U.S. EIA's International Energy Outlook 2021
Good morning, everyone, and welcome, virtually, to the Center for Strategic and International Studies. My name is Joseph Majkut. I am the director of our Program in Energy Security and Climate Change, and we are so very glad that you are joining us today.
Everyone in our program, myself included, are very excited to welcome our guests. We have with us several analysts and leaders of the U.S. Energy Information Agency, which is today – October the 6th – launching its International Energy Outlook for the year 2021. This is a benchmark product and an important function of the EIA providing projections, not forecasts, of energy markets and energy supply for the coming decades. As our world considers the critical issues of energy security and climate change, this type of analysis provides critical input for policymakers, analysts, and the general public. So we thank them for their work and we are delighted to host them today to introduce this new report.
We have a great program for you. As we get through the presentation in the – in the first half, the latter half we will have a question-and-answer session. You may use the chat function or the Q&A function on our broadcast to give us questions. I will moderate those with a panel of experts, so you are allowed to go as technical as you would like and we will try to exercise a little editorial judgment to make sure that we have a really rich discussion here today.
With that, I would like to welcome our first presenters.
Steve Nalley is the acting administrator and deputy administrator of the EIA. He will give us a brief overview of the – of the function of this report and the EIA itself.
And then we will be joined by Angelia LaRose, who is an assistant administrator for energy analysis, who will walk us through the headline results of the report being released today.
Thank you again. Please use the Q&A functions that are available to you. And I look forward to a robust and rich discussion.
With that, Stephen, take it away.
Stephen Nalley: All right. Well, thank you, Joseph and CSIS, for hosting this year’s International Energy Outlook release event. And good morning, everybody. Thank you for joining us.
It’s my great pleasure to have the opportunity today to share the key takeaways from this year’s International Energy Outlook, or the IEO. The IEO represents the culmination of the yearlong efforts by many EIA energy professionals, several of whom you’ll hear from today later in this presentation during the panel discussion. So I wanted to acknowledge their extraordinary efforts.
I also want to remind our audience that the full IEO document and the graphs from today’s presentation, as well as a full set of data tables, will be available for download on EIA’s website following today’s presentation. And the recording of this presentation will be made available on both the CSIS and EIA websites.
Next slide, please.
Three broad themes emerged from our – from our IEO 2021 analysis and modeling, which you can see on this slide.
First, under the influence of current energy trends and relationships and without significant policy changes or technological breakthroughs, EIA projects that both global use of energy and production of energy-related carbon-dioxide emissions will increase through 2050. If these current trends continue, we project that global energy use will increase by nearly 50 percent by 2050. We project that GDP and population growth will offset near-term pandemic-related declines in energy use and long-term energy efficiency improvements, driving increases in global energy consumption.
Second, although adoption of renewable technologies will grow significantly through 2050, if current trends continue we see power from natural gas, coal, and batteries will allow the electric grid to meet demand through that period. Under current laws and regulation, these fuels are often the least-cost resources to meet reliability needs and provide energy when renewable resources such as wind, sun, and water are not available.
And third, under current trends we project continued growth in oil and natural gas production mainly to support increasing energy consumption in Asia. A driver of this growth is that we believe that the developing economies of the non-OECD Asia in particular will import more liquid fuels because those economies lack sufficient production capacity to meet growing demand. When we adjust our model for faster global economic growth or for higher oil-price trajectory, we actually see unprecedented levels of petroleum and other liquid feedstock production into the future.
Next slide, please. I don’t know if the slides are advancing or not.
The IEO cases that we are releasing today are the first we have developed that incorporate the facts of the worldwide COVID-19 pandemic on global energy use and energy-related emissions. Economies responding to the COVID-19 pandemic had a significant effect on global energy consumption in 2020, which you can clearly see in the graphs of OECD and non-OECD energy consumption shown by economic sector. We expect energy consumption in non-OECD countries to return to the 2019 levels by 2021 as a result of rapid economic recovery in transportation and industrial sectors, and especially in heavy manufacturing. In OECD regions, we don’t see energy consumption returning to the 2019 levels until 2030.
I’d like to call two milestones to your attention brought on by the pandemic that mark shifts in the consumption of energy for transportation. First, 2020 marked the first time more energy was consumed for transportation in the non-OECD economies than for the OECD economies. And second, for OECD economies we do not expect energy consumed from transportation to return to the 2019 levels through 2050, particularly because of increasingly strict fuel economy standards, though we project that energy consumed in transportation will return to the 2019 levels by 2022 in the non-OECD economies.
Next slide, please.
These are the topline findings we are releasing today. Please keep in mind that our IEO is not designed to predict what will happen in the long-term international energy markets, but rather to provide model projections of what may happen given certain assumptions and methodologies. And by varying those assumptions and methodologies, the IEO 2021 illustrates important factors contributing to the future energy production and use across the world.
