Nick Rolli – Vice President-Investor Relations and Financial Communications Jacek Olczak – Chief Financial Officer

Michael Lavery – CLSA Bonnie Herzog – Wells Fargo James Bushnell – Exane Pam Kaufman – Morgan Stanley Judy Hong – Goldman Sachs Chris Growe – Stifel Adam Spielman – Citi Thomas Russo – Russo & Gardner

Good day, and welcome to the Philip Morris International Third Quarter 2016 Earnings Conference Call. Today's call is scheduled to last about one hour, including remarks by Philip Morris International management and the question-and-answer session. [Operator Instructions] Media representatives on the call will also be invited to ask questions at the conclusion of questions from the investment community. I will now turn the call over to Mr. Nick Rolli, Vice President of Investor Relations and Financial Communications. Please go ahead, sir.

Welcome. Thank you for joining us. Earlier today, we issued a press release containing detailed information on our 2016 third quarter results. You may access the release on our website at www.pmi.com or the PMI Investor Relations App. During our call today, we will be talking about results for the third quarter of 2016 and comparing them to the same period in 2015, unless otherwise stated. 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 with the potential to reduce individual risk and population harm in comparison to smoking cigarettes. 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. Jacek.

Thank you, Nick, and welcome, ladies and gentlemen. Following the comprehensive business review that we provided during our recent Investor Day event, I will focus today on a brief summary of our 2016 full-year outlook and third-quarter results, and highlight some important developments in select geographies. A detailed discussion of our third-quarter and year-to-date September results can be found in our earnings release. Let me begin with our full-year outlook. As announced this morning, we are reaffirming our 2016 reported diluted EPS guidance, at prevailing exchange rates, to be in a range of $4.53 to $4.58. Our guidance includes $0.35 of unfavorable currency and continues to represent a growth rate, excluding currency, of approximately 10.5% to 11.5% compared to our adjusted diluted EPS of $4.42 in 2015. As a reminder, in the fourth quarter, we anticipate strong currency-neutral net revenue growth, driven primarily by the annualization of price increases and the growth of RRPs. We also anticipate a favorable cost comparison, driven by the significant investments that we made in the fourth quarter of 2015 behind iQOS and our cigarette brand portfolio, including those related to the implementation of the EU Tobacco Products Directive. Moving to our third-quarter results, net revenues and adjusted OCI increased by 3.6% and 4.3%, respectively, excluding currency, reflecting favorable pricing, partly offset by negative volume/mix, particularly in the EEMA Region. Adjusted diluted EPS increased by 4.0%, excluding currency. Our favorable pricing variance of $440 million in the third quarter was driven by positive contributions from all Regions, notably EEMA. During the quarter we announced or implemented price increases in a number of markets including those shown on this slide which should further support favorable pricing in the fourth quarter. We recorded currency-neutral financial growth in the third quarter despite an organic cigarette shipment volume decline of 5.4%. The decrease was due mainly to lower cigarette industry volume in Argentina, Indonesia, the Philippines and Russia, as well as lower cigarette market share, notably in North Africa, the Philippines and Russia, partly offset by market share growth in the EU Region. Year-to-date September, our cigarette shipment volume declined by 3.9%, consistent with the year-to-date August decrease that I referenced at Investor Day. We continue to anticipate a similar decline for the full-year 2016. Our shipment volume for heated tobacco sticks reached 2.1 billion units in the third quarter, an increase of approximately 900 million units compared to the second quarter of this year. While principally driven by Japan, all iQOS launch markets contributed to this growth. While our cigarette market share, excluding China and the U.S., declined by 0.6 points in the third quarter, due notably to North Africa, the Philippines and Russia, Marlboro's cigarette share increased by 0.3 points, driven by the EU and Asia Regions. Let me now move to select geographies, beginning with the EU Region. Regional cigarette industry volume declined by 1.4% in the third quarter, contributing to a year-to date September decrease of 0.2%. For the full-year, we continue to expect a decline of around 1%. Our Regional cigarette share in the quarter increased by 0.4 points, driven by the continued strong performance of Marlboro, which grew share in four of the Region's top six markets by cigarette industry volume, notably Germany and Spain. In Indonesia, we recorded sequential market share growth in the third quarter, driven by our strong performance in the full-flavor machine-made kretek segment, notably following the geographical expansion of U Bold in June. To further enhance our position, in September we launched Marlboro Filter Black kretek in 25 cities. Although it is early, we strongly believe that this offer will set the standard for the full-flavor machine-made kretek segment. In addition, this month the Indonesian Ministry of Finance issued its 2017 excise tax regulation, which will result in a weighted-average excise tax increase of approximately 10% industry-wide. As you may recall, the weighted-average increase in 2016 was approximately 15%. In Japan, our cigarette market share continued to stabilize, and was essentially flat compared to the third quarter of last year. This performance is notable given the impact of HeatSticks cannibalization on our cigarette share, and was supported by recent Lark and Marlboro launches. iQOS continues to perform exceptionally well, with HeatSticks market share reaching 4.3% in the last week of September. In the third quarter, HeatSticks market share was 3.5%, an increase of 1.3 points compared to the second quarter of 2016. Importantly, these share gains have been achieved despite the limits we have placed on iQOS device sales since June. PMI's total combined market share, including cigarettes and HeatSticks, was 27.9% in the third quarter, reflecting an increase of 2.5 points compared to the same period in 2015. In North Africa, our third-quarter market share was 24.6%, representing a sequential increase versus the second quarter. This growth was driven by the improving performance of Marlboro in Algeria, highlighting the early success of the initiatives that we shared during Investor Day. In Russia, the Ministry of Finance submitted a proposal earlier this month to raise the specific and minimum excise tax levels for 2017 by approximately 10% above what is currently in the tax code. If the proposal is approved by the Duma and signed into law by the President, it will result, assuming full pass-on, in a total tax pass-on of approximately RUB11 per pack in 2017. The ad valorem rate of 13% remains unchanged in the proposal. I will finish my discussion of important developments with an update on the cigarette market in Argentina. While cigarette industry volume continues to be impacted by the significant excise tax-driven price increases in May, declining 14.1% in the third quarter, adult smokers appear to be adjusting relatively quickly to the higher retail prices given the broader inflationary environment. The average monthly decline in August and September was 8%. We continue to expect a full-year cigarette industry volume decline of around 12% in 2016, and a return to its low single-digit historical decline rate during 2017 once the May 2016 tax increase has been lapped. In conclusion, our third quarter results were in line with our expectations, and our full-year outlook remains strong. Thank you and I will be happy now to answer your questions.

