葛蘭素史克 (GSK) 2017 Q2 法說會逐字稿

So welcome, and good afternoon to everybody here, and good morning to those of you who are either watching or listening in from the U.S. Today, we have announced our Q2 results, and I'm also setting out the priorities that we will drive to improve our performance and our returns for the long term, with the first phase focused on the next 3 years.

Of course, for the whole presentation, we start off with our cautionary statement as usual.

And in terms of the agenda this afternoon, Simon and I are going to share our priorities and financial implications for the next 1.5 hours, and we'll have a short break, followed by an extended 1.5-hour Q&A with the broader group.

Because today, we have a fantastic team from GSK here: Patrick, of course, who leads R&D; David Redfern, Chairman of our HIV business and also Chief Strategy Officer and leader of the strategic Pharma work.

Brian and Luc are here, the relatively new heads of our Consumer and Vaccines business.

I also have with me Jack Bailey, President of our U.S. Pharma and Vaccines business; Eric Dube, who leads our global Respiratory business; Deb Waterhouse, the new leader in our HIV business; and her head of R&D, John Pottage, and they've just landed here today directly from attending the IAS meeting in Paris.

And then to stay with our R&D team and to answer more specific questions you may have either in the later session or off within the reception this evening, we've also got another 8 key R&D leaders here in the room.

So I'm going to spend most of this presentation talking about our pathway to better returns.

But I want to start, standing in front of you for the first time as the new CEO, by sharing with you my view on the purpose of GSK.

This company does important work.

Our purpose matters: to help millions of people every day do more, feel better and live longer.

We prevent and cure disease in more than 150 countries around the world.

We keep hundreds of millions of people well, from the newborn babies through to the elderly.

Now this is the reason that most people join GSK.

It's certainly true for me.

And it's what drives the discretionary efforts of our employees and partners all over the world.

So this purpose will remain our ultimate priority under my leadership.

We have a long history, traceable back maybe 300 years in a variety of company configurations.

A history of helping change the burden and impact of disease through innovation.

And as we strengthen and modernize this company, this business, to set a platform for its future success, we must make sure that we drive performance, first and foremost, through applying great science to impact human health.

The demand for health care and innovation is more intense than ever as life expectancy, of course, continues to improve.

But we are, no doubt, in a challenging and fast-changing industry and environment.

The good news, of course, is this is a market full of opportunity, first and foremost, with fundamental advances in science, perhaps most notably in genomics, but also exciting new frontiers such as our own bioelectronics.

And in fact, with this acceleration of science and technology, we should all expect some material shifts in the way our industry operates and who our competitors and partners are as we use digital data and analytics to fundamentally transform the way we discover and develop medicines, the way we interact with patients and consumers and health care professionals.

Now as life expectancy expands, of course demographics drive demand.

But equally, they put pressure on budgets.

Now pricing pressure is absolutely not new in this industry.

It's long been a concern in Europe.

But it is now a major factor in the U.S. as both public and privately funded bodies look at ways to tackle the affordability of medicines.

And for us, this pressure is most obvious now and as we look into next year in the inhaled respiratory market.

And finally, it's hard to believe that the industry is not poised for more change.

Potential tax reforms in the U.S., pressures on performance on R&D output could all be drivers that affect their consolidation over the next few years.

And this could have, in turn, significant implications.

So it's important that we're agile and have flexibility if these changes should materialize.

So in this kind of health care world, GSK offers the opportunity to capture value and withstand the pressures across our 3 businesses.

We participate in a broad spectrum of the health care market.

If you think about it chronologically, from prevention with Vaccines, self-medication and wellness in Consumer Health, to high-end therapeutic treatment with strong life cycle management in Pharma.

So we are not reliant or over-dominated by one therapy area or product.

Portfolios in each of these businesses have innovative and established products, with leading positions in major therapy areas and categories, whether it's meningitis in Vaccines; Respiratory, pain relief and oral health in Consumer; or again, of course, Respiratory and HIV in Pharma.

Our recent transaction in 2015 created new scale for us in Vaccines and Consumer, and all 3 businesses now benefit from a global footprint capable of accessing growth in established and Emerging Markets.

This includes significant scientific, technical and regulatory expertise for the manufacturer of high-quality products.

And earnings and cash flows for the 3 businesses provide balance and a level of sustainability for the group's performance and its ability to invest in future growth and in returns to shareholders.

Recent performance has demonstrated the benefits of our 2015 transactions.

And in the last 18 months, new product growth in all 3 businesses has been stronger.

Good delivery of our new inhaled Ellipta portfolio and Nucala, our new biologic asthma treatment, is driving an effective transition in our Respiratory business.

And the success of Tivicay and Triumeq has reestablished GSK as a world leader in HIV.

Both are showing strong and encouraging Q2 performance.

In Vaccines, new meningitis vaccines, Bexsero and Menveo, are both growing strongly.

And whilst this business will continue to see volatility based on tender agreements, our supply position is stabilizing, providing good further Opportunities for growth.

In Consumer Healthcare, growth in our 7 power brands is ahead of the market, driven particularly by well-loved brands like Sensodyne and Voltaren.

We are seeing a marked slowdown in consumer market growth, and that's clear in our own Q2 numbers.

But while growth has slowed, we still very much believe we have a portfolio capable of responding to this challenge, and we definitely have the profitability levers in place to continue to deliver our target margin improvement of at least 20% by 2020.

That's around double the pro forma margin of 2014.

Sales growth and cost-saving measures, together with exchange rate movements, have seen cash flow and margin improvements for all 3 businesses in the last 18 months.

And the adjusted operating margin has improved by 400 basis points since 2015.

However, while this recent delivery is encouraging, it has to be set against the longer-term context for GSK's performance.

The reality is, longer-term performance has been weaker than we all would have liked.

Although recent new product momentum has been better, sales growth on the longer-term basis has been limited due to some slow starts or missed launches, the reducing contribution to profits and cash flows from our largest product, Seretide/Advair, and insufficient R&D output to compensate.

And supply issues have also been a clear factor.

All of this is clearly reflected in our long-term TSR performance, which has been uncompetitive when viewed against our peers'.

Our R&D performance is an important part of what we need to change and improve upon.

In fact, in output, we have had a significant number of launches, especially in the last 3 years.

And we've been amongst the industry leaders on volume metrics and have had some undoubted successes.

However, peak year sales per asset, the chart on the right here, have been among the lowest of our sector peer group.

So we have not consistently translated the output into commercial success.

Arguably, this is a combination of asset choices made, issues with the competitiveness of our labels and the competitiveness of our commercial execution.

In all cases, this requires a much more integrated partnership between commercial and R&D.

What is also clear is that comparatively, we've allocated capital in R&D to a much broader range of projects versus our peers.

And as an average spend per project was also low, we arguably didn't back some key assets with sufficient resources for strong enough data packages.

And development cycle time for GSK also appear to be longer than the average for the industry, notably in Phase II clinical development.

And we know that pace is critical in such a competitive environment.

So all of this points to the need for a significant overhaul and reevaluation of how we develop our clinical assets and, and most importantly, our commercial and R&D interface.

So alongside this challenge of addressing R&D development, there are several other key things we have to get right over the next 3 years.

All businesses -- all 3 businesses need to perform, but our priority is clearly Pharma.

It will start with some key changes to our operating model and portfolios.

We must maximize value from recent and new product launches.

We must address the pricing exposure in our portfolio.

This is acute in the inhaled respiratory market.

We have to ensure our pricing strategies and our cost base reflect the reality of the therapy areas and geographic markets we're operating in.

We must make the right choices to invest and develop early clinical-stage pharma pipeline for our next wave of innovation, which will come in the early 2020s.

This pipeline appears promising, but it is early and much of it is still unproven.

So over the next 3 years, we will start to see clinical data on many assets.

And how we get to the best decisions on investing behind them is, of course, of critical importance.

Beyond Pharma, we must realize the benefits of our newly scaled Consumer and Vaccines and, across all 3 businesses, put more discipline into our capital allocation processes

Our distribution of returns to shareholders has also been ahead of cash flows in the past few years.

So improving our cash generation for more flexibility to invest in future growth is also key.

As R&D data reads out, we have to make sure we can back our winners.

At the same time, we need to be ready for inorganic solutions to strengthen our pipeline or take advantage of other opportunities to strengthen our company such as the consumer put option.

Today, we are repeating and updating the 2020 outlook for the group, which was set out for investors in 2015.

And to deliver this outlook, we have a lot of work to do.

All of these objectives need to be met.

And as a reminder, our outlook is for sales to grow at compound annual growth rate of low to mid-single digits over the 5-year period to 2020 and for adjusted EPS to grow mid to high single digits.

And these outlooks were all given using constant exchange rates with 2015 as the base year.

And they do anticipate the launch of at least 1 Advair generic, adding to an already strong market pricing pressure in the near term.

But another very important point I want to make today is that these outlooks are also, given the potential development options in our pipeline, subject to data-driven R&D investment opportunities.

Of course, we've already made some assessments of the R&D investment that will be required going forward and built this into our plans.

But I want to be clear and transparent with all our investors that if additional investment is required to maximize full value of a particular asset, we will act with the long-term interest of the group's performance in mind.

So let me start to outline some of the changes we are and will be making to achieve these important goals.

First, we're going to have fewer and more focused priorities for GSK.

The environmental trends I already outlined at the beginning of this presentation are going to lead to fundamental changes in our industry.

Responses to them require agility and our own change of pace.

But they also need to be set alongside the long-term nature of R&D and responsible use of our capital.

So under my leadership, all 3 businesses will be tasked to be focused on 3 long-term priorities: innovation, performance and trust.

Now innovation is important to all 3 of our businesses.

But again, the top priority is Pharma.

We are very focused on maximizing commercial opportunities for our recent and near-term launches.

We've made mention of the 3 important launches to come.

The closed triple.

We know around 1/4 of COPD patients end up on triple therapy, and this will provide a simple, single-inhalation way of delivering it in a device that allows a move from single to dual to triple as needed.

The first of our HIV 2 drug regimens, potentially reducing the burden of drug treatment with a well-tolerated and highly efficacious combination.

And then lastly, Shingrix, we believe a step-change in efficacy for the prevention of shingles, a debilitating disease that's going to impact 1 in 3 of us in this room; 1 in 2 of us once we reach the age of 85.

Beyond the near-term launches, the big priority is to strengthen the pipeline.

And as I said earlier, our performance here has been hampered by allocation of resources and suboptimal commercial/R&D interface.

A combined commercial and R&D team have reviewed the full asset portfolio and investment priorities.

Now the result, we have developed a priority list of assets to invest behind and develop.

But of course, this list will continue to evolve as the data reads out.

We will continue to invest behind areas we -- where we already have strong leadership positions, Respiratory and HIV, expanding potentially from HIV into broader Infectious Diseases with our ongoing work in antibiotics and hep B. But these 2 are our core therapy areas.

And we will also focus on near-term investments into very specific assets in 2 potential specialty therapy areas, depending on data, of course: Oncology and Immuno-inflammation.

Now we have a number of programs in both these areas but, if clinically successful, could provide us with options to generate value ourselves or with others.

We're targeting 80% of our R&D capital to be allocated behind these 4 areas over time.

And we intend to pursue disciplined business development to strengthen the early-stage pipeline in these priority areas.

We will also continue investing in our core capability differentiators: our expertise in developing targets through genetics, modern medicinal chemistry approaches and advanced manufacturing.

We believe, for example, our cell and gene therapy technology platform could be a key manufacturing asset for GSK, and we intend to focus on deploying it behind our Oncology portfolio.

We will be part funding this reinvestment in more focused priorities by stopping investments in areas and assets where we see less opportunity for GSK.

And we've already made the decision to terminate, partner or divest 13 clinical programs and around 20 additional preclinical programs.

We've completed a strategic review of our Rare Diseases business and, for the assets currently in our pipeline, intend to secure opportunities for further development and generation of financial returns outside of GSK to give them the very best possible chance of successful clinical development.

We've also recognized that we need to improve the governance of our pipeline decision making.

And this starts with changes in processes and people.

Luke Miels, who's going to be starting with us in September, is President of our global pharma business, and his experience at Roche and Sanofi and AstraZeneca will be invaluable in helping to build commercial rigor into the way in which we prioritize our pipeline programs and define competitive target meds and profiles.

We've also recently appointed Tony Wood, who is joining us from Pfizer and will be responsible for leading platform technology and science.

We're undertaking a deep review of the efficiency and effectiveness of our drug development function, and we stepped up a new Development Advisory Board alongside and as well as the new Board Scientific Committee, to which I know our recently appointed board member, Dr. Laurie Glimcher, CEO of the Dana-Farber Cancer Institute, will be a fantastic addition.

So let me just give a few more details on the priority assets selected at this stage.

Our clear aim here is to develop the next wave of innovation, a new pharma portfolio for the early 2020s and beyond.

This slide lays out on the right-hand side the assets we are currently prioritizing for development in the 2 core and 2 potential areas.

Now clearly, again, we are very data dependent on progressing development on many of these earlier-stage assets here.

But in all cases, we believe they have the exciting potential to offer new, differentiated innovation to patients and payers and sizable peak year sales contributions to the group.

