Global energy demand will plateau from 2030, oil demand will flatten from 2020 to 2028 and go to a significant decline thereafter, the shift to renewable energy will be quicker and more massive than most people realize, yet the energy transition will not be difficult to finance. These are some of the momentous conclusions of a set of major new reports from independent energy consultancy DNV GL, under the name Energy Transition Outlook (ETO). They are based on an independent model and output from hundreds of the company’s experts who work in all sectors of the energy industry. Energy Post spoke with project leader Sverre Alvik and lead author of the renewable energy section of the report, Paul Gardner. They agree: “changes are coming so fast they will surprise many people.”
When a company like DNV GL makes projections of our energy future, this carries some weight. The company employs some 3400 energy experts, who are active across the entire energy value chain, from generation, transmission and distribution to sustainable use, covering sectors including renewables, storage and grids, as well as oil and gas. DNV GL prides itself on its “technology neutrality”. It has no special interests, and no axes to grind. Moreover, its experts work all over the world. In fact, the company’s first Energy Transition Outlook report includes different projections for 10 major regions in the world, so it has a truly global perspective.
“Changes are coming so fast, they will surprise many people”
It is precisely the unique position DNV GL has that motivated the company to develop its own Energy Transition Outlook, says Sverre Alvik, who works in the Strategic Research section of DNV GL in Oslo, and is head of Energy Transition research within the company. “We have a history of over 150 years of building trust and providing forecasts. We believe it is important for all energy industries to understand the current reality.”
That reality, according to DNV GL, is about to undergo an unprecedented transformation. “Changes are coming so fast that it will surprise many people, even inside the industry”, says Alvik.
More efficient
The first major change that will have huge implications for the energy sector is that, perhaps for the first time in human history, global energy demand is about to plateau. And this will happen in the 2030s, according to the Energy Transition Outlook.
This conclusion seems to fly in the face of most other energy scenarios. The oil companies in particular count on continued demand growth many decades into the future. “It is a rather unique finding, yes” says Alvik. “That is to say, you do find this in 2-degree scenarios, such as the IEA’s 450 scenario, but not in any reference scenarios or even scenarios that take into account existing climate policies.”
“Many people I talk to are saying their next car will be electric”
There are three main reasons why DNV GL expects energy demand to peak, explains Alvik. “First, we assume a somewhat lower population growth than the UN median forecast. We use the IIASA/Wittgenstein Centre for Demography and Global Human Capital model, because it seems to better take into account the influence of rising urbanization and education levels on fertility growth. Second, we assume lower GDP growth than the IEA and the oil companies do, although our projections are in line with other sources such as McKinsey, PWC, MIT and Statoil. Economists like to see growth rates of 3 or 3.5%, but we think that’s unrealistic.”
“And thirdly, we have a more optimistic view on the continued cost reduction potential of solar and wind power and on improvements in energy intensity than many of the reference scenarios used in the industry.”
Alvik notes that “the growing electrification in combination with the growth of renewables will make the energy system more energy efficient, leading to important energy savings.”
Oil demand peak
The second major transformation that the Energy Transition Outlook foresees is a renewable energy revolution. Although the model takes into account that cost reductions will take place in the oil and gas sector, and assumes only a modest carbon price (at most $60 per ton in 2050), it finds that solar and wind will rapidly become cheaper than fossil fuels. As a result, renewables will expand to make up 44% of primary energy supply by 2050 and 85% of electricity supply, at the same time that the share of electricity in total energy supply will rise dramatically from 18% today to 40% in 2050.
The amount of renewable energy in the system by 2040 will be way beyond any existing experience
Natural gas is the only fossil fuel that will see growth (14% from now until its peak in 2035), while oil demand will decline by 38% and coal 73% until 2050. Biomass andnuclear are projected to remain fairly stable.
DNV GL has oil demand flattening from 2020 to 2028. “Not so very different from what Shell and Statoil have said they expect to happen”, notes Alvik “but considerably earlier than BP or the IEA have said. We know this conclusion will be challenged and I should say that continued investment in oil production will be necessary to make up for depletion.”