The IEO starts with our U.S. projection from the Annual Energy Outlook 2021 which we released in February, and it assumes U.S. laws and regulations current as of September 2020. From there, we incorporate current international laws and regulations, including climate policies, into our IEO side cases. This is the same practice we followed in the past, but with today’s release we are also publishing a report titled “Climate Considerations in the International Energy Outlook” – IEO – “2022” to enhance transparency and communication about how we model climate policies in this year’s IEO.
In a moment, Angelia LaRose will provide more information about our findings and assumption(s), but I want to briefly mention two key drivers of long-term energy consumption in our IEO projection. Those are global economic growth and world oil prices.
In our reference case, the average annual growth rate of GDP globally is 2.8 percent per year over the 2020 through 2050 period. And by 2050, we assume that world oil price is $95 per barrel in real dollars. Our side cases vary these assumptions, and in this year’s IEO for the first time we expand the publication to include sectoral consumption, electricity capacity, and electricity generation results for all of our side cases.
In addition, the COVID-19 pandemic continues to have a significant influence on the behavior of world economies. The pandemic – the pandemic-related energy consumption varied by country, fuel, and sector. As the situation continues to evolve with emerging developments about long-term supply-chain impacts on manufacturing or potential permanent changes to commuting and work patterns, we’ll address these COVID-19 impacts in our short-term energy outlook forecast in our international energy statistics database, as well as through many of our other statistical and analytic products. So I encourage you to take a look at those as they’re available on our website.
Next slide, please.
And finally, I want to make a couple quick points before I turn the presentation over to Angelina. First, a reminder that by law EIA’s data and analysis and forecasts are independent of approval by any officer employed by the U.S. government. The views of this report, therefore, should not be construed as representing those of the U.S. Department of Energy or other federal agencies. Our information is intended to help inform energy policy discussions and we never advocate for any particular policy. Though our forecasts cannot predict the exact future of energy markets, I believe that EIA’s analysis, forecasts, and projections bring context to the data trends and help inform a healthy national and global dialogue on energy-related issues. In that context, the IEO plays an important role in EIA’s mission as the nation’s premier source of independent and impartial energy information.
And with that, I’ll turn the presentation over to Angelina.
Angelina Larose: Great. Thank you. Thank you, Steve. And thank you, CSIS, for hosting today’s event.
Before we’re going to jump into more of the details behind IEO 2021 highlights, I wanted to take a moment to emphasize the importance that assumptions have on our projections. So this slide provides a brief overview of our GDP and world oil-price assumptions in our reference case as well as our four side cases, and this illustrates the range of uncertainty that comes in making long-term energy projections. Also, one thing to note on this, as well as on future slides: You’re going to see that the IEO uses the OECD designation as an approximation for the developed and emerging world.
So global GDP is one of the most sensitive inputs because small changes in economic growth can significantly affect projected energy consumption. The growth in non-OECD’s GDP, which almost quadruples from 2020 to 2050 in our high economic growth case, is notable – especially compared with OECD countries, which almost double during that – during the same period in that case. And although GDP is one of the strongest drivers in energy consumption, we also considered cases with different world oil-price trajectories. So the graph on the far right shows the range of energy consumption across the cases depending on assumptions. The highest 2050 energy consumption is in our high economic growth case, followed by our high oil-price case which also assumes a higher level of economic growth, and then the reference case.
Next slide, please.
So, as Steve mentioned, one of the main themes of our IEO is that there’s continued growth in population, growth in GDP, and consequently there’s growth in energy consumption and emissions through 2050. As you’re going to see in this section, there are two competing forces that determine energy consumption and carbon emissions in the IEO analysis.
First, demographic and economic pressures such as GDP and population growth generally increase energy consumption and emissions. Population increases by 2 billion people by 2050 and there is significant economic growth in certain regions. Conversely, fuel economics and laws and regulations can lead to increases in energy efficiency and consequently decreases in fuel growth. During the projection period and assuming current laws and regulations demographic pressures are dominant, leading to an increase in both energy consumption and energy-related CO2 emissions.
Next slide, please.
So total energy consumption, which you can see on the graph on the far left in this slide, continues to rise through 2050 in both the OECD and non-OECD regions. And this is largely a result of increasing GDP, which you can see in the middle graph, as well as the increasing population shown on the far right. As standards of living increase, most notably in non-OECD Asian countries, demand for goods and the energy needed to manufacture those goods also increases. So although energy markets are undergoing efficiency improvements, the levels of GDP and population growth more than offset any efficiency improvements under current laws and regulations. And me mentioning current laws and regulations will be a common thread through the course of the presentation.
So next slide, please.
So this slide illustrates the fuels that underlie that growth in energy consumption. Under current laws and regulations, we project that growth in liquid fuels consumption, which you can see in brown on this slide, continues at a near-constant pace through 2050. So as travel increases, the post-COVID-19 mitigation efforts partly ease, the majority of passenger and freight vehicles continues to have – continue to have internal combustion engines fueled by gasoline and diesel. Industrial use of petroleum and other liquids, particularly for chemical feedstocks, also increase during the projection period.
