Right. Well, a big hello, everyone, and welcome to the launch of the 2018 edition of the Energy Outlook. We've got a full auditorium here in London, and I believe we have now over 11,000 sites linked up on the website. So good morning, good evening. We have people on the line from Canada to China, so a big welcome to London.
Thank you for your interest. We appreciate you taking the time to be with us. I know how busy we all are right now. I think everyone I talk to 6 weeks into the year, it feels more like 6 months everywhere. I was in Cairo last week where we just started up our first major project of the year.
We're producing gas from a field called Atoll to power homes and businesses in Egypt less than 3 years after making the discovery. And if you're not in the oil and gas industry, you may not know it, but that's really fast from discovery to production. Our industry is really moving fast too. And it feels like we used to have more time to think and to plan, didn't we? Everybody tells me that as well.
Whereas now understandably, we're all preoccupied with the day to day and future gazing may therefore seem a little bit like a luxury. But it's important to look ahead to the bigger picture. And that's the beauty of the energy outlook. It's why we asked Spencer Dale and his economics team to take the time to think about and ponder, reflect, consider the trends that may shape the future energy landscape. Professor one time at Caltech, Berkeley said to me, to have a good organization, you have to have good people, you have to give them tools and you've got to give them the time to think.
And it's always made me think, are we giving people enough time to think? That said, you'll hear from Spencer in a moment, we don't pretend that the outlook provides all the answers. But it does give us insight into the direction of travel and it's a great gift to have. As you'll also hear, we've made a few changes to keep improving the outlook's contribution to our understanding. One of those is to look a little bit further ahead, this year out to 2,040, which extends our view of the landscape over now over 2 decades.
And I'd like to thank Spencer Dale and his team of economists and the many contributors to the report, and many of you are on the line as well. He and his team have once again produced a first class publication that I hope readers will find useful. I'll hand over to Spencer in a few moments his overview of the findings. I should say some of the findings because as you'll see, there's even more information packed into this year's report than usual. First though, I'll very briefly set the scene by outlining the 3 main themes that struck me when I read the report.
And the first one is just the speed of the transition that's underway. I mentioned how fast the industry is moving right now, and it will continue to change as government policies, technologies and social preferences change and alter the way in which energy is produced and consumed in the future. Predicting to a degree of certainty how these changes will turn out is tricky business. It's why this year's outlook will consider a number of different scenarios, 6 in fact. It laid all of our overall understanding of the uncertainties that we face as a company, I hope it will for you as well.
The second theme is that competitive pressures within global energy markets continue to intensify as demand for energy continues to grow. However, technological advances mean our ability to produce energy is growing even faster, be that in unconventional oil and gas or in renewables like wind and solar energy. The continuing rapid growth of renewables means that in a couple of decades, we can expect the fuel mix to be the most diverse ever seen. This combination of diversity and abundance is going to mean the marketplace will be highly competitive for some time to come. I firmly believe the focus on efficiency, reliability and a very disciplined approach to cost and capital is here to stay in our industry.
My third observation is that the speed of change and the diversification of the fuel mix is having a significant impact on carbon emissions. The evolving transition scenario, it's one of the 6, Spencer will outline, which assumes policies, technologies and preferences evolve in a manner and speed seen over the recent past suggests that the growth in carbon emissions is slowly slowing markedly. Now that's the rate of growth. And that's good news. However, the slowing falls well short of the sharp drop in carbon emissions thought necessary to achieve the climate goal set out in Paris.
So there's a lot more to do. The outlook also shows that policies focused on specific fuels and technologies around renewables and EVs are unlikely to be sufficient on their own. As I tell almost every audience I speak with, we need to remember that a race to renewables will not be enough to meet the Paris goals, nor will it keep the world moving. The race needs to be to lower emissions. So we should be agnostic about fuels and focus instead on reducing emissions.
What's needed is a comprehensive approach that encourages both improvements in how efficiently we use energy as well as the continuing shift to a lower carbon fuel mix. In BP, we continue to believe that carbon pricing must be a key element of any such approach as it provides the incentives for everyone, whether that's producers and consumers alike to play their part. So those are the 3 themes that I took away from this report, the speed of the transition, the intensifying competition and most importantly, the need to keep downward pressure on carbon emissions. While no one can say exactly how these trends will play out, the outlook can help manage the uncertainties ahead to ensure we are fit and ready to play our role in meeting the energy needs of tomorrow. It means we need to be ready to meet society's demand for more energy.
And we should be proud of that effort as it will help to lift 1,000,000,000 more people out of low incomes. But as we do this, we need to continue adapting and working together to play our part in achieving the transition to a lower carbon energy system. That's our dual mission. And I'm confident this industry is up to the challenge. I know BP is up for it.
And I hope you find this energy outlook a useful contribution to our discussions today and for those where you are around the world. And think about it in the subsequent actions you take. Over to you, Spencer.
Thank you, Bob. Good afternoon, everyone. Let me add my thanks for everybody for sparing the time to come to today's launch of this year's NEG outlook. Everybody here in St. James, London looking around, there's lots of familiar faces.
Thank you all very much for sparing the time again. And also around the world via the web, as Bob said, we've got a record number of people registered today. Thank you very much. Please stay with us. I'll let you into a secret.
The really good bits come right at the end. So don't leave before the end. And please do send in your questions. So if you hear me say something interesting, even better, if you hear me say something which sounds wrong in most couple of years, write in with your questions flash up on the screen and we will try and answer as many as we can during the Q and A session. Also take this opportunity just to thank everybody in BP involved in producing the outlook over the last few months.
That obviously involves the rest of the economics team, many of them here today, many of them also watching via the web. Thank you, guys. A busy last couple of months, but hopefully worth it. But the idea of the economic of the energy outlook is it doesn't just focus on economics. It tries to have a more holistic view of all the factors affecting global energy markets, technology, commodity markets, upstream, downstream, across the regions.
And to do that, we try and pull in insights from across the whole of the BP family. So this is very much a team BP effort. So thank you very much to everyone involved. And if I may, just a particular thank you to Robert Spicer and Ayesha Dullywell, who work in the downstream strategy part of BP. And they've worked with us incredibly closely over the last couple of months, working on the work we've been doing on EVs, electric cars and new mobility.
More on that, in fact, much more on that work later. But a special thanks to Robert and Ayesha for the work that you've done with us. Thank you. Before jumping into the detail, particularly for sort of seasoned watches of the Energy outlook, I wanted to mention and pick up on some of those changes that Bob mentioned we've made to the format and structure of this year's outlook, 3 in particular. First, as Bob said, we've extended the outlook for 5 more years and we're now going out to 2,040.
So looking ahead, 24, 25 years or so. And this has a benefit of bringing into sharper focus some features of the energy transition, which are more of a blur if you look ahead to only 2,035. 2nd, we've tried to focus even more this year on the range of possible paths and outcomes the energy transition may take. As I fear I bored many of you in the past, the value of forecasting is not to try to predict the future. Any point forecast will be wrong.
Rather, the value of exercises such as the Energy Outlook is to understand better the nature of the uncertainty we face and the key judgments and issues which really shape that uncertainty. So in particular, in this year's outlook, we consider 6 or 7 alternative scenarios designed to explore a series of what if questions relating to the energy transition. The scenarios vary in terms of their implications for both energy demand and the fuel mix shown here on the left hand chart, and as a consequence, the implied path for carbon emissions shown on the right. Now don't worry without looking at all these scenarios in detail at this stage. I will come back and look at all of them over the next 40 minutes or so.
