BP PLC Energy Outlook 2035 Presentation

Feb 10, 2016 AM EST
BP PLC Energy Outlook 2035 Presentation
Feb 10, 2016 / 03:00PM GMT 

Corporate Participants
   *  Bob Dudley
      BP PLC - Group Chief Executive
   *  Spencer Dale
      BP PLC - Group Chief Economist

Conference Call Participants
   *  Daly Sporten Kroosh
      MFC - Analyst
   *  Jason Gammel
      Jefferies LLC - Analyst
   *  Hugh Lee
      Ebico - Analyst
   *  Chris Weldon-Smith
      Oxford - Analyst
   *  Paul Newman
      ICAP Energy - Analyst
   *  Megan Darby
      Climate Home - Analyst
   *  Bruce Duguid
      Hermes Investment Management - Analyst
   *  James Skinner
      New Economic Foundation - Trustee Emeritus
   *  Amcorta Dehaat
      Centre for European Reform - Analyst
   *  Norman Sully Spencer
      - Analyst
   *  Neanda Salvaterra
      Wall Street Journal - Media
   *  John Feddersen
      Aurora Energy Research - Analyst
   *  Ari Ucting
      Bloomberg News - Media

 Bob Dudley,  BP PLC - Group Chief Executive   [1]
 (Video playing). See, even energy can be fun. (Laughter)

 Hello, everyone, and welcome to BP, and thank you all for your time. Welcome to the launch of the latest edition of the Energy Outlook 2035, and a very warm welcome to everyone joining us online from around the world. We had a figure earlier today of 5,000 people had signed up for the webcast today, which is 40% more than last year. They probably want you, Spencer, to tell them what the price of oil will be. (Laughter)

 We are a global business, and this is very much a global publication. We couldn't produce it without help from people in so many countries around the world, and we share our forecasts in return. So, thank you, everyone, who has helped create these projections all around the world.

 But I guess you might be asking: Why are we looking at energy in 2035 when we have so many challenges right now in 2016? We are in the middle of a prolonged and very severe downturn in oil and gas prices; among the worst I've known in more than 36 years in the industry. It's been a very painful process adjusting, and the continued weakness in the oil markets is going to be with us for some time.

 For some time, but not forever. This is not the first crash, and it won't be the last, so we need to take the long view. And in fact, it's at times like these that you really need, I think, an authoritative and very fact-based picture of the road ahead.

 It's one thing to adapt to tough conditions today. That's important. But it's important to adapt in a way that means you're ready to meet the next set of challenges, and there will be many. That's why we need a clear sense of how the energy landscape might evolve over the next 20 years. And that's the role of the BP Energy Outlook. It's about lifting the focus from the here and now, to look at where you're heading in the future.

 With that in mind, let me pick just three trends that come out of this for me -- out of this outlook. They really stood out from Spencer and the team's work. The first of these is the projection for continuing growth in the demand for energy.

 There are strong forces driving this projection. In particular, the global economy will continue to expand, driven by the fast-growing Asian economies. As it does, more energy will be required to fuel these higher levels of activity and living standards. Our forecast is a rise of about one-third by 2035.

 Faster gains in energy efficiency are curbing the rate of energy growth, and GDP growth is subject to some uncertainty. But even so, significantly more energy will be required over the next 20 years to enable the world's economy to grow and prosper.

 The second notable trend I see is the fuel mix itself. Fossil fuels are likely to remain the dominant source of energy, powering the world's economy, supplying 60% of the energy increase out to 2035, and almost 80% of total energy supplies by that date.

 Now, within these figures, gas remains set to become the fastest growing fossil fuel, by far. Gas is very much the growing fuel of the early 21st Century. There are ample gas resources, especially with shale. It is often competitive with coal, and it is increasingly being transported internationally. And it is the cleanest fossil fuel, with major reductions in CO2 possible when it replaces coal in power generation.

 In contrast, the growth of global coal consumption is likely to slow sharply, as the Chinese economy rebalances towards a more sustainable pattern of growth. Renewables are set to grow the fastest of all, as their costs continue to fall, supported, of course, by the pledges made in Paris. But their current scale is small -- the point of realism. So even with rapid growth, we foresee only about 10% of primary energy, or just under that, coming from renewables other than hydro by 2035.

 The third big message for me is that the outlook for carbon emissions is changing significantly and positively. The rate of growth of carbon emissions is projected to more than halve over this outlook period compared to the past 20 years. Now, that's not emissions; that is the rate of growth. This is driven both by faster gains in energy efficiency or energy savings, and by the shifts towards lower carbon fuels or energy switching.

 Despite this slowing of the rate of growth, carbon emissions are still likely to increase, indicating the need for further policy action. The trends we observe are not meant to be what we want to see, but they are what we see from the data. As you probably know, we believe carbon pricing has an important part to play, as it provides incentives for everyone to play their part.

 I think business is very good working within the framework of rules that it understands. It can allocate scarce capital; it can do it quickly. But we need the frameworks for which to be able to make these long-term decisions.

 Those are just three of the fundamentals, and I'll hand over to Spencer in a moment to take you through the detail and the data. But let me just say the Energy Outlook has a very practical application in BP.

 Take the three trends I've just mentioned. They are all reflected in BP's businesses today, and our plans for the future. We're continuing to invest in a very disciplined way, especially now, to help meet growing energy demand. And you'll see that in a great set of projects starting up over the next five years, and the strong portfolio we have to choose from for the future.

 Looking at the energy mix, our portfolio is already around half natural gas, and is heading towards more gas with some very big projects on course, on budget, in Egypt's Nile Delta, in Khazzan, in Oman, and the giant Shah Deniz 2 project in Azerbaijan, combined with the southern corridor pipelines to Europe. We could be 60% gas by the end of this decade, and more beyond that.

 We are very aware of our part in the transition to a lower carbon economy. More gas will help with that, and we have a sizeable and growing renewables business, as well as making inroads on efficiency, big wind farms in the United States, big biofuels in Brazil.

 So, sure, 2016 looks at to be another tough year for our industry. But we mustn't lose sight of the longer term. Energy is a long-wavelength business, and you need a long-term perspective, and that's what the Energy Outlook gives us.

 I'd like to thank Spencer and his entire team that are here today, and all of the contributors across BP, ministries around the world and governments. They have worked very hard at gathering the data, and analyzing it for the benefit of industry and policy makers. So again, thank you.

 We hope this edition of the BP Energy Outlook can make a useful contribution to all your discussions about the future energy needs and trends going forward. Thank you very much.


 Spencer Dale,  BP PLC - Group Chief Economist   [2]
 Thank you very much, Bob. Let me extend my thanks to all of you for sparing the time to attend today's launch of the Energy Outlook, both all of you here in person in St. James, London, and also, as Bob said, the many of you around the world joining via the web. You're all very welcome.

 I also want to just extend my thanks to the many people in BP who have been beeving away over the last few months producing the Energy Outlook. That obviously includes the rest of the economic team, so thank you very much, guys. But it extends far beyond that, with contributions from right across BP -- upstream, downstream, technology, policy, trading.

 The Energy Outlook is very much a team BP effort, and I must admit I think that's one of its key strengths. The Energy Outlook isn't just based on the views of a few nerdy economists sitting in head office. It reflects the views of experts right across BP, so many thanks to everybody who contributed.

 I stood here a year ago presenting last year's Energy Outlook, and at that point I'd only been in BP a few months. I found the whole process of working on the Energy Outlook an enormously fascinating thing, and learned a huge amount. But at that point I wasn't sure if much of that was just because it was all new to me, and for everybody else it was all old hat, and it was just me discovering all of these things that everybody else knew.

 I have to say, a year on, I've learned even more this year around than the previous year. And I think that's just a reflection of the huge forces and trends which are reflecting our markets and our industry at the moment, and the impact they are likely to have over the next 20 years. The Energy Outlook just provides a fantastic window into those forces and trends.

 That said, I must admit it's a bit like what Bob was saying, it might seem odd to some to spend time discussing the outlook for the next 20 years when the energy markets are in such a state of flux. How, you might well ask, can we have any confidence in forecasting the demand for energy and fuels over the next 20 years in such an uncertain environment? And the simple answer is we don't.

