Warn anyone in the USA about the coming energy crisis and you’re likely to see eyes roll. “What energy crisis? That was half a century ago! Markets and technology won. Today we’re back among the top oil suppliers!”
All true, but the response gives a false sense of security that has policymakers and publics sleepwalking toward a cliff. An energy crisis is likely ahead, no matter our rank (currently third) among oil supplying nations. Seeing the coming crisis requires looking beneath the veneer of oil supply claims and asking some deeper questions.
The issue of energy scarcity is important because energy, fossil or otherwise, is tightly tied to economic output. A prolonged energy crisis—one in which substitutes for scarce energy are too expensive, environmentally harmful, or beyond humanity’s technological capabilities—would likely put an end to growth of industrial economies. While degrowth to a steady state economy is what steady staters seek, a lengthy and substantial period of degrowth would be a nightmarish outcome that would produce unnecessary suffering and conflict.
Revisiting an Energy Price Shock
Fifteen years ago, the world suffered an economic downturn that required intervention by central banks and governments to prevent a depression-level economic collapse. Economies in much of the world had been expanding since 2001, but deregulation of mortgage lending produced a bubble that sent shockwaves through the global financial and economic system. This much is well understood, but what is little recognized is the pin that burst the bubble. The pin was the price of oil, which doubled between 2007 and 2008.
Oil prices rose because world petroleum output could not keep up with demand. Because oil is a “master resource” used for energy or as a feedstock in practically every economic sector, rising prices caused the global economy to slow. The over-leveraged housing industry, already vulnerable because of a lack of credit-worthiness of some buyers, began to unravel. And although government intervention has been extensive, GDP growth after the 2008 crisis remained tepid.
But the economy recovered as energy extraction picked up, underlining the critical importance of energy to the economy. With higher oil prices and a period of low interest rates, extraction relying on a new technology—fracking—was added to the oil industry’s toolbox. Fracking increased oil supply by opening access to so-called “light tight oil” from mid-continental U.S. shale deposits. This period of increased supply is known as the “shale boom.” It made the USA a major supplier, and concerns about energy supply slid into the rear-view mirror. Indeed, headlines about ”Peak Oil” that were common before the Great Recession of 2008-09 soon disappeared as fracking opened up supply.
The last decade, however, has brought new attention to the limits of energy availability and has shown that the shale boom may be short-lived. Because of oil producer obfuscation (particularly on the part of OPEC) we are still unsure of total global oil reserves, and by a terminological sleight of hand (described below), what was once considered oil has changed meaning, adding to confusion about reserve totals.
An Accounting Problem
Oil accounts for about 40 percent of total global energy consumption. Given its critical importance to economies worldwide, you’d think estimates of the remaining stocks of oil would be a settled matter. Yet experts have offered a wide range of estimates for decades. Determining the remaining recoverable reserves of countries and of the world is difficult for several reasons.
First is a lack of transparency; producers are reluctant to disclose the extent of their assets, or they wish to exaggerate them for greater global influence. Oil analysts have been suspicious of some producer claims for many years. A recent analysis suggests that OPEC reserves are overstated by 300 Gb (billion barrels), and FSU reserves by 100 Gb. (The reduction in OPEC reserves would align with the long-held theory to explain the “mystery” of sudden reserve additions in the 1980’s—the additions were likely a maneuver to increase export quotas.)
Fossil fuels continue to dwarf renewable sources. (Our World in Data)
Another problem in counting oil reserves results from conflating heavy oils with conventional oil. Heavy oil resources are plentiful, but less economically useful than conventional oil, and extracting them is economically (and environmentally) costly and difficult to scale up. Yet heavy oils are counted in production as though they were equivalent in quality and accessibility to conventional oil. In fact, because they are harder to extract, their “flow rate” is limited and they cannot provide significant spare capacity in times of need.
Shale oil also complicates the oil accounting question. The USA is endowed with the best oil shale deposits (for oil production) on the planet and has more than doubled its production over the past ten years. This output has boosted all-liquids fossil fuels production and helps to explain how world demand has been met over the last decade. As seen in the figure, conventional oil has plateaued. Nearly all new additions to consumption have come from U.S. tight (shale) oil.