We also no longer have a base or central case. The probability of any single scenario materializing is negligible. Remember, any point forecast will be wrong. That's always been the case, but I fear not always fully understood. And I was worried that the use of a base case has confused that even more.
However, for the ease of explanation, much of the presentation today and in the booklet is framed around what we call the evolving transition scenario or ET scenario. Now before you ask, no, the ET scenario has nothing to do with cute creatures from outer space, sorry. Rather, the idea of the evolving transition, the ET scenario, is to consider the broad path the Global Energy System might evolve along if government policies, new technology and social preferences continue to change in a manner and speed seen over the recent past. A sense, if you like, of the broad path we're all traveling along absent a shock to either technology or policy. To be clear, this is not a business as usual case.
The world of energy is very different today to what it was 10 or 20 years ago, and the forces giving rise to those changes are assumed to continue in the ET scenario. Indeed, in many dimensions, the pace of change has been increasing in recent years and we've tried to build that acceleration into the scenario. But just to repeat, just with all the other scenarios, the likelihood of the ET scenario materializing exactly as projected is tiny. The third change we've made to this year's format is to look at the forces shaping the energy transition through 3 distinct lenses or windows to energy by sector, by region and by fuel, shown here for the ET scenario. Viewing the changing behavior and energy needs of the world from each of these perspectives helps, well, I hope it helps, to shed light on the underlying forces shaping the transition, how energy is used, where it's used and in what form, the how, where and what of energy.
And I will come back to these three windows in a moment. But before I do, just to summarize the economic backdrop underpinning the outlook, and this is common across all the scenarios we consider. Global GDP is projected to grow by around 3.25 percent on a PPP basis, which you can see here is pretty much broadly in line with the world growth over the last 25 years with global GDP more than doubling by 2,040. Some of this growth is driven by increases in population, but around 3 quarters is driven by increases in productivity, shown by the dark orange segments here, particularly in fast growing and developing economies. And you can see the importance of productivity in, for example, China, where it's all productivity growth, which is driving growth and in India, where productivity growth is driving the vast majority of that growth.
Increases in productivity output per head translates into increasing prosperity, income per head. And as a result, more than 2,500,000,000 people, a third of the world's current population are lifted from low incomes by 2,040. The increasing prosperity of the developing world is one of the most important forces shaping economic and energy trends over the next 25 years. If we go back to my 3 energy windows, my how, where and what, I'm going to focus on the next few slides on the evolving transition scenario. The increase in global activity and prosperity causes energy demand to increase by about a third by 2,040.
That equates to an annual average growth rate of around 1.3%. So quite a bit slower than the 2% or so seen averaged over the past 25 years, reflecting quickening gains in energy efficiency. The ever increasing ability of the world to produce more with less energy is another key force shaping the energy landscape over the coming decades. In terms of the different perspectives, I'm afraid I only have time today to give you a brief glimpse behind the curtains of each window today. But I'm hoping that this quick peek behind the curtains will unleash enough of the nosy neighbor in you for you to want to look at for the full story in the Energy Outlook booklet.
You can get the booklet that you can download that free of charge from bp.com. And for those of you here in the room today in London, there's a copy waiting for you to pick up when you leave. So if we briefly look, 1st of all, behind the curtains of the sector window, no expense spared at BPOT, where growth of energy demanded broad based across all sectors, industry, buildings and transport. Industry is a major consumer of energy. It accounts around half of all energy consumed today and around half of the growth over the outlook.
Within industry, the non combusted use of fuels, particularly as a feedstock in petrochemicals, is the fastest growing source of demand, shown by the light blow bars here on the right. Now the ET scenario assumes that increasing environmental pressures on the use of some petrochemical products, particularly single use plastics and packaging dampens growth. But even so, the non combusted use of oil and gas is the largest contributor to their growth in the latter part of the outlook. So the use of oil and gas as a feedstock rather than a source of energy becoming the main driver of their growth. The growth of energy used in buildings, shown here on the left, is concentrated in Asia, Africa and the Middle East, which together account for around 90% of the growth of energy used in buildings.
The point to note here is that the relatively warm climate covering much of these regions, Asia, Africa, Middle East, means there is little new demand for heating. Rather, the majority of the demand is for air conditioning as well as for additional lighting and appliances. As a result, almost all of the increase in energy consumption in buildings over the next 25 years is in the form of electricity, shown by purple. A good example of whether how energy is used really matters. The right hand chart here shows the growth of energy used in transport, which slows sharply relative to the past.
This is a good example of that doing more with less trend I mentioned a moment ago. In the ET scenario, global demand for both passenger and freight services more than doubles, but this is largely offset by improving efficiency, this purple bar here. So energy used within transport increases by only about 25% over the outlook, much slower than in the past. If we go back to the 3 Energy windows again, and in this time, if we look through the middle regional window, as may be quite familiar to many of you, all of the growth in energy consumption is in fast growing developing economies, driven by the increasing prosperity and improving living standards I mentioned earlier. Energy demand within Neo ECD, shown by the green bars here, is essentially flat over the outlook.
It's important to remember that plentiful supplies of energy enable this increase in prosperity. Without access to secure and affordable energy, the improvement in global prosperity and living standards would be severely stunted. That's one half of the dual challenge facing the global energy system that Bob just mentioned. The other half of the challenge is to provide energy in a sustainable way, taking account of the implications for the environment and climate. More energy, less carbon.
And I'll come back to carbon emissions later. As shown by the 2 blue charts on the 2 blue bars on the right here, China and India are the biggest growth markets for energy, each accounting for around a quarter of the increase in global energy demand, with India showing the light blue bar overtaking China as the main engine of growth from the early 2030s onwards. And as you can see, Africa, showing here in red, starts to play a meaningful role in driving energy demand towards the end of the outlook. That's one of those features of the energy transition which comes into sharper focus once you think by going out to 2,040. As the global pattern of energy consumption changes, difference in the fuel mix across regions has an important bearing on the energy landscape.
The 2 main centers of growth, China and India, both start with relatively coal intensive energy systems. In the ET scenario, China's coal intensity declined sharply with coal consumption falling in outright terms. You can see that gray bar getting smaller. The fall in China's coal consumption is more than offset by a large rise in renewable energy, with renewables overtaking oil as the 2nd largest energy source in China by 2,040, a quite remarkable transformation. In contrast, India's coal consumption increases more in line with its overall energy demand such that the share of coal within India's fuel mix falls only slightly.
Turning finally to the final window on the energy transition in terms of how different sources of energy are evolving over time, the watt of energy. In the ET scenario, renewable energy in Orange is the fastest growing source of energy, accounting for over 40% of the increase in energy supplies, with its share in primary energy increasing from 4% today to around 14% by 2,040. Natural gas, in red, grows much faster than either oil or coal, overtaking coal and converging on oil by the end of the outlook. In contrast, global coal consumption essentially flatlines over the outlook with its share falling to its lowest level since the Industrial Revolution. The rapid growth in Renewables means in the ET scenario, we are heading for the most diversified fuel mix the world has ever seen, with oil, gas, coal and non fossil fuels, each providing roughly around a quarter of the world's energy.
