Source: Bloomberg New Energy Finance. Note: Battery-electric and plug-in hybrid light duty vehicles
Rapid technology change forecasts vs. more cautious predictions based upon energy transition history
If you happen to come across a Big Oil executive with a long face today, his or her gloom is probably caused by reading the newest report from Aurora Energy Research, which forecasts peak oil demand by the mid-2020s and $19 trillion in lost revenue for oil companies thanks to rising fuel efficiency and the rapid spread of electric vehicles. My advice? Buy that poor executive a scotch and show them this column.
Just how bad could things get for the global oil industry? Well, according to the Oxford, UK-based researchers, pretty awful:
Demand destruction from electric vehicles and improvements in fuel efficiency leads to oil demand peaking in the mid-2020s. As it becomes clear that oil demand is in decline, and a barrel of oil may be worth more today than in the future, OPEC’s strategy changes from one of restricting supply in order to prop up prices, to a strategy of increasing production to gain market share. With demand declining in the 2030s, the Burnout scenario predicts that oil prices could fall to $32 per barrel in 2040 (in today’s money) – in line with the long run historic price and less than half the current price.
“Our new analysis points to a possible energy future of mass electrification, digitalisation, and new technologies, in which the rise in electric vehicles and continued improvements in fuel efficiency lead to peak oil demand occurring in the mid-2020s, and oil prices falling to less than half their current level by 2040,” Richard Howard, head of research at Aurora, said in a press release.
“Indeed, this flips the very idea of ‘peak oil’ – previously hypothesised for the supply side – as electricity grows in importance as a transport energy source.”
Talk about a worst case scenario for Alberta.
A recent Canadian Energy Research Institute study forecast that by 2038 Alberta oil sands output could rise from the current 2.8 million b/d to 7.5 million b/d under a high price scenario or 5.5. million b/d under a business as usual scenario.
The CERI scenarios require global oil consumption to continue rising as forecast by the International Energy Agency and many industry forecasts (e.g. BP, Shell, ExxonMobil) from current consumption of 100 million b/d to as much as 117 million b/d by 2040.
Which of these two scenarios is most likely? The Apocalyptic end of the crude oil world envisioned by Aurora (described as “a leading independent energy market modelling and analytics company founded in 2013 by University of Oxford professors and economists”) or the long-term expansion fueled by rapidly growing Asian economies assumed by CERI?
Which scenario you think is most likely depends upon how one views technology change.
If technology behaves like smartphones – which experienced phenomenal sales growth after their introduction in 2007 and came to dominate the global market within just a few years – then technology adoption in the modern age has sped up exponentially and S-curves have flattened out.
Dr. Howard and his Aurora colleagues are clearly in this camp, as his analysis suggests “rapid technological shifts are likely to prove more effective in combating climate change than the national carbon targets set under the Paris Agreement.”
He has plenty of company, academics like Tony Seba (who in a 2017 study argued that by 2030 95% of American miles traveled will be in autonomous EVs and the Mobility as a Service business model) and popular hypers like Bloomberg New Energy Finance (“Our latest forecast shows sales of electric vehicles (EVs) increasing from a record 1.1 million worldwide in 2017, to 11 million in 2025 and then surging to 30 million in 2030…”) who pump book sales or pricey subscription models with ever more outlandish EV adoption forecasts.
Then there are sober researchers like Dr. Jessica Jewell, a research scholar in the energy program at the International Institute for Applied Systems Analysis in Vienna, Austria and a visiting associate professor in the field of energy transitions at the University of Bergen, Norway.
She argues that energy transitions historically have happened very slowly (Prof. Vaclav Smil estimates energy systems change at only 3% a year) and this one appears to be no different, especially the electrification of transportation by EVs.
“When we run our models lots of different ways, the power sector always leads because it’s always cheapest to lead with the power sector. And the transport sector always lags,” she told Energi News in a 2017 interview.
“So it’s hardest to squeeze out carbon from the transport sector unless we have some huge battery breakthrough, which is not totally unheard of. I mean, we could, but we have to build a totally new infrastructure, whereas with renewables, you’re just basically plugging it into an existing infrastructure and then increasing the smart grid’s capacity a little bit.”
The pace of technology adoption is determined by the complex interplay of “accelerators” (those things that speed up adoption, e.g. declining cost curves for EV batteries) and “constraints” (those things that slow down adoption, e.g. EV range anxiety caused by low energy density of current battery technology).
As adoption of new technology speeds up over time, we would expect accelerators to become more dominant and constraints less so.
For the foreseeable future, the constraints to EV adoption are much more powerful than the accelerators. Given the commitment and high levels of capital investment by governments, automakers, and technology firms the pace of EV diffusion will almost certainly speed up significantly sometime in the future, probably in the late 2020s or early 2030s when the next step change generation of batteries appear in electric cars.
Then it will take decades for EVs to displace internal combustion engine vehicles from the global auto fleet and electricity may begin to displace oil during the 2050s or 2060s.
This is the future as envisioned by Dr. Jewell and many of the scientists, economists, and EV analysts interviewed by Energi News over the past few years.
If nothing else, we should balance the fantastic predictions of Aurora Energy Research, Tony Seba, and Bloomberg with the more cautious approach of Dr. Jewell.
Public policy and private investment decisions – like the billion dollar decisions routinely made by oil and gas executives – should be based upon evidence, data, and common sense, not hype.
Note: A number of readers have objected to my conclusion, pointing out that it sounds like I’m dismissing the Energy Transition and the threat EVs pose to oil. I’ll concede the point. Let me clarify with the forecast I describe in my speaking presentation, “Electric Vehicles and the Future of Alberta Oil.”
Under a business as usual scenario, I would expect a series of technology step changes in EVs that would make electric cars really competitive with ICE vehicles in the late 2020s/early 2030s, meaning EVs would likely arrive at the tipping point on the adoption S-curve around 2035. From that point on it’s a question of how quickly EVs replace ICE vehicles in the global auto fleet. If we assume that ICE cars average 15 years on the road and the fleet would have to turn over twice, then EVs become market dominant (70% to 80% market share) around 2070.
But there are four disruptions (super-batteries, Mobility as a Service, the impact of China manufacturing, policy support for electrification of transportation) that could significantly change that timeline. If all four disruptions happened, and their effect was widespread and deep, then EVs might become market dominant sometime in the 2040s, maybe the early part of that decade.
Under that scenario, peak oil demand probably does happen mid-2020s and the global petroleum industry is in big trouble. If an over-supply of just 1.5 million to 2 million b/d collapsed oil prices in late 2014, what would the effect be on the industry of losing five or 10 or 20 million b/d by 2030?
The better conclusion to this column is that predicting how those disruptions – and others that might arise in the near future – will unfold is tricky business because it’s still early days for the EV industry and policymakers.
But I think the better way to think of the diffusion of electric vehicles and their impact on the oil sector is as a continuum from 2030 to 2070, then to unpack the stories behind the disruptions, which will enable us to at least say something intelligent about the topic rather than simply repeating the conclusions of (suspect) studies.