Even though we’re projecting that renewables consumption, in green on the slide, remains below liquid fuels consumption levels, renewable energy consumption more than doubles between 2020 and 2050. The rise of renewables, which accounts for 27 percent of 2050’s global energy consumption, in the reference case, results from falling technology costs and government policies. Consequently, we project that the electric power sector will increase its use of renewable energy sources to meet growing electricity demands.
Although natural gas consumption, in blue, grows by nearly one third through the projection period in the reference case, the growing shares of renewables limits the national gas share of global energy consumption, which decreases slightly from 2020 to 2050. Lower relative prices of national gas in the near term, as well as the need to supplement intermittent renewables, are important drivers in natural gas consumption.
Coal’s share, which you see in black, of global energy use steadily declines through 2050. Coal consumption declines in absolute terms through 2030, partly as a result of low near-term natural gas prices. Policies and slowing investment in coal-fired power plants also play a role in this near-term decline. However, coal remains a significant player in the energy mix through 2050, under current laws and regulations, for multiple reasons. And this includes expansion of coal-reliant heavy industry in India, the availability and security of local coal supply in some regions, and the projected growth of coal-fired generating power plants in non-OECD Asia to fuel that region’s growing economies.
Next slide, please.
So as noted in the previous slide, liquid fuels consumption increases over the projection period. And liquid fuels remains the largest source of primary energy in the reference case. This chart shows the global liquid fuels consumption across all five of the evaluated cases. As a reminder, the side cases represent alternative assumptions about macroeconomic growth and crude oil prices. In all five of these cases liquid fuels consumption is higher in 2050 than in 2020, reaching a high of 50 percent over 2020 levels in the high economic growth case.
The IEO 2021 assumes current laws and regulations, including for carbon emissions, and implicitly assumes some level of geopolitical stability and cooperation in order to reach – in order to reach the necessary levels of production required to meet this level of consumption. Our cases are designed to provide a baseline of observed patterns obtained solely on the foundation of economic activity, population growth, technology change, current laws and regulations, and least cost, for which alternative scenarios can be compared.
Next slide, please.
Even with continued growth in liquid fuels consumption, across the cases the previous slide reflects a significant increase in the number of electric vehicles from 2020 levels. So the IEO reference case – in the IEO reference case there’s increased economic activity, population, and private mobility following the pandemic-related woes of 2020, increasing the existing global light duty vehicle fleet through 2050. Although the 2020 global light duty vehicle fleet primarily consists of conventional gasoline and diesel internal combustion engine vehicles, and you can see that in the center graph, sales of electric vehicles grow in the IEO as the conventional stock is slowly replaced.
In our reference case, electric vehicles including both all electric vehicle, as well as plug-in hybrid electric vehicles that run on liquid fuels as batteries become depleted. The electric vehicle’s share of sales grows most quickly in OECD Europe, where we project that about 80 percent of the passenger light duty vehicle sales in 2050 will be electric. Worldwide, we project that electric vehicles will account for 31 percent of the global light duty vehicle fleet in 2050, 34 percent in OECD, and 28 percent in the non-OECD. Significant growth in electric vehicle sales, as well as the share of sales during the projection period, causes internal combustion engine fleet of light duty vehicles to peak in 2023 in OECD and 2030 globally.
Next slide, please.
So with continued growth in energy consumption and increasing fossil fuel use, global energy related carbon dioxide emissions rise through 2050 in our reference case. And you can see that from the graph on the left. As a reminder, IEO cases represent laws and policies current as of May 2021. So mandated efficiency, fuel, and technology regulations are generally more stringent in the OECD cases. Energy- related CO2 emissions grow much more rapidly in the non-OECD, largely as a result of increases in energy demand associated with population and economic growth. Non-OECD countries’ energy-related CO2 emissions increase by 35 percent in 2050 over 2020. And this is compared to the 5 percent emissions growth in the OECD.
Although energy- related CO2 emissions tend to follow GDP and population growth, changes in the fuel mix on energy efficiency directly affect the degree to which emissions correlate with energy consumption. So carbon intensity, which is the carbon emitted per unit of energy consumed – and is shown on the center graph – is largely determined by a region’s fuel mix. And it decreases in both the OECD and non-OECD through 2050. Larger declines occur in non-OECD Asia because of an increasing share of renewable energy, as well as technological efficiency improvements.
However, the average carbon intensity across non-OECD countries remain higher than those in the OECD through 2050, mainly because of a higher retention of fossil fuels in the non-OECD. So for example, on average the share of electricity derived from coal across the non-OECD countries is more than twice that than would have been the OECD over the projection period. Energy intensity, which is the energy consumed per dollar of GDP – and you see in the graph in the far right – also declines through 2050. And non-OECD countries experience the fastest reductions. Energy intensity between OECD and non-OECD countries become more comparable in the latter years of the projection period because technology use becomes more aligned as regional economic composition changes.
So to close up this section, population and GDP are key drivers in our long-term energy consumption and emissions trends. Although these key drivers are partly offset by current efficiency laws and regulations, the overall trend is increased consumption and, consequently, increased emissions.
Next slide, please.