Thank you. We will now conduct the question-and-answer portion of the conference. [Operator Instructions] Our first question is from the line of Michael Lavery with CLSA.

Good morning.

Hi Michael.

Could you just talk about Russia for a minute? We've seen the consumer there be pretty resilient in the past; you are now calling out some share pressure from pricing ahead of competitors. Are you seeing a change in the consumer approach and are they – is there any change in their willingness to spend? Especially, if so, could you give any sort of estimate of what you think the impact of the proposed tax increase might be as well?

I'll start with the last part. I mean that the impact as I said is about RUB11 at pass-on so it's around the same level as we had last year.

I'm sorry, could I clarify? I guess I meant with that increase do you expect any different category impact this year versus last year next year versus this year?

No, not really. I guess it's sort of the same increase as we had last year and there was this 1 percentage point change in ad valorem, but I don’t think it’s going to change the dynamics between the segments. What you see in Russia on the consumer level is, yes, the consumer is looking for value. He is looking for a value. No don’t question about it I mean that the premium segment, if at least judged by the share of Marlboro and especially Parliament, I mean it will hold pretty strong pretty well. There is down-trading. We observe a down-trading in the market, but more between the mid and the value segments. This is where the extra price promotions or delays in the price implementations are most frequent. We're also launching our own initiatives in this segment. So I guess also the share pressure which we have in Russia this year I think we should start being on a positive share territory in 2017.

Okay, that's helpful. And then, with respect to Japan I know it's maybe too early for you even to know but do you have a sense with the proposed indoor smoking ban, if that might apply to iQOS as well?

Not at this level of detail that go into these discussions, but I can say that we have a number of places in Japan where the existing smoking bans have been lifted due to the fact that iQOS is not a combustible product; therefore, there is no burning, there is no smoke. So we will have to see how this is going to evolve. But in a current environment with the restrictions for smoking, et cetera, as you noted, iQOS continues to gain the penetrations in the markets, so I don't think it's going to stall the growth prospects of iQOS in Japan.

Thank you very much.

Thank you.

Our next question comes from the line of Bonnie Herzog with Wells Fargo.

