菲利普莫里斯國際 (PM) 2017 Q3 法說會逐字稿

Good day, and welcome to the Philip Morris International Third Quarter 2017 Earnings Conference Call.

Today's call is scheduled to last about 1 hour, including remarks by Philip Morris International management and a question-and-answer session.

(Operator Instructions)

I will now turn the call over to Mr. Nick Rolli, Vice President of Investor Relations and Financial Communications.

Please go ahead, sir.

Welcome, and thank you for joining us.

Earlier today, we issued a press release containing detailed information on our 2017 third quarter results.

You may access the release on www.pmi.com or the PMI Investor Relations app.

During our call today, please note the following unless otherwise stated.

First, we'll be talking about results for the third quarter of '17 -- 2017 and comparing them to the same period in 2016.

Second, all references to total industry, PMI volume and PMI market share performance, reflect cigarettes and PMI's heated tobacco units for those markets that have commercial sales of iQOS.

A glossary of terms, adjustments and other calculations as well as reconciliations to the most directly comparable U.S. GAAP measures are at the end of today's webcast slides, which are posted on our website.

Reduced-risk products, or RRPs, is the term we use to refer to products that present -- are likely to present or have the potential to present less risk of harm to smokers who switch to these products versus continued smoking.

Today's remarks contain forward-looking statements and projections of future results.

I direct your attention to the forward-looking and cautionary statements disclosure in today's presentation and press release for a review of the various factors that could cause actual results to differ materially from projections or forward-looking statements.

It's now my pleasure to introduce Jacek Olczak, our Chief Financial Officer, for the last time on our quarterly earnings calls, as I'm sure most of you know that he will be assuming the duties of Chief Operating Officer on January 1, 2018.

Jacek?

Thank you, Nick, and welcome, ladies and gentlemen.

We are pleased by our third quarter performance, notably reflecting very strong currency-neutral financial results, including growth in adjusted diluted EPS of 11.2%; sequential improvement in our total shipment volume performance, supported by both cigarettes and heated tobacco units; higher total international market, excluding China and the U.S.; and the continued positive momentum for iQOS in all geographies, particularly Japan and Korea.

However, industry-wide dynamics in Saudi Arabia and Russia that we had flagged previously are putting pressure on our results and moderating our growth outlook for the year.

In Saudi Arabia, the significant excise tax increase in June, which resulted in the doubling of retail prices, is currently driving higher-than-anticipated declines in cigarette industry volume, especially in the highly profitable Premium segment, where Marlboro is the leading brand.

In Russia, cigarette industry volume is also softer than expected, while net pricing in the market remains constrained by the competitive environment.

We are, therefore, revising our 2017 reported diluted EPS guidance to a range of $4.75 to $4.80 at prevailing exchange rates.

Our guidance also now includes approximately $0.17 of unfavorable currency compared to $0.14 previously, due to principally -- due principally to the Egyptian pound.

Excluding currency and the favorable $0.04 tax item recorded in the first quarter, our guidance represents a growth rate of approximately 9% to 10% compared to our adjusted diluted EPS of $4.48 in 2016.

Our full year outlook continues to reflect a total shipment volume decline of around 3% at the low end of the 3% to 4% decline range that we expected earlier this year as well as currency-neutral net revenue growth of over 7%.

We do, however, anticipate a moderate decline in our full year adjusted OCI margin, excluding currency.

This primarily reflects the impact of the industry dynamics in Saudi and Russia, coupled with higher investments supporting the commercialization of iQOS, consistent with our aspiration for a smoke-free future.

Additionally, for the fourth quarter, we estimate a positive currency variance on our reported diluted EPS at prevailing exchange rates.

This is due to a favorable comparison related to the Egyptian pound, which had an adverse transactional currency impact on our results in the fourth quarter of 2016 due to its significant devaluation versus the U.S. dollar.

Let me now take you through our third quarter results in greater detail, beginning with our total shipment volume, which declined by 0.5% or 1.3% excluding inventory movements.

The sequential improvement in our total volume decline notably reflected heated tobacco volume growth, driven by Japan and Korea, as well as cigarette volume growth in Indonesia and Pakistan, coupled with a deceleration in the cigarette volume decline in the Philippines, one of our largest cigarette shipment volume markets.

We expect total volume growth in the fourth quarter, driven by heated tobacco units and despite the anticipated cigarette volume drag from Saudi Arabia, where industry volume declined by over 30% in the third quarter and should remain weak into 2018 and other Gulf Cooperation Council markets, which are expected to implement a tax structure similar to that of Saudi Arabia.

We recorded very strong currency-neutral financial results in the quarter, building up on our sequential quarterly momentum in the first half of this year.

Net revenues increased by 9%, driven by higher heated tobacco unit and iQOS device sales, notably in Japan, as well as favorable pricing of our combustible tobacco portfolio.

Adjusted OCI increased by 6.8%, primarily reflecting the impact of higher net revenues, partly offset by the increased investments supporting the commercialization of iQOS, particularly in the EU region.

Adjusted diluted EPS increased by 11.2%, supporting year-to-date September growth of 7.1%.

Please note that our third quarter financial results on a reported basis were impacted by the Egyptian pound, which depreciated by approximately 50% versus the U.S. dollar since the third quarter of 2016 based on average quarterly rates and contributed approximately $0.08 of the total $0.12 negative currency impact in our EPS.

Thanks to the exceptional performance of iQOS, our third quarter net revenues for RRPs reached $947 million and accounted for nearly 30% of our total net revenues.

Please keep in mind that the portion of this net revenues are from iQOS devices, which yield a negative margin due to introductory discounts offered in the initial commercialization phase to accelerate adult smoker switching.

While we remain in the early stages of our transformation to a smoke-free future, the size of our RRP net revenues confirms the exciting progress that we are already making on this journey.

Our pricing variance of $309 million in the quarter reflects positive contributions from all 4 regions and was driven by Asia and Latin America & Canada, in particular.

Our September year-to-date pricing variance of $1.1 billion came despite essentially no net pricing in Russia.

Turning to market share.

We recorded the second straight quarter of strong sequential growth in our total international share, excluding China and the U.S., driven by both our cigarette and heated tobacco brands.

