Good afternoon, everyone. Good morning, good evening, depending where you're joining us today. I'm Spencer Dale, BP's Chief Economist, and thank you all so much for sparing the time. So here at BP's HQ in St. James's, London, we have a pretty full house with lots of familiar faces. It's good to see everybody again, and virtually around the web, where I think we have upwards of about 20,000 people joining today's cast. So you're all incredibly welcome, and thank you again for sparing the time to join us for the launch of this year's Outlook. While I'm sharing it up, let me also give a big shout-out to the rest of the team that have been really busy over the next few months, over the last few months preparing this year's Outlook. So thank you very much, guys. Busy time, but also good fun. So thank you very much.
This is actually the 13th BP Energy Outlook. The first one was published in 2011, complete with a picture of the very youthful-looking Bob Dudley. He looks fantastic, doesn't he? Very youthful. That 2011 Outlook considered the prospects for energy markets out to 2030. Looking back, it's striking that some of the profiles and analysis in that first Outlook are quite close to today's. That's especially true for developments in mature, well-established fuels: oil, natural gas, hydropower. But other developments in the energy system, especially those related to new and emerging energies and technologies, weren't really on the radar in 2011. The exponential growth in wind and solar power, ditto supplies of US tight oil and natural gas, the electrification of transportation. Perhaps a lesson here is that these types of Outlook can provide valuable insights.
The inertia of the global energy system, with long lead times and even longer-lived assets, means it's possible to identify some key trends which will shape how we shape energy markets over the next 10 to 15 years. But we also need a big dose of humility. The one thing we know for sure is that over the next 25 years, the energy system will change in ways that we can't even imagine today, which, for anyone producing an Outlook like this, is both scary and exciting. That first energy Outlook was written against a backdrop of growing awareness of the challenges posed by climate change. It had a section asking what was needed to bend the trend in emissions, a question many people are still asking today.
The COP, the 2011 COP, COP17, which was held in Durban, South Africa, established a so-called Durban Platform for Enhanced Action, which ultimately led to the historic Paris Agreement 4 years later. 2011 was also the year of the Fukushima tragedy. That disaster, perhaps more than any other world event before or since, demonstrated the value of globally integrated energy markets. Cargoes of liquefied natural gas, LNG, diverted from Europe to help meet the massive shortfall of Japanese energy as its nuclear power fleet was closed down. Fast forward to today. Those same 2 overarching themes, energy sustainability and energy security, remain huge issues shaping the energy landscape. Despite the hope and ambition at the time, carbon emissions have risen in every year since the Paris Climate Goals were agreed, other than the COVID-induced drop in 2020. The carbon budget is running out.
The fallout from the war in Ukraine means the importance of energy security has once again come to the fore. In sharp contrast to 2011, the 2024 version of energy security highlights the dangers associated with globally integrated energy markets and dependence on imports. This updated version of energy security instead stresses the benefits of increasing reliance on domestically produced energy, or at the very least, on obtaining energy from resilient, trusted, diversified resources. I know the concept of the energy trilemma, the importance of energy systems providing energy which is secure and affordable, as well as sustainable, may feel a little stale, a little out of date. I would argue its relevance has never been greater. Any successful and enduring transition needs to address all three elements of the trilemma. Those three needs play a central role in informing and shaping this year's Energy Outlook.
This is our plan for the next 40 minutes or so. I will start by briefly describing the two scenarios which provide the basis of this year's Outlook. I will then use the scenarios to identify two types of developments or trends in the energy system out to 2050, trends which are common across both scenarios and those which are more differentiated depending on the pace of transition, and I'll explain why I think this distinction is interesting when we get there. I will then discuss the implications of delaying a shift to a faster decarbonization pathway and conclude by using the scenarios to try to draw up an action plan of key initiatives which could help achieve a shift to an accelerated transition.
At that point, you'll be pleased to know I will stop bombarding you with charts and numbers, and we can switch to questions and answers, which is always the most interesting bit. For those of you online, please feel free to submit your questions at any point during the presentation. For those of you here, a little bit more restraint, please, until we get to the questions and answer session. For those joining online, we'll also invite you to take part in a real-time poll on some of the issues raised by this year's Outlook. So in short, we have an action-packed 70 minutes or so ahead of us, so please don't go anywhere. Don't go off to make a cup of tea. Don't check your emails. Stay with us, and you won't miss anything.
So, first, in terms of the scenarios we're going to discuss, this year's Outlook uses two main scenarios: Current Trajectory, shown here in green, and Net Zero in blue. Current Trajectory, as the name suggests, tries to capture the broad pathway along which the current global energy system is currently traveling. It places weight on climate policies already in force and on governments' aims and pledges for future decarbonization. At the same time, it places weight on the difficulty of actually meeting those aims and pledges. As you can see here, carbon emissions stop rising in the middle of this decade, but the pace of the decarbonization afterwards is pretty shallow, such that by 2050, carbon emissions are only about 20% below their current level. In contrast, Net Zero explores how different elements of the energy system might change to achieve a faster, sustained reduction in carbon emissions.
In that sense, Net Zero can be thought of as a what-if scenario. What elements of the energy system might change if the world collectively acts to reduce emissions by around 95% or so by 2050? Net zero assumes a significant tightening in climate policies. It also embodies shifts in societal behavior and preferences, which further supports gains in energy efficiency and the adoption of low-carbon energy. Now, it's not possible to directly infer the consistency of Current Trajectory and Net Zero with the Paris Climate Goals, not least because the Paris Goals refer to rises in average global temperatures in 2100, and the scenarios only extend out to 2050. But it is possible to make an indirect inference by comparing the cumulative carbon emissions in the two scenarios with the corresponding carbon trajectories taken from the IPCC Sixth Assessment Report, and that's what's shown in this next chart.