Good morning.

Good morning Bonnie.

My first question has to do with your volume and I guess I was hoping you could drill down a little further on the declines in the quarter. And then what gives you the confidence that you are going to still be able to hit the down around 3.9% for the full-year? Then curious to hear how that compares with your expectations for the industry volume for the full-year?

We wouldn't change the expectations for the industry volumes, which we have shared during the investors day. I think the trends are pretty much known at this stage I think there's some improvement will continue into the 2017. Specifically to us, I think and then I start maybe with the Latin America and Argentina. I think the consumer is adjusting to the prices pretty fast, obviously to the price increase. I think this high inflationary environment somehow, somehow helps. So I think sequentially we should see some less of the pressure from the volumes coming from or the clients from Argentina, Algeria which created quite the pressure on our volumes in Marlboro including Marlboro this year, we start seeing the first signs of recovery. It's going to take a while, but at least on the trend line we crossed the lowest point. I think already in Q4 and going into 2017 I think the pressure from Algeria and North Africa should ease Indonesia. I think the tax proposal, which came from the Ministry of Finance is manageable. You're looking for the just tax pass-on of about say 10% versus the 15% increase which took place this year, so I think this should be helpful. In additions to the fact that I think at least judging for the last few quarters performance of our share there, I think we should have a less – in fact, all share losses in 2017 than this year and hopefully on a bit stronger market than we have had in Indonesia this year. Philippines is a completely different story because it's all questions: how much we manage our mix to the positive territory. And so far we are very successful with up-trading the markets to Marlboro, which continues to grow and it's a very strong growth. Actually, we would like to enjoy this momentum for a longer while. Even if there are some pressures on the total industry volumes coming from the lower parts of the market, I think we would be fine with this. Pakistan, which also put quite a dent in our volume this year – it's presumably the most difficult to predict the market, because of the other forces which obviously play a role there. The contraband I think recently has reached about a third of the market, so the question is: where is the level of separation and that's it or we can expect worse than this. But again, from a financial perspective, it doesn't have that much of the impact. It doesn't have that much of a financial impact. So and Russia, the tax increase is also pretty manageable on the high side, but comparable, like we had this year. And as I said earlier, I think Russia should also turn the corner when it comes to share declines, so next year I guess on a similar sort of the market dynamics, we should have less of a volume leakage due to the negative share performance.

Okay, that was really helpful and I appreciate that. Then I do have another question I guess on your other brand volume, which was down considerably in the quarter. So I was hoping you could talk a little about your strategy with some of these brands. You touched on having a greater – you remember at the investor day you touched on having a greater focus on a fewer number of brands going forward, so I guess I'm assuming this is consistent with the sharper declines we have been seeing for some of the smaller brands. And I guess I was also hoping you could confirm that although these brands are declining rapidly, they are contributing nicely to margins as they are likely promoted less.

Yes, these brands will, in most cases, actually have a pretty low margin. The two big brands which are in our categories is to be a fortune, and very much a Jackpot in Philippines. So these are the brands which have a near zero sort of the margin territory and a lot of consumers are going from these brands to Marlboro, so we actually, through the positive mix, are improving pretty fast our financial situation there – our financial results there. We will have also brands in Indonesia in this other category which are not necessarily on the higher end of the margin spectrum. So I would rather say brand like a Jackpot in the Philippines we would presumably treat or we are actually treating this more tactically. Fortune is a little bit different. It's a brand which we do manage and we stay focused, but we do also recognize that, at least at this stage, Marlboro due to the closer of the price gap is a more attractive proposition for the consumer. And we are absolutely fine with that. In Indonesia, many of these local brands are the brands which we do support but we also try to focus on what is less dilutive to our margins.

Okay. And then maybe my final question, Jacek. You guys definitely walked through iQOS a lot at your investor day, but could you possibly highlight a few possible updates in the last few months on the improvements you've seen with conversion and maybe cannibalization rates?

