or actually, are we going to have to start rehiring, particularly in places maybe like the U.S. going forward, where it will turn out that the cuts were a little too deep and the long-term impact to end markets was not quite as bad as maybe you feared it would be in early Q2?
Well, clearly, some of those cost savings are not sustainable forever and some could even be reversed when you are taking, for example, days off which provides significant savings in a given month or a given quarter. So next year, you, of course, have a base for comparison, which is tougher. So I wouldn't take for granted that all the savings which we have made are sustainable and which should flow into 2021 and 2022 P&L. Some of them are one-off. Again, travels, part of travels, salary cuts and some of them could even be reversed. And again, the horizon of the mid-term model of Legrand is -- remains at 20% EBIT and at 22% EBIT. So I don't know if it answers your question, but...
But just to be -- not exactly. I suppose what I meant was sales are flat year-on-year. Headcount is down 9%. How do you actually...
We have to -- be careful, be careful, be careful. Flat is -- I mean, sales are flat, Q3. And the headcounts are down 9% over 9 months. So the minus 9% of headcount has to be compared to the minus 10% in like-for-like sales. So we have had the ability to decrease our sales. So this is the number of people, almost as much as volume. Again, sales -- but so if you want to be even more precise, the minus 9% in terms of headcount has to be compared with about minus 11% in volume. This is the comparison to be made. And this minus 9% also includes, of course, some direct labor which is somehow proportional to top line. So...
But going forward, the minus 9% has to be compared with the fact that sales are flat now. So do we need to bring people back on if sales are going -- now that we're actually recovering as much as we are. And I appreciate Q4 might be a bit weaker, but surely where we are today is better than where you thought you would be 6 months ago.
So a couple of comments. Number one, and I think I made it clear, but I don't believe that this plus 0.1%, which we recorded in Q3 is sustainable over Q4. So I don't expect our sales to be flat in Q4. I expect our sales to go down in Q4, number one. Number two, part of the minus 9% in headcount is somehow variable. So take, for example, direct factory workers. Today, you do a plus one and you have to hire back a number of people. And that's actually what we did with temporary workers in a number of European countries, for example, in Q3, because in some countries, the demand was up again 2%, 3%, 4%, 5%. And we had to hire back some direct workers, even though we do it mostly with temps because of the lack of visibility. Number three. So number three, you don't lower -- we haven't lowered structurally our cost base by 8%. They are -- should the economy recover, depending on the pace of the recovery, there are a number of people that we will have to hire back in order to lead the group. So I would not take for granted that we have lowered structurally our headcount and our cost base by 8% to 9% and that it will have a mechanical, a very strong impact on profitability going forward. Part of that is not sustainable. Part of that is linked to the decrease in volume. Part of that will have to be reversed. So that the economy will be better.
We do not have any more questions for the moment. [Operator Instructions] It looks like we don't have any more question. Back to you for the conclusion.
Well, thank you very much, ladies and gentlemen, for taking the time to go through our presentation. I will not highlight again what we released this morning. Should you have more questions, please don't hesitate to call Ronan, Samy, Franck or myself. We'll be happy to answer. And I look forward to meeting you and discussing with you in the months to come. Thank you very much.
Thank you, ladies and gentlemen. This concludes today's conference call. Thank you all for your participation. You may now disconnect your lines.