 But that's not the point. The value of exercises such as the Energy Outlook doesn't stem from the fact that we kid ourselves that we have some magic crystal ball. Rather, it comes from providing a consistent framework which can be used to explore and analyze the forces shaping energy markets over the next 20 years.

 A key component of that value is providing a discipline and coherence to our thinking. It seems almost certain, for example, that renewables are likely to grow very rapidly over the next two decades. Likewise, China's economic growth looks set to slow materially.

 But understanding the likely consequences of these trends is far less certain. Do they signal a point of inflection or a hard stop? Qualitative trends are easy to talk about. Quantifying their likely impact provides the real value.

 Once you have a base case which captures these forces, you can then ask two further questions. One: backward looking. How have events over the past few years turned out differently from what we expected, and what signal do they contain for the future? Learning the lessons from the past is a hugely important process when trying to think about the future.

 The other question is forward-looking. What are the key assumptions and judgments shaping the outlook, and how would it differ if we changed those assumptions? As we all know, we live in a hugely uncertain world. Understanding the nature of that uncertainty is crucial for a company like BP undertaking multi-billion-dollar long-term investments.

 The Energy Outlook booklet you will receive on the way out today, or you can download from BP.com, contains all three of these elements: a clear base case, a review of past revisions, and a series of alternative cases exploring key uncertainties. Now, I'm not going to be able to do anywhere near full justice to all of that in today's presentation. But what I hope to do is do enough to whet your appetites, so you'll go away and read the outlook in full. What I plan to do today is I'm going to try and focus less on precise details and facts, and instead try and highlight some of the key issues and questions raised by this year's outlook, and what the next 20 years have in store for us.

 Okay, to set the scene, annual global GDP growth is expected to average around 3.5% over the outlook. So, that's just a touch softer than growth over the recent past, and it equates to GDP roughly doubling over the next 20 years.

 Some of that increase is driven by population growth, but the vast majority is driven by increases in productivity, shown here by the orange bars, especially in emerging Asia, as those countries gradually build up their capital and adopt best-practice techniques from the West. So, this is a pretty conventional story of Asian tiger-led growth, with China and India accounting for almost half of global growth.

 This doubling in the world economy causes a demand for energy to grow by around a third. Virtually all of this increase in energy is consumed in fast-growing emerging economies. Energy consumption in the OECD, shown by the green swath on the left-hand side here, is essentially flat over the next 20 years.

 Growth in energy demand is projected to average just a touch below 1.5% per year over the outlook, which is quite a bit slower than the growth over the past 20 years or so, reflecting both a sharp deceleration in China's energy demand and this plateauing in energy consumption within the advanced economies, shown by this green line.

 But can we really be sure that energy demand will keep increasing? Isn't this big oil simply talking its own book? What drives energy demand? The simple answer is economic growth.

 As the world economy expands, more energy is required to power higher levels of economic activity, especially in emerging economies in Asia and Africa, where large parts of the population still don't have access to electricity. Plentiful supplies of energy enable economic growth, raising living standards and lifting people out of fuel poverty.

 The amount of additional energy required to fuel the global expansion is offset by reductions in energy intensity, the average amount of energy needed to produce each unit of output. And in particular, energy intensity over the outlook is projected to fall more quickly than in recent history. Increasingly less energy is required to fuel economic activity. And that reflects both the continued rebalancing of the Chinese economy, and more generally, worldwide efforts to improve energy efficiency, post Paris.

 The impact of falling energy intensity is illustrated very clearly here in the left-hand panel. Global GDP, shown by the red line, more than doubles, but energy demand increases only by a third. The widening wedge between the two reflects the impact of declining energy intensity. But even with this rapid improvement in energy intensity, energy demand still increases by a material amount over the outlook.

 There are two basic reasons why that bottom-line conclusion may be wrong. If global GDP grows more slowly than expected, the red line grows by less. Or, if energy intensity falls more quickly than projected, causing an even bigger wedge between GDP and energy demand to open up. And both of those possibilities pose risks to the outlook.

 To explore the first risk, the possibility of slower global growth, we consider an alternative case in which China is assumed to grow more slowly than expected in the base case. In particular, we construct a scenario in which Chinese GDP growth averages just 3.5% over the outlook rather than the 5% growth assumed in the base case.

 Taking account of the implications of this slower Chinese growth for the rest of the world leads to global growth of a little below 3% per year. So that's about 0.5 percentage points lower than in the base case, in which you can see if you look across to the right-hand side will be comparable to one of the weakest periods of economic growth seen in recent history.

 Energy demand in this alternative case, shown in purple, is projected to be just 1% per year, weaker than at any point in recent history, and around a third slower than in the base case. But even in this alternative case, energy demand still increases by around 25% by 2035. Significantly less than in the base case, yes, but still a marked increase.

 What about the second possibility, that energy intensity may fall even more quickly than in the base case? As I mentioned, the projected pace of improvement in energy intensity in the base case, shown here in purple, is significantly quicker than seen in history. The green bar on the left here shows the average pace of decline since 1965, and the red bar, the quickest pace of improvement in any 20-year period over that time. So, a simple reading of history might suggest that the risks to energy intensity are actually skewed towards a slower rate of decline, which, other things equal, would point to stronger energy demand.

 Of course, history may not be a good guide to the future. In particular, the changing structure of the world economy, together with the pledges made in Paris, mean that energy intensity is likely to fall more quickly than in the past, which are exactly the judgments built into the base case. But as shown by the yellow bar here, for energy intensity to fall sufficiently quickly to offset any increase in energy demand, it would need to decline at more than double the rate seen over the past 20 years, which seems highly unlikely.

 So, to return to the question I posed -- what drives energy demand -- the simple question is the increasing need for energy as the fast growing economies of the world grow and prosper. Increases in energy enable that growth. Rapid declines in energy intensity are likely to limit the required increase in energy. And the extent of the increase will be highly dependent on the strength of global growth. But unless the world economy grows far more slowly than most commentators expect, we are likely to see a marked increase in the demand for energy over the next 20 years.

 In terms of the fuels meeting this increased demand, the outlook points to a significant shift in the mix over the next 20 years. Fossil fuels remain the dominant source of energy powering the world economy, providing around 60% of the increased energy, and accounting for almost 80% of total energy supplies in 2035.

 Gas is the fastest growing fossil fuel, overtaking coal as the second-largest fuel source by the end of the outlook. Oil grows steadily, although the secular decline in its share continues.

 In contrast, coal suffers a sharp reversal of fortune, with its growth slowing to just 0.5% a year, less than a fifth of the rate seen over the past 15 years, and its share within primary energy declining to its lowest on record. Renewables, shown by the orange line, are the fastest-growing fuel source, with their volume almost quadrupling, and their share within primary energy increasing from around 3% today to 9% by 2035.

 These shifts in the global fuel mix over the next 20 years reflect a myriad of factors, and there are perhaps four key issues in particular worth highlighting when thinking about that. One of those is the outlook for carbon emissions, and I'm going to come back to that later.

 The other three are the prospects for shale oil and gas with a key sub-theme of what have we learned about the US shale revolution; China's changing energy needs as its economy continues to rebalance; and the prospects for renewables and non-fossil fuels more generally. What I plan to do over the next 10 minutes or so is say a little bit about each of those, starting first with US shale.

 These charts show forecasts for US tight oil and shale gas since 2013, which is when we first started to produce separate forecasts for shale oil and gas in the Energy Outlook. So, just for a moment, just think about that. It's only three years ago which we started producing separate forecasts for tight oil and shale gas, which is a stark reminder of just how recent a phenomena the US shale revolution is. It's only three years ago.

 As you can see, we've been repeatedly surprised by the strength of US shale. Technological innovation and productivity gains have unlocked vast resources of tight oil and shale gas, causing us to revise our projections successively higher. And this year is no exception, with sizeable upward revisions to the profiles for both tight oil and shale gas, shown by the green lines.

 For US tight oil, the falls in investment and rig count over the past year mean that production is currently around 0.5 million barrels a day lower than the peak levels seen last spring, and is likely to fall further through much of this year. But as the oil market gradually rebalances, and prices eventually firm, we expect US tight oil to increase by almost 4 million barrels a day, stabilizing at around 8 million barrels a day, and accounting for around 40% of total US production.

