Rémy Cointreau Q2 Sales 25/26
Thursday, 30th October 2025
Transcript produced by Global Lingo London - 020 7870 7100
https://www.global-lingo.com
Rémy Cointreau Q2 Sales 25/26 Operator: Hello and welcome to the Rémy Cointreau Q2 Sales Conference Call. Please note this conference is being recorded. And for the duration of the call, your lines will be on listen-only. However, you will have the opportunity to ask questions after the presentation, and this can be done by pressing Star One on your telephone keypad to register your question. You are kindly asked to limit yourself to two questions only. If you require assistance at any point, please press Star Zero and you'll be connected to an operator. I will now hand you over to Luca Marotta, CFO, to begin today's conference. Thank you. Luca Marotta: Good morning, everyone. Thank you for joining us today. As highlighted in our press release, H1 sales decreased by 4.2% organically. This performance includes a -11% decline in Q2, which should be, as expected, the lowest quarter of the year. It reflects some adverse phasing effect in a still challenging macroeconomic environment. This result stems from mixed regional trends, mainly driven by, on one hand, strong sales growth in the US cognac division for the second consecutive quarter, supported clearly by low comps but also improved sequential depletions. They are definitively improving, depletion, but less than expected, and they are still negative. On the other hand, depressed cognac sales in China affected by an increasingly difficult market and unfavourable calendar effects due to the shift of the Mid-Autumn Festival and some residual disruption in travel retail. China. This effect waving for 0.7 points in Q2 or in H1 1.4 points at group level.Specifically, the Q2 sales decline breaks down as follows. Volume decrease of -4.7% and -6.2 in price mix effects, largely driven by the underperformance of high-end brands cognac and some price adjustments. Now looking into the overall sales performance by region. Americas recorded quarter, plus 12.8% sales growth in H1, including a slight growth in Q2, mostly driven by a solid, robust performance in cognac. In parallel, the lack of spirits division turned negative, affected by adverse phasing in Q2, but following a very strong Q1 and despite a resilient depletion environment.
APAC, Asia-Pacific, sales decreased by 14.8%, with Q2 strongly impacted by China, which is facing tighter market conditions and an unfavourable Mid-Autumn Festival calendar effect. The same time, the rest of Asia generated a mid-teens growth over the quarter. EMEA, big EMEA region, declined by 9.2, posting in Q2 a similar performance to Q1 in an environment, global markets still affected by subdued consumer demand. This was sell in.
Now let's talk H1 value depletion estimation, a group level; so the best approx, what is the final sell out. In the US, value depletion declined by mid to high single digit year on year including a decline of mid-single digit in Q2. So better improving compared to what we recorded in Q1, but still negative. Compared to pre-COVID six years ago, H1 value depletion are down mid to high single digit, but stripping out the VSOP range, they are +40. In China, value depletion were down mid-teens year on year in H1 and up by high teens versus H1 '19-'20. Beyond unfavourable calendar effect in China, this performance is clearly disappointing for us and reflects tougher, more complicated market conditions. in EMEA, value depletion decreased by mid-single digit year on year and they are performing more or less the same level, negative mid-single digit compared to H1 '19-'20.
So what you can say overall in terms of selling sell out equation at group level, we can say that H1 value depletion fell by high single digit year on year, more or less -8, underperforming,
clearly, especially sell-in trends that were a -4.2. Why? Because essentially the US restocking from a very low base without increasing the level of stock with the positive on sell-in even if depletion dynamics were still improving but still negative.
To conclude on this very first slide, considering all that, we have decided to lower our full year organic guidance, and I will come back to the main drivers at the end of the presentation. All in all, we now expect the organic full year sales to be between stable and up low single digits, while we expect organic full year comp to decline low double digit to mid-teens. The later one clearly includes the estimated impact from tariffs in the US and price undertaking in China.