In some cases, we already have encouraging early efficacy signals, be it with cabotegravir when given by monthly injection, maybe even 8 weeks, as you would have seen in our data presented at IAS, or the exacerbation reductions in specific COPD populations, so for PI3K delta or danirixin, a good example where we've just recently accelerated our trial times by more than a year with real-time data analytics.

In our 2 potential therapy areas, we've also had some encouraging recent news, whether that's positive Phase IIb data on a new topical agent for psoriasis and atopic dermatitis, tapinarof, or in a strong early efficacy readout in multiple myeloma for BCMA.

About 20% of our discovery and development spend will be allocated to areas outside these 2 core and 2 potential areas.

So you can see here that there are still 2 late-stage development assets in other therapeutic areas which we believe, with sufficient differentiation, could still offer good commercial returns.

And of course, we continue to invest in discovery across a broad range of therapeutic areas.

We believe attaining this broad reach of the earlier stages of R&D is important for our longer-term success.

And Patrick and John and the team can answer all of your specific questions on why these assets are our priorities for us later on.

Their successful commercialization will, of course, be critical.

And with an improved R&D and commercial interface, I expect us to be making better decisions going forwards.

We will not be closed at all to exploring and achieving different routes to market for these assets, particularly with our potential new specialty medicines.

As we develop this next pipeline, there is a final critical point I would like to emphasize.

This portfolio will have the advantage of being developed at a time when the group's patent exposure is relatively low once we've digested Advair, with nothing material until the mid-2020s, something we believe is attractive when considering GSK's risk profile.

So I said we are reinvesting behind priorities and partly funding that by terminating, partnering or divesting assets that are better progressed by someone else, where our new data perhaps shows they have a lower likelihood of success or we do not see them to be of sufficient scale for us.

The 13 decisions that have already been made are listed here, and there are more, of course, to come.

We've also made the decision to terminate, partner or divest around 20 preclinical programs, as I've said.

And this is going to free up immediately some R&D spend and up to, should it be appropriate, 500 people to put behind the priority programs we've identified.

And just as the priority list will continue to evolve as new data come through, we should expect this list to evolve, too, with an ongoing reduction in the cost and time to stop.

And lastly on the pipeline, this provides a time line on data visibility, although clearly not visible in the room but is published.

And these data will inform our development choices and our investment strategies, both organically and in business development, should it be appropriate.

As they become available, we will be updating investors and presenting them to the scientific community where appropriate.

And obviously, the levels of capital we deploy in R&D will depend on these data.

Our second priority is a new, company-wide focus on performance.

And it must be sustainable, ethical and, over time as near-term pressures are digested, more competitive.

We need to drive growth with competitive products and competitive commercial execution and, more clearly, internally aligned objectives across all businesses and fuel that growth by managing our costs and improving our cash generation.

A key change will be a stronger focus on execution, driven first by a much better internal alignment, including of incentive, with integrated strategies for each of our 3 businesses, including across R&D, commercial, supply chain and any corporate support.

We will implement 1 P&L and 1 strategy with innovation, performance and trust priorities for each business.

We will be much more disciplined in our focus on priority markets and make operating model changes where it will improve results.

And Simon is going to talk much more on our improved capital allocation later.

But essentially, we are prioritizing resources to the strongest assets and geographies in our portfolios -- in our portfolio and moving capital and resources away from those that offer more limited opportunities for the company.

As you saw last week, we've initiated a strategic review of our CTLA-4 in antibiotic business with an option to sell and announced the divestment of some smaller nutrition brands in Consumer.

We've also decided to exit Tanzeum.

We will progressively withdraw support to allow us to complete necessary clinical studies and allow prescribers to transfer patients onto alternative treatments.

We have decided to terminate our rights to sirukumab as part of our efforts to focus our portfolio and reshape our Pharma business.

Across the whole company, we need to fuel our growth with a renewed cost discipline.

Overall, we believe we can generate a further GBP 1 billion in annual cost savings from existing and in new programs.

And a key part of this will be driven by improved efficiency in our supply chain.

We've already announced plans to reduce our site network across the group by 9 sites, and we're continuing to review and look for further opportunities to simplify the network in coming years where it makes sense.

Simplifying our portfolio is a key enabler of improving our network efficiency.

In the next 12 months, we expect to divest or exit more than 130 non-core tail brands within the Pharma alone, brands that could create complexity for our supply chain.

And these brands just in Pharma account for around GBP 400 million in annual sales and the plans were already in place for 90 brand exits.

Overall, we are looking for a 22% reduction in the number of Pharma brands we have.

We're also looking to reduce our overheads in manufacturing across the whole group with a simpler network and improved productivity.

And we will reduce our supply base in manufacturing by 25% by 2020.

And we're going to also generate greater year-on-year savings through better procurement, which we're going to unify for the first time in a single, company-wide procurement organization across all 3 businesses.

At the same time, we want to invest in our performance.

We intend to put commercial support behind our upcoming critical launches, and we're deploying capital in our supply chain behind our core growth areas.

As we announced last week, for example, we expect to invest more than GBP 140 million just here in the U.K. at sites supporting production of HIV and respiratory medicines.

And performance is, of course, driven by people.

You would have already seen that we've made a number of new leadership appointments, and I expect more to come across our top 200 leaders as we look to strengthen capabilities across the group.

One of the key capabilities I'm going to be most focused on is digital, data and analytics.

Our priorities here are to improve clinical outcomes and reduce time and cost to discover and develop drugs.

We're invested in several partnerships on large-scale genetics data and are seeing with learnings from the pioneering Salford Lung Study many possibilities and real-world evidence and also investing in digitally enabled clinical trials.

We want to develop valued real-world data at scale and, of course, step-change our customer and consumer engagement across all of our businesses.

This will be the starting point for a much wider intervention across the group as we look to harness technology to improve our performance.

And I was very excited yesterday to announce the appointment of Karenann Terrell as part of my team, as our new Chief Digital and Technology Officer.

She comes to us after being CIO of Walmart, will be a key partner to accelerate and modernize our capabilities here.

I'm a bit stuck on the slide.

Can we move it forward?

Thank you.

So it's true to say the biggest change in our company is going to be felt across our Pharma business, which didn't benefit from the catalyst to change that Vaccines and Consumer had with the 2015 deal.

And of course, Luke, when he arrives, will layer on to this Pharma change in due course as he appoints his own new leadership team.

We're focusing our commercial efforts on driving an improved performance in the U.S., which is, without doubt, the priority market.

But the biggest change geographically, in fact, is going to be about being more competitive in our Emerging Markets business.

And this has been a historic strength for GSK given our footprints, our strong brands and the very talented people we have working in these countries.

However, global and local competition have impacted our returns, and we need to structure more effectively and efficiently for long-term, profitable growth.

We need a model that can competitively drive what is today a largely classic branded product business with brands like Augmentin, but also one that can successfully launch more new innovations such as the Ellipta portfolio or Nucala.

To do this, we're going to create a new, single, dedicated, end-to-end operating model for Emerging Markets, spanning commercial, supply and R&D, for life cycle management.

The group will have its own dedicated governance model and the right commercial structure for each market, whether that's a stand-alone business, a cluster of similar markets or a distributor-led model.

Each market will be resourced accordingly, and we remain very committed to the access of our medicines.

Successful delivery of our performance priority also requires an improvement in our cost base.

The GBP 1 billion of annual cost savings by 2020 will be used to support investments in strengthening the pipeline and commercial execution.

The savings will also help protect our margins, which we expect to face sustained pressure as a result of pricing pressure, particularly in our inhaled Respiratory business.

And our third and final priority is trust.

We have to acknowledge there is a trust deficit in the world, be that in governments or in business.

And in our industry, where trust matters more than most, if we don't close that gap, it will impact long-term value creation.

And it has to start by getting our fundamentals right.

The most important thing we can do for trust is to discover and make great medicines that people need.

Then we must get our medicines to the people that need them, which means improving reliability of supply and our customer service levels.

Commitment to quality and safety across GSK is critical, and we will continue to invest in people and capability to deliver competitive and top levels of performance here.

And after some difficult compliance issues in recent times, we're also committed across the company to doing our best to prevent breaches and, should they occur, to respond to them swiftly and transparently.

But to build trust beyond the basics, the fundamentals, we are very focused on the way we engage with stakeholders and our contribution and responsibilities to access to medicines.

It's absolutely critical that our partners, customers and society trust our science and our intentions.

The way we engage with health care professionals is something we're both proud of and believe in.

We've evolved this already several times since its introduction, and we will continue to do so to make sure our engagement is competitive and trusted.

We also want to keep improving transparency with our other key stakeholders, and that, of course, includes our investors.

I much appreciated your transparency with me as I begin this job, and the team and I look forward to building a constructive and transparent dialogue over the years ahead.

I've already referenced the tough pricing environment, and it's clear we have to deliver differentiated innovation to deliver -- to achieve value and returns in this context.

When it comes to setting our prices, we have consistently taken a balanced approach with the recent launches of our new products, and this will continue.

We will look to support access where we can and generate sufficient returns.

We are very proud of our reputation and our long-term commitment to global health, but we will focus our resources to drive more global impact, especially in needs such as malaria and HIV.

And for our people, who are our greatest ambassadors and on whose talent and teamwork it all depends to support their everyday engagement, we intend absolutely to increase efforts to adopt modern, progressive employer practices focused on improving diversity and inclusion, supporting their personal well-being and being a flexible, life-friendly workplace.

Everyone in the room who are listening or watching knows the power and importance of company culture.

There are many special elements of GSK's culture, which includes very deep attachment to our values and our purpose.

But there are also some dynamics we have to change.

To be successful, we need to have a culture that is driven in equal measure by competitive performance and strong values, a culture where we always bring the outside in and we learn it from the very best, wherever they are in the world, and a culture that is focused on better decision making and cost and cash discipline.

So there are some key changes we will be making.

This will include new appointments, some of which we've already announced, with more to come over the next 12 months.

And I've also highlighted some of the operating structure changes we're making.

We're changing how we set objectives internally and how we manage and measure performance.

We will be introducing new company-wide incentives next year better aligned to our IPT priorities and our overall business performance and, for example, including cash flow measures across the company.

And while we will be keeping our values, we are going to be changing the company expectations of employees with 4 new expectations.

For everybody working in GSK, we will expect courage of decision making, accountability for results, prioritizing of people development and effective teamwork.

Expectations, of course, would have to start at the top of the house.

Before I conclude, I want to outline our capital allocation framework.

These choices will be key and again start with the IPT focus, driving improved cash generation.

And our priorities for use of cash are laid out here.

First will be to invest in the business in 3 key areas: the Pharma pipeline, organic but also early-stage inorganic; realization of the consumer put option should it come, which, as I said earlier, would strengthen our position in Consumer Healthcare; and further investments to expand capacity in our Vaccines business.

Our second priority will be to deliver returns to our shareholders through payment of dividends.

We continue to expect to pay a dividend of 80p for 2017, and we've also announced our intention to pay 80p in 2018 as part of a new dividend policy established for 2018 and beyond, where regular dividend payments will be determined primarily with reference to key -- free cash flow generated after meeting our requirements for investment in the business.

And lastly, we will use cash for business development purposes, with M&A obviously dependent on the right kind of return profile.

So in conclusion, we're setting out a clear path forward over the next 3 years.

Across the whole company, we will prioritize focus on innovation, performance and trust and by striving to get the right balance of change and strong commercial execution.

We expect to deliver the outlook set out for 2020, with adjustments to the Pharma margin targets and the Consumer sales growth rates.

This will reflect good execution of the plans we have laid out today, combined with investments we expect to make in R&D and the divestments and exits we've also outlined.

We will navigate some near-term pressures and the structural portfolio changes through the next couple of years especially and, at the same time, build a new platform for future growth for the decade beyond that.

Our aim is to have all 3 of GSK's businesses delivering competitive performance with clear pathways for delivery of long-term, sustainable growth.

Improved R&D and strengthened innovation will be at the core of this, using our science, technical know-how and our talented people to produce differentiated and much needed medicines, vaccines and brands that will make meaningful differences to patients and consumers and achieve good returns for our shareholders.

With my colleagues here today, and on behalf of everybody else working at GSK, I want to say to our investors that we are committed to achieving this.

And with that, I'm going to hand over to Simon, who's going to take you through the detail of our Q2 performance and the financial implications of these longer-term outlooks.

Thank you.

Simon, over to you.

Thank you, Emma.

So before moving to the outlook for 2017 and 2020, I'm going to comment briefly on our Q2 results.

As a reminder, I covered these results in detail in a video issued alongside our press release earlier today, both of which you can find on our website.

So I'll keep my Q2 commentary over the next few minutes in a relatively high level in order that we can focus time on the future and the implications of the announcements we've made today on our financial outlook.

So first on our total results.

These include some significant charges that reflect better prospects for the group and the implementation of our new business priorities.

Specifically, we have again increased the estimated valuations of our Consumer and HIV businesses as well as the level of contingent consideration we expect to pay to Shionogi in relation to the HIV business, and to Novartis as the next vaccines milestone becomes more likely.

Just as a reminder, this relates to non-U.

S. sales of Bexsero, which are clearly growing strongly.

Total results this quarter also include charges of approximately GBP 450 million relating to our decision to withdraw support for Tanzeum over the next 12 months or so.