 The resource base for US shale gas looks far greater than that for tight oil, such that US shale gas grows rapidly over the outlook, averaging over 4% per year. By 2035, US shale gas accounts for around three-quarters of total gas production, and almost 20% of global output. And measured in comparable units, US shale gas by 2035 is more than twice as big as tight oil.

 If we widen the lens to look at the global production of shale oil and gas, US output, which is shown by the green bars here, dominates the growth in global production of both tight oil and shale gas. But as you can see, production gradually diversifies, so that in the final 10 years or so of the outlook, production from the rest of the world accounts for around half of the global increase of both tight oil and shale gas in those final 10 years, with marked increases in China, Argentina and Canada, amongst others.

 In the base case, global tight oil increases to around 10 million barrels a day, so roughly double its current level. But that still accounts for less than 10% of the oil market. The role of shale gas is more pronounced, with its share in global gas production increasing from a little over 10% today to almost 25% by 2035.

 So, this is a pretty upbeat story on tight oil and shale gas. But I'm also very conscious that we have been repeatedly surprised by the strength of the shale revolution. Have we really learned our lessons this time? What happens if we continue to be surprised?

 To explore this possibility, we consider an alternative, stronger shale case, in which global shale resources are assumed to be far greater than currently thought, and productivity is significantly higher than in the base case. Now, the full details of this alternative case are described in the booklet, but just to give you a sense of how this might change the outlook, the dotted lines here show the revised market shares for tight oil and shale gas in this stronger shale case.

 Tight oil production is around 10 million barrels a day higher than in the base case. Some of this is at the expense of other types of oil production, which is crowded out, and the rest, just over half of the additional increase, is matched by higher demand. Likewise, shale gas production is around 75 Bcf a day higher than in the base case, with shale gas in this alternative case accounting for more than a third of global gas production.

 Although this is not my central case, we do need to remember that we, and most others in industry, have to date repeatedly underestimated the strength of the shale revolution. And so, perhaps this provides some food for thought.

 Turn next to the issue of China's changing energy needs. At the risk of stating the obvious, China really matters for energy. It is the world's largest consumer of energy, accounting for over a quarter of the world's total energy demand. And it's the world's largest growth market, accounting for almost 60% of total increase in global energy consumption over the past decade.

 But as China's economy rebalances to a more sustainable path, its energy needs are likely to change, with potentially quite profound implications for global energy trends. In particular, we expect China's demand for energy, shown here by the purple bars on the left-hand panel, to grow on average by less than 2% per year over the outlook, which is around a quarter of its rate over the past 15 years.

 Much of this reduction is driven by slower economic growth. Chinese GDP growth is projected to average around 5% per year over the next 20 years, around half the pace seen over the past 15 years.

 But growth in China's energy demand is expected to slow by more than GDP. That partly reflects increased energy efficiency. It also reflects a changing pattern of economic growth within China, with growth shifting away from traditional heavy industrial sectors which are highly energy-intensive, towards more consumer-facing service sector growth which requires far less energy.

 The most marked change from the past is in China's coal consumption, which, after years of strong growth as it fueled China's rapid industrialization, barely grows over the outlook, and is declining towards the end. The sharp break in China's coal consumption is compounded by strengthening environmental policies aimed at encouraging a shift away from coal towards cleaner, lower-carbon alternatives. As you can see here in this chart on the right, as a result, China's reliance on coal falls sharply, with the share of coal in China's primary energy declining from around two-thirds today to less than 50% by 2035. So, big shifts in the world's largest energy market over the next 20 years.

 Turn finally to the prospects for renewables and other non-fossil fuels. This chart shows forecasts for renewable power from the past five Energy Outlooks.

 As you can see, especially if you have got very good eyes and can look at colors, but if you trust my word, just as with US shale, we have repeatedly revised up our outlook for renewables, as we have been consistently surprised by faster-than-expected cost reductions and more rapid deployment, with this year posting the largest upward revision to date, reflecting both rapidly falling costs and the agreements reached in Paris. And as I mentioned, renewables almost quadruple over the next 20 years, accounting for over a third of the growth in power generation.

 The one thing I found really interesting when looking at this bit of the analysis is that these repeated upward revisions to renewables have not led to significant revisions to the overall level of non-fossil fuels. In particular, the upward revisions to renewables, shown here by the orange bars, have been broadly offset by downward revisions to nuclear energy post-Fukushima, and to biofuels, reflecting slower-than-expected technical progress.

 The key takeaway for me from this chart is that, just as with fossil fuels, the risks and uncertainties around the growth of non-fossil fuels are also two-sided. You can get it wrong on both sides; it's not just one way. That's what I wanted to say on those three key factors helping to shape the fuel mix.

 If we briefly turn now to look at oil and gas in just a little more detail, starting first with oil -- relates a little bit to what Bob was saying. I must admit, one of the great joys of working on the Energy Outlook, not only does it introduce you to a whole range of fascinating issues, it also provides a great excuse not to spend your whole time worrying about what the oil price is going to do next week or next month. So, I'm going to try and hang on to that today, and not get dragged into it -- so, apologies.

 But in very broad terms, the judgment underlying the Energy Outlook is that the market gradually rebalances over the next couple of years, as the current low level of prices boost demand and dampen supply. As Bob has said in the past, lower for longer, but not lower forever.

 Further out, we expect the global demand for liquids, which is shown on the left-hand side here, and which includes biofuels and other liquid fuels as well as oil, to increase by around 20 million barrels a day to reach just over 110 million barrels a day by 2035. All of this increased demand comes from emerging economies, with China and India accounting for over half of the increase.

 This is matched on the supply side, on the right here, by an increase in both non-OPEC and OPEC production. All of the net increase in non-OPEC supply comes from the Americas, led by the increase in US tight oil I was just talking about a moment ago. In terms of OPEC behavior, which has obviously been the subject of considerable attention recently, OPEC is assumed to maintain its market share at around 40%, increasing its production by around 7 million barrels a day over the next 20 years.

 If we look a little more closely at the factors driving the increase in oil demand just for a moment, around two-thirds of the increase in oil consumption reflects higher transport demand, as the number of vehicles outside of the OECD triples over the next 20 years to around 1.5 billion vehicles. The impact of this increase in the global car fleet on fuel demand is offset by large extent by gains in vehicle efficiency, which is assumed to improve even more rapidly than in the past. So, in old-fashioned parlance, if you like, an average passenger car in the vehicle fleet in 2035 is expected to travel around 50 miles per gallon, compared with around 30 miles per gallon today. So, really significant gains, 50 versus 30.

 Those efficiency gains are achieved by substantial improvements in the efficiency of the internal combustion engine, including far greater hybridization, smaller vehicle sizes, and improved vehicle design more generally -- so, for example, using lighter materials to build our cars. Importantly, the relatively high cost of batteries means that the penetration of electricity within transport demand is projected to remain very limited over the outlook.

 There's obviously huge uncertainty as to the precise pace at which battery technology improves and is able to start really competing with oil in the transport sector. Our central view, based on the analysis set out in BP's Technology Outlook, which was published last year -- and for those who haven't seen it, it is a must read, especially for issues like this. Based on the analysis in there, is we think that the batteries won't be able to compete, really compete -- the point at when that's going to happen, it's going to happen largely beyond the 2035 outlook. But there's obviously considerable uncertainty here, and it could come sooner, and so there's an obvious caveat there.

 If we turn next to natural gas, as I mentioned, gas is projected to be the fastest growing fossil fuel, growing at almost 2% per year. On the demand side, growth is supported by both gas gaining share relative to coal in the power sector, aided by ample supplies and strengthening environmental policies, and also by its increased use in industry as emerging economies industrialize. On the supply side, this increase in gas demand is met roughly evenly by increases in conventional production, with marked increases in the Middle East, China and Russia, and from shale gas as we were discussing earlier.

 A key feature of the gas outlook is a sharp increase in global supplies of liquefied natural gas, LNG, which is expected to more than double over the outlook. Around 40% of this increase in LNG supplies occurs over the next five years, as a series of in-flight projects are completed, led by Australia and the US. It's quite mind boggling.

 This roughly equates to a new LNG train coming on stream every eight weeks or so for the next five years. Quite a mind-boggling growth. And this strong growth spurt, as this market comes of age, causes LNG to surpass pipeline gas as the dominant form of traded gas by the end of the outlook.