Pages number from 3 to 5, I would like to come back very briefly on the main marketing initiatives of the quarter. Let's start with LOUIS XIII, which is extending its brand universe beyond testing with the launch of its very first Art de la Table collections. The initiative is fully aligned with our long-term strategy to reinforce the Maison positioning at the very top of luxury, to enrich and complete the consumer journey, and to create new opportunities for differentiation across our freestanding stores, boutiques and key markets. To bring this project to life, LOUIS XIII partnered with chef Alain Passard, a three Michelin star chef at L'Arpège in Paris. This collection was created in partnership with the French porcelain house J.L. Coquet. Each collection consists of six pieces to start with, designed and crafted for by over 40 artisans. They translate two of the Maison founding pillars, Terroir and Time.
From a commercial standpoint, the launch - this launch also play a pivotal role in animating our boutiques, enhancing visibility and driving traffic. Overall, Art de la Table reinforced the brand's cultural equity, expands its experiential ecosystem, and underlines our ability to innovate within the codes of ultimate luxury.
Page number four. I'd like to highlight our last innovation. And it belongs to the brand Cointreau, is entering Cointreau in a new territory this year, with the launch of this very first ready-to-serve spritz in the United States. Why are ready-to-serve rather than ready-to-drink? Because ready-to-serve - ready-to-drink are ready represented close to 12% of the alcohol market. Ready to serve remains a fast-growing niche, today worth more than half $1 billion in the US, and actually outpacing ready-to-drink subcategory in terms of growth and potential. Ready-to-serve cocktails are strongly associated with hosting occasions when consumers want both convenience and the ability to impress. We see clear spikes in sales around the holidays, confirming that ready-to-serve has become a trusted solution for entertaining. Looking ahead, this segment should continue to grow through premium premiumisation, flavourful proposition and ease of consumption. Our objective with this launch is to extend the Cointreau footprint into new daytime occasions to recruit younger consumers who are seeking convenience to modernise even more the image of the brand, and to gain additional shelf visibility in a highly competitive and crowded environment.
The product itself has been carefully designed as a ready-to-share 750ml bottle at 10.5% ABV. It comes with three variants; orange and blood orange, lemon and lime and grapefruit and tangerine. Each recipe is crafted with Quatreauly French white wine, cyos juices, and natural flavours, but without any artificial flavours or colorants. Consumer testing results are excellent. We launched end of September 2nd flavours across nine key US markets with a retail price of 1999. Distribution is already confirmed in more than 1000 points of sales in terms of numeric distribution through major liquor chains, and sell-in is ongoing to support a national progressive iterative rollout, including the third flavour in March 2026.
Lastly, for the marketing and colourful experiences from the business and brands, on page number five. A few comments on a minor point, which is Mid-Autumn Festival in China, 25% in terms of weight of our sales in China. So it's very important one, was clearly joking. During the Mid-Autumn Festival, our priority was to sustain demand and engagement in what is a softer consumption environment even if first results showed that we continue to gain market share. I'll be back on that point, even more in Q&A session. We therefore focus our effort on a few high-impact, cost-efficient activation designed to keep our brand visible and relevant during these key consumption moments.
Starting clearly with the Rémy Martin Club, we celebrated the brand's 40th anniversary with a strong Integrated campaign running from August to October. We leveraged our brand ambassador, Li Xian, across digital, social, live stream channels, building strong reach and engagement at the limited cost. The limited edition design and creative gift with purchase mechanics have stimulated the sales across key markets. So mainland, but not only also Hong Kong and Macau. Campaign generated a lot of press clippings and in a high or immediate return GMV from live stream session alone confirming a good consumer traction despite the muted and depressed environment. At the same time, Rémy Martin XO gained exceptional visibility in Mid-Autumn Festival in China with the launch of Anish Kapoor limited edition, a creative collaboration reinforcing the brand's prestige and desirability. Altogether, this initiative allowed us to maintain strong brand visibility, support our partners during the peak season, and reinforce confidence among distributors. They also demonstrate our ability to stay disciplined and impactful in our marketing investments, keeping our brands aspirational while driving the best we can sell-out efficiency, efficiency during a very and more challenging festival season.