Our second highlight is the growth of renewable energy. So in the IEO we project that renewable generation grows significantly and accounts for nearly all global electricity generation increases through 2050 in all cases. However, as more intermittent electric generating capacity, such as solar and wind, is incorporated into a region’s electric grid, a range of generating sources will be built or maintained to provide backup to the variations in electricity generated from these intermittent resources.
Next slide, please.
So globally electricity generation grows through 2050. Electricity generation grows slowly in the OECD regions and almost double in the non-OECD. Incremental electricity generation comes largely from renewable resources beginning in 2025, because technology costs continue to fall and favorable laws and regulations remain in place. As renewables, particularly solar and wind, become cost competitive, the reference case projects that nearly all post-2020 electricity generation growth in the OECD region will come from these sources, and displacing an increasing share of existing non-renewable, mostly fossil fuel-based, sources. And you can see that on the center graph.
Governmental policies support this growth in some OECD regions because even though it is cost-competitive, renewables generation has less opportunity to grow without policies to encourage it, because electricity demand is generally slower in the OECD. In non-OECD, we project that electricity generation from renewable resources accounts for about 90 percent in generation increases from 2020 to 2050. And you can see that on the graph on the right. Because electricity generation grows at almost twice the rate of the OECD region – twice the rate in the non-OECD regions compared to the OECD regions in the reference case, the non-OECD regions add over double the generation from renewable sources than the OECD.
As intermittent generating capacity increases, other generating sources are needed to maintain reliability of the electric grid. World coal-fired generation, which you can see in black on the chart on the left, generally declines through 2030 in the reference case. But it remains a significant part of the worldwide generation mix because it continues to provide reliable capacity and supplemental energy. Because we expect natural gases to generally rise after 2030, growth in natural gas-fired generation, shown in blue, slows, and generation from existing coal-fired power plants becomes more – become more economic. Although incremental generation is largely powered by renewable sources, dispatchable generation remains an important part of an electric grid.
Next slide, please.
So several factors determine which generating source are used to provide the least-cost energy and a reliable grid. And this includes the predictability of intermittent resource, as well as the cost and performance characteristics of the other technologies used to meet demand. And this slide provides an illustration of that. So, for example, solar generation predictably occurs only during daylight hours, regardless of the location and installation site. The regularity of this resource, which often overlaps daily and season demand patterns, is better supported by battery storage.
In India, which you can see in the graph on the left, we project that solar generation will make up almost half of the 2050 electricity generating capacity in that country. To accommodate that shift, we project that India will install 330 gigawatts of batteries to support a system with such a high level of solar-powered generation. In contrast, wind generation tends to vary widely through the day and across seasons, requiring the use of more conventionally fueled resources that are not energy-constrained, such as natural gas-fired combustion-driven plants.
Canada, which you can see on the right, has a high latitude and limited sunlight in the winter and fall seasons, which makes wind a more economic choice to build and operate over solar. By 2050, wind accounts for more than one-fourth of Canada’s electricity generating mix. In this case, natural gas-fired generating capacity pairs better with wind to provide energy and support grid reliability.
Next slide, please.
So our final highlight is that supply of petroleum and other liquids continues to increase in both OPEC and non-OPEC regions to meet the growing demand through 2050 across cases, particularly in non-OECD Asia.
Next slide, please.
So regional economic growth is a key driver of long-term energy consumption. The fastest-growing economies in the IEO 2021 reference case are largely non-OECD countries in Asia. And we can see that on the graph of on the far left. India’s economic growth is the highest. And although China continues to grow at an average rate equal to Africa, as well as other non-OECD Europe and Eurasia, its growth notably slows throughout the projection period.
Together, these top five growth regions were home to 71 percent of the world’s population in 2020, and 44 percent of its GDP. By 2050, these shares grow to 73 percent and 58 percent, respectively. In the reference case, liquid fuels consumption grows the most in non-OECD Asia regions, where consumption almost doubles from 2020 levels. And you’re going to see that on the following slide. Transportation and industrial sectors account for most of this productive growth.
Next slide, please.
So within IEO this liquid consumption growth primarily occurs in the rapidly developing economies of China, India, and countries in our other non-OECD Asia region, which includes Indonesia, Thailand, Vietnam, and a number of other countries. In contrast, crude oil production is relatively flat in the largest Asian countries – China and India – through 2050. And you can see that on the graph on the right. And production declines in other non-OECD Asia after 2030. This disconnect between consumption and production is most notable in India, where 2050 consumption levels are three times higher than 2020 levels. Yet, crude oil production will be lower in 2050 than it is in 2020.
Similarly, although China increases crude oil production slightly, this increase will not meet liquid fuels consumption growth over the production period. To compensate for this regional imbalance between oil consumption and production in the reference case, non-OECD Asia will need to supplement its local production with increased imports of crude oil or finished petroleum products, largely produced in the Middle East. This varies somewhat for natural gas, and actually plays an important part in the market.
Within our reference case, the United States remains the largest producing region of natural gas. Russia stays the top natural gas exporting region, and its exports more than double from 2020 to 2050. And non-OECD Asia as well as OECD Europe’s imports of natural gas grows significantly of 2020 to 2050. And this also supports this higher level of Russian natural gas exports.