The rest of my comments will be on our adjusted results.

The results we reported today reflect another quarter of strong operational delivery as well as continued investment behind key future growth drivers in each of our businesses, particularly new product support and R&D investment in Pharmaceuticals.

Over the last several quarters, we've stepped up Pharma R&D spending as we advanced our pipeline.

As Emma has highlighted, HIV is one of our core therapy areas, and during the second quarter, we took the decision to invest for the first time in a Priority Review Voucher to accelerate the FDA's review of a key asset, our first 2-drug regimen in HIV.

The GBP 106 million cost of the PRV was charged to R&D expenses in Q2, and, with an impact of 5% on adjusted earnings per share in the quarter, was the key driver of the reported decline in adjusted earnings per share of 2%.

Turning to sales growth.

In Q2, we continued to drive good growth in Pharma and Vaccines, more than offsetting the declines from older products including Advair and despite a 2% drag from divestments to reported Pharma growth in the quarter.

Consumer had a more disappointing quarter with sales flat overall after an estimated 2% impact from divestments and destocking in India ahead of the introduction of GST.

The flat result was despite some strong performances from our power brands.

However, this was not enough to offset a broader slowdown in consumption in key categories, particularly in international but also the U.S. Our U.S. Consumer performance was also impacted by tougher competition in our allergy business, which particularly affected Flonase.

While we are taking steps to address these changed market conditions and improve our competitiveness over the balance of the year, we also expect to face additional pressure in the second half from the introduction of generic competition to one of our legacy Novartis products.

The impact of this on second half sales is expected to be up to about GBP 40 million, and a full year impact in 2018 would be around up -- GBP 80 million.

Given these various factors, we're not now expecting much growth at the top line from the Consumer business this year.

Also, unless the market backdrop improves, we would not expect more than low-single-digit growth in sales next year, especially when you factor in the drag from divestments, GST and the ex Novartis generic.

The slower growth we're seeing this year and the impact on 2018 are key drivers of the updated outlook for the Consumer business out to 2020, where we now expect a top line percentage CAGR over the 5 years of low to mid-single digits.

We're controlling costs tightly while ensuring we continue to invest in our power brands.

We still expect to make significant progress this year towards our operating margin target for Consumer of 20%-plus by 2020.

On to operating margin more broadly for the group.

As I mentioned before, the PRV investment is a big one-off impact in the quarter, taking 1.6 percentage points off our adjusted operating margin for the group in Q2 at constant exchange rates.

Across the rest of the cost base, we're driving operating leverage in the P&L with a 90 basis point CER improvement, excluding the PRV, by driving sales growth in Pharma and Vaccines in particular, as well as the benefit in mix and continued tight cost management in all 3 businesses, even while we invest behind our new product launches and priority assets in R&D.

In the quarter, royalties were up, and we continue to expect around GBP 300 million for the full year.

From a free cash flow perspective, we've driven an improvement of over GBP 300 million compared to the first half last year, even after investing in the PRV.

This has been through a stronger operating performance, tight management of CapEx, lower payments for restructuring and continued currency gains, offset by some catch-up in dividend payments to minorities.

As we saw last year, the first half has seen a significant build in inventory as we prepare for seasonal sales, but also this year, an additional impact as we get ready for our upcoming launches, particularly Shingrix and the closed triple.

We're continuing to improve the efficiency of inventory utilization with working capital balances 10 days lower than they were this time last year.

As the inventory build unwinds, we expect free cash flow to be significantly higher in the second half of last year -- sorry, second half of this year, as we saw last year.

This was very much the pattern and reflects the changing mix of the business post the Novartis transaction.

So where does that lead us for the year?

We did not want to fund the PRV by cutting back on other key priorities and drivers of future growth.

And as a result, the PRV and other accelerated launch spend for the 2-drug regimen will impact our previous expectations for growth in adjusted earnings per share this year by around 2%, and we're updating our 2017 guidance range accordingly.

With no Advair generic expected this year, we now anticipate adjusted earnings per share to grow by 3% to 5% in 2017 on a constant currency basis.

If exchange rates remain at current levels, then we would expect a full year FX tailwind of around 8% to adjusted EPS.

Looking further ahead, the commercial environment remains highly competitive, especially in our Respiratory business, where we're also seeing continued pricing pressures.

We're in the middle of the contracting round for 2018, and every indication suggests that as we move into next year, there's going to be no letup in the pricing pressures we've been seeing for Advair but also the new Ellipta respiratory product as payers anticipate an Advair generic sometime in 2018.

Against this market backdrop, it is critical we continue to invest to grow the market share of our recent launches as well as prepare for those that we expect to launch shortly if regulatory approval is given on the time lines expected.

It's also important that we continue to invest in the newly prioritized pipeline.

We're going to use our financial architecture to ensure that the delivery of our strategic priorities of innovation, performance and trust translate into clear financial goals that we can embed across the company.

The goals of our financial architecture are to drive stronger growth in sales through improved innovation across all 3 businesses; to drive earnings per share faster than sales through better performance; driving operating leverage through the business from tight cost control and continued financial efficiencies; and to convert more of those earnings into cash, which can either be reinvested in the business or returned to shareholders, all of this achieved the right way, consistent with our values and our objective of building trust in GSK.

We'll use the architecture and its common goals to help create step change in the alignment of our operations around 3 fully integrated businesses.

This will enable new end-to-end emphasis on cost discipline and cash consciousness.

It will also allow us to be clearer, both strategically and operationally, on how we invest and allocate our capital between our different businesses.

I'd now like to run through each of these goals in a bit more detail.

We've already driven annual cost savings of more than GBP 3 billion from our combined integration and restructuring program.

This has helped to offset major headwinds over the past few years from pricing and the decline of profitable, older products such as Seretide/Advair, Avodart, Lovaza.

It's also allowed us to restructure our cost base to create significantly greater flexibility to reallocate our resources to where we see the greatest returns before we have to add additional funding.

The continued stability in SG&A expenses and the leverage that this produces despite significant promotional investment going in behind our new products is evidence of the effectiveness of these efforts, but there is clearly more we can do.

We anticipated at the time of the Novartis transaction that we would invest approximately 20% of the savings, and our tracking since that time suggests we've invested slightly more at around 1/3 of the benefits.

This has gone mainly to promotional support and supply chain improvements and then, more recently, to R&D.

2/3 has been used to offset the margin pressures we've been experiencing from pricing and the decline of the older products.

We've now identified another GBP 1 billion of annual savings from the same program, primarily in the Pharma business, through supply chain efficiencies, simplifying our operations, improved procurement savings and a more streamlined functional model aligned to the new business priorities that Emma has outlined today.

We again intend to reinvest approximately 1/3 of these savings into supporting our already launched new products, key near-term launches and the R&D portfolio.

Similarly, the rest will be applied to protecting our margin by offsetting some of the pricing and competitive pressures we continue to face in the Pharma business.

While those pressures will remain a meaningful drag on our operating margin over the next several years, particularly in the year in which an Advair generic in the U.S. does arrive, we expect to begin to see a better balance going forward as the operating margin benefits from stronger sales growth from new products and the drag from Advair reduces.

Remember, though, that we've always said we would manage the margins to deliver more sustainable earnings per share growth for the longer term, and so margins may move around quarter-to-quarter as we invest to drive that growth.

Important to delivering a sustainably better margin is that the tailwinds also go beyond restructuring benefits, even though those have been a significant contribution.

Across our 3 businesses, as well as at the center, we plan to ratchet up cost discipline with a particular focus on improved supply chain efficiency to drive a more competitive cost of goods, tighter control of G&A, including capping growth in functional and other support capacity, and strengthening our procurement capabilities.

In our supply chains, we're reducing complexity by divesting some of the tail, cutting the number of manufacturing sites and lowering the number of SKUs.

And beyond our internal manufacturing network, we're also focused on simplifying the number of contract manufacturers we use, which will improve utilization and contracting terms as we leverage better scale with the contractors that remain.

Generating additional savings from procurement will be a significant focus for us.

And we've announced today that we'll consolidate all of our procurement activities into one unified organization to leverage scale and best practices.

And we're targeting this organization to deliver mid-single-digit percentage savings in material and indirect costs as well as non-production spend to improve our cost of goods and allow us greater investment flexibility across the P&L.

In establishing our new business priorities, we've also identified opportunities to reduce nonworking SG&A further, simplifying and hobbing the back office elements and support functions, including medical and regulatory, and to reinvest this behind customer or patient-facing SG&A.

We expect this will enable us to cap the growth of our nonworking spends behind sales growth, improving operating leverage and move us into a more competitive position relative to our peers and other relevant benchmarks.

To underpin this additional cost discipline, we will govern the performance of each of our 3 businesses through a fully integrated P&L.

Whilst we've had this in place for Consumer Healthcare for the last couple of years, this is very new for Pharma and Vaccines.

And it's driven a step change in decision making for Consumer and should do the same for Pharma and Vaccines mainly by aligning supply chain, commercial, R&D and functional teams to a single set of objectives and allowing clearer trade-offs, improving performance and profitability as a result.

We also need to improve our cash generation, and to do so, we will build cash metrics more directly into employee incentives.

We'll also implement end-to-end cash flow accountabilities within each of the businesses to allow us to drive cash consciousness much deeper into the organization.

We will ensure the 3 businesses have specific cash targets and are accountable for managing their cash flow directly.

We will focus particularly on driving further working capital efficiency, managing our CapEx in a more integrated way, more closely aligned to the priorities of the 3 businesses, and reducing restructuring spend.

As an example, we plan to reduce our holdings of inbound raw materials by more than 30% by 2020 by shifting to a much more vendor-managed or consignment stock model.

On CapEx, we're now prioritizing pipeline investments and capacity expansions for new products much more directly and turning off other investments around older products.

We've also recently established new ways of working on our procurement of capital to ensure not only competitive purchase costs but also more standard equipment and infrastructure.

For example, we previously announced increased investments behind Ellipta and have leveraged our strong and strategic relationships with preferred suppliers for the manufacturing equipment across all of our Ellipta-producing sites.

This drives better capital efficiency, ease of technical transfers and an improved capability with this equipment internally.

As the equipment is now identical across all of our Respiratory sites, we know how it performs better, resulting in less deviations and wastes and reductions in unit costs for each Ellipta device.

We have a specific program for Ellipta that is targeting a 20% reduction in unit cost for an Ellipta device by 2020.

To strengthen our allocation of capital across the group and ensure that it is allocated to where the best returns are available, we're implementing a clearer set of priorities for our capital and creating a new board to govern the allocation of capital between our businesses.

To support this new approach, we have, for the first time, allocated all of our invested capital between the 3 businesses so that we can track the overall returns each of them make and be able to allocate between the 3 much more deliberately.

The priorities for the new board are: firstly, to invest in the business and drive growth by strengthening the Pharma pipeline and backing winners as data reads out; taking opportunities to strengthen the company such that consumer put, should it come; and then thirdly, expanding capacity in Vaccines, particularly in support of our new meningitis and Shingrix products.

Secondly, to improve shareholder returns, which I'll return to later.

And thirdly, pursue targeted business development focused on bolt-ons and partnering.

Managing these capital allocation priorities will all be done with the parallel objective of continuing to strengthen our credit profile and protect our target short-term ratings of A1/P1.

We will expand the use of cash flow-based return metrics beyond individual project assessments, which is how we've used them previously.

And we've now introduced a consistent cash return on invested capital methodology to prioritize investment across the group as a whole so that we can compare the returns from each of the 3 integrated businesses as we allocate capital between them.

We will regularly benchmark ourselves to our peers.

R&D spend will be subject to the same allocation process to ensure we are looking at R&D returns in the context of the 3 integrated businesses.

As a result, we will no longer publish a separate measure of R&D returns.

So turning to the 2020 outlook.

To start with, at a group level, we're maintaining the overall outlook for sales and earnings that we gave you back in May 2015.

This reflects the overall balance of positive and negative factors that we've seen since then, including high new product sales, earlier delivery of the original integration and restructuring benefits and the extension of that program, offset by greater pricing pressures in the Pharma business, particularly in Respiratory, lower Consumer sales growth, as well as upward pressure on the tax rate.

The benefits to the upside are also partly offset by the minority interest in the additional growth we have seen in HIV.

When reviewing this outlook, it is important to understand we've kept the same goalposts.

The 2020 outlook we're giving you today uses exactly the same methodology we used in May of 2015, and that includes assuming that the sales and earnings compound growth rates are at constant exchange rates over the 5-year period and the exchange rates we've used to estimate the 2020 positioning, including the 2020 margins, are, therefore, 2015 rates.

And if sterling stays at current levels, we'd estimate a positive tailwind to the 5-year EPS CAGR outlook we've given you of around 5%.

We're also continuing to assume Advair will encounter generic competition between now and 2020, and we'll only have GBP 200 million to GBP 300 million of residual sales in the U.S. by then.

Hopefully, this gives you a clearer idea of what has and what hasn't changed.

But you should also note that this unchanged outlook is despite an impact on group sales of around GBP 900 million from divestments and exits over the period, including the withdrawal of Tanzeum.

Looking at each of the businesses briefly.

Firstly, in Pharma, we've made good progress with new products, more than offsetting the decline in Seretide/Advair so far.