As Bob mentioned, this increasing diversity will further add to competitive pressures within global energy markets. That's all I have time to say today about these three different perspectives on the energy transition, the how, where and what. As I said, I hope your curiosity has been sufficiently piqued that you do want to look behind the curtains in fall, and you can do that by downloading the full Energy Outlook booklet. What I want to do for the rest of today's presentation is explore some of the issues and uncertainties highlighted in this year's outlook by considering 5 questions relating to the energy transition. What have we learned about electric cars and the mobility revolution?
When is global oil demand likely to stop growing? Just how fast will renewable energy grow? How resilient is the outlook for natural gas and is the transition to a lower carbon energy system happening fast enough. So starting first with what have we learned about electric cars and the mobility revolution. I pose the question in this way since it feels well, at least it feels to me that we're all very much still learning about the best way to think about and analyze these issues.
As such, the analysis and issues outlook is very much offered in the spirit of our work in progress at the moment. One area in which our thinking has moved on is how best to measure the penetration of electric cars in the passenger car market. Almost all the current discussion on commentary on electric cars is centered around how quickly the number of electric cars will increase. Will the stock of electric cars reach 200, 300, 400,000,000 cars by 2,040. But there are two problems with this focus on the number of electric cars.
First, it treats all electric cars the same, but that's clearly wrong. A pure electric car, a pure battery electric car is powered solely by electricity. In contrast, a plug in hybrid electric car is powered by a combination of electricity and oil. Very roughly, you can think of a plug in hybrid today as being half powered by electricity and half powered by oil. Simply adding battery EVs with plug in hybrids is comparing apples and oranges, hence the different colors.
The second problem with focusing only on the number of EVs is that it takes no account of the different intensity with which electric vehicles might be driven, especially as shareability increases. If there are, say, 300 electric vehicles on the road, but they are driven twice as much as conventional cars, that's more akin to having 600,000,000 EVs on the road. So the approach followed in the outlook is rather than focusing on the number of EVs is to consider the share of vehicle kilometers, VKMs, powered by electricity. That measure, the share of VKMs powered by electricity takes account of both these problems, the different types of electric car and the intensity with which they are driven. Great expense, very good.
That's a good way of measuring things. But does this matter? It matters a lot. In the ET scenario, the share of EVs in the global car park, the number based measure here on the left, reaches around 15% or so by 2,040, 300,000,000 cars or so in a car park of almost 2,000,000,000 euros But the share of passenger car VKMs powered by electricity is over 30%, twice as much. The share of passenger VKMs powered by electricity is shown in the purple bar in this chart.
The key driver of this greater penetration, I. E. 30% rather than 15%, reflects the interaction of EVs with shared mobility and autonomy. So just to be clear upfront here, by shared mobility, I'm focusing here on the ownership of the car. So where shared mobility refers to traveling in a car, which is not privately owned by that individual.
So that includes conventional taxis, ride hailing services like Uber or Lyft or any form of car rental. They all count as shared mobility. In the ET scenario, fully autonomous cars start to become available from the early 2020s. But the high cost of that technology means in the first instance, they are largely purchased by companies offering shared mobility services rather than by individual families or households. The cost saving associated with no longer having to pay for a driver, which could reduce costs total cost by as much as 40% or 50%, leads to a surge in the importance of kilometers driven by shared mobility autonomous cars in the 2030s, shown by this light blue bar here on the right hand side here, which is showing you the growth of kilometers driven by shared mobility autonomous cars.
Importantly, and here is the link to electric cars, the relatively low running cost of electric cars compared with conventional internal combustion engine cars means a vast majority of these shared autonomous and more intensity used cars are electric. Substantially boosting the share of VKMs powered by electricity relative to measures focusing simply on the number of EVs. Measured in this way, roughly half of the increasing penetration of electricity in the car market reflects the growing number of EVs and half the increasing intensity with which they are driven. And the game changer for intensity is autonomy. So what does this mean for oil and other liquid fuels used by cars?
This is shown in this next chart here, which you read from left to right. Passenger cars currently use around 19,000,000 barrels a day of liquid fuels, so around 20% of the total liquids market. And that's shown here in the 2016 column. In the ET scenario, the distance traveled by passenger cars more than doubles by 2,040, shown by the green bar, largely driven by 100 of millions of families in the developing world being able to afford their first car. But the impact on oil demand is largely offset by tightening vehicle efficiency standards shown by those combined yellow and red block in the middle there.
In addition, the increase in shared mobility powered by electric cars further crowds out the demand for liquid fuel shown here in blue. The net result is that in the ET scenario, despite the distance traveled by cars more than doubling over the next 25 years, the amount of liquid fuels used by cars is essentially unchanged. Now, the evil eyed among you would have spotted that electric cars enter this chart twice, once via their shared mobility effect in terms of the intensity and once in terms of the tightening in the vehicle efficiency standard in the red bar there. We just discussed the role of electric cars in shared mobility, reflecting that intensity effect. That's the blue bit.
In terms of the role of electric cars within the vehicle efficiency standard, this reflects the increasing number of EVs. As vehicle efficiency standards tighten around the world, car manufacturers can meet these standards in 1 of 3 ways. They can sell a greater proportion of small conventional cars rather than big conventional cars. They can undertake investments in general efficiency improvements, such as more general hybridization or light weighting, or they can sell a greater share of electric cars, hence the role of electric cars, the number of electric cars as part of that tightening vehicle efficiency standard. But there's an important point to note here as there's a key interaction between the number of electric cars sold and the other forms of vehicle efficiency, which I'm not sure is always fully appreciated by some of the analysis I read outside.
In practice, car manufacturers will sell electric cars for a variety of reasons other than simply meeting their vehicle efficiency standards, such as responding to customer demands or as part of a long term strategy. But for a given vehicle efficiency standard, if a car manufacturer does sell a greater proportion of EVs, it means they could invest less in the other two types of efficiencies and still hit the overall efficiency standard. As such, without any change to efficiency standards, the impact of oil demand of simply selling more EVs would tend to be largely offset by smaller gains in the other types of efficiency. For those of you who like to think in graphical terms, Global Vehicle Efficiency Standards define the size of the combined yellow and red block. If for whatever reason car manufacturers sell more EVs, the size of the red block increases, this is likely to be offset by the yellow block getting correspondingly smaller as less investments are made in other types of vehicle efficiencies.
As such, in the ET scenario, the main impact of electric cars on liquid fuels used by cars in net terms comes via the intensity with which they are used and their interaction with shared mobility, the blue bar, reducing oil demand by around 2,000,000 barrels a day. The final issue I want to consider in terms of electric cars is a possibility that regulatory changes or technological improvements cause electric cars to grow far more quickly than assumed in the ET scenario. Since many of us met in this room a year ago, a number of countries have announced potential bans on internal combustion engines. Car manufacturers have set more and more ambitious targets for electrification. Suppose we're just miles behind the curve here.
So to explore this possibility, consider a scenario in which there's a worldwide ban on the sale of all internal combustion engine cars, ICE ICE cars, by 2,040. So global worldwide ban on the sales of all ICE cars from 2,040. We call this the ICE ban scenario. In fact, the scenario assumes that sales of plug in hybrids are also banned, So the correct name should really be the ICE and plug in hybrids banned scenario. But that wasn't quite as snappy.
So we start with ICE ban scenario. For those of you who don't follow this area closely, this type of ban would be far more stringent than any potential ban any country has announced so far, and we've applied it across the entire world, from Delhi to Doncaster, Moscow to Montana.
