This is not the introduction for the event. Safety briefing. Okay, we generally start by saying that there will be no alarm this morning. We don't know about that. There will be one. At 3:00 P.M. there will be a fire drill. It's just an alarm. We will wait here at 3:00 P.M. It's a test. Be aware of it. Nevertheless, if anything happens, there will be an evacuation system, a voice system, which will be activated telling us to evacuate the building. From that on, we'll be escorted by staff from ETC. Nevertheless, important to know that the fire exits and the emergency exits are right out of this door, when you get out of this room, just ahead. Otherwise, to the right, same direction as the bathroom, leading you to the outside mustering points.
I think those were, in addition to the normal observations, those were the main points on safety. Unless you have questions, we'll wait for 1:00 sharp before we start the event. Good afternoon, all. Good morning. My name is George Menane, and I'm a Vice President at Shell Investor Relations, and I'm delighted to welcome you all here in London and on the phone and on the web to our first Chemicals Investor Briefing. The purpose of today's session is to shed some more light on our chemicals business, which is one of our growth priorities and an important component of our world-class investment case. The format of the session today will be a one-hour presentation, followed by a one-hour question and answers. Questions will be from the room, but also from the phone line. At the beginning of the Q&A session, we'll announce the practicalities for that.
For this Q&A session, we'll invite you to think and focus your question maybe on the topic of the day, which is chemicals. Having said that, we are very pleased today to have John Abbott, our Downstream Director and member of Shell Executive Committee, and Graham van't Hof, our Executive Vice President for Chemicals. Having said that, with no further ado, I would like to welcome John to take us through the first part of the presentation. Thank you very much.
Thank you very much, George. Say good afternoon to all of you in the room, and I'd like to extend a particular welcome to those of you who are on the line. Probably good morning for some of you in the U.S. and Canada, and probably some of the investors I met in Canada, in Toronto and Montreal recently, so good morning to you. As George has said, we're here today to talk about chemicals and give you a deeper dive into our chemicals business. But as always, before I start that, I do have to highlight our disclaimer statement, which I know that you are intimately familiar with now. Having reread that, I'll move on to the energy challenge. This slide really looks at some of the large and important trends that are really affecting our energy sector and influencing it.
It's not just the energy sector, it's influencing our society, and it will absolutely shape our industry, and it will shape Shell in the next decade and beyond. There are ongoing long-term changes in society that actually affect that landscape. I know that many of you are familiar with those, but it's important when we get to our chemical story. Increasing population, increasing urbanization, an increasing number of people moving into that middle income bracket. Really, that is rising living standards and additional vehicles. All of this adds up to an increasing energy demand that could double between the year 2000 and the year 2050. At the same time, of course, quite rightly, the world has to take action to reduce its greenhouse gas emissions.
Energy efficiency and growth in renewables will inevitably impact our energy demand growth and the sources of energy supply on which we rely. Let's go to our strategy. Let's Make the Future actually speaks to Shell's purpose to provide more and cleaner energy to society. This is going to be absolutely key to our progress for decades to come. Our strategy is based actually on four aspirations, and each one of those guides every one of the businesses that is part of Shell's portfolio. The first one is to create a world-class investment case. This is about delivering higher, more predictable returns and growing free cash flow per share. Underpinning all of that is a conservative financial framework for which Shell is known.
We also want to maintain our scale and strengthen our leadership position to sustain our ability to influence key issues in the energy sector, both today and going forward in the future. We have an ambition to reduce Shell's carbon intensity. This is absolutely inevitable, and it's something we are working very hard on today, and we'll continue to put emphasis on this, as well as on the broader challenge of not only navigating but thriving in the energy transition to which we are now all part of. We also need to establish beyond all doubt that Shell provides shared value, that we're a force for good in society, and that we create value not only for our company and business in its own right, but for our communities and our stakeholders where we operate.
This is actually quite hard to do in terms of metrics, but it's also quite hard to do in terms of practice and getting it right. These aspirations are not actually independent of each other. We need to succeed in all of them to deliver on that world-class investment case that I've talked about. This is an ambitious set of aspirations that speak to me personally as well. I like this level of ambition that we now have in our company and believe that this is the basis for differentiated company performance. I really do. Now, many of you, because I can recognize the faces in the room here, will be very familiar as to how we now segment our portfolio in the Shell group into a number of strategic themes, but I think it's always worth reminding ourselves.
The first category, the cash engines, these are needed to deliver strong and stable returns and strong and stable free cash flow that can cover our dividend and buybacks throughout the whole of the macro cycle and leave us with enough money, of course, to fund the future. Our growth priorities have a clear priority and pathway towards delivering strong returns and free cash flow actually in the medium term. Graham will talk later on about the chemicals part in that. Our future opportunities, of course, should then provide us with material growth in cash flow per share in the next decade. Through all of this, our intention is to be in fundamentally advantaged positions. This actually demonstrates in my mind that Shell is preparing for and investing in the challenges and opportunities that the energy transition will inevitably offer us.
With that word, let me now quickly transition to the downstream. In downstream, you should have no doubt in your mind that the priority remains the continued improvement of our financial performance and ensuring future resilience through the energy transition. It was actually, as many of you will know, and I certainly know that it's an intensely competitive sector with fundamentally narrow margins. That has been the story of my career, working in fundamentally narrow margins within downstream. We've been restructuring our portfolio, and in the last decade, we've actually exited 23% of our refining capacity. That's 900,000 barrels a day. There has been a lot of change, but of course, we continue to scrutinize and upgrade our portfolio as part actually of our ongoing business. Many of those parts of the downstream business you'll be aware of.
In marketing, which we also class as one of our cash engines, the business are delivering resilient and increasingly improving results. This is due to three things. It's due to our investment in strong brands, the Shell brand, our differentiated products and offerings, and our premium fuels and lubricants, which hopefully all of you purchase. To build on this, we plan on continuing to selectively grow our networks in rapidly growing economies, and that means Asia and beyond. You'll hear more about this in the future. In refining and trading, which we also class as a cash engine, we've done a lot in the past couple of years to improve performance of our refineries through many self-help programs, but also through a much stronger integration with our world-class trading organization, both on crude supply and on all products.
Crude flexibility within our refineries has been key to creating that integrated value between refining and trading. Now in chemicals, advantage feedstocks have become a really competitive advantage for us. Chemicals is a growth priority for Shell. This strategic position recognizes the attractive and different risk profile of the chemicals business, which Graham will talk about. This characteristic is underpinned by strong market fundamentals of high growth rates, different end market exposures, and attractive returns when projects are able to access those advantage feedstocks as I've just talked about. As I say, Graham is actually going to give you new information in this area today. In summary then, overall, our strategy to develop and maintain hubs where trading, refining, and chemicals can be fully integrated to date has certainly proven to be the correct one.
It has proven especially helpful during the last few years as well, as we've seen periods of high volatility, low volatility in the market. The resilience that the downstream has given us has been quite notable. This, of course, has been certainly supplemented by our large, growing, competitive, and differentiated marketing business. On that point, I will now hand over to Graham, who will give us a deeper dive into our chemicals business. Graham, over to you.
Brilliant. Thanks, John. I've met a number of you in the past when I've played bit parts in participating in investor meetings. I think behind that is the realization that what we haven't done in announcing chemicals as a growth engine in Capital Markets Day in 2016, we haven't provided the depth of understanding behind why does that make sense for us, why does it make sense for Shell, and what is the underpinning of how we see growth in value being developed over the coming years. For those of you who don't know me, I've been in Shell for a little over 30 years. Sometimes a bit of a depressing thing to say. Nevertheless, most of that has been in the chemicals business. Not all. My last role, I was head of Shell in the U.K.
I was also head of what we call CO2 and alternative energies. I've been doing this role since the beginning of 2013, coming up to five years. In seeking to explain where we are in the chemicals business, first of all, I'm going to start off with the whole demand story. The thrust of the demand story essentially is this is the demand story for hydrocarbons compared with uses as fuel. Where do we see our positioning within the chemicals industry? Why is it that Shell seeks to participate? How does this make sense based on Shell's capabilities? Our strategic positioning, the exact strategy that we have and how we translate that strategy into execution.
I'll go through that with a whole series through a series of examples, get into the growth story about projects to come, what does that look like in terms of our financial future and the aspiration that we have in terms of development of cash flow going forward into the 2020s. Let's kick that off, you may look at this and say, "What on earth is this?" This is a thing called the Solar Impulse for those who are not familiar with it. This plane circumnavigated the world in 2016 based only on solar power, so went the whole way around the world based only on solar power. You might say, "Okay, that's very interesting. Now what's the so what?" The so what is 90% of it is basically made from chemicals.
It's just a simple example as to where the chemical industry and the backbone of chemicals going into the chemical industry is a part of the growth as to where society is going and the solutions that society is seeking to bring to the table. If we look at that in a slightly more classic way in terms of looking at the growth profile going forward, here you just see a chart in terms of the growth by region of total petrochemicals demand. The punchline is you're seeing something like a 50% increase in demand between the present day and 2030. Then you see sort of highlighted where's the engine room of that demand coming from, and the punchline is it's coming from Asia, and that's about population growth, and it's also about people where household income is fundamentally increasing.
Maybe for some of you who are less familiar with the chemical story, you might sort of go chemicals, and then you might jump to maybe plastics, and you might jump to plastics and go to sort of plastic bottles, and that might be your sort of line of thinking about is isn't that what the chemical industry is all about? I think one of the things that I wanted to illustrate with this picture is the huge level of diversity in the economy that petrochemicals go into. One of the things you will see is actually a very substantial amount of this goes into medium to long-term use. It's medium to long-term material use as opposed to short-term use. Obviously by contrast with the fuel side of hydrocarbons where in essence pretty much everything gets burnt, we're putting this into long-term materials consumption.