So you can see here, the cumulative carbon emissions in Net Zero shown here are broadly in line with the IPCC C2 and C3 scenarios. That C2 scenarios are judged to be consistent with a greater than 50% probability of returning global warming to 1.5 degrees after a high overshoot, and the C3 scenarios with a 67% probability of limiting average global temperature rises to 2 degrees. Now, neither of these categories are directly comparable to the Paris Climate Goals of trying to limit global temperature rises to well below 2 degrees and pursue efforts to limit temperatures to 1.5. But the correspondence of Net Zero with the IPCC C2 and C3 scenarios means I think it might be considered you can think of Net Zero as being broadly consistent with the Paris Goals.
In contrast, Current Trajectory, as shown here, relies pretty much in line with the IPCC C5 scenarios, which refer to a greater than 50% chance of limiting global temperature rises to two and a half degrees. That suggests that Current Trajectory, which remember is one in which the world continues along a pathway similar to the current one, is not consistent with the Paris Climate Goals. If we go back to the two scenarios, and apologies to those who've heard me say this a thousand times in the past, but I'm going to keep on saying it, both of these scenarios will be wrong. We can't predict the future. We know we can't predict the future.
Rather, the two scenarios simply provide a rough sense of how the global energy system might evolve under different assumptions about the future: one in which the global energy system continues along its current pathway, and another in which the world takes actions to drive emissions to Net Zero by 2050. The chances of the energy system evolving exactly in line with either of these scenarios is virtually nil. So why bother? Why are these scenarios useful or interesting? I think there are a couple of reasons. First, although we could see transitions more or less rapid than these scenarios, taken together, Current Trajectory and Net Zero span a wide range of the possible outcomes facing the global energy system. And as such, they can help inform a strategy which is resilient to a wide range of the uncertainty that we face.
Second, we can identify trends or developments in the energy system which are common across both scenarios and those which differ significantly. I think this distinction between common trends and differentiated trends is potentially helpful when trying to assess the confidence we might have to different trends or features of the energy system materializing. In particular, if you think about it, if the same broad trends are apparent both in a pathway akin to the one the world is currently on and on one in which the world rapidly decarbonizes, it may suggest those same trends will also be seen in a whole range of other pathways which lie somewhere in between those two big scenarios. In contrast, if other elements of the energy system are very different in Net Zero to Current Trajectory, it suggests they are far more dependent on the pace of transition.
That's what we're going to look at next, focusing first on the common trends, future developments in the energy system which seem less dependent on the speed of transition. The first common trend I want to highlight concerns a historic challenge facing the global energy system. As you know, low-carbon energy, shown here in orange, has increased significantly in recent years, boosted in particular by wind and solar power, which has almost doubled since 2018, accounting for a third of the growth in primary energy, causing emissions to rise less quickly. But low-carbon energy is not yet growing sufficiently quickly to keep pace with increases in overall energy demand. As a result, the consumption of unabated fossil fuels, fossil fuels used without capturing and storing their emissions, is continuing to increase alongside the growth in low-carbon energy. You can see that in the gray bars here.
The world is in what might be called an energy addition phase of the energy transition, in which both types of energy, old and new, unabated fossil fuels, low-carbon energy, are both growing. We see previous examples of energy additions in history, in the mid-19th century when coal grew rapidly and supplanted wood and other biomass as the world's primary energy source, and roughly 100 years later when the rapid growth of oil meant it overtook coal as a dominant energy form. In both those previous periods, the world remained in the energy addition phase, consuming similar or growing amounts of the old energy, even as it adopted the new. The world has never reduced its consumption of any fuel on a sustained basis. It's just consumed more of everything. That's the historic challenge facing the global energy system.
If the world wants carbon emissions to fall, it can't remain in the energy addition phase indefinitely. The energy system needs to move from energy addition to energy substitution, in which the growth of new energy, in this case low-carbon energy, exceeds the increase in total energy demand, causing the old energy, unabated fossil fuels, to decline. The common trend is that this historic shift from energy addition to energy substitution happens in both scenarios, albeit with different timing and intensity. In Current Trajectory, the energy addition phase continues in the 2020s. You can see in the 2020s as a whole, the growth of low-carbon energy increases even more, but still not enough to meet all of the growth in energy demand, and unabated fossil fuels, shown here in the black bar, continues to rise.
That changes in the 2020s and 2030s and 2040s as we start to see unabated fossil fuels start to decline. But the extent of this energy substitution is relatively limited. Unabated fossil fuels still account for two-thirds of primary energy in 2050 compared to a little over 80% today. The shift to the substitution phase in Net Zero happens sooner and with greater intensity. The energy system moves into energy substitution in this decade. You can see those black bars already falling in the 2020s, and then gathers pace in the 2030s and 2040s, such that by 2050, the share of unabated fossil fuels in primary energy declines to just 15%. Two factors underpin this quicker, more pronounced shift to energy substitution: a sharper acceleration in energy efficiency, dampening and eventually reducing the world's overall energy needs, and even faster growth in low-carbon energy.
The source of this increase in low-carbon energy, and the second common trend I want to highlight, is the rapid growth in electricity powered by wind and solar. Electricity demand increases 75%-90% in Current Trajectory and Net Zero. Most of this growth, around 80% or so, occurs in emerging economies as living standards improve and access to electricity increases. This growth in power demand happens alongside an increasing electrification of the energy system, and you can see that in this chart here. So today, just a little over 20% of the world's final energy demand is electrified. That increases to a little over 35% by 2050 in Current Trajectory and to over 50% by 2050 in Net Zero. Over 50%, more than half of the world's energy use is electrified.
This increasing role for electricity is reflected across all end uses: industry, transport, and buildings, as more and more of our everyday processes and activities are electrified. The growth in power generation is matched, or more than matched in the case of Net Zero, by the rapid expansion in wind and solar power, shown here by the blue bars on the right-hand side, with the share of wind and solar in global power generation increasing from close to 15% today to between 50%-70% by 2050. It's this growth of wind and solar power which is driving the increase in low-carbon energy and the move to energy substitution in the two scenarios. The surge in wind and solar power is underpinned by continuing cost reductions, especially over the first 10 to 15 years of the outlook. Although necessary, these cost reductions are not sufficient on their own.