We continue to expect this shift to continue over the 5 years with a low-single-digit CAGR for sales despite higher headwinds from divestments and exits over the period than we originally expected.

We also originally expected 2020 margins to be flat at just under 30% at constant exchange rates, with leverage from sales growth being offset by other factors.

We now expect we can do better on margins to deliver low 30s at 2015 exchange rates, thanks to the additional cost efficiencies that I've already talked about as well as additional leverage from new products performing ahead of expectations.

Next, Vaccines continues to be on track with our original expectations both for sales growth and margin.

You've seen the strong results Vaccines has driven over the first part of this 5-year period.

And looking forward, we have high expectations for Shingrix in particular as we move through 2018.

This key launch will clearly require significant investments in both supply chain and the necessary sales support, which may impact margin in the short term, but we still expect our 2020 margins will be 30%-plus.

So an improvement of at least 5% compared to where Vaccines' pro forma margin was in 2015.

The Consumer business continues to be an attractive business with clear synergies within the group.

However, given the expected impact from the market slowdown I've already discussed, as well as the impact of shorter-term competitive challenges, divestments and GST, we now expect the 5-year CAGR in sales to be low to mid-single digits.

Importantly, our margin, we are still targeting to get to the operating margin of 20%-plus that we previously indicated by 2020.

In the bottom half of the P&L, we'll continue to drive financial efficiencies.

On tax, we now expect moderate upward pressure over the next few years, and we've already seen some of that this year given the group's changing mix, particularly geographical mix, and a more challenging tax environment.

Overall, we continue to expect adjusted EPS to grow mid- to high single digits over the 5 years to 2020 in constant currency terms.

We understand how important regular dividends are for many of our shareholders.

As a result, returns to shareholders are a key priority for capital allocation but after first meeting the investment requirements of the business necessary to support long-term future growth.

We continue to expect to return an 80p dividend for 2017, and we've also announced today an expectation that subject to any material change in the external environment or our performance expectations, we'll pay a dividend of 80p per share in 2018 as part of a new dividend policy for 2018 and beyond.

This new policy sets out our objective to distribute regular dividend payments that will be determined primarily with reference to free cash flow generated after funding the investment necessary to support growth.

Over time, the group intends to build free cash flow of the dividend to a target range of 1.25 to 1.5x before returning the dividend to growth.

Starting in 2019, we will return to our previous approach of setting dividends on a quarterly basis rather than continuing to provide a medium-term outlook.

And so back to our overall vision out to 2020 and beyond.

We will focus our efforts around new priorities to strengthen innovation, improve performance and build trust for all 3 businesses, and Emma has set out some of the specific actions we're taking to do this.

Based on these plans, we continue to expect to deliver our 2020 financial targets at a group level, reflecting the balance in upsides and downsides that we've covered.

Importantly, we are also laying a clear platform for growth beyond 2020, including continued cost discipline, a deeper cash consciousness across the company and a clearer capital allocation framework to enable investment behind the continued success of our new products and the development of a stronger pipeline.

With that, thank you for your attention.

We're now going to take a short break for about 15 minutes, after which we will return to take your questions.

And there are refreshments outside.

Thank you very much.

(Break)

Welcome back.

We're now moving into the Q&A session for up to 1.5 hours or longer, if you've got more questions.

As you can see, I am joined by my team here on the stage.

And thought just to remind you, in case you don't recognize them from their photos, from right to left, we've got Simon, RSV; Patrick Vallance, Head of R&D; David, who is Chair of our HIV business and the Chief Strategy Officer; Deb and John, who lead HIV business; Luc and then Brian: Luc runs our Global Vaccines business; and Brian, Global Consumer; then we have Jack, who is President of our U.S. both Vaccines and pharma business; and last but not least, Eric Dube, who runs our Global Respiratory business.

So I will be chairing the Q&A session, but obviously sharing out the answers to your questions with them.

And then in the front row, on both sides, we have many of our R&D leaders as well.

So just in terms of the logistics of this session, you're probably all extremely familiar with them but for those of you that are in the room, could you please raise your hand and then switch on the red button in front of you when I signal to you to ask your question, and then please, please switch it off, so that we can answer and then move on to the next person.

And I'll also be taking questions that are going to come in over online and on the telephone line and via the webcast.

Last request, please do, as usual, try to restrict your questions to 2 to 3 at a time, so we can get around as many people as possible.

Okay.

So who would like to kick-start?

That's an encouraging sign.

Let me start in the front row.

Cheers, Andrew, go ahead.

Thank you.

Andrew Baum, Citi.

Two areas.

First of all, you highlighted oncology as a potential platform pending the readout of the datas.

Does that preclude any significant transactions within the oncology space ahead of that time?

And then in the same vein, could you give some further color on the divestments of the established products business in terms of the consideration you may expect to get -- financial consideration from the sale of those revenues, just thinking in terms of financing a bolt-on transaction and strengthening what you could do with your balance sheet?

And then a question for John, and apologies for the predictability, given the IS data on the resistance mutations that were shown, particularly the integrase resistance mutation.

Particularly from a competitive perspective with Gilead being able to leverage that data and to shy people away from adopting it, how do you deal with that, especially in the U.S, where your attraction with KOLs, the 2 drug regimen is somewhat less than as in other territories in the world?

So thank you very much, Andrew.

We'll come back to John and maybe Deborah also, talking in terms of the commercial competitiveness question in terms of our HIV business.

Simon, do you want to comment on the divestment consideration?

There's a pretty wide range of businesses within that mix, but typically, for that profile, you'd expect 1 to 2x sales, something like that.

And in terms of your question on oncology.

I mean, you're right, we have very clearly highlighted that as a potential area dependent on data.

We do want to see whether we have an anchor asset for us, ourselves or whether or not we should go move out with any other partners, and there are several (inaudible) that we're already in partnership on certain cynical that we could consider, where indeed, as you're aware Novartis has the right view on asset -- from an asset by asset basis.

So I don't think we would be looking at any material transactions ourselves proactively until we see more data there.

So can we just get to the answer on the expected and important HIV question?

John, would you like to respond to that first?

Sure, I'll give you a little prospective, so you're asking about the ACTG 53 study that was presented at IAS on -- yesterday.

And this is a pilot study that actually followed the first pilot study that went forward for the 2 drug regimen of dolutegravir plus 3TC treating treatment naive patients.

That was the paddle study, and actually at the meeting we saw the 96-week data.

And of those 20 initial patients really the durability has really stood up over the 96-week period.

Following the first report of that, the ACTG went forward with a larger study of 120 patients being treated with dolutegravir 3TC and they were reported on 24-week data at this meeting.

And actually, the meeting or the data that was presented was pretty spectacular because you had 90% of patients at 24 weeks with the 2 drug regimen being fully suppressed.

Now as you note, there were 3 patients that did have urologic failures.

One of whom who the investigator described as chaotically nonadherent, to paraphrase him, did develop resistance or emergence of resistance mutations to both the 3TC and also to the integrase drug with a kind of what we would describe as a minor resistance mutation.

Now I think it's notable for all the previous treatment naive studies with dolutegravir, we have not had a patient develop that.

We have had reports of patients who are treatment naive develop integrase resistance mutations, but that's been a very rare event, and as I said, not in association with clinical trials.

This is a pilot study, and I think the real telling of the tail here will be looking at the results in the Gemini study, which would be over 1,200 patients being studied with this.

It's obviously always disappointing to us to see a patient develop that, but clearly, someone who's not adherent it's not that unexpected and that's something we take in and really look to see the larger body of evidence as it develops going forward.

I do think you really have to come back to the overall performance of the regimen compared to that really before you really can assess at how it stacks up against other regimens along the line, because development of resistance actually occurs with all regimens.

And so I think that we'll just have to see the data as it plays out.

But I think that overall, the data there is very encouraging and pretty exciting to us.

Deb, do you want to comment on the competitiveness?

(inaudible) you know that the failure that John has just talked about will be kind of graft by our competitors, but I also believe that Gemini is a very, very important study.

And that is where you will really see the strength of dolutegravir 3TC.

From my perspective, we've got a pipeline of 2 drug regimen products with rilpivirine, dolutegravir coming out first, followed on by dolutegravir 3TC, and then we move into the long-acting era with cabotegravir.

So I think we've got a very strong proposition.

Gemini will be key.

We've obviously already got SWORD 1 and 2, which we shared at CFROI.

For me, we have a very competitive offering in the U.S. and globally.

And I think we're very much prepared to match our competitor, share our voice wise with our sales force, in the marketing space with our medical sales forces, which are the same size in a few places even larger than our competitors.

So I think we're set up to be very, very competitive both in the U.S. and beyond.

Feedback I heard yesterday both from KAEs in Europe and the U.S. is that they're very excited about the 2 drug regimen portfolio we've got that, but they're very much waiting for Gemini.

And I think Gemini now becomes a much more important milestone from a data perspective for us to judge just how successful we will be with dolutegravir 3TC.

Okay.

We'll go across the front row.

Vincent?

Can you speak up a bit?

Yes.

Vincent Meunier from Morgan Stanley.

So the first question is a follow up on oncology.

I mean, how do you think you can become a top oncology company in the context of you restarting investing in that area after the divestments quite recently to Novartis and quite a very competitive landscape with many big companies investing in new technologies for several years?

So how -- what is your value proposition here?

The second question is on the dividend.

Can you explain us how already that is work?

I mean, should we consider that the 80p for '18 is the base, maybe a floor?

Should we expect at some point maybe a dividend cut at the end of the decade, if you did not cover the dividend or if you constantly make a big acquisition?

And then the free cash flow is impacted?

Okay.

Thank you very much for the questions.

I'll respond on the dividend, we also have a question online from Mark about getting on the outlook dividend after 2018, so I should be combining those 2 together.

And I'll give a quick word on oncology and perhaps ask Patrick as well to pick up on that.

And so on the dividend, we know it matters.

We're expecting 80p in 2017, and we expect it in '18 because we know that was a key question.

After today, we wanted to give visibility on it.

We also wanted to give visibility by announcing a policy that will be based -- distribution will be based off free cash flow with a target cover of 1.25 to 1.5, so we give clarity that we're working towards rebuilding that cover before increasing the dividend.

That's really a very important message that we want to have understood, if you want to keep investing in the business for its future growth.

And our intention is to be rebuilding that cover off an 18' base.

That said, I'm not going to stand here and say the dividend will never be cut if some circumstances happen to say that, that is required and appropriate.

But that will be a board decision at the time, and our intent is absolutely to be rebuilding the cover, as I said, from that basis, and we are not going to be pronouncing on the dividend in the medium term.

Having said this, we will then move back to '19 into quarterly declarations.

So I hope that is crystal clear to everybody.

And then on the oncology question, I just -- we -- in the deal with Novartis, which we all feel was a fantastic way to get value for the commercialized oncology assets that we had and build up to really world-leading scale business in Vaccines and Consumer is important to say and it was perhaps not sufficiently understood at the time, but we absolutely retained our R&D capability in oncology with some great R&D talent and some exciting early-stage assets that we think do have potential to bring real value for GSK.

As I've said already, how we get that value is still to be confirmed depending on what the data says.

And Patrick, I don't know whether you'd like to comment a bit more on specifically on the aspect.

Obviously, Luc Miels will also have a meaningful role to play here in terms of thoughtfulness about what is possible, what is right in terms of our right to win and some of the commercialization in the specialty area, which is a bit different in terms of building commercial capability.

Patrick, if you want to add to that?

(inaudible) we kept our discovery at first in oncology, and we kept it in a very focused place, which is where we thought we had deep expertise.

One, in epigenetics where we were early in the field and remain very deep in the science we have and the second in immuno-oncology based on a very strong immunology presence at GSK and the immuno-oncology expertise of Axel Hoos, as the leader of that area.

So those are the 2 areas where we're not trying to be an all encompassing oncology play, we're pursuing largely those 2 areas.

And Axel -- I don't even want to comment on any of the recent things, because our pipeline is now beginning to declare itself in terms of where we are in our clinical readouts.

And indeed, in terms of the combinations that we have, which are either unique or at the forefront of the next wave of some of these areas.

Do you want to comment, Axel, on anything specific?

Thank you.

No.

But what I can say, that our pipeline is entirely based on innovation.

We have delivered a lot of value in a relatively short period of time.

If you think about the way this deal was structured with Novartis, there was only discovery assets left in the oncology pipeline post-Novartis.

But now, 2.5 years later, we have 11 assets in the clinic, we have several of those that carry a lot of promise.

And the lead asset is actually just revealing itself to have a level of efficacy that you could call potentially blockbuster efficacy.

So with a 60% -- approximately 60% response rate in refractory multiple myeloma with our BCMA antibody drug conjugate, we already doubled what daratumumab had shown at the same stage of development, and daratumumab has become a blockbuster asset.

And if you compare it with other combination of work, pomalidomide, dexamethasone and daratumumab together achieve about the same level of response rate that we have as a monotherapy.

So if this continues, which is what we're expecting, then, once we enter combinations, I think we will have a strong stance for multiple myeloma patients.

So having said that, that's the lead asset in the portfolio where we clearly have efficacy data now that is meaningful and there are other assets pushing to produce additional value.

Where, as Emma says, we still have to wait for more readouts.