 The significance of this is that the greater ability of LNG to respond to price signals is likely to cause regional gas prices to become increasingly integrated over time. If prices move too far apart in one market relative to the other, LNG supplies can be diverted to bring them back into line. The way I think about it, I think of LNG acting like a glue, joining regionally segmented markets so they move increasingly in unison.

 So, we're likely to see quite significant changes in pricing dynamics in global gas markets over the next few years. That's all I wanted to say in terms of oil and gas.

 The final issue I wanted to touch on this afternoon is the outlook for carbon emissions, where the big-picture message is that the outlook for carbon emissions is changing very significantly relative to the past. This next chart is designed to explain what's going on. So, the way to read this chart is you start with GDP, in the orange bars, and the growth of GDP over the outlook, shown on the right hand of the chart, is expected to be broadly similar to the past 20 years.

 But as discussed earlier, energy intensity over the outlook, shown in purple, is expected to fall more quickly than in the past. Moreover, the shift in the fuel mix means carbon intensity, the average amount of carbon emitted for each unit of energy used, shown in turquoise, also falls far more decisively than in the past. And as a result, the growth rate of carbon emissions, shown in gray, is expected to more than halve relative to the past 20 years, less than 1% growth compared to more than 2% growth.

 Some of this break with the past was in previous Energy Outlooks, but is more pronounced this year in the wake of the pledges and agreements made in Paris. That said, despite the slowdown in growth, emissions are projected to continue to grow with the level of carbon emissions increasing by around 20% by 2035. The widening gap between the projected path for emissions and, for example, the IEA 450 scenario, which is often used as sort of a benchmark scenario consistent with achieving the 2-degrees-C ambition, that widening gap illustrates the scale of the remaining challenge.

 From an economics perspective, a meaningful global price for carbon is likely to be the most efficient mechanism for responding to that challenge. One of the key benefits of a price in carbon is it provides incentives for improvements on both sides. On the demand side, reducing energy intensity; and on the supply side, reducing the carbon intensity of the fuel mix.

 Now, this next chart provides an alternative way of illustrating what's going on. When I first saw this chart, I found it really confusing, but bear with me for a minute. It's a neat chart if you can get your head around it, so I'll try and explain how to read this chart.

 So, the chart looks at different combinations of average declines in energy intensity and carbon intensity. Those were the two factors we just saw a minute ago, which were driving the slowdown in the growth of carbon emissions. As you move from the top left to the bottom right of the diagram, signaling a faster rate of decline in both energy intensity and carbon intensity, the average growth of carbon emissions declines.

 As we just discussed, the average rate of declines projected in the base case are quite a bit faster than seen over the past 20 years. And that's a comparison of the green dot, which is the base case, relative to the history, shown by the blue dot. But they still fall well short of those needed to get anywhere close to the IEA 450 scenario.

 It seems clear that we are in a transition to a lower carbon world. A major uncertainty when trying to produce an outlook for energy over the next 20 years is the speed of that transition. To explore this uncertainty a little further, we consider an alternative case in which there's a faster transition to a lower carbon world, shown here.

 This faster transition case is based on a number of judgments, a carbon price rising to $100 per ton in the OECD and China, and at least $50 a ton elsewhere. In transport, tougher vehicle standards, together with other measures, leading to more efficient cars, smaller cars and reduced mileage. And in industry and buildings measures to drive through significant gains in energy efficiency.

 As a result, as you can see on the chart on the right here, both energy intensity and carbon intensity are projected to decline at historically unprecedented rates. Emissions peak in 2020, and by 2035 are around 8% below their 2014 levels. That still falls short of the IEA 450 scenario, but goes well beyond the pledges contained in the Paris INDCs.

 In terms of the outlook for energy in this faster transition case, energy still grows, but around two-thirds of the rate in the base case. Non-fossil fuels supply all of the increase in energy. Fossil fuels fall slightly, with their share in primary energy falling from around 86% today to around 70% by 2035.

 The main casualty is coal, which falls sharply, hurt in particular by the high carbon price causing it to lose share in the power sector. Renewables are the main beneficiary, with their share in primary energy increasing to 15% by 2035 compared with 9% in the base case.

 Now, a rise in the share of renewables from 3% to 15% may, to some, not sound that impressive, but fuel shares are really slow moving. So, to put this in some sort of historical context, the increase in the share of renewables in the final 15 years of this alternative case is comparable to the gains made by oil in the period after 1908. That was a period which included the Texas oil boom, the discovery of oil in the Middle East, the British Navy switching from coal to oil, and the Model T Ford starting mass motorization. So, this is pretty extraordinary gains in renewables.

 But it's worth noting that even in this scenario, even with that growth, the demand for oil and gas grows, albeit at a reduced rate. So, hopefully, if you're anything like me, this sheds some light on how the energy landscape might be affected by a faster transition to a lower carbon environment.

 Let me stop and conclude before throwing it open to questions. In concluding, not surprisingly, I would echo the three key takeaways that Bob highlighted in his introductory remarks.

 First, global demand for energy is likely to increase markedly over the next 20 years, as the world economy expands, and more energy is required to power higher levels of activity and rising living standards. Increased energy enables that growth.

 Second, the fuel mix is likely to change significantly, with coal losing ground, renewables gaining, and oil and gas combined broadly holding their own. And finally, the outlook for carbon emissions is changing significantly, with emissions likely to grow far less quickly than in the past. But it is not changing quickly enough, suggesting the need for further policy action. Thank you.

 Bob Dudley,  BP PLC - Group Chief Executive   [3]
 Well, thank you very much, Spencer, and the team, again. And a big welcome to everyone who's here in London with us, and those of you who are on the web.

 We'll open it up for questions, as I know some questions here in the room, I think, in London. And we have a screen full of questions here, but we will pick as if we know the answer. (Laughter)

Questions and Answers
 Bob Dudley,  BP PLC - Group Chief Executive   [1]
 Let's start right here, and the second one, there was a hand here. I'll try to --

 Daly Sporten Kroosh,  MFC - Analyst   [2]
 [Daly Sporten Kroosh] from MFC. Thank you for yet again a brilliant presentation. Very well done to you and your team. Going back to shale gas and tight oil which formed a big part of your presentation. When you look at this, what kind of recovery factor improvements do you see over time?

 And if these recovery factors are increasing, how do you see this impacting on oil price? For example, we see with the technology present today, that shale oil at $40 to $45 a barrel, could open up fully again. The rigs will be out there drilling fresh and employment rising in this part of the US oil industry. If you could comment on that, please?

 Spencer Dale,  BP PLC - Group Chief Economist   [3]
 I thought that wasn't going to be a question on oil prices. So I think the issue of recovery rates is really key here. Why does tight oil -- US tight oil -- start to plateau about 8 million barrels a day towards the end of our outlook? What's the constraint? Why doesn't it keep growing and growing? It's because of the resource base. It's because we just don't think the resource base is big enough.

 Why have we revised it up from last year, is because some of the very significant improvements we've seen in productivity has unlocked some of those previous -- has unlocked more resources. And so that resource constraint has been relaxed, which is why we think we're going to see higher growth.

 If those recovery rates increase materially, suddenly that resource base could improve really quite significantly. So the numbers I have in my head is often people in the past talked about a recovery rate of 5%, perhaps as much as 10%, we start to see that drifting up. But how much further that could go I think is an absolutely key judgment.

 We have some further improvement built into the outlook, which is why we're able to get that growth. But it's not huge and so if it starts to get anywhere close to what you would see in a conventional well, then I think that would provide even more upside risks to the outlook.

 In terms of prices, I should at least try a little bit. I think, and I've talked about this in other environments, I think the speed at which shale responds to price signals means it has a very significant impact in terms of price volatility, acting to dampen price volatility. And so in the near term, by how much how quickly shale falls off, will have a big impact in terms of how quickly the market starts to turn. And how quickly shale then comes back on again will have a big impact in how quickly prices recover.

 Do I think that shale is going to have an important bearing on the trend level of prices over the next 20 years? It seems less obvious to me. The point I showed you is even with this growth, it's only counting for 10% of the market. It's not obvious to me it's the marginal barrel of oil in the long run. So it's not obvious to me it has a big impact on the average price level over the next 20 years. But I do think it has a very significant impact in terms of the responsiveness of prices to shocks and the dampening of that volatility.