Obviously you can see the whole range of different types of products and all the exciting solutions that come to society as a result. Where and why do we seek to play in the value chain? Let me explain this chart. Let's just start on the left and the right, first of all. On the left-hand side, you can see feedstocks, so hydrocarbons coming from whether it's gas-based, liquid-based, out of refineries and so on and so forth. That's the start of the petrochemical industry. In the far right, obviously at some point, we end up with you and me as consumers. Across the top, and I'll provide a bit more detail in a second, showing a little bit more breakdown of lots of products and so on, but here we'll just stay at the big principle level.
Across the top, we start off with base chemicals. We move into what we call intermediates, then getting to more performance products, and then moving very much to solutions and almost service-based over on the right-hand side, as we come into servicing the consumer. What's fundamentally different between playing on the right and playing on the left? Well, let's start on the right. Classically, what are we seeing here? We're seeing businesses where we're playing a lot with tailoring the product. We're seeing lots of SKUs, so a lot of complexity over there. A lot of things like very people-intensive in terms of tailoring products, in terms of technical service, in terms of salespeople. In general, if you look over on the right-hand side, you see a very people-intense capital light business compared with what you see on the left.
What you see on the left is the complete opposite. You see this is all about advantage feedstock. It's around access and optimization of hydrocarbons. It's around world-scale assets, so very large world-scale, capitally intense. It's actually over on the left-hand side, it's people very light. There's a certain dichotomy between participating on the left and participating on the right, which when you actually look at the competitive landscape, which I will show you in a moment, you can see how actually the industry tends to move somewhat towards the left and towards the right. Why does participating where we participate make sense? Well, if you look at these fundamental bullets underneath where we see the pecking order here on the left, you can see how this links to Shell's basic capabilities.
As we go through some more slides, I'm going to actually add a lot more content behind these bullets, but I'll just explain them to start off with. Obviously, access to advantage feedstocks is fundamentally about access to the right hydrocarbons. We are a hydrocarbon company. That's what we do well. Finding the right hydrocarbons in the right places at the right value to create advantage is something we do naturally well. Process technologies, I'm going to give you lots of examples as we go through, but fundamentally, process technologies is actually a core to our chemical business, but it's actually a core to other parts of Shell as well. It goes across the Shell portfolio. Obviously, perhaps a simple example you'd be very familiar with is our GTL facility in Pearl in Qatar.
It's the same basic chemistry underpinning goes to support development of Pearl, as does support and development of our chemical business. Big integrated site is absolutely critical in terms of getting both optimization of hydrocarbons, optimization services, things like utilities, but also optimization of the cost base in terms of running them very much as an integrated base with a single management structure, and then obviously large scale. Let's now look at some more detail behind that. Now I've thrown a whole bunch of content into the same kind of view that you were just looking at. You're still looking at the same left to right that you were looking at before in terms of feedstocks on the left, coming ultimately to consumers on the right. Perhaps just start off for a second and just stay with the colors in the middle.
What you'll see is in the yellow ones, this is where Shell's playing. The blue ones is where Shell is not playing. You see illustrated the point I was making before about us being on the left-hand side. What you will see to start off with is very strong in the base chemicals, moving into a certain amount of intermediates where we have fundamental advantages, and only occasionally moving into the performance space where we feel the need to go one step further in order to have advantaged monetization of what we create on the left-hand side. The other thing I'm going to tee up here, which I'm going to come onto with a bit more detail later, is what's interesting about the feedstock piece.
One of the things you will see here is there are a number of different feedstocks which can get you into this petrochemical chain. You can have obviously ethane, or you can have naphtha, for example, going into ethylene crackers. That gets quite interesting because of the diversity of valuations between those feedstocks, and we'll talk about that. As those valuations change, it also changes the competitiveness of the value chains that are running across here. That changes the competitiveness of those value chains and the products in them relative to maybe competitive products out in the market. You could think of something like polyester competing with cotton, for example, or natural rubber competing with synthetic rubber. It also affects the competitiveness across the value chains themselves. You can think of polyethylene competing with polypropylene, or polypropylene competing with polystyrene.
As these hydrocarbon valuations in the feedstock shift, they shift the underlying competitiveness of what's happening in the middle, that changes the demand patterns of what's going on in the middle. This all creates quite a lot of churn and volatility in terms of the margin environment, which is a little bit different than the chemical industry that you saw 20 plus years ago. Where actually there was a lot of stability and continuity in the valuations on the left-hand side. You didn't get all these disconnects, which we're now seeing in terms of both the values of the feedstocks, also then the perturbations that they then create in terms of different strengths of competitiveness of products on the right-hand side. This is a very important point in terms of diversification of portfolio.
Let's say a little bit more about strategic positioning, I think this is an extension of the picture that I was showing perhaps in one dimension before. On the X-axis, you're looking at moving from left to right in that value chain picture that I was looking at previous, so from basic chemicals over to solutions. On the Y-axis, you're looking at having either generic feedstocks or moving to advantage feedstocks. Clearly, the IOCs tend to play towards the top left of this. That's a natural place for an IOC to play. You also see that the positioning within that for Shell tends to be towards the top left. Seeking to have a lot of advantaged feedstock positions. Also, we've taken the strategic decision to keep our portfolio relatively simple. That means that we can operate lean.
We tend to have a low cost base, I'll come on explaining that in more detail a little bit later. Clearly, there are other positions around here. Perhaps the one that's most obvious you might go to is the bottom right, you could think of a company like Bayer or some of the classic chemical companies. Obviously, we've seen a bunch of restructuring of that in the last 10, 20 years, somebody like a Bayer, BASF, may be a little bit more towards the left-hand side of that in terms of strategic positioning. What is our strategy? The strategy should start to probably be fairly obvious from the tee-up that I've given so far. We see four building blocks to our strategy. It starts off with great opportunities for advantaged feedstock.
That advantaged feedstock can be refinery, refining based, or it could also be gas-based coming out of an upstream position. There's no point having an advantaged feedstock if you don't then effectively monetize it. Effectively monetizing it consists of a number of steps. The first is you better have a world-class footprint in terms of the manufacturing base and platform that you have. Secondly, you better have excellent products to put into the market to make sure you're fully monetizing. Align that with an excellent customer portfolio, and then you better operate everything extremely well in terms of what we call excellence every day, or you could think of as operational excellence, safety, reliability and so on. If you characterize that very simply, we start on the left with access to advantaged feedstock, and we move to the right, which is monetized with competitive advantage.
The underpinning for that in the long term is nearly always technology. Continued investment in technology and having advantaged technology positions is your fundamental long-term advantage position. There's something else I've put on this slide which is quite important, which is the word competitive advantage. Philosophically, we do not believe that you can or will get rewarded in the petrochemical industry if you do not bring some level of competitive advantage to the table. Just doing me-too investments in me-too locations, we see as not likely to achieve a really good rate of return. If you look at the investments that we make, and I'll speak to those investments later on in the presentation, we only make final investment decisions on investments where we say we can see a fundamental basis for competitive advantage. I will talk you through what those bases for advantage consist of.
How do we make that happen from an organizational perspective? What you see in this chart is kind of our organization at two levels. Obviously, the top level here is a kind of hierarchical view, if I can look at it like that. The bottom one is kind of a representation of how we make this come to life in practice. You can see the chemicals business here. The reality is we make this happen through integrated sites. We run single integrated sites. Sometimes it's chemicals alone, if it's an advantaged gas feedstock, or chemicals with a refinery in a mega complex.
That's run with a single management structure, a single set of optimization of hydrocarbons, a single set of optimization of utilities, it keeps it very lean in terms of the fundamental cost base as a result, it makes sure that we optimize always across the whole wherever the best advantage needs to flow to. The optimization of hydrocarbons coming in and out of that, and particularly in, to an extent out, we optimize with our trading colleagues. Our trading colleague, Andrew Smith, who runs the trading business, and I should refer to Lori Ryerkerk, who runs our refining and manufacturing business. Then over on the left-hand side, where we're looking for advantaged gas positions, we work with our upstream and integrated gas colleagues, Andy Brown and Maarten Wetselaar. You can look at existing positions in places like the North Sea.
We have a shared cracker with ExxonMobil in Scotland, that's fed from the North Sea. We work very actively with our upstream colleagues in terms of accessing advantaged wet gas to get cheap ethane to feed into that cracker. Also, if we get to our future portfolio, we look at the projects we're seeking to bring to fruition in the medium to longer term, a lot of this is being worked with our colleagues in upstream or integrated gas. As I said, in the long term, the underpinning of all of this is technology. Also how do we bring projects to fruition, mega projects, mega scale projects? A lot of capability required to do that really well. That's in our projects and technology organization. I'm sorry, I thought that was an alarm going off. Which is run by Harry Brekelmans.
The point being is to illustrate two things: the interdependence of working, also this is an illustration of the synergies with the rest of the Shell Group. It also adds to what the group is doing itself. We're using the same capability base to work on chemicals that we are to working on the rest of the group. All right. I'm now going to step back. If you remember those four building blocks of our strategy, you'll see those building blocks across the top right here, I'm going to walk you from the left to the right in terms of those building blocks and make points under each one of those building blocks. We'll start on the left in terms of advantage feedstock.
The chart you're looking at is looking at cracker margins in the three different regions, fundamentally, you see those three different regions with the US Gulf Coast being based off ethane or ethane to our US colleagues, Europe and Asia being based primarily off naphtha cracking. Then if you look at the chart, what do you see? We've called this out in terms of these gray arrows. Fundamentally, pre-basically shale gas, you saw that ultimately these margins equilibrated as the hydrocarbons fundamentally equilibrated. Ethane and naphtha naturally equilibrated based off flexible feed crackers in the US, that always kept ethane and naphtha cracking fundamentally in step. That equilibrated across the planet. Post-shale gas revolution, this dislocated.
This dislocation is incredibly important, both in terms of then providing opportunity for value from cheap gas streams, also then in terms of the subsequent trade flows and the phenomena that I was describing previously in terms of the intermaterial competition and the lack of predictability that has come into the petrochemical industry compared with the period pre this point. That's a really important point both in terms of the ability to predict the future compared with what we saw previously, as well as the need for portfolio diversification, because it's difficult to predict. You might kind of ask yourself, well, where is this going in the future? I think obviously you can go and look at all sorts of analyst predictions.