Shale oil is beset by several problems, however. One is the daunting and capital-intensive nature of the extraction process. Unlike extraction in conventional fields, fracking shale for oil is a constant effort involving drilling down one to three miles, then laterally for miles more before hydraulically fracturing the shale (injecting fluids under tremendous pressure containing large quantities of sand to keep the fractures open), and finally, pumping the liberated oil out of the deposit. This must be done continually to exploit a field.
A second problem with shale is the nature of the “oil” produced. Analysis by petroleum geologist Art Berman indicates that fully 30 percent of reported oil production in the USA, much of it from shale, is natural gas liquids—light hydrocarbons that have significantly less energy content than conventional crude. The light grade of oil is not suitable for heavy transport that relies on diesel. So, much of the fracked shale oil produced by the USA cannot be used in the country and must be exported, so it does not contribute to U.S. energy supplies.
Growth in U.S. oil production since around 2005 has come from hard-to-get sources (blue and yellow). (Art Berman).
The other problem for shale is the spectacular decline rate of a typical well. A conventional well may have a decline rate of 6 percent per year after peaking, but fracked wells plummet dramatically from the start, with a decline rate of 60 percent in the first year and 25 percent the second.
This means that companies need to drill new wells continually just to maintain production at a constant level. This is “The Red Queen” predicament, after Lewis Carroll’s “Through the Looking Glass,” in which the Queen advises Alice to run as fast as possible just to stay in place. It’s a very apt metaphor for fracking.
If world oil demand continues to increase as expected by energy advisory bodies such as the International Energy Agency and the U.S. Energy Information Agency, fracked deposits will have to perform increasingly well in the years to come. Yet production at two of the major oil shale plays—the Eagle Ford in 2013 and the Bakken in 2020—has apparently peaked, leaving only the Permian Basin as a prospect for expansion.
In sum, given the plateauing of conventional oil, the likely exaggeration of some countries’ reserve levels, and the rapid decline of fracking as a strategy for boosting conventional output, a peak in total liquids should be of urgent concern to policymakers and the public.
The Net Energy Cliff
The lengths humans will go to extract oil illustrates its value as an energy source. Gone are the days when an explorer could stick a pipe in the ground and hit a “gusher.” Today we drill miles deep in the ocean, mine dirty oil sands, and crack open deep rock (fracking) to find oil. But these increasingly extreme measures themselves require increasing inputs of energy.
This raises a key question: How much energy is being expended to get various forms of energy? What is the energy cost of energy production? Analysts studying the question developed the concept of “energy return on energy invested,” or EROEI, to answer this question. The measure expresses the energy in the extracted resource compared to the energy cost of its exploration and development.
The Net Energy Cliff. (Adapted from Wikipedia)
For example, extraction of 50 units of energy in oil (as in historic oil and gas fields) may require one unit of energy, for an EROEI of 50 to 1. But over time, as oil extraction requires increasing effort, oil’s EROEI might fall to 30 to 1, then 15 to 1. Declining EROI is precisely what characterizes the current state of fossil hydrocarbon extraction, as the graphic shows.
The implications are staggering. A declining EROEI reveals that extraction of energy will be increasingly expensive and eventually, cost-prohibitive. Hydrocarbons will still be in the ground, but the costs of their extraction will continue to climb. This also means that, barring the development of some new type of energy source, society will have to adapt to a much lower energy future. And it suggests that the monetary costs of extraction will erode GDP growth and eventually cap economic expansion.
Action is Needed Now
In 2005, just a few years before the rising price of oil triggered the 2008 economic crisis, the U.S. Department of Energy commissioned a report from the think tank SAIC titled “Peaking of World Oil Production: Impacts, Mitigation and Risk Management.” It’s clear from interviews that the authors were shocked by the implications of soon-to-arrive global Peak Oil, which they termed “an unprecedented risk management problem.” Analyzing the supply and demand side of the oil scarcity challenge, they concluded that at least a decade, and more likely two, would be needed to prepare for Peak Oil and prevent social and economic upheaval.