Turning now to slide number six. So turning now to numbers. H1 sales amounted to €489.6 million, representing year-on-year decrease of 44.1 million, or -8.3% on a reported basis. This performance was shaped by the following factors. First of all, an organic decline of €22.3 million, which is -4.2. This performance is split between +2.4 of positive volume effect and -6.6 of price mix. Price mix negative impact results from a slightly negative pricing effect and low- to mid-single digit negative mix effect. Why? This is linked to the underperformance of high-end products inside any given brand, and clearly by the cognac division performance compared to the weighted average.
Second and negative currency translation impact of 21.7% or -4.1% loss, mainly driven by the deterioration of the US dollar, which accounted for -11.3 million, and the Chinese renminbi, which accounted for €7.9 million.
Let's now turn to slide number seven. Delve into organic trends by region. Let's start with the Americas, in which organic sales increased by 12.8% in H1, i.e., down more or less 15% on a six-year basis. This year-on-year performance includes a mid-teens growth in volume and a low single digit negative price mix impact, reflecting an unfavourable mix, first of all, and some adjustments on VSOP. In the US, sales grew by mid-single digit in the quarter, Q2, driven by a strong performance in cognac, linked to a low base of comparison and a continued sequential improvement in value depletion, not as much as expected, but improving nonetheless. In this context, inventory levels in the US remained close more or less to four months at the end of Q2. Canada sales were down mid to high single digit in the quarter, impacted by phasing effects between Q1 and Q2. Why? Because overall, sales were up high single in H1. And LatAm, Latin America, sales were also affected by phasing effect between Q1 and Q2, down strong double
digit in Q2, but up strong double digit in H1. So Canada and LatAm a bit of phasing between them between Q1 and Q2. End of September, Americas big region accounted for 39% of group sales, so increasing five points compared to the previous year, clearly plus 12.8 and top line comparing average performance -4.2.
Turning to APAC, Asia-Pacific. Organic sales declined by 14.8% in H1, but clearly increased by more than 20% on a six-year basis. So on the short term, China is performing and APAC negatively compared to the Americas. But if you compare the two biggest regions to six years ago, the dynamics are reversed. Analysing the volume value equation of APAC, the performance was impacted by high single digit volume decline, while the value part was negative and more than mid-single digit, driven by the underperformance of the high-end brands and more promotional activity. In China, sales were down approximately -25% in Q2, impacted by tighter market conditions, including stricter discipline and austerity measures, which should lead to a global overall as a market soft Mid-Autumn Festival. This performance also reflected an unfavourable calendar effect and some residual disruption in travel retail. So if you compare this technical effect to APAC level only, not a group level, we have 1.5 points negative impact in Q2 and 3 points in H1. This technical effect will normalise from Q3 onwards.
By channel, direct e-commerce was the only growing channel in China, with sales up more than 10% in this quarter, bringing the overall e-commerce ecosystem penetration rate at the end of the H1 at more or less 25%. This was sell-in. In parallel, global value depletion in China were down mid-teens year on year and up 18 versus '19-'20. But once again, during Q&A, we talk about the calendar effect and what is a more normalised performance. Given depletion and in line with sell-in trends in H1, inventory levels in China remained healthy at more or less at the same level as before at the end of September. Compounders are saying the same things in terms of stock coverage. As were in the region, rest of Asia showed a strong improvement compared to Q1, posting a mid-teens sales growth in Q2, led by cognac and to a lesser extent by the Liqueurs & Spirits division with two growth regional engine of this quarter Australia and Japan. End of September, APAC accounted for 39%, so the same weight of the Americas, but in this case is down five points compared to the prior year.
Last but not least, EMEA, in which organic sales were down 9.2 in H1, and more or less were down -10 compared to six years ago, primarily reflecting a negative value effect. Inside that, we need to analyse by subclusters, sub-region within Europe. Let's start with what we call the third-party distributors region, 3PD; recorded a mid to high single digit sales decline in the quarter, impacted by Germany and Greece. In parallel, Czech and Poland, Czech Republic showed good momentum in the quarter. Overall, talking about brands, Metaxa was strongly up and Cointreau gained market share in many markets, but the category is declining, partially offsetting the rest of the portfolio.