The one thing I would say is obviously even if you look at today's environment, if you pick a gas price of something like $3.50 per million BTU, an oil price equivalent is $23 a barrel. In the U.S., you're something in the $20 a barrel level in terms of your feedstock cost, whereas obviously if you are operating based off naphtha, you're operating somewhere off a crude price equivalent of some kind. That is still quite a big gap, and now you can use that gap up in terms of things like shipping costs or capital costs, but there's still quite a significant gap. I think how this pans out is inherently what makes the future extremely interesting in terms of the future of petrochemicals. Sorry, I missed something. Thanks very much.
The point on the right is to say that before this 2008 period, the balance of our portfolio was very substantially liquid-based. One of the major corrections we had to make in the late 2000s was our U.S. portfolio was pretty much liquid cracking based with some exception. We needed to very quickly move that over to gas-based, and that's what you see in terms of the shift from the top right to the bottom right from 2007 to 2016. We are roughly 50/50 between sort of oil-derived, naphtha-derived, and gas-derived. Roughly speaking, we see that balance staying into the future, and I'll show you that a little bit later. Let's look at our portfolio. What you now see, but you would not have seen 10 to 15 years ago, is a lot of balance between our regional exposure. Quite balanced across them.
Fifteen years ago, you would not have seen very much in Asia, and that's the main growth that we've taken and put into place in the past decade, has very much been adding to our Asian footprint. If you look across the top, you see something very interesting, which is how our footprint has changed in the last 15 to 20 years. In 1998, we had 133 manufacturing plants, and now we have 15, and we are a substantially larger business than we were in 1998. What was that all about? Well, if you went back to our competitive positioning chart, in 1998 you would have seen that pecking order rather further to the right. We were participating further down in the value chain, but we were also participating in a number of areas where actually we had no competitive advantage.
We drifted into the space where fundamentally we were not using the core capabilities and strengths of the Shell Group in a useful way. As a result of that, we made the decision to actually divest a number of businesses. I happened to be running one of those businesses in the late '90s, which was our global polystyrene business. We decided we had no fundamental advantage in that business, we divested it. As a result, we were able to drastically simplify the base of our portfolio and reduce our cost structure. With a lot of the cash that was generated at the time, we were able to then strengthen the core that we had. If you looked at the assets that we had in 1998 that we retained, a lot of those were not yet world scale.
Between 1998 and the mid-2000s, we've moved those to world-scale assets and got to the world-scale type of footprint that we now have today. The other thing that we did was then strengthen our position on the right-hand side of this in terms of what we put into place in Asia. Obviously, the investment we made in what we call Nanhai, our joint venture with CNOOC in Guangdong Province in China, with a petrochemical complex there, which we put into place in 2006. The addition of a cracker in Singapore in 2010, a world-scale MEG unit, and then subsequent debottlenecks of those. Also, the buyout of the joint venture that we had with BASF in SMPO in Singapore. We've substantially increased our footprint across Asia as a result of this.
Also illustrated, which I referred to before here, is a number of where we've got these core locations with refinery petrochemical mega sites. Let's look at now the product and the customer portfolio. You might not realize, but we've actually been in this business since 1929. 1929, if anybody is a chemist of any kind, you'll realize that a lot of this chemistry was only just getting discovered in this period, somewhere between the '20s and the '30s. Shell has been active in this space all the way back through then in terms of technological development and development of businesses. We have about 1,000 customers. I don't know if that sounds a lot or a few to you, but in this industry, it's relatively few. Within that 1,000, actually, probably 90% of our business, plus/minus, is in the top 100.
That is very consolidated relative to a lot of competitors. What that means is we're able to have quite consolidated positions. We really operate our global business from only three hubs. It's Houston, it's Rotterdam, it's Singapore. We don't have lots of marketing offices around the world. We don't need to. The reason we don't need to is because we've got very large major relationships with major partners, which we've had for 20 years and we will have for 20 years. These are very deep relationships embedded in deep contracts, long-term relationships, but relatively lean as well. We talked about the product portfolio. Let me illustrate a few points, and you'll see that the main of our products are all proprietary technologies that Shell developed. In styrene monomer, we have this process of styrene monomer and propylene oxide. Styrene monomer, we're number one in the world.
In propylene oxide derivatives of polyols, we're number three in the world. In higher olefins, we are number one in the world. In detergent alcohols, we're number two in the world. In monoethylene glycol, we have the world's leading and a leading market share in catalyst and licensing provision and a major position in the markets. More recently, we've developed a new technology for a product called diphenyl carbonate. Diphenyl carbonate is the critical precursor to get yourself into the polycarbonate chain. We've got that technology now to the point where we are able to start commercialization, and that is something that we are progressively working on because we see this technology having a fundamental cost advantage versus the alternatives as well as being fundamentally safer than the majority of the alternatives, which use a chemical intermediate called phosgene, which is very nasty indeed.
Obviously from a process safety and safety concern is not something that people and governments tend to want to see proliferating. Moving now on to the operational excellence space. This is a space which I would say has been a challenge but is a fundamental opportunity in terms of our value creation going forward. What I'm looking at here is I'm looking at two dimensions of unplanned downtime. Perhaps focus first of all on the red one. The red one is what you might call normal unplanned downtime. This is perhaps a trip of an asset. It's out for one day, two days, three days, four days. It's kind of normal, let's say, recurring reliability challenges that one has. Clearly, one wants to drive that down over time to make sure you've got the maximum capacity and the maximum utilization.
What you see is the general pattern in the red is this pattern has been going down steadily over the last several years. Quite pleasing in terms of the underlying that we've got in terms of the general reliability. What has not been pleasing has been the yellow, which we call big hits. That's perhaps a reference very much to, as we look at it from the chemical business, perhaps it's less material at a total group level, but certainly from the chemicals level, significant. You'll see we had outages in 2014 and 2015, particularly, two different outages in Moerdijk, one related to our SMPO-2 assets, the other related to our cracker in Moerdijk. Then the other significant one which we'd speak to was the outage that we had in our cracker in Singapore.
Three significant events, quite a lot of outage time impacts both in terms of our financials, which I'll get onto in just a second, but also obviously impacting our customer base and some of those deep relationships that we have and us needing to work very hard in terms of making sure we're sustaining those. As would be fairly obvious to you, we've invested a substantial amount of time in understanding the different root causes behind this and putting in place a number of corrective actions to make sure we get to a better place, which is broadly speaking what you see here and what you see coming into 2017. That's our move to then what we see in terms of financials. I'm going to make a number of points around the underlying financials.
Let's stay on the top left in terms of the earnings chart, first of all. First of all, look at the gray line, which is basically illustrating a pretty flat earnings performance. As I was just talking about, if you actually went to 2014, 2015, and 2016, on average, you would be tending to see that gray line more towards the $2 billion mark rather than the mark that you actually saw, and that's attributed solely to what I was talking about previously in terms of those outages that cost us substantial money as a chemical business. That would move our underlying performance that we've seen in the last several years to about the $2 billion mark. The other thing that's quite interesting about this is looking at the two red lines.
Between 2011 and 2014, you've got an average crude price of $104, 2015, 2016, 2017, $49 a barrel. What's interesting about that? Well, obviously one of the sources of profitability has been, let's say, the gas-derived part of your portfolio. That's most profitable when you've got a high gas oil differential. In that period through sort of 2011 to 2014, very high gas oil differentials, very high profitability coming out of that part of the portfolio. The interesting thing is, even as that has collapsed, actually our performance has actually strengthened. The businesses, the part of our portfolio which was making the most money in 2013 and 2014, was actually making the least in 2015 and 2016, and vice versa. The ones that were making the least in 2013, 2014, were making the most in 2015, 2016.
Clearly, one can spend a lot of time strategizing around this stuff and trying to pick the winners. For us, having an appropriate balance in our portfolio going forward is extremely important to be able to catch whichever are the right waves as you go. Clearly, we're very pleased with the financial performance we've had year to date, and here we've expressed it on a 4 quarter up to the 2Q rolling basis, clearly looking extremely healthy. As we move to the right-hand side and we look in terms of our return on capital performance, we're very pleased to see the uptick, which frankly we would've liked to see more in the last 2 or 3 years, was depressed because of these big hits. Moving into a much better position in terms of our return on capital performance.
If we sort of stop at that point a second and say, okay, where does this go forward? How are you going to drive value improvement and value growth going forward into the future? We have 3 fundamental levers that we look to to drive value going forward into the future. The first speaks to just improve the base that you already have. I'll come back to that in a second as to exactly what we're doing to it. We talk about adding to your existing portfolio, so the footprint that you've already got, adding new units to it, and so on. The third one is adding new pieces to your portfolio completely, absolutely completely new parts to the portfolio.
In the coming slides, I'm going to speak very much to what we're doing in the second 2, so projects that we're bringing to fruition in terms of additions to the existing portfolio and fundamental new footprint. Just for a second, I'll speak to the one at the top. At the beginning of 2016, we formulated a plan to improve our underlying profitability performance by half a billion dollars, based on where we were at that level. The fundamental levers that we were driving for that are a mix between cost and value. Our intent is to get to $500 million improvement in our income performance. At the end of 2018, so embedded in our 2018 plan, we will have got to about 80% of that, of which is a mix between about 40% is fundamentally cost-based and 60% value-based.
If I give you a sort of illustration of what's going on there. On the cost-based side, some you have in the fixed cost side. If you look at my own organization, for example, I will be running with 10% less people at the end of 2018 than I was at the end of 2015. A reduction in terms of the basic cost base. Also working on things like unit rates, optimization of what we'd call the cost to serve, so you're more on the variable cost side, is where all those costs are coming from. On the other side, in terms of the value side, the things that we're most interested in is driving volume of our core assets. Creeping very cheaply, keeping the capacity that we've already got.