So I think it's fair to say this is
a fairly challenging scenario. The impact of the ice band is shown here on the green line with sales of battery electric cars as a share of total car sales reaching 30% by 2,030, 2 thirds by 2,035, 100% by 2,040. As a result, electricity powers around 2 thirds of passenger cars VKMs in 2040 shown here in the right, 2 thirds in the ice band scenario compared to just 1 third in the evolving transition scenario. For those of you scratching your heads at this point thinking, well, why don't electric cars account for all passenger car VKMs in 2,040? That's because some conventional cars sold before 2,040 would still exist and still be in use.
In terms of the impact of the ice bath scenario on oil demand, as we discussed As we discussed a moment ago, it depends heavily on the extent to which vehicle efficiency standards are altered and hence how big the offset is with other forms of efficiency. If we assume that vehicle efficiency standards are tightened by a corresponding amount such that there is no offset at all with other forms of efficiencies, you get the biggest possible effect on oil demand. Oil demand in 2,040 would be around 10,000,000 barrels a day lower than in the ET scenario. The ICE band roughly halves the use of oil used by passenger cars by 2,040. Now that's obviously a fairly sizable reduction in oil demand.
But even in this ice bound scenario, the level of demand in 2,040 is higher than it is today at around 100,000,000 barrels a day. The suggestion that the rapid growth in electric cars will cause oil demand to collapse just isn't supported by the basic arithmetic even with really rapid growth. Perhaps even more striking is the impact on carbon emissions. I'll come back to the details of this chart in a moment when considering the question on carbon. But just briefly, the blue line here shows carbon emissions in the ET scenario, which increased by almost 10% by 2,040.
That's much higher than the almost 50% fall in carbon emissions thought necessary to achieve the Paris climate goals, shown here in orange by the even faster transition scenario. So how much difference does the ice band make in terms of closing that gap? Almost none. Carbon emissions in the Ice Band scenario increased by around 7% compared with 10% in the ET scenario. And this is despite assuming that all the electricity needed to power the additional electric cars is produced by renewables and so doesn't generate any additional carbon emissions.
The simple point is that although electric cars may bring important benefits in terms of urban air quality, reducing oil demand by 10,000,000 barrels a day, although welcome, doesn't really move the dial in terms of carbon challenge. And I'll come back to what really does move the dial a little later. I'm sorry, that was an extraordinary long answer to my first question. At this point, you all think, oh, my goodness. Just to briefly recap, the short answer is, I think it has moved on quite a bit.
I promise I won't try and summarize again everything I just went through, but perhaps just to highlight three points. The penetration of electricity in the car market depends on both the number of EVs and the intensity with which they are used. In terms of intensity, autonomy is a potential game changer. And the potential impact on EVs on oil demand over the next 20 or 30 years, even with really rapid growth, looks relatively modest. Enough on EVs.
We turn back to my 5 questions, and I do I promise my answer to the other questions will all be a lot shorter. In terms of the second question, when is oil demand likely to stop growing? The very short answer is it all depends. If we flesh that out a little bit and start with the evolving transition scenario, the demand for oil and other liquid fuels grows over much of the outlook, increasing by around 13,000,000 barrels a day to reach 109 by 2,040. But as you can see from this chart on the right here, the pace of that growth gradually fades over the outlook, with liquid fuel demand declining over the final 5 years.
This slowing growth largely reflects the fading stimulus from the transport sector, shown by the blue bar here. As I mentioned earlier, global transport demand more than doubles in the ET scenario, but the impact of this on energy demand is largely offset by accelerating gains in efficiency. The other scenarios considered in the outlook, which have significantly different profiles for oil demand, are all focused around more aggressive tightening in environment and climate policies, and so all point to an earlier peaking in oil demand. The ice band that we just talked about and 2 faster transition scenarios, which I will come back to, the faster transition scenario in dark green and the even faster transition scenario in orange. In the I SPAN scenario, the demand for liquids starts to fall in the early 2030s.
And in the even faster transition scenario, which as I just mentioned is broadly consistent with meeting the Paris climate goals, oil demand starts to fall in the late 2020s. Just to highlight a couple of points from this chart. First, I don't think you should conclude from this chart that the risks to oil demand are all skewed to the downside relative to the ET scenario. As I said, the alternative scenarios we focused on in this year's outlook are all focused around more stringent climate and environment policies, hence this downsized SKU. But if, say, efficiency of new vehicles improved at the same pace as the past 15 years rather than significantly faster as assumed in the ET scenario, demand for liquid fuels would be going well into the 2040s.
There's uncertainty either side of the ET scenario. 2nd, it's important to keep this peaking in order market perspective. Even in the even faster transition scenario, consistent with a sharp fall in carbon emissions, liquid fuel demand is close to 85,000,000 barrels a day in 2,040, more plateau than peak. Indeed, to put that in perspective, this line shows a level of potential oil supply if the world was to stop investing in new oil production today, and existing production declined at an average annual decline rate of around 3% a year. And you see this big and widening gap between potential supply and any of these demand profile tells us that even in a scenario consistent with meeting the Paris Climate's goals such as the orange line there, the world is likely to need significant levels of new investment in oil production for many years to come.
Turning briefly to the question of where this oil production will come from. In the ET scenario, much of the initial growth in oil demand is met by increases in U. S. Tight oil. And as U.
S. Tight oil starts to flatten off in the early 2030s, the pattern of growth is then handed to OPEC members. In particular, OPEC members are assumed to gradually diversify their economies, reducing their dependency on oil, allowing them to adopt a more competitive strategy of increasing their market share. A key uncertainty when thinking about the precise split between U. S.
Tight oil and OPEC is the potential for U. S. Tight oil to keep growing. In the ET scenario, U. S.
Tight oil grows by around 5,000,000 barrels a day from current levels, peaking at close to around 10,000,000 barrels a day in the early 2030s. This is consistent with rigs remaining around current levels shown here on the right with rig productivity increasing by around 40% or so. But there's significant uncertainty concerning both the pace and duration of tight oil growth depending on the availability of finance and other inputs required to support rapid expansion in the short run, the pace. And over the longer run, the total volume of resources that can be economically extracted, the longer run duration of that growth. One possibility shown by this early peak scenario is that the availability of finance and resources allows a more rapid expansion in production with output peaking around 12,000,000 barrels a day in the mid-2020s.
But if overall production is the same as in the ET scenario of around 70,000,000,000 barrels produced by 2,040, this would then be followed by a more rapid decline. So this is purely a timing issue here rather than telling us anything about the underlying resource. Alternatively, if resources were more plentiful, such that total cumulative production is, say, around 50% higher than in the ET scenario, shown here in this greater resource case, this would allow U. S. Tight oil to potentially grow to around 15,000,000 barrels a day by 2,030 and remain around that level for the rest of the outlook.
One point to note here is this chart on the left, these profiles for the 1st period of the scenario are all pretty similar. And as a result of which, it may take some time before we get a clear sense of actually what path we're on. Is Tidore growing rapidly because lots of finance is available? Or is Tidore growing rapidly because the underlying resource base is very strong? We won't know for a while.
That's all I want to say on oil demand and its implications for supply. If we turn next to the question of just how fast will renewables grow, the starting point here is that we, along I think with almost all other forecasters, have been repeatedly surprised by the strength of renewables in recent years, consistently increasing our outlook. To give you one example, although they're not directly comparable, the ET scenario if I compare the ET scenario with the base case in the 2015 energy outlook, solar energy in 2,035 is projected to be over 150% higher as solar costs have fallen far faster than anticipated. Some of that rapid decline in costs stems directly from the pace of technological progress. Some also reflects the interaction with stronger policy support, particularly in China and India, which account for more than half of that upward revision, enabling solar energy to move more quickly down the learning curve.