Our international market share was also up slightly versus the third quarter of 2016.

I will now discuss a few of our key geographies, beginning with the EU region.

Total industry volume in the third quarter declined by 4.5%, in part due to estimated 2016 trade inventory movements related to the Tobacco Products Directive, mainly in Italy, France and the U.K. September year-to-date industry volume declined by 2.7%, consistent with our full year decline forecast of 2% to 3%.

Our regional market share, including cigarettes and heated tobacco units, was essentially flat in the quarter.

Share in Germany and Spain remained under some pressure, largely due to Marlboro's move about -- above round price points, which I have discussed in prior quarters.

However, France and Poland recorded strong market share gains, driven by Marlboro and Chesterfield, respectively.

Share in Italy increased slightly, driven by the strong growth of HEETS.

We have now grown our shares sequentially in Italy for 3 consecutive quarters.

Regional adjusted OCI in the quarter declined by 7.6%, excluding currency, primarily reflecting higher investments behind the commercialization of iQOS.

We expect a return to currency-neutral adjusted OCI growth in the fourth quarter, driven by higher heated tobacco unit volume and a favorable cigarette industry volume comparison.

Moving to Russia.

Total industry volume declined by 7.9% in the quarter, due largely to the impact of further excise tax-driven price increases as well as the recent growth in illicit trade.

For the full year, we now anticipate a decline of around 7% compared to a range of 5% to 6% previously, mainly reflecting the growth in illicit trade and the lower expected trade inventory movements at year-end due to a shift in the planned 2018 excise tax increase from January to July.

Our August quarter-to-date cigarette share increased by 40 basis points versus the same period last year.

The growth was driven notably by Philip Morris, largely reflecting the successful portfolio consolidation of low-priced local brands as well as adult smoker downtrading in the market.

Our quarter-to-date share also increased sequentially, growing by 10 basis points versus the second quarter.

As noted earlier, net price realization in Russia is a challenge this year due to the ongoing competitive environment.

In the Philippines, our profit growth continued in the third quarter, driven primarily by higher pricing.

Importantly, price increases at the bottom of the market, albeit delayed, have further narrowed the price gaps of lower-priced brands to Marlboro and Fortune.

Marlboro, in particular, has benefited from the narrowing price gaps, which contributed to a share increase of 3.5 points for the brand in the quarter.

While our total cigarette share declined by 2.4 points, it was up by 1.6 points versus the second quarter, reflecting share gains for both Marlboro and finally, Fortune.

In Indonesia, cigarette industry volume in the third quarter grew by 6.5%, primarily reflecting a favorable comparison related to inventory movements mainly associated with the timing of Ramadan.

Excluding this movement, industry volume was stable.

For the full year, we continue to anticipate a cigarette industry decline of around 3%, due mainly to the soft economic environment and related pressure on consumer spending.

Our cigarette market share declined by 60 basis points in the quarter, due primarily to the Sampoerna U and Sampoerna A, mainly reflecting the impact of price increases, partly offset by the strong performance of Dji Sam Soe Magnum Mild.

Share for Marlboro increased by 20 basis points, driven by the continued growth of our machine-made kretek Marlboro Filter Black offer, up by 1.7 points following distribution expansion, partly offset by the decline of Marlboro in the White segment.

This was mainly due to its price increase above the round price point of IDR 20,000 per pack.

In Japan, the spectacular performance of iQOS continues to drive our results.

Our total market share increased by 5.3 points to 33.2% in the third quarter, with HeatSticks up by 8.4 points to 11.9%.

HeatSticks is currently our largest brand in Japan and the second-largest brand industry-wide.

September year-to-date's total industry volume decreased by 4.1%, excluding inventory movement, consistent with the secular decline range for cigarettes prior to the introduction of iQOS.

Our retail offtake shares in Japan further highlight the success of iQOS, irrespective of geography and the presence of competitive smoke-free product.

HeatSticks closed the quarter with a weekly offtake share of 14.6% nationally, up by 1.9 points versus last week of the second quarter, with share gains across all areas.

Importantly, we are beginning to fully supply the Japanese market with HeatSticks and build normal inventory levels commensurate with the growth in demand, a process that we expect to continue in the fourth quarter.

As part of this effort, we began to process -- the process of shifting our HeatSticks shipments to Japan from air freight to sea freight during the third quarter.

However, we effectively remain supply-constrained in the market due to iQOS device capacity.

This limitation should gradually ease over the coming months, in part due to the increasing contribution of devices from our second supplier.

We expect to be able to fully supply the market with devices in early 2018 based on our current demand forecast.

The current constraint on devices also reflect a growing number of consumers, who chose to own multiple devices or who upgrade to the latest device model sooner than we had initially assumed.

Turning to Korea.

The exceptional heavy performance of iQOS continues.

National market share of HEETS reached 2.5% in the quarter despite the relatively limited distribution focusing on Seoul and other major cities.

The success has been driven in large part by high iQOS awareness, which exceeded 50% among adult smokers nationally within just 4 months of launch.

In fact, before iQOS was even launched in Korea, its awareness had reached around 20%.

Another measure of the early success of iQOS in Korea is the high level of full and predominant conversion, which reached 83% in September.

This is already above the 70% to 80% range generally observed in our more established iQOS launch markets.

Looking now at some of our iQOS launch markets in the EU region.

We are approaching and even exiting, in the case of Greece, a national market share of 1%, with solid growth compared to the third quarter of 2016.

As our weighted distribution in this market still only ranges from around 35% to 75%, this clearly implies higher shares within the areas where we are focusing our marketing and distribution efforts.

Additionally, in all 5 markets presented on this slide, we increased our sequential market share compared to the second quarter.

In EU, EEMA and Latin America & Canada region launch markets, where our focus remains more targeted, such as those presented on the slide, we're also pleased with our overall progress.

With the exception of Spain, where iQOS was only launched in the fourth quarter of 2016, we grew our focus area offtake share by at least 50 basis points in each market over the past year and also increased our share sequentially compared to the second quarter.

It is important to note that the positive momentum for iQOS outside Asia has been achieved despite the more challenging environment for building iQOS awareness and product comprehension among adult smokers, which is due largely to the stricter limitations on consumer communication.