We've got several great examples that we've already put in place in the last year and a half. If you look at the expansion of our output in the Norco cracker in Louisiana, with no real major capital, we've improved, added probably about $75 million of output from just that Norco cracker. Similarly, the cracker we have in Bukom, in Singapore, continued creep. Both have exhibited production records in the past year. We've added to our EO capacity in Singapore, again, with extremely cheap investment, just a few million dollars adding to substantial amounts of capacity. Beyond that, we've deepened our working with our trading business in terms of looking for where can we add value, both in terms of the feedstock optimization, but also in terms of the optionality that we create on the sell side of our portfolio.
Now I'm going to look at the second two bullets in terms of those growth projects going forward. Perhaps before I do so, I'm just going to talk a little bit to how our portfolio has developed over time, as you can see the capacity sitting in this big red chart in the middle. You see as we move from 2000 to 2005-2006, Nanhai, the CSPC joint venture in China, shifting our total capacity up. We went through a certain level of retrenchment period, which was all around restructuring our U.S. Gulf Coast position, which was based on these liquid feedstocks. We needed to shift from liquid feedstocks to gas feedstocks, but in doing so, the absolute output also got reduced at the same time. We started up the cracker in Bukom, and then subsequently, we've been working down the debottleneck of those assets.
What have we got to come? Three major projects are past final investment decision. The first one consists in that middle category of adding to the existing footprint, which is the addition of a fourth alpha olefins unit in our Geismar, Louisiana facility. I'll provide more detail on each of these in just a moment. The second is a second petrochemical complex sitting alongside the first petrochemical complex that we already have with CSPC in China. The last one is a brand-new petrochemical complex, ethylene to polyethylene, in the heart of the Marcellus Utica in Pennsylvania. Let's talk first of all about the first of these, which is the addition of a fourth alpha olefin unit in Geismar, Louisiana.
This is a world-scale facility. Beyond this, it will bring the site to a total capacity of 1.3 million tons, which is 25% of the world's capacity for alpha olefins is all on one site of ours. This is a highly integrated site, moving from alpha olefins to internal olefins to alcohols, also EO ethoxylation. The whole chain between creating alpha olefins but also potentially moving into detergent intermediates is all in one very highly integrated and massive site. That is then highly integrated with the feedstock side, which is either Norco, which is just down the road in Louisiana, or alternatively, supplies from Texas, all in pipelines that Shell owns themselves.
This is based on our proprietary technology. Actually, we're working on another generation of the proprietary technology and looking to bring a pilot plant on stream next year for the next generation after this one, which is just a reinforcement of how we're continuing to invest in technology and have the best technology in the world. This project's been going extremely well. We're extremely pleased with it. Perhaps a contrast between some of the challenges that some of the big projects in Texas have been experiencing in terms of productivity. We're beating all our aspirations here. This will be on stream in the second half of 2018. In China, we're adding a second petrochemical complex to the existing petrochemical complex that we have there. We've been extremely pleased with the operations of our existing joint venture. It's been operationally outstanding.
The workings of the joint venture and our relationship with Sinopec has also been outstanding. Based on that relationship and the existing success of the existing platform, we agreed together to add a second petrochemical unit. We closed that deal during 2016. That will add over 1 million tons of capacity, bring a number of Shell proprietary technologies to fruition in that asset. That will be, when complete, the largest petrochemical site in China. We will see it being the most competitive petrochemical site in China, highly integrated to two refineries. Two refineries, two petrochemical plants. This is a very impressive mega site also in the heart of South China, which is perhaps the less competitive of the competitive landscapes between South China and moving to East China and Shanghai. The final growth platform is our investment in Pennsylvania.
As I said, this is the heart of Marcellus Utica. The concept is to take cheap ethane, which otherwise we'd have to find alternatives elsewhere, particularly moving that ethane down to the very strong demand that's pulling on it now with the investments in the U.S. Gulf Coast. The valuation on that has to be at a discount to the U.S. Gulf Coast. We're converting here in a one and a half million ton cracker into 1.6 million tons of polyethylene in this space. You can see these pictures down below, and maybe we could have even provided one which was pre-today, which was looking at the site as we took it on. One of the things you'll see is, as of today, it is in extremely good shape.
For those of you who followed this closely, we took the decision to manage affordability last year around the whole BG acquisition funding to actually slip the start of what we call main works, the main construction, to the back end of this year. That's just about to start. The tremendous advantage we got from that was actually having that much more time to do what we call early works, which is getting all the foundations in place, getting all the underground cooling water systems, getting concrete poured. What you would see in terms of the readiness to start main works construction, most project directors would not recognize because of the advanced nature of the project and our ability to hit the ground running as we go into main works. On the right-hand side, you see as the site will look in the future.
There are three fundamental bases of a competitive advantage that we bring to the table here. The first one, as I referred to, was this ethane advantage. In the ethane that's coming out of Marcellus Utica that you see in the bottom right, otherwise having to find a demand source somewhere else, it could be some export limitations on the East Coast. In the main, it has to go to the U.S. Gulf Coast, demand down on the south coast of the U.S. The second one is if you look at the demand for the product, in other words, polyethylene, and you see in these yellow bubbles where is the demand. The fundamental picture you come away with there is the demand is towards the northeast. You've got two logistic cost advantages here.
One logistic cost related to the feedstock side, the other logistic cost related to the product side. That adds up to a substantial amount of cash cost advantage versus placing a cracker complex to polyethylene on the U.S. Gulf Coast. That's why we like this project so much in terms of the basis for competitive advantage that we have. I can tell you also that the prospective customers like it a lot as well. The structure at the moment is relatively consolidated. They're extremely interested to see a player like Shell with our reputation coming into the market, not just as Shell, but also because of the positioning of where we are, because they can see this having advantages in terms of the time to supply compared with what they have to have at present today.
That offers them potential in terms of things like working capital advantages versus today. Clearly, that's an opportunity for us together to turn that into value potential. What does this look like in terms of how we see the money adding up as we go forward? The left-hand side of this chart in the middle, you've already seen. I've just shown it to you. That's exactly the same. What we're showing on the right is where we see this business moving to as our aspiration into the mid-2020s. The fundamental here is we are aspiring to move this business to an earnings of $3.5 billion-$4 billion, something approaching double. In cash flow terms, that's roughly between $5 billion and $6 billion.
You might note that the base CapEx underneath that, once we get into a steady run and maintain mode, is about $1 billion to $1.5 billion. You can see what the kind of free cash flow potential that you have as a result. Something perhaps I should have said before, which I'd like to reinforce here, is we do assume, as we make this projection going forward, that some of what we experience today is above cycle margins. As we think forward into the future, we do not expect that we sustain all of the margin environment that we have today, because if you look at it on a cycle average basis and you go through different products, clearly, on average, we are somewhat above cycle average at the moment.
As we think about the sort of staircase as to going forward into the future, the first thing we do is we actually assume a reduction as a result of that, then we build our way back in. Where are we in terms of how we see our trajectory towards this? What I would say is about two-thirds of this aspiration is currently baked in. What I mean by baked in is between the improvements that we have in terms of the base business for that $500 million aspiration I referred to earlier, and the already final investment decisions we have taken, and we are then in the construction phase between those buckets. That gets us about two-thirds of the way through to reaching this aspiration. Beyond that, where does the final third come from?
That comes from we have a substantial portfolio of new business development. Some of those examples you would be familiar with of things that we've announced in terms of MOUs or LOIs and so on. We have a range of opportunities in our funnel of new business development. Clearly, for any one of those opportunities, they have to end up making sense. We will bring them hopefully to final investment type decisions. We will only invest where we see fundamental competitive advantage, something which is fundamentally going to add value to the Shell Group. Of course, at the point where the EC and board make those decisions, they will also optimize those decisions as a function of what other opportunities there are across RDS as a total portfolio.
The exact, let's say, steering and fine-tuning of this trajectory beyond the two-thirds to a third is something that the, let's say, we can tune as we go, depending on how we mature this portfolio and how we mature other parts of the RDS portfolio. If you step back from that in total, what is the point we've just talked about? Hopefully, this has been a useful explanation of the whole of sort of how do we make money, what's the basis for our strategy, what's the underpinning for our strategy. The punchline, I think is we see this providing a substantial amount of improvement in value of Royal Dutch Shell going into the mid-2020s. The result of that is adding up to a world-class investment case. With that, why don't I pause?
I'm sure you've got a vast amount of questions scribbled down. Let's give you the opportunity to ask those questions, George.
Thank you very much, John. Thank you very much, Graham. Now it's time for Q&A. A few house rules on the Q&A. We schedule for roughly one hour. Remember, sharp at three o'clock, there will be an alarm here in the room through which we'll be muting the line for a few seconds. First point. The second point is one question is one question. If you could stick to the discipline, that would be really helpful. The other point is, I think, to maximize the value of today's event. If we could keep the questions as much as possible focused on the chemical business. That will also be the best way to leverage the value of today's event. In practice, we'll be alternating question in the room here, but also on the line.
Before we start, operator, could you please remind us and remind people on the line how they should proceed to ask question virtually?
Thank you.
Thank you
It is time we begin the question and answer session. If you have a question over the phone, please press star one on your push button phone. Each participant may ask one question. We'll pause now to allow everyone an opportunity to ask questions.
All right, we start with the room. Please wait for the microphone before asking your question so that people on the phone can follow. Thank you very much.
Thanks. It's Lydia from Barclays here. Could I have one and a half questions? Go ahead. On the safety and operation reliability side, as you identified, that's been quite a, sort of, that has held back profitability. Can you walk through what you've done on that to improve it and sort of the confidence that you have in those processes going forward? Partly related to that, John, within the whole downstream side, how does that work? There's the other part. With that $500 million improvement from the base, is that assuming that you, in terms of there's recovery just from that process side, or does it explicitly do that side of it?