Second subcluster, UK and Nordics, sales turned positive at a low single digit in the quarter, showing solid sequential improvement versus Q1, led by Cointreau, Mount Gay and The Botanist. The performance reflected a rebound in cognac versus Q1, which was almost flat in Q2. Rebound still cumulative on the H1 negative quarter. Benelux and France, sales were clearly declining by strong double digit in the quarter, impacted by competitive promotional pressure in cognac and softer trends in liqueurs and spirits. Last but not least, in MI and CIS, sales were down low double digit in Q2, impacted by Rémy VSOP performance while the launch of Rémy DS in South Africa and Nigeria is giving and bearing some promising fruits. In H1, value
depletion, so talking about sell-out, declined the mid-single digit year on year. And on CS basis, excluding Russia, is a more or less the same performance. Overall, inventory levels remain healthy across most areas. And end of September, the EMEA region accounted for 22% of group sales, which is stable compared to the previous year.
Let's now turn to slide number eight and the analysis by division. And we start with the Queen division, which is cognac. Cognac division posted an organic sales decline of -7.6 in H1, driven by a 0.7 increase in volume. So volume cognac were positive, and a negative price mix of 8.4. End of September, cognac accounted for 61% of our sales, down 2.5 points compared to the previous year.
Starting with the APAC, only cognac. In mainland China, sales declined by around -25% in the quarter. So it's the same performance, if you consider the global portfolio or only cognac in China, i.e., a mid-teens decline in H1, but up the more than 60% versus H1 '19-'20. Comps clearly overall over this year were building some blocks that are clearly high. This performance has been affected by tighter market conditions, including stricter discipline austerity measures, which clearly do not allow consumer confidence to recover quickly. In addition, the sharp decline includes an unfavourable calendar effect from the late month, as well as recent travel retail disruption, now on a path during the Q3 on normalization. In this context, an indirect channel remained under cash pressure, while direct e-commerce was the only growing channel turning up, increasing its performance more than 10%. Elsewhere in China, Hong Kong and Taiwan reported weak performance in both sell-in and depletions impacted by same challenges in China. Macau, even from very low figures, was strongly up after clearly favourable phasing, some promotion, but a little more dynamism. Overall, H1 value depletion in China were in line with the sell-in, so down mid-teens. On a six-year basis, this is equivalent to a 20% growth. The remaining part of Asia, sales were up strong double digit in quarter in cognac, led by Australia and Philippines.
In Americas, let's start to dig in with North America. So the combination of US and Canada. Cognac sales were up by mid-teens in the quarter, supported by a low base of comparison and continued sequential improvement in depletion, mostly in volume. Q2 specifically for the US value depletions declined by mid to high single digit, of which down low to mid in volume, mostly driven by Rémy Martin VSOP improvement. Given this factor, cognac inventory coverage still close to four months at the end of Q2; and on 12 months basis, value depletions, so include minus 5 points or negative price mix effect year on year, but on a six-year basis, price mix remains very up at +11 points. Latin America, the remaining pieces of Americas, sales were down for cognac by strong double duty in Q2, impacted by negative phasing. Sales were up by a very strong digit - strong double digit in H1 driven by VSOP and LOUIS XIII. And then the third region inside the Cognac in terms of weight, EMEA, in which sales declined by mid-teens in Q2, affected by very strong and competitive promotional pressure across most major markets and soft demand. UK and Nordics inside that were flat in the quarter, driven by Louis XIII, Rémy 1738 in a category marked by intense promotional activity. Europe third party distributor was down by double digits, mainly due to Germany, where their market remains very soft and highly promotional. And at the end, MI sales declined by mid-teens, even if early results are encouraging for Q3 following the launch and the follow up of VS in South Africa and Nigeria. Last but not least, H1 EMEA value depletion were down for cognac double digit year on year, down very strong double digit versus six year ago.