The rapid expansion of renewable power requires wind and solar capacity to be deployed more quickly than ever before. That's possible only if a number of other enabling factors (grid infrastructures, planning and permitting, social acceptability) scale at a corresponding pace. It also requires power systems to bolster their resilience to the increasing variability of power generation as the importance of wind and solar increases. Some of that enhanced resilience is achieved by greater demand-side flexibility, with demand incentivized to match fluctuations in generation. It's also met on the supply side by growing amounts of battery storage, gas and coal-fired generation combined with CCUS and low-carbon hydrogen. The significance of these supply-side measures doesn't derive from the volume of the energy they provide, which is pretty small in both scenarios. Rather, their significance derives from the value of being able to provide energy on demand as and when needed.
Resilience is a value game - right time, right place - not a volume one. The third common trend I want to highlight is for oil demand, which declines over the outlook in both scenarios, although, as you can see, the speed and extent of that decline varies quite significantly. In both scenarios, oil continues to play a major role in the global energy system over the first half of the outlook, with the world consuming somewhere between 80 and 100 million barrels a day of oil in 2035. As you can see here, in Current Trajectory, the profile for oil demand is pretty flat over the first half of the outlook, with a level of oil demand in 2035 pretty much exactly in line with the level of oil demand today.
Then we see a gradual decline over the second half of the outlook, with the level of oil demand reaching around 75 million barrels a day by 2050. In contrast, the pace of decline in Net Zero is far quicker, with oil demand falling to between 25 and 30 million barrels a day in 2050. That's around 70% below the current levels of demand. The single biggest cause of this decline in oil demand is the falling use of on-road transport, shown here in the yellow bars. This is initially driven by the improving efficiency of the global vehicle fleet. Further out, there's a growing impact from increasing electrification of cars and trucks.
The number of EVs increases to between 1.25 and 2 billion by 2050 in the 2 scenarios, with the amount of oil used in road transportation falling from a little over 40 million barrels a day today to between 5 and 25 million barrels a day by 2050 in the 2 scenarios. Three trends which are common across both Current Trajectory and Net Zero - trends which might occur in other pathways which lie somewhere between these 2 scenarios. The energy system makes a historic shift from energy addition to energy substitution. There's a sharp acceleration in low-carbon energy led by wind and solar, and the demand for oil begins to decline. There are significant differences in the timing and magnitude of these developments, but the underlying trends are common to both scenarios.
I want to switch now to features of the energy system which are more differentiated across the two scenarios, more dependent on the speed of transition. The first is the outlook for natural gas. Now, we don't know much about how the energy system will evolve in the future, but I think for most fuels and energy vectors, we at least have a pretty good idea about how the level of demand in 2050 will be relative to today. So, oil and coal, lower. Wind, solar, electricity, biofuels, low-carbon hydrogen, all higher. But that's not the case for natural gas. In natural gas, in Current Trajectory, natural gas demand increases over the whole of the outlook, such that it's around 20% higher in 2050 relative to today's level.
But in contrast, in Net Zero, natural gas increases in the near term but then starts to fall away, such that by 2050, it's around half its current level. These contrasting outlooks reflect the impact of two competing forces. The dominant factor in Current Trajectory is the increasing use of gas in emerging economies as they grow and industrialize. In contrast, in Net Zero, natural gas demand is crowded out as energy systems around the world increasingly electrify and the share of wind and solar in power generation increases. These two opposing forces are operating in both scenarios. In Current Trajectory, increasing demand in emerging markets dominates. In contrast, in Net Zero, that's swamped by the growing electrification and the increasing role of wind and solar.
Given that much of the growth in natural gas in emerging economies is met by liquefied natural gas, LNG, it follows this uncertainty concerning the future level of natural gas demand is also reflected in uncertainty about the level of LNG trade, which you can see here in Current Trajectory increases by about 80%, where in contrast, in Net Zero, it falls by around 40%. That range of possible outcomes for LNG, both up and down, adds to the uncertainty associated with investments in LNG facilities, which have long economic lives. The other feature of the energy system that I want to highlight as being particularly dependent on the speed of the transition is the use of low-carbon energy vectors to help decarbonize processes and activities which are hard to abate, or put more precisely, hard to electrify.
The charts here focus on low-carbon hydrogen on the left and sustainable aviation fuel on the right. But the same general point holds for other low-carbon energy vectors and technologies, such as ammonia and methanol in marine, carbon capture, use, and storage, direct air capture. As you can see here, low-carbon hydrogen in Net Zero by 2050 gets to around 400 million tons per annum. That's around five times the amount of industrial feedstock currently provided by gray hydrogen. As a result, the use of gray hydrogen as a feedstock is virtually eliminated in 2050 in Net Zero. Low-carbon hydrogen also plays a significant role as an alternative to fossil fuel energy in heavy industry and long-distance transportation, as well as a smaller but important role in power systems, mainly as a source of long-duration energy storage.
In contrast, as you can see, in Current Trajectory, low-carbon hydrogen goes to less than a quarter of that level, with close to two-thirds of industrial feedstocks still being sourced from gray hydrogen in 2050. On the right, a similar story also holds for SAF, which accounts for almost 80% of total jet fuel in Net Zero in 2050, compared to just 20% in Current Trajectory. It's also worth noting that even in a relatively rapid decarbonization pathway such as Net Zero, these low-carbon energy vectors really only start to increase in the second half of the outlook. In Net Zero by 2035, low-carbon hydrogen is a bit less than 100 million tons per annum, and then you see this very significant increase in the second half of the outlook, the same in terms of this very strong growth in the second half of the outlook for SAF.
