Julián González - CEO & Director Andreas Schneiter - CFO

Jörn Iffert - UBS Investment Bank Edouard Aubin - Morgan Stanley Jon Cox - Kepler Cheuvreux Paul Bonnet - Bank of America Merrill Lynch Mischa Rölli - Crédit Suisse Peter Testa - One Investments Rebecca McClellan - Grupo Santander Gian Werro - MainFirst Bank

Thank you very much for the introduction. Good afternoon, and thank you for participating in the call.

Participating - presenting the results here today are myself, Julián Díaz; and Andreas Schneiter, CFO. As in previous calls, we are going to use the presentation disclosures today in our website.

Let's move to Page 3 of the presentation. On Slide 3, you can see the topics for today's call.

I will first review our operational performance in the period, and Andreas Schneiter then will present the financials. And finally, I will return for a trading update on October and my closing remarks.

If we move now to Page 5 of the presentation with the highlights these 9 months, we managed to post a good set of results despite the adverse conditions we experienced in some operations during the peak summer season. Turnover for the 9 months grew by 4.6%, with the largest share being contributed by organic growth reaching 3.1%.

Most of the operations performed well, except for Spain, Brazil and Argentina, who have slowed down and impacted the organic growth in the third quarter. Excluding the impact of these three countries, organic growth will have amounted to positive 2.8% for the third quarter and to positive 6% for the 9 months.

We will talk about it in more detail in the next slide. We continue to have a very healthy flow of new estate being added to the business.

Up to September, we added over 18,000 square meters of gross new commercial space across 121 shops. We expect further additions as we already signed a contract which will contribute around 16,000 new square meters of commercial space in 2018 and throughout 2000 - or beginning 2019.

We also continued working on the renovation of existing shops by refurbishing over 27,000 square meters of retail area during the first 9 months of 2018. Moving to the gross profit margin.

We reached 59.9% compared with 59.4% last year. The improvement of 50 basis points is mainly due to the renegotiation and better terms with global and local suppliers.

Our EBITDA margin expanded by 42 basis points to 12.3% from 11.9% in 2017. And EBITDA, in absolute amounts, reached CHF806 million, 8.5% higher compared with previous year.

Despite the challenges we have seen in the third quarter, we have confirmed our capability to generate resilient cash flow. Our free cash flow has increased by 33.1%, reaching CHF620 million, and equity free cash flow increased by 59.1%, reaching CHF430 million.

Cash EPS also increased in the 9 months by 4.5% and reaching CHF6.07 per share. The contribution of the business operating model implementation was fundamental to these results.

Efficiencies totaling CHF33 million are already reflected in these 9 months' financials, which is already more than previously estimated. For the full year results, we have therefore increased our expectation to around CHF40 million efficiencies to be reached in 2018, leaving the remaining CHF10 million to benefit the P&L in 2019.

Now we also confirm again the CHF50 million to be generated in savings within implementation of the business operating model. Last but not least, we have already completed on 31st October our share buyback program, CHF400 million originally expected to run for 12 months.

I will discuss the return to shareholders in the following slides. Let's move now to Slide 6.

Regarding organic growth. Looking at organic growth development on Slide 6, you can see the top chart, that after a very strong performance in 2017 and an ongoing positive development in early 2018, organic growth started to slope in the second quarter of 2018 and turned negative in the third quarter.

Looking at organic growth by division, we see that the slowdown is mostly concentrated in the division of Southern Europe and Africa and division Latin America. The chart also shows that in our other divisions, we have positive growth.

This is also reflected in the organic growth performance, if excluding the impact of Spain, Brazil and Argentina, which amounted, plus 2.8% positive for Q3 and plus 6% for the 9 months, but let's move to the next slide, where we will go in more detail, division by division. In Slide 7, division Southern Europe and Africa.

Organic growth in our division, Southern Europe and Africa, fell by 2.1% in the first 9 months, while in the third quarter, organic growth is lowered to minus 5.2%. Spain, which accounts for about 2/3 of the division and which saw record years of growth in 2017 and before, had a weak performance with single-digit negative growth in the third quarter.

While the number of passengers and tourists were largely flat, spend per passenger was affected by an unfavorable nationality mix of passengers, which caused a decline in spend per passenger. In detail, we show a higher number of British, German and French tourists going to Turkey and other Mediterranean destinations in the peak summer months instead to Spain.

To drive growth going forward, we have launched some pilot initiatives in 4 regional airports, an agreement with Aena, the Spanish airport operator. These initiatives are focused in improving sales and expected to provide best practice for other airports.

Performance in other operations in the third quarter was as follows: Africa, after a very positive first half of the year with double-digit organic growth, negative environment in the third quarter with double-digit decline in Morocco and Ghana; single positive performance in Egypt and Nigeria; very positive middle single-digit growth in Turkey; low single-digit growth in Italy; and good positive double-digit growth in France and Malta. We move to Slide 8, Division 2, U.K.

and Central Europe. We reached 3.6% in the first 9 months and 4.2% in the third quarter.

Those figures exclude the negative impact from the closing of our operations in Geneva last year, starting in October 2017. Therefore, it was the last quarter of disclosure to affect organic growth.

Overall, trading continued to be healthy in the region. In the third quarter, we even saw growth acceleration mainly due to the pickup in the United Kingdom, where several operations benefited from refurbished stores and intensified in-store marketing efforts.

Elsewhere in the division, performance continued to be solid as follows: Switzerland, with positive middle single-digit organic growth for the third quarter and first 9 months; Sweden continued with its positive performance; while Finland was flat in the first 9 months. If we move now to Slide 9, division Eastern Europe, Middle East and Australia.

Organic growth in the first 9 months for division - for this division increased to 15.2%. In the third quarter, organic growth decreased - also - reached 4.4% positive.

Even the high comparatives of the previous year as well as the annualization of the reopening of the Melbourne operation after its complete refurbishment, and this is, in our view, a solid performance. Within the division, Russia, kept the good performance seen in the year and grew middle single digits.

Bulgaria, Serbia and Armenia all had good growth. We saw a strong performance in the Middle East, with most operations growing double digit, Jordan, Kuwait and India.

Sharjah also performed well with single-digit growth. In Asia, we saw lower organic growth due to the high comparatives, resulting in the following performance for the third quarter: Macau and Cambodia, with a strong double-digit growth; and South Korea, with high single-digit growth; China, Hong Kong and Singapore were negative following the same pattern of the previous quarter due mainly to the closing of some operations in these countries; Indonesia was slightly negative after several quarters of double-digit growth; and Australia continues to grow double digit, driven by the full renovation of the stores.