So next slide, please.
So in conclusion, the IEO 2021 projects increasing energy consumption, emissions, liquid fuels consumption, as well as renewable consumption through 2050. Intermittent renewable generation, although significantly increasing, continues to rely on fossil-fuel-fired generation to maintain grid reliability. And growing demand in non-OECD Asian countries drives production and trade of liquid fuels and natural gas.
Next slide, please.
So one final note on our IEO 2021. I’d like to highlight that we will be releasing three issues-focused papers later this year that explore areas of uncertainty in international energy. This include energy implications of potential iron and steel sector decarbonization pathways, effects of changes in coal supply and demand on international trade and electricity generation, as well as the changes of composition of economic growth in China. So stay tuned to our website for these insightful analyses.
So thank you for your attention this morning. Thanks again for CSIS and thanks again to our many analysts in EIA who contributed to this report. I look forward to your questions to our panel of the IE experts. Thank you.
Dr. Majkut: Thank you so much, Angelina and Steve. That was really a fun presentation.
You know, personally speaking, I’ve been looking at these EIA graphs for so much of my career, and it’s a real pleasure to be in a position to be sort of like a conduit to others for them. I really – I’m enjoying myself, and I thank you for your presentation.
One note for our audience: We also have finished on time, respecting the professionalism of our colleagues, and so that gives us a lot of time for discussion. So please do insert questions using the Q&A or the chat function that is available to you, and we can have a robust discussion with a panel of experts who have contributed to the technical side of this report, right? So we’ve seen the highlights, but we can also kind of dive in with some of our EIA analyst guests on the more technical or the more material aspects of this, like, very large analysis that has been presented today. With that, I’d like to introduce our panel of experts, and begin lobbing some questions over to them.
Erin Boedecker – Erin Boedecker is the team lead for energy consumption and efficiency modeling at EIA. Chris Namovicz is the team lead for electricity, coal, and renewables modeling. We’ve got Mindi Farber-DeAnda, who is the team lead for petroleum and natural gas modeling, and Aaron Bergman is the team lead for macroeconomics and emissions.
So as think about your questions, you know, you can try and fit them into any of those areas of expertise. I’ve got plenty myself, and maybe I’ll take moderator’s privilege to begin.
One of the things that we saw in the presentation, and one of the things that’s – was a mention of this, this new addendum report, how the energy outlook will treat climate policies internationally. For any analyst, it’s hard to hit a moving target. Climate policies change all the time.
Chris, what is the EIA doing to try and keep updated, and how do climate policies that are on the books or conceived influence the analysis in the final report?
Chris Namovicz: Maybe I’ll let Aaron Bergman field this one. Aaron?
I think Aaron might have frozen out.
Dr. Majkut: Have we –
Mr. Namovicz: The question is how we account for – are you back on, Aaron?
Aaron Bergman: I am back on. I don’t know what happened. I apologize.
So as I’ve heard in the prior presentation, I mean, EIA is very much focused on existing laws and regulations. We do this both in our annual energy outlook and in our international energy outlook. This doesn’t – one reason we do this; it doesn’t put us in a position of sort of speculating. It also provides a very good baseline. You know, if we see just what’s on the books right now, we can, you know, provide a baseline from which people can do further policy analysis off of that. So that’s one of the reasons why we do that this way.
You know, as you mentioned, we produced this report that sort of outlines some of the policies that are currently – are included, the ones that are on the books. You know, we don’t – for example, if someone has made like NDC – Nationally Determined Contribution – as part of the Paris Agreement, unless that is codified in the law or regulation, that would not be included in the International Energy Outlook.
Dr. Majkut: Yeah, thank you for that. As a brief follow-up, you know – so let’s take an example, right? So the EU has a cap-and-trade system which governs emissions from several source categories or, like, let’s think – Canada has a national carbon tax that covers various sectors. How do those fit into the modeling and end up influencing the final results?
Mr. Bergman: So we do – I believe we include the Canada carbon tax; also, I believe, South Korea has a cap, and we try to capture the largest – we capture many aspects of the European trading system. Obviously, you know, there are limitations on what one can do in a model, but we do our best to capture all of those policies, to the extent we can, and we’re always looking to improve our model to be able to capture, you know, more details of the policies and things like that.
I don’t know – if there’s specific questions about individual policies, you know, we can discuss them. But generally, as I said, at least for those three, we do try to capture them, to the extent that the model is capable.
Dr. Majkut: Great. Excellent. I think that, you know, as I said, it’s tough to hit a moving target and it’s great that that’s – that fits into the analysis.
So maybe we could talk a little bit about some of the results. Right? One of the keys that I just saw is that for – to meet growing energy consumption, particularly in non-OECD countries, there’s this – there’s significant growth in the power sector, and particularly it seems like there’s a lot of growth in renewable generation. But there needs to be also some – there’s still a lot of natural gas and coal on the system at the end of the projection.
You know, for Chris, what influences that outcome? How do we get there? And, you know, given trends in the costs of these relative technologies over time, how do you view that outcome in the projection?