Okay. Shall I start and you add, John? Yeah. Perhaps the first thing is to say is that if you go into the root causes of particularly those three different incidents, the root causes are actually all very different. That points you into a different series of directions. Some very technical, some more operational. Perhaps two things I'd particularly speak to in terms of interventions that we've taken. Some of you might be familiar with, on the personal safety side, one of the interventions we made a number of years ago was what we call life-saving rules. Where what we've fundamentally done is diagnosed how come people end up dying. If you looked back to the root causes, you found a number of fundamental commonalities.
What we put in place was what we call life-saving rules, which were all about helping people to understand, and then obviously training and then putting compliance processes in place so that people did not, let's say, take those actions like, let's say, falling from heights. That's a very simple example. You have to be tied off. In essence, that has been a very successful intervention. The reason people still get hurt today tends not to be because of those life-saving rules, you have to move on to the next generation of it. In effect, we've done exactly the same thing on the process safety side. Which is to step back and look not just at these incidents, but actually a little bit to your point, look back at 10 years of incidents across the whole of our downstream, and what is all of those.
Wind those back to, okay, what are fundamental operating practices that we can tune in and make very simple for people to understand and operationalize? We put that intervention through what we call process safety fundamentals in place, in the last 12 months. The other major intervention is to strengthen, let's say, our depth of expertise bench. Making sure that when we look at any particular process unit that we say, do we actually have sufficient longevity or tenure of people? Of course, you have people coming through and adding new insights from elsewhere, but you need enough strength of longevity, of tenure. Let's say strengthening the system that makes sure that that gets in place and puts it really strongly in place. John, any-
No. Thanks, Graham. I think everything that Graham has said is relevant. I think, Lydia, I'd even take it up to the group level. I think the important message is that our asset integrity process safety management systems are the same across the whole group. I think that is an important message because that's a transition that we've actually made in the last few years. What that offers us, apart from a consistency of how we manage our asset base, it means that we can feed the learnings back across not only upstream, but integrated gas and downstream. We certainly do that, and we are learning from the upstream in the same way that upstream are learning from downstream. I think that is important, and there's a lot of rigor around those global processes.
The other thing as we reflected, and Graham is absolutely right, that these incidents, there was no common theme. Which in some respects is good news, in other respects it's bad news because it means you have to manage them as separate incidents. A thing that it really drove us to do, and actually I personally was the instigator of this process in oil sands, because this was a program that we first trialed in the Scotford Upgrader. That is where you essentially get all your drawings out and say, "Look, I understand all the historical issues, but what is in there that's coming next?" I think we are very good at looking back and learning from it, and we can always be better.
The new skill I think we learned with this process was get all the drawings out, go through the original design and say, what is the unknown unknowns, and how can we be proactive in managing those? That has been exceptionally powerful and successful in the Scotford Upgrader.
We then applied that to Port Arthur, following the expansion of Port Arthur. Of course, that is not in our portfolio now. We then subsequently applied it to Pulau Bukom and Moerdijk. There is the what can we learn from the past, how do we stop these things happening in the future? That has been an important development, which we have now implemented globally.
Please.
Thank you. It's Brendan Moran from BMO Capital Markets. If I can just refer you back to slide 11, I'm glad you're saying we're going to be in demand for petrochemicals, can you sort of just step through some of the threats to the demand growth that you talk about? Can you talk about your exposure to say, you say 20% of demand is packaging, just in terms of threats of either substitution, recycling, I know it's thematic at the moment, countries banning, say, plastic bags.
Yep.
Is this real? Does it impact your business? I guess just lastly, in terms of other areas where we're seeing either substitution or demand destruction, please.
Yep. Clearly, I think what you refer to is what we might more broadly call the circular economy, and clearly, obviously, you've got a phenomena related to climate change. In general, climate change offers growth opportunities in petrochemicals. The reason is things like, as I showed on the chart afterwards, one of the clear advantages in terms of reducing vehicle efficiency or improving vehicle efficiency is taking weight out. The easiest way to take weight out is to put plastics in replacement of glass and steel, because that's how you get a substantial reduction in the weight, because about 50% of what you use as energy in an existing car is actually moving the mass of the vehicle and the single driver around. Similarly, if you look at what is at the front end of the abatement curve for CO2 is insulation in buildings.
If you want to do insulation in buildings, you go straight to petrochemical solutions. You've got all those underpinning drivers. You're absolutely right. Of course, the circular economy is asking questions around recycling, strengthening of recycling, maybe not using plastic bags and so on. I mean, all I would say is, and you see underneath this, it says IHS and Shell Analysis. We've also worked with people like McKinsey in this analysis. I would say that we will never be completely right. All of that is factored in terms of what you are seeing as a projection going forward. Clearly, we can all create scenarios in terms of assumptions for a faster trajectory of recycling or a slower trajectory of recycling, you can change the numbers a bit.
I don't think you fundamentally change the conclusion, particularly when you looked at the next slide in terms of the different range of products, because yes, you can play that game particularly sort of maybe in the sort of bottle side or something like that. You've also got to also ask yourself about alternatives. First of all, if you look at bottles, what you might not realize is a typical plastic-based bottle has actually usually got something like seven different layers. It's got seven different, or not quite seven, but it's got different products. Actually, your ability to recycle that back to where you started is fundamentally undermined.
If you make it from a single product, the quality just isn't there in terms of the aspirations that you've got in terms of things like barriers for oxygen, barriers for water, depending on what the product is. There are some real impediments in terms of your ability to recycle, let's say, from the product that you're starting with and you want back again. What you can do is you can recycle it back to, let's say, something that might get used in kids' playgrounds. Clearly, that kind of recycling you can do, and clearly, let's say the recycling in terms of back into the energy system is quite possible.
Recycling from a kind of virgin-based product and product performance back to the same is incredibly difficult because of complexities like that, as well as the fact that as you go through it, you actually lose a lot of performance. The other issue that you end up needing to take into account, which is an interesting sort of philosophical one for governments and society to end up grappling with, which I don't think is completely realized yet, is potentially the clash between how we think about climate change and how we think about circularity. If you think about climate change, you would say, 'Well, let's take the example I just used, which is putting more polymers into cars to reduce weight, to reduce CO2.' Climate change would drive you to use much more polymers.
If you put a circularity hat on, well, you'd say, "No, I like glass, because glass I can use, and then I can melt down, and I can have another glass." What's the clash between those points is both the CO2 output of the use. It's also the CO2 consumption of the recycling process. Actually, you use an awful lot of energy to end up recycling glass. Society's going to have to end up grappling with that whole system, and let's say the sort of intellectual infrastructures for working all that out is going to be very complex to actually think through and put into place. For now, I think the punchline answer to your question is, as best we can, we, but also obviously tapping into other minds outside in terms of external thinkers, have factored it in as best we can.
The only thing I would add from a downstream point of view, and it may be just a small anecdote, but it's a useful anecdote, is we've discussed this between our chemicals and retail business. One of the major pieces of feedback we got from our 30 million customers who come through our 43,000 sites every day is, "We want you, Shell, to participate more in the recycling and collection of waste," which we will do. By the way, you can now, within our retail sites, get these little red thermos flasks, which are 100% recycled. We are thinking about it in our own company as to how we can facilitate this challenge that Graham has described.
Where should we go?
We just have one in the room, and then we also maybe go onto the line.
Oh, there's one. Yeah.
Thanks. Gordon Gray from HSBC. You were talking about product development as being, give or take, a third of your ambition to grow the business. Can you give us a feel just for a few of the areas that you get most excited about in terms of new technology, new products that you think could be real breakthroughs, real game changers in the next few years?
Yeah. I think if you go back to what I was showing in terms of Solar Impulse and so on, I think you've got the sort of basis for that, which is, let's say, enabling things like solar to be really effective. I see the petrochemicals being a fundamental foundation of that. We are not really at the ultimate tailoring of products space. As I talked about in terms of our positioning, we're really in the front end of this. What I get a little bit more excited about, which is perhaps not what you so much wanted to hear, is more about the process chemistry, which is to come out with new process chemistry, which gives us a fundamental advantage in terms of coming into what ultimately will be an existing intermediate, which is like diphenyl carbonate, for example.
I get very excited about the fact that we've got blow-away technology. Obviously, it took us like a decade to bring it to fruition. To me, that's very exciting. That enables us potentially to move towards the polycarbonate value chain. The polycarbonate value chain is a very interesting one. Polycarbonate's behind things like LED lights. As, again, we move down the climate change, tackling a lot of demand for things like LED lights, polycarbonate in replacement of glass in automotives. That's a very interesting value chain to look to get access to. That's the kind of place I'm really excited about. Maybe we go online with a question.
We're going to the phone on the line. As a reminder, please press star one if you wish to ask a question.
We're back in the room. Graham, do you want to take another question?
Thank you. James Tolbert, Citi. I think you've indicated that ethane unit, large capacity, is at the bottom of the cost curve. Can you tell us about the top of the cost curve? I presume that's where the price is set. What is the price differential between that and that Pennsylvania site? Sorry, cost differential between that and Pennsylvania. How is the top of the cost curve changing? How quickly is that being displaced by new ethane units coming through?
Yeah, great question. Fundamentally, yes, you're right. Obviously, when you look at, let's say, ethane based, obviously you do need to look at where is the feedstock coming from. If you go to, let's say, the very original Middle East investments, they were based off a gas price of something like $0.75 per million BTU. That was only in place probably through the 1980s. You can see the likes of the Saudi government then progressively actually moving that price up as they have less surplus gas available. Obviously, that's tended to be the bottom of the cost curve. Nevertheless, as you look at current gas prices in North America and probably what you might expect them to be in the future, clearly North America has got a cheap gas price.
I referred to if you took $3.5 per million BTU, you're at something like $23 a barrel equivalent if you work in, let's say, oil price terms. Obviously then, if you jump to then using a naphtha-derived feedstock, you're based off whatever the crude price is, plus or minus with a spread from naphtha to crude. Within that, you would see, generally speaking, naphtha-derived products being the top end of the cost curve. Which within that? Well, it's the small, old pieces of kit. Where do you see substantial restructuring? Well, places like Japan. A lot of restructuring you've seen in the last several years. Consolidation of the Japanese chemical industry, which has been seeking to deal with not only excess capacity, but also excess uncompetitive capacity. The same is true to an extent in Europe.