If we move now to Page 10, Latin America, Division 4. As you can see, in the top of the left chart, here is where we see the most pronounced change in terms of growth.

Organic growth came from positive 9% in the first quarter to minus 11% in the third quarter. This means a change of 20 percentage points in a few months.

As mentioned before, the main reason for this decline is the extreme currency volatility in Brazil and Argentina. In the third quarter, the Brazilian real devaluated by 20% versus the U.S.

dollar, while the Argentinian peso lost 45%. These types of movements typically reduce the purchasing power of customers even if we do sell our products in hard currencies and considering the fact that in Brazil and Argentina, our main customers are Brazilians and Argentinians.

Thus, they are directly impacted by the currency devaluation. And just looking at the performance, we see organic growth mirroring the movement of currencies.

This development also started to show some effects in other South American countries. On the other hand, performance in Central America continued to be very strong.

Our cruise division continues to grow a high double-digit growth, benefiting from the several new contract wins and openings realized in the last few quarters. We also saw good performance in Dominican Republic, Puerto Rico, Jamaica, among others.

Mexico has the exception of staying flat in the third quarter after a strong half year 2018 and single-digit positive performance in 9 months. If we move to Slide 11, Division 5, North America.

Organic growth in North America remaining very strong at 7.5% in the first 9 months and 7.1% in the third quarter. The Duty Paid part of the business, which is mainly convenience, continues to grow strongly at the back of a steady growth in the number of passengers, increase in the spend per passenger as well as positive impact of the new concessions.

Our duty-free shops are also performing well with middle single-digit growth. If we move to Page 12 of the presentation.

Regarding passenger growth, I will now explain both charts and tables that show a very strong picture. The global increasing passenger number has continued to be very strong and the expectations of industry experts remain very solid.

This is a good sign that the industry keeps its good and resilient fundamentals. Looking at our own footprint, passenger growth expectation is different: first, as we are still underrepresented in Asia and we are currently unable to control the full scope of these lows; and second, in the third quarter particularly, we have been impacted by the three markets disclosed before, Spain, Brazil and Argentina.

If we move to Slide 13, with effect to driving organic growth, two main pillars are the opening of new retail space and the refurbishment of our existing stores. In the 9 months 2018, we have opened a total of 18,000 square meters of gross new retail space, and we are targeting 26,000 for the full year 2018.

The main highlights of the new openings so far are as follows: in Madrid, new Hudson shops; in Malaysia, the first downtown shop in this specific country; for cruise lines, I would like to name the two new ships that we have opened in early summer, with cruise lines such as Holland America, Carnival and P&O; and finally, 23 new stores in several North American locations. Looking at the refurbishments, we have been very active on the renovation of our operations.

And up to September, we have renovated 58 stores, equal to 27,000 square meters of retail space. Among the main projects are as follows: Malaga, three stores; Heathrow Terminal 3, New Generation Store; Glasgow, main store; Bali, two stores; and Cancun Terminal 3, New Generation Store.

When we do these refurbishments, we consistently also implement our digital strategy elements, I mean, at improving the shopping experience for our customers and driving sales. Good examples are the two New Generation Stores we'll launch in Heathrow Terminal 3 and Cancun Terminal 3, with an extensive use of digital technology.

Here, we have also already considered the first learnings of the other New Generation Stores opened last year. In this context, we have also equipped ourselves, associated with the tables, with a lot of considerable improvement in customer service through better product information.

In Slide 14, new contracts and pipeline. Among the new space signed so far within the year and totaling 15,000 square meters, the main contracts already disclosed are the new airport operation in Perth, Australia; the 10 new stores in Philadelphia; one U.S.

store in St. Petersburg; and two new stores in Chile; as well as several shops at Chicago Midway and Boston Logan in the U.S.

Moreover, we have an important project pipeline of close to 40,000 square meters we are currently working on. This potential opportunities are currently mostly located in Southern Europe and Africa, Division 1, as well in the Division 2, U.K.

and Central Europe. If we move to Page 15 for commenting on the business operating model.

I would like to update you on the current status of the implementation of the business operating model implementation, one of the most important internal projects as it's allowed us to generate a total of CHF50 million of operating efficiencies, which will be reflected in our P&L in 2018 and 2019. The business operating model is currently being implemented in several countries in Europe, Middle East and Asia.

So far, implementation has been started in 47 countries, of which 23 have been - already received the first certification, and 11 of those have passed the second certification. The benefits are already visible in the P&L, and we expect to already see an impact in the first CHF40 million in full year 2018, which is considerably higher than the CHF26 million originally planned.

The remaining CHF10 million will be bill-up and be reflected in 2019 results. If we move to Page 16, as already mentioned at the beginning of the presentation, in this page, I would like to give you an update on the cash return to shareholders within 2018.

First, on May 17, 2018, we paid a dividend of CHF3.75 per share, for a total of CHF200 million. With respect to the dividend payment to shareholders, we have committed to pay at least the same amount as previous year, equal to a minimum of CHF200 million and up to a sustainable return level of 40% of cash net earnings as a target.

Additionally, on October 31, we have just completed ahead of time the share buyback program of CHF400 million with what is scheduled to last 12 months until May 2019. In total, we spent CHF402 million and purchased over 3.3 million shares, which we intend to cancel, as already announced.

Our digital strategy. In 2018, we have opened two additional New Generation Stores, one at Heathrow Terminal 3 and one at the Cancun Airport Terminal 3 in Mexico.

In both stores, we have also been able to implement the first learning from the previous installations in Madrid, Melbourne, Cancun Terminal 4 and Zurich. The next New Generation Stores to be opened will be Buenos Aires and Amman in 2019.

Our online reserve and collect platform had been extended to 21 countries, covering 81 airports, by the end of September, and we will fully increase coverage by the end of the year. On the platform, we have also further improved the assortment and the value proposition.

The footprint of RED by Dufry, our customer loyalty program, has also been extended and covers now 36 countries and 160 airports. And last but not least, we have equipped our sales associates of 40 stores in 14 countries with a sales target, which allowed to considerably improve the service provided to the passengers.

And now I am passing to Andreas Schneiter for continuing with the financials.

Thank you, Julián, and good afternoon and good morning, everyone. Let's move to Page 19, where we have the growth components.

Reported growth was positive in the third quarter due to positive FX translation effects. As Julián already explained in detail, there was a slowdown in organic growth in the third quarter due to challenging external factors in certain markets.