Thank you.

I'm going to go to the phone.

Kim from Bernstein, can you ask your question please?

No, maybe not.

Okay, there's one question online that, Simon, I'm going to ask you to take up question on whether the slowdown in consumer make us think that probability with Novartis or perks -- being a perk us next year, whether that probability has increased?

Probably a question for Joe.

But if so, are we comfortable we can fund this as well as retaining flexibility for other M&A, whether that's in Consumer?

Thanks, Emma.

I mean, clearly, the first part of the question is a question for Novartis.

And they've made that position pretty clear in their own recent earnings call, but they will have to decide when they want to exercise their put.

It is their put, not our call.

We've been very clear we would like to acquire the rest of the business as in when they decide to exercise, and we're very comfortable we can fund it exactly how we choose to fund it.

I think it's very premature to get into that discussion as clearly, that put could be some way away.

Questions in the room.

Keyur?

It's Keyur Parekh from Goldman.

You spoke 2 questions.

The first when you spoke a lot about what you are changing.

My question is on what you're not changing, and that is your financial targets for 2020.

A lot has gone in your favor since you first issued those targets.

You announced a GBP 1 billion incremental cost savings program today and yet what we are seeing is unchanged EPS expectations.

So just help us think, are you running faster just to stand still or is there inherent (inaudible) to what you're laying down today?

That's question number one.

And then question number two, apologies, the dividend thing is still not crystal clear to a lot of us, so I'm coming -- I'm going to come back to that.

The press release says, over -- building dividend cover over time.

I would like to understand what over time actually means in kind of your mind?

And inherently, that's driven on the fact that if one looks that it'll be a you define free cash flow for 2016, the dividend cover was 0.8x free cash flow.

To take it to 1.25x (inaudible) number, it is unlikely to get there by 2020.

So is that an appropriate context to think about over time?

Or should we be thinking about a longer-term cycle than that?

Thanks very much, Keyur.

And again, to reiterate, part of an assessment of where we are came from listening to a lot of very transparent shareholders and lead analysts, so I should thank you for that and request it on an ongoing basis.

Perhaps easier when a CEO is new, I'm counting on it for the long term.

So I'm going to ask Simon to come back on the definition of over time.

But just to give me my headlines on the 2020 guidance, I think it's very important and you did have it, in his slide on the puts and takes that make up landing on the same outlook.

Just a reminder, since 2015, we will be taking out GBP 900 million of turnover and divestments so that is was what the base was in 2015, of which GBP 500 million is still to come, I think GBP 400 million of which is within pharma in terms of the to go.

We are talking about investing more in R&D, but the big focus agenda in R&D where we look at Q2 or the assets, and we've taken into account, the assets that we think we want to bet on that will be continue to be invested in.

Slight pressure on the tax rate and some slow downs or pressures, whether that's an adjustment in consumer, both for environmental and one-off or also the pricing pressure, which is near term, still very wheels [sticking and held] respiratory.

So we are making some important and quite aggressive changes to make sure that we can still provide a reasonably competitive outlook for 2022, when we look at it through our 3 individual business units.

With a marked forward on margin expectations within pharma, and most importantly, investing in where we create the strongest value for the long term, which is delivering a pipeline that is valued, not only by patients, but by the market.

So Simon, do you want to comment on the dividend over time question?

Yes, I think just to pick up the point that we had earlier also about the 2019 position, what we're saying in terms of how we're trying to indicate or guide people on the dividend going forward is that from 2019, we will go back to what is normal, of declaring and telling people about our dividends each quarter as we going forward, and that's what we used to do.

We created this bridge across the Novartis transaction.

And now today, we've announced a new policy to set the framework for declaring dividends going forward against the baseline that, as Emma described, we're establishing for 2018.

And the intent is then that we will grow into cover over time, and I think that's probably several years rather than 1 year.

I'm not going to get into specifically exactly how many years, but it's important against the other priorities we've got.

As we said, our first and foremost priority is to make sure we're investing behind the future growth of the business, so that we secure that cycle of growing into the cover going forward.

So I think that against that expectation, you should not expect the dividend to grow in the short term.

Clearly, none of us can say that we'll never ever consider a reduction in the dividend if some reason dictates, but we should do that.

But what we're not saying is that, that is sort of in some vacuum in 2019.

There's very clear framework to guide how we go forward here.

And hopefully, that's clear.

But if not, then we should go through it again just to make sure everyone is clear.

Okay?

Okay.

A question for Patrick that's come up online.

With regard to our target to reduce drug development times relative to peers -- it's looking to reducing to be more competitive.

So can you talk about balancing the inductions with our third long-term priority of trust?

Patrick?

Yes, I'm not quite sure what this question is getting at, whether the implication is that we're going to cut corners to do it or whether it's about our class priority to actually deliver medicines.

I mean, clearly, we're not going to cut corners, and I will talk about what we are going to try to do to reduce our development time.

So we know that we do very well in respiratory and HIV, and we know a lot of the reasons we have longer time lines in other areas, are to do with things like how we got our R&D commercial interface, which is being addressed head-on and decision-making to try and reduce the wide space between trials.

So point number one is that we're going to address the R&D commercial interface, particularly when we're thinking about areas where we haven't traditionally had that strong interface that we need to get right.

We've got processes coming into trials, which are reducing trial time, so Emma has alluded to one, which is in the danirixin study, we have real time data capture, which allowed us to make a decision about a year earlier than we otherwise would have done on what we thought the result was and therefore trigger the next phase of study.

So expect to see more in the way real time data capture, real time information from multiple outputs from patients and the use of digital also to do things like site selection and the ability to time delivery of drug substance, where in one study we have would reduced trial times by 6 months.

So I think when we look forward, it's a combination of decision-making, R&D account -- R&D partnership with commercial in order to make sure we don't circle, particularly in the Phase II and some strong digital approaches to try and improve performance of trial delivery and times to deliveries, it's absolutely not about trying to reduce the standard of the data we generate in the evidence we're putting together.

Quite the opposite, in fact, we want to increase that.

Okay, back to the room, please.

It's Matthew Weston from Credit Suisse.

Two quick U.S. commercial questions, if I can and then one bigger picture on R&D.

A couple of the points, Emma, that you made around U.S. commercial were I think an increase in investment and support and also you've talked about changing incentives.

So your predecessor clearly went out on a limb with a new commercial structure in the U.S. and was confident that the industry would follow, it didn't seem that they did.

So could you let us know whether or not we're now seeing a change back at GSK to a more traditional incentive-led sales force model in the U.S. or whether or not you're just tweaking the existing one?

And then also I was surprised to hear about the Ellipta pressure across the board in 2018.

One of the arguments previously on Ellipta was that doctors loved it.

It was better for payers because you could switch patients much more efficiently and you could save cost.

So I wonder, Jack, whether there's any real change in terms of -- is it really just Advair and is that Ellipta story still holding up?

Or something changing?

And then finally, just on R&D.

Emma, again, you referred to a lot of focus, a lot of focus on improving output and you talked about leadership changes in the top 200 of the business.

But are you sure that you've got the right people on the ground in R&D to actually deliver all these changes?

Or are we actually going to see significant turnover of employees to really get the maximum efficiency and output out of the GSK R&D organization?

So I'll make a couple of comments and then I'm going to ask Jack to comment on the competitiveness, both of the Ellipta portfolio and also our commercial policy.

So quickly, in terms of R&D.

As I said, people drive performance.

We all know that.

We've made some changes.

I think more will come.

We wouldn't expect it to be announced ahead of time, so -- but we absolutely know that we have got some fantastic scientists in the company and we undoubtedly will see some renewal as across many of the areas of leadership.

In terms of the incentives, the incentives I was talking about actually was much more broadly across the whole company as opposed to specifically sales rep incentives.

And we do plan to announce internally for 2018, some fairly significant adjustments to help align behind both the strategic priorities and the performance objectives there to really make sure we're all pulling much more strongly to be competitive versus the marketplace, and there's definitely a big opportunity on that.

What you allude to in terms of our sales force policies.

As I've said, GSK has been quite proud of its patient-first principles to try and remove any perceived conflict of interest.

At the same time, we've already evolved those policies, not least under Jack's leadership to make them more competitive and frankly, simpler and easy to execute against, particularly in line behind critical launches, and we will continue to do that.

But the fundamental principle of us being a values-based, trust-based company, where people can trust our science and our intentions and having very productive relationships with actually ATPs are absolutely critical to who GSK is and will continue to be.

Jack, do you want to comment, please?

Okay.

Thank you, Matthew.

Just to underscore on the patient-first, we continue to be very values-oriented, but we're going to balance it with competitive performance, and so we monitor and we continue to adjust as we have the last 2.5 years on that.

Speaking specifically to the Ellipta portfolio, it has performed well.

I think when you look at the overall portfolio over the past year nearly doubling, products like Breo more than doubling.

Breo is now the #1 prescribed ICS LABA with pulmonologist.

We had our best semester ever in Q1 and Q2 of this year with Breo and our second best with our AC containing products in New England.

So performance is very good.

The reality is both by -- driven by market pressures, potential regulatory changes and potential legislative changes, both at the federal and state level.

I think that's what the industry is exposed to.

We know it is exercised on a class by class basis, and those classes that tend to be more retail oriented and tend to have multiple products in it, tend to be under the most pressure.

So I think that's what you've heard from Emma and Simon is recognition that isn't going away.

But in the meantime, we will drive -- continue to drive strong share of market performance across the entire Ellipta portfolio.

Thanks, Jack.

All right, back to the room, please.

It's Graham Parry from Bank of America Merrill Lynch.

Just first question on the guidance.

You qualified your reiteration of the midterm guidance with the potential need to invest in R&D, which implies that should the right opportunities come along, that guidance could potentially come down.

So how should we handicap the potential for that need?

And do you think current R&D spendings are sufficient to achieve long-term objectives as you brought in external assets, would that mean you have to lower guidance?

And how much of the GBP 1 billion of savings you think would be allocated to, to offset that?

And secondly, another question on R&D.

GSK has made much noise in the past about R&D structures, processes, we've had [seds], we've had DPUs, we've had R&D investment boards.

And so in your analysis of what haven't worked in R&D, why didn't they work?

So can you just help us to understand what's changed this time, because for those of us who followed the stock for a long time, it feels like we're hearing some of the same messages again that we've heard under previous new CEOs as they started.

And then thirdly, you said you're interested in owning all of consumer health, including the Novartis stake, and that's previously been viewed from the lens of Novartis putting that stake to you.

Their latest communication seems to be tilted more towards, let GSK come to us.

So could -- in the context of that, could you help us understand your desire to pursue Novartis for a transaction rather than the other way around and give us some kind of a reassurance of the value that you would get for GSK shareholders in such a transaction?

Okay.

So just on the put, I think as we have already said it's their put to us and that's going to be their decision.

I think, Patrick, actually, you would be the very best place to say about what we think is going to be meaningfully different, in terms of the operating changes because we are quite deliberately not choosing to do major structural research, which we think will just cause more delays as opposed to improve output, and then I'll ask Simon to comment on your first question.

I think we're actually not changing DPU model.

And we believe the output from those DPUs has been extremely good.

So the discovery organization, I think, continues to innovate and produce really high-quality output, and we've got some of the DPU heads here who can speak to that.

John Bertin, for example, I think is an absolutely recognized world leader in his field.

I think where we haven't done as well, and it was clear from some of the statistics shown, is in some of the areas of development where I think we failed to focus enough.

And as a result of that, we had too many things progressing too slowly in the development organization.

And not only have they developed too slowly, but I don't think we've had the partnership right for the R&D and Commercial, which meant they landed in a prepared partnered organization that could drive them to full value.

So by focusing down, and it's a very significant focus down on the 2 areas plus too emergent.

I think it gives us a vertical integration, all the way from target selection through to commercial ability to deliver on it, in which we can drive things through in a way where we've got a recipient and partner organization.

So I do think the R&D and commercial interface is very much tighter in this new design.

And I think, obviously, the arrival of Luke is going to help them further with that.

I think in development, we also absolutely need to not only make sure that we do things fast, which we've talked about, but we make sure the evidence generation is aligned with what's really required.

And I think by spreading more thinly and going across too many therapy areas, I think we often had quite good molecules, which didn't end up getting the right evidence generation to be commercially successful.

So again, that narrowing the commercialization and development is a very key change to how we're thinking about things.

Thank you.

Simon?

Yes, Graham, I think the answer to your question starts with the Keyur's point earlier that in terms of looking at the outlook to 2020, key paths of the trade-offs we've described is making sure we've got flexibility to invest behind the newly prioritized R&D pipeline.

So there is a reasonable amount of allowance in there for R&D spending.

You've seen R&D spending coming up quite quickly over the last several quarters, clearly we're going to start annualizing some of that.

And so I think the way you should think about it, is that we will make specific investment decisions, if the data justifies it, and there's a clear evidence points to bring in front of you, and we'll adjust accordingly with an obvious opportunity sitting in front of us.

So that's -- and I wouldn't build in other adjustments at this point.

I think it's all within the outlook to 2020 that we've given you.

Okay.

More questions in the room, at the back, please.

Could you put your -- or someone else needs to turn off maybe.

Can't hear you.