Now let's talk of the remaining more or less 40% of the sales, because 61 was a cognac in which we have for 37% Liqueurs & Spirits. Slide number nine. Liqueurs & Spirits Division reported a +4.1 organic sales growth in the H1, driven by very solid volume increase at +5.2 and a slight minor negative price mix effect on -1. End of September, Liqueurs & Spirits accounted for 37% of sales, up three points.
And let's now review the division performance by subregion. And let's start with Americas and North Americas, in which sales were down by mid-single digit in the quarter, as were affected by adverse phasing after a very, very dynamic Q1. Sales were up high single digit in H1. Both key brands, Cointreau and The Botanist, delivered solid performance over the semester, supported by resilient depletion, the success of the latest campaign, the recent launch of the ready-to-serve Cointreau citrus spirits.
In parallel, Cointreau and The Botanist Q2 US value depletion, so best across sell-out, were flat year on year. And in terms of price mix was down only one point compared to last year for the 12-month period ended September, and increased 18 points on a six-year basis. In Latin America, sales were down by strong double digits in the quarter in a softer consumer environment for Liqueurs & Spirits.
Second region in terms of weight for Liqueurs & Spirits is EMEA, in which sales decreased by mid-single digits in Q2, so declines affected by all subregions except the UK, while H1 value depletion were up slightly year on year. Breaking down sales down further, the U.K. and Nordics posted a sequential improvement, with sales up by mid-single in Q2, led by Cointreau, which is gaining market share, Mount Gay, which is benefiting also from a lower ABV for version of Eclipse and Botanist, which continues to secure new distribution listing. In Europe third party distribution clusters, in which sales decreased by low to mid-single digit in Q2, impacted by Germany, Greece and Spain. In parallel, Poland, Czech and Czechia, and Romania posted strong momentum. Overall in terms of brands, the growth, the solid growth from Metaxa, Cointreau, and Mount Gay partially offset the impact of the softer consumer environment.
Finally, Benelux and France were strong - were down by strong double digit in Q2, while MI and CIS was down only by mid-single digit. Third region by weight is APAC, in which China sales were up by low double digit in Q2, driven by Brucladio and Botanist. In terms of volume depletion, H1 value depletion were up slightly. And remaining part of Asia was up by single -mid-single digit in Q2, driven by Australia and Japan, strong momentum for Bruichladdich, notably the release of the Octomore series 16 in Japan and the further market share for Cointreau in Australia and New Zealand. This was 37% of our sales Liqueurs & Spirits division. So if you combine 37, Liqueurs & Spirits 61, which is cognac, it doesn't give 100 because we have non-group brands which represent now more or less 1.5% of group sales. Sales were down 0.5 points year on year, 1.5 compared to two, and they recorded a very strong decline in H1 of -35.7%, mostly affected by the negative - sharp negative performance in group brands in Benelux and the UK.
To conclude, let's now turn to slide number ten, which I think is an interesting one. For a few comments on our lower guidance for the full year '25-'26. The deterioration of market conditions in China and the weaker-than-expected sales rebound in the US, even if is positive, but is weaker than expected, have led ourselves to revise our assumptions for the year. For China, we had already adopted cashless assumption for China and were anticipating a slight annual decline. However, market conditions have further tightened following the implementation of the
new alcohol consumption restriction for official while increased, strongly increased promotional activity is also prompting us to show greater flexibility on pricing.
Regarding the US, we still expect the strong growth through the year, supported by the continued ongoing improvement and depletion. But it is clearly not enough to maintain the initial forecast for the US. Consequently, we now expect group organic sales growth to range between flat and low single digit; and before was mid-single digit. In this context, we now expect group's organic comp to decline between low double digit to mid-teens; before was mid-single digit decline. Beyond the revised top line, which is clearly the trigger of the margin deterioration, I'll be back on that point on Q&A, because we have to remember the weight of the top line, more or less 1 billion and the weighted operating profit last year, more or less a bit more than 200 million. So we are on a scale of 1 to 5. Behind the revised top line, this new guidance reflects the fact that we intend to support the recovery of the underlying depletion sell-out, which is clearly improving in some key regions like the US, continuing our investment in A&P with the sales ratio maintained at the level before pre-COVID. So we are not doing also mad things as well. Keeping a tight control over our overheads costs while maintaining strict operational discipline, but remember that in the last two years, we already made a strong cost cutting. So at the end, we need also some people and operational OpEx to be able to pulverize and contribute to the rebound, or otherwise the rebound will be weaker.