Furthermore, adult smokers outside Asia who purchase iQOS generally have similar high levels of product conversion.

Turning now to shareholders' returns.

In September, our board approved an increase in our quarterly dividend to an annualized rate of $4.28 per share.

This marks the 10th consecutive year in which PMI has increased its dividend, representing a total increase of 132.6%, or a compound annual growth rate of 9.8%, since PMI became a public company in 2008.

Before concluding, let me share a few comments on the management changes and new geography segmentation announced on September 28, which are intended to drive the company's transformation towards a smoke-free future while maintaining its financial performance.

These changes should enable faster decision-making and a greater focus on both parts of our business, i.e.

combustible and the reduced-risk products.

Effective January 1, 2018, PMI will operate in 6 geographic regions, up from the current 4, as you can see on the slides presented.

A detailed split of the markets by region is included in the glossary of this presentation.

We will begin reporting results based on the new regional structure as of the first quarter of 2018 and plan to provide 3 years of historical data reflecting the new structure no later than our first quarter earnings release in April next year.

To conclude, we recorded very strong, currency-neutral financial results in the quarter, supported by a sequential improvement in our total shipment volume performance.

The strong momentum for iQOS continues.

To date, we have launched iQOS in key cities in 31 markets, and more than 3.7 million adult consumers have already stopped smoking and switched to iQOS.

Our revised 2017 EPS guidance reflects a growth rate of approximately 9% to 10%, excluding currency and the favorable tax item, compared to adjusted diluted EPS of $4.48 in 2016.

The strong full year outlook reflects currency-neutral net revenue growth above 7%.

Finally, we remain focused on generously rewarding our shareholders with our robust cash flow.

For the year, we continue to target operating cash flow of approximately $8.5 billion and capital expenditures of $1.6 billion.

Thank you, and I will be happy now to answer your questions.

(Operator Instructions) Our first question comes from Adam Spielman with Citigroup.

I have a couple of questions, please.

First of all, you -- so first of all, you've obviously seen good growth in Japan.

But nonetheless, slower growth than you saw in Q2.

And I'm talking about the market share growing 190 bps quarter-on-quarter on the IMS basis versus 290 last quarter.

And I was wondering, is that entirely due to the fact that you have supply constraints on the devices?

Or is it perhaps because you're now so large as you're growing market share is harder?

Or is it somewhat to do with competition?

That will be my first question.

I think the largest weight, I would put on the supply of devices, compound by the fact that we observe more and more consumers -- converted consumers who decided to own more than one device.

So that obviously creates additional bottlenecks because devices, which we ship instead of going to the new consumers, if you like -- I mean, they go to the existing consumers.

Obviously, there is some impact of some competitive products.

As you know, they are not nationally available.

I'm very pleased that even in the places where they are available, I mean, iQOS continues growth.

As I said, I would put the highest weight to the device availability rather than our dynamic in the market.

But you're right to say that the higher you are, presumably, it's more difficult to grow.

But I still will think we have quite the runway to deliver in Japan.

And then turning to Europe.

One of the things that strikes me is the big contrast in growth rates between Europe and East Asia.

And I was wondering if, looking forward, there is anything you can do to accelerate the growth or whether the strategy is to do what you continue to do, which is to grow market share but at a rate that's much slower than in East Asia.

We'd wish to grow the market share of iQOS in Europe at the level which we have in Japan and Korea.

Just the operating environment is different.

I mean, the marketing communication and the channels for the communication, which are available to us in both Asian markets, are not necessarily available -- are not available in Europe.

So we are in discussions also with the regulators, and we're doing our efforts because it is mainly based on a 1:1 market, if you like, and obviously takes time.

So if I would compare the efforts to -- which we're putting in place and the productivity of these efforts in Japan and Europe, they're pretty high.

Now I know that we -- and I am, in particular, in the company very excited about the iQOS.

But I have to, with my experience in this company, reflect on one thing.

All European markets, despite the fact that some people might perceive the growth of iQOS to be lower than, I don't know, expected than in Asia, are well ahead of any comparable product launch in a combustible category.

So we know that we're pushing the right product in the market, that we focus on the right product in the market, and there is no discussions that will continue with that.

I think 2018 -- as you noticed, Italy, after some period of time, is now crossing one share of the market.

I think 2018, in many markets in Europe, will be a turning point.

But we just have to continue to stay focused, invest in what we think is working, conversion rate -- rights from those who we approach, who purchase device on a very high level.

So I think it's just a matter of the effort and the results going to come, and there will be a tipping point -- or a turning point in these markets, where word of mouth will start playing a much higher role to the extent as we observed in Japan and Korea.

And one final question for me.

You said you will build inventory further in Q4.

Really, the question is, when we're looking at 2018, do you feel you need to build inventory of iQOS any further?

Or will you be comfortable that it's at the levels you want in East Asia and I guess, in Europe as well?

Yes.

First of those questions, we have highlighted this $2 billion -- a little bit more than a $2 billion inventory movement on -- essentially in Japan -- driven in Japan.

This number contains -- as you might have figured it out yourself, contain reduction of inventories on the combustibles, clearly because the volumes are going down, and bringing the inventories of iQOS to something that you all consider the sustainable level, which is the moment when we start unlocking device capacity, et cetera, which allow us to continue to supply the markets without any disturbance.

This is not the typical inventory buildup, which you have ahead of the launch and later on or next period, you take adjustments for.

I do not expect that the pipelining of inventory to Japan to the normal level, which would correspond to our anticipated demand for the product, would have to be reversed in 2018.

So this is not a classical thing, which we had in a traditional conventional business, that this inventory movement to the positive in one quarter or one period will resulting in a payback period -- negative payback period in the next period.

I do not expect this to happen with this one.

So yes, we're highlighting that we're shipping the product ahead of IMS, of in-market sales.

But we need to build these inventories.

Otherwise, we cannot run the normal operations.

And they should stay in our base, also in the base of our revenue growth for 2018.

Are you going to add to the inventory build in '18 I think is what you're assuming.

No.

We will build inventory to the desired duration.

And as you know, when you're targeting duration, it's always the question of the forecasted future sales.

We know what the outlook for the demand is.