This effect clearly shows in the like-for-like growth that turned negative 0.9% in the third quarter. As to net new concessions growth, Q3 was below previous quarters mainly because of some of the projects annualized in the second quarter, such as Melbourne, where the expansion went live at the beginning of Q3 last year.

Generally, net new sales contributions can still fluctuate because of the timing and the size of such projects. Based on our current project portfolio, we expect that net new concessions growth will accelerate again in the fourth quarter.

Overall, Q3 top line performance has been lower than what we originally expected. Having said this, and considering the pressure on the top line, we have delivered good results overall and our other KPIs as we will see shortly.

There has been a good gross margin progression, we could improve EBITDA margin, and we have generated a very good free cash flow and equity free cash flow. The performance of the third quarter illustrates that we do have a very robust business model that can be managed tightly.

And although we're not entirely happy with the performance, we do feel that we have managed the business well and delivered good results given circumstances. If we move then to Page 20, I would like to talk about the FX impact.

In the case of Dufry, the FX impact that you see is the translation effect converting into Swiss francs all the different currencies where we sell our products. You see our main currencies in terms of sales are at the bottom of the - bottom right chart.

These are the U.S. dollars, the euros and the British pounds.

In this context, let me make one additional comment because we do price in those currencies, our organic growth is also measured in these currencies, e.g. this means our organic growth in Latin America is based on U.S.

dollar sales and not based on local currency sales. Then on the top left chart, we see the evolution of the FX impact on turnover.

In the 9 months, we had a positive FX translation effect of 1.5%, and we had a similar positive impact in the third quarter of 1.3%. At the bottom left chart, we see the evolution of our main currencies versus the Swiss francs, which point to a negative translation impact in Q4 due to the strengthening of the Swiss franc against the euro and the British pound.

Then if we move to Page 21, to the income statement, and let's move directly to gross margin as we already talked about turnover. Gross margin improved by 50 basis points, driven by negotiations with suppliers and also initiatives that we developed together with the brands, mainly related to promotions and the so-called brands plan.

There was also a slight benefit from mix effect. Concession fees increased by 30 basis points to 28.1% in the 9 months.

Of this increase, about 2/3 is attributable to the Spain contract, where the lack of growth, in combination with the increase in minimum guarantees, led to a relative increase in concession fee charge. For all the rest of the business, the net increase was about 10 basis points.

Personnel expenses and other expenses together improved by 30 basis points in the period. This is mainly due to the efficiencies from the business operating model as explained by Julián, and you can see well the improvement in the general expenses.

The positive impact of the business operating model and the personnel expenses is not fully visible as there were some underperformance in some operations and some increases in personnel expenses in North America offset this improvement of the business operating model. As a result, EBITDA grew to CHF807 million and EBITDA margin improved 40 basis points to 12.3%.

Then moving on, depreciation and amortization was slightly higher than in previous quarters at 6.4% of turnover. This is due to an increase in depreciation, which is related to our continued investments in new space and refurbishments.

Amortization was almost stable in Swiss franc turns and slightly decreased as a percentage of turnover. Just as a reminder, more than 80% of the amortization is related to acquisitions.

Linearization was a CHF27 million charge for the 9 months. As a reminder, linearization comprises of the noncash elements of the Spanish contract, this means the straight lining of the minimum guarantee increases over the period of the contract terms as well as the prepaid concession fees.

Due to the seasonality of the Spanish business, the linearization is positive for Q3. For Q4, there will be a charge again.

For the full year, the linearization charge will be around CHF50 million. Our operation result for the 9 months was CHF32 million.

Of this amount, about half is related to new projects and startup as well as restructurings and closings. Financial results improved by 25% to CHF99 million.

This is mainly due to the refinancing that we executed in 2017 as well as the lower net debt that we had along 2018. Income tax was CHF92 million in the 9 months, and there was a big increase year-on-year of about CHF55 million.

Of this amount, CHF47 million were incurred in the first half of the year. And as we mentioned at the time, the first half year increase related to deferred taxes that are noncash in nature, about CHF35 million, and of those CHF35 million, about CHF20 million are one-off charges, which are, in majority, related to the restructuring in the U.S.

that we did due to the Hudson IPO. Then in the third quarter specifically, the increase in taxes was CHF8.5 million, and the tax rate moved to about 27% from 23%.

So the third quarter is more representative as there were no material one-off items, although the devaluation in Brazil and Argentina resulted in an increase in deferred tax charges. In addition, there is a mix effect, whereby we accrued more profits and taxes in those operations where we grow faster.

Now taxes are quite difficult to forecast. Our best estimation at this stage is that for full year 2018, we will have a tax charge of about CHF100 million in the income statement.

Moving on. Noncontrolling interest were CHF49 million, of which the largest part is due to our business in North America.

The CHF49 million already includes the Hudson minorities from the IPO that we did in February this year. The result then basically is cash earnings, which improved by CHF7 million to CHF319 million.

Let's then please move to Page 22, where we have the cash EPS. If you look at the top chart, we show a bridge from last year's to this year's cash net earnings.

As you can see, the tax charge negatively impacted the cash EPS for 2018. As I mentioned beforehand, because the largest part of the increase in taxes is noncash and includes also one-offs, the actual result does not fully reflect the real performance.

The effect of loan from one-offs is about CHF0.40 per share. Then if we move to Page 23, we can talk about the cash flows, and we have shown here our KPIs in that context.

For the 9 months, we had a record cash generation with free cash flow at CHF619 million and an equity free cash flow at CHF430 million. We will feel the details in the coming slides, but before we go there, let me make one comment.

2018 is the first year for some time where the cash flow has not been impacted by any exceptional items and that, that reflects the cash generation of our business. In our view, 2018 is actually a good proxy for future performance.

Moving through the numbers. I would like, first, to refer on the seasonality chart at the bottom of the page.

As we had mentioned 3 months ago in our latest - in our last results conference call, some of the positive impacts which normally makes Q3 the strongest quarter in terms of cash generation were anticipated in Q2, specifically, the positive impact of net working capital during busy season. The key points here is that we have a very strong cash generation for the 9 months and we shouldn't necessarily look just at the future.

Despite a lower top line growth, our expectations for the full year remains largely unchanged compared to previous calls. For the free cash flow, we expect to reach an EBITDA conversion at the higher end of our range of 50% to 55%.

And for the free - for the equity free cash flow, we expect to end up at the lower end of the previously indicated range of CHF350 million to CHF400 million for the full year 2018. Now we can move to Page 24.

We will look at the free cash flow components in more detail. In the first 9 months of 2018, we saw a positive net working capital inflow of CHF94 million, of which about half is due to changes in core net working capital as you can see at the bottom of the right chart.