 Bob Dudley,  BP PLC - Group Chief Executive   [4]
 A question here, here, and then we'll go to the web.

 Jason Gammel,  Jefferies LLC - Analyst   [5]
 Hi, it's Jason Gammel with Jefferies. Maybe we could continue on the theme of volatility and how much volatility do you actually expect to see in both supply and demand growth, maybe addressing two specific issues. The first, the level of capital that's being directed into new sources of supply, given the low level of oil prices right now.

 And the second level of demand growth giving that emerging markets have generally been the source of most of the demand growth. And now there's a huge trend of capital flow shifting so it's not going from developed markets into emerging markets because of the pressure on commodity prices.

 Spencer Dale,  BP PLC - Group Chief Economist   [6]
 So both good questions and I guess I should just be open up front. In terms of -- so when my children were young, they used to do drawing, and when they were very young, they used to have these really fat crayons. And then as they got older and older they used to do this get close and do fine art stuff.

 This is like fat crayons. This is looking at the next 20 years. This is not the right vehicle to try and think about the near-term aspects over the next -- how the market is likely to respond over the next couple of years. It's just not the right vehicle. So it's hard to use this to try and ask questions about -- precise questions -- about the near term just because this is like a fat crayon over the next 20 years-type story.

 On the capital side, the observations I would make is we've seen a very significant reduction in the level of capital expending. The level of capital spending -- CapEx spending -- by the end of this year could be something like 30% down on the level it was in 2014, so a very significant impact.

 The two questions then, which is in terms of how much, how quickly, will that come through, and so two points. Question number one is how quickly would that come through. And the source of analysis that I've seen suggests it doesn't come through very significantly in the first year or two, for the simple reason that companies like BP -- unless Bob tells me incorrectly here -- is once you started the project, you don't stop it, you follow through. And the big impact of these reductions is delaying FIDs of future projects. And so a lot of that peak may not come through until 2020 or 2021.

 Secondly, the nominal spending has fallen by around 30% or so, a number, 30%, 35% or so. There are many estimates out there produced by the consultancies which suggest that costs have come down by 20%, 25% or so, so far and could fall further this year. And so the reduction in real spending is an awful lot less than reduction in nominal spending.

 Because very simply, a dollar of investment today buys you an awful lot more investment than it did a year ago. And I think it will buy you even more in a year's time. So we have to remember when thinking about the scale of these things, so think in terms of real terms reflecting the cost reductions rather than nominal terms.

 In terms of the demand story clearly, financial markets from one month to another can affect how individual emerging markets are faring. The underlying story here is one of global GDP expansion driven by fast-growing emerging economies, particularly in Asia. And that fast growth then underpins the growth of energy, and that's the underlying story here. I'm not sure I see anything in terms of near-term fluctuations in financial markets which detracts from those underlying drivers which should drive that global expansion over the next 20 years.

 Bob Dudley,  BP PLC - Group Chief Executive   [7]
 Okay, where is the second question? Second question here. Then I'll go to the web and then we'll come back.

 Hugh Lee,  Ebico - Analyst   [8]
 [Hugh Lee] from Ebico. As you say, the carbon emission projections are very depressing. Have you taken into account the effect on GDP of the extreme climate events and the mass migration that those carbon emissions are going to cause?

 Spencer Dale,  BP PLC - Group Chief Economist   [9]
 The simple answer is no. We just don't have -- I think there's two points to that. One is, I don't think the sophistication of this type of exercise allows us to do that. I think perhaps secondly, is somewhat constrained by the fact I'm only looking ahead 20 years.

 So as I understand it, and I'm not a climate expert, many of those are factors, people would argue, would build up over a period of time. Many of them happening off stage relative to our 20-year-ahead outlook. But the simple answer is no, we haven't.

 Bob Dudley,  BP PLC - Group Chief Executive   [10]
 Okay, we've got a question from Bernard Davis in Hong Kong. Not sure we do have the answer for this but how will the proliferation of electric vehicles affect BP or the industry, if at all?

 Spencer Dale,  BP PLC - Group Chief Economist   [11]
 So as I said, our central view in the outlook is the penetration of electric vehicles and electricity more generally is likely to be pretty limited over the next 20 years. That's based on the analysis contained in the Technology Outlook published last year, that it takes time until battery technology is able to compete -- really compete -- aggressively with oil in the transport sector.

 The other thing to bear in mind is it roughly -- a rough rule of thumb I have in my head -- is it takes around 15 years for the car fleet to turn over. So even if by 2030 or 2035 battery technologies had improved more rapidly than we are expecting, such that they were really able to compete in a worldwide sense, then by 2035, the impact that would have on average consumption would still be relatively limited. Because you'll then need to turn that car fleet over and that will take you another 15 years or so.

 Does this mean we're saying electric vehicles and electricity will never penetrate? No, the central message from the Technology Outlook is if you look ahead 30 or 40 years, then it will start -- there's good reason to think that it really will start to be able to compete. But over the next 20 years of our outlook, we think it's likely to be pretty limited. But as I said, that's said with a huge humility with knowing that these things are very uncertain and there's an obvious caveat around that.

 Bob Dudley,  BP PLC - Group Chief Executive   [12]
 Question over here. Maybe two and then a third and a fourth. Then we'll go back to the web.

 Chris Weldon-Smith,  Oxford - Analyst   [13]
 [Chris Weldon-Smith] from Oxford. You showed a very rapid rise in renewables, but we know as renewables arise, it will become increasingly technically difficult to accommodate them and they will cost more and more. Have you tried to model that and put in that negative feedback in any way?

 Spencer Dale,  BP PLC - Group Chief Economist   [14]
 Yes, we tried to take that into account both in the base case and in particularly in that alternative case where renewables are growing so much more. So the particular constraint for example, is within the use of renewables within the EU, where it's starting to get 20% or 30% or 35% of power. And then that starts to become quite a significant constraint.

 You need to think about solving the intermittency problem and so on. That acts as one of the constraints for the penetration of renewables within there. I'm sure there's more sophisticated ways of doing it but we tried to capture some of those effects, both in the base case and in that faster transition case.

 Bob Dudley,  BP PLC - Group Chief Executive   [15]
 Two rows back. One row back, sorry.

 Paul Newman,  ICAP Energy - Analyst   [16]
 Thank you, it's Paul Newman from ICAP Energy. Spencer, I wanted to ask you -- thank you, I'm sure we're all very encouraged to see your modeling of the gradual replacement of coal with gas. I'm sure everyone is very concerned about the amount of coal we've been using recently. I hope, obviously, that trend will continue but my question really is this.

 Have we really got a reasonable chance of getting anywhere in terms of Paris and carbon unless we really get more active and more serious about nuclear? Because at the end of the day, while gas is a lot better than coal, it's still creating about half the carbon footprint that coal is.

 Spencer Dale,  BP PLC - Group Chief Economist   [17]
 One of the nice things about the Energy Outlook is we don't try and predict what the right thing should be or the optimal way. What we try and do is just do some objective analysis saying based on these trends what do we think is the most likely thing to happen. Based on current trends, we don't see a very significant role for nuclear energy. It was picking up more in that faster transition case but it wasn't a complete -- it wasn't really bursting through.

 And that's in some sense -- and you know this, Paul, more than I do -- is there's two things going on. In the central case the share of nuclear energy within the Energy Outlook is like a flat line. Sometimes when you see a flat line you think, oh, God, that must be really dull. But this is one of those flat lines where there's two things going on either side of it and they're balancing out.

 On the one hand, nuclear energy within the traditional markets of the past, the EU, the US, in those markets we expect nuclear capacity to be declining over the next 20 years as many of the reactors are coming up passing their 60-year lives and they will likely be decommissioned. And the combination of both the politics and the economics of nuclear means that the levels of investment are likely to be quite limited.

 On the other hand, you have this amazing growth of nuclear energy within China. The LNG factor I was telling you earlier, the other amazing fact I carry around in my head is the nuclear program in China is equivalent to a new nuclear reactor being introduced every three months for the next 20 years. A new nuclear reactor every three months for the next 20 years. For those who lived through Hinckley, where it seems like we've got two or three years of parliamentary commissions just to get through one, (laughter) new nuclear.