And therefore, obviously, 3 months of inventory -- for example, 3 months of inventory today in Japan is much higher in absolute terms than the 3 months of inventory a year ago.

And that obviously creates optically or, in fact, creates the inventory movement.

But essentially, I'm just trying to keep the inventory -- build the inventory to what we think is the right level, taking into considerations that the product is subject to the surges in the demand much higher than any other launch, new products in a conventional business.

And also bearing in mind that, in the meantime, until we have a full capacity in other locations than Bologna, that we have to manage the risk of continuing to have supply in the market.

Our next question comes from Judy Hong with Goldman Sachs.

So Jacek, I wanted to just start with the combustible business.

So arguably, the competitive and pricing environment this year has been more challenging in some of your markets like Russia and Germany.

So I just wanted to get your color just in terms of what do you think is different this year and how you are thinking about whether you need to really shift your strategy there.

I'm sorry, the -- on a -- purely on a pricing front.

The bigger surprise to the negative, unfortunately, came from Russia.

Lack of a pricing for that long in the year -- net pricing -- net manufacturers' pricing just barely passing the January tax came as a surprise.

And due to the size of the market and our position in that market, that clearly put the big pressure on our numbers.

I mean, I have to admit that the beginning of the year, we knew that the pricing is getting a bit -- not as we would expect.

But I admit that at the beginning of the year, we didn't thought that it's going to last for that long.

If you want to see a rainbow, maybe you have to go for the rain.

So I would think it presumably should all go well for 2018 because the prices are moving slowly in Russia, so at least '18 should look much better versus '17 as we have now.

But Russia was clearly the only market on a pricing, which got -- which took us to the negative by surprise.

If I look at the German and a few other markets, if you're referring to some delay in the price increases, et cetera, it was always like this.

So this is not -- it's not that I am happy about it, but it's not something which is surprising.

I've seen it, so it's a little bit of the norm here and there.

And then, obviously, this whole GCC and Saudi Arabia, in particular, in the quarter, which is a mix of -- the prices went up.

So the industry passed on this very heavy -- very high tax increase.

But the market is reacting as it's reacting to the 100% price increase.

I think many investors have underestimated or underappreciated the size of the GCC in terms of the volume, but also profitability.

Saudi 2016 was a market of, if I remember correctly, around 32 billion units.

PMI, Philip Morris, has about 41% share.

Marlboro is half of 28%.

Actually, more than half of that share is Philip -- is Marlboro there.

Margins -- and this is where the devil sits, if I may use this word.

Margins in Saudi are well above PMI average margins.

And clearly, Marlboro is well above the Marlboro average PMI margin.

So when the market grows, at least at this stage, by 30% or plus -- or more than a 30% down, that starts putting quite a significant pressure on our financials.

I have to say it like this.

We have put at the beginning of the year this guidance of 9% to 12% EPS growth, and I know it today that if not Saudi and Russia, we would be flying well above 12%, well above 12%.

Knowing what has happened to iQOS, which comes very strongly, if I wouldn't have Saudi and Russia, we'd be flying well above 12%.

Okay, that's helpful.

So then on iQOS, I mean, it seems like you've upped the investments in EU pretty significantly sort of year-to-date.

I just wanted to get a sense of if iQOS profitability is sort of tracking breakeven, as you kind of indicated before.

And then as we think about the expansion plans in EU, how should we think about the phasing of some of these markets in terms of going to a fuller distribution and sort of the ramp that we should be expected to see in that market in 2018?

We have said that iQOS will breakeven in 2017.

And actually, in this quarter, iQOS already -- in the quarter, iQOS already contributed total PMI to the bottom line.

So as -- we're actually doing a little bit better.

Can you quantify how much?

Judy, maybe one day I will, but not today.

But we said that we breakeven.

We've -- essentially, we're coming to the breakeven in Q2, about -- and Q3 confirmed that we -- this is a net contributor to the bottom line, which is very handy, if you like, also from a perspective that we had this against Saudi and the Russia events.

So despite that severe adverse situation, we can still deliver, in my opinion, pretty strong result.

There is one component when we talk about the investment in iQOS.

There is this component of devices, right?

And we have been saying from the very beginning, these devices, to some extent, they're inflating, if you like, our revenues, but we're not making margins on the devices.

Quite opposite, okay?

Devices alone are a drag on our margins.

And if I would do the calculation and exclude the devices from, take last quarter, Q3, results, my margin -- my OCI margin would be up somewhere in the range of 80 to 90 basis points.

And what went -- on the one hand, it's a positive because if consumers are buying a second device, it means that we really built better even loyalty of consumers to stay with iQOS, which is very good also versus a potential competitive entries offering in the market.

But on the margins, it's put a little bit of pressure.

As you know, as of mid of the year or second quarter of this year, we started to lift the prices of device and targeting about $110 worldwide.

It will take us a time when the device price will go up in a market.

So we're trying to address.

But obviously, devices will not be as accretive to our margin than the bottom line as the HeatSticks or combustible product.

So we also have to be careful when we look at the numbers.

Got it, okay.

And then my just last question.

In Korea, so I know they're supposed to vote this week on taking the iQOS tax up to 80% or 90% of the combustible.

So I'm just curious, as you think about other markets and sort of the discussions you're having on the tax structure, is sort of the conversation shifting to there is a relative, I guess, the risk difference, and so there is some level of tax differential that the governments are pursuing?

Or do you think that the current tax rate going up to this 80% or 90% eventually is actually a little bit more disappointing?

Well, we're always wearing the opinion that the product warrants the different tax treatment than the combustible classical cigarettes.

So on the one hand, yes.

I mean, if Korea tries to move the tax up, I think they're recognizing that the product should have -- or should enjoy the different taxations than a combustible.

So we'll see where we land.

The decision is not final.

But my understanding is that the product will have -- or will enjoy, if you like, a lower taxation with combustibles, which, taking into consideration the risk profile, et cetera, of this product, is absolutely understandable from any aspects, public health policy, smokers, et cetera.

So as you know, in some countries, the tax differential is very attractive on RRPs or iQOS versus combustibles.

I don't think in the long, long term that level of a tax differential will be maintained if it stays perfect.