Of course, the dependence of these low-carbon energy vectors on the pace of the transition is hardly surprising. Yes, the cost of low-carbon hydrogen and sustainable aviation fuel decline over the outlook, helping to support their growth further out. And yes, the nature of their learning curves means the faster they grow, the quicker their costs decline. But they remain expensive relative to the incumbent fossil fuels they are displacing. As such, their longer-term growth is dependent on the willingness of society to bear that additional cost, the willingness of society to support a rapid transition. So, two key features of the energy system which are particularly dependent on the speed of the transition, whether natural gas increases or decreases over the next 25 years, and the pace and extent to which less mature, higher-cost, low-carbon energy vectors, such as low-carbon hydrogen and sustainable aviation fuel, penetrate the energy system.
I want to turn now to the implications of delay. The motivation here is at the recognition that the longer the world remains on something like Current Trajectory with this very slow and shallow pace of decarbonization, the harder and potentially more costly it will be if governments and society at some point decide to take actions necessary to meet the Paris Climate Goals. That then begs the question of how we should think about the implications of a delayed switch to a faster transition path. Now, the IPCC does not produce estimates of carbon budgets which equate specifically to the Paris goal of limiting temperatures to well below two degrees. But it does helpfully provide an estimate for a carbon budget consistent with a high probability, an 80% probability, of limiting temperature rises to two degrees.
That's equal to around 900 gigatons, shown here on the right-hand side chart. Given that it's not possible to use a post-Paris consistent target, we base the analysis and the outlook around this 2°C carbon budget. The cumulative carbon emissions implied by Net Zero stay within this carbon budget. The way to understand how these charts work, I've taken the flow of carbon emissions in Net Zero here and then plotted out their cumulative pathway here. You can see here that stays within the carbon budget in 2050 for Net Zero. In contrast, for Current Trajectory, that exceeds the budget in 2030, 2040s. Hence the concern, the longer the world remains on a pathway like Current Trajectory, the harder and potentially more costly it would be to stay within anything like a 2-degree budget.
To explore that risk, we developed a third scenario, Delayed and Disorderly. Now, the setup for Delayed and Disorderly is highly simplified, based around just two stylized assumptions. First, it assumes the global energy system continues to move in line with Current Trajectory for a period, after which sufficient policies and actions start to be undertaken to begin an accelerated fall in carbon emissions consistent with meeting the 2-degree budget. The second assumption is that there are limits to the pace at which it's possible to decarbonize the global energy system in an orderly manner. That is, without having to resort to policies and actions which have outsized economic and social costs. The idea that there is some limit to the pace at which the global energy system can decarbonize in an orderly manner seems fairly intuitive to me.
Trying to calibrate that fastest orderly pace is much harder. For illustrative purposes, in the outlook, we consider two possible approximations. One possibility is to assume this fastest orderly pace broadly equates to the pace of decarbonization achieved in Net Zero. This assumption implies that at the latest date the energy system could continue along the Current Trajectory and still achieve an orderly transition consistent with remaining within the 2-degree budget would be around the middle of the next decade. So, you can see that here in this delayed Net Zero purple line, where Net Zero starts around the turn of the decade. It's coming down pretty much in parallel with the Net Zero line. As you can see, the cumulative pathway for that Net Zero just remains within the 2-degree budget.
The second approximation was to assume the maximum orderly pace of transition was equal to the very fastest speed of model decarbonization across the many IPCC scenarios. So, by searching across those IPCC scenarios, we take that very fastest model pace. That's somewhat quicker than Net Zero, suggesting a shift to a rapid decarbonization pathway by the mid-2030s and still remaining within a 2-degree budget. So, that's shown by that steeper green line on the chart on the left, again, just remaining within the carbon budget on the right. These approximations, which, I stress again, are highly stylized, suggest that if the move to an accelerated transition path is delayed much beyond the early to mid-2030s, there would be an increasing likelihood that it wouldn't be possible to have an orderly transition and still stay within a 2-degree carbon budget.
If that, in that case, to stay within a 2-degree budget, costly or disorderly measures would be needed to reduce or curtail the use of unabated fossil fuels and so achieve an even quicker pace of decarbonization. If, like me, you find that arithmetic pretty sobering, it sort of begs the question of what the world needs to do to shift from the current pathway to something more akin to Net Zero. That's the focus of this final section. This is the same slide I showed you at the beginning of today's presentation with the carbon emission profiles for Current Trajectory and Net Zero. The goal now is to compare the two scenarios to identify the key actions and developments driving the faster transition in Net Zero relative to Current Trajectory.
Based on that comparison, what needs to change to move the world off its current relatively slow, shallow, geographically uneven pathway onto a faster Paris-consistent one? Which sectors make the most difference? What needs to happen in those sectors? Where does the world need to focus its efforts? The single biggest difference between the two scenarios is the quicker pace of decarbonization of global power systems in Net Zero relative to Current Trajectory. And that accounts for around 40% of the difference in the speed of decarbonization in two scenarios shown by this purple wedge. So, the way to think about this purple wedge is if the pace of decarbonization of power markets in Current Trajectory was the same as Net Zero, that would move, for everything else stay the same, that would move the green line down to the bottom of the purple wedge.
It counts for that much of the difference. Global power generation in Net Zero is entirely decarbonized by 2050, whereas in Current Trajectory, that pace of decarbonization lags by around 15 years or so. Digging deeper, the main factor accounting for this difference, so why are power markets in Net Zero able to decarbonize more quickly relative to the other one? The main difference is the pace at which emerging economies decarbonize their power systems. Even in Current Trajectory, power generation in the developed world largely decarbonizes by 2050. But the rapid growth of electricity demand in many emerging economies means it's far harder to grow low-carbon energy sufficiently quickly to both meet the growth in power demand and displace existing fossil fuel generation. It's that same Energy Addition versus Energy Substitution point again. Some, but only partial progress is made in Current Trajectory.
In contrast, in Net Zero, the growth of low-carbon energy, especially wind and solar, is significantly faster, helping it to crowd out coal in particular. Away from power markets, across the so-called end-use sectors, the sectors in which energy is ultimately used, industry, transport, and buildings, the most important sector accounting for the faster decarbonization in Net Zero relative to Current Trajectory is the industrial sector, shown here in red. Now, that's not that surprising. The industrial sector accounts for more carbon emissions than the other two sectors combined. It's where most of the action is. Some of the faster pace of industrial sector decarbonization reflects quicker gains in energy efficiency in Net Zero relative to Current Trajectory. That reflects both the efficiency of the industrial processes themselves and also wider energy conservation measures, helping to reduce the demand for manufactured goods and materials.