So have we learned our lesson? Renewable Energy is certainly projected to grow rapidly. In the ET scenario, renewables in power are the fastest growing energy source, accounting for over 50% of the increase in power generation, with their share in global power increasing from 7% today to around 25% by 2,040 and growth becoming increasingly broad based with China and other parts of the developing world taking over from the EU as a main engine of growth. In the ET scenarios, renewables gained share in the power sector faster than any energy source in history, the closest parallel being the rapid buildup of nuclear power in the '70s '80s. But despite that, there must be a chance that renewables continue to surprise on the upside.
In the ET scenario, subsidies are gradually phased out by the mid-2020s, continuing the recent trend we have seen with renewable energy in most parts of the world increasingly able to compete against other fuels in the power sector. But what would happen, say, if rather than those subsidies being gradually phased out, government support remained around current levels for the entire outlook? What would happen then? This is shown here in this so called renewable push scenario. This extra policy support means renewables account for more than 90% of the growth in Global Power over the outlook, with the share of renewables within power reaching over percent by 2,040 compared with that 25% I just told you about in the ET scenario.
So plenty of scope for policy to continue to surprise us to the upside. It's interesting to note that although this policy support reduces carbon intensity in the power sector, the reduction is only around half that achieved in the even faster transition scenario. So this chart shows reductions in carbon intensity. This is an evolving transition scenario. This is the greater reduction in the renewables push and this is even greater reduction in that even faster transition scenario.
And this smaller impact in the renewables policy here is symptomatic of the diminishing effectiveness of policies focused solely on encouraging renewables, reflecting the increasing cost and difficulty of dealing with intermittency issues as a share of renewables increases. I think this gets exactly to the point that Bob was talking a moment ago about racing for renewables. In contrast, the high carbon prices in the even faster transition scenario as well as supporting renewables also encouraged increased coal to gas switching and the large scale deployment of carbon capture and storage and so lead to a far more pronounced fall in carbon intensity. I think, Bob, this is exactly the point you're doing. Let's not raise renewables.
Let's try and reduce carbon intensity. There are different ways of doing that and more effective ways of doing that. So are we going to continue to be surprised by renewables growth? I hope we've learned our lesson, but I think we should recognize that the outlook for renewables is unusually uncertain. We include some analysis in our main booklet, which you can look at afterwards comparing our outlook with other external forecasts.
The growth of renewables in the ET scenario is towards the top end of that of the external forecast. So we may draw some comfort from that. But it's striking that the dispersion of views across external outlooks is greater for renewables than for any other energy source. This is an unusually uncertain area depending on both the pace of technological progress and the form and persistence of government support. Watch this space.
We turn next to the resilience of natural gas demand. One feature of the energy transition, which almost all commentators agree with in one form or another, is that gas demand is likely to grow pretty robustly over the next few years and certainly more quickly than either oil or coal. Given that broad consensus, if you've got a job like mine, the obvious question to ask is, how could that be wrong? The ET scenario certainly conforms to the consensus view. Natural gas demand is projected to grow robustly by around 1.6% per year, reflecting broad based support across both sectors and regions, helped by growing supplies of liquefied natural gas, LNG, increasing the availability of gas around the globe.
Natural gas accounts around onethree of the entire increase in global energy demand in the ET scenario. In terms of how this could be wrong, rather than thinking about growth of gas demand in terms of sectors or regions, it's possible to separate the growth profile into 2 alternative components. First, growth stemming from gas gaining share relative to coal and also relative to oil in transport, which I've called switching and growth caused by other effects, mainly economic growth. As you can see, very roughly, around half of the growth in gas demand in the ET scenario stems from these switching effects, gas gaining share relative to coal and half due to these other sources of growth. In terms of this switching effects, some is driven by simple economics, in particular, the increasing availability of low cost gas in the U.
S. And the Middle East, allowing gas to gain share relative to coal. But some of this switching is driven by policy measures, promoting a shift to a lower carbon fuel mix, especially in Asia and the EU. One potential risk to the outlook for gas is that these policy measures fail to materialize, causing coal demand to be stickier, limiting the scope for gas to grow. To get a potential to get a sense of the potential impact of that risk, consider a scenario in which there is no coal to gas switching at all in Asia and the EU, the 2 key regions which this policy induced switching is most pronounced, as well as virtually no oil to gas switching in transport anywhere in the world.
The growth of gas in this less gas switching scenario is around 1.1%. So around a third slower than in the ET scenario, but still pretty robust relative to the outlook for either oil or coal. Switching to the other end of the policy spectrum, another downside risk to gas is that climate policies rather than tightening by less than expected, tightened by more. For example, in that renewables push scenario that we just talked we just thought where government support for renewables stayed a lot higher, the stronger growth in renewables crowds out natural gas from the power sector. So it's our annual gas of growth in gas demand in the renewable push scenario also slows to around 1% shown here by the lilac bar.
Gas demand is more subdued in the faster transition and even faster transition scenarios, reflecting the impact of a more comprehensive set of climate policies, leading to significant improvements in energy efficiency as well as providing strong support for renewables. But as you can see here, even in the even faster transition scenario, the level of gas consumed in 2,040 is pretty much in line with the level of gas consumed today. So in terms of that question in terms of the resilience of natural gas, the outlook for gas is exposed to the possibility of climate policies being both less or more stringent than assumed in the ET scenario. The growth profile looks relatively resilient to scenarios in which the policy surprise is restricted to particular fuels, such as less coal to gas switching or greater support for renewables. It's more vulnerable to comprehensive set of climate policies, encouraging greater energy efficiency as well as a switching to renewables.
But even in that even faster transition case, gas demand in 2,040 is similar to current levels. So turning to my 5th and final question, is the transition to a lower carbon energy system happening fast enough? Based on the ET scenario, the clear answer is no. Carbon emissions are projected to increase by around 10% by 2,040. Although this growth is far slower than in the past, it falls well short of the almost 50% decline in carbon emissions thought necessary to achieve the Paris climate goals, shown here by the even faster transition scenario in orange, which follows the same fall in carbon emissions as the IEA's sustainable development scenario.
To remind you, the ET scenario is designed to capture the broad path the global energy system might evolve along if government policies, new technology and social preferences continue to change in a manner and speed seen over the recent past. The clear message from the energy outlook is that to achieve a sharp reduction in carbon emissions, we need a far more decisive break from the past than recent momentum in policy and technology implies. The EFT scenario illustrates one possible configuration of policies and outcomes that brings about such a fall in carbon emissions. With a sharp rise in carbon prices operating within the power sector and increasing regulation incentivizing more rapid gains in energy efficiency and fuel switching in Industry, Transport and Buildings. Now I think it's fair to say we're getting close to the limits of our modeling capabilities with scenarios like this.
So I wouldn't want to use this scenario to provide advice about the precise policy mix for bringing about such a transition. Rather, the focus and the outlook is more on what impact such a transition might have on the energy system. Global energy demand continues to grow, but less quickly than in the ET scenario, reflecting more rapid improvements in energy efficiency. As you can see here on the chart on the left, much of the additional abatement carbon abatement in the EFT scenario takes place in the power sector. The key role played by the power sector in reducing carbon emissions is a pretty common feature across other external scenarios with similar falls in carbon emissions.