Conversion, I don't I have a fresher number than what we have showed at investor day. There may be a few numbers which I still remember because some of these KPIs we are tracking weekly or even more frequently. I think at Tokyo – if you might remember, the Tokyo market share of the offshore at the investor day which we shared was 7%. We now stand at 7.3%, so we continue to growth and the same is for Japan. I think a national off take share the last week or the third week of September which we have shared during the investor day. 5.2 now turns to be 5.5. You see the pretty rapid share advancement; you’re talking 20 basis points, 30 basis points in a span of a couple few weeks. I have the same in Italy. When I think the share in Italy, the launch area was 60 basis points, 0.6. And now I think we lifted this last week of – sorry, first week of October I think the reading would be 0.8. Switzerland we lifted from 0.4 to 0.5. The penetration rate, the number of devices sold continues to grow. So I think in all geographies, from the top of my head, there was a continuous growth momentum very much in line of what we have shared with the investors at investor day.

All right, thank you very much.

Thank you.

Our next question comes from the line of James Bushnell with Exane.

Hi, good morning. Thanks for taking my question.

Good morning.

I have two questions, please. The first one is on iQOS in Japan. Clearly, you have limited the expansion because of the supply constraints you have. I just wondered if you could give us an estimate of what proportion of consumers don’t have access to the product right now and whether the existing users over-index to smokers of a particular strength or taste of cigarettes. That was my first question.

We see less and less of a proportional representation of former Marlboro smokers compared to the underlying share of Marlboros. But, clearly, the iQOS is making more or less significant inroads into the competitive combustible portfolio. Now how much really we have left unhappy consumers in Japan at this stage, I know that they are waiting release. I know that the consumers have to enroll by name of this release, but to be frank, I don’t have a number at this stage. The number is somewhere, but clearly we will see how much of that backlog, if you like, of unserved consumers we will have the moment when we leave the cup which we estimate to happen somewhere in the end of the Q1 next year when we will be very comfortable with our HeatSticks supply. But my gut feeling is we’re not talking about a 5% or 10% unsatisfied demand. I think the number is much higher than that.

Okay, great. Thank you. And then my second question, more of a modeling one, which is your associate and minority lines moved quite a bit since Q2. I just wondered if the numbers we are looking at in Q3 we could consider as more normalized, post some of the changes you’ve had in those areas.

I think so. There might be some movement. I think Russia should be – the Megapolis should be okay. I think maybe there’s some movement in North Africa around some of the structures which we have in Egypt, but I think we should be pretty much in line of what we could expect there. There are some currency – the currency movements might have an impact on this line. As you know, we have had some exposure in Egypt on the – we have to take a dollar exposure in Egypt and there are some limitations in availability on the dollars in the country, so we might have some distortions of these lines coming from North Africa, Egypt as I mentioned.

Okay, great. Understood. Thank you very much.

Thank you.

Our next question comes from the line of Matthew Grainger with Morgan Stanley.

Hi, this is actually Pam Kaufman on for Matt. I just wanted to ask on pricing, it improved sequentially to about 6.5% but needs to accelerate further to reach your benchmark of 6% for the full year, and presumably would help bridge the gap to strong fourth quarter earnings growth. Can you review where you have seen incremental pricing recently that would allow you to deliver on this target for the full year?

We should have in Q4 of this year the strongest pricing variance compared to all the quarters. We are looking in well in the north of $500 million pricing variance in the quarter and actually maybe even closer to $600 million. We should have – all-in-all in the fourth quarter we should expect above the 10% uplift on the revenue quarter on quarter. On an ex-currency basis. As I said, this – the growth in the fourth quarter, on the one hand, will be coming from the favorable comps which we have on the cost, while we still continue spending obviously more behind in iQOS, but this is more on the combustible part of the spending compared to the Q4 last year. But we also are looking into – we expect to have quite a significant uplift in the net revenue and corresponding contributions to the bottom line.

Okay, thank you.

Thank you.

Our next question comes from the line of Judy Hong with Goldman Sachs.

Thank you. Hi, Olczak.

Hi, Judy.

So my question is on actually the mix impact, because in addition to pretty healthy pricing, it sounds like the third quarter also benefited from strong mix. So how much is this geographic mix impact? And then, when you think about the impact on iQOS, either on price mix, is there a way to kind of quantify how that would’ve impacted either the price or mix in the quarter?

The raise is doing well so obviously due to the margin, geographical margin split – on a geographical basis the mix is working in our favor. If you go into individual countries, the countries with the worst mix in the quarter were actually Indonesia and, if I recall, actually was UK. But overall – but that’s it essentially. When it comes to iQOS, yes, obviously we’ve already start adding the support to the revenue growth and it’s going to be even stronger in Q1 of this year already. I guess when we close the year we will start talking more about the financial impact of iQOS, at least at the revenue level, but there is already some volume impact. You know we have a capacity of about $7 billion available this year. We sold a bit less than $4 billion, so I think we’ve been very transparent what one can expect from the volumes of iQOS in the fourth quarter and that’s going to help lifting the revenue growth quite a bit.