Rapid shift to EVs
The oil demand peak is mainly caused by a third major transformation forecast by DNV GL, which is a very rapid shift to electric vehicles (EVs). “This is a change that looks sensible”, says Paul Gardner, segment leader storage at DNV GL and the lead author of the renewable energy supplement of the report. “Many people I talk to are saying their next car will be electric.”
According to Gardner, EVs will win out because of their superior performance and – in time – lower costs. “EVs have so many advantages.”
“There is a significant risk that regulation will hold up some of the changes that we are foreseeing”
The additional electricity demand as a result of the EV revolution will not be an insurmountable problem for grid balancing, says Gardner. “We are not talking about a doubling of demand. The charging can be handled. However, the amount of renewable energy in the system by 2040 will be way beyond any existing experience. There will be costs and changes need to be made to the system.”
New regulation is particularly important, notes Gardner. “We have seen many times that electricity regulation is a hindrance to the expansion of renewable energy, because it can’t keep up with the speed of change. There is a significant risk that regulation will hold up some of the changes that we are foreseeing.”
“We have a lot of wind, a lot of solar and a lot of gas plant that spends a lot of time doing nothing”
One example Gardner mentions is in his own specialty, storage. “In many countries network operators are not allowed to go into the storage business, because that’s the preserve of generators. But storage is useful for network operators. This kind of regulation has already got in the way of energy storage projects.”
Gas peakers
Is a 100% renewable energy future realistic, according to DNV GL? There has been quite a lot of debate about this in the U.S. recently.
“It is possible to build a secure system with a very high level of renewable energy”, says Gardner. “But in our projections we don’t quite get to 100%. We see quite a bit of ‘peaking’ gas generation to go with renewables by 2050. We have a lot of wind, a lot of solar and a lot of gas plant that spends a lot of time doing nothing. Even when we include the backup costs for the gas-fired power, this still looks attractive.”
“That’s a pleasant surprise: we can afford the transition”
Gardner does add there is one thing that the model does not yet take into account, namely the effect that temporary surpluses of wind and solar power will have on the economics of renewables. “We have not yet shown what the impact is of that.” However, he says, “we will also need to decarbonize heat, so it makes sense to store the surplus renewable energy, for example to heat water with it or to convert it into gas, to be used for heating. It is not yet clear what the most economic route will be.”
Good and bad news
The implications of the energy transition for the energy sector are obviously huge. “One important implication of peak energy demand”, says Alvik, “is that competition will increase. The energy market will become much more competitive and cost-driven.”
Another, perhaps surprising, finding is that, thanks to restrained demand, overall investment in energy won’t have to increase, despite the transition. “Major investments need to be made”, explains Gardner, “but the amount of money the world spends on energy does not change much. The total number will be lower relative to GDP. That’s a pleasant surprise: we can afford the transition.”
That’s the good news. The bad news, says Alvik, is that in spite of all these huge and rapid changes, they still won’t be enough to meet the goals of the Paris climate agreement. “Still more extraordinary efforts will be demanded from the energy sector to tackle the climate change challenge.”
DNV GL’s Energy Transition Outlook: no additional investment needed
Unlike for example the IEA’s World Energy Outlook, DNV GL’s Energy Transition Outlook does not present various scenarios. It presents just one “most likely” energy future, and this is based mainly on costs comparisons: it assumes that over time prices will follow the same trend as costs. It also factors in long-term climate policies. For example, it assumes that fossil fuel subsidies will be gradually phased out over the next few decades and renewable energy subsidies will continue, but slowly decrease over the coming decades.
Further, it assumes that cost learning curves for solar and wind power will be maintained until 2050. That may seem optimistic, but according to Sverre Alvik the learning rate applies to every doubling of capacity. Since it will take longer to double capacity in the future, it will also take longer to achieve the same cost reduction as before.
Just looking at capacity additions in the electricity sector, the Energy Transition Outlook makes the following forecast:
Globally, investments in fossil fuels will more than halve from around $3.4 trillion/yr today to $1.5 trillion/yr in 2050, according to DNV GL, while non-fossil energy expenditures show the reverse trend, increasing fivefold from around $500 billion/yr today to $2.7 trillion/yr in 2050.