In fact, we ourselves took out a small cracker that we had in Germany, in Rheinland. We had two small crackers, one with advantage feedstock, one without advantage feedstock. The one without advantage feedstock we shut down at the end of 2011, and the reason was because we saw it fundamentally being at or moving to the top end of the cost curve and not really an economic basis to move it into a much more competitive basis. In Europe has
let's say, a number of small assets which are at the top end of the cost curve. To an extent, the fact that Europe has been waiting for a wave of imports, to an extent, for years coming from the Middle East, but now it's waiting for imports coming from North America as the North American investments come on stream, is also looking at potential. Inevitably, people are looking at where is what going to have to get shut down in Europe as a result of that, and people can speculate on that. Clearly, there's top end of the cost curve there. Interestingly enough, if you look at something like polyethylene in North America, as we were working through the maturation of our Pennsylvania Chemicals project, one of the realizations we made was a substantial amount of existing footprint in polyethylene in North America is actually pretty small scale.
It's old. There's quite a lot, even around 100,000 ton per annum level in terms of polyethylene facilities, but a lot of stuff between 100,000 and 200,000 ton, and we're putting 550, 600-plus thousand ton units. These have a very substantial difference in actually in the fixed cost base, less in the variable cost, in the fixed cost base of something like $100 a ton advantage. You have a mix in terms of the scale effect of world scale as well as the variable cost element coming to, let's say, the total cash cost. In summary, in general, you will see marginal assets in Europe and old Asian assets, perhaps particularly in the Northeast Asia area, tending to be the ones that get shut down. The only twist on that, I'd say, is that, of course, you do have the odd other technology in place.
The investment in the amount of ethane cracking has tended to add a lot to the C2 chain, but then mean there's less propylene. One of the things that the world is tending to do is to add propane dehydrogenation units to meet that gap in terms of propylene capability. Now, depending on what your assumptions around LPG pricing are at any one time versus naphtha, those propane dehydrogenation units could also become the marginal asset as well. Does that answer your question?
Yes, thank you.
Great. Please.
Edward Firth, Exane BNP Paribas. Yeah, just wanted to ask a question in terms of your projects and ultimately the returns you're targeting there, specifically in relation to your three major projects. Additionally, perhaps how they vary between the greenfield investments and the brownfield, and then perhaps how this affects your chemicals division return on capital.
Yeah. I'll answer at a portfolio level. Our portfolio aspiration is to sustain a return on capital of ±15% through a cycle, we will look for the portfolio of investments that we make to achieve that objective. We won't speak specifically to the returns of individual projects, but what I can say, I think, which is fairly obvious, is there is a hierarchy between, let's say, a very cheap amount of creep in an existing asset to, let's say, a full-scale debottleneck of that asset through to a new unit on an existing site through to a greenfield. Fundamentally, what's taking place as you move through that spectrum is the amount of existing kit and investment that you are leveraging is high at the front end of that.
As you move to the right end of what I just described, you are having to put more and more of that underlying infrastructure in place. You're having to put things like utility systems, the base concrete foundations, the underlying cooling water, electricity, all this underlying infrastructure on a greenfield, you're having to put all those in place from new. When you start with just a little bit of creep in existing asset, you don't have to do pretty much any of that. Inevitably, what you will then see is the greenfield. You will see a relatively lower return, and you'll see the very highest return in terms of the simple creep. That's why when we went to the $500 million, I realize, Lydia, I didn't answer your question now.
The $500 million, one of the things that we prioritize most of all is very simple, very easy, cheap creep of your existing kit, because that's a real boost to your return.
That's essentially what you see across all manufacturing facilities, whether it's refinery or chemical plant, you'll always get your marginally higher return by creeping the existing assets. Of course, there's a limit to the cash flow then you can generate from debottlenecking.
Perhaps just a very quick one to finish off the answer to Lydia's second question, which is to say in the $500 million that we aspire to into improvement of the base, that excludes any improvement from big hits. Please. Oh, sorry. It's flowing with the mic.
Hi, Michele Della Vigna from Goldman Sachs. I wanted to come back, if I can, to your slide 18, which shows the margins of naphtha-based petrochemical versus ethane. You make the point that there's been a disparity in the last seven years, which is fair, but it looks like in the last couple of years, there's been a reconvergence of margins with the naphtha-based improving and the ethane-based falling. I was just wondering, given that there's such a disparity in the cost of feedstock, why has that happened? Do you expect it to continue?
I actually think it's quite temperate, to be honest. I think it just happens to be there as a snapshot of where we happen to cut this off. I don't see this as, let's say, a fundamental structural issue, and I think there's enough volatility going on in the different regions that will cause this to, let's say, not to equilibrate going forward. I'll just give you a simple example of that. Obviously, in North America, we've got quite a lot of capacity coming on stream. Each of Dow, CP Chem, and ExxonMobil are bringing on ethylene to polyethylene cracker complexes in, I don't know, plus minus the next six months or so. Inevitably, that has to have some form of depression effect.
Certainly, one would expect in the polyethylene margin, which have actually been spectacularly high and have propped up the profitability of, I think, those companies very effectively in the recent two to three years. I think you would logically have to expect some form of depression on those. How that flows between ethylene and polyethylene, that's a more complex dynamic. That then you've got to get into the nuances of where the level of net investment is between the cracker side and the derivative side, and is there a surplus or deficit between the cracker investment side and the derivative side. That will affect where the balance in terms of that margin has gone. In the last two or three years, the amount of margin that has been captured between ethylene and polyethylene in North America has been huge.
In that period, the amount of margin between ethane and ethylene has been relatively small, and that's part of what you see in terms of this significant decline. Exactly how that plays out going forward, I think, is a little bit unclear. I actually read something that I think WoodMac put out, which was expecting maybe a little bit of balancing in that space, which may see this move to a different place. I think you have to look at what the third parties are saying there. Europe and Asia have tended to be strong. Europe, I think, always the question mark for Europe is, will it see a flood of imports coming from North America? That's the big question in Europe. Europe, otherwise, has actually tended to be quite stable. Not a lot of changes in the fundamentals in Europe.
I think that some of the uncertainty is really around where this surplus capacity from North America is going to go. I think one of the question marks that might be there is some of the capacity coming on stream is actually held by players who have substantial positions in Europe. Whether they're going to move that product into their own markets or not, I don't know. I think that will certainly influence how Europe changes. Asia's got a reasonable amount of capacity coming. I think in general, probably some softening might take place. The picture I'm drawing is a reasonable amount of uncertainty, but I don't see structurally here why these are going to sit in the same place. I think there's inherent volatility, which I attempted to describe in terms of all of these competitive, both feedstocks and also value chains.
The fact that these keep moving in different places, I think, keeps uncertainty. It may sit close together, but I think I expect volatility going forward.
Thijs Berkelder of ABN AMRO. Last week, we saw a new Dutch government doing two key announcements on implementing CO2, let's say, tax, carbon tax, bringing back like the U.K., I would say. Secondly, allowing CCS projects. In what sense is that threatening your European operations and the competitive feedstock situation for your European facilities? Secondly, on your, let's say, base CapEx scenario for the mid-'20s, does it include CCS type of projects already, or is that excluded from that base number?
Okay. Maybe I'll start, Graham, and then you can talk. I mean, of course, CO2 pricing is not just a European phenomenon. By the way, I was with the Dutch government last night, so I get quite a little bit of insight on their thinking. Of course, a lot of this legislation still has to be passed. We should be very well aware of it, and we should be in a position where we can respond to it. As I say, this is not a European phenomenon because you also see the Singaporean government proposing a $20 a ton CO2 tax in 2019. We do need to be ready for that. As we look across our portfolio, there are still a number of questions, of course. Is the degree to which some of those taxes ultimately get passed through to the customer, yes or no?
You can have your own view on that. I think increasingly as well, we are looking as we build new facilities, to what degree can we make them CCS ready? Also, as we build new facilities, and the Pennsylvania cracker is just one of those, we are building new facilities at top quartile CO2 performance.
We would not authorize a new FID without it being the best it could be in terms of competitive CO2 production, because at the end of the day, the assets that will be at risk are those which are less competitive in terms of their CO2 emissions. Maybe you can add a little bit more detail, Graham.
Well, I was just going to go somewhere else first, then I'll add some detail, which is philosophically, of course, as a company, we do believe that the world needs to tackle climate change. In tackling climate change, then somebody has to get the ball started. Getting the ball started does need some regulatory intervention, and therefore some governments, like the European governments and the Singaporean governments, it is actually a good thing that they are moving this forward. The challenge for an industry like chemicals is if the result of that intervention just ends up with, for example, a shutdown of supply in those regions, and actually that supply then gets met by somebody who's investing in lower energy efficiency, actually, the world has achieved nothing.
As part of our advocacy, we are actually naturally supporters, as you know, both in terms of carbon pricing, but also of support for CCS as we see necessary, but also that there is appropriate support given to an industry like the petrochemical industry, which does not end up with it just being shut down to solve this problem, and then the same supply being provided from somewhere else with a poorer carbon footprint. Inevitably, there will be puts and takes as this works through, and we'll see how it pans out. As John said, in terms of the investments that we make, we aspire for all of our new investments to be first quartile in terms of energy efficiency, and we have an eye on then the whole idea of being CCS ready.
Does it force you to switch from naphtha to ethane in Europe or do you have the same as the U.S.?
The question was, does this force you to move from naphtha to ethane in Europe? I think the economics of use of ethane in Europe is complex. Obviously, there are a number of examples of which our existing footprint with ExxonMobil is one which is in Mossmorran in Scotland, which is based off ethane cracking, but from the North Sea. That goes back to where there was surplus wet gas in the North Sea, an investment made in the 1980s. Clearly, where you have crackers, particularly those crackers which are probably coastal crackers and had an existing ethane capability, supplementing that with ethane from North America may well make sense. The degree to which you then move to, let's say, a naphtha cracker based on the coast converting to ethane, the problem is you've now got to start investing in quite a lot of new things.