You also can see there that we already have core net working capital below 5% in the second quarter this year. The other half of net working capital improvement was due to a better noncore net working capital.

Cash taxes were CHF81 million compared to CHF69 million last year. Compared to the taxes in the income statement, cash taxes are much more stable and growing more in line.

For capital expenditure, we reached CHF181 million in the first 9 months. You can see the evolution in the bottom left chart.

As a percentage of turnover, CapEx is slightly below 3% versus a normal level of 3% to 3.5%. Then moving to Slide 25.

Here, we have mainly 2 items to be discussed. Interest and minorities.

Interest paid was CHF127 million for the first 9 months, CHF30 million lower than one year ago. As already mentioned, the improvement is due to the refinancing executed last year as well as lower debt levels.

Cash flow related to minorities were CHF56 million, and this is CHF22 million higher than last year. About half of this increase is just the timing difference, whereby we paid dividends to minorities in Q3 this year, whereas, last year, we paid the same dividends in Q4.

On Page 26, then we have the bridge from equity cash flow to net debt. On one hand, we have the proceeds from the Hudson IPO, which is CHF665 million, and on the other hand, we have the cash that we returned to the shareholders, i.e.

the share buyback, the purchase of treasury shares and the dividend payment that we made in May. As for 30th of September 2018, we returned about CHF600 million to our shareholders since the beginning of the year.

After the completion of the share buyback last week, the total amount for 2018 will be CHF720 million. Then on Page 27, we have the balance sheet.

There is no major comment. The asset side hasn't really changed.

Concession rights continue to decrease as we amortize them. On the liabilities and equities side, we have an increase in equity, which is due to the Hudson IPO, and the decrease in net debt, which is due to a combination of IPO proceeds and the cash generation as explained beforehand.

Then on Page 28, we have net debt evolution, which reached CHF3.01 billion after September 2018. Our covenant was 2.92x net debt-to-EBITDA.

With this, we are within our target range of 2x to 3x net debt-to-EBITDA and well below the maximum threshold of 4x leverage that we have agreed with the banks. As mentioned in our earlier calls, we have a long-term financing in place and there are no maturities before 2022.

In October 2018, we extended our RCF facility by one year to 2023, so we further improved our maturity profile. To conclude, a few comments on IFRS 16.

If we move to Slide 29, we have a refresher on the different aspects of IFRS 16 that I don't want to go into details there. This is just for those people that may not have all the effects at the top of their mind.

On Page 2017, you may know that we have IFRS 16 being implemented for 2019, so the first time reporting will be Q1 2019. The current status is that we have now - that we are now in the final stretch of our review.

Compared to our previous indications, we have a lower impact on the balance sheet and to a lesser extent, also on the lines in the income statement. Our current estimate for the balance sheet effects, currency trends towards the CHF4 billion mark and amortization impact is expected to be below CHF1 billion.

As you may remember, from 2019 onwards, EBITDA will not be a meaningful KPI anymore. We will focus more in cash flow metrics as they remain largely unchanged, and we also will attempt to provide some additional metrics that will allow to track performance of our business before and after the change to IFRS 16.

This concludes my part of the presentation, and I'd like to hand back to Julián again.

Thank you, Andreas. I will now move to Slide 31 - sorry, 32.

In the first four weeks of October, regarding the trading update, net sales were gradually improving, with organic growth close to plus 1%. The improvement in organic growth is due to a number of factors, including seasonal lower exposure in Spain and better performance in this market too; generalization of the Geneva closing last year in October; further improvement in performance in Asia, U.K.

and U.S.; and the contribution of the new openings mainly in Asia, such as the Hong Kong MTR train terminal operation and the new shops in Perth. For the full year, we therefore expect an ongoing year-on-year performance improvement for the group overall, with the information as follows: an organic growth for the full year of between 2% and 3%; an EBITDA margin between 12% and 12.3%; and a confirmed level of equity free cash flow, as Andreas mentioned, of CHF350 million, CHF400 million.

In Slide 33, what we have, as a conclusion, is to summarize the performance in the first 9 months. I would like to really remark the resilient set of results realized despite the adverse conditions in some of our key markets.

Despite the slowdown in sales, we have further expanded our margins at gross profit and EBITDA levels, and we have confirmed our cash generation capability. We have reached an advanced implementation stage of the business operating model, and we have completed the share buyback program.

And we are executing as planned our other goals set for the year 2018, which are a further expansion of the initial initiatives in the context of E-Motion as well as to continue focusing in cash generation. I think, so far, the situation in 2018 has been challenging, especially in the third quarter, but we are, especially in this set of financials, showing again the resilience of the business model.

That's all from our side, and if possible, we can now move on the Q&A session. Thank you very much.

And there would be 4, 4 quick ones, please. The first one is, looking into first half 2019 on organic growth, I mean, can you roughly share with us what is your expectations on net new shop extensions going through the first half '19?

And any other initiatives for service things that you can return to around mid-single-digits organic growth in the first half '19, or will it remain at the low single digit? Give your current and best guess.

Second question would be on the EBITDA margin. If I read it correctly, you are guiding that the Q4 EBITDA margin is down around midpoint of the guidance, 50 basis points.

Is there something special you should consider on the concession fees across profit margin? And then here's also a question, how should we think about margin progress in the first half '19?

And third question, on the cash flows, I mean, you said 2018 should be a good proxy for 2019. Don't you see the risk that you have to do prepayments for concession renewals in 2019?

And then the last question, very quickly on your strategy, the U.S. - you see the Hudson proceeds through the share buyback.

Does it mean that there is a limited M&A prospect then in Asia? Or how shall I read it?

Okay. Thank you for the questions.

Let me start with the like-for-like in Q1. It still is very early so because even we don't have the budgets finalized.

But I see 2018 in - sorry, in 2019 Q1 in a different mood. Last year, we grew 7.5%.

As in this year, obviously, in terms of comparable, it's going to be more difficult to compare. In order to reach the level of 7.5%, a lot of things will change, for example.

What is going to happen in Brazil? What we have seen during the last 10 days, 15 days is a significant turnaround.

After the elections, the exchange rate normalized, and as a consequence, the situation in terms of performance, sales performance, is improving. Argentina doesn't move a lot.

Spain is improving. And I think, obviously, it all depends on the circumstances, but I think the toughest comparable next year will be Q1.

For the year, I cannot tell you anything, but I can, obviously, so yes, as soon as we know the information, we will provide the market information with more accurate facts because now it's just a perception. Regarding the EBITDA margin for the Q4, I think you need to - obviously, there are some elements that may limit EBITDA margin development.