Okay. So if we think about it, I think iQOS added about 1% to your total volume so the revenue impact would’ve been higher in the second quarter? I’m so sorry, in the third quarter.

Well, the pricing is at the Marlboro level, as you remember, plus remember in the revenue line we also have devices because we are selling the devices as well. So you have an uplift from the volume at the Marlboro price in a given geography, plus obviously the taxes and also the price at which we have – the volume at which we’re selling the devices to consumers. They both contribute to the revenue.

Okay. Then just, in Japan, it seems like your obviously contribution from iQOS is still very strong, but your combustibles share has also gotten better. So can you just walk through how sustainable that scenario of actually seeing combustible improvement plus the iQOS contribution? Because it sounds like maybe Japan your volume would have been up high single digit, ex inventory, if you add the iQOS volume.

Yes, I think if you will take the full-year impact, iQOS clearly is already a significant contributor to our business there. Look, I have to actually thank you for raising these questions, because I should say how proud we are that in one year we managed to sort out the share pressure on our conventional business in Japan, despite the fact that part of that share we had to gain through the cannibalization to iQOS and, in addition, we saw successful with iQOS. I think the market share – the biggest dynamics now which you will have in a market I guess is the cannibalizations in both months between a combustible category, including own portfolio and iQOS. But the share leakage which we have to other taste segments or new taste segments, it seems at this stage that we have very successfully addressed. So it’s pretty remarkable results actually for our Japanese organization.

Okay. Then just finally on currency. I know at the investor day you talked about currency in 2017 being less of a headwind versus 2016. So just if we kind of hold spot rates at these levels, is there any kind of guidance you can give? A little bit more specific guidance on 2017, particularly around the yen, just given your hedged strategy?

Yes, I guess two questions actually. Sorry, two facts which would help maybe better predict the currency on your side of ocean. On the right is your elections impact decision about the interest rates so I am a little bit more remote from this one. I guess the number of currencies, if I look at how currencies were playing against us this year – and remember we started Q1 I guess it was at $0.19, Q2 we were $0.08 on EPS, Q3 we are $0.04. So clearly we could see that the number of currencies started to put less pressure on us. A currency, which I singled out during the investors day, which many of us sometimes tend to forget, we had some exposures in North Africa and in particular in Egypt. So the Egyptian pound is a little bit of the worry. I don't know how they are going to relax the currency rules and how much it's going to translate into the devaluation. Other than that ruble is doing pretty okay. Euro, Japanese yen, some of you know that we hedge; we already hedged some positions in 2017, but the yen already started to contribute positively in the last few months, in the last quarter definitely. So – I – we’ll have to wait until February when we will be post elections, post the Fed's decision and presumably these two big unknowns will be removed we will see how the market price the currency at that time.

Okay thank you.

Thank you.

And our next question comes from the line of Chris Growe with Stifel.

Hi good morning.

Good morning Chris.

Good morning. I had just two questions, if I could. I wanted to ask first of all, as we think about iQOS in Japan in particular and just given the constraints you have to your capacity right now, should we foresee further market share expansion in that market? Or are you kind of at a limit right now on how much you can sell in that market incrementally?

Well, we are still selling devices, not to the level where there is the demand for the device. As you know, without the device you cannot trigger the consumption of HeatSticks. But we somehow manage the growth or limiting the growth to the corresponding growth in the available production volume which we have. The capacity which we have, 7 billion, this year is not the 7 billion divided by 12 in equal installment per month. It's very much in adding back. So the peak of this annualized capacity is going to be in December this year. So, yes, we can add some devices, but we are not satisfying the whole demand for the devices from the consumers. And this demand is a default is newcomers – people who want to leave the combustible category, quit smoking and they move to iQOS, to heat not burn – but also existing consumers who want now a second device, et cetera, because they, for the convenience, whatever other purposes, they want to have more devices. By the way we are selling more devices week after week, quarter after quarter but not the extent or the demand is for the devices in the market.

Okay. Then my second question, just in relation to volumes overall, there was a lot of timing differentials in certain markets this year. Did overall the timing differential lead to a plus or a minus for volume this quarter? Do you have a good sense of what volume would have been?