And this reflects the potential effectiveness of carbon pricing in the power sector where fuels compete side by side against each other. And so comparatively small changes in relative prices can have a big impact on the fuel mix. A great example of this is the example of the carbon price floor introduced here in the U. K. On reducing coal fired power generation in recent years.
This relates to that turning the dial on carbon emissions I've mentioned earlier. My general approach when talking to officials or politicians is if they really want to make a significant difference to the carbon outlook, we should start by talking about the power sector, then we should talk about the power sector, and if we have time, we should talk a little bit more about the power sector. If you're having conversations with politicians or officials and they want to do this and they're not talking about the power sector, ask them why. In terms of the fuel mix, the biggest gainer is renewable energy, which more than accounts for the entire growth in global energy in the EFT scenario, with its share increasing to around a third by 2,040. But if renewables account for around a third of the world's energy needs by 2,040, it suggests some other forms of energy needs to account for the other twothree, even in the case of that much faster transition.
The majority of this additional the remaining 2 thirds is provided by oil and gas shown together here shown by the green and red bars to the right here, which together provide a little over 40% of the world's energy in 2,040. This is less than the 55% or so implied by the ET scenario, but suggests that oil and gas are likely to continue to play a material role in the global energy system over the next 20 or 30 years, even in a transition path consistent with the Paris climate goals. I've gone on miles too long. Let me stop and conclude and then throw it open to the questions. And please, for those of you remaining on the web, please do send in your questions.
The Global Energy System is in transition. Before we throw up our arms in a sort of despair of uncertainty, I think there are some aspects of this transition which we are relatively confident about. Growth in energy demand is likely to be driven by increasing prosperity in fast growing developing economies. That demand growth is likely to slow as we learn to do more with less. Global energy supplies are likely to be characterized by increasing abundance and diversification.
Renewable Energy is likely to play an increasingly important role in global energy. Nevertheless, oil and gas are set to continue to play essential role in the global energy system at least for the next 20 or 30 years. Those features seem pretty robust across many scenarios. But there are other aspects of the energy transition which remain very uncertain. What impact will electric cars and autonomy play in reshaping the transport sector over the next 20 or 30 years?
What role will natural gas and renewables play in the transition to a lower carbon energy system? And most important of all, will we achieve the decisive break from the past necessary to bring about a sharp reduction in carbon emissions? There are no simple answers to these questions. All forecasts are wrong. But hopefully, this year's energy outlook and the many outlooks to come will help us gradually to understand better the energy we face the uncertainty we face.
Thank you.
Make an even faster transition to the seats. Well, thank you. Thank you very much, Spencer. And thank you all very much here in London and those of you who stayed with us on the lines from around the world. I found this year's presentation, 1, you have to pay attention to as much as any year.
You've introduced new things like ICE Bands and VKMs that are like a language we all maybe need to learn. But I want to thank you. We've got we have I'm not sure how much time. I think we have about 30 minutes or so. We've got already a lot of questions here.
There's some really, really good ones here. I'm seeing them for the first time. I'm going to start though here in the room in London with 2 here, then I'm going to go here, and then I'll go over here and here. So first two from London, if you could just mention what organization you're from.
Hello. My name is Nora Kalinski. I'm from the Cambridge Forum on Geopolitics. I have a question pertaining to Russia, Russian companies like Gazprom and Rosneft. So Russia has been attempting to assert itself more and more in the European market and certainly with projects of Turkish Stream and Nord Stream 2, Russia's presence in Europe is likely to increase.
Russia is also asserting itself for attempting to do so in the Middle East. So my question is, what challenges may this assertion of Russian companies in Europe and the Middle East post to BP? And how may these challenges be turned into opportunities? Thank you.
Thank you. Well, first, I think the outlook here shows that the world needs all forms of energy. I mean, that's just clear. And there's going to be an increasingly competitive world and resource producers will want to find their markets. And I think that's what we're seeing here with many, many countries and many companies around the world, not just Russia.
So I think the other thing that Spencer's energy outlook talked about is the role of gas. I mean, it to me has to be part of the transition, it's not a destination fuel for reducing carbon emissions. So getting natural gas into the market, I think, is broadly good for the world. So I'll stay away from the geopolitical comments about it because I think there's all kinds of pressures around the world everywhere in geopolitics right now. But I so I'm going to turn to the next question and about the energy outlook.
It's Morten Froesch from MFC. First of all, Spencer, thank you very much for yet another brilliant presentation, this time with much more detail than in previous years with all these scenarios. You mentioned that you had 5 questions. There was a 6th question, which I was missing. In this now transition period, and we've got now the next 3 to 5 years, the formative period of that transition period.
And looking, for example, at North America, where you touched upon this, you talked about finances to be available for drilling. But you also need a lot of money in North America for a midstream infrastructure, which needs to be built quickly. Then to China, the switch from coal to gas, again, what we have seen this winter is going to cost large very large amounts of money. And again, India also big transitions. To what extent have you considered availability of capital to pay for this?
Have you put in any capital constraints in any of these scenarios?
So the technical answer in a modeling sense is no. So the technical way we do this is we try and work out what is the economic efficient outcomes of different scenarios, and then you assume that capital will flow to make sure that occurs. Now that's clearly a risk. And I think that was one of the risks which underpinned the scenario I was talking about where climate policies it's tightened by less than people expect. When if you think about the amazing transformation which is happening in China at the moment, there's a significant shift away from coal towards other cleaner forms of policy.
The reason why coal today in China is a dominant source of energy in China is because it's cheap. If you try if you encourage you to move away from that fuel to other sources of fuel, that has many benefits, but it means you're asking your households and your families and your citizens and your industries to use a more expensive form of energy. And that acts as a drag on growth. And so there's a challenge and a balancing act that the Chinese authorities and many authorities around the world need to balance in terms of the longer term benefits in terms of shifting to a lean lower carbon fuel mix and the near term fiscal imperative of trying to use energy as cheaply as possible. And I think I don't we don't try and sort of quantify that calculus, but what we try and that was one of the sort of underlying reasons why I think it's conceivable.
It's quite possible and I'm hopeful that climate policies are stronger than assumed in the evolving transition scenario. A clear message from that was we need climate policies being stronger than that if we're going to have any chance of getting carbon emissions falling sharply. But I also think there's a risk that they could be they could be tightened by less than we expect because of that sort of fiscal implications it has.
Steph, I'll go to a couple of questions here. I'll go here in a second. A footnote on that, Spencer, someone is I can't verify the accuracy there, but I think it's not far off. Someone said that in addition to capital constraints, there are other issues that governments need to plan on. And I think the figure in Europe is $240,000,000,000 a year is collected in fuel excise tax.
And so the switching away from liquid fuels to EVs, there's a whole set of policy that needs to be thought through on the cost of that as well for governments. I don't have an answer on that. Do you have any?
No, I think my ICE band scenario may be very popular in some quarters. It may not be other so popular in other parts of government when you do think about the tax implications. I mean, I think in some sense, you have to see these things in the round and understanding the fiscal implications and where your sources of taxation will come from if you start to radically reduce internal combustion petrol cars is one part of that, I think. Yes, absolutely.
Now can I put you on the spot here with these fascinating questions? It's come from the U. S, Renewable Energy Magazine. How has the United States pulling out of the Paris Climate Accord influenced your 2018 outlook?