Number one is the new businesses that we are open - that we are opening and we opened. As you know, and I commented on that, the cruise lines and some of the businesses in Asia are lower margin, even that, obviously, they have an IRR expected at the same level and sometimes higher than the business that we are operating today.

This will have a, let's say, an impact in the margin. And the second one is Spain, Brazil and Argentina.

If Spain, Brazil and Argentina are not turning around as we are expecting, especially Spain and Argentina, the fixed cost will be over-proportional to the situation and also main, but the percentage. But within this range of EBITDA that I mentioned, all these things are considered.

Regarding the cash flow...

So on the cash flow, I think, you were mentioning that there should be any prepayments. Look, I think, we highlighted many times that prepayments are actually not normal course of business in our industry.

And as far as we know, from today's perspective, there's no such things there. So I would reiterate or confirm that based on what we know today, there shouldn't be anything one-off in 2019.

And regarding the M&A, I don't want to obviously repeat myself many times, but I have seen, and I am still repeating that the business - travel retail is very fragmented, that the best way of allocating capital in this company is to continue with this type of growth, especially in Asia. And is - the question is are you going to consider a delay in the strategy?

The answer is no. I think, the opportunities will happen when they happen.

And obviously, the conditions are related to the vendors. If they decide to sell, I think, we should consider the opportunities and one of the, let's say, basic principles of our business operating model is to continue with the growth of consolidating the business.

Okay. Only last one, Julián, would you really - is it 100% sure that you are canceling the shares or could you use the shares you bought back also for M&A prospects in 2019?

The idea of these shares was not originally that. The idea is still not originally that.

We acquired these shares in a situation where we saw the price, we saw that it was a package of good value, because we saw that the value of the share price at this time was really low and we thought it was a good deal for the company investing in their own selves, that's all and still they are. If in the future they are using M&A transaction, why not.

But today, it's not the intention.

Just three quick ones for me. Just to follow up on the space growth, so sorry, if I missed the answer, but it came in below expectation in Q3.

Could you be reaccelerating the space growth in '19 towards the 2% historical average? That's number one.

Number two, on the gross margin, I guess, that was the positive, so one of the positive surprise of Q3, which is a 90 basis point expansion year-over-year and you've mentioned some of the drivers of that. So that came ahead of your annual guidance, I guess, of 30 to 40 basis points.

Based on what you said regarding EBITDA in the fourth quarter, should we expect a much less significant expansion of the gross margin in the fourth quarter? And what should we be expecting towards 2019?

Again, should we come back to the 30 to 40 basis points? And lastly, on - I know, it's a bit premature to talk about EBITDA for '19, but given that you've pulled forward some of the benefits of the BOM program, should we be looking at an EBITDA margin in '19 more or less in line with what you - in percentage terms, of what you are likely to post in '18?

Okay. Regarding the new space, I think, it's very similar to previous year, it's 4.2%.

And the difference is close of operations along the year. I don't see any difference with previous year that we have been always between 3% and 6%.

And probably what happened in this quarter is that there are annualized FX, especially in Cancun and in our Greece Division in Asia. But this is something that happened, because we're opening at the beginning of the year, especially that we now are at the - at comparable level impacting less in the positive side, but I still believe that between 4% and 6% of new growth space is possible every year.

Regarding the gross profit margin, obviously, is - in my view, is a sustainable situation, because the gross profit margin has been, for a while, one of our main drivers. We have been negotiating with suppliers locally and globally, especially locally, for improving the gross profit margin and, I think, during the Q4, we are going to see a similar increase.

Regarding EBITDA for 2019, it's difficult. As I said, we have not even finalized the budget.

I cannot comment on the specifics until I have a more clear and obviously transparent information. I prefer to avoid the discussion about full year 2019 until we know what is going to happen.

So Julián, on the gross margin thing, so I mean, if your guidance implies a contraction of EBITDA of around 40 to 50 basis points in Q4, what you're saying is that you can still - you should be in a position to post the 90 basis points or something gross margin improvement in the fourth quarter?

I think, this is a realistic information in terms of gross profit margin. In terms of the Q4, obviously, as a consequence of what I said regarding the full year is a consequence, I am going to - I want to clarify that, because I repeat it.

There are new businesses that we are consolidating with lower EBITDA margin, especially the cruise lines. And I said that at the beginning of the year.

And we have added - we are going to add this year close to 13 new cruise lines. It's a different business, but in terms of IRR, it's, as I said, it's similar or higher, and I'm now going to avoid new concessions for new businesses, because the EBITDA margin is lower than higher.

It is a different business that is consolidated. There are also other businesses in Asia with lower EBITDA margin that started to be consolidated in the 1st of October.

And this is happening. And also, it's happening, what I said, there are operations like Spain, Brazil and Argentina, that due to the lower, in this case, sales have a fixed component of the costs that are impacting also the possibility of continuing with obviously the same level of gross profit margin - sorry, of EBITDA margin.

Those are the reasons.

A couple of questions on my side. Just looking at your comments, again, about reiterating at least CHF200 million on dividend - on the next dividend, the way I look at it, it has to be over CHF4 per share just to get to CHF200 million, because, obviously, you've reduced your share count.

When I look at consensus, it looks like it's around CHF3, CHF3.8, do you think that consensus dividend expectations for next year are too low? That's the first question.

Second question, I wonder if you could just give us a bit of an idea on the size of those new businesses, which will obviously have a negative impact on mix, just roughly? Is it like a couple of percent or whatever it may be, those new businesses coming through, so we can try and think about what the margin of those businesses might be coming on.

The second question. Third question, just a nuts and bolts on working capital and CapEx evolution.

Core net working capital continues to remain well under control, any thoughts on where it could go next, Andreas? Or do you think it probably will remain now roughly where it is?

Same on CapEx, you mentioned below 3% in Q3. What is your best guess for the year as a whole?

And then just a last question, if I can, on Spain and Aena, and that contract coming up in 2020, you said, you wouldn't sign a similar contract in the future. I'm sure the negotiations are pretty intense.

Just wondering where we are there, particularly, as you are launching new projects with Aena, which we think that you are in the driving seat, at least, to try maintain those contracts. Yes, any update there, that would be much, much appreciated.

You want to start with it?

I'll start with the working capital and CapEx. Look, I think, we are currently at a good level, in the sense, look, we will try to improve the working capital as we can, but I - as we always said, look, the 5% that we have now seems like a realistic target.

