(PM)
(PM)
Nicholas Rolli - Vice President, Investor Relations
Andre Calantzopoulos - Chief Executive Officer
Martin King - Chief Financial Officer
Michael Lavery - Piper Sandler - Analyst
Robert Rampton - UBS - Analyst
Vivien Azer - Cowen and Company - Analyst
Adam Spielman - Citi - Analyst
Pamela Kaufman - Morgan Stanley. - Analyst
Chris Growe - Stifel - Analyst
Gaurav Jain - Barclays - Analyst
Operator
Good day and welcome to the Philip Morris International Fourth Quarter 2019 Year-End Earnings Conference Call. Today's call is scheduled to last about 1 hour including remarks by Philip Morris International management and the 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.
Nicholas Rolli - Vice President, Investor Relations
Welcome and thank you for joining us. Earlier today, we issued a press release containing detailed information on our 2019 fourth quarter and full year results. You may access the release on our website at www.pmi.com or the PMI Investor Relations App. A glossary of terms, including the definition for reduced risk products or RRPs as well as adjustments other calculations and reconciliations to the most directly comparable US GAAP measures and our business transformation metrics are at the end of today's webcast slides which are posted on our website.
Unless otherwise stated all references to IQOS are to our IQOS heat-not-burn products. Comparisons are presented on a like-for-like basis, reflecting pro forma 2018 results, which have been adjusted for the deconsolidation of our Canadian subsidiary Rothmans Benson & Hedges Inc. Effective March 22, 2019, 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 Andre Calantzopoulos, our Chief Executive Officer. Martin King, our Chief Financial Officer will join Andre for the question-and-answer session. Andre.
Andre Calantzopoulos - Chief Executive Officer
Thank you, Nick, and welcome, ladies and gentlemen. Our 2019 results continued to reflect strong underlying business performances from both our combustible and smokefree portfolios. IQOS is performing strongly across a broad array of geographies and remains firmly on track to meet our 2021 heated tobacco unit shipment volume target of 90 billion to 100 billion units. Meanwhile, our combustible business continues to perform well, underpinned by solid pricing. We achieved several important milestones in our transformation to a smoke-free future. This notably included the authorization for a version of our IQOS product from the US Food and Drug Administration through the pre-market tobacco product application pathway and the subsequent commercial launch in Atlanta and Richmond under our licensing agreement with Altria. IQOS is now commercially available in 52 markets worldwide.
Let me now take you through the main elements of our full-year results, starting with volume. Industry total volume for cigarettes and heated tobacco units declined by 2%, broadly in line with the historical trend and slightly better than our prior forecast of around 2.5%. Our shipment volume declined by 1.4%, in line with our previously communicated forecast decline of 1% to 1.5%. The out performance of the industry was driven by heated tobacco share gains of 0.6 points, which helped offset cigarette declines, partly impacted by heated tobacco unit cannibalization.
Focusing on the fourth quarter, I would also like to highlight that while the effect of inventory movements across the year was small, the greater shipment volume decline of 4.4% in Q4, 2019 partly reflects unfavorable combustible cigarette inventory movements of 2.3 billion units compared to the prior year quarter, notably due to the European Union region and Japan. We anticipate a partial reversal of this effect in these geographies in the first quarter of 2020.
Fourth quarter in-market sales volumes declined by 3.1%. As previously anticipated in our third quarter remarks, this in-market sales decline primarily reflects the impact of recent pricing in Turkey and the Philippines. In Turkey, our cigarette volume decline was due mainly to the impact of two price increases during the year, taken in April and August totaling TRY5 per pack or 44%, which disproportionately impacted our share given the timing of our pricing vis-a-vis the competition, in addition to the industry decline. In Q4, 2019, our share improved on a sequential basis, but the total market was weak due to a rise in illicit cut tobacco consumption. In the Philippines, our cigarette volume decline mainly reflected a lower total market in the immediate aftermath of industry price increases, though from a volume/mix perspective this was partly compensated by a strong share performance from Marlboro.
Heated tobacco unit shipment volume increased by 44% to 60 billion units in 2019. This performance, including shipments of over 17 billion units in the fourth quarter, reflects broad-based growth across our launch markets, with notable contributions from the EU region, Japan and Russia, and keeps us well on track to deliver our 2021 shipment volume target of 90 billion to 100 billion units. This excellent performance means that heated tobacco units now make up nearly 8% of our total shipment volume, as compared to almost nothing in 2015. We expect this proportion to grow further as our positive momentum of reduced risk products continues.
Turning to our 2019 financial results, net revenues increased by 6.4%, excluding currency, driven by strong pricing for our combustible tobacco portfolio and growth in heated tobacco units. RRP net revenues reached $5.6 billion or close to 19% of PMI's total net revenues, with IQOS devices accounting for approximately $0.7 billion or 13% of RRP net revenues. As highlighted previously, this lower proportion of device revenues primarily reflects favorable geographic mix of heated tobacco unit volume, the longer lifespan of the latest IQOS devices and the impact of device retail price changes in select markets.
We recorded a strong combustible tobacco pricing variance of 6.5% in 2019, better than our initial expectations for the year, helped by a broadly rational excise tax and competitive environment. There were notable contributions from Germany, Indonesia and Russia, in addition to the aforementioned pricing in the Philippines and Turkey. This was in line with the average historical variance since the 2008 spin. On a currency-neutral basis, adjusted operating income increased by 11.2%, while adjusted operating income margin grew by an excellent 170 basis points. This margin expansion was driven by RRP scale effects and favorable geographic mix for the heated tobacco units. Additional drivers include pricing in combustibles and the impact of our cost initiatives, where we are firmly on track to reach over $1 billion of efficiencies by 2021, helping to finance growth investments behind RRPs. To this point, we implemented all our $400 million planned incremental RRP investments in 2019, with the net increase in spend partly offset by higher efficiencies realized as part of our overall cost program.