Yes, it's always a good question. So when you think about the profile for carbon emissions in the Energy outlook, most of the really big significant falls in carbon emissions are coming from outside the OECD. They're coming, in particular, from China, other developing economies. So that's in terms of quantitatively, a lot of the actions happening from outside the OECD. Even within the U.
S, if you look about what the key drivers of the U. S. In terms of the big falls in carbon emissions in the U. S. Over the last few years, that's largely been driven by economics, not policy.
That's the availability of low cost gas causing huge falls in coal consumption with gas gaining share in the power sector. So my instinct is that a shift in administration doesn't directly affect those types of mechanisms hugely. It seems to me what is significant is when I sort of observed the Paris process from the outside, what was very significant was the leadership role that the U. S, together with China, played in bringing everybody around the table with a desire to try to reach an agreement. And I think the most worrying thing for me is whether if the U.
S. No longer plays that leadership role, will we still have that momentum? Because remember, Paris, if achieved anything, was a first step, only a first step towards a road to a more sustainable outcome. We need to make many more steps along the way. I don't think I think it's too early for us to build in a significant shift in that sort of political momentum, and we haven't built in a significant shift in the political momentum.
But I think that's where the issue here is, less to do with direct policies in individual countries, but more whether that political leadership role and momentum fails if you haven't got such a strong support from the U. S.
Your charts did show a continuing reduction in emissions in North America
moving forward? Yes, indeed. And I think, again, that's driven because with gas, the increasing role of gas within U. S. Power sector.
I think it's I think I'm right in saying that U. S. Coal consumption has fallen by around 20% or so over the last 2 years. The falls in U. S.
Coal consumption have been as big, if not bigger, than we've seen in China, And that's directly with coal and coal and gas switching as the as just the availability of cheap gas has gained share in the power sector.
One more question from abroad. Because we didn't talk about this very much and around the world policymakers don't talk about this enough or I think fully understand the positive implications, but it has a high cost CCUS. What is the future of CCUS? Will the price of carbon ever be high enough to make CCS or CCUS viable?
I certainly hope so. In that even faster transition scenario, that was tried to sort of say what's an efficient way of bringing about that still significant reduction in carbon. And in that world, carbon capture and storage or carbon capture use and storage plays a material role. In the power sector, but I think increasingly as technology evolves also within industry as well. Are we is that evolving is it playing a big role in the evolving transition scenario?
No, it's not. But that's in a world where carbon emissions are beginning continuing to rise. So I think I would like to see a more significant role for CCE S because I think that will be a good a fairly good indicator. We're starting to get on the right path of carbon emissions falling very sharply. But at the moment, the level of carbon prices we observe around the world are just nowhere near enough in order to incentivize that.
Hence, we're just not on the right path. And that sort of, if you like, is 1 is symptomatic of this evolving transition path not being on the right path consistent with getting carbon emissions down to the levels that we all want to in terms of achieving those Paris climate goals.
Okay. Back into the room here in London. So 1, 2, 3, and then we'll come back in a second. So right here in the second row.
I'm Hughley, I'm a consultant in the coal industry. My question was going to follow on from that one anyhow about CCS. Given the enormous expertise that BP has in the technologies that are needed for CCS or parts of the CCS chain, Is BP willing to have a break with the past and invest in CCS now even when it doesn't look economic because it's got to be economic for us to save the planet?
Well, I'll mention that BP had 4 CCS projects at scale. We did have to step back from that after the Gulf of Mexico. And we have had a a capital framework that has been quite tight, as you'll know, now for 8 years. So we have the expertise in the company. We do do some CCS type projects with injection of CO2 back into reservoirs.
We can do more on this, but it's simply been we've had to redirect our efforts. And I know that there's some big projects in Norway. We've had I think, government in the U. K. Is looking at some big projects in the U.
K, and we'll see. That's probably all I can commit right now.
And Bob, am I right in also saying the OGCI, which BP is part of, has also identified CCS as one possible area which you may well invest in as well?
That's right. The OGCI is the oil and gas climate initiative of 10 companies has put together and started the beginning of a $1,000,000,000 fund jointly with the companies take over time. We want to invest it wisely and well at all kinds of CCS, including even cement companies, which use a lot less CO2 than the current processes around CO2 or cement emits, as you know, an enormous amount of CO2. So it's investing in a broad set of things that we're putting our technologists on. 4th row and then back and then across to the 3rd row.
Rohit, hello. Thank you for very interesting set of scenarios. In constructing the scenarios, can I ask about 3 other factors? One was you assumed continued economic growth. So do some of your scenarios factor in the potential for economic volatility?
Secondly, you talked about the phasing out of vehicles by 2,040 or ICE vehicles. Did any of the scenarios look at an earlier timescale, say 2,030 with all ICE vehicles out by 2,040? And finally, did any of the scenarios look at the potential for severe climate change disruption in that period and what that might do to both energy demand and the energy mix?
So the short answer is no. But they're all perfectly good ones. And in some sense, that first and third go together in some sense. So in previous scenarios, we have looked at the implications of slower growth in global economic growth. And as I was saying here, I mean, the key factor there is if productivity growth in China and India doesn't grow like we expect it to do, that would have a huge impact on global economic growth, and that will then feed through to energy demand.
And we've studied that in the past, and it's sort of there. And I think that sort of interaction with that third dimension there is if you start to build in sort of climatic changes affecting economic growth, then that 1st and third can do it. Now we haven't done that. We have done economic growth in the past and just sort of tried to see it. And what's interesting there is some of the impacts it has on the fuel mix, because sometimes slower economic growth will actually start to squeeze out the growth in renewables, partly because of the merit ordering, but partly also what we have found in the past.
When economies are struggling, some of those fiscal pressures we were talking about earlier really start to bite. And that sort of that need that if you like, the ability to switch away from a high cost fuel to lower cost fuel becomes less easy. And so you bite often on things like renewables. I haven't didn't look at a nice ban where we buy by 2,030. The one we've done here is pretty extreme.
If you start saying, well, let's just get to the point where there are just no ICE vehicles by 2,040 at all, that's pretty significant. One is you would need to start ramping up the sales of electric cars now, very rapidly now. So they're starting to account for very big shares really rapidly now. Now the one aspect of that is there's a significant cost to that. Electric cars cost an awful lot more money than internal combustion engine cars.
Ultimately, in almost any economic model I know, the ultimate burden of that cost will fall on households and families. Sometimes people say, Well, I'm going to impose a tax on companies. Sometimes governments do it. Ultimately, the burden always ends up falling on citizens. And so that will be a significant cost to citizens.
There's also a significant implication in terms of infrastructure costs, in terms of the ability to do that. Now most people in this room, when I say infrastructure costs, will be saying, well, if I start driving my car, if I get to a motorway, will it be a fast charger? Yes, that's one problem. The problem faced by a Bangladesh family in the middle of the countryside is something completely different and an order of magnitude more important and more significant than whether you have to wait for 20 minutes or half an hour to recharge your car. So there are some significant limits about how quickly you can do this.
I think the ice band we thought about was pretty extreme. Doing it more, I think if you want to think about doing it more, you've got to think about that cost burden and the implications that has not for people in the infrastructure implications in rich Western countries, but in some of those developing countries if you're going to do a worldwide ban, which is sort of what we've done here.
Okay. Mike, in the 3rd row and then the 2nd row, and I'll come back over here in a second.
Yes. Thanks, Bob. Bob, I was wondering it's Michael Ritchie from Energy Intelligence. Bob, I was wondering if you could expand a little on why you think gas will grow so fast over the next 2 decades to overtake coal and converge with oil by 2,040? And how does that how is that influencing BP's strategy going forward?