It's hard to material improvement, so that's why I would think that the levels that we have today for working capital and CapEx are actually sustainable. If we can improve it marginally, obviously, we'll try to do that, but I wouldn't expect any big change, and thus as such, and also, if you want the cash flow generation, equity free cash flow and free cash flow shouldn't materially change from a structural point of view, that's why we feel that the 2018 numbers actually are a good proxy as I have tried to explain beforehand.

Sorry, and the dividend, yes, look, that's probably our mistake. When we think about it, it's like the same amount, so I would argue the CHF3.75 are probably the base case.

The board hasn't decided anything yet, so I think what - the proper way of forecasting that is to take the CHF3.75 type of share that is outstanding, so it's just below the CHF200 million that Julián mentioned. But this is our mistake, because we still think on the full number of shares.

Okay. From my side, Jon, regarding the size of the new business, obviously, depends on the performance, but is between 3% and 5% of the total sales.

If we are able to open everything on time, this is the full year impact. And then if we are, obviously, depending on the months - or depending on the number of days you open the business.

Regarding the Spanish contract, I think, I have been very transparent with that. The conditions are not the best conditions in a contract for us.

In fact, this is a very tough contract, especially due to the minimum annual guarantee in two of the lots, lot number one and number lot number two. This is public information.

What we have done is we have agreed with Aena. The implementation in five airports in Spain of a project, that we'll consider the traffic flow, the configuration of the shops, the brands, the assortment, the pricing policy, the digital approach and also new opportunities of businesses that may arise during the discussion in order to show up that after six years or five years of operating this contract, the company, as you know, the contract was awarded to the former World Duty Free operation that now is operated by us.

And we believe that with these changes in the operation, there is a possibility to really overpass the problem with a minimum annual guarantee. We just started the renovation of some of these airports and we expect that during the next 3, 6 months, we will be able to show up Aena that this is a different, obviously, way of doing things.

As you know, when you agree, and I'm not talking about specifics of the contracts, because, as you know, it was a very difficult moment in time when World Duty Free agreed about this situation due to the economy in Spain and due to the situation of the Spanish airports at that time. The original business plan was based in assumptions and was based in the current market conditions in - at this time.

To date, we believe that with all the changes that happened, not only in the airport environment, but also in domestic market in Spain and in the countries where the passengers are coming from, the commercial offer in the Spanish airports in the retail part, in the duty-free I'm talking about, requires a different approach. And as a consequence, the test is probably the best example, because it's not the elliptical discussion.

It's practical and specific discussion, comparable with previous performances in 2018 and 2017. If the results of this test, as we expect, I think, it's obvious that we will be very interested to continue.

And I think this is the assumption that we only work with at this stage of the process. Depending on this performance, everything else is possible.

But I think, first of all, we need to weight out the performance and see what the reality is.

So just to confirm, the - you think that what you've done there on this pilot programs should deliver better revenues? And if that's the case, then you're happy to, sort of, maintain the form of the current contract, including a new minimum annual guarantee?

There, I'll say yes.

Yes, just three quick questions from me. The first one is actually on the business operating model, just wanted to know a little bit, because at your Capital Markets Day, earlier in 2018, you said you were expecting to save CHF26 million in 2018, now you say CHF40 million in 2018, but I guess, the business operating model, the whole idea behind it was to save on the operating expenses, right?

And when I look at the operating expenses and the first 9 months of this year, actually, I only see a little bit of leverage on the operating cost, but otherwise pretty much everything is unchanged. And so the 40 bps increase in the EBITDA margin actually comes fully from the gross margin increase.

Should we actually expect any savings - cost savings on the operating expenses? So that is my first question.

Then my second question is actually I thought one of the disappointing thing about the results actually came from Asia, and you pointed toward the tougher comparison base. So effectively, the comparison base is approximately 6 percentage point tougher, so between Q2 and Q3 last year, but actually the performance of the business deteriorated from 23% to 4% between Q2 and Q3 this year.

So where does the 12 percentage point underlying deceleration come from? It's the second question.

And then the third question is on the net new concessions. I just have a little bit of hard time to reconcile exactly how this works, because, in the last two quarters, we have seen approximately 2% that was coming from that.

But the reality is, if my understanding is correct, when you open a new concession, this should last approximately, well, four quarters, right, before this is in the base. So I don't really understand how come we had only two quarters of around 2% and before that we had below 1% and now have below 1% again.

So if you can enlighten me a little bit on that, that would be much appreciated.

Okay. Regarding the business operating model, I think, I have repeated this many, many times, is, we target CHF50 million total input in the P&L in 2018 and 2019.

Initially, we forecast an impact in the P&L 2018 of CHF26 million. Due to the acceleration of the implementation of most of the initiatives of the business operating model, we have explained during the presentation that - is that to - is that expected CHF26 million in 2018 will be CHF40 million.

And the reason is the acceleration of obviously the implementation of this business model. So far, in the P&L 9 months, what you've seen is CHF33 million.

So can you just point to me where those CHF33 million are, because the only thing I can see is the general expenses went from 4.9% of sales to 4.6%, which for me, is CHF80 million, not CHF36 million.

Yes. So - this is Andreas, Paul.

Look, I think, you have, on one hand, the general expenses that is one thing. The other - and that's what I tried to explain in my speech is like you have personal expenses and I know that you don't see an improvement there, but I think, you have two elements that go against each other.

On one hand, you have the business operating model improvements, but on the other hand, you have two things: on one hand, you have a negative impact, for example, in Spain, because of the lower performance of the business and then you also have a negative impact in North America from increases in personal expenses. So although you don't see it there, reality is that the business operating model has delivered there as well.

But it's not something where you just can say, look, let's go straight there, because there are these two elements that - if you want to offset the improvement.

Okay, fine. So in the underlying business, in the normal course of business, we should have seen around CHF33 million to CHF36 million, but now we're only seeing, like, CHF20 million, because that was the leverage due to the lower organic growth.

Is it what you mean?

The lower organic growth and the cost increases in the North America, correct.

Okay. I'm Asian.

Okay. And what about the two others, please?

Yes, I've seen in Asia, I think, I commented during the presentation. You have number one is the annualization of Australia.

Number two is the annualization of the Asian cruise line business. And number three is the slowdown in some operations in Asia, especially in Indonesia, that during the first quarter, was double digit growth and in the third quarter was almost flat.

Net of new concessions, I think, the net of new concessions, is, as I said, I comment on that before. If - new concessions contributed 4.2% of gross of the sales.

Closed concessions 2.9% during the first 9 months of 2018. And this 1.3% net added to the FX impact - sorry, added to the like-for-like, CHF1.8 million result, an organic growth of 3.1%.