Adjusted diluted earnings per share increased strongly, growing 9.9% excluding currency, and exceeding our prior forecast of around 9.5%. As explained in our third quarter results, the lower currency neutral growth in adjusted diluted earnings per share compared to adjusted operating income reflects higher non-controlling interests and a higher like-for-like tax rate. Our strong operating cash flow of $10.1 billion increased by $0.6 billion, benefiting from a number of working capital initiatives and the timing of certain cash costs related to our Berlin factory optimization. Capital expenditures of $0.9 billion came in slightly below our full-year assumption of around $1 billion, benefiting from production efficiencies.
Turning now to market share, our total international share grew by 0.1 point to 28.4% with lower share for cigarettes, including the impact of cannibalization, more than offset by higher share for heated tobacco units, which reached 2.2%. The share of our cigarette portfolio declined by 0.5 point, reflecting continued adult smoker out-switching to IQOS, particularly in the EU region and Japan, coupled with lower share notably in Argentina, Indonesia, Korea and Turkey. Further, despite this out-switching, Marlboro's share of the cigarette category increased by 0.3 points to an all-time high of 10%, driven by Indonesia, the Philippines, Saudi Arabia and Turkey.
IQOS is now commercially available in 52 markets, representing 44% of the total international market, where our weighted-average geographic coverage within these markets is approximately 60%. This follows the addition of eight new markets in 2019, including Hungary, Sweden, the United Arab Emirates, the United States and the fourth quarter launch in Mexico, initially focused in select areas of Mexico City. In the later part of the year, we took another important step in our journey towards a smoke-free future with the launch of IQOS 3 DUO, which is now available in all of our international IQOS markets, supported by our new Simply Amazing brand campaign. The Simply Amazing campaign focuses on everyday relatable moments and emotions by showcasing to adult smokers who would otherwise continue to smoke the benefits of switching to IQOS. This latest addition to the IQOS family was designed with enhanced features to help adult smokers switch more seamlessly from cigarettes. IQOS 3 DUO allows two consecutive uses without recharging the holder, while its charging time is significantly faster compared to IQOS 3 and IQOS 2.4 PLUS. We have seen positive initial effects on rates of conversion and consumer satisfaction from the roll-out of DUO, which offers us encouragement on consumers' response to innovation as we start 2020. We continue to nurture a strong pipeline of future product improvements and innovations for the core IQOS platform.
Turning now to a more detailed discussion of RRP performance, we estimate that there were nearly 14 million total IQOS users as of year-end, representing the addition of 4 million adult users over the course of the year. Based on our current momentum, we expect this rate of acquisition to significantly increase in 2020. We further estimate that 71% of the total or close to 10 million IQOS users have stopped smoking and switched to IQOS, with the balance in various stages of conversion. This reflects widespread user growth, with highlights including a near doubling of users in Italy and Germany, a near trebling in Poland, very strong growth in Ukraine and the addition of 1.7 million users in Russia to reach a total of 2.5 million.
Let me now take you through the performance of IQOS in 2019. The overall share performance of IQOS heated tobacco units continues to see excellent progress. Indeed, in international markets where IQOS has been commercialized, IQOS heated tobacco units were the third largest tobacco brand in the fourth quarter with 5.5% share, increasing from the Number 4 position in third quarter. This has been achieved despite not having full national distribution in a number of markets, as mentioned earlier. The continued excellent growth of IQOS reflects progress across a broad range of markets. Our commercialization approach starts with a focus on the key cities within a market, building adult smoker awareness, trial and conversion. On this slide, we see the strong offtake share momentum in a number of key cities. This gives us further encouragement as to the potential future growth at the national level as we see increases in both our geographic reach and rates of consumer awareness and trial.
In the EU region, fourth quarter share for HEETS reached 3.2% of total cigarette and heated tobacco unit industry volume, an increase of 1.5 points or near doubling, compared to the fourth quarter of 2018. On a sequential basis, share growth accelerated in the quarter, increasing by 0.7 points. In-market sales volume also grew 16.1% compared to Q3, 2019. This growth reflects continued strength across a broad range of markets, as detailed in this slide. It is worth noting that IQOS is only present in geographic areas representing approximately 57% of industry total volume in the region. The quality of consumer acquisition is also high, with conversion rates and registration rates improving over time.
IQOS continued its strong performance in Russia in the fourth quarter, with HEETS share up by 3.3 points to reach 5.0%. On a sequential basis versus the third quarter, HEETS share increased by 1 point, while in-market sales increased by over 20% to reach 2.9 billion units. HEETS share growth in the quarter was consistent with the pace of adult smoker adoption and our geographic expansion. We are now commercializing IQOS in cities representing approximately half of the market by total industry volume, compared to an estimated 40% at the end of the second quarter. In Japan, our total reported share for heated tobacco units increased by 2.4 points to reach 17.6% in the fourth quarter, supported by the launch of IQOS 3 DUO and line extensions in Marlboro HeatSticks and HEETS line-ups.
As we have previously mentioned, total industry and share metrics in Japan are currently somewhat distorted by the low-priced cigarillo category, which grew rapidly over the past two quarters to reach a total tobacco share of 6% in December. These products presently enjoy a significantly preferential tax treatment, though the tax council of the ruling party has recently announced the closing of this gap over two steps in October 2020 and October 2021. While the growth of this category is likely to be temporary, we plan to enter with the Philip Morris brand in the current quarter to capture our fair share.