So I think it was similar to the stories we've had in this. I think there's 3 or 4 things which are causing underpinning this strong growth in gas demand. One is there's broad based support across sectors. So we see strong growth of gas, both in the power sector and the industrial sector. So the world continues to industrialize, particularly in the developing world and increasing electrification of the world.
They provide 2 sources of growth for gas demand. Also, it's pretty broad based across regions as well. So you see increasing growth of gas demand in the OECD as well as the non OECD. That's pretty rare for a fuel. A second significant issue here is the growth of LNG.
As the growth of LNG in Carizon or LNG Carizon increasing around the world, that's increasing the availability of gas around the world and so allows a sort of latent demand in some important demand centers to be met, which other which wouldn't be able to be met anyway as easily or as cheaply? And third, is this sort of environmental and climate pressures encouraging a switch away from coal towards cleaner, lower carbon fuels, natural gas being a beneficiary of that as well as renewables. And we showed you that switching effects can account for around half of the growth of natural gas. Now not all of that coming via policy. Some of it's economics, some of it's policy.
But I think policy also is playing a role here. So there's a number of factors, I think, underlying the support for natural gas. Why? Which is why that sort of that broad consensus is people expect a pretty strong growth of natural gas over the next 20 or 30
years. And while we're at a footnote, a very similar question came in from the UAE. Basically, it's our outlook, consistent with what you said, LNG supplies will more than double going out to 2,040. And most of the LNG will come from North America and the Middle East, with Asia and Europe being the key markets that will be the buyers for it. So second row and right here and then over here, here and then we'll go back to another set of good questions.
Hi, I'm Anna Huttenstein from Bloomberg. There's been a lot of interesting points made and changes made to this year's outlook, everything from oil demand slowing faster to revising up EV uptake quite significantly. So my question is how is this adjusting your company strategy?
Well, in many ways, first, you will see us, we continue to go gas production. We've been growing it for some time now. This last year we've had 7 major projects, 6 of them are gas. We have 6 major projects coming on this year, 5 of those will be gas. We, of course, do what we do well with advantaged oil production.
So we're also, not just because of this year's, but over time, we're continuing to develop business models. But what we see is this what will be the rapid transition to lower carbon energy. So we're spending a lot of time, team's done a great piece of work on the strategy of developing business models, making sure our technologists and our scientists are involved in many, many types of different joint ventures. We can understand, be involved, be ready and make bigger investments when we're ready. We have historically invested a lot of money in renewable energy.
And to this day, we have 2 big renewable businesses in wind in North America and biofuels in Brazil. But we'll be ready at the right time. Number one strategy for us as a company must be get our balance sheet strong, give us the flexibility and at a point which we're ready and we really know where to invest big, we'll be ready to do that with big bets. And right now, we're not quite there. Yes.
Hiran Mochandani from Aurora Energy Research. Congratulations on a great product, Spencer, and thanks for giving us opportunity to work with you guys on the power sector modeling. In the internal discussions we had at Aurora for the scenarios that we ran with you, there was some talk of the different trajectories for the carbon prices in all the various different runs we did. I guess my question now is what is BP's view on the likelihood of a global carbon price convergence? After all the scenarios and the talks and discussions we've had, what
did you guys land on
in the end? Thank you.
So thank you. So for those of you who aren't familiar with Aurora, I think Aurora based in Oxford or one of the sort of I think the smartest energy consultants consultancies around. If you want to do some serious modeling, far more serious that we can do, we go to Aurora. So thank you for Aurora helping us again this year. If one could a is.
You impose a global carbon price. And there may have some other bits as well, but a global carbon price would be a significant element of any sort of sensible economic solution. The world isn't always jumped to the most sufficient sensible sort of economic solution. I think my instinct would be, in the first instance, different regions and countries imposing meaningful carbon prices would be a good first step, and then you can find different ways to make sure that you can deal with those carbon leakage issues. In the outlook, we have carbon prices rising in some parts of the world to about $40 real.
The message from the energy outlook is $40 carbon per ton is nowhere near high enough to get us onto that sort of even faster transition scenario. And you need to be having carbon prices of $100 or more per ton before we start getting serious implications. Dollars 100 per ton would be just 2 points: a) very would now would then allow CCS to play a far bigger role. I should also say, dollars 100 per ton would be relatively cheap relative to the carbon abate cost of carbon abatement associated with most electric cars at the moment. So it's a high price in one level, but nowhere near highest price in some in another dimension that we're facing at the moment.
Thank you. Lia?
Thanks. It's Lydia Rainforth from Barclays. Just coming back to the vehicle kilometer miles on the ICE engine side, what's the relationship between the take over BVs and the autonomous sharing thing? Is it a case of if you go from that 2,500,000 barrels a day to 5,000,000 barrels a day impact on the EVs, does the autonomous impact go from 2 to 4? Is it that direct relationship or does it impact sales at some point on the autonomous side?
So what we have done for simplicity in this case is we kept that intensity effect just fixed in the ICE band just because we were getting to levels where it was too much. My hunch is, if anything, the effect from the autonomy should come down a little bit. So the intuition for this being essentially that intensity effect works by squeezing out miles driven by internal combustion engine cars. If there's less and less internal combustion engine cars on the road, there's less oil for you to squeeze out. So and my hunch would be, if anything, it would come down.
So the overall fall in oil demand would be even less than the 10,000,000 barrels a day. I was very conscious when we were doing the I SPAN that I didn't want to look like I was doing anything to make it look in any way smaller. People would always say typical BP, they've done some trick or some stick over their sleeve. So what I tried to do was do everything as it was as sort of clean as possible. So the impact of the ICE ban has started to work straight away.
The shift in vehicle efficiencies allowed all of the impact to go through. The impact all the additional electricity was powered by renewables, so there's no abatement. So I was trying to do everything as absolutely as clean as possible, and I also kept the intensity effect there as well. So all if anything, that 10,000,000 barrels a day is all an upper bound in terms of the impact on oil demand. But I thought that was the sort of right way of presenting the scenario.
So, Spencer, our masters of the web have said one more question. We have a lot of questions from around the world. I'm going to put one question to you because I think we have shown in orange renewables is a big block. So we have a question from Emilio Velarde in Peru. Of the renewables, which is the one driving the biggest impact towards 2,040, solar, wind, biofuels, title, what is it?
So in absolute volumes, it's pretty much identical between solar and wind in this year's energy outlook. In growth rates, solar is growing far more quickly than wind, but it's starting at a smaller level. But in absolute times, in absolute, they're more or less they're both in volume terms trying to do producing the same amount. I think in terms of that pace of growth going forward, we have and we've talked about in the past, we have wind solar catching up with wind pretty quickly just because the pace of technological change and the steepness of that of the learning curve for solar looks quite a bit steeper than that for wind. So at this scenario in the evolving transition scenario, pretty similar in volume terms.
But in growth rates, solar is the one which is really rocketing ahead and it's solar the one where we've been most surprised by the sort of the pace of growth over the last few years.
So ladies and gentlemen, thank you all very much. Thank you all who have joined us from all around the world. Very patient. It's remarkable the number of people that have been in contact with us around the world. Thank you all for spending your afternoon with us here in London.
And please take a copy of the book as you go out, if you still read and do it in paper, I do. But thank you again. Spencer, again, thank you to your team.