In previous quarters, what we had is a participation of 1.9%. The reality here is, again, the comparable with previous quarters, because now what we have is, and this especially, the cruise line business in Asia.

In Australia, the last year was the first time that we opened with a full renovation of shops, is comparable. And now, it's moving to, obviously, like-for-like impact when you have...

Okay. So considering that the comparison base actually gets another around 15 percentage point offer into Q4, are we going to see negative organic growth in Asia into Q4?

Is it a possibility?

No way. I think, In Asia, we have very good new operations and we start in the 1st of October - not 1st of October - end of October and November, that we'll mitigate any possible negative impact, will be positive.

Okay. And so what about the phasing of the net new concessions, which is my last question.

Yes, new concessions.

So look, I think, as you pointed out correctly, I think your concept is absolutely right. So whenever we open or close a new business, obviously, that will go into the net new concessions lines for four quarters.

Now obviously, depending on how big the project is and when it starts, you may have every quarter openings and closings. So to give you an example, in Q4 2017, we closed Geneva.

That's why you would see actually quite a strong drop in the quarterly net new concessions growth in Q4 last year, okay? Now for example, Melbourne, that was a big one that we opened beginning of Q3 2017 that will roll-off now - or that has rolled-off actually in Q3.

I think, because it's project by project, there is no, if I can call it that way, systematic tier on what the growth can be depending on what comes off and what goes on even the numbers may fluctuate. And I think, again, that is something that I wanted to explain in my speech, so you cannot just take any given quarter and take that as a run rate.

There is some volatility there. The key point for us here is like Q4, again, it should accelerate, because there's a number of new projects, which are actually quite sizable, and Julián commented about it, they go online or they have gone online in Q4.

Yes, just a couple from my side. First one, just as a clarification, with your Q4 guided implying around 46 bps of EBITDA margin decrease, I wondered, I mean, does this also - does this guide also imply a mid-single-digit decline in organic growth in Division 1 in Q4?

I'm asking, because, obviously, the Aena winter guide, capacity guide has also some C-capacity data - is suggesting quite remarkable recovery in Q4, but also Q1 '19, so is this already - does it still imply around mid-single-digit decline?

Can you - so - the line cutoff several times during the time that you were asking the question. Can you please repeat, like, we cannot hear the question.

Sure. So with the Q4 guide implying 40 to 50 bps of EBITDA margin decrease.

I wonder does this guide also imply a mid-single-digit decline in organic growth in Division 1 in Q4, so - because, obviously, we have seen - there's some C-capacity data and also the Aena guide for winter capacity suggesting quite remarkable improvement in traffic growth. So that will be my first question.

And then my second question is, I think, in the press release, you speak about structural issues in Spain. Are those comments simply, basically, related to the minimum annual guarantee?

Or what else do you mean with structural issues? Has there been any change to the duty-free environment there?

And the third question would be, I mean, I guess, you guys have quite a good view of how the Chinese consumers are feeling in general, so in addition to what you already commented on APAC growth, is there any, sort of, slowdown for a Chinese in any sense?

Okay. Regarding the Division 1 organic growth in Q4, we don't expect that double-digit organic growth in Q4 for Division 1, in fact, the Spain, that is the main - has been the main reason of the negative organic growth, is improving significantly, number one.

Number two, the structural changes in Spain means and I think during the presentation, there was a specific comment about one issue that is very relevant, the growth in the traffic in Spain - international traffic in Spain increased by, I think, 1%. The difference is the profile of the passenger in this increase.

British passengers and other nationalities decreased by around 4%, 5% during the period that we are talking about. As a consequence, because these passengers were substituted by Spanish passengers, the spend per passenger has been significantly impacted.

But what the problem is, the problem is that, one, let's say, British has around €10, €11 per head, and 1 Spanish passenger is €2 per head. This is the main structural problem that I tried to explain and probably is not well explained.

They are - in the Chinese customers, what we have seen, in general, is not any slowdown, it's a drop in the spend per passenger. And what the reason is, depending, obviously, the destination, but the main reason that we have seen is that the type of Chinese passenger traveling today is a bit different than 2 or 3 years ago.

We are - we'll comment in several of the locations where we have Chinese passengers, different passengers' profile, lower profile than before. And I think, this is something that has been repeated in most of the locations that we have been operating.

It's not the impact in number of Chinese passengers, it's more of a change of profile of the Chinese passengers.

Just three things. One, maybe, just following up on your comment on the Chinese passengers, is that something you saw happen, say, during the second half of the year?

Or is that a longer-term comment that you're making? And then just two things on mix to make sure I understand.

On the Spain, Brazil and Argentina factor, I guess, because of the season, Brazil and Argentina have a higher weight in Q4, so even though the difficult trading happened in Q3 and it's a bit recovering in Q4, that higher weight is what gives you the point on the operating leverage that you are making? And then the third question is just on the cruise side.

It's about 3% of sales. I was wondering if you could give us some sense based upon the business, one, what, sort of, base of sales that would represent just to give us some context for that opportunity to be able to understand your comment about it?

Yes. Regarding the Chinese passengers, I think, this is a gradual change that we have seen over the past years, but essentially in 2018.

There are more Chinese passengers traveling with, obviously, the address - profile lower than before, where obviously, most of the passengers were high-level. Now we have also mass-market Chinese.

The second one is the seasonality. Spain, as you know, is very seasonal, especially in Q3.

And the recovery of Spain that is, in fact, happening now made no impact at the level, obviously, needed in order to accelerate organic growth, but is not going to that much the - performance. Brazil and Argentina are different cases.

Brazil, depending on what we have seen, is an improvement compared with Q3. And Brazil has a very, let's say, a good performance during Q4.

It's not like in northern hemisphere, it's like more in the southern hemisphere. Argentina is not an improvement.

Argentina remains the same and is going to be tough, as a consequence. Spain, even in the case that recovers, is not having the important impacts that we have lost during the Q3.

Brazil, if Brazil recovers, we will have a significant good impact in Q4, because it's an important quarter and Argentina is not moving a lot. I don't think that we should consider Argentina recovering.

Okay. And then on the cruise point, just so that I understand the context for what you're talking about?

Obviously, what we disclosure in our information is around 3% generated by the cruise business in the top line in the P&L. But I have this commented on before was that we have opened during this year and most of these cruises have been opened at the beginning - at the last part of September, beginning of October and along the last quarter is new 30 cruise lines.

The contribution of these in a full year basis will be between 3% and 5%. And I think depending, obviously, gradually, the implementation and the starting of this business will increase even, obviously, the percentage of participating in the total sales.