 So two points on that is. One is we all know that when we take one big number away from another big number and we get a flat number, you only need to get one of those wrong to get a quite sizeable shift. So it may be those two things won't be leveled out.

 The other perhaps potentially exciting thing is up until now, where renewables have gained by this really amazing learning curves and the costs are coming down, if anything there seems to be negative learning on nuclear energy with the cost rising rather than falling. But it's possible that during this huge expansion you're seeing in China, you could get some very significant learnings by its doing. And that may spill over into some other parts of the world as those costs of the economics of nuclear energy change.

 I guess the final point, which is the slight implication of your question is, is what do you think we need to do? The truthful answer is I have no idea, and that's the benefit of a carbon price. The beauty of a price is you don't have to pick winners and losers. You don't have to say to policy makers now, pick the winners and losers for the next 20 years about what fuel is going to work and what's not going to work.

 I have no idea which one will work and no idea at the pace in which we achieve movements in energy efficiency, how quickly costs will go down in different fuels. The wonderful thing about a carbon price is you let the market do it for you. The market will find the most efficient mechanism and you don't have to play the game of trying to pick winners and losers over the next 20 years.

 Bob Dudley,  BP PLC - Group Chief Executive   [18]
 So one from the web and then second row up here. Spencer, what is your vision on carbon capture and storage at CCS? Did you take it into account in your projections?

 Spencer Dale,  BP PLC - Group Chief Economist   [19]
 I'm sure I've told this thing last time. It is really striking. You have these questions in front of you here. It's like you have an exam paper and you've just gone into the exam room. You've turned it over and gone, oh, my goodness. And then it's even worse because at least when you turn an exam paper over you get to pick which question, but here your boss gets to pick which question (laughter).

 So carbon capture story. Within the central case, we only have a very small role for carbon capture and storage. And that's simply is based on the view that the economics of that and the need for scale, the amount of money you need to achieve the scale of that, in the current environment in that central case just means that it doesn't have a major role.

 Interestingly, in the faster transition case, as the carbon price gets up to $100 real, you're starting to see a more significant increase in carbon capture and storage, particularly towards the end of the Energy Outlook. And that's why, if you like, that's one of the factors why that carbon emissions line is starting to bend down quite a bit more quickly towards the end as carbon capture and storage plays a more significant role. In our central case not very much at all but more of a case within the faster transition, the economics being helped by the high carbon price.

 Bob Dudley,  BP PLC - Group Chief Executive   [20]
 That was a question from Sarah [Brudenniss] in the UK. Now second row, and then we'll go to the third row for a couple of questions.

 Megan Darby,  Climate Home - Analyst   [21]
 Megan Darby, Climate Home. As a climate analyst and journalist, it's interesting to see you expand your scenarios and consider a faster low carbon transition. But as you acknowledge, it's still some way off the IEA case.

 Last year your shareholders asked you to consider -- to stress test your business against a 2-degree limit on global warming. In Paris, the goal agreed was well below 2 degrees and pursuing efforts for 1.5 degrees. So why is that 2-degree analysis not visible here in this outlook and when are we going to see it?

 Spencer Dale,  BP PLC - Group Chief Economist   [22]
 What we do with the Energy Outlook is try and produce a global outlook. It's not designed to talk about BP's own portfolio. Let me explain to you what we did when we were thinking about the faster transition case.

 What we didn't do is work back from a particular aspiration and then cast back and work out what did you need to do. In part, that was because many people have already done that. The IEA's 450 scenario do that in great detail.

 Instead, what we did was we sat around with our key policy experts, technology experts, carbon solution experts within BP and said working within the current realms of policy -- so without needing a fundamental regime shift in policy -- and working within the current realms of technology and how we think technology is likely to evolve, without assuming a very significant shift in technology, what do you think would be a scenario which would be very stretching but would be within those scenarios of within the current policy regime and within the current gambit of what we understand as technology. And that's what that faster transition case came up with.

 As you can see, I think relative to history, that is still quite an exceptional change relative to things in the past. That diagram showed you the rates of declines of energy intensity. The rates of declines in carbon intensities are off the scale relative to anything you've seen in history. The penetration of renewables is extraordinary relative to what you've seen in history.

 So I think it's interesting, number one, to see what the energy landscape is in that world. And I also think you just underlined to me just how unbelievably challenging that 450 scenario is relative to what you need to do just to get, if you like, half way towards that, which is crudely where this faster transition case gets you.

 Bob Dudley,  BP PLC - Group Chief Executive   [23]
 I think a conclusion is that the technology has not yet been invented to have the breakthrough to get there. I think that's what we conclude with what we see. Third row, there's two gentlemen there and then some.

 Bruce Duguid,  Hermes Investment Management - Analyst   [24]
 I was going to ask a similar question. It's Bruce Duguid from Hermes Investment Management. To build on that discussion, you are advocates for a 2-degree policy outcome, together with other companies in the oil and gas climates initiative. And you are also advocates for a cost of carbon as the main instrument to achieve that.

 So have you painted the picture of what type of cost of carbon you would need to see in order to get that IEA 450-type scenario that's more than faster transition? And what impact that would then likely have on your markets, oil and gas in particular?

 Spencer Dale,  BP PLC - Group Chief Economist   [25]
 No. The only scenario we've done is the one I've shown you. I haven't done another one which I've got just in case somebody asks. So the only one I've done is the one we've done. And as I said, what we designed this to do was the framework which I just explained rather than working back from an aspiration.

 I think technically, this scenario is still consistent with the 2-degree-C scenario if carbon emissions were to fall sharply enough beyond that. It's not designed for that. But the way you end up here is not very far apart from the IEA's breach scenario which was designed to be just about. So does this roll out 2-degree-C? No, it doesn't, as a technical point. But it wasn't designed to do that.

 I guess what I'm really nervous about is saying I know today what carbon price would be necessary to achieve 450. Because frankly, if somebody tells you they do know the answer to that, I wouldn't believe them because we just don't know enough. We don't know how the world will evolve. We don't know how R&D will change once you start introducing a carbon price and so on.

 What this tells you, I think, is a carbon price rising to $100 is starting to have some real implications. It starts that trend, that upward trend is bending and is coming downwards again. But exactly what the right number is and can anybody work out some optimal trajectory for a carbon price to get you 450, my own personal view is if somebody thinks they can do that, I would be somewhat skeptical.

 James Skinner,  New Economic Foundation - Trustee Emeritus   [26]
 James Skinner, New Economics Foundation, Trustee Emeritus. You haven't said very much about costs and about capital costs and investment costs and production costs of energy. It seems to be pretty clear that the costs of producing renewable energy have fallen very, very sharply. And the projections are that they are to continue to fall as they get much larger production.

 Conversely, the long-term -- medium- and long-term trend for fossil fuels is almost inevitably going to be upwards, given that tight oil is now taking over increasingly from conventional oil. So if you've got a trend of rising costs in fossil fuels and the falling costs in the renewables, does it really make -- leaving aside the question of how many people are killed by mining and burning gas, oil -- does it really make commercial sense to go, to seem to be backing what looks like being the wrong horse?

 Spencer Dale,  BP PLC - Group Chief Economist   [27]
 I think a little bit here relates to this point -- and I apologize, this is a hobby horse -- between qualitative trends and quantifying trends. The story you just said about relative prices should mean that the share of renewables should be gaining over time and the share of fossil fuel should be declining over time. That's exactly what I just showed you in our outlook.

 And why are renewables rising over time? Because their costs are coming down, and in the outlook we talk about the cost curves and how quickly those costs will decline over time, which allows renewables to more than quadruple over the next 20 years, more than quadruple over the next 20 years. That is only possible because their costs are falling very sharply. So those qualitative trends, at least on the renewable side, is exactly what's built in.

 But the whole point here is you don't have to say, well, what impact does that actually have? That's what the bet is now. I'm more than happy to say somebody else can produce a different forecast with slightly different outcomes, but I'm not flying in the face of that at all. I'm not backing the wrong horse; I'm backing exactly that horse. I'm just then using a disciplined framework to think about how significant that would be.