So first, could you elaborate on the reason why you kind of divested sirukumab, which seemed reasonably a solid asset?

And now you're more dependent on a few late-stage assets after the choices, so do you include some buying or in-licensing of late-stage assets in addition to the effort you're making internally in R&D?

So I think to be very brief on those.

Our focus on R&D is really going to be on early-stage stuff, but should through -- we are planning to revitalize off the R&D BD team fairly meaningfully and should the scans that we do suggest a late-stage asset is worth of looking at, then we will, but that priority is on early stage strengthening.

And in terms of sirukumab, it's simply a question of with our allocation of resources, what do we think we are the best commercial leaders of, in terms of it being the most competitive thing that we are capable of executing against.

And our judgment was that was not to be the case.

We're still supporting very strong partners (inaudible) terms of the work that lies ahead, but we think that's our decision.

Would you like to add anything, Patrick, on that?

No.

You're right.

And it's just about us putting our resources where we can make them most successfully.

Okay.

Next question please.

Naresh Chouhan from New Street Research.

Sorry to go back to the dividend, but I'm sure as you're well aware, it's very important to the current valuation.

If it's going to take a number of years to grow into the dividend and free cash flow is going to have to grow about 50% to get you within the dividend target range, is it fair to -- and would you agree that you're over distributing and do you think that, that current situation is sustainable for a number of years to come?

And then secondly, on R&D, would it be fair to assume that your return on R&D has fallen since you last updated us on the IRR?

I know you don't update us on the IRR, but just give us a sense directionally on where that's gone.

Obviously, you had Breo and Anoro sales performance being somewhat disappointing and some late-stage failures offset by better [b] performance.

And if so, can you help us understand how you came to the capital allocations since you got to with Vaccines and Pharma and other R&D is, obviously, not your choice as you stated?

Okay.

So Simon, I'd like you to comment on the IRR and capital allocation choices.

Just on the have we distributed ahead, yes, I think I said that in my opening comments.

We did distribute ahead.

But we are really focused, as Simon has said several times, on rebuilding our cash flow and our cover, and I'm not going to reiterate again the principles that we're working towards.

But we are -- we do want to get our balance sheet back into a stronger position, and we are quite focused on the A1 P1 rating at the same time concurrently.

So that will be my comments on that.

Simon, do want to discuss?

I think just to be clear, we're not saying that there's no cover for several years.

We're saying we don't get to comfortably into the range that we've set out, and that is going to take some time.

And then, clearly, we have to also, along the way, make sure we're funding the investments we were just talking about in terms of R&D pipeline, managing our balance sheet, strengthening our credit profiles, we have flexibility if other things come along.

And I think trying to balance those while also maintaining something that shareholders have said to us is very important, I think, it's why you get the picture that you get.

But it's not so there no cover for the next several years.

And I think in that context, when you look at the capital allocation framework, we want to look at the R&D for each of the businesses within an integrated return for each of Vaccines, Consumer and Pharma because they all look quite different.

The relative returns on Pharma, on Vaccines, R&D compared to the numbers we've previously published are not coming down as new products kick into that mix.

Clearly, as we go through the next sort of wave post the upcoming launches we've got have dual generics and the close triple, then you might expect to see it do.

But that, again, is why we want to look at it in the context of the business as a whole.

And Vaccines and Pharma from a return point of view over the medium term look relatively similar.

Clearly, today, Vaccines are still [beginning] with some quite big investments behind relatively new products, which we're putting a lot of capacity behind, and we inherited a significantly loss-making business from Novartis.

So it didn't start -- it didn't start back in 2015 in a very good place, but it's moving pretty quickly through that.

And that's why we think it's right to allocate capital because fundamentally, and Luc will tell you, we can sell pretty much every [day as] we can make.

It's Sam Fazeli from Bloomberg Intelligence.

I have 3 questions, if I may, knowing that you said to.

First one is on Consumer.

In terms of what drives the put value.

We've had a -- you've had a quarter, which wasn't particularly strong and you've got this generic that's come along, where I don't know whether you are aware of it before or not, but it's eating into the growth profile.

How does this lead to still an up valuation for the put option?

Second one is, would your guidance be different if there was a second and the third generic Advair, because I think you've said you're assuming a generic Advair launch in the guidance or maybe at least a generic Advair launch, but just...

That would be a -- that's a correction on that one.

And thirdly, on Tanzeum.

What do you think was done wrongly or whatever -- what happened there in terms of the process of developing it and taking it into market?

What I think most observers were, probably, viewing the product [MABA] as the strongest product in the class, which has -- been something that you've learned from taking forward.

Okay.

So I'll take on those 2 questions.

Just on the put, and Simon can answer, but the -- in more detail.

The revaluation is related to FX on some of the smaller currencies, and also the kind of valuations of consumer companies have moved up a little bit.

And I don't know, Simon, if you'll add anything more in terms of the put?

The valuation is based on forecast of the business, not just for the next 12 months, but kind of medium-term forecast.

And we've seen, obviously, big swing in the main trading currencies last year.

We're actually seeing more benefit than we previously planned for from a lot of the smaller currencies, so we've done a bit of catch-up in this quarter as well as quite a big shift, actually, in the comparable multiples.

So that's why we've seen a significant move this quarter.

And on your question on Tanzeum, I'm going to ask Patrick to take that as well, because it's a really important one and to the earlier point about what's going to be different now.

We actually did quite a detailed review of where we haven't necessarily got this right.

And Tanzeum would be a good one, in terms of having a truly competitive asset with the right kind of full alignment of what winning looks like between the developers and the commercial executors.

And so we just spent some time looking into that.

So do you want to comment a bit more on lessons learned from Tanzeum and so?

I mean, as Tanzeum, A, went through a rather interesting different development path in GSK, which was set as a different vehicle, which I think we won't do again.

So I think there was a specific way of doing it, which was outside a lot's of the normal governance process.

But I think the simple answer to your question, Sam, is that it wasn't called early enough when the data told us that it wasn't going to be commercially successful in our hands.

And I think we should have called it much earlier.

I think that it was evident after some of the early trial readouts.

And I think at that point, we should have called it and stopped it.

Okay, back to the room.

In the back, please.

Kerry Holford, Exane BNP Paribas.

3 questions, please.

First, can we just go back to the incremental price pressure?

So clearly, this has been a reason to trend the guidance for 2017.

But I'm keen to better understand what has got incrementally more difficult.

Since the beginning of the year, you've referenced respiratory a number of times that generic Advair is not coming through next year now, so what's got more tough this year?

And in context of that, could you just talk to us a little bit about rebating in HIV, specifically today, and how that looks out into 2018?

Secondly, a question for Simon on the former margin.

The midterm margins guidance is, so that's now been around low 30s.

And now based on 2015 FX rates, if I understand correctly, so you referenced the growth -- the earnings growth over that period, if you would do just currency-owned, you might also get the same for your expectations for pharma margins on today's currency in 2020.

And then thirdly, a question for Patrick.

On the pipeline refocus, I guess, it makes sense given your struggle to strengthen HIV and respiratory to continue to focus here, but I might argue that those 2 disease categories are relatively well served by medicines in the market today.

So where is the room for disruptive therapies in those 2 indicators?

Okay.

So we'll come back to Simon and Patrick on your second and third questions.

Just on the update to this year's guidance, just a slight correction, the adjustment for this year's guidance is related to the investments in the PLV and the associated costs with launching, therefore pulling forward a bit the launch behind the dual therapy.

We have referred to pricing pressures several times because they're very real.

And we now have a bit more visibility on '18 contracting as well.

But I'll let Jack and then Deborah to comment on, first of all, research room in the HIV environment.

Yes.

In terms of the pricing pressures, as we talked about earlier, it's really multifactorial.

I think the environment is as dynamic as we've ever seen it 25 years in this industry to see the competitive landscape and the market-driven pricing pressures with ongoing pair consolidation, et cetera.

As I said, regulatory, just as recently the last few days, obviously CMS has put out its own proposed rule as it relates to 340B.

And then legislatively, we've got over 30 states that have -- that are trying to enact legislation on drug price, either transparency or procurement laws.

So it really is multifactorial.

It will continue to be intense, given everything from state budgets to the federal deficit to the ongoing market structure in terms of consolidated payers and some of the actions of competitors.

So that's what we're continuing to see going forward, especially in some of the classes, as Emma mentioned, like retail inhaled respiratory.

Deb?

So HIV is a very different marketplace, so it'd be a therapy area where -- what medicine each patient receives is extremely individualized.

In terms of how the market is split for us, so you've got 6% (sic) [60%] of patients in government-funded scheme and then 40% sitting with the commercial insurers.

So for us, contracting pressure is completely different in the commercial insurance.

In the government-funded space, at the moment, we're obviously watching what's happening in the emerging American health care environment.

Potentially, there could be pressure on Medicaid.

But as I'm sure many of you know, if there is pressure on Medicaid and the expansion of Medicaid in some states is reversed, people don't fall out of care.

They go into the Ryan White safety net program.

So what you see is commercial space, which is currently not contracted, and then you will see a government space where there is some uncertainty due to the health care environment within the U.S. There may be some pressure in Medicaid, but due to the safety net system that we have, patients will still receive their medicine.

And I think it's very important that we understand the difference in this particular disease area, not only about the individualized choices that physicians need to make per patient, but also the very active patient group -- patient groups that we see who are very active in lobbying across the world, particularly in the U.S. And that's something the payers have been reluctant to face up to.

In terms of paradigm shift in HIV from a treatment perspective, John, why don't I let you handle that one?

I would just comment on you might say, yes, the market HIV patients are well served today.

But there's actually 2 dynamics that one really worries about.

And so as an infectious disease physician, I'm always worried we're up against a very tough foe that replicates very rapidly, very sloppily and develops resistance.

So we're always worried that there's emergence of resistance, and we don't want to be behind the eight ball, so to speak with that.

So we're always worrying about that.

So the need for developing new drugs to treat the development of resistance if that comes forward.

The second thing, though, is that we've turned this disease going from a death sentence where no one was being able to be treated and survived with it to one where people live many, many decades.

60 years is often talked about now if someone gets infected, say, in their late teens or early 20s.

So it's a different population.

As they live longer, you're now having to deal with all the comorbidities of aging, diabetes, hypertension, other diseases that actually require other medicines.

And so we also have a real need to develop medicines that have no interactions or don't have effects going forward with that.

So I think it's really -- it's a dynamic disease, and I really most worried that people get lulled into, yes, everything is done.

We're done here, let's go look at something else because this is something that will move forward.

And I think that, that's really what drives us as we tried to produce better and better medicine.

So Patrick, would you add anything in terms of paradigm shift?

Well, it's precisely why discovery is broader than development because new areas come from discovery for 15 years time, so we retain that broader discovery.

In terms of respiratory, I think it's true, there are many areas of, say, asthma, COPD, where needs are met.

There are many areas that still aren't.

So subcategories like severe asthmatics not driven by eosinophils remain an unmet need, such as the need for oral treatment to simplify treatment regimens are not met.

And so I think there are categories of both asthma and COPD, where there's still quite significant need, and particularly in smaller patient groups where I think there's an opportunity for areas where there wouldn't be the same pricing pressure.

And in respiratory, we're looking very carefully outside those 2 areas as well.

So we recognize that there's an increasing number of medicines coming through in asthma and COPD.

And pulmonary fibrosis is an area that we're very interested in growing in respiratory field as well as acute lung injury.

Just to add on, on HIV, I mean, long acting is clearly making a difference.

There's clearly the possibility of things like broadly neutralizing antibodies coming along, and ultimately, people are working on something, which may be very, very difficult, but we change, which is obviously whether you can get very long-term remission and cure.

So there are still other areas to go after.

And as Emma said, we're including in HIV also the broadening into other infectious disease areas, and hepatitis B is one I would highlight there.

Okay.

Simon...

Okay.

Yes, on the margin, it's slightly depends on how the mix of the business plays out between now and 2020.

But if you assume similar mix to what we have today and you take the quarter-end rates at the end of Q2, you'll be about 2.5% on top of the margin guidance that I gave, so somewhere between 2% and 3%.

Okay, more questions.

Please.

Richard Parkes from Deutsche Bank.

First, I'd just have to push a little bit more on the capital allocation dividend policy.

If you look at the framework that you gave, I think the Consumer option came -- it was prioritized ahead of dividend.

Obviously, that's an unknown potential cost.

You said that you can fund that through your current balance sheet.

But would it be your intention to fund that through your current balance sheet and maintain the dividend?

That's the first question.

And the second question is just on R&D productivity, maybe for Patrick.

There's been a lot of kind of discussion about streamlining your focus and improving decision-making, but maybe the bigger challenge is in improving your -- the scientific leadership and thought leadership within R&D in pharma within Glaxo.

So is that just improving that -- is that just a function of increasing the investment behind your core areas of focus?

Or is that something that you can also improve through business development and in licensing acquisitions?

So Patrick and Simon?

Go ahead.

So on the Consumer product, it's clearly on the chart as a capital priority.

We don't know when it's going to arrive, but we anticipated it might arrive from the spring of next year.

And so we've anticipated it also in terms of thinking about funding structure going forward and the outlook for the dividend that we described.

So I think while it's a little early to say precisely how we're going to fund it, we wouldn't expect it to have any impact on the dividend profile that we've just already described for you.