Other than energy efficiency, there's no single silver bullet accounting for the faster decarbonization of industry in Net Zero. It's more to do with policy incentives and support driving faster and more widespread adoption of different low-carbon technologies in different sectors. Increased electrification in light industry, greater deployment of CCUS in the cement sector, ditto low-carbon hydrogen in steel. All these technologies exist today. The challenge is to apply them at speed and scale. The remaining difference between the two scenarios is accounted for by the faster pace of decarbonization in transport and buildings. The faster pace of decarbonization of the transport sector in Net Zero largely reflects the quicker electrification of road vehicles, again, especially in emerging economies led by China and India.
For the building sector, the main drivers, again, are greater electrification, especially the use of heat pumps, and faster gains in energy efficiency, both in terms of the efficiency of electrical appliances and in the design and construction of buildings. So, what should we make of this? If we stand back from the detail, what are the key actions, the key areas of focus needed to accelerate the energy transition? For me, I would highlight four in particular. First, and perhaps most obvious, to achieve a fast decarbonization pathway akin to Net Zero, the whole world needs to decarbonize. It's not enough for the developed world to focus on decarbonizing their own domestic energy systems. They also need to help and support emerging economies to transition. Second is the importance of energy efficiency. Energy efficiency is often overlooked in policy discussions and debates.
It sometimes feels like the poor cousin of climate discussions. But the ability of the world to move quickly from energy addition to energy substitution depends on a significant acceleration in energy efficiency. Energy efficiency is also the best possible response to the energy trilemma. Third, decarbonize global energy markets. Many of the benefits associated with increasing electrification of the energy system rest on that electricity rapidly decarbonizing. Finally, application, not invention. There will undoubtedly be huge amounts of innovation and technological progress over the next 20 years helping the world to transition, many of which we may not even be able to think of today. But the big differences in Net Zero relative to Current Trajectory don't stem from new inventions or technological breakthroughs.
Rather, they come from the faster and wider application of known technologies: wind and solar in power markets, electrification, low-carbon hydrogen, carbon capture, use, and storage. So, where does all that leave us? As I said at the start, this is the 13th BP Energy Outlook. Looking back like we did at the beginning, it sort of makes you wonder, what will my successor be saying as they launch the 26th BP Energy Outlook in 2037? I hope, and I indeed think, that some of the common trends we've discussed today would have materialized. In particular, that the world would have successfully made the shift from energy addition to energy substitution with the use of unabated fossil fuels and with carbon emissions on a clear downward trajectory.
I also hope that the pace of that transition is sufficiently rapid that the specter of a delayed and disorderly pathway is seen as yesterday's problem, the delay finally over. If that is the case, I've no doubt this will have been helped by some breakthroughs and developments that we haven't even considered into today's outlook. But whatever those breakthroughs turn out to be, my hunch is that the transition will have at least some of the four elements from our action list: the whole world decarbonizing, developed and emerging economies supported by accelerating energy efficiency, decarbonized power systems, and the widespread application of many of the low-carbon technologies that we know about today. Unfortunately, the one thing I know for sure is that in 13 years' time, they won't be making any reference to a very youthful-looking Spencer Dale. But you can't have everything. Thank you.
Okay, now for the fun bit. Before we go to the Q&A, I just want to flag the three survey questions we're going to ask today. As I mentioned, these will be available to everyone online to answer. Again, apologies to those of you in the room, but you can think about what your own answers are and then compare it with what our online respondents say. For those online, the questions will come up in a sidebar on your screen, and you will have two or three minutes to answer each question. Then we're going to come back and report your answers before the end of the webcast. These are the three questions. First, when do you all think the world will move from energy addition to energy substitution?
We're giving you five options here, starting from pretty much now, with the latest being in the 2040s or later. So, when do you think that's going to happen? Second question, what do you think is the single most important enabler for a faster energy transition than the world is currently on? What we've done here is given you the four options that I came up with in terms of my action list. But if you don't like any of these, you can pick the final one. The final question was, when you all come back in 2037 and we're looking at the 26th BP Energy Outlook, what technology or trend would have surprised us most? We had great fun as a team trying to think about what our options are.
So, the four options that we came up with were AI-enabled improvements in energy efficiency, new ways to electrify high-temperature heat, nuclear fusion, or climate engineering technologies. But again, you may not like any of those, and you can pick the final one. And as I said, we will show the results at the end of the Q&A. I must admit, I'm really intrigued to find out the answers. Okay, that's all of that. Now we're going to go to the Q&A session. And to help moderate that session, I'm going to be joined by Doris Fujii. Doris, why don't you take the stage and introduce yourself?
Thank you, Spencer. Hi, everyone. My name is Doris Fujii, and I'm the Head of Hydrogen and CCUS Analysis in Spencer's team. I'm going to be moderating the Q&A today. It's great to be here in person here in London because I'd normally be waking up at the crack of dawn to watch this from Houston. Speaking of Houston, we want to send all our best to all those who were impacted by Hurricane Beryl this week and wish you all a speedy recovery. A few housekeeping items before we get started here. We're going to be taking questions both online and in the room. I'm going to be starting in the room after a few other items. Please do get your questions ready. When you do have a question, please raise your hand, and someone will come to you with a microphone.
Then please state your name and the organization you're from, and then ask your question. For those of you online, thank you for all the questions so far. Please do keep them coming in. We'll try to get to as many of them as possible in the 30 minutes or so that we have. There's an option to submit questions anonymously if you'd prefer. Yeah, that's. Oh, okay.
They can vote.
They can vote. Thank you, Spencer. There's an option to upvote using the little thumbs-up sign if you really want a question to be answered.