You've clearly got to invest in some way. There's got to be the export capability and ships and so on. You also need the import facility. You need the ethane storage regasification, and then you've got to convert your naphtha cracker in some way. The back end of the naphtha cracker is not optimized for ethane cracking and neither are the furnaces. You need to make a substantial number of changes to that piece of kit as well. That might make sense. It's probably starting to be rather hard to work at this point. If you move inland from there, now you've got to have some infrastructure to move the ethane around as well, and your cost is just going through the roof. The punchline in my mind is, does this make sense for some people? The answer is yes.
Is there a limit to it? I think the answer is also yes, because you get to a point where the amount of capital you've got to throw at the problem doesn't make sense anymore relative to the differential you've got in place.
Base CapEx?
Sorry, I don't understand that. The question is the base CapEx. I'm not sure, the base CapEx of what?
Does it include CCS technologies?
In terms of the scope of what we've got? The existing projects that we've got right now do not include the scope of CCS right now. Yeah.
I think the other thing I would add, and this is the same in the Singapore agreement, is I'd advise you to go and read the small print in the Netherlands coalition agreement because there are different recommendations regarding power and industry. Read the small print. I think the important thing, as Graham quite rightly stated, putting a market price on carbon is something which the Shell Group supports and has been advocating. Yeah.
Hi. Biraj at RBC. I have a question on slide 22. You show your returns profile versus a few peers, one of the things that jumps out is that the volatility relative to some of your peers is quite high over the last few years. The question is that just relating to the big hits over the last couple of years, or is there something systemic in your portfolio that would make you more volatile than your peers going forward?
Perhaps two answers to that. I think the primary basis for volatility is indeed the big hits. Yep. You would see this tracking more stably across this period. Actually, as we saw in the NIBAT profile, more with an upward trajectory going from 2013, 2014 through to 2016, 2017.
I'll put a however in there, which is that the however is we tend to obviously look at from a high level, you look at petrochemical companies perhaps as being alike. If you actually get under the hood, their portfolio is very different. The product portfolios that they've got are substantially different. Clearly, if you look at ourselves versus ExxonMobil is very large in polyethylene. It's also very large in terms of its footprint in North America. It has a substantial weighting to, let's say, the polyethylene to ethane differential in North America. We do not have that exposure today. We obviously are moving towards that exposure in the future. We have a whole range of different exposures and a different set of product slates. We've got a little bit more diversity in terms of the different value chains that we go into, and less concentration.
Fundamentally, our geographic footprint and our product footprint is slightly different. That inevitably will end up with, let's say, some volatility in the differentials. Some of that volatility may come from them, some of the volatility might come from us, and it depends what the future turns out. Do we have any questions online yet? No. Okay. We'll keep in the room then.
Thank you. It's Irene Himona, Societe Generale. I have a question on your aspirations on slide 28. Again, I'm not sure what base CapEx is, but assuming this is total CapEx, you're looking to generate $4.5 billion or so of free cash flow. Trying to put it in the context of the group, obviously, we don't have group targets for 2025, but in terms of the current targets, you appear to be showing that chemicals will be absorbing roughly 5% of group CapEx. It will be generating free cash flow covering a quarter of your current dividends or 20% of your current 2020 group free cash flow targets. This is an extremely ambitious plan, I would suggest.
My question really is, I appreciate one-third of that aspiration is new products, but in your mind, what are the key two or three, let's say, risks to that aspiration, which appears very ambitious?
Irene, I will let Graham talk about the risks, maybe ahead of that, I think we were very clear in June 2016 on Capital Markets Day around our group's free cash flow aspirations and returns. We talked about by 2020, the $20 billion-$25 billion of free cash flow and double-digit returns. We were, I think, very transparent in segmenting those free cash flows by strategic theme. I probably wouldn't go too far in terms of extrapolating beyond 2020. From a group perspective, when we're ready to share that information, we will certainly do that. I think it's important to not speculate about how we'll split that going forward, because we clearly have options here as well. No, you're absolutely right that this is a strong growth. I would reiterate something that Graham said earlier.
We're talking of moving the NIBAT from a $2 billion level to a $3.5 billion-$4 billion and a cash flow from operations from $3 billion to $6 billion. The important thing that Graham said is two-thirds of that is essentially related to final investment decisions that have already been taken. A base business. I think you need to understand the percentage of that that has a very high certainty. I can tell you that we have a rather large funnel of options that we have available to us, and those options go beyond the financials that we've just described here. We have already risked, in putting forward these numbers and tabling these numbers, the fact that some of these options will proceed and some of them will not proceed. You can add to that maybe, Graham.
Yep. Perhaps something to make really razor sharp as part of this. The economics on the left, and particularly let's call it the base CapEx level, assumes, let's say, we move towards a more cash generation strategy in the mid-2020s. We've got a growth and investment phase, and then we move it to a sort of cash generation phase. The assumptions that you're making around free cash flow generation have embedded in those assumptions that we make that shift rather than, for example, continue to invest. That obviously would be a different choice, and we're not making that choice now. That will be a choice that, let's say, the leadership will be making in the early 2020s. That's just to make sure we kind of clarify how we make assumptions around this left-hand box. Clearly, let's just be sort of really transparent in this.
Clearly, our own performance is incredibly important. It's incredibly important in terms of successful delivery of projects, successfully bringing them on stream safely, reliably, et cetera, operating them safely and reliably going forward. That's an underpinning all the time. That's why we reinforced it in terms of our base business, but it continues to be true also in terms of your growth trajectory. Clearly the, let's call it the hydrocarbon environment and the different environment in the different product chains that you've got. It's a fact of life, this is always going to be uncertain. That's frankly, deliberately why we produced a little bit of vagueness on the right-hand side of that, which is to reflect exactly this phenomena. I referred earlier to the fact that we are well aware that on average, we've, in the past 12 months, say, experienced some above cycle margins.
Clearly, we assume that comes out as we work our way forward. There's a certain assumption that gets built into whatever we have, where we happen to be in the mid-2020s in terms of that cycle position. Who knows at this point, and it may be actually well above this, or it may be somewhat below it. Obviously, there's a certain level of volatility in terms of petrochemical cycles, which you've seen in those previous charts. You can work out from then how they can influence some of the cash flows, the degree to which, let's say, some of those perturbations are symmetric or asymmetric.
In general, because of the different competition between the feedstocks and the value chains, it's tended to create a certain amount of asymmetry in those cycles, which is very different, that is what has created some of the stability you've seen in cash flow here. That's rather different from if you looked at, let's say, the cash flow of a petrochemical business in the 1980s, this would look very different. You would have, let's say, a rather depressing phase, then you have a boom bust type of phenomena. You get that to a certain extent within individual product chains, but you're now not getting it within the whole, if you have a sufficiently diverse portfolio. I would say both the level of oil gas differential and probably the volatility in that oil gas differential will have an influence.
I think the more there is volatility in that oil gas differential, it will probably tend to undermine the investment clarity by which people will make investment cases. If I take an example, if you had a period of low oil price, it will put the brakes on people investing in North America. It will probably tend to encourage people to invest in China. If you have the reverse, you will tend to put the brakes on people investing in China, and you will accelerate investment in North America. You can see this, if you look at something like coal to chemicals in China. If you go back two, three years ago, a lot of investment in coal to chemicals, where coal was basically cheap relative to crude, and so looked really strong and attractive from a Chinese perspective.
Some of that is coming on stream, actually the rate of new investment to that is slowed down dramatically, the reason is because actually at current crude type of prices, it's not really competitive. I think volatility in the hydrocarbon environment, in the feedstock hydro environment is inherently good because it puts a number of brakes on the investment clarity. A long period of stability at a certain level would probably be relatively less good in my mind.
Hi, it's T-Pan from XAN. I just wanted to follow up. It seems to me the Pennsylvania cracker is one of these big-ticket items in the Shell portfolio, you did mention affordability pushing back the investment a little bit. I want to understand why retain such a high stake. Just sort of for a classic oil and gas analyst, how do we go about, in your view, modeling Pennsylvania and particularly in terms of the contribution to cash flow? I know you've got this sort of bridge to 2025, the market may be a bit more short term than that. I was just wondering when should we forecast sort of start up a contribution to cash flow? Thank you.
That was very well posed, wasn't it?
You're absolutely right. This is a big-ticket item in the Shell Group. We had numerous discussions, and Graham and I remember those, with both the Executive Committee and the board, in ensuring that this really did rank at the group level. I think as well, the other point I would add is that this project has been a long time in the development. Whilst we may have only brought it to your attention more recently, the number of years that we have been looking at this, the number of years we've been getting to a minimum technical scope, the number of years we've been negotiating the gas contracts to make sure that they're competitive and cover all the scenarios that you can imagine, that has taken quite some time.
I would even argue that whilst Graham has been absolutely transparent around the fact that for group affordability reasons, we chose to delay the entry into the major works and focus on the early works. This is the best project we have seen in terms of readiness to go into early work. I think we are very well, sorry, into main works. We are very well positioned for a project that will be delivered on time, on budget, and the likes. Your question around why do we want 100% participation, which is how I read it. If we have a project that is fundamentally advantaged Has very attractive returns. I would ask why would we want to share that with a partner?
Perhaps an added twist on top of that is that what is a contrast in the downstream versus the upstream is when you dilute, in effect, you create a competitor, for antitrust reasons. You create a joint venture, and that joint venture is a separate legal entity, and at a formal legal level, that legal entity is now competing with any base business you have. That is not a good place to go strategically, because you've now set up a business which is now competing with potentially what you're trying to do for yourself. That's really not a nice place to go strategically, and you then become somewhat out of control. Obviously legally, there are limits to what influence you can provide because you have to keep appropriate levels of separation.