On a total tobacco view including cigarillos and adjusted for trade inventory movements, the share for our heated tobacco unit brands increased by 1.5 points versus the prior year quarter, and by 0.6 points sequentially to 16.9%. 2019 in-market sales volumes for our heated tobacco unit brands grew 4.2% compared to a total tobacco market, including cigarillos, which declined just over 3% after adjusting for the estimated impact of trade inventory movement. This helped drive growth of the overall heated tobacco category to a fourth quarter total tobacco share of almost 24%, including cigarillos and adjusted for trade inventory movements.
In Korea, there remains a lingering impact on the heated tobacco category of the consumer confusion caused by the Korean FDA's 2018 communication on tar. The category also remains highly competitive, particularly in the area of non-menthol flavors and related new taste dimensions that are also present in the cigarette category.
HEETS' share in the fourth quarter declined by 2.2 points, or by 1.4 points on an adjusted basis. However, on a Sequential basis, fourth quarter adjusted share was essentially stable, and our segment share continued to grow, supported by IQOS 3 DUO and recent launches that expanded the flavor line-up. While we are encouraged by this trend, we still have a lot of work to do to reinforce the heated tobacco category's benefits and build upon IQOS' leadership position.
A key development in 2019 was the launch of IQOS in the US through our commercial arrangement with Altria. The first IQOS retail stores opened in the initial launch markets of Atlanta and Richmond, marking an historic milestone and providing better alternatives to the 40 million men and women in the United States, who continue to smoke. While it remains early days in the lead market commercialization, we are excited about this significant opportunity. As a reminder, IQOS is currently the only heat-not-burn product on the market authorized by the US Food &
Drug Administration through the PMTA pathway as “appropriate for the protection of public health.” We also plan to seek an additional marketing order for the IQOS 3 device in the coming months, which will further support the efforts to convert US adult consumers who would otherwise continue to smoke.
Building on the success of our heated tobacco portfolio, we plan to launch our e-vapor product IQOS MESH 2.0 this year. Following consumer confusion around e-vapor in the later part of 2019, we took the decision to postpone our initial commercial launch. Unfortunately, these misperceptions, although moderating, still persist and as such we now plan to launch in the third quarter of this year when we will also reach the optimal capacity to deploy at scale. As a result, we plan a faster acceleration of commercial roll-out through the second half of 2020.
It's worth pausing here briefly to consider the dynamics of the international e-vapor market in which IQOS MESH 2.0 will operate. The existing market is concentrated in a small number of geographies, with 10 markets making up around 70% of the approximately 25 million international adult users, which despite numbering almost double that of heated tobacco contribute less than half its retail value. This reflects the heavy existing skew to open tank systems, a low degree of product differentiation and a low rate of full conversion relative to heated tobacco products. Of these 10 markets, seven also have a meaningful limit on nicotine concentration. The IQOS MESH product has been designed to address these dynamics and will be able to leverage our existing RRP commercial infrastructure and capabilities to drive consumer trial and repurchase.
In summary, we are optimistic that IQOS MESH will deliver a superior experience and drive further growth in our business over time.
Lastly, in terms of the commercialization outlook on RRPs, we also recently announced a global collaboration agreement with KT&G for the commercialization of their smoke-free products outside of Korea. The key rationale for this agreement is to accelerate the growth of the smoke-free category, which will require multiple products providing a wide array of brand, taste, price and technology choices to adult smokers. The capabilities we have built around RRPs in term of commercial infrastructure and know-how, technology, scientific substantiation and regulatory engagement are best-in-class. These capabilities can be leveraged to broaden our strong portfolio and innovation pipeline to further drive category growth, including partnerships with others when it makes strategic and economic sense to do so. KT&G has a range of smoke-free products, which we see as complementary to ours. We have responsibility for all elements of commercialization of products supplied under this agreement, and we intend to apply a market-by-market approach to deployment. The agreement will run for an initial period of 3 years, with the intention to expand the market footprint based on commercial success. While we don't disclose the financial terms of such arrangements, this is a royalty-based agreement. Products sold under the agreement will be subject to careful assessment to ensure they meet the regulatory requirements in the markets where they are launched, as well as PMI's high standards of quality and scientific substantiation of their harm reduction potential. We are now working towards the first launches later this year. There are no current plans to commercialize KT&G products in the United States. With this agreement and our own IQOS MESH launch in 2020, we have an active and exciting year ahead.
Turning to 2020 guidance, we forecast reported diluted earnings per share to be at least $5.50 at prevailing exchange rates, compared to reported diluted earnings per share of $4.61 in 2019. On a like-for-like basis and excluding an unfavorable currency impact of approximately $0.04 at prevailing exchange rates, this forecast represents a projected increase of at least 8% versus pro forma adjusted diluted earnings per share of $5.13 in 2019. In terms of quarterly phasing, we expect earnings per share growth and net revenue growth to be particularly strong in the first quarter and notably softer in the second quarter 2020. The first quarter will benefit from the timing of pricing in the Philippines, and a favorable comparison with first quarter '19, which included the absorption of a substantial excise tax increase in Turkey.
Conversely, performance in the second quarter will likely be impacted by both the comparison to a strong performance in the second quarter '19, and the most pronounced effect of the tax-driven pricing in Indonesia. It's also worth noting that the higher weighting of growth to markets with significant non-controlling interests seen in the second half of 2019 will continue into the first half of 2020 due to annualization effects.