I don't think a quarter is actually a good way of looking at things. There's always distortions. If I take nine months this year, year to date, I think the total distortion, if I remember, were in the tune of 200 million units by total volume. It's a lot of things which will net of over the period time, the timing of the shipments, the inventories, et cetera. If I take Q3 on the year-to-date basis versus Q3 year-to-date basis last year I think the total due to the shipment, timing of the shipments inventories would be about 200 million units.

Units, okay. If I could add just one quick one, in terms of the next nine or 10 markets where you have to announce for the launch of iQOS, should we expect that to come soon or would that be a separate announcement or? Certainly by the end of the year you indicated you wanted to be in 20 markets, correct?

Yes

Okay

But we’ve added recently Denmark. I think we added – we're about adding Holland, Canada I think we talked about this I think at investors day. It's getting dynamics to this point that I think we will be more giving of updates on a quarterly basis, what we have done in a quarter, than proactively start talking where we go because it's becoming a little bit like a map of United Nations.

Right. Okay, that's very helpful. Thank you.

Thank you.

Our next question comes from the line of Adam Spielman with Citi.

Thank you for taking my question.

Hi, Adam.

I just want to be a little bit – make sure that I have understood what's happening in the fourth quarter correctly. If I've understood it correctly, you said, first of all, there will be pricing variance of close to $600 million and then on top of that there will be benefits from reduced costs, firstly, relating to the European Tobacco Products Directive and, secondly, less spending on reduced-risk products. And then, finally – and I'm just checking that I heard you correctly – your spending on conventional products will not change. Is that correct? In the fourth quarter versus last year's fourth quarter?

You are almost correct, but not fully correct. The pricing variance will be up to the $600 million – $500 million plus up to $600 million you’ll have about $500 million or so in the comps. Not necessarily it means that the costs will be get down, but in the comps, Q4 on Q4, you will have in debt $500 million higher spending on iQOS, or RRPs, but less spending in that quarter on combustibles. But I think, for modeling purposes, I would just think you all have uplift in revenue through the price and support of the volume coming from either some improved trends, but very much for the iQOS and better comps on the cost Q4 on Q4.

Can you just – less spent on combustibles; is there any way of quantifying that, sort of percentage change?

No, but look, we have taken some costs ahead of the tobacco product directive implementation – there were some related costs in a number of markets in the EU. We accelerated that spending in Q4 2015 and most of these initiatives have already taken place. And in many countries the tobacco product directive now is in the physical product implementations in the shops of all infrastructure retooling of machines and all the other associated costs that have already been incurred. Therefore, optically it looks like we have less spending on combustibles in Q4, but this is in comparison to Q4 last year.

Okay, thank you very much.

Thank you.

Our final question comes from the line of Thomas Russo with Russo & Gardner.

Hi, Jacek.

Hi, Tom.

Hi, can you just remind us what the year-end likely capacity will be for iQOS stick productions end of 2016, end of 2017, and end of 2018? If you can give some measure of those. And did you say that the production lift that we are looking at present will kick in by the end of January of next year? And if so, when do you think you release the restrictions on device sales in Japan and other markets if we're up and running by the end of January?

I think the release on the cap on devices will happen towards the end of Q1 next year – of 2017. Capacity is always two numbers to – it's a pretty dynamic situation the two numbers to remember. We have available total capacity this year over 7 billion, but in December annualized, our capacity times 12, is 15 billion.

Okay.

From December that wants to figure it out, what is my underlying capacity for the month of January, for example? The first month after December it's 15 divided by 12. And then we adding 15 January, February and every month after that, so we will have a commercial capacity for the full year of 2017 of 32 billion, hopefully a bit more than 32 billion, and we will close 2017, i.e., December with 50 billion capacity. If you take January 2018, my monthly capacity is 50 divided by 12.

Perfect.

But if you remember then will be in the mode that we can add 4 billion capacity per month, if needed, if we start committing the capacity in the early part of 2017.

Yes, thank you.

Thank you.

Perfect.

There appear to be no further questions at this time. I would like to turn the floor back over to management for any additional or closing remarks.

That concludes our call for today. Thank you for joining us. If you have any follow-up questions, please contact the IR team, thank you again. And Have a great day.

Thank you, this concludes today's conference call. You may now disconnect.