That's a very different phenomena than the one you experience in upstream, where actually that whole phenomena doesn't, because it's crude, which disappears at a flange. You don't have this whole issue of actually moving into the market and your products competing with another investment. Those are the basis. We said, "Well, do we want to create a competitor whereby all the hard work we've gone to de-risk this project, we actually give half of it away?" Well, really that doesn't feel a great idea. We'd much prefer to keep this 100%. We think it's a good thing. We need to manage the affordability, but we keep it strategically under our control.
Timing on?
Early 2020s. We are just moving into the main works right now. Many of you can work out how long major projects like this take place, and you can figure things out for yourselves.
Hi, I'm Alex Tassone with Citi. Just as a follow-up to that, do I understand that you use a different discount rate in the downstream, or specifically chemicals on the major project than you use in the upstream? Is that the right way to think about it?
I don't think that is the right way to think about it, actually. No.
it's the same hurdle rate across the organization for all major-
I mean-
capital projects?
I'm not going to go into particular numbers. I think the important thing is that you're absolutely bang on that all of these projects we rank and discuss at the Executive Committee. We have a monthly process by which we look at the potential projects and rank and risk them. We have a consistency of financial parameters. Having said that, though, of course, the risks to upstream and downstream projects are quite different. It's multidimensional when you make a decision. It is absolutely multidimensional. I think you need to recognize that there are many factors we take into consideration when we do look at these projects. What is the longevity of the production? What is the risk profile in the country? Et cetera. The risk profile is very different.
I think it's fair to say, John, isn't it, that in terms of the way the group in total looks across the portfolio, we've added a lot of sophistication to-
Yes
the way that we look at robustness-
Absolutely
of projects and their different types of risk profiles as a way of then plotting them against each other. Obviously, one element is pure expectation of profitability. Another one is then looking at what are the inherent risk profiles around that.
Absolutely, yeah.
As you hint, one way of tackling this is obviously fiddling with discount rates, another one is to get down to the fundamentals of the underlying risk profiles and look at that.
That's where we go. I think we're pretty sophisticated in the way we assess our projects now.
Can I just ask a follow-up? Would it be fair to say that with the fall in oil price, that you're more agnostic to downstream or chemicals major investment than you maybe were at $100 oil? Is there a greater % of downstream-related projects that are coming up the queue?
You can see that for yourself in the data we shared with you on Capital Markets Day in 2016, where I think we were quite clear about how much we are spending in each strategic theme. In terms of our spend and our profile, that's very clear up to 2020. I think you can put things into maybe three categories. You can put things into, is this an oil category? Is this a gas category? Or is it a GDP-related category? Clearly, much of what we are doing in the downstream, not all of it, but much of what we're doing in the downstream chemicals is a GDP-related category. That doesn't mean to say that we will not continue to invest significant amounts in terms of oil exploration and oil projects.
You may argue that one of the biggest downsides that any integrated oil company has as well is under-investing in oil, and you see an up spike in the oil price. Again, I don't think it's quite as generic and simple as you indicated. What I can tell you is that the downstream, and that includes chemicals, marketing, refining, and trading, has the capital that we require to grow the business and to maintain the cash and returns that we're looking for. It is not starved of capital, but also we're not throwing capital at this business either. As we go forward of the downstream portfolio, chemicals is the most capital-intensive part of our business. We are not rushing out building another refinery soon. Trading is essentially a working capital business, and marketing tends to have a relatively low capital intensity.
This chemicals business is the highest capital intensity within downstream. We currently have access to the capital we need to deliver on this profile that Graham has shared today.
I'm not sure if I'm breaching George's one question only. I wanted to bring back, you showed your footprint on slide 19, and you referred to world-class a number of times, Graham. Just which of your assets would you describe as not world-class, that are sort of still work in progress? If you can just step through a couple of your key sites where you need to see further improvement.
Yeah. I'm not going to go into too much detail here, the one place I would speak to in terms of continued attention that we put in is in terms of our US Gulf Coast footprint. The crackers that we've got in place there, we've basically got 4 crackers. Three of those were originally designed to use heavy liquid feeds. Clearly, around the shift in the gas oil differential in North America with shale gas, we needed to convert that to be able to use gas feeds, also refinery gas streams. I think we did that extremely effectively. Clearly the kit that you've got on the ground is not what it was actually designed to do.
That brings with it some optimizations in terms of energy as efficiency, to a certain extent, footprint, because if you look at crackers, you basically have what you call the front end and the back end of a cracker. The front end kind of basically gets you to the basic products, and the back end separates them out. Clearly, with an ethane cracker, the main product you're pulling out is ethylene, therefore, the amount of separation you need at the back end of the ethane cracker is relatively simple. With a naphtha cracker, you're actually producing a much longer chain of products from ethylene, propylene, butadiene through to a benzene and beyond benzene. Therefore, the size of the back end on a liquids cracker is substantially greater.
When you go from a liquids cracker to an ethane cracker or convert it there, you end up with a whole bunch of surplus capital. That surplus capital is still sitting there, in some ways probably using some level of utilities and so on, which is inherently inefficient. That footprint is clearly not as efficient as we would like it to be. We've looked at it very closely. We do not see it makes sense to, let's say, put a bullet in it and put in brand new capital replacement. It doesn't make economic sense.
Clearly what does make economic sense is sort of ducking and weaving, if I can call it that, around, let's say, is there a part of the back end or is there a part of some of the subsequent downstream conversion infrastructure that we've got in place, which is not really capitally efficient? Maybe we need to prune in some way, either operationally or beyond that. I'd say that's the main area that we have as a focus of what's not quite in the right place, that we would like it to be in the right place. Now, I'd also say philosophically, we always believe in continuous improvement. We always look at the benchmarks, and we're always looking for how do we get to Moerdijk to the next level? We've got a creep project which we're working on right now in Moerdijk.
We've just done a creep project in Singapore. There's always that activity going on and continuous improvement. I'd say if I point you to anywhere, that's the place I point to in terms of the one we go, how do we get that to keep moving that to a better place? Here, I think.
Lydia.
To Lydia.
Here.
I don't want to cut us time, very quickly, just when you talked about the cash flow aspirations, you did say that they included lower margin assumptions than we have at the moment. Are you able to quantify how much you've taken off that? Because I think one of the difficulties we have modeling it is that sensitivity to the margins.
I think I gave you at least a clue, when I talked about, let's say, the underlying business being around about the $2 billion mark. If you can kind of look a little bit between where we are today and the $2 billion mark, you have a certain clue.
Just before the alarms. Thijs Berkelder , ABN AMRO. Slide 21, I'm missing the implant downtime, the plant downtime 17, Hurricane Harvey.
Yeah. We didn't choose to put the data on.
That's third quarter data, by the way.
Oh, thank you.
Put it this way, Deer Park was not designed to operate underwater.
It didn't operate underwater.
It didn't operate underwater.
Yeah. Which is in the public domain.
Yes. Oh, here we go.
Attention, please. Attention, please. The fire alarm is about to be tested. Please take no action. Attention, please. Attention, please. The fire alarm is about to be tested. Please take no action.
Doesn't work. Now it is taking action, that's right.
Attention, please. Attention, please. We are investigating an alarm condition. It may be necessary to evacuate the building. Please await further announcements. Attention, please. Attention, please. We are investigating an alarm condition. It may be necessary to evacuate the building. Please await further announcements.
We don't know how long this goes on for, do we?
A few seconds, I would have thought.
Seconds.
Okay.
Attention, please. Attention, please. The fire alarm test is now complete. Please respond to all future fire alarms.
Very good.
Okay. Let's give me a little bit more time. Maybe not.
Attention, please. Attention, please. The fire alarm test is now complete. Please respond to all future fire alarms.
They said seconds, they didn't say how many seconds. Great.
Okay. Any other questions? Do we have any more questions online? No. Okay. Do we have anybody online? Yes. Yes, we do. Thank you. I would say that if you don't have any more questions at this point, we will move in now to mingle. Maybe before we close, I'd just like to leave you with a few parting comments. At the Executive Committee level, I think we've spent a lot of time articulating the importance of chemicals as a growth theme. Certainly, it was our intention today to give you a deep dive into our portfolio and give you new information and be open and transparent about our portfolio, not only today, but the portfolio going forward and the options we have.
Graham, I'd like to personally thank you for being very open on your responses. In terms of the key messages, we have a very competitive and strong portfolio today. I think you've seen that in terms of the NIBAT that we're delivering today. Take it from me that our number one priority remains ensuring that the base portfolio is run safely, it's run reliably, and it generates the cash, which it is clearly capable of delivering, as you've seen. The second priority is very much to deliver on the projects that are in execution and will come online between now and the early part of 2020. The Pennsylvania cracker, the alpha olefins unit, and the doubling of our capacity in the Guangdong province of China.
Those are very important to us. Graham and I are absolutely focused on delivering those in line with what we shared with you today. Indeed now, I think after today, you've seen our capability to grow our NIBAT from the nominal $2 billion per annum up to $3.5 billion to $4 billion in the period that we've indicated, also to move our cash flow from operations from the $3 billion to potentially up to the $6 billion level. I think what you've also heard, though, is that a significant proportion, two-thirds of that, is related to those three projects, and it is related to our base business. I think that is also an important message to take away. We have significant options ahead of us to further develop the portfolio.
I think as you try to calculate yourself, this business, as we go into the early 2020s, will generate an awful lot of cash. That gives us a nice choice to make at the group level around, do we continue to invest in the portfolio and use at least part of that cash to reinvest into the chemicals business, or do we make that choice to take the cash and invest in other things? I think the important thing is that it is a very nice choice that we have. We have a strong portfolio of options ahead of us, should we choose to invest some of that cash to continue on a growth trajectory. That's an option. That's an option that we're very well aware of as an Executive Committee.
We will come back to you as we consider those options going forward. Thanks once again for your questions. I look forward to maybe further questions in a group or one-on-one basis as we mingle for the next hour. Thank you.
Thank you.
Graham, John, thank you very much. Thank you very much all for joining today and your flow of questions. The next opportunity for engagement will be on the 2nd of November for third quarter results. We look forward to seeing you and to hearing you then. Thank you very much.