If it looks like a really difficult question, don't give a thumbs-up to that one.
Yeah. Those are the ones we want. Yeah. So, that's all I had in terms of housekeeping. So, let me turn to the room now and kick this off. Who would like to ask a question? Oh, we've got several hands up. Great. Let's go to this person right here, and then maybe the person next to you after that.
Hi, hello. I'm Martijn Rats from Morgan Stanley. First of all, I want to congratulate you with such a detailed and thorough piece of work. It's a fascinating read every year. So, I'm very grateful for BP and your team to keep the series going. I want to ask you two questions. In the presentation of forecast, what is always a particularly interesting element is the revisions from last year. And you haven't said much about it, but I was wondering if there are elements of what you've just presented that are now different from the 2023 outlook, things that happened over the last 12 months where you would say, actually, frankly, we needed a bit of revision there, particularly with regards to the oil demand trajectory. I'd be very interested in that.
And the second question I wanted to ask you, not an easy one, but there's a lot of emphasis on energy efficiency. But in the history of global energy statistics, there is also a very strong effect of efficiency improvements bequeathing more utility to that energy, and then people end up using more of it rather than less. So, I was wondering how you think about that, if any of that is sort of captured in the outlook. It's an incredibly difficult and somewhat unfair question, but is it really a solution? Thank you.
Thank you, Martijn. Good to see you. So, two things. One is big changes from last year, and then how does that apply to oil? And I'll tell you both bits. So, there's hundreds of changes from last year. And actually, let me just plug the booklet. For those in the room, you don't get to answer questions as we go. You don't get to do the poll. You do get to have a hard copy as you walk out, right? So, as you walk out and you see the hard copy, there's a new section at the beginning, which we sort of summarize recent developments in the energy system. We haven't done that before. And it's tried to capture this sort of sense of the news. I think one bit of the news is energy demand has grown more quickly than we thought. And it's just come back.
There's just greater momentum in global energy demand. And a number of things flow from that. One is the demand for pretty much all types of energy is just higher than we thought, at least in the near term than we expected. That's true for fossil fuels. It's also true for the wind and solar and some of the low-carbon technologies. So, that's one. Two corollaries of that. One is that energy efficiency, which comes to your second point, has disappointed. Energy efficiency, on average, over the last 10 or 15 years has grown between 1.5% and 1.7% a year. As you know, the COP pledge is for it to grow at 4% each year out to 2030. The average growth over the last four years is 1%. So, it's slowed. And that's another issue.
I guess another third point is just completely surprised by just the strength of solar power. That's driven in part by these huge falls that you know fully about. I mean, we estimate that solar modules, the cost of solar modules has fallen by about 60% over the last 4 years. We expected strong growth in solar energy. We didn't expect really, really, really strong growth in solar energy. So, that's a third bit. That's the main factors affecting oil as well. The only other bit on oil, I think we've taken a look at our modeling of the impact of oil as a feedstock into pet chems. And I think we thought we were on the low side. I don't think this is any great news. I think it is more us head-scratching and thinking, will it really come off quite as much as it does?
And so, we've revised up our little bit of the growth of oil as a feedstock. And that's pushed up a little bit the oil profiles, particularly in the second half of the outlook. Energy efficiency and sort of Jevons paradox, which is the idea the more efficient you become, the more you use energy. I don't know how to think about it really, Martijn. I mean, I think there are clear components of how the energy system which is evolving, which should lead to improvements in energy efficiency. So, the greater adoption of electric vehicles, the greater adoption of heat pumps, that should all lead to greater improvements in energy efficiency. I also think the component of energy security, one component of energy security was to worry far more about how much energy you're using. And so, that also leads to improvements in energy efficiency.
But as you say, it's a hard thing to see. We have, on our Current Trajectory, energy efficiency is increasing a little bit more than the past, but not much. It's about 2%. And in Net Zero, it's quicker. Again, not as much as 4%. It's more like 3.5%. But the way to think about that 3.5% of energy efficiency in Net Zero is sort of saying, what does the energy system need to do to get to Net Zero? And you can't do it all by building more and more and more low-carbon energy. In part, you do need to see an improvement in energy efficiency. So, for Net Zero, I think it's a necessary part of the package that we need to see. Sorry, I'll make my answers short next time.
Okay. Thanks, Spencer. I think the person right next to you. And then, oh, there was somebody over on this side. Sorry. Can you raise your hands again over here? Yeah, maybe right here. And then I'm going to take a question online after that. And I'll get to you guys after that. Don't worry.
Sorry, was that me? Okay, sorry. Jonathan Stern from the Oxford Institute for Energy Studies. This is really a definitional question. Could you say more about your definition of low-carbon hydrogen? To what extent does it include zero-carbon hydrogen, electrolyzed from renewables or nuclear? How much natural gas with CCUS? And therefore, that presumably adds to your natural gas demand. So, that would be useful to kind of get a sense of, because you made that a big component, what you have in mind for that.
Yes. Thank you, Jonathan. Good to see you. Again, in the outlook, we explain a bit more. Roughly, roughly, so it's dominated by green and blue. There's very little. There's a little bit of bio-based hydrogen, but not very much. And no significant nuclear-based hydrogen. So, it's largely green and blue, with that blue dominated by natural gas. And in the outlook, by 2050, it's sort of roughly 60% green, 40% blue. That's the rough component. And the way we ended up thinking about this is you look at the world today, and blue hydrogen looks like it has a cost advantage in many parts of the world. But then we also recognize that the cost of transporting hydrogen is very expensive. And so, this ends up being a regional market.
We end up with that 60/40 split, not by doing a sort of function of lowest cost, global cost, but essentially looking at the regions where we think hydrogen is going to grow significantly, say China. In China, green hydrogen is going to be cheaper than blue hydrogen. That 60/40 split is based more on a regional built-up from a regional picture rather than from a global picture overall.
Great. So, we're going to take a couple of questions online now, and I'm going to go to this side of the room. So, there's a question from Biraj Borkhataria. I hope I pronounced that correctly. How sensitive is your scenario to population growth over time? What happens if global population stagnates versus growing to 10 billion+?