Let me now explain the dynamics in Indonesia in more detail. As we said in December, the industry faces in a typical year of catch-up on excise tax and pricing. The 2020 excise tax took effect on January 1st, implying a weighted-average excise tax increase of 24% industry-wide, with a 46% increase in the minimum banderole price. While the potential tax pass-on is relatively steep at 14% of the weighted average price, in the context of a two-year stack with no excise tax increase in 2019, the average percentage increase is broadly in-line with historical levels. The tax increase and the higher banderole price should also give us the opportunity to address the price gaps, which impacted our 2019 share performance. So far between October and January, we have announced pricing of 8% on a weighted average basis, or approximately 60% of the pass-on. To date, some of our competitors are lagging on pricing, and will require a larger catch-up to comply with the new minimum retail price that is effective April 1st. We should also highlight the one-time impact on our pricing variance from the lack of any excise tax increase in 2019. With minimal pricing taken in the first nine months of last year, we will not benefit this year from the usual annualization effect of increases taken in the prior year. This exceptional dynamic effectively creates a one-off drag on our 2020 pricing variance of over $200 million. In line with our prior comments, we expect the total industry, measuring 307 billion units in 2019, to decline by around 6% to 7% in 2020, with the highest impact likely in the second quarter as the new minimum retail selling price takes effect and the new prices work their way through the trade. With the resumption of pricing this year, we expect market dynamics to improve in 2021.
Our 2020 guidance reflects certain key assumptions. The most significant, as we noted, is the pricing roll-out in Indonesia which at this stage we assume to dilute our combustible pricing variance to approximately 5% in 2020. While lower than our 2019 pricing variance, this is still very robust and is supported by a generally rational excise tax environment. In combination with the projected total market decline of 6% to 7% in Indonesia, at this stage we cautiously assume currency neutral net revenue growth of around 5%. However, I would reiterate that we remain very confident in our guidance of at least 8% currency neutral earnings per share growth. One of the reasons for this confidence is the positive margin outlook, where our efforts to deliver over $1 billion in annualized cost efficiencies by 2021 are bearing fruit, led by important initiatives related to productivity, cost category management, a project-based organization model and other cost efficiencies. In 2020, we expect cost efficiencies to fully offset expected net incremental RRP investments. With the growing scale and geographic reach of our RRP business, additional investments now primarily reflect variable costs linked to new consumer acquisition and existing user retention, plus new market entries. Consequently, the cost per user is improving, with an expected decrease of 25% in our acquisition cost per user in 2020. We expect the increasing leverage of our commercial platform to contribute to an increase in currency neutral operating income margin of at least 150 basis points in 2020.
Moving to the industry volume backdrop, our estimated total market decline is around 3% to 4% for the year. While this is weaker than the historical average of 2% to 3%, this very much reflects the temporary situation in Indonesia, which represents around 11% of total industry volume. An additional factor comes from the present growth of the cigarillo category in Japan, which is not included in the cigarette total market calculation. As shown on the chart, the assumed 6% to 7% market decline in Indonesia is an expected effect of around 0.7 points on the global industry, with a further 0.3 points from the growth of cigarillos in Japan. While there are the usual puts and takes elsewhere, together these two temporary impacts fully explain the deviation from the historical average, with underlying fundamentals remaining unchanged. An additional upcoming development included in our industry forecast is the Tobacco Products Directive ban on menthol cigarettes in the European Union, effective May 2020. Menthol and menthol capsule variants make up for around 10% of consumption in the region, and while we do not expect a significant impact on the overall industry volume decline of 1% to 2%, there is a further opportunity for IQOS given that menthol HEETS variants are not covered by the ban. We expect PMI shipment volumes in 2020 to again outperform the industry trend, and anticipate a total cigarette and heated tobacco unit shipment volume decline of approximately 2.5% to 3.5%. We expect our shipment volume trends to be slightly better in the second half compared to the first half due to the dynamics already mentioned in Indonesia and Turkey.
While we are not providing a specific 2020 target for heated tobacco unit shipment, we expect continued broad-based growth and remain well on track to meet our shipment volume target of 90 billion to 100 billion units by 2021. We anticipate a full-year effective tax rate of approximately 23%, consistent with last year, and a relatively stable net interest expense compared to 2019. We are targeting 2020 operating cash flow of approximately $10.5 billion, subject to year-end working capital requirements and currency movements. This includes the expected one-time adverse impact of new excise payment rules in Australia, and the timing of certain cash costs related to the Berlin factory restructuring. Together, these two factors account for an expected unfavorable cash impact of approximately $350 million. We project total capital expenditures this year to be approximately $1 billion.
Summing up a strong year, we made very significant progress in our transformation to a smoke-free future in both our organization and our 2019 business results. Just over 4 years since the first full-scale IQOS commercial launch in Japan in September 2015, we already have $5.6 billion in RRP net revenues, making up almost one-fifth of our business. Importantly, RRP revenues are increasingly accretive to our profits, as evidenced by the strong margin growth seen in 2019, and expected in 2020. IQOS continues to perform strongly, with 35% growth of heated tobacco unit in-market sales volumes from an increasing number of markets where close to 10 million adult smokers have already stopped smoking and switched to IQOS. We project the number of new users over the course of 2020 to be significantly higher than 2019 and at a cost per user that is lower, as the large RRP infrastructure investments of the past several years are mostly complete. We are also very focused on maintaining the leadership of our combustible tobacco portfolio, which is delivering robust performance and pricing power. Despite the temporary headwinds in Indonesia, we are confident in our 2020 guidance and remain well on track to deliver our 2019 to 2021 compound annual ex-currency growth targets of at least 5% net revenue growth and at least 8% adjusted earnings per share growth.
Finally, we are confident in our strategy for a smoke-free future and are convinced that our current and future RRP portfolio continues to provide us with the single largest opportunity to accelerate our business growth and generously reward our shareholders over time.
Thank you. Martin and I are now happy to answer your questions.
Operator
Thank you. We will now conduct the question-and-answer session of the conference. [Operator Instructions] Our first question comes from the line of Michael Lavery from Piper Sandler.