But I do think that there will be -- that this product will enjoy a lower taxation than the conventional cigarette.

Remember also is the growing understanding of the harm reduction principles in which the product innovation plays the important role.

Even the most recent FDA announcements squarely fit into this direction.

And I would assume that the tax policies, as many other regulatory policies, will do differentiate both categories to the benefit of the reduced-risk product.

Our next question comes from Vivien Azer with Cowen.

So I just wanted to revisit the guidance change, please.

Jacek, I very much appreciate your candor around the pricing development in Russia, seeing you're going in expectation as well as the big drag from the GCC.

But since, everywhere, you've opted to hold the 9% to 12% guide, so it seems to me that perhaps there was another lever that you were hoping might materialize as an offset to justify holding the high end.

So if you could just comment, please, on your decision to hold the high end of guidance up until now, given that some of these negative factors that you've called out kind of were known over the course of the year.

Well, if Russia pricing environment would improve faster or earlier, we essentially would be squarely in our 9% to 12% or in the -- we wouldn't have to go to the 9% to 10%.

Russia alone would clearly -- or shorter duration of this price situations in Russia, if you like, would keep us in the upper end of the guidance.

I said before that GCC and Russia, if this wouldn't have happened this year, we would be well above 12%.

I want to quantify this one, but I'm not talking 10s or 20 basis points.

We'll be flying above 12%, assuming that iQOS, obviously, stays as is, which is coming strongly -- is coming very strongly.

So as I said, it's very handy, on the one hand, that we have this great performance of iQOS, and it is coming at the moment when we have to somehow mitigate financially, if you like, the pressures which we have this year in Russia and Saudi.

Saudi will continue, at least, from the annualization perspective for the next year.

We'll have to see where the markets stabilize.

There are some other parts of the GCC -- of the Gulf Cooperation Council countries to change the excise.

UAE is changing now in October.

But Saudi alone is a 65%, and profitability is even more.

So I think the biggest identified impact and drag, we already have now.

Okay, that's helpful.

And then my next question, just on pricing more broadly.

I fully appreciate the nuance around Russia given the multiyear tax change, which has been quite aggressive.

But if I look at your business over the course of '17, with the exception of Latin America, it does look like, on the combustible business, price/mix realization has kind of decelerated across the board.

And this is a question I get from investors a lot in terms of just the sustainability of the pricing model.

So given that we're seeing that decelerating price/mix across a number of your geographies, I was just wondering if you could comment on the durability of the pricing model, please.

It's just Russia.

If I would add what normally would have -- what should have been, if you like, the level of pricing in Russia now, our pricing variance would be squarely in what we're always saying, 6%, 6.5% -- 6.5% actual.

That's presumably -- no.

We know that we're now below 6%.

It's still pretty strong price realizations.

It's just the missing component of Russia.

I mean, if all our countries -- I mean, there are countries which, for the ineffectiveness of tax structures, et cetera, pricing is always a bit of a questionable because it's a big trade-off between the volumes and the pricing, like Italy and France, for example.

But there's other -- the knowns type of a market.

Other than that, yes, I mean, it's -- there was some delays in pricing in Germany.

Frankly speaking, every time when we have -- over the last -- both the few years, have been taking prices in Germany, there were always delays.

So as I said earlier, it's not that it makes me happy, but this is not surprising.

I mean, Russia, I don't think anybody achieved a success in Russia with what has happened in Russia.

Let's put it that way.

Understood.

And my last question, just on iQOS.

Specifically looking at Germany -- and the reason I'm focused on that market is just because the premarket clinicals were so good relative to other European market.

It looks like perhaps you lost share on iQOS in Germany, if I'm reading the chart right.

You had 60 basis points of share in the third quarter.

I believe it was 80 basis points in the second.

No, no, no.

I don't think we lost share in any of this place.

The only place, which we have highlighted, which is growing -- has a growth, but this is a very, very -- compared to others, small growth, is Spain.

But that's the different thing.

But Germany, if you look at -- I mean, Q3, the numbers were, whatever, 0.6 for the quarter, 0.4 quarter before.

So remember that we just -- we do not present really in Germany.

We present it in a few cities, okay?

So the focus is there.

I've been recently in Munich, and I think the thing starts rocking slowly here.

Germany is not the -- I don't want to offend nobody.

It's not the fastest market in the world.

They produce the fast cars, but they -- things takes time in Germany.

Therefore, I'm very much looking to my new role that I can focus more on markets like Germany and others.

Our next question comes from Matthew Grainger with Morgan Stanley.

So if I could go back to just the heated tobacco category in Japan, generally.

You talked a lot about device constraints and some of the limiting impact that may be having in the short term on progression of market share, which, regardless, is still very good.

Could you -- I know this is a difficult topic to elaborate on, but could you give us any observations you have about the interaction between iQOS and competitive products that are launching in the market?

I think it's clear that it hasn't stopped your progression, but I'm sure that consumers want to sample the various products that come out.

Are you getting meaningful observations from consumers who are using iQOS and one of the peer products, and then making a decision potentially using them in a dual manner or switching back to iQOS fully?

Just any context you can share on that.

Well, there are obviously some interactions from iQOS users with competitive products and vice versa if they cannot -- especially if they cannot buy iQOS.

The feedback we're receiving is that taste-wise and overall experience, taste is very important because people are using the product for satisfaction, and satisfaction is taste.

IQOS seems to come very strongly.

So at this stage, to be very frank with you, I'm not that much worried about the competitive offerings.

But obviously, there always will be a trial, it's a new category.

Why people should believe in one product presentations in a category, other products in a categories, we always will have that sort of trialing, trying the other propositions.

But it's nothing really which would stop me at this stage.

As you know, it's innovation.

We are working also on further improvements of our product, and I guess competitors is exactly focused on the same aspect.

So the game is over.

So far we're winning, and I hope we will continue winning.

Okay, great.

And just with respect to share repurchases -- share repurchase and the dividend increase you just put into effect, could just give us an update on how you're thinking about the potential for repurchases next year?

I mean, currency, at least on our estimates, looks pretty much neutral.

And my sense is that you've always been hoping that it would sort of swing in a more positive direction before you would feel very comfortable pursuing repurchases.