The report garnered a great deal of attention at the time, as did other warnings of energy limits. But the subsequent “shale revolution” changed everything. Instead of being recognized as a last domain of exploration and recovery, the media framed shale and fracking as an energy elixir. The intervening years have not produced the preparatory planning that Hirsch warned should occur.
Perhaps a peak visible on the horizon will draw attention to the predicament we are in: Perpetual economic growth cannot be reconciled with energy limits. The longer we wait to act, the higher the cliff, the more painful the landing, and the more difficult the transition to a steady state economy. Some local communities have been planning for energy scarcity, and I will share their work of conservation and adaptation in a future post.
Comments
Welcome to the new Resillience commenting system!
We have moved from the former Disqus commenting system to a new system as of September 10, 2026.
We made this change because of privacy concerns regarding Disqus. It's also part of a set of changes to bring more services to the Resilience community under a single login: commenting, (coming September 2026), and online courses (coming 2027). Unfortunately, your old Disqus comments may not reliably link to your account in this new system. Please contact us if you have any questions or concerns.
Good information, & thanks for the valuable perspective. I just wonder whether, by “involuntarily,” you mean “voluntarily,” although I am confident the reduction in use of FFs will be involuntary.
I did mean “involuntary”. As I see it, there is no way to rapidly reduce fossil fuel use without catastrophic consequences to economies and lives, so it will never happen voluntarily (the reason it hasn’t happened already). Only an involuntary collapse of fossil-fueled modernity has any chance of saving the climate.
But after reading James Hansen’s latest papers I fear it may already be too late (two very important papers – Global Warming Acceleration: Causes and Consequences; Global Warming in the Pipeline). In any case, the sooner carbon emissions stop, the less damage done.
Ideally, I always prefer to back up claims with evidence in the form of documentation and citations. But sometimes finding the relevant data can take a really long time! So this time I’ll merely say that some experts in EEE (energy, economy, ecology) have been saying that a very, very small sliver of world population is simply overwhelmingly responsible for “excessive emissions” (greenhouse gas emissions from fossil fuel use). I believe Jason Hickel is one of these people. https://www.jasonhickel.org/about
Kevin Anderson (the climate scientist) has said similar things about the percentage of the world population which contribute the bulk of emissions — a tiny portion of the total of us, mostly.
I’m not remembering with clarity what those percentages are, but it struck me as evidence against the premise that a deep cut in energy use would necessarily result in economic, social and political catastrophe. It seems that we have a good deal more wiggle room than many commentators suspect. After all the overwhelming majority of current emissions are produced by a tiny sliver of world population, and a fraction of the population even in the Global North / rich countries.
I’m hoping someone here will remember the data sources on this.
A rapid reduction in fossil energy use by itself would accelerate warming–once we lose the shading/cooling effect from combustion particulates.
The waste issue is largely a policy problem. If you remove uranium and stable isotopes from 1000 kg. of spent fuel (not counting cladding), you’re left with 10 kg. of heavy transuranics (we are developing fast reactors to consume these) and 8 kg. of fission products that are still active after 10 years. Some of those have uses, but even if we decide to put all active fission products down Deep Isolation boreholes (a viable option which is currently illegal) by 400 years, it will be about as radioactive as some common granites.
There are several Gen 4 designs which have more promising economics than old-tech nuclear. In the U.S., Kairos is the frontrunner for a Gen 4 SMR. They think they’ll be able to hold the initial cost of their fuel to under $10,000 per kg. U, with 20% burnup. That would translate to a fuel cost of about $2.25 per megawatt-hour for heat, or $6.65 per megawatt-hour for electricity (at 34% conversion efficiency). So their starting point on fuel cost would be competitive against hydrocarbon fuels, and they think they could eventually get the fuel costs down to under a fifth that.
Plant cost will take a while to pin down, but molten salt reactor developers seem to be generally aiming for under $4000 per kW(e) capacity for build costs plus enough part and refurb costs to get the plant to 50 years operation. Because heat is the native mode for nuclear, that would make it more cost-competitive for replacing fossil heat directly. Using nuclear heat to generate electricity, about 2/3 of it will be rejected. Most Gen 4 reactors will run hotter than our water-coolant reactors, which makes the reject heat more usable for secondary applications.