Michael Lavery - Piper Sandler - Analyst
Can you touch on IQOS and some of what's going on in really sort of two different ends of the adoption curve I guess. On the one hand, you see a market like Hungary, for example, that has 4 share points plus in just a year out of nowhere. But then you're also talking about the lower costs of acquisition per user that you've mentioned in the past comes from scale and growing word of mouth. I guess I'd love to just understand both dynamics a little bit more. One is, how are you driving some of the rapid adoption in select markets? And then second, what are some of the key drivers of your efficiencies as they get a little bit more mature?
Andre Calantzopoulos - Chief Executive Officer
Yeah. Well, I guess as a learning organization, every time we launch in a new market we apply what we learned from the previous one. So the time when we start growing faster is shorter and shorter. And I think we showed a chart like this recently and we'll explain that in CAGNY next week. So it's a learning organization. The deployment of more digital skills that makes that every new market we open starts growing faster than the previous one, and that's a fact of organization knowledge. Now as we have also the infrastructure in place, clearly the cost per user will decrease and thus decrease this year in 2020. We have to invest a little bit more money in retention. I would say although the cost of retention will also go down as we have more people. But clearly - and the cost of acquisition is higher.
So it makes sense now that we have enough people in our CRM base to invest a bit more there because losing a consumer they have acquired is much bigger loss than not acquiring someone by definition, okay? And also because now we have the digital infrastructure, we can start enlarging beyond over time the offering beyond accessories in just special editions and normal products, two more things that can help us build a different type of business over time as we have data on consumers and so on. So I'm optimistic that although we will continue investing, Michael, incrementally in the RRP. So if we grow the number of users, this cost per user is going to go further and further down, and, obviously, we retain more consumers. So that's all on the positive side, and that's how I see it, if I was clear.
Michael Lavery - Piper Sandler - Analyst
One just quick follow-up related to that, when you look at the evolution of IQOS usage of consumers, you've shown the charts that pretty consistently show around 70% of users who fully convert. But obviously, you're also adding new users constantly as well who have, obviously, just gotten started. What amount of time does it typically take for somebody to transition and maybe related to saying it kind of a different way,for users that began using the product, say, a year or two ago, is the amount that's fully converted more like 75%, 80% or 85% or how should we think about that dynamic?
Andre Calantzopoulos - Chief Executive Officer
Well, for the new users, clearly on average, we say two weeks to a month depending on the individual sometimes, people take much more time to fully switch, okay? They start using the products immediately, obviously, because they bought it. Now regarding retention over time, it's fairly stable I would say, and I think with the new programs we have in our CRM will increase this retention. Now depends also on the market as you know, in markets where you have a lot of competitive activity, we get some dilution of consumption because people buy other devices, typically vastly cheaper from competition and they try a few products, but if I compare markets where the competition is not yet successful, I think we have very good retention. And I think we can further enhance this retention as we now deploy a much better CRM program during the year. So I am pretty happy with what we have and we can improve from here.
Operator
Your next question comes from the line of Robert Rampton with UBS.
Robert Rampton - UBS - Analyst
Hello. three questions for me. The first would be in outside Korea and in Japan, are there any markets where IQOS volumes have declined sequentially and then on the flip side, are there any markets where sequential growth has accelerated?
Andre Calantzopoulos - Chief Executive Officer
Okay. I mean, except for Korea, I would say we have growth in every market, okay? Korea is a particular phenomenon because they - a confusion of consumers around the heat-not-burn category that is not dissimilar to what happened in many markets from the e-vapor category and that happened at the beginning of last year in June when the Korean FDA, although they found a very significant reduction in the toxicants of heated tobacco product compared to cigarettes, very similar to what we have communicated, they kind of confused consumer by say, yes, but the power of this product is very similar to cigarettes, although we're comparing apples-with-oranges because there's two different paths. The one is water and glycerine and the other is but chemical. o this has stalled the category and I think will take a bit of time to restart the process. I think sooner or later, there is a realization by the authorities in Korea that that's not the right thing to do with consumers and the alignment we have now much more with KT&G after the agreement I think that will help grow the category. But except for Korea, I don't see any other place where we have an issue.
Robert Rampton - UBS - Analyst
And then just in terms of the margin expansion. So if we're thinking about how much of your 2020 expectation is driven by say cost saves, pricing and IQOS, is it fair to think that the impact of IQOS is offset by cost savings and the incremental increase is pricing or how should we think about that?
Martin King - Chief Financial Officer
Yeah, the margin expansion is driven by a number of factors, first of all, it's partly mix in that we're growing very rapidly now in the EU, which has very good margins on IQOS heated tobacco units. The second piece is, as you point out, our cost saving initiatives have stepped up significantly. At this point, the investments that we're making. behind RRP are now fully offset by these cost savings initiatives and that's where you get the big benefit for margin expansion and the cost saving breaks down into a couple of buckets. In the operations area, we have various productivity especially as we get better and better at producing the consumables for heated tobacco units with scale, better up-time on the equipment for waste. We're improving the footprint across the whole manufacturing area.
You saw the factory optimizations we've done in the last year, which begin to pay off in 2020 and beyond and we also have initiatives across all of our spending categories to find ways to become more efficient intelligently and rotate money out of less efficient areas and be able to put it into reduced risk products. So that whole effort started to pay off, for the last year you saw our margin expansion, last year was very strong at 1.7 percentage point increase and then this year we're committing to at least 1.5 percentage increase, and this is something that we will continue to focus on in the out years going forward as well.
Andre Calantzopoulos - Chief Executive Officer
And regarding pricing, I would say, as I explained, we foresee pricing to be a bit below this year than 2019, that is essentially due to Indonesia, okay? Because if we took 2019, we had annualization of pricing from '18 in Indonesia, which at this stage we assume is not going to happen. Obviously, one, there was no real pricing in '19. So you have no annualization in '20 and we have to be a bit cautious about pricing and how it rolls out in Indonesia, there's a pretty steep increase, but we assume that this situation will resolve during 2020. So in 2021 we have annualization of pricing in Indonesia and come back to normal.