Thank you, Biraj. And I think I've made this joke before, but it's definitely true. This is like sitting when you turn over your exam paper and you suddenly see the question come up in front of you, and you're going, "Oh, my goodness, goodness." So, it does have some impact, Biraj. But I think the big picture story here is the growth in global GDP over the next 25 years, and with it, the growth in energy demand over the next 25 years is far more dominated by increasing prosperity, not by increasing people. So, essentially, economic growth has to have one of two things. Either more people or those people become more productive. That's it. Economics is pretty straightforward, right? And so, in the next 25 years, the far more dominant factor, counted around 70% of the growth in economic growth and in energy demand, is increasing prosperity.
That's a story of essentially an emerging middle class in Asia. That's what drives economic prosperity over the next 25 years. That's what drives the growth in energy demand. I think that's interesting when thinking about the expansion of energy demand. If energy demand is growing because people are becoming wealthier, not because there are just more and more people, that changes the type of marketplace you're operating in. Because as people become wealthier, they can become more discerning in terms of the types of goods and services they provide, more differentiated. The story for the next 25 years for economic growth generally and for energy demand is driven by increasing prosperity, less to do with economic growth.
Great. Thanks, Spencer.
Population growth, sorry.
Okay. We have another question from Mark Howard. Do you believe the increase of CO2 emissions due to technological advances, AI, artificial intelligence, make Net Zero achievable? In parentheses, see Google and Microsoft emissions.
Now, I've got a sneaky feeling my team was running a book of how long we could get into the Q&A until we had an AI question. So, whoever's won, I'll say my time says here, 1:56. So, how do we win that time? So, the direct impact of AI on energy demand is we can sort of see in terms of that increasing demand for data centers. And trying to guess and estimate how much that will be is obviously an incredibly difficult thing to do. But at least you can sort of form a view about that. And we have tried to build in an impact of that into our outlook. Just to give you a sense of that, that increase in electricity demand or power demand going to data centers because of AI does make a significant impact in the growth of electricity in some markets.
So, particularly, for example, the U.S., you see a pickup in the growth of power demand over the next 10 years. And much of that we can attribute to AI demand. Although it's worth keeping in proportion, if you look at the global system as a whole, that's true for the growth rates. In terms of levels, the amount of electricity consumed by data centers accounts for around 1% of global electricity today. And that increases to about 3% over the outlook over the next 10 or 15 years. So, important in growth rates, but important to keep in perspective in terms of absolute amounts. That's a direct impact.
What's far harder to calculate, and why I think, Mark, I'm a little bit more optimistic than you in terms of the impact of AI, is how AI can improve energy efficiency, how it can start to help us digitize grids, how it can help us improve industrial processes, how it can start to find new materials that we can use for and advances that we can use for CCS and so on. And so, that's likely to reduce that direct impact. How much? So, the net impact of AI, I think, is going to be less than the numbers you see at the moment, which only look at the positive side.
In some sense, there are many commentators out there that actually think that in the medium term, the net impact of AI will be to reduce energy demand, helping the world become more and more efficient and more than offsetting that positive impact. I think the truthful answer is I don't know. I would implore people to make sure they think about the offsetting effect from potential improvements in energy efficiency and not just focus on those direct effects.
Right. Thanks, Spencer. So, let's take a question over on this side of the room here. Maybe the person on the end there.
Yeah, good afternoon. I'm Nick Wayth from the Energy Institute. Fabulous presentation, Spencer. I was really struck by how linear your Net Zero line was. I intuitively would have expected a curve to sort of be more S-shaped and sort of accelerate, accelerate, accelerate. And then maybe at the bottom, it gets harder and harder for that last 10%-15%. Could you sort of say a bit more as to why it sort of looks so linear on the chart?
Yeah, I think there's, I mean, it's a good question, Nick. And actually, I was going to, so, Nick Wayth, for anybody who doesn't know, the CEO of the Energy Institute, who's now producing the statistical review. And I was going to highlight that very high growth of carbon emissions in your statistical review as one of the surprises in response to Martijn's question. I think there's a couple of things going on, Nick. So, in some sense, what would give you that faster decarbonization further out is the fact that some of those technologies start to come through more quickly. And so, that helps. The counter is, in some sense, the way we do these scenarios, you do the easy-to-abate stuff first. And so, you get after the power sector.
It's a lot easier to get after the power sector than it is to get after steel or CCUS in cement. In some sense, you can see from those, although you had a really bad seat, didn't you? You can see from those wedges, the power sector does a lot of work. So, I think it's a net of those two things. As further out, you have two benefits. The costs of some of those technologies decline. Moreover, some of those fixed assets are turning over. And those two things together really help. The counter is, by then, you're on the hard stuff. And the hard stuff is more difficult. And then I think those two things net out to that sort of somewhat linear picture.
Okay. I think we had right next to him there.
Thank you. Christopher Kuplent for Bank of America. Spencer, 13 years. This may be a lucky number. And maybe your answer to my question is too soon. Because COVID and the energy crisis that we've lived through, particularly in 2022, aren't that far away. But maybe still now, an opportune moment, perhaps better than last year when we were still in the midst of things, to again ask you, how has your view of the Current Trajectory changed because of those two events? Because I remember years ago, us sitting here saying Current Trajectory looks like 4 degrees global warming. We're down at 2.5. And maybe if I can add another question, what do you think is the risk that actually, in a few years' time, the Current Trajectory points to a worse outcome?
Because it seems that so far, it's accelerated the pace of the energy transition whenever we look at those revisions. Thank you.
Thanks, Chris. It's a good question. I think this is my eighth or ninth energy outlook, which is always a worry because then people can remember what you said or that period ago. So, my instinct, Chris, is this is a personal view. I don't think I'm seeing a significant shadow or scarring from COVID. I thought I would. As we were living through it, it felt like a defining thing of our generation. And it doesn't feel like that. I think the war in Ukraine is, though, and does and has affected things.