In parallel, we have updated tariffs net impact to 25 million on COP in line with our lower expectation in sales, of which 5 million in China and 20 million in the US net impact. These estimates include the mitigation plan because the gross effect is clearly higher, which accounts for the 55% more than the gross impact, as well, as I already said, an increased A&P support in China and the US, to favourise the rebound. In addition to this organic performance, there are also additional negative currency effect, which remain very negative and highly volatile. While our hedging policy helps to mitigate the adverse impact, the recent evolution of the dollar and renminbi leads us to expect now on sales between -50-€60 million in terms of translation top line which is unchanged compared to our previous estimation. 40% of this impact should occur in H1 and 60% in H2. And operating profit between -25 to -€30 million before was -15 to
-20. Two thirds should be booked and realised in the H1 and one third in H2. Exchange rate volatility is likely to persist throughout the year, which is why I will continue to keep you updated in any occasion that we will talk so six times a year. Thank you for your attention. And now I am happy and open to answer to your question.
Questions and Answers Operator: Thank you, sir. And as a reminder, if you would like to ask a question, please press Star One on your telephone keypad. If you change your mind and want to withdraw your question, please press Star Two. And again, we kindly ask you to limit yourself to two questions only. And please ensure your lines are unmuted locally when you'll be prompted when to ask your question. So again, to join the queue for questions, please press Star One on your keypads. The first question today comes from the line of Laurence Wyatt from Barclays. Please go ahead, Laurence. Laurence Wyatt (Barclays): Good morning, Luca. Thanks very much for the questions. A couple for me, if that's okay. Just firstly on the guidance, I was wondering if you could give us a bit more detail on what specifically changed between the end of Q1 and now, and whetherthat was really the disruption in Q2 that caused you to make the change, or is it more about your expectation for the second half of the year. And similarly, you talked about specifically the US being a bit weaker, China being a bit weaker, travel retail not recovering in the way that you were hoping in the US. I was wondering if you could say if any of those were dramatically more than any of the others in causing you to make that change. And I suppose you didn't really mention Europe in terms of your expectations versus where they were at Q1, with regard to the guidance change and whether the changes in Europe have caused you to change your guidance as well.
And secondly, you sort of flagged it when you were making your comments around the ongoing normalised performance in China and where that currently is. I was wondering if you could tell us where you think the current normalised performance is in the market, and whether you're seeing any let up of the government crackdown on alcohol consumption at all, if that's sort of reducing towards the end of the year; what you're seeing in the travel retail channel, if you're seeing a bit of an improvement in sell-in, and whether you think the soft Mid-Autumn Festival is going to lead to a soft Chinese New Year when it comes early next year. Thank you very much.