Robert Rampton - UBS - Analyst
Apologies. Just one follow-up on that. In terms of IQOS operating margins, can you give us some color on how it compares to cigarettes in say Italy at one end and Russia at the other?
Andre Calantzopoulos - Chief Executive Officer
Well, IQOS margins are higher than cigarettes, okay? So that's basically clear and a lot of the margin improvement is coming, as Martin explained, from IQOS essentially.
Operator
Your next question comes from the line of Vivien Azer from Cowen.
Vivien Azer - Cowen and Company - Analyst
So you guys sound very constructive on the momentum on IQOS, that is clear. I was hoping that you could comment on your key competitors' commentary this morning for the RRP category in Japan. Specifically, they may be have a different view than you, specifically saying that they're looking for more moderate growth to capture 25% of industry volumes as competition intensifies, can you comment on that?
Andre Calantzopoulos - Chief Executive Officer
I think it's very difficult to comment at this stage what is going to be the precise growth of the category, maybe I see it higher than them, but we see both growth and that's the important part. And I think we are well positioned with all the initiatives we have to capture the largest part of the growth. So we'll see at the year-out call where exactly we end. But the key thing is that we remain focused. IQOS still has a lion's part of the segment share and the momentum is positive and accelerating. We're increasing segment share and overall market share. So that's the key thing to retail.
Vivien Azer - Cowen and Company - Analyst
Just a follow-up on Japan. Can you offer any color on what your HTU mix is between HEETS and Marlboro at the end of 2019?
Andre Calantzopoulos - Chief Executive Officer
I don't have the exact numbers in my head, but I would say the cannibalization is stable. We had a little bit of down-trading I would say after the price increase because Marlboro increased prices more than HEET but now it's back to stable. And they are both growing but Marlboro is growing again.
Martin King - Chief Financial Officer
I think, Vivien, we haven't given the exact split, but I think it's safe to say that Marlboro's share, Marlboro HeatSticks share ended up being more than we had originally expected. We thought that HEETS would a bigger component by now, and Marlboro HeatSticks has held up better than we expected given the pricing. On the other hand, HEETS did its job of expanding their category and making the offering available to a broader array of consumers that were more price sensitive. So we're satisfied with its performance and an overall it did its job. However, more heated tobacco users opted to stay with the Marlboro HeatSticks than we originally planned, which is positive, of course, for our margins and for the profitability.
Vivien Azer - Cowen and Company - Analyst
Now that's a great outcome. Last one from me, Andre, I appreciated your commentary on some of the e-vapor concerns that are spreading seemingly to the EU. That's certainly consistent with what we heard from one of your competitors yesterday. Are there any principal markets that we should be watching out for, in particular, where consumer concern or confusion around e-vapor risk is taking hold most dramatically?
Andre Calantzopoulos - Chief Executive Officer
Well, I think the maximum was after the unfortunate cases in the US with the lung disease and death that were spun by some people in the wrong way as you know. I think this is getting better, but we are not there out yet, because there are still people that spread confusion despite all the efforts. Myself and others are making to explain that this had nothing to do with the vapor category and more with misuse with Cannabis oils and so on. My - our perspective on this was in any case, our plan on e-vapor was to build the capacity and go into the bulk of the market, that is the European Union towards the year end, okay, because then we have capacity at scale. We had planned a couple of markets earlier on a trial basis. I think it's better not to put more oil in the fire, especially since we will put the e-vapor under the IQOS brand.
So to minimize confusion, I prefer that we let a little bit time fly, we build capacity and it doesn't move in reality, our plan for the European Union, which was towards the end of this year. So that's my perspective. And I think that it's about time there are more voices here to stop misleading people. And I was reading the recent Q&A by the World Health Organization and it's not my commentary, but It is so much of misleading. And I would say malicious statements in there that some of the top advocates in public health came out and say this is unacceptable. So they modified their version, but they are still misleading. So I think short-term that may appear not a good thing, but long-term the more people say insane things, the more credibility they lose and the more the alternatives will become more credible because they lose their credibility. And I think we are in this phase now where frankly speaking, you see more and more of these nonsensical things and more and more voices saying let's stop this, have a reset now of the situation and stop talking seriously because these alternatives are better for people who would otherwise smoke. But as far as the commercial part is concerned, that's it. That's - it's what I said and I explained just now. Hope I am clear.
Operator
Your next question comes from the line of Adam Spielman from Citi.
Adam Spielman - Citi - Analyst
As Vivien was asking about Japan, can I just ask one quick question there before turning to IQOS. The question on Japan is whether in your pricing variance, you're baking in any assumption for the - around the tax rise that we're, obviously, going to get in October this year? So that's the first question.
Andre Calantzopoulos - Chief Executive Officer
Well, any, pricing is all in the guidance, obviously, Adam. So we'll always hope there is an opportunity in Japan when there is a tax increase to get pricing, we make some assumptions, we'll see how much they materialize or not.
Adam Spielman - Citi - Analyst
So the real question, the core question is, obviously, around IQOS. And so, one, - so thinking about the European to begin with, the European Union to begin with, in Q3, you had 0.1% quarter-on-quarter market share growth and then you've had a much higher, really, really strong growth in Q4. And I was just wondering how you should think - how we should think about EU market share progressing going forwards, whether there's any sort of seasonality in that and how you think we should sort of smooth out those few things, slightly disappointing in my view in Q3 and then a really strong performance just in Q4.