But given good earnings momentum and the fact that it doesn't appear to be a headwind next year, does that make you more optimistic about the prospect of reinstituting a bit of buyback?

I think our focus will continue to be on the dividend even at the times when our balance sheet was or is under pressure.

We're still under pressure.

We still haven't recovered what we should recover for -- to be in line with our prescribed ratios coming from a credit ratings.

I don't think, in the near term, we will get back to the buyback unless there is a sudden reverse on the currency.

Currencies would have to come each time positive to us, i.e.

we would have to be, in the times of a very weak dollar, that we would recover fully the desired balance sheet strength and start thinking how to deploy the cash to our shareholders in other forms than just dividends.

I think our focus will remain on the dividend and reward shareholders with a dividend and hopefully, dividend growth.

Our next question comes from Michael Lavery with Piper Jaffray.

I was wondering if you could elaborate a little bit on some of your thinking on iQOS spending.

Has that increased?

And specifically, I'm curious if that's part of the guidance adjustment.

Or since you held your revenue guidance the same, is it just that your guidance has maybe -- it doesn't have an upper end, it's open-ended?

So is it just that you're closer to 7% then you were previously?

Or is it that you've kept the revenue guidance, seen the weakness in Russia and Saudi Arabia, but you've got an offset somewhere else that's lower-margin?

What's the right way to reconcile all those pieces?

Well, definitely, we'll deliver revenue above 7%.

And the upper end of the guidance always is on the revenue.

It's always a fact of how many devices we'll be able to ship and how many HeatSticks we'll be able to ship, et cetera.

I think we have a good view now on what's going to happen to the volumes in -- and the price in Russia, GCC, et cetera, so it's just on this one.

Remember, I think it was a quarter or 2 quarters ago, I said that we're approaching a time when we're above 7% in revenue.

And I do hope -- and actually, I think it will become a new norm for PMI.

So pretty -- as a mix of the volume growth and a pricing growth, I think it's a good quality top line growth, which the company is capable -- is able to offer now.

And obviously, we may be much higher than 7% this quarter was -- you consider quarter revenue growth.

Even if I was to be very aggressive and adjust the entire revenue growth by the growth, by contribution of devices, which, as I said earlier answering one of the questions, drag on the margins, at least at this stage, I still would be well above our past algorithm, if you like, the 4% to 6%.

The thing we're approaching with PMI at that time -- with this type of a top line growth should become a sort of a norm.

Not necessarily on quarters.

As you know very well, we never manage the business quarter-to-quarter.

We focus on the longer period of time.

But I think this is what we can -- when we deliver what we can achieve.

And just on the iQOS marketing investment spend piece of it.

Is that incrementally a higher number?

Or is that -- are your plans there pretty much the same as they were and it's just the event?

Higher than last year, obviously.

We're not increasing versus what we have planned for this year.

We're just implementing the plans.

There is this variable component, as I mentioned earlier.

It would seem that investments, one should also count the devices, if you like, or some loss on the devices.

And that's a variable component because consumers decides to buy more devices than one per person and obviously, somehow increases the investment.

But as I said, we're working for the pricing, and our schemes will address it.

But on the other hand, we're essentially -- we have a first manifestation of the loyalty of the consumers because why consumers will buy -- will purchase more than one device if they are not sure to stay with that product within a category?

So I would read it very positively for the future periods.

And that -- it means that we have a consumer, and the consumer is happy with the proposition.

As you know, we're rolling out the new versions of the devices.

I think this will become a sort of a norm that every period or so, there will be additional devices.

There is a growing demand for accessories for the personalization, et cetera.

I think in Japan, obviously, we'll be on the forefront because it's the largest, at this stage, consumer group to serve.

So there will be other revenue and hopefully also, margin, frankly speaking, opportunities which are ahead of us.

So the things which are coming a little bit faster than expected and may put some pressure, again, this multiple device ownership this year was a negative margin, fine.

I mean, we take a bit of pressure, but I think it's very good for the next period.

And just back on -- you've touched on how you've been able to get enough capacity to start building inventory in Japan and save on the air freight.

But do you still expect to be capacity-constrained through this year?

And if so, obviously, then that would mean that HeatSticks...

On the HeatSticks -- on HeatSticks, we should know -- remember, we have targeted year-end installed capacity, 50 billion.

And now with the latest plans and installations, et cetera, I think we should be above 60 billion year-end installed.

So we're already lifting the numbers.

We'll update the investors community in February of what is our outlook for the capacity for next year as we're still locking in some plans.

But that -- I think on the HeatSticks side, we start getting comfortable, okay?

Devices -- device, I think, as of this quarter, as of last fourth quarter, we should also be okay.

We activated the second supplier, which offers us even a bigger capacity than the first one.

So I think as of beginning of next year, assuming the -- our forecast for demand is, what we have is right, I think we should be -- we should have a first year of a sort of a normal operations when it comes to the supply side of the equation.

Okay, that's helpful.

And just one last one on Indonesia.

Your competitor, JT, obviously, is making a bigger push there with an acquisition.

Does that impact any of your thinking on what some of the competitive risks in that market might be?

Michael, I think you mean the Philippines, right, with JT?

Well, Indonesia specifically, where they bought a distributor and a...

Okay.

But what -- I don't think that -- no.

Well, I'm not saying no to acquisition.

I'm just saying that I don't think it's going to drastically change the dynamic in Indonesia.

Philippines, yes, actually.

It's a positive.

It happens to be also JT -- JTI.

So you see my focus is more on the Philippines than JTI in Indonesia.

Our next question comes from Bonnie Herzog with Wells Fargo.

A lot has already been asked and discussed, especially the stepped-up spending behind iQOS.

But I guess I wanted to go back to something you mentioned in your prepared remarks in terms of the required higher consumer spend to increase awareness.

I'm trying to understand why this might be different in some of the markets you're pushing deeper in versus some of your more established iQOS markets.

Because not in all markets we can say what we know about iQOS and what I think we should say to consumers so they can have access to the full information, what this product does versus combustible cigarette.

As you know, these -- regulations always are somehow retroactive, if you think -- if you like, in a sense that it's difficult to regulate the things which are not in the market.

They are in a market, and then regulators start thinking what to do.