Luca Marotta: Thank you so much, Laurence, for your question. They are very broad and wide, so I'll be a bit long. I will try to talk slowly because they are very important answers and questions as well. So the change in the guidance, which is sales and profit warning, because they are combined, has been driven by the fact that the some key compounders to be able to deliver the sell-in were not supported enough, even if improving, compared to our trajectory. In terms of importance drivers of the sales warning, because, I repeat, the operating profit is a consequence, but the main triggering point is the sales performance, which is not going the way we expected. It is, first of all, China, because Q2 was clearly worst in terms of performance of what we expected until mid to end of July and is - I will be back on this point, so I will remain - I don't want to confuse you, so I will remain very factual now. First is China. Second is the US. Even it is growing, you have noticed that is still growing. Cognac is performing. But depletion even if improving are still negative, and with the mix with a slightly negative compared to volumes. So it is inferior. Our expectation, the H2 bigger rebound based on the heels of the mathematical restocking, if you don't have, I repeated many times, a solid and at least marginally positive depletion is more fragile. So in this moment, in terms of dynamics of top line, we are well aligned, but you need to align the depletions as the reality is, which is improving but not still positive, with the top line to come. So China, second US, and third, you're right, I didn't mention because it is less important in terms of impact compared to China and the US. But EMEA is going the same way negatively. So clearly EMEA even if has a lower impact, it is not respecting its budget footprint as well, but is less important in terms of compounders, and, sorry, I don't want to say that for the European people, but is less strategic in terms of the quality of the mix. So I repeat, these are the mechanics of the warning in terms of sales, which is the first triggering point.Second point, it is the mechanics of the P&L, because I read some first notes by some analysts today. Not you specifically, but some. So please remember, the relation 5 to 1. Now we are company which is 1 billion more or less top line, 200 and something million in operating profit last year published rates which is the organic base of this year. 5 to 1. When you decline top line of - if you consider zero is 50 million from 5 to 0 is equal to 30 million, you consider our
gross margin expected, the decline should have witnessed in bottom line is higher than what we are highlighting. It means, compared to some comments, like Edward for Jefferies, for instance, there is not a deleverage, quite the opposite, because the mechanical one should have been bigger than the switch from -5 to organically, because then we will talk about forex, which is something else. It is - there is some leverage even if we are maintaining our key investment.
So in terms of mechanics, you do the exercise. You have 30 to 50 million. You apply our gross margin and you see that at the end we are declining to low double to mid-teens. If you consider that you stretched the maximum, which is mid-teens, it is 10 points compared to -5, 200 is 22 million. So mathematically you have a leverage of something which is between 10 and 15 million. So let's talk very analytically. So you can do your model. Clearly, you are very aware of the fact we decline the impact of tariffs. Tariffs are giving everything equals 5 million because it was 30 million. So this is some marginal additional leverage.
So after the maths, which was important, let's relook at the main engine of the warning. Sales, I repeat China, US and to a lesser extent EMEA. Gross margin, because even if the 5 million are playing positively compared to visual guidance, in terms of mechanics compared to previous year, China is more promotional. We want to continue to influence our improvement in depletion everywhere in the world. Also in the US, even if we are not right - at the right spot we want to be, we cannot give up looking only to the bottom line, because bottom line is influenced by the top line. As I said, the top line is the first reason why we are reducing that. So we don't want to fuel too much. And on top, don't forget, the China in terms of cost of operation, is more costly than previous year because price undertaking has been a relief compared to the anti-dumping, but is clearly more costly. Don't forget that tax duties last year were 5%, this year 10% because it was an exception. So the cost of doing business in China, everything equal has increased. A&P ratio, we are focusing to try to maintain it stable compared to the top line in terms of weight, which is mechanically means that we need to save some money because otherwise the rigidity of this line, it is bigger than the volatility of the top line. And last point, we continue to apply strong pressure and control over our overhead costs as a percentage of sales, but we don't have any big material, further savings in absolute value considering the big, big effort we have made in the last two years. You remember 230 million, of which more than 100 million overheads. And part of that, they were temporary. So there is an integration this one off. So all in all, these are the mechanics of the warning, I repeat, is the sales. And there is some leverage on the structure of the profit and loss.
Now let's back to the your question, which is what is China and what's happening. Please forgive me, I need to be a bit longer on that as well. Trying to remember all the points in China. And let's start with the negative evidence. So, overall, Mid-Autumn Festival on a market perspective is soft. Is soft because the temperature is more cloudy than expected, and the austerity and measure has put a more complex environment and less confidence for the wholesalers and for overall, the ecosystem. So we are operating in a more complicated environment. Value depletion, where we I highlighted in Q2, more or less a mid to high teens negative. But there is a mechanical effect of delay, more or less three weeks of Mid-Autumn Festival. Despite the all-negative situation and also impact on sell-in, when we measure, we try to measure the real performance, stripping away this kind of calendar effect, real performance are less negative. We are still negative, but not at this extent. If we consider only specifically our Mid-Autumn Festival performance, we are positive. So we are realizing volume depletion specifically for