Andre Calantzopoulos - Chief Executive Officer
Okay. It's in broad strokes, okay, there is seasonality in summer, pretty obviously that's favoring cigarettes, hence a little bit less market share on the heated tobacco unit. The second thing although IQOS was very marginally affected by the whole situation of e-vapor in the US, it was just the moment in the quarter, we had a bit of a slowdown in September I would say but we recovered as of October, November, and I think we're entering 2020 with a pretty good momentum. Number of users is increasing, as I explained. I think the menthol is an opportunity for us at the - towards the second half of the year. So I'm very optimistic about the momentum and outcomes in the European Union.
Adam Spielman - Citi - Analyst
And so just to summarize, I think you'll say, you expect to get more new users in the EU in 2020 than you did in 2019, partly because of a medical issue, but also the CRM that just the whole momentum is so good.
Andre Calantzopoulos - Chief Executive Officer
Yeah, that's what I'm saying.
Martin King - Chief Financial Officer
Hey, Adam, I would also just - around the IQOS heated tobacco unit share in the EU, there is seasonality there. If you look at, for instance, in 2018 you see the same basic dynamic where the increase from Q2 to Q3 is relatively small with a bigger increase from Q3 to Q4 and then again in 2019 you see the same dynamic. In addition, this year we had even more pronounced effect that Andre mentioned with the other news flow, but this is back to the seasonality piece. Cigarettes in EU are going to be higher because it's warm weather, people go outside and they could can use it more and then lower in the winter months, whereas IQOS you could use much more easily, frequently without any worry as far as what the weather outside is, so you don't see the volume changes quarter-to-quarter as much as you do on cigarette. So it's really the cigarette seasonality that's driving these share numbers to be impacted.
Andre Calantzopoulos - Chief Executive Officer
And overall, I would say, Adam, we see a significant increase in the number of users overall across the market, okay. And I think it's also due to the fact that we're getting scale, getting momentum. We see more word of mouth happening in the markets and the tools we use now are most scalable. So I hope we will see this accelerate.
Adam Spielman - Citi - Analyst
And then quickly turning to Russia, which is, obviously, a very large market and you are gaining share very fast. Roughly speaking, well, in Q3 last year you had roughly 1% market share. You've now got 5%, so in five quarters you've gained 4 percentage points of market share. I guess the question is, do you think that rate of growth can continue, or would you expect it to slow slightly?
Andre Calantzopoulos - Chief Executive Officer
Well, first of all we need to compare comparable things, okay. We are still growing in cities where we started before, there is some, obviously, some expansion in there that helps. And we still have scope for expansion. Now it's pretty clear in my mind that at a certain stage, we would need probably a second price point in the portfolio in Russia because in certain cities, we will be getting the essence of the premium and somehow a large part of the more, I would say, wealthy part of the mid-price segment. I don't know but we will need this in 2020, but we have to plan at least for 2021, okay? And that's how I see it. So we don't see a slowdown, but we need to prepare for reaching the limits in terms of affordability over time in the big cities. Now in the cities, we just started, clearly we have more runway in front of us.
Operator
Your next question comes from the line of Pamela Kaufman from Morgan Stanley.
Pamela Kaufman - Morgan Stanley. - Analyst
So over the last several years, you've quantified a level of incremental IQOS investment in each year. How much incremental investment do you anticipate in 2020 relative to 2019?
Andre Calantzopoulos - Chief Executive Officer
Well, in my view it's going to be a bit less than 2019. And it's all variable. At the end of the day, it all depends on the number of new consumers you acquire. So I would say, would be between 300 and 350 depending on the number of new users that we have because there are certain costs that are purely variable but as Martin clearly said, all this is offset on a comparable basis by all the savings we have from our cost saving initiatives. So that helps clearly the margin expansion, plus the geographic mix of IQOS. That is helpful.
Pamela Kaufman - Morgan Stanley. - Analyst
And I wanted to better understand your decision to partner with KT&G and how you see their products competing with IQOS and given that this partnership expands your portfolio to e-vapor and hybrid products, do you also see a need to fill the gap in modern oral?
Andre Calantzopoulos - Chief Executive Officer
Yeah, I think at the end of the day, I've said many times, I don't think we will do all the new products on our own or invent everything and we're very open to partnership, and this is a very simple principle, it's what can we add to the partnership? I think we have technology, we have IP that's complements, we have the scientific assessment capabilities and, obviously, the route to market. And in my view, KT&G has an interesting product like the Hybrid product, which by the way is a peripheral heated tobacco product that vaporizes at the same time. When we think there is potential in certain markets and certain consumer segments, their heated tobacco product is the second more performing heated tobacco product and it's also heating from the inside and I think with the know-how we have, if necessary we can improve this product. And I think we decide market-by-market what are the right price points for this product, so that they are complementary to our IQOS strategy. And as we discussed in Japan, for Japan, there may be some cannibalization, but from Japan, for example, we did two price points with our own products. We didn't see much cannibalization. We're still much more incrementality. So that's how I see it.
For the e-vapor product, we have not major plans at this stage, but are much more interested in their hybrid and the heated tobacco with the team that has potential in these markets and that's how I see it. We also have more alignment I would say, overall, in Korea, that's important for the Korean market. And we also have access from a technology point of view to a country like Korea that is one of the centers for electronics. So I think it's a win-win for both companies. It's a license agreement, the financial works for both of us. So I look forward to working with them and expanding in the right markets, offering consumers a much more wide range of products.
Pamela Kaufman - Morgan Stanley. - Analyst
And can you give an update on the impact on new user adoption in Japan from your device price reductions and other initiatives to accelerate IQOS growth in this market over the last year? Are you seeing increasing heated tobacco adoption and trial among the harder to convert later adopters?
Andre Calantzopoulos - Chief Executive Officer
Certainly, we see some. I think the IQOS DUO will help because consecutive use for some people is important. We addressed this problem with the MULTI, but now we address it also with the main product and we see better adoption.