 The only point where I take exception to you would be it's inevitable that the price of fossil fuels will rise over time. That's not so obvious to me. I think one of the -- I guess it's not so obvious to me -- I guess for twofold. One reason why I think it's not, and again this comes out from the Technology Outlook, is the world has a very large amount of fossil fuels available, and there's more fossil fuels under the ground than I think anybody ever thinks will ever be used. So there's the idea we have a supply shortage of fossil fuels and that I think is no longer seen as very credible.

 Moreover, we've seen unbelievable productivity gains in some parts of energy sector recently. One of the amazing things about tight oil and shale gas has been the unbelievable gains in productivity you've seen there. So if you look particularly at tight oil, productivity gains of a rig in tight oil measured by initial production has averaged over 30% per year, each year, for the last seven years. Unbelievable, mind-boggling increases in productivity.

 If you can start to take some of the those learnings and put it in conventional oil, it's not at all obvious to me that the price of fossil fuels have to rise over time. But your basic point that renewables are becoming increasingly competitive and they should be gaining share is exactly -- and I agree 100% -- and that's exactly what's built into this outlook.

 Bob Dudley,  BP PLC - Group Chief Executive   [28]
 A question from Australia. Spencer, you've already answered it but looking for more specifics. Do you expect coal demand to come down over time? And how much? And under what conditions is it likely to happen either earlier or later?

 Spencer Dale,  BP PLC - Group Chief Economist   [29]
 So we do expect coal. Coal continues to grow in our central case but just by an awful lot less and an awful lot slower than we've seen in the past. So it's growing on average by around 0.5% per year over the next 20 years compared to the past. Over the past 20 years, it's averaged 3%, so it's one-sixth of the growth rate seen in the past.

 A big reason for that is this shift in China, as I was telling you in the past. To a very large extent, China is coal. And China's needs are changing very substantially, and as a result of which, this -- in the past, as coal surfed the wave of Chinese industrialization, as that changes and the economy start to slow and rebalance, that will reduce the demand. That's reducing the growth of coal.

 Moreover, as we generally see, renewables become more competitive, environmental policies tighten, gas supplies become more and more available, coal will lose -- it's losing share relative to gas -- coal is losing share relative to gas and renewables within the power sector.

 What can make that happen more quickly, I think if any of those mechanisms go more quickly than expected, if the Chinese economy rebalances more sharply or they were able to achieve a quicker shift in its fuel mix, or if we see renewables or the supply of gas grow or as costs come down with renewables, the supplies of gas become more significant. Or if policy makers start to actually achieve a higher price for carbon, that would also bring about a quicker shift away from coal into other lower-carbon fuels.

 Bob Dudley,  BP PLC - Group Chief Executive   [30]
 And another one from James Golden in the US. You've talked about renewables, Spencer, but you haven't actually talked about different kinds. So what do you think are the most promising ones as you look out to 2035?

 Spencer Dale,  BP PLC - Group Chief Economist   [31]
 I think in growth terms, in growth the most significant, the best, the most significant one will be solar. That's largely just because it starts further -- it's earlier in its transition and so the cost reductions you're likely to see in solar are likely to be far more significant, so its growth rates will be lower.

 But in terms of absolute amounts it starts at a lower base relative to wind. So in terms of by 2035, I think wind is still likely to have a bigger share. But in terms of growth, it's solar, as you get very sharp reductions in the cost of solar, allowing the growth rates to be very significant.

 I think, I should know off the top of my head, I think for solar, the growth rate we have is close to 9% or 10% per year. I may have got that wrong, but somebody is nodding at me so that may be right. So I think for solar is something like 9% or 10% a year growth. Really very significant.

 Bob Dudley,  BP PLC - Group Chief Executive   [32]
 So before the third row up here on this side. But before I do that, we have a number of questions from all over the world, the United States, Europe, even from Belize. Different forms of a question: now that Iran's oil industry is opening to western markets, what are BP's plans? I think we'll put that aside today (laughter) so I don't want to raise any expectations. But thank you.

 Amcorta Dehaat,  Centre for European Reform - Analyst   [33]
 Thank you. My name is [Amcorta Dehaat], the Centre for European Reform. Last year I recall you showed a slide on the impact of geopolitical volatility. And I wondered if you on purpose didn't mention it today, or whether it's in the booklet we will receive outside. But could you say something about the impact of geopolitical volatility in this year's Energy Outlook?

 And related to that, if I recall correctly, you estimate that by 2035 OPEC will add another 7 million barrels a day. Where is that going to come from? Is that Iran and Iraq? Or is there something that we're missing? And how does Russia factor into that? Thanks.

 Spencer Dale,  BP PLC - Group Chief Economist   [34]
 Not very much in terms of the OPEC. I'm being facetious.

 Amcorta Dehaat,  Centre for European Reform - Analyst   [35]
 No, geopolitical terms.

 Spencer Dale,  BP PLC - Group Chief Economist   [36]
 So you're very clever. You asked three questions there (laughter). In terms of supply disruptions, there is no significance in not putting that up other than the fact I had too many slides already. But it's the same story as last year.

 Just to remind people, what we had done, which we built into last year's outlook, was we thought the level of geopolitical disruptions were causing the level of -- with taking about 5% of OPEC supply off the market. And then that was only gradually coming back over time, but just reflecting huge uncertainties. And we've assumed very similar assumptions for that this time around.

 For where does that 7 million barrels a day of production in OPEC come from, I'm told by my OPEC expert that we're not supposed to break it down and explain to people exactly bit by bit. But I don't think that's a very helpful answer (laughter).

 Crudely, Iraq is increasing by around 1 million barrels a day. We have Iran's production gradually increasing. It gets back to its pre-sanction levels by about 2020, early 2020s and then carries on growing. And we also have Saudi increasing its production. The rest is spread around but those three are doing quite a bit of the work.

 In terms of Russia, when I stood up here a year ago, somebody asked me a question about Russian production. And I said we had Russian production being broadly flat over the next 20 years. Some people were looking at me quite quizzically and said, well, how on earth can it do that with these sanction levels and what was going on in the Russian economy.

 And our answer then was, well, we think Russian production will be pretty stable for three main reasons. One is their reserves of conventional oil resources are very deep. The sanctions affect their ability to undertake unconventional projects, but they don't stop them in any way affect their ability to do conventional projects. So if anything, we just divert CapEx to that.

 Secondly, their fiscal regime is vanished in such a way to make sure that companies continue to have an incentive to work there. And third, the reduction in the ruble has also helped to support the supply chain.

 When I said that last year, people still at the end looked a bit quizzical. As you know, Russian production has actually increased slightly this year. And our view going forward -- and I think those mechanisms which I outlined a year ago underpinned it -- and our view going forward is it remains pretty flat going out for the next 20 years, again for those same reasons.

 Bob Dudley,  BP PLC - Group Chief Executive   [37]
 There's a question in the back row over here. Then we'll come over here to the fourth row, and then we'll come in here. Question was up. So while we're doing that I won't ask you one of the questions here because this is even deeper water, but people would like to know how you see the power of OPEC in the future. I think we should just leave that (laughter).

 Norman Sully Spencer,  - Analyst   [38]
 My name is Norman Sully Spencer. I have a question about exploration, if I may. Given the scale of the world's proved reserves that you've alluded to, the target of permitting global warming to 2 degrees and the continued development of the technology powering the shale revolution, is there any rationale for continued oil and gas exploration?

 Spencer Dale,  BP PLC - Group Chief Economist   [39]
 In our central case, I'm not sure I'll be able to bring it up now. Actually, there's some people in the back who have a really nice job today because they don't have to work very hard. So let's see if they can, if they can bring up slide 14 from the pack, from the actual book thing. Normally all they're doing is typing up the really tough questions, so well done.

 If I draw your eye to the chart on the right hand side here, this looks at the growth of different energies over the last 20 years in terms of the outlook. Oil and gas combined are the green and red bars. What you can see is in our central case, we expect the combined growth of oil and gas over the next 20 years to be almost the same as the combined growth over the last 20 years.

 So we've still got -- in our central case, you've still got significant increasing demand for oil and gas. Even in that faster transition case, what we were talking about earlier, as I said, oil and gas continue to increase in that scenario. Overall, fossil fuels fall, but oil and gas together still rise. Even in the IEA 450 scenario, by 2035, the level of oil and gas is pretty much around the same level as today. Which, if you think about the very sharp decline rates, means you have to keep on investing even in that -- what we were just saying earlier -- really quite extreme case in terms of the 450.