One potential secondary application would be direct-air CO2 removal. Some DAC systems are already down to around 1.5 MWh(th) per tonne CO2 captured, so a megaton-scale DAC plant would require around 170 MW average power–mostly as heat to separate the CO2 from the capture medium, and some for moving large amounts of air. So a DAC system that piggybacks on, say, a 170 MW(e) air-cooled nuke, could get most of its energy needs met from reject heat and the cooling fans–potentially enough for nearly 2 million tonnes per year CO2 capture–assuming no further improvements in DAC efficiency. (Generating the same electricity using coal releases more than a million tonnes CO2 per year on avg.)
I’m pretty sure the metal requirements for a small nuke would be much less than for an average container ship, and there are currently around 50,000 of those in service. Concrete requirements should be much less than for river dams, and there are 57,000 of those. If we were to build 50,000 small nukes, that might be something like 8 or 9 terawatts (e) of capacity, which is roughly what you suggested. If half of those were paired with DAC systems, that could be as much as 50 billion tonnes CO2 removal per year. If we can get to 40 billion net-negative tonnes per year, we could return to pre-industrial CO2 levels in 25 years at that pace. It would not be an easy lift, but we’ve done larger-scale production in multiple areas before–based on profit motive alone. I certainly don’t see it as hopeless.
Removing uranium from irradiated fuel is reprocessing, the dirtiest technology anywhere and a key way to extract plutonium (which makes nuclear weapons). Putting transuranic in reactors is not only very dangerous but it doesn’t make them go “away” any more than dumping garbage into a river makes it go away. Transuranics have several decay products they transform into over time, sometimes lots of time.
Putting fission products into deep holes could never contaminate groundwater, nope, no way, honest, trust us … except there are many examples of this problem happening in reality.
It also takes a lot of fossil fuels to make reactors, mine uranium, mill it, enrich it, operate reactors and then how much energy will it take to babysit the waste for longer time periods than industrial civilization has existed? Pu-239 has a half life of more than 24 thousand years. The arrogance is surreal.
By dissolving spent fuel in molten salt, uranium could be selectively extracted by fluorination (uranium hexafluoride is a gas, which can be easily separated from a liquid). This is very different from aqueous processing–which is indeed messy.
The uranium could be removed without separating out any plutonium.
Reactor-grade plutonium does not make nuclear weapons. They use weapons-grade Pu for that.
They could be consumed completely in molten salt fast reactors. After they are consumed, all that will remain are the fission products. Some of these are born stable. Many will decay to stable quickly. Over 80% by mass will not be radioactive after ten years. Of the remainder, nearly all the hazard would reside in just two isotopes–cesium 137 and strontium 90. We actually have uses for these, but even if we just want to dispose of them, by creating them in molten salt, they will immediately be bound into the salt (as chlorides or fluorides) and will never have the volatile, highly-mobile, elemental phase which makes them such a notable hazard in today’s spent fuel.
The transuranic isotopes would cease to exist once they are fissioned. That is nothing like dumping garbage in a river.
If we wait for them to decay to stable on their own, that process will take many billions of years. Or we could quickly eliminate them by fissioning them.
Cesium and strontium can be immobilized by locking them in solid salt, which can then be incorporated into synroc (like glass) molded to fit into Deep Isolation-style canisters, which will be emplaced in clay mud (to provide balancing geostatic pressure) inside of steel wellbore casing, inside of geologic formations which have been tested to confirm they have been stable and isolated for many millions of years. And the cesium and strontium will become a thousandfold less radioactive each 300 years. Can you think of a plausible scenario whereby cesium and strontium could spontaneously break their salt bonds, escape the synroc, escape the canister, migrate through the clay mud, penetrate the bore casing, and then travel up to a kilometer upward through impermeable geologic formations to reach groundwater–all in less than 300 years? There is probably nothing in your life that has that level of safety.
Even today’s reactors need less materials and less embodied energy than windmills or solar panels per unit of electricity generated. And better reactors could greatly improve on old reactors.