Shifting investments to renewables, where the investment is upfront capex, implies a shift from an energy system with a 60/40 split between opex and capex to one with the inverse split of 40/60. In dollar terms, global opex declines from about $2 trillion/yr in 2015 to $1.5 trillion/yr in 2050. Conversely, capital expenditure almost doubles from $1.8 trillion/yr in 2015 to $2.6 trillion/yr in 2050. These figures do not include the cost of grids and energy efficiency investments.
According to the Energy Transition Outlook, the energy transition can be undertaken without any increase in overall energy investments. Global energy expenditures will stay approximately constant over time. With the forecasted increase in global GDP of 130% the next 33 years, this implies that energy expenditures will fall to less than half of current levels of global economic output – from 5% to a little over 2% of gross world product.
This means that the energy transition is not likely to be jeopardized by a lack of funds, and, while adjustment will be needed to cater for the heavier capex load from renewables, the transition is unlikely to prove financially disruptive. On the contrary, there may be scope to accelerate the pace of change.
[adrotate banner=”78″]
S. Herb says
I am curious about the role of the increasing contribution of oil-fired generation capacity shown in the graph. Where is it, and is it being used at a low or high capacity factor?
S. Herb says
Most likely the color-coding for gas and oil has been reversed, so that the graph shows increasing gas generation capacity?
Helmut Frik says
I think for the question how much above 80% renewables share can be reached, the development of grids, especially HVDC-Grids will be very important.
If one assumes a countrywide grid of a small or medium sized country, 80% could be at the upper limit. If one assumes a strong worldwide grid, 100% are a nobrainer, because the sun never settles and the wind never stops blowing in this grid – supply and demand are practically flat around the clock. Reality is and will be somewhere in between. But few people understand how useful strong grids are, as it seems.
Wessel Bakker says
The Energy Transition Outlook states that there likely will be an increase in number of large transmission network projects including strengthening and extending both national and transnational Interconnectors. Question is whether these can be built in time and first time good. See my blog on this I published yesterday on LinkedIn: https://www.linkedin.com/pulse/can-europe-expand-power-transmission-time-wessel-bakker
peter vaessen says
The percentage of variable renewables that can be acommodated or hosted by the electricity grid depends on power, energy and spatial distribution of the variable renewables, the capacity of the grid and the implemented technologies to make the grid more flexible and smart, like:
– Strong regional interconnections
– Use of (distributed) storage and microgrids
– Activated inverter capabilities of grid connected renewables
– Rapid response of aggregated loads (Active Demand)
– installed technology for power flow steering and control
Given the fact that the present overall T&D system is only “lightly” loaded, energy wise, and the limited implementation of above mentioned technologies one can conclude that even a more than 100% share of renewables is possible.
Alastair Leith says
SEN have modelled for the grid in south-west Western Australia called the SWIS grid and it has no interconnectors to the National Energy Market. We’ve found that to go to 86% RE by 2030 would cost no more (and probably considerably less if renewables continue to get cheaper) than business as usual replacing coal and gas with new coal and gas as plants are retired due to life cycle. To go to 100% is possible, but adds significant costs because of the winter wind droughts that can last up to a week or more long accompanied by heavy cloud cover across the entire grid. So biofuel gas turbines or biomass (cheaper to burn but air quality issues) begin to make sense at that point rather than a massive overbuild of wind, solar and energy storage. But it is possible.
Most countries already trade energy with neighboring countries, which may be a way forward to cover bad weather periods.
Bob Wallace says
” As a result, renewables will expand to make up 44% of primary energy supply by 2050″
OK, not enough information to tell what they are forecasting. Using “primary energy” creates confusion. What is their forecast for the percent of “used energy” coming from renewables? Over half of the energy in primary energy is simply wasted, never does a bit of work, and there is no need to replace it.
60% of the primary energy (coal) that is consumed in a coal plant turns into waste heat. We use only 40% of the energy in coal.
80% of the primary energy (oil) that goes into our cars and light trucks turns into waste heat. We use only 20% of the energy in oil.