Mr. Namovicz: So what we’ve seen over the last decade or so is that the cost of renewables have, of course, come down tremendously; the cost of gas varies a lot but, in general, with the shale gas revolution it’s come down a lot as well. But renewables are really at the point where, in most parts of the world, depending on your access to resources, et cetera, renewables are easily cost-competitive to build – for new capacity growth. If you need additional generation on the system, your choice is between building the capital cost investment for a coal or a gas plant plus the fuel for coal and gas, versus essentially it’s all capital costs for wind and solar. That costs – renewables compare very favorably in that transaction. When the comparison comes, though, to displacing the coal or gas utilization, or especially the capacity, the capital investment of that coal and gas has already been made many years ago, in a lot of cases. It’s a lot more difficult for that capital cost of renewables to displace that capital cost through just the operating, just the fuel cost for coal and gas, absent policy. In areas that we see, especially in the OECD where there are strong policies – that oftentimes tips the balance and renewables can displace existing stuff. But especially in fast-growing areas like non-OECD Asia, areas that might not have a lot of access to lower-cost fossils, you know, in their own country, it’s – you know, it’s very difficult to displace the incumbent generation on a cost-competitive basis but still very easy for those resources to displace the new generation.
Dr. Majkut: Yeah, so it’s a – you know, the graph we saw of, like, the growth of renewables – in particular solar in some of the non-OECD countries – was, like, actually really quite impressive, for my own priors. You know? But one of the other lessons we saw was that there’s this continued demand for consumption in your model, and then particularly I’m wondering, you know, in the U.S. there’s a lot of attention to natural gas, natural gas exports, and participation in international gas markets.
So, Mindi, maybe it would be interesting to hear more about how this model sees international trade in fluids like gas or LNG particular, or oil, as you’re looking at this sort of, like, shifting consumption pattern.
Mindi Farber-Deanda: Thank you. We see growth in natural gas trade all throughout this model. We do actually highlight it in our tables and it is one of the few energy sources where you actually see trade tables in the IEO. And what you’ll see is given the regions – the greatest growing region is non-OECD Asia, and a lot of that gas is directed towards them. It can come from a variety of places. I mean, even though the United States is the largest producer of natural gas, we are not the largest net exporter of gas in the world. Russia has that. And so, as a result, you’ll see gas coming from middle – I’m sorry, OPEC Middle East, which is a region – we have select regions that we show in our outlook, so OPEC Middle East, Russia, Australia, and the United States supplying these regions. There’s also a lot of pipeline growth, so, you know, while LNG is significantly growing, it won’t displace all of the pipeline growth that is occurring because Russia is also enabling pipeline supply into Asia as well as OECD Europe. So we are seeing movement throughout.
Dr. Majkut: Excellent. Yeah, thank you.
Switching from fluids to electrification – right? – one of the – there’s this – early on in the presentation we saw this sort of – the offsetting effects of growing efficiency in the economy and growing consumption. Building electrification or electrification of other services I perceive as being an important element in overall gains in efficiency.
Erin, how does this model see? What does it see for building electrification over the next decades? And how does that influence this efficiency trade-off?
Erin Boedecker: Thank you. So there are different types of electrification, both electrification as moving from other fuels to electricity and then also increasing access to electricity. In a way, we include both in the projections. Electricity use is expected to grow faster than any other energy source in buildings, both residential and commercial, and by 2050 it accounts for 54 percent of all delivered energy used in buildings. And this is accounted for by increases in standards of living, households experiencing greater access to things like air conditioning, electric cooking, electronics – right? – all that plug load that goes in the wall. So the place where we see it growing fastest is in India, where there is a growing population, rising income, and a strengthening commercial sector, the growing population, and with more income, of course, they’re going to ask for more services. So we see all of that contributing to a broader electrification over the next 30 years.
Dr. Majkut: OK. Excellent.
Well, we’ve got a lot of questions coming in; please keep them coming. I’ve – they’re filling my chat window here and there’s a lot of stuff. You know, I’m trying to basket a few of them together on the macroeconomic assumptions in this modeling, and so, you know, Aaron, maybe you’d be in a position to start here, or Chris, but, you know, there’s a question here about the relative demand for oil in a world where, like, vehicle electrification and the energy transition seems imminent. Can you explain why you see fairly high oil prices in the projection, even as those technology shifts are proceeding?
Mr. Bergman: So I’m not sure – I mean, I can talk certainly about sort of the demands for things like that, I mean, that are GDP projections, so we do take our GDP projections from Oxford Economics; they provide us with the model that we then sort of integrate into the model. So the GDP growth projections are based on sort of the judgment of their analysts, you know, combined with our projections about energy prices, and so things like population and GDP and household income that feed into the projections for demand, for consumption for things like fuel consumption, liquids consumption, electricity consumption, and things like that.
With respect to the oil price – I don’t know, Mindi, maybe that –
Ms. Farber-Deanda: Got you. Thanks, Aaron.