Now we have over time an expectation of building balance sheet capacity to fund the different things that we've talked about today.

So when it arrives is obviously key a part of exactly how we choose to implement them.

Okay.

I think, I mean, obviously it's true that scientific leadership is a key part of R&D.

I think we've got some outstanding scientists, and some of them here and you can speak to them afterwards.

We've got some real world leadership positions.

We've got some other areas where we need to bring in new scientists.

And we've already indicated, one such hire, Tony Wood, is coming in, who I think is an outstanding medicinal chemist and leader in his field in the area of product development in terms of CNC and so on.

And we know that we've got very broad connectivity across certain areas in academia that we're going to build on in terms of accessing new science.

So accessing new science, whether it's through [BD], which we already alluded to, we're going to revamp our BD organization in terms of connectivity, whether it's through the sort of deals we've got with venture capital firms, where we're limited partners and seeing access to new things started, the academic links we've got and models we got, which are leading to, I think, significant inflow of new ideas.

Or the leaders in our own DPUs, I think, continually refreshing our science is an absolutely key thing.

And I think we've got strong leadership positions in some places, I've said, and some areas where we do need to look and make sure that we've got cutting-edge science.

And that's always going to be the case, and we will refresh and continue to refresh the leadership there.

Thank you.

So we've got Tim, I think, back on the phone unmuted.

Tim, would you like to ask you questions please?

Yes.

You talked about ongoing price pressure various points.

Some companies give us price volume, foreign exchange information when they report results like Lilly did yesterday.

Can you say or quantify what pricing was for Glaxo across your whole book of business in Q2 or in first half?

Second question is on late-stage pipeline opportunities with closed triple and ZOSTER vaccine.

It would be great if Glaxo could kind of put a stake in the ground and give us a rough indication of how big you think those product opportunities could be.

It seems especially tricky with a closed triple, given the pricing pressures and generic entrants coming in respiratory.

And then last question, Slide 20, Emma, in your deck has one mention new emerging market operations.

I'm not sure what that means, but I think about the level of disclosure by Glaxo in emerging markets has gone down, I think since the start of 2016.

And I'm wondering kind of what's going on in that part of the business, what's going to change going forward and are you going to start to disclose more granularity, so we can track performance?

Okay.

So I'm going to ask Eric in a minute, too, to comment on closed triple sort of competitiveness and probably also Luc on Shingrix.

But just to say that we don't forecast value sales for our assets.

We have said that Shingrix will be a contributor around a 1/3 of our growth, and we do believe that could be our biggest vaccine.

We're very excited about that.

We will hear from both of them on those assets.

Simon, could maybe give you the net price and volume numbers.

So I think we did this...

(inaudible) as a whole, it's about minus one in terms of net price, but we have indicated that in the most recent quarter, Tim, so we expect just to keep giving you some guidance on that.

But obviously, it's an aggregate level.

And coming back on the emerging market point, Tim.

As I said in my opening words, this continues to be an important business for us, but it is -- and it has contributed to growth, although in certain countries, as you well know, we've had some difficult times in recent years.

But we expect it to continue to contribute to growth for the company.

It's around 1/4 of the business.

But we needed to do so more profitably without removing, in any sense, the access to medicines that we know is part of our responsibility and purpose.

We just need to have a much more fit for purpose, particularly from a cost structure point of view, operating that, because 90% of the business is still in branded generics.

It's also notable that we haven't been as good as we should have been at launching some of our innovation.

So we want us to be better at rolling out innovation, but has cost structures and topology and archetypes of markets with appropriate structures around them.

So we are going to be making some meaningful shift there as well as running on a, frankly, an integrated P&L with supply chain.

This is an area where, frankly, our supply chain both in terms of service levels and probably flow in a number of factors has not been where it should be.

So we're going to be doing a lot of work on that.

And as Luc points his leadership, we should see ongoing contribution but more profitable growth from that part of the world.

So maybe I can ask Eric to comment on closed triple, please.

Yes, thank you for the question.

We are very excited about the closed triple opportunity.

If we look at a lot of the emerging evidence within COPD, it addresses one of the major challenges that we have, which is these patients continue to progress and remain symptomatic, continue to have a high rate of exacerbations.

We believe that the future, just as many experts reiterate, that the future treatment of COPD is dual therapy, the LAMA/LABA class as well as of the triple therapy.

And we have seen an incredible profile begin to emerge with our closed triple from our FULFIL study and we eagerly await a landmark study, the IMPACT study later this year to be able to further reinforce that profile.

If we just look at how patients are treated today, about 1/3 of patients are on triple therapy now.

And so we believe that, that is a strong base of business to be able to shift to close triple.

However, when we look more broadly at patients that are on ICS/LABA, which now has been demonstrated inferior to closed triple as well as LAMA/LABAs.

That's a base of business and a big segment of the market that is still symptomatic and can benefit from either an oral with the LAMA/LABAs or the closed triples.

So with the efficacy profile as well as the challenge of complexity that these patients face, many patients that are on open triple today are on 2 different devices, oftentimes 1 once a day, 1 twice a day, it's a real challenge.

And we believe this meets a risk significant need that both physicians and patients have expressed for us.

I don't know, Jack, if you want to talk a bit about the pressures on pricing that you would expect and how we can address.

Actually, I -- could just to build on the closed triple and our excitement in the U.S. affiliate for it.

This is the last piece, if you will, in terms of our inhaled portfolio.

We will be the only company that will be able to run the breadth and gambit in terms of these products all on the same Ellipta platform.

Certainly, when we look back at Breo launch versus this launch, first of all, we'll be first to market with this one versus the fourth ICS to market.

Second, we will experience a much stronger installed base of Ellipta users.

One in 4 patients in this country is now started -- to meet the ICS/LABA, is started on Breo.

And as I said, Breo is now the #1 pulmonologist prescribed ICS/LABA.

One in 3 patients who needed a dual is starting on a GSK Ellipta products.

So you got this very much more installed base of Ellipta users, which makes the jump up to closed triple in the same device on the same device platform much easier and much more attractive.

So certainly, from logistics standpoint, we won't get into the details, but we will be fully resourced to make sure we are highly competitive from a share of voice standpoint.

And the last thing is because the FDA change in guidance, we have engaged payers much earlier than we did with our earlier Ellipta products because of the new guidance.

And so that's enabled us to really get a good beat from a payer perspective, and there's a lot of excitement there just like the physicians.

#1 term we hear from physicians and market research is finally.

Finally, they have a close triple option.

All right, thank you.

And Luc on Shingrix.

Come up.

(inaudible) on Shingrix.

As you said, we will not share any specific forecast, but let's give you a bit of a perspective with the potential here.

We've said that it will deliver 1/3 of our growth from '15 until '20, and we're well on track on delivering this mid- to high single-digits growth with the Vaccines business overall.

If you know that today with the current used products vaccine, in the U.S. only, they make $780 million a year and only 30% of the potential population is covered with that.

And as I said, 80% of that is U.S. only, whereas we will do a first in the U.S. launch, but then a global launch of Shingrix.

And it's really a response to what Emma laid out as the criteria for real innovation.

It's a highly efficacious, sustained efficacy of 90% across all ages and highly differentiated versus what is today.

So it really has the potential to set a new standard.

And if you have seen all the press release around the June ACIP, where we shared our revaccination data and telling you that we're on track, actually, on every single milestone as to what was launched, that should give you the confidence -- that gives us the confidence that this is indeed a potential of big vaccine.

Thank you.

Please.

It's Michael Leuchten from UBS.

Just going back to your cost consciousness slide, you mentioned that at the moment the only consumer has its own P&L.

Does that mean ViiV does not?

And then for the businesses that don't have their own P&L at the moment, what systems are required to make that happen and how long will that take?

Good catch on ViiV.

But, Simon, you answer.

Talk about the system...

You're absolutely right, the ViiV does have its own P&L, [for at least] 2 of our shareholders sitting in there.

So it was more -- it was more a question about thinking -- thinking about the integrated pharma and vaccines businesses, which we have not pulled together in that way before.

We don't need any systems upgrade to do that.

We're using the model that we developed to our consumer now to have that capability in place and we'll have implemented by the end of the year.

So we're ready to go.

Please.

(inaudible) A follow-up question on the (inaudible).

Are you actually moving out of (inaudible) because you're so enthusiastic about the assets you would want as early as possible?

And the reason for the question is really that the CEO of Novartis has stated publicly that the company is in no hurry to put as long as the business is going well.

So presumably, for a small fee, you could actually move back the first time that they can put by quite a bit, and thereby buy a lot of flexibility for that period when you might not have ideal dividend cover, and that would then enable you to actually commit to a progressive dividend, which is very much the norm in the industry and actually the reason why a lot of investors invest in this industry as opposed to just having to live with that risk of a potential dividend cut.

Second question is quickly on cost cutting.

I mean, it does sound like you're finally getting to the stage where you might be taking some risk with regards to the business.

So for example, cuts to regulatory or cuts to sort of changes to the manufacturing of commercial drugs to reduce the COGS.

So is that a correct perception that you feel that you need to take some risks to prop up the margin?

Or is that a misperception?

And how do you generally mitigate that risk?

And if I could just as a very quick follow-up on R&D where we really, really appreciate your candor.

I just want to make sure I understand correctly.

You don't think there's any issues with the science itself, so all of the problems were upstream and you think that the R&D NDA is actually broad enough to deliver continuing flow into the pipeline?

I'm just asking because listening to some of your competitors speak, it just sounds like they have a lot more technologies, a lot more internal databases; whereas listening to GSK, it always sounds very focused around specific areas as expertise like epigenetics, but maybe that's just the communication issue.

So any clarity there would be great.

Okay.

So perhaps (inaudible) for questions from the floor, Simon.

It's a great question on Novartis, but you might imagine, if we want something, then they're going to react, okay?

So we have to make it clear that it's an important capital allocation priority for us that we would like to own the whole business.

So you set up then a public dynamic, which probably not very helpful to try and to resolve it sensibly for both sides to continue to debate it in the open.

At the right point, they will be ready to sell and we will be ready to buy, and we need to buy at the most effective price for our shareholders and vice versa.

So I'm not sure we can really go backwards and forwards very much more on this other than to make it clear that we're very happy with how it sits today, they're very good partners.

We don't need to do something tomorrow.

But if they want to exit, we're very happy to buy it.

So I would like to correct the point we're taking risk with the regulatory or quality and manufacturing.

That is absolutely not our intent.

In fact, when you look back over the last few years of history, where we've had some, both us and others in the industry have suffered for some major supply issues, which frankly are extremely expensive, much more than any benefit you get by kind of cutting short site delay.

And that has often been because it was under invested in these fundamentals.

So please do not walk away with us -- the thought of us taking risk on quality or safety or regulatory.

What we are trying to do is get more competitive around our costs, around working our capital and around the productivity of our factories and around an end-to-end view of our supply chain.

And that's not, to me about risk taking.

That's about understanding what good looks like and holding that bar in the right place for us and we have got a very mobilized supply-chain organization, looking in a lot of detail in that.

That said, I have alluded a few times, that I'd like us to be a bit more of a courageous company in terms of placing some of the bets fully and in most obvious area in that is going to be in R&D.

So coming back to your question on R&D, I'll ask Patrick to overlay.

We have been and we are really focused on our development processes because the reality is that after this period up until 2020, I would like you to be able to have a renewed confidence and valuation of our pipeline for the next wave.

As a reminder, it doesn't need to come through until the mid-20s, great if it does and we would like to advance things as much as possible, that would be a part of the work.

So we are very focused on development.

Patrick's mentioned the renewal and ongoing quest to renewal of scientific expertise, whether it's internal or connectivity externally.

We've also alluded to a few areas just in terms of platform technologies in a more fundamental level, that we think we are competitive.

We might like to comment on those whether that's going after targeting much more efficiently in terms of genetic evidence or whether that's in our medicinal chemistry or, in fact, in some of our more advanced manufacturing technologies.

So I don't know if you want to comment a bit on the expertise from a discovery point of view as well.

Yes, I do.

And I think, Emma, you're right, we haven't spoken about it as much as we should have done.

And I want to pick up on a few things, but if I want to pick up on chemistry, we're the first company to get encoded library technology working full screening, which I think has made a big difference to screening.

We're very advanced in terms of the PROTAC technology, which allows them to pull out protein in cells chemically, which is a hot area in medicinal chemistry.

We have, I think, the world-leading proteomic organization in our Cellzome part of the organization in Heidelberg, which I think is well recognized as being able to do things in terms of looking at proteomic interactions with molecules isn't available elsewhere.

And they've had, as many other parts of GSK, have multiple publications in nature covering that.

In terms of genetics and genomics, we've got undoubtedly the biggest collaboration, which we stimulated with the European Bioinformatics Institute and the Sanger around genetics, which is led together with what we're doing with UK Biobank to deal where Regeneron have joined us to do the screening thing.

I think Regeneron has seen us as leaders in that.

So I think in terms of the science infrastructure, we've got actually very leading platforms across chemistry in particular, less good in some of the antibodies technologies, but we're definitely competitive, we're just -- I wouldn't claim we're absolutely out there at the front in those, some on gene therapy, I think we're the cutting edge of what's being done there with the first product to prove and some quite interesting approaches to how you make them.