At the moment about 90% of the primary energy we use globally is from fossil fuels. We need to replace less than half of that with low carbon energy. If we were to produce 44% of our current primary energy use from renewable sources we’d not be using any appreciable amount of fossil fuels.
I really wish we could post graphics here. Pictures and thousand words. But, so – look at 10.1 at
https://goo.gl/8cbAX4
The US wasted 68.2% of its primary energy in 2016.
44% of our current primary energy use is approximately the same as 100% of our total energy consumption.
Or are they saying that only 44% of our used energy will be from RE in 2050? Extremely hard to believe.
Now, the time required to clean up our act…
Out of our 100% primary energy used today about 90% is high carbon. We need to replace about 40% and we can forget about the other 60%.
Install enough renewable capacity in a year to produce 1% of the primary energy we use today and we can ‘retire’ more than 2% of our fossil fuel primary energy. Install enough RE in one year equal to 1% of our current primary energy use and we basically eliminate FF in 40 years.
Install 2% RE per year and fossil fuels are only serving niche purposes 20 years from now.
I’d like to see some solid analysis that ignores primary energy and deals only with our actual energy needs and what it would take to be largely fossil fuel free in 20 and 30 years.
I suspect that due to the rapidly dropping cost of wind, solar and storage along with growing concerns about climate change we will ramp up to about 2% new RE (based on today’s primary) in about five years. Especially as EVs start selling in high numbers.
Let’s say 2025. But we won’t stop at 2% because 1) RE will continue to become cheaper, 2) concern about climate change will continue to increase, and 3) we will appreciate living with cleaner air and being sick less frequently and will insist our urban air is free of fossil fuel pollution.
And 4) we’ll be spending less for electricity as we add wind and solar.
Bob Wallace says
“we assume lower GDP growth than the IEA and the oil companies do, although our projections are in line with other sources such as McKinsey, PWC, MIT and Statoil. Economists like to see growth rates of 3 or 3.5%, but we think that’s unrealistic.”
We can have 3% or higher growth. As long as the energy capacity we add is RE and not FF. And if it’s RE then we have to add only half as much to get the same amount of GDP growth.
It’s likely that our transition to renewable energy will spur large amounts of economic growth. Millions of good paying jobs will be created around the world as countries ramp up their wind, solar and storage industries.
Energy cost will fall making the cost of manufacturing lower.
Air pollution will decrease which will lead to far fewer health care expenses and lost work days.
Bob Wallace says
“Another, perhaps surprising, finding is that, thanks to restrained demand, overall investment in energy won’t have to increase, despite the transition. “Major investments need to be made”, explains Gardner, “but the amount of money the world spends on energy does not change much. The total number will be lower relative to GDP. That’s a pleasant surprise: we can afford the transition.”
It’s 33 years from now until 2050. Very few of the coal, oil and nuclear plants in operation today will be in operation in 2050. Stuff gets old and wears out. The average lifespan of a coal or nuclear plant in the US is about 40 years.
We’re going to replace those worn out plants with something. The installed cost of wind and solar is much smaller than the installed cost of coal plants.
Then when you add in fuel and health care savings we should be spending far less for energy per year than what we have been spending and would have spent were we to continue down the fossil fuel path.
Steve Heins says
The Paris Agreement is a voluntary accord, with many unreliable countries, no accounting standards or independent verification.
Helmut Frik says
And pressure from the whole world when some states do not comply with their promises. How many states will value higher CO2 emissions so much that they are willing to run in trouble with everybody else around the world? Trump already could not move anything in his favor at the last G20 meeting. Being Paria is a problem even for big states – and at todays costs for renewables not even producing any economic benefit.
Alastair Leith says
Most unreliable proving to be the US of A. But the mayors and governors will stand into the breach.
Nigel West says
“As a result, renewables will expand to make up 44% of primary energy supply by 2050 and 85% of electricity supply, at the same time that the share of electricity in total energy supply will rise dramatically from 18% today to 40% in 2050.”