Yeah, our oil price comes from our world oil price that’s established as part of our Annual Energy Outlook exercise that occurred earlier this year, and it’s an exogenous series that we insert into the model. So what happens with this is we are looking at what’s happening as of the end of last year and the prices were – a price path was established that took us out to 95 in our reference case, and that is what we’ve used in the IEO. We also have a high-price and a low-price case, and those are worked through to see what happens to the economies in that – in that way. But keep in mind, for big changes in the transportation sector we’re going to expect to see some type of policy change, and that policy change hasn’t been fully implemented such that we could reflect it in our IEO. So unless those things happen, we don’t see the huge impacts and destruction of oil consumption that I believe the questioner was asking about.
Dr. Majkut: OK. Interesting.
So, then, there’s a – there’s a follow-up question here which I’m actually personally really interested in, is: To what extent does the model incorporate the changes we expect to see in climate over the – over the same 30 years, right? So if changing temperature patterns or precipitation patterns alter either resource availability or resource demand as, like, you know, for instance, cooling and heating demand changes.
Ms. Bergman: So I can speak to sort of on the GDP side and let Erin speak on the demand side. Other Erin.
With respect to GDP, Oxford has recently implemented feedbacks on sort of global temperature to do their GDP projection, sort of taking into account the latest research on climate impacts from econometric modeling. So that is included as part of the Oxford projections there, so we do see them. So that is one of the causes of differences between the GDP projections in the prior IEO and this International Energy Outlook, is this sort of recent addition of Oxford to the model.
And maybe, Erin, do you want to speak a little bit about the temperatures?
Ms. Boedecker: So, yes. Although in the IEO on the U.S. side we do include projections for heating including degree days, which in turn translates into impacts on heating and cooling loads, in the IEO we don’t have that explicitly modeled. However, we do have the changes in economic drivers, and of course we do have what the basics are for different regions. So in India, we would expect to see electricity growth for cooling as incomes rise. And so we’re taking into account the general regionality without having explicit changes in temperature on the demand side.
Mr. Namovicz: I think there is also a resource aspect to that question, and we do not have – we’re not an integrated assessment model and we don’t have a detailed enough model, of resources anyway, to account for any changes that you would see in rainfall or wind patterns or cloud cover.
Dr. Majkut: Yeah. Interesting. I mean, those effects, I think, are probably – we would – we might expect them to be more significant over time, but I don’t know that they’re going to be first-order effects compared to the things that your report highlights, right – population, GDP, et cetera?
Question specific to India, though happy to entertain an answer that’s more general. The results show fairly significant growth in battery storage to offset the daily variability of the significant growth in solar resources there. Is that in this model a result of government policy? Is that fostered by government policy or is that something that falls out of the economics? And how do you think about sort of the multiple policy interventions that can go on in an energy system maybe more globally – more globally?
Mr. Namovicz: So India, and in fact the rest of the world, we – in the electric power sector, we’re trying to model the policies that, you know, we’re aware of. It’s a little bit difficult sometimes to distinguish what a – in the U.S. we know, if Congress passes a law, it’s a policy; we’re going to model it. If the president just says something off the top of his head at a press conference, maybe not so much. But for the rest of the world, we don’t necessarily have those kinds of insights.
Nevertheless, for India we have a number of renewable energy targets in the electric power sector. We’re not modeling a storage target. So the storage that we see in India is primarily based on economics, and I think it’s fair to say that it’s largely in response to the substantial buildout in solar. We see that in a lot of our modeling with storage, that when you have a large buildout of solar oftentimes a large buildout of batteries will accompany that because they pair very well together from a – sort of a technology usage perspective.
Dr. Majkut: Very interesting.
We’ve got a high-level question here that I think deserves some investigation. Your report shows a fairly significant growth in renewable generation and electricity consumption over time, but at the same time shows what I would say is like an increase in greenhouse gas emissions or a relatively significant increase in greenhouse gas emissions over the next 30 years. What explains that, those two going together?
Mr. Namovicz: I think if you look at the slides we showed, demand for energy – consumption of energy grows by about 50 percent and greenhouse gas emissions grow by less than 25 percent. I don’t have the exact figures off my head. So, in fact, we do see a lot of impact from the growth in renewables and such in reducing the carbon intensity that the – we showed the carbon intensity on one of the slides, and it is declining. That being said, demand is still growing and there are still portions of demand that are – that – you know, most economically satisfied through burning of fossil fuels, so you do get some emissions.
I don’t know if Aaron Bergman wants to add anything to that, or?
Mr. Bergman: I mean, I think that’s fair. I mean, while you see – do see renewables’ penetration in the electric sector increasing tremendously and reducing the carbon intensity of electricity generation, you know, the opportunity for decarbonization in the transportation sector and the industrial sector, as you know, is not nearly as prominent. And those are also very large drivers of emissions.
Dr. Majkut: So as –
Ms. Boedecker: Right. So – oh, sorry.
Dr. Majkut: Go ahead. No –
Ms. Boedecker: I was just going to jump in with a little detail and say that keep in mind that air and freight travel are both highly dependent on those economic drivers that Aaron just talked about, and they do have limited options for switching to alternative fuels, away from fossil fuels.