This is the size of the magnitude.

Okay. So that 3% to 5% is cruise alone on top of the normal gross openings?

Yes.

Can you hear me?

Yes, yes, perfectly.

Yes. I have got four questions for you, please.

Firstly, the organic growth of 1% that you saw in October, was that a combination of improvements in the like-for-like and the new space? Or is it, sort of, more reflection of the new space from the cruises as well as reduced, sort of, dilution from Geneva?

My second question is can you remind us to what the Mac inflation in 2018 and 2019 is? And are you hoping to, sort of, get an inflationary-neutral or less inflationary Mac situation on this, sort of, new contract terms?

My third question is what do you think has driven the improvement in Spain in the last weeks? And finally, you talk about cruise in Asia, only business is being slightly dilutable or the mix effect - the margin being lower.

Is that at the growth or is that predominantly at EBITDA?

Okay. The organic growth increase is - we are talking about days, in the first weeks of October, the 1% is mainly due to the new space and the comparable with Geneva last year.

We haven't seen all the detail, but still it's very premature to say something else. In my view, there are also two aspects that could be positive: one is the sales in Spain that are not negative so far or were not negative during this day so far and the improvement of the operation in Brazil, that is still negative, but is better than during the Q3.

Those are the main changes in the organic growth during the first days of October. Regarding the inflection in the market in Spain is public information.

This year has increased - the minimum annual guarantee increased by 6%. And next year, I'm not sure, but I think, it's another 6% in 2019.

Regarding the - if we were - if we are going to be able to overpass this inflection in terms of sales, it's obvious, in 2018, it's not going to happen, but I think, with the initiatives just started and the trends that we are seeing now, in 2019, may happen. Why?

Because the - obviously, the spend per passenger during the time that we have initiated this project that I comment on, the sales results are positive, but nothing concrete, because it still is very short time. The cruise business have two different levels: one is obviously impacting the top line, but also the EBITDA margin, as a business, is a lot lower than the EBITDA margin in any standard duty-free airport operation.

For division - and if you remember, I comment on that at the beginning of the year, we want, I think, 40 cruise lines and these cruise lines are going to be consolidated during 2018 and beginning of 2019 and this is - may impact the margin. But this is reflected in the guidance that we have commented on in Brazil.

Okay. So Julián, just two things, firstly, so the cruise, is it a lower gross margin as well as EBITDA margin?

It's not a gross margin, it's the concession fee.

Okay, excellent. And secondly, just about the Spanish Mac inflation, I mean, it's difficult to tell, obviously, with Aena or et cetera, but I am assuming that one of the things that you probably want to erase from any future contract is significant inflation in the Mac, right?

Because that's been fairly penalizing.

Yes. It's a very important factor, because in this contract, as you know, gradually, the Mac increases significantly, the first year that we took over the contract, the increase in the minimum guarantee was around 15% or 16%, and gradually, 6% per year.

We have seen this in '17, '18 and I think in '18. Obviously, the key point here, Rebecca, is if - that we can really drive more sales per passenger.

And personally, I am convinced that we can do it. The difference is that, in this case, we need to really show up and test it.

We - and we are doing this project together. They are very, obviously, interested that the top line is increasing significantly, I would say.

And we expect during the next 3 or 6 months, good results on this initiative. The first result, and I don't want to really identify these good results, because - the new initiatives, because we've been doing a lot of things during 2018 and probably some of these projects already impacted the sales.

The sales results are very - are positive. We turned around the situation of going negative.

Now the situation is flat to positive week by week.

Okay. But the improvement in Spain recently isn't - I mean, it's too small, right?

It's just pilot. So the improvement in Spain is what, because there's been a pickup in international passengers or?

No, the international passengers, it is more or less the same. As you know, the British passengers that is probably the most - not probably, is the most important customer we have in Spain, have swept from Spain to other destinations.

And to substitute these passengers with lower spend per passenger, like, it's happening with the Spaniards, is a big challenge, because you have, obviously, this is the proportion that when I said 10 to 2, it's the real proportion, it's not something - I am not guessing here, it's a real proportion.

Just 1 or 2 questions for. The first one regarding Latin America.

Do you expect a spillover effect in Latin America, also influencing the growth and organic growth in Mexico? And do we think this growth could also shift into negative territory, maybe in Q4 and also in H1 2019?

And then the second question, maybe, more hypothetical, but assuming that you intend to further increase your cash EPS also in the future, what would be an optimal level in net debt-to-EBITDA to start a new share buyback probably if you don't find a suitable target in Asia soon?

Okay. Regarding Mexico, being fair, I don't think that Mexico has the race to go through our process of negative EBITDA.

What happened during the Q3 - sorry, negative organic growth? What happened during the Q3 is a very more straight point is we have opened many shops, the traffic, in different terminals.

The type of traffic in Cancun has changed from terminal to terminal. We have been under renovation for a long time.

We have opened two new big shops, one of them was in a current terminal, the other one in a new terminal and this impacted the organic growth. This is - what I have seen over the past 2 or 3 weeks is, again, they are reinitiating the same growth done before.

Then is the classification of net-debt...

Yes, so look, I think, as I said, we are currently at around 2.9x, so that is at the higher-end of the range. So if you do the very simple math, so if we were not to do any acquisition for the next couple of years and assuming that we will keep the dividend as we have indicated now, we probably would be at around 2.5x in a couple of years or just below 2.5x.

So ultimately, this is a decision by the Board of Directors. This is not our decision.

But I think from a management point of view, I would feel comfortable that we say, look, we, kind of, give ourselves a year or a couple of years before we need to decide or before we really have - feel on pressure to do another share buyback. So I think we're in the sweet spot in the sense that if we can do M&A, we have financial flexibility to do so, but we also still have enough leverage that it will make sense to delever if there is nothing in the short term.

Julián, as you speak about the mix of changing with the rising shares of cruise lines having lower margins, this is also resulting that you put on the review your medium-term EBITDA target of 13%?

As I said, Jörn, it is very difficult to say today what is going to happen in the middle term EBITDA margin. I don't have the answer.

I don't know. I think what we have today is as a consequence of what we've done over the past year and I think the two acquisitions at that time justify the EBITDA that you mentioned.

The reality in terms of margin for the future, I prefer to have a clear understanding about 2019 and the impact of the new businesses in order to answer properly the question. I don't know.

Okay. That's all.

Thank you very much for participating in the presentation and in - and the call. All the questions were very interesting.

Thank you very much. Bye-bye.