So we're not operating in the regimes where it's clearly recognized that this is a heated versus combustible tobacco product.

And therefore, we can make XYZ-type of claims.

So we have to somehow adjust to the local market regulations conditions, et cetera.

We know it very well that the consumers who switch to iQOS, who adopt the iQOS, pretty quickly, they experience -- if they fully switch out of the conventional cigarette, they realize the benefit of switching to the product, okay?

I mean, the symptoms which they are experiencing are very much similar to those who have quit smoking, okay?

So they are positive.

So this is the word of mouth.

Consumers talk about these things.

But this obviously takes a bit more time.

The second thing is that in different markets, you have a different market in China that's still open, being the communication channels where I can advertise, where I can talk to consumers, et cetera.

Some markets, we only -- I give you the example of Germany.

In Germany, we can use outdoor, but very few things we can say on this outdoors by regulations.

I think there will be some development soon.

But for the time being, we are a little bit restricted.

In Italy, theoretically, we can say more things -- or practically, we can say more things.

But we don't have outdoor or print, and we're essentially restricted to the communications in the tobacco units.

Tobacco units is nothing else than the convenience store when people are coming for 20, 30 seconds transaction time.

So all of these markets, each of them has a different type of challenges, and we have to adjust our structures and infrastructure in this market to take the battle.

And as I said earlier, essentially, with our consumer one by one, knowing that there is a consumer number X which will be a tipping point, and then the whole thing will create a snowball effect, which we have created in Japan or very fast, actually, in Korea.

Okay, that's helpful.

And then just a final couple of quick questions on Japan.

I guess, first, could you update us on how the new HeatStick variance you introduced into the market are performing?

And do you anticipate more introductions?

And then on your combustible business, could you give us an update on the price increase and how you expect that to play out for the remainder of the year in that market?

The new variant, menthol with a flavor, I mean, it was very well received.

I mean, we're producing -- sorry, we're shipping what we -- we're selling what we produce.

So it's also behind -- there is obviously some, if you like, internal cannibalizations.

And now in the presence of more variants, consumers are switching from other variants.

Variant is responding.

That taste variant is responding to some taste preferences, which were existing -- coexist in a combustible business.

So we will try in some markets to, if there is an important segment, in terms of the taste segment, preferences, we'll try to cater to this need.

But so far, it's essentially Japan.

And with regards to pricing in Japan, well, you know the history, right?

We tried.

We ended up with a Marlboro price increase.

And I guess for the near term, it's going to stay, unless there are some changes around the taxations in Japan.

But I think for the long -- for the first time in the long history of PMI in Japan, frankly speaking, Japan is missing nothing in terms of generating the growth of OCI in the current situation.

Thanks to the fabulous volume performance, frankly speaking, we have a phenomenal OCI growth in Japan and somehow, pricing.

If it comes, it will be a nice cherry on the cake.

We obviously know our positions on pricing in a market.

But what was always the bottleneck in Japan that you cannot grow the market without the pricing, we actually found the solution that you can grow the profitability in the market absent the pricing, which normally should have happened but is not taking place.

Our next question comes from Chris Growe with Stifel.

So just a quick question for you.

As you think about the number you gave, roughly 60 billion sticks available -- HeatSticks available at the end of the year and that growing through 2018, I just want to understand how we should think about launches for iQOS in other markets, particularly in Europe, where you've got a base, if you will, for the product.

And I guess related to that, how do we think about Platform 2 and when that may launch or may, at least, go through test market?

Platforms -- I start with the second part.

Platform 2 will go to the test market this year, but we still have not announced where we're going to do it.

The 60 billion plus -- above 60 billion is the year-end annualized capacity.

So we do nothing on the capacity for next year, which obviously is not the plan.

60-plus billion, we already have in the pocket.

And obviously, we -- I think the machinery and activating the other factories than just Italy.

We actually have the first shipment from -- or productions and shipment from Romanian factories.

So I mean, it goes as per planned.

So as I said, we'll update on the capacity in February.

The capacity -- remember, initially, we targeted 100 billion year-end 2018.

And because now we will start the year with a higher capacity, it's obviously logical that the year-end installed capacity 2018 already will be lifted, at least proportionate.

Okay.

So I think with -- from that standpoint, with 15 billion sticks of availability in the fourth quarter, is that sufficient to start a national launch in another market?

Obviously, you have Korea building as well here, so it's becoming larger.

Yes.

You will have the market, well I mean [I could].

In all of these markets, we always balance where is the right moment when we want to go national, okay?

And we try to build a much bigger, larger iQOS communities within a territory.

And therefore, as I said very often, this national sort of market share are not really reflecting the pockets in which we focus.

And we have places in Italy and Switzerland, in Germany, et cetera, with local, if you like, market share, 2%, 3%, 4%, 5%, depends on the location.

But this is how we want to build the whole thing before we start spreading the resources too thinly around -- across the broader geographies.

You will have a market also.

We started already some developing markets, which are also very important in our strategy.

I think on the chart, we had Colombia, which came to the 1.4% market share in Q3, very strong start.

So we will be in a number of geographies trying to either open the market for the first time, usually starting with the 1 city, 2 cities, capital cities and in cities or in a market, where we are already presented with some cities to assess in which markets -- we think these markets are now ready for the national expansions, taking into consideration also availability of the marketing channels, et cetera.

So how labor and capital, if you like, intensive this might be for us to the next year.

But I think what we have created through to 2017 is a very nice base of a market to select from for 2018 expansion plans.

So I think this was our objective.

And some markets may go full speed 2018, some will go 2019.

We'll have now that luxury, if you like, of choosing how we want to push the pedal there.

Okay.

And then just a quick follow-up maybe on your total cost in the quarter.

Constant currency were up pretty strongly by our estimate.

Does that -- do you expect that to continue in the fourth quarter?

And should it continue to 2018 as you consider more of these launches and just more investment behind iQOS, in particular?

There is investment behind iQOS.

Remember, in the total cost, you also have just the impact of iQOS, right?

So on a purely COGS -- due to cost on the COGS -- due to cost, the COGS will be flat, marginally different, due to COGS.

Now if you take the volume impact, you will have a positive on the combustibles because we produce less.