Dr. Majkut: Excellent. Thank you. You know, we’ve learned so well how to do these virtual events, but the conversational aspect is always like a little tough. So apologies for interrupting.
There’s a question here that – regarding carbon capture, right – so the use of fossil fuels, but some intervention at the technical level to capture CO2 out of flue gas and either deposit it or store it. Does your model see a future for that technology? What role does it play in the outcomes that you see?
Mr. Namovicz: Right now the only areas that we’re seeing in the electric power sector that use CCS technologies are, I think, Canada, where they have a pretty strict, you know, carbon cap – and it’s not – you know, unless you have a fairly stringent restriction on carbon, there’s not much economic justification for carbon capture. It’s strictly in response to a, you know, policy, and the policy needs to be pretty stringent to get it in the money.
Dr. Majkut: Yeah. Interesting. Thank you for that answer.
There’s a question here about – you know, well, so natural gas prices are on a lot of people’s minds, particularly in the international sector at the moment. What does – you know, what does the model say international gas prices might look like over the – over the long term?
Ms. Farber-Deanda: Thanks. So we do expect natural gas prices to decrease a little bit after the current runup, but again, we don’t see the seasonality that you see in the current market. So in the long term we do expect a reduction, but then a slow and steady increase. Our Henry Hub price, which is the centerpiece of our natural gas prices – I mean, we do include European and the Japan-Korea price as well – but with that we do expect prices to slowly increase over time.
What causes that? Demand. We do see some interplay between natural gas and coal. There are certain thresholds beyond which coal becomes more competitive. And as a result, we will see changes over time and, you know, that natural gas price does impact what happens in our outlook.
Anybody want to add anything, or?
Dr. Majkut: We’ve got a second technology-specific question related to how biofuels may be used in the transportation sector.
Ms. Farber-Deanda: So, unfortunately, in our current outlook biofuels was handled exogenously, so it’s analyst judgment. And I’ve been working biofuels for a while now, and so we track and we see what’s going on. We did not dig as deeply into it in the IEO as, say, we do in our Annual Energy Outlook domestically. But it’s – again, it’s very policy-driven. And unless those policies are in place already and are really spurring on penetration by biofuels, you’re just not going to see the level of growth. I mean, there is some growth when you look at our tables, but it’s not as large as I guess some would perhaps like to see. But again, that’s – it is a policy response that we just don’t have in there yet –
Dr. Majkut: Got it.
Ms. Farber-Deanda: – because those aren’t in place. Thanks.
Dr. Majkut: So related – I think this question is going to – also touches on policy issues and might be a little forward-looking, but somebody in our audience is asking how trade is treated in this international model and, in particular, the EU conceptions around carbon border adjustments or other, you know, rising tariffs in multilateral trade. How does the model treat those types of instruments, or how might you expect it to treat them over time?
Mr. Bergman: Certainly, you know, to the extent that trade, you know, affects the macroeconomics situation, Oxford takes into account tariffs and things like that in their model. They update their model, you know, multiple times a year. You know, we take – I think we froze the – our current one we took was from back in February, and so as the Oxford sort of conception of the GDP impacts globally of these things are encoded in their model, which we use as part of our GDP projections. So to the extent that there are any changes in that going forward, you know, Oxford would certainly incorporate them in their model and projection and that’s how we would incorporate them sort of on the macroeconomic level.
Dr. Majkut: And, you know, as you think about bringing climate policies into the International Energy Outlook, would things like CBAMs, or carbon border adjustments, kind of fit into that climate policy class of instruments?
Mr. Bergman: Yes, those are certainly on our mind as we sort of work to figure out how to best incorporate climate policies in the model. Definitely. Keeps us up at night a little bit.
Dr. Majkut: Yeah, I can image. I think CBAMs keep a lot of people up at night, and figuring out how to incorporate them into our thinking and modeling is probably on a lot of people’s minds.
We’re coming very close to the end. And maybe, Chris, you know, I’m interested in your thoughts, if you feel anything’s been missed in this conversation or there’s anything that you think deserves a little bit of extra treatment or commentary at this time.
Mr. Namovicz: No, I thought the questions have been very good and there’s – it’s a huge – trying to analyze 30 years into the future for the entire globe is quite a challenge, so there are just so many paths that we could go down that I think the audience did a good job of picking particular paths to go down.
Dr. Majkut: Well, thank you for your time and for your work. I know that it is a difficult task and we appreciate that you are one of the groups that really puts things out there. It’s part of your mission but it’s not easy to post the tables online and try to do your best judgment. So, you know, I thank you. I think CSIS is really happy to have hosted you all today. Our presenters Steve and Angela – or Angelina, excuse me, were great. Our panelists Aaron, Mindi, and Chris were very informative.
To our audience watching live, or recording in the future, we thank you for your kind attention. All of this material will be available from the EIA; they make it very public and very accessible. We encourage you to go take a look, try and learn from the EIA’s efforts. And we look forward to seeing you in the future at the next report. Thank you kindly. And with that, we sign off.