 So my view is based on the central case, the demand for oil and gas looks like it's going to continue to rise in the future at pretty much the same rate it has risen over the last 20 years. Thank you, at the back.

 Neanda Salvaterra,  Wall Street Journal - Media   [40]
 Hi, I'm Neanda Salvaterra for the Wall Street Journal. This question is for Mr. Dudley. In terms of Brexit, what is your view on that?

 Spencer Dale,  BP PLC - Group Chief Economist   [41]
 (Laughter) you get the hard questions.

 Bob Dudley,  BP PLC - Group Chief Executive   [42]
 (Laughter) just like I know when you're negotiating oil and gas contracts it's best to be quiet in the middle of negotiations. So I think we've said this as a Company. We think there are reforms that can make Europe more competitive as a place to do business.

 We support the improvements that we think can be made. We think it's good for BP if the UK is in Europe, and we think it's good for Europe if the UK is in the -- I think that's probably enough said (laughter). But thank you for the question (laughter).

 We had a question right here, sir. Third row in the middle, as hard to get to as you can get at. It's like tight oil.

 John Feddersen,  Aurora Energy Research - Analyst   [43]
 Sorry. John Feddersen, Aurora Energy Research. Many thanks for the presentation and for a great Q&A, I think, as well. My question is about the Paris climate talks. I think you attributed a lot of the changes between last year's outlook and this year's outlook to Paris outcomes.

 Could you say a little bit more about the drivers there? I know not everyone is unanimously in agreement that it was a particularly positive outcome. What gives you faith enough to revise your baseline scenario coming out of Paris?

 Spencer Dale,  BP PLC - Group Chief Economist   [44]
 Very happy to, John. And John Feddersen, who works for Aurora, for people who don't know Aurora, we worked quite closely with them on some of this work in terms of some people are modeling, doing very serious modeling of energy. I think Aurora, one of the best outfits out there. Thank you, and it's good to see you, John.

 The key thing for us was, we base our outlook on the most likely path for policy over the next 20 years. So when we produced our outlook a year ago, we knew Paris was going to happen and we built some of that into our outlook. So I can't compare -- so the impact of Paris cannot be looked at by comparing this outlook with the previous outlook because the previous outlook already included our best guess of Paris.

 So the best way in which I think -- and I know this is not quite answering your question -- I will answer your question. But the best way for me to think about what is the impact of Paris and everything around Paris is by looking at how we expect the next 20 years to be very different to the last 20 years. As I showed you, in terms of carbon emissions, we expect the next 20 years to be very different to the last 20 years.

 In terms of your specific question, in terms of how did Paris do it, I think for us, I think the level of -- partly the precise INDC pledges, but also the degree of agreement and the ratchet mechanism, which was discussed within Paris as well, gave us greater confidence that we would see different policies in one form or another which would lead to greater energy efficiency.

 Which is why we've reduced slightly the level of overall growth of energy demand in this forecast. And also a shift in the fuel mix, which has led us to crowd out coal a little bit more than last time and also crowd in a little bit more renewables in particular. So they were the mechanisms and changes relative to last year.

 Bob Dudley,  BP PLC - Group Chief Executive   [45]
 Spencer, given the pace of technology development -- and this is a question from Mauricio Bermudez Neubauer, here in the UK -- how do you forecast or try to account for unknown unknowns, technology disruptions and energy supply and demand that far out?

 Spencer Dale,  BP PLC - Group Chief Economist   [46]
 So one of the features of all economic forecasts -- and I've been forecasting for 20 years in one form or another -- is economic forecasts are always smoother than reality. And they are always smoother than reality because you can't predict the unpredictable. So when we were producing this outlook, do we -- almost by definition, I can't take build in the impact of an unknown unknown.

 So what we will do is, we will look at our best guesses, the pace at which technology has diffused over time and evolved over time, and make guesses about how that technology is likely to continue to improve. But almost certain is technology won't follow the nice smooth straight line that we have built into our forecast and it will come in fits and starts. The out-turns will be far more uncertain than we thought.

 And in some sense, that's exactly why we were doing the analysis in this term's outlook to try and keep learning so we can learn from that and build into the future. So two areas where you see technology work at its most over the last far or five years: renewables and shale oil and gas. That was the reason why I showed both of those charts to show you look, absolutely we have been surprised by that and we're trying to learn by that surprise.

 And so all you can do, I think, is how do we try and do this is just keep on being very humble, try and learn from the past and build in those learnings to the future. But for certain, the world will be far more volatile and far more unpredictable than our simple base case assumes.

 Bob Dudley,  BP PLC - Group Chief Executive   [47]
 Question on the end? Yes. Third row first, I think, and then second row.

 Ari Ucting,  Bloomberg News - Media   [48]
 This is [Ari Ucting] from Bloomberg News. I just want to know if there are any revisions that you have made in this outlook compared with the last one? And if yes, what does this tell you about the energy world that we live in now?

 Spencer Dale,  BP PLC - Group Chief Economist   [49]
 Yes, we do. Again, just to keep the guys in the back on their toes, if we bring up slide 64 from the pack. This chart demonstrates the revisions we've made and the way to read this is that the shaded bars at the back show you the range of annual revisions in past annual outlooks. Just to give you a way of gauging what's a big revision and what's a small revision.

 The key revisions we made are very much along the lines of the question I just gave to John. We've revised down energy demand a little bit and that's largely due to energy intensity. GDP growth hasn't been revised but it's an energy intensity story. Coal has come down very significantly and that's because, particularly, we revised down slightly the profile for growth in China and that's had a big impact. And also the impact of Paris and then this very significant upward revision in renewables.

 The one which I think was, perhaps what I found most surprising, I'll just comment on that, is gas. I would have expected the impact of Paris to have increased levels of gas. There's two offsetting effects going on here. One is the strengthening environmental policies which we expect post Paris, helped boost gas, in particular in terms of visibility to gain share relative to coal.

 But the other big impact of Paris in terms of improving energy efficiency, particularly in sectors where gas is quite predominant, push back in the opposite direction. And the net of those two is a slight downward revision to gas. But that's just because those two things are going the opposite directions. This chart is in the booklet, and there's some analysis surrounding it. Thank you, guys.

 Bob Dudley,  BP PLC - Group Chief Executive   [50]
 Ladies and gentlemen, we have gone a little bit over time. That's a little bit difficult with a webcast and I know some of you have some obligations. Spencer will be here, maybe they can grab you where we missed a few questions here, but you can catch Spencer here at the end.

 A big thank you for coming. Thank you for taking your time to personally come here, and those of you on the web, who are really spread out all over the world, thank you for joining us. We'll be back next year.

 We'll see how much we revise it next year, based on what will be an interesting year, no doubt. We will be back in six months with the statistical review as well, which is much more short-term in terms of supply and demand and use of energy. So thank you all very much.

PRELIMINARY TRANSCRIPT: "Preliminary Transcript" indicates that the 
Transcript has been published in near real-time by an experienced 
professional transcriber.  While the Preliminary Transcript is highly 
accurate, it has not been edited to ensure the entire transcription 
represents a verbatim report of the call.

EDITED TRANSCRIPT: "Edited Transcript" indicates that a team of professional 
editors have listened to the event a second time to confirm that the 
content of the call has been transcribed accurately and in full.

Thomson Reuters reserves the right to make changes to documents, content, or other 
information on this web site without obligation to notify any person of 
such changes.

In the conference calls upon which Event Transcripts are based, companies 
may make projections or other forward-looking statements regarding a variety 
of items. Such forward-looking statements are based upon current 
expectations and involve risks and uncertainties. Actual results may differ 
materially from those stated in any forward-looking statement based on a 
number of important factors and risks, which are more specifically 
identified in the companies' most recent SEC filings. Although the companies 
may indicate and believe that the assumptions underlying the forward-looking 
statements are reasonable, any of the assumptions could prove inaccurate or 
incorrect and, therefore, there can be no assurance that the results 
contemplated in the forward-looking statements will be realized.

Copyright 2018 Thomson Reuters. All Rights Reserved.