And you have a negative, if you like, on iQOS because we produce more.

And actually, the negative on iQOS offsets the positive on a conventional.

Our next question comes from Jon Leinster with Berenberg.

A couple of questions, if I might.

Just for -- just a very specific one.

On -- in the Asia region in Q3, on the sort of conventional business in iQOS, the price/mix in Q3 seems to be -- to have sort of decelerated very sharply.

So I can understand EEMA with Russia and so on and so forth.

But why has the price/mix in Asia for the conventional business suddenly decelerated so sharply in the third quarter?

It would be driven by Philippines on volume, Australia on the volume/mix, and the rest would be on the positive side.

I'm just thinking about what drove the OCI and where the variance is coming from.

So you mean, Philippines volumes up, Australia volumes down?

Yes.

But Australia is presumably also the mix component.

Philippines will be on a -- should be on a positive mix because Marlboro is actually growing there, but volumes are lower.

And I guess there was in Australia -- if I remember, there might be some other smaller markets altogether.

Okay.

And on the -- if you're constrained on the device side, does that mean on the HeatSticks side in the third quarter, you could have produced considerably more?

On the HeatSticks side in the third quarter?

Well, if you're saying you were basically constrained by volumes and you were saying that the constraint is on the device side, does that, therefore, mean that you could have done well above the sort of just under 10 billion units?

Not really because the production is still ramping up.

If I will take the year-to-date production versus what we have shipped or sold, if you like, I think the delta is literally a couple billion, I guess.

So this is somewhere in the pipeline and the factory warehouse, et cetera.

So now we're shipping to -- because now we start having some level of inventories in Japan, Japan is not much restricted on the HeatSticks sales.

But obviously, they can supply the market with the HeatSticks.

But if you don't generate the new consumers because you cannot have devices, I mean, that is becoming the bottleneck.

So I guess -- as I said, Q1, I assume when the devices should be in the free supply and the HeatSticks will have an inventory and we'll continue to be in a full supply, we should start seeing the first quarter of operations result in a constraint on neither the device nor the HeatSticks.

Right, okay.

And also in discussion with Platform 2, I mean, should -- I mean, the capacity that you're talking about, the sort of 100 billion or, indeed, more, you have 60 billion at the end of this year, 100 billion or more end of next year, that presumably applies to Platform 1 or iQOS plus Platform 2 because it's sort of made in the same way.

Is that correct?

And is that likely to be actually a meaningful number for Platform 2 in '18 or not really?

Well, to some extent, yes because the -- there are 2 processes in making the HeatSticks.

One is the tobacco processing, but that process is shared between Platform 1 and Platform 2. So capacity can be shared.

When it comes to making the final product, the stick, the HeatSticks and the Platform 2 will have to have a different machinery.

So partially, the capacity is within these numbers.

Partially, capacity for the P2 is within this number.

But remember, we'll go to the test market this year.

We'll learn the -- it's fair to assume that we may need to modify something, and then we'll start thinking what are the right markets when we start launching the P2.

So I don't think it's that much of an issue for us, 2018, that we'll obviously have some capacity for 2018.

Then we'll have to decide how we want to expand that capacity earmarked only for P2 going into 2019 and beyond.

Just to be clear, though.

If the capacity -- if P2 -- on the secondary side, can you use the traditional conventional machines to produce the stick on the P2 so you could (inaudible)?

No, no.

It's different.

No.

We can use the same -- machine-wise, it's different.

We're always -- let's say the concept of putting a few components together and the formula route, which is a cigarette or HeatSticks, is the same.

Therefore, we're trying to -- we -- our objective is to do it as much as possible within existing facilities because we can use our people and the skills.

And therefore, the learning curve is clearly really picking up.

But machine-wise or CapEx-wise, if you like, this will require a separate program, a separate investment.

Right, right.

So if P2 picks up, we can expect another build-out of capacity -- a different build-out of capacity?

Yes.

But I think we also have learned how to build the capacity on P1.

So once you build the first RRP factories, remember that the quality assurance processes are different than in the combustibles, et cetera.

So very likely that P2 will be produced somewhere along the P1.

So I think it's a lot of -- part of the investments, which were made behind the P1 will serve the P2, okay?

By the way, net-net, it will require a separate, dedicated equipment.

And lastly, just in the switch from air freight to sea freight, particularly for -- obviously for iQOS, I mean, when it comes out of a factory, it's sold.

So it goes presumably to [TNS] or whoever it is.

And at that point, who gets the benefit of the lower costs of sea freight versus air freight?

Is that a margin benefit to you?

Or is that a margin benefit to TNS ? Or is it split somewhat?

This is Philip Morris International Inc.

We take the full benefit of switching.

So it's a lower cost issue.

Okay, fine.

It's us.

It's our cost because we cover the air freight.

This doesn't change anything on the revenue recognition because always somehow the factories, warehouse, et cetera, plus/minus.

But this is always at the center of manufacturing.

It technically creates -- or optically creates the higher inventories, which somebody was alluding to at the beginning of this call today.

Because when I have air freight inventories, how long is the plane flight is the flight from Italy to Japan, okay?

You could assume next day you have the cigarettes on the other side.

When I put the cigarettes on the boats, on the ships, then obviously, you count the 8 weeks or so of idle inventory, revenue is recognized, but inventory is not accessible to the market for further commercialization because it's on water.

So this is just a technicality of the trade.

But this is -- we're trying to put the HeatSticks to the same model -- or operations models which we have combustibles.

It's a significant difference in air shipments and the sea shipment -- air freight and...

Yes.

It's you that gets the benefit of the shipping?

Yes.

It's all to us.

It's nothing to whoever distributes or anybody else in the supply chain.

Ladies and gentlemen, we have time for one more question.

Our final question today will come from the line of Owen Bennett with Jefferies.

No, sorry.

My question has been asked.

Okay, have a good day.

Thank you.

Well, thank you very much.

That concludes our call for today.

If you have any follow-up questions, please contact the IR team.

We're here in Switzerland.

Again, thank you, and have a great day.

Ladies and gentlemen, thank you for joining the Philip Morris International Third Quarter 2017 Conference Call.

You may now disconnect your lines, and have a wonderful day.