And I think the reason why it's done that, and it's a similar argument to what I made last year. I think that when I sat in capitals having conversations in Brussels or in D.C. before the war in Ukraine, almost all the conversations were focused on the need to get to Net Zero, the urgency of decarbonizing the energy system. And they sort of forgotten about the importance of energy security and energy affordability. It sort of was just there was a real focus. And I think what happened in the war in Ukraine is it reminded everybody about the importance of that energy security and energy affordability. And in some sense, that's why I raised the point about the trilemma at the beginning.
I know the trilemma, people sort of roll their eyes and go, "Oh, my goodness, not another person talking about the trilemma." But it really matters. And what was interesting is you go to China and India, they never had forgotten about energy security. And I think the West has remembered energy security. And I think a feature of energy security is an apologist for the economic jargon. Energy security is a public good. The benefits from energy security derive to accrue to everyone, not just you and I who are doing our trade in energy, but to everyone. And what we know about public goods is you can't just rely on markets to provide public goods. Governments need to play a role in the design of those markets and the implementation of those markets. And I think you can see that today.
I mean, I don't think the U.S. IRA is a sort of coincidence. I don't think the U.S. Green Policy Deal is a coincidence. I think green industrial policy does feel different and significant. And I think that is related to the increasing awareness of the importance of energy security. And I think that's quite a defining thing in terms of how the world has changed. Will the Current Trajectory go backwards? I mean, I think there's lots and lots of reasons for being optimistic in terms of just the pace at which low carbon technologies are declining. And it is a common trend in those two scenarios. And so, I quoted solar modules falling by 60%. I think over 10 years, they've fallen by about 80%. Stationary batteries, similar 80%. And so, the cost of decarbonizing the world is getting cheaper and cheaper. So, that's a good thing.
But there is a danger that as governments get to the point where they've made their pledges and then they start to get to the point where they have to turn those pledges into actions, they become more difficult. And I guess that's the risk, OK? I'm not sure. It's not my central case. But the risk is it's a lot easier to make pledges than it is to follow through in the actions. And it may be that that's the concern. But I hope that's dominated by the fact that the ability and the cost of decarbonizing is getting quicker and cheaper over time, I hope.
Okay. This has gone by really quickly. I think we just have time for one more question. And we're going to.
This gentleman's been having his hand up all the time.
OK. All right. Well, let's take this question then. All right.
Thanks, Spencer. It's Alastair Syme. Spencer, do you have any view on how you should think about price around your different scenarios, the price of energy? Is it flat, inflationary, deflationary in real terms? And do you think within that consumers will be willing to pay a price for carbon?
Let's pick a different question.
Okay.
So, we do have a view about what happens to prices. But I'm not going to share it with you. And the reason why is because I don't think it's particularly interesting. And if I did, because it will be wrong. And if I did share, you can't do a scenario like this without a price. That's how you balance demand and supply. So, I can't pretend to you I haven't got one. Of course, I've got one. Otherwise, my scenarios don't make sense. But if I start sharing that price with you, I know for sure that all it will be is BP says X or Y. And it's not X or Y because I've told you all those scenarios are wrong. I started by saying all the scenarios are wrong. I also know within five meters of St.
James, all those scenarios are wrong, are gone, and BP expects. So, we do. But I don't think I'm going to share it with you. The interesting question, the other bit, not both bits are interesting. Just, I'm not going to tell you the first bit. The second bit is, will society pay for cost of carbon? And I think that's a more interesting and it's a difficult question. There was always an economist like to think, well, we know the answer to how to solve this. We do it with carbon pricing. It's silly. It's obvious. And then when politicians don't do it, we all roll our eyes and go, politicians don't understand, do they? And then, so you think we just need to explain it to them more. Most of the politicians I met are all very smart people. They understand the economics of carbon pricing.
It's just the politics of carbon pricing are really brutal. So, I think some of this will be done by carbon pricing. But I also think some of it will be done by other types of regulation and mandates. And I think for some sectors where the cost, the shadow carbon price is quite high or very high, like in steel or cement, it may well be better to do that via different types of mandates or regulations just from a politics perspective rather than trying to do it by pure carbon pricing.
OK. Great. I think that's all the time we had for the Q&A.
Oh, now we get to do the fun bit.
Yeah, I wish we had time to answer more questions. Maybe you don't. But yeah. So, thank you, everybody, for all the really thoughtful questions. And I think we're going to move to the poll result.
Now we're going to do the.
Right.
And so, just for Nick, and in the spirit of diversity, I'm going to go to this screen. So, the first question is, when will the world move from energy addition to energy substitution? And we gave you those different possibilities. And so, what we've got here, the second half of the 2030s. So, it's a combination I think it's essentially the 2030s together is the biggest winner, which I think is pretty much what Current Trajectory was saying. So, that was the argument there. So, that was the first one. The second one is, the question was, if you remember, what do you think is that key technology to help the world move to a faster transition pace? And oh, we have a clear winner there. All right. The clear winner is the whole world needs to decarbonize.
We can't just focus. The developed world can't just focus on decarbonizing its own energy systems. It needs to help the world. It helps the need to help emerging markets to decarbonize as well. And the third one is, for those of you who are here in 2037, what technology would have surprised us most in terms of relative to the energy outlook today? Oh, OK. That is a split. So, OK. So, we got AI-enabled improvements in energy efficiency. So, a few people agree with me, Martijn. Nuclear fusion. OK, pretty much. I think that's across the board, which is sort of what we should be, right? If it was obvious, we should have put it in the energy outlook now, I think. So, thank you very much. I'm not sure if that accorded with people's views here or not. We can see.
I think that pretty much wraps it up. So, let me, I guess, just stop by just thanking everybody here in the room. And you do get the prize of a hard copy if you want to take one back with you. Everyone who stayed online, thank you very much, especially those in Asia. I know the time is getting very late. So, thank you for staying with us. And thank you very much, everyone. Thank you.