The pricing is also in general to avoid all these promotions and promotional money, we think that's the right price of the product and we always prepare the portfolio for the next generation that will always come at a premium. So we always reduce the previous version price in view of the forthcoming new version. And as I explained many times, we have a strategy of having a major innovation every two, three years and significant upgrades every year in the portfolio and that's what we're doing in Japan. So say that there is enormous reduction - reaction just because we reduced the device prices, it helps, but there is a whole mix that works. We see actually more reaction in Russia where disposable income is really an issue. By having still the 2.4 Plus on the market and IQOS 3, when still a lot of consumers buy the 2.4 because they are more price sensitive market. I think the answer is not only pricing, it's much more resolving constantly the consumer pain points that are a barrier to entry in the category and that's why we're working on the current version and the forthcoming one. So that's a little bit the philosophy we apply.
Operator
Your next question comes from the line of Chris Growe from Stifel.
Chris Growe - Stifel - Analyst
Just had a question for you and a bit of a follow-on from a earlier question. As you said, your cost per user and acquiring users in IQOS is going down, and I guess I'm interested in that scale effect you have in RRPs. You, obviously, had to spend more money this year but I think at the same time, I'm just trying to understand how you can leverage that infrastructure you have as you think about the launch of Platform 2 or MESH. Is a lot of the same or are a lot of the same resources being used for that? Are those - I assume they are figured as part of that $300 million to $350 million you cited for this year as you think about the launch cost of those new products.
Andre Calantzopoulos - Chief Executive Officer
Well, obviously, these new products, if they go through our infrastructure, which means our own retail, third-party retail that is kind of franchises, our own digital platforms now that are put in place, our coaches, obviously you can use the infrastructure you have in a much more efficient way.
So I would say that still the vast majority of the investment during the year is still on IQOS heat-not-burn and we assume even when roll out a bit more massively P4, it's going to be on a per user basis much cheaper compounded by the fact that we don't have to explain the category to people. We just need to explain to people what the difference is and what the benefits of MESH are compared to existing e-vapor product. So there is no category explanation effort behind. So it will be much cheaper in my view or less expensive than the heat-not-burn product.
Chris Growe - Stifel - Analyst
And then I had one of the follow-up question. And this gets to a question - a bit of a question earlier and your response. But you have seen this - a number of markets in the EU, where IQOS market share has really accelerated and we talked in the - couple few years ago about in Japan when you got over that 2% to 2.5% threshold, you start to see market share really accelerate. Are we at that point in the EU? Do we see that in the fourth quarter? Because some of those market share increases are quite notable. Are we at that stage is my question and are you seeing that in more markets, or do you hope to see that in more markets in 2020?
Andre Calantzopoulos - Chief Executive Officer
Well, we hope we'll see in more and more markets in 2020. As I said, we expect the number of users to grow, which means compared to - we had 4 million users, we expect a much higher number than that. So I call this acceleration on a same market basis, okay.
Look, it's the beginning of the year, we feel good about the fourth quarter. We have good momentum going into 2020. We'll keep you up to date on how we progress and that comes at a much lower cost. So I think we're building it pretty nicely.
Operator
We have time for one final question and that is from Gaurav Jain from Barclays.
Gaurav Jain - Barclays - Analyst
My question is on that Canadian subsidiary, so can you tell us where we are with the RBH creditor protection negotiations which are happening?
Andre Calantzopoulos - Chief Executive Officer
Well, they are progressing, but I cannot give you more details, okay. Still, there is a stay of all litigation and there are discussions to resolve the issue. Once we have something concrete, I can give you more details, but things are progressing.
Gaurav Jain - Barclays - Analyst
A follow-up on the IQOS MESH launch, so would you consider applying for a PMTA considering Altria changed some of the language around their JUUL agreement, if JUUL is not allowed in the market for a year and other things?
Andre Calantzopoulos - Chief Executive Officer
I think we have a lot of plans for P4. So one day, we will probably apply for PMTA in the US, but we still have to finalize a few studies on P4, for we have the same type of file as we had for IQOS, but that's important also for many other countries. And in due time, we will discuss if and when we want to enter the US market. Now the FDA is going to be very busy in the US over the next few months on all the PMTAs. I don't think that's the appropriate moment to do anything.
Gaurav Jain - Barclays - Analyst
And if I can ask one last question, can you comment on the sell-in versus sell-through trends for IQOS, both in the last quarter and for the full 2019?
Martin King - Chief Financial Officer
I'm sorry, the in-market sales versus the shipments or in your words, sell-in versus sell-out, were approximately equal. If anything actually the IMS was a little bit more than the shipments. So there was no inventory change on the year.
Andre Calantzopoulos - Chief Executive Officer
Yeah, it's in line with what we said for beginning of the year.
Operator
That concludes the Q&A session of today's call. Presenters, do you have any closing remarks?
Andre Calantzopoulos - Chief Executive Officer
Yeah. I would like to thank you all for joining first of all. I would say we had a pretty good 2019. We're entering 2020 in a very good shape. We outlined the one unusual thing that is Indonesia. We think we'd see how this unfolds. We prefer to be cautious about this at the beginning of the year. And just something that is in the air just now, I mean, this whole - this issue around the coronavirus. We have no problem today.
But clearly if that continues, there may be some impact on travel and the duty-free and in terms of supply chain, I think we're in pretty good shape for a few weeks. But if that persists for a very long period of time like everybody else, we may have some issues with device supply. Now that's not a continuation of this like in any company, is not baked into our guidance, but we all hope this is going to be soon resolved. So we put it behind.
So, thank you again and I look forward to a pretty good 2020.
Nicholas Rolli - Vice President, Investor Relations
So thank you very much. That concludes the call. If you have any follow-up questions, please contact the Investor Relations team here in Switzerland. Have a great day. Thank you.
Operator