And I would bring that through also to some of the approaches in big data from things like the Salford Study, but more importantly, 2 years ago, we appointed a lead for data across R&D from another industrial sector completely who's moved us from being able to access 20% roughly of our internal data to now over 95% of our internal structure data in a way that scientists can access it through 1 place.

I think if you ask around, I don't think anyone has managed to achieve that yet.

That's a huge data resource in terms of scientific input.

So we are very focused.

We clearly haven't spoken about it enough in terms of the discovery platforms we have with, I think, some really cutting-edge science.

James Gordon from JPMorgan.

3 short questions, please.

First one was commercial, just about China, can that return to being a big growth driver?

And if so, what's the plan there?

The second question was about doublets.

We've got the first doublets approval coming out quite soon.

But what does that do to the profitability within pharma?

I assume -- or is it fair to assume they are going to be significantly lower priced, and then you're going to have significant pay a way to J&J as well?

And then all the growth that's going to come from ex-U.

S. or from Europe?

So are we going to see significant margin pressure there and are there ways you can offset that?

And then just the third question was about the divestment today, and you talked about that roughly (inaudible) in terms of revenue.

So would it be very high profitability?

So does that create a bit of a free cash flow gap?

And is this -- is the streamlining done or could it be an ongoing partner to GSK for the few years that have more streamlining?

Okay.

Thanks very much, James.

It's a brief question.

Simon, could you pick up the divestment, please, and then I'll come over to Deb on the doublet.

Generally, below average profitability.

And also in the disposals we're making, we are divesting more capital-intensive businesses than our average, so we're saving ourselves some CapEx going forward.

So there should be net-net, material contributions to cash generation.

Can you take on the (inaudible)...

(inaudible) asset that we've got coming through the pipeline.

So there is an impact on the margin overall.

But actually, what we'll be doing is driving a greater volume of the business that we're really focused on our top line sales number and really driving that from a shared perspective as far as we can.

And what I would say is this is where we are today in terms of dolutegravir share when you add in dolutegravir plus rilpivirine, and then you add in dolutegravir plus 3TC.

We aim to have a higher overall share.

The dolutegravir after those 3 -- those 2 assets for launched on top of Tivicay and Triumeq when we still have significant business.

So it's a real share volume sales play, which ultimately will drop positively to the bottom line, but the margins does take a little bit of a hit.

And just in terms of your question on China.

Obviously, we took a big hit in China.

Absolutely delighted that the board continues to support investments in the market for obvious reasons.

It's important that we participate, not just commercially, but from a manufacturing point of view and also from an R&D point of view in China, with China, for China.

And I think we have to be patient in terms of seeing materiality of the contribution, but we're still very much reporting our progress there.

Back to the room for any more questions or others back online.

Okay.

So online, we have a couple.

So first for Simon.

Repeat question from (inaudible) asset management on what is our view on large M&A and the rating commitment?

Are we more focused on bolt-on.

So I think as we made clear in the presentation, the focus is very much on bolt-on partnership type deals, and we think we can accommodate the likely flow of those within the current balance sheet capacity.

Big scale, you would never rule out, but we've talked before about the disruptive nature of those.

And so the bar is very high for those.

And certainly, that's not something that's on the immediate agenda.

Okay.

And then Steve Scala, can we get a question from you, please?

Sure.

In his first quarterly release in July of 2008, Andrew laid out a strategy of growing a diversified global business, delivering more products of value and simplifying GSK's operating model.

He also focused on improving shareholder value and focusing on new strategic priorities to address the changing health care environment.

Today, you mentioned the DPU strategy is not changing.

How is the big picture strategy you are providing today different than that of nearly a decade ago?

Or are you saying that the strategy is the same, but execution needs to improve?

And then the second question is, what initiatives would GSK put in place to blunt an Advair generic, such as multiyear contracting and/or authorized generic?

Okay.

So I'll come back to your very important first question in a moment, but I'll ask Jack to pick up on the Advair generic question, since he's leading the U.S. business.

I appreciate the question.

As Simon had referenced in multiple meetings, right, there's still uncertainty around whether generic Advair will come, but certainly, especially with the Mylan and Hikma complete response letters, but the acceptance of the Sandoz, NDA.

So at some point, it will arrive.

We do have a whole array of different tactics that we will employ leading up to and through the presence of any generic Advair, when and if it does come.

So I feel like we've got the whole toolkit at our disposal.

Okay.

And so on your first question.

I think the -- you're right that the key part of it that is execution based.

And particularly near term, we need to be very competitive in our execution focus with the 3 launches that we've talked about and then ongoing shifting environment.

So I think I would probably highlight the 2 most important shifts.

The first one is putting innovation first and innovation within pharma first.

And -- never right to comment on previous leaders strategies, but when we talk about more products of value, there was quite a strong push around the volume agenda across the full diversified business, and we want to make sure that we're really focusing on building growth through volume and value that is driven through innovation.

We want to put R&D and science absolutely front and center of GSK for its next period.

And that is going to be visible in our capital allocation and focus on getting that business to competitive performance.

And that's the second big shift, I think, which is bringing more edge to the culture and the performance focus of the company through more focused choices, particularly in the portfolio through the people that we put in place and the changes that we want to make with the culture.

So portfolio of therapy areas, assets and market unashamedly putting the U.S. front and center whilst putting in a fit-for-purpose operating model for the emerging markets to drive the access that we have the responsibility to deliver.

Any more questions in the room?

Come on.

Thank you.

Graham Parry, Bank of America.

Just a couple of product questions and (inaudible), I'm wondering where that fits in the new strategy.

Does it fit in any of your core areas?

If you think at a positive outcome readout on that, is that something you'd be looking to partner out on that data?

Or would you look to rebuild a new franchise around that product?

And secondly, in Bexsero, you previously as a firm alluded to that as potential multibillion-dollar vaccine.

It wasn't on your list of key products.

And then thirdly, just going back to the IAS data and the Tivicay/Triumeq competitor from Gilead's bictegravir.

The physician feedback there was very much focused on softer issues that may be harder to detail against in the commercial market, so patient reported CNS symptoms, the tolerability of backbone, CV concerns, which have been, I guess, disapproved over time.

And so could you just run us through your commercial strategy for pushing back against that perception issue that's out there in the market?

So I'm sure there's only a degree to which we want to run through our commercial strategy for competitive reasons.

But I'll ask Deb to comment on that in a moment.

And -- but let me start also on Bexsero, it wasn't listed on the product because that was just a development product, not the assets that were listed on that chart, not the assets that are currently in market.

It's absolutely key for us.

And actually, our meningitis portfolio, we expect to contribute sort of 1/3 on Shingrix, we expect 1/3 to come from our meningitis portfolio, which is performing extremely well, and maybe Luc, I'll ask you to add a comment on that.

But let's start with Patrick on that great question.

Can you explain why we decided that, that was potentially still an asset that could bring real value to GSK?

Yes.

And so I think we've got a very, very good molecule there.

If you look at the dosage, it's about 5 milligrams, most patients going to end up on that substantially lower than competitor molecules.

We know it works, we know we can dial up and down hemoglobin in the Phase III trials.

Those Phase III trials are recruiting faster than expected, so we're ahead of the time on that, and we believe we've got a very clean molecule in terms of its on-target and off-target effects.

For example, nothing on prolyl collagen hydroxylase.

So I think -- we think we've got a good molecule, and it will read out in due course.

In terms of how we best commercialize that, that's an ongoing discussion as to how we best achieve that, whether we did alone or in partnership.

Luc, do you want to comment on Bexsero and then we'll finish with Deb on...

(inaudible) how it contribute exactly to this 1/3 of growth to the future.

So it's the meningitis portfolio, so it's Menveo, [ABCWI] and meningitis, Bexsero.

So it's growing 30% year-to-date versus last year.

And if you know that before the acquisition of Novartis, we were actually only selling less than 900,000 doses.

Year-to-date, we've supplied already more than 15 million doses.

So every dose we make.

And as Simon laid out, we are investing in capacity here to make sure we can support that demand.

In the R&D space, we are also working on the life-cycle management of providing a combination product, ABCWI , the full alphabet of meningitis, so which is the pentavalent vaccine.

So it is absolutely a key priority for Vaccines to be 1/3.

Thank you.

And then Deb -- well maybe join actually on the HIV question, you may want to answer that.

Let me just give a couple of principles and then we can work the argument through that.

So when we talk about CNS adverse event, it's usually a mixture of symptoms, whether it's insomnia, sleep disorder, depression, headaches and there's a whole list of things that often go to that list.

And you'll find the different companies, different groups define that differently.

In terms of the integration inhibitors, which we're talking about dolutegravir and bictegravir, it is a class effect and you do see it with all of them at around the same percentage.

So what we thought IAS, we're the first 2 of 4 Phase III studies for bictegravir.

And so the database that we see is fairly small for bictegravir.

Now the 2 studies they presented were in treatment naïve patients.

I think the interesting one is more of a direct comparison of dolutegravir and bictegravir, both with the same backbone, which was half and FTC.

And so when you look at the CNS adverse event and look at the long list of them, actually, the 2 drugs were fairly similar, but it's interesting in terms of patients discontinuing therapy, there was 1 patient in the bictegravir group who left therapy because of sleep abnormalities and none in the dolutegravir group.

The one that's a little more difficult to really get a handle on was the comparison of bictegravir, TAF and FTC against Triumeq.

And so in this study, they presented a patient reported outcome instrument.

Now the investigator who is presenting the study didn't go into great detail.

I do believe they then listed a whole host of symptoms where they looked at on particular days and it almost looks like you could be cherrypicking a little bit of what was going on with the patients.

I think we really have to get a better handle on that and take a look at the data there to see if there's any real differences.

But I think at the end of the day, my opinion is I don't think you will once we see additional data coming with bictegravir.

We do have this huge database that we have with dolutegravir, where it's really well-established, and again, we need to see a much more detailed approach there.

So I'll turn it over to...

Probably just to comment on bictegravir data generally (inaudible) .

Yes, I think that obviously, statistically, it was not inferior to dolutegravir.

But if you look at these 2 studies, and the first one I commented on, which was the more direct comparison in terms of treatment effect, dolutegravir was numerically better than what you saw with bictegravir for both the studies.

And so I think it really showed really what we know very well with dolutegravir with it being a very substantial drug for patients.

Let me just spend 2 minutes on competitiveness because it comes up a couple of times this afternoon.

So with dolutegravir, we have delivered a very strong data package into the therapy area where data is really the key driver along with guidelines as to how physicians choose to treat their patients.

So dolutegravir has got a significant Phase IIIb, IV program, which we almost completed.

5 studies where we demonstrated superiority versus competitors within the integrated class, but also versus the other kind of sedations that are used, so really strong data, that ultimately drives with an HIV physician behavior.

However, if we talk about on top of that competitiveness, you've now seen (inaudible) versus Triumeq and you've seen bictegravir versus dolutegravir.

And actually, they look fairly comparable.

But we're moving the market on, again, with the 2 drug regimen pipeline that we have starting with rilpivirine and dolutegravir, moving into dolutegravir 3TC and then moving into long-acting injectables with cabotegravir and rilpivirine.

So we feel that we're moving forward with our pipeline that we will move the market forward whilst our competitor is still in the 3-drug regimen paradigm.

Now let's see how that plays out.

And then obviously, in the future, we've got other strong molecules in our discovery portfolio, which, again, we believe will continue to move things further on.

And we do believe long-acting is going to be phenomenally important.

And on that basis, I think that, plus very strong performance commercially, is going to lead us to be winning from a market share perspective.

It's really the innovative option that we're developing rather than more of just incremental changes.

Thank you very much.

So I'm going to take one more question from online.

The last question coming to Simon.

Of course, we are all available to pick up on any other further questions that you may have.

The last question is from [Paul Casley].

Can you please provide some color behind the revised outlook for upward pressure on the group tax rate over time?

What's changed?

So I think as I highlighted in the presentation, as the mix of the business changes and particularly given the prioritization to the new asset that Emma commented on, we are putting more of the revenues and profit streams into a higher tax jurisdiction.

And that obviously puts to one side where there might be tax reform in the U.S. or not.

But on the current tax rates, that is creating this upward pressure.

Plus in a post-spec world, we are seeing much more activity from tax authorities and challenges and disputes, which, ultimately, we will seek to resolve in an appropriate way and a balanced way.

But they will quite often require provisions in anticipation of quite long periods of disputes.

So I think that will also be part of the drag going forward.

It's mainly about the shift to the U.S.

Thank you, Simon.

So thank you to you all for coming here today or listening in or watching.

Thank you for your thoughtful questions.

We can obviously continue some more discussions now.

But I just wanted to say a couple of very brief words on how you should expect us to be updating you on our performance going forward.

You're obviously going to hear from me, but also consistently members of the leadership team at our quarterly earnings calls.

And we will be holding more regular meet the management sessions, so that you have the opportunity to meet and get to know the board or team from different parts of the business.

And lastly, obviously, critically, we'll be updating you as of our pipeline progress, particularly around the assets that we've highlighted today, sharing an important data as it comes through and making specifically our R&D leaders available to you all for questions.

So thank you very much, and please do join us for some more refreshments.