Is DNV really predicting world electricity supply will comprise 85% renewables by 2050? If they are the major economies would need to commit to that target now and few have. Decarbonising to that extent using intermittent sources would be a monumental task, as noted by critiques of Jacobson’s studies. Also auction processes for renewables deployment now in use are slowing down deployment of grid scale renewables.
In the best locations solar’s CF is 20%, wind around 35%. Storage is not up to the task or economic in the northern hemisphere to enable renewables to reach a contribution of 85% of electricity supply. Politically a worldwide supergrid is not possible either even if the technical and economic challenges could be overcome.
Renewables costs based on LCOE have now reached attractive levels compared to FF. However it would be a mistake to assume that a simple comparison of LCOE cost of generation will solely determine future generation investment. LCOE is not a good method of determining the overall cost of integrating generation sources into a system.
Countries that have adequate generation capacity to meet peak demand can replace closing coal-fired power stations with intermittent renewables. That is a window of opportunity for renewables but is limited as countries must have adequate firm generation capacity to meet demand. The UK has reached this point already. UK Government announced this week that off-shore wind Round 2 CFDs have hit £57/MWhr. Building renewables only would not work for the UK so calls to abandon new nuclear are misplaced.
Firm gas fired plants must have a big and continuing role to play in replacing closing coal and nuclear capacity to cover for renewables intermittency.
Bob Wallace says
“Decarbonising to that extent using intermittent sources would be a monumental task, as noted by critiques of Jacobson’s studies. ”
Between now and 2050 (37 years from now) most of the world’s fossil fuel and nuclear plants will wear out and have to be replaced. With something. The issue is not one of a monumental task, but which technologies to use for the replacement we will have to do.
“In the best locations solar’s CF is 20%, wind around 35%. Storage is not up to the task or economic in the northern hemisphere to enable renewables to reach a contribution of 85% of electricity supply. ”
Your numbers are incorrect as is your claim. US utility solar with single-axis tracking has a CF of roughly 30%. US wind turbines installed after 2013 are averaging above 40% CF. In the best locations CFs are over 50% and heading to 60%. Very little grid storage will be needed to reach 80%/85% RE penetration. If we develop nothing better pump-up hydro can give us the long term storage we need.
” LCOE is not a good method of determining the overall cost of integrating generation sources into a system.”
That is correct, but at this point in time a grid based on new wind and solar with storage and NG fill-in is cheaper than new coal or new nuclear with their backup needs and load matching difficulties.
ect.
Hans says
Also the CF of PV is defined in a misleading way. For other power generators the CF is defined relative to the maximum power output to the grid. For PV systems it is defined relative to the nominal power of the PV modules. However, this is not the maximum power output of the system. The maximum power output is determined by the maximal AC-output of the inverter. This is mostly lower than the nominal module output. Therefore, if the CF of PV would be relative to the actual maximum output of the system, and thus defined consistent with that of other generators, the CF would be higher.
Nigel West says
You cherry pick renewables data based on the best solar and wind conditions seen in a few parts of the world to make your long term predictions. California is not northern Europe.
Wind turbines that are hitting high CFs in Europe are under rated machines for off-shore duty. In future machines will not be so constrained as they will have higher ratings. They will produce more energy, but their CFs will be <50%. The intermittency problem remains as the weather can't be changed by improved technology.
Decarbonising the world cannot be done with renewables alone. Nuclear is needed too. That's not just my view but the view of 21 leading US academics:
https://blogs.scientificamerican.com/plugged-in/landmark-100-percent-renewable-energy-study-flawed-say-21-leading-experts/
BTW this expert report does not support your fantasy views on pumped storage development in the US.
Bob Wallace says
I was replying to this statement –
“In the best locations solar’s CF is 20%, wind around 35%.”
I replied –
“Your numbers are incorrect as is your claim. US utility solar with single-axis tracking has a CF of roughly 30%. US wind turbines installed after 2013 are averaging above 40% CF. In the best locations CFs are over 50% and heading to 60%.”
Now, as to the “21 experts”.
You might want to read Jacobson’s rebuttal to their paper. It appears that they are not very expert. Or truthful.
http://web.stanford.edu/group/efmh/jacobson/Articles/I/CombiningRenew/Line-by-line-Clack.pdf