We can mine uranium using electric pumps with in-situ leach mining–which is how the majority of uranium is mined today. And techniques for mining from seawater have been improving quickly over the last 20 years, with the latest advance nearly reaching cost competitiveness with land mining. And molten salt fast reactors could use spent fuel and depleted uranium for fuel, so they wouldn’t need fuel mining for hundreds of years.
That’s already an electric process. It’s as clean as electric cars.
That actually produces energy.
Once fission products are permanently sealed in a deep geologic formation, they won’t need any further looking after. And civilization has lasted a lot longer than the few hundred years it would take for fission products to lose nearly all their radioactivity.
Not if we consume it in molten salt fast reactors.
Passing judgment prior to understanding is, itself, a kind of arrogance.
How about those small nukes? $89 per MWh(e).
https://www.reuters.com/business/energy/cancelled-nuscale-contract-weighs-heavy-new-nuclear-2024-01-10/
But even if all your cost numbers pan out, generating the electricity is only the start of the transition. Also needed are tripling of the electrical grids, replacing all space heating with heat pumps, replacing all ICE vehicles with EVs, etc.
Transitioning to a non-carbon energy system for a global industrial civilization is a big project, one that should have started decades ago at the latest. It’s far too late for an energy system re-do. It’s going to take every morsel of fossil fuels we can find just to maintain what we have now and even that’s looking iffy. Peak oil is still looking like it was 2018. The energy cliff is looming.
https://dothemath.ucsd.edu/2011/10/the-energy-trap/
Yeah, I don’t think small Gen 3’s are going to work out. But that says nothing about the prospects for small Gen 4’s.
Yep. That still leaves industrial heat applications, heavy transport, and the big one: removing hundreds of billions of tonnes of CO2 from the atmosphere. But nuclear energy could be adapted to help with each of those.
That’s at the latest for an ideal and smooth transition. But we can only go from where we are. So it likely won’t be an ideal and smooth transition.
But it isn’t here yet. As Murphy notes, the critical measure is not EROI but energy payback time (which is where fossil fuels have their advantage–since most of the fossil fuel infrastructure has already been built). Several of the advanced reactor teams are aiming for 2 year build times and less than 1 year energy payback time. That reduces the deficit time to 3 years, after which, each reactor could power the build of another reactor every two years–assuming 50% of the energy output goes into general usage and 50% is dedicated to fleet growth. And with each new reactor, the fleet growth-rate capacity would also increase, even while the amount of energy going into general usage would also be increasing. If this growth rate can keep up with the decline in fossil fuels, the energy cliff (and the energy trap) can be avoided.
3. Risks beyond waste disposal
4. A very long wait from planning stage to completion of construction. Last I heard, this takes a minimum of ten years. It may now take longer.
5. Energy costs of construction, mining and processing of uranium.
6. Toxic effects of uranium mining and processing
Item 4 is of crucial importance, because ten years is a very long time for continuation of business as usual. The emissions over this ten year period are decisive for future climate conditions. We need to drop emissions immediately and dramatically, not ten years out.
Hence my reference to new ways to derive energy from old sources. New kinds of nuclear could avoid or greatly reduce the major problems with old-tech nuclear.
Problem is electricity is much less useful than oil, even if the dream of nuclear power were possible. The musk monster loves to feed these fantasies, (for financial and ego reasons).
This article restates much of what’s been written for over a decade, though it is a fairly good summary. In that decade, none of the basic physics** has changed, even though people still believe in saints Tesla and Tech.
_______________________
** e.g., a mole of electrostatic bonds in a battery physically contains about 10x less energy than a mole of covalent bonds in a hydrocarbon.
Nuclear electric plants produce both electricity and rejected heat. Reject heat can supply a large portion of the energy requirements for direct-air carbon removal–which could help to offset the CO2 emissions from oil. Carbon Engineering (acquired last year by Oxy Petroleum) has also demonstrated combining CO2 with hydrogen to produce syn-crude, from which several liquid fuels can be refined. The synthetic diesel and aviation fuel analogs burn much cleaner than their fossil counterparts (which could help reduce contrail persistence, and the warming that causes). A lot of the heat energy used to drive syncrude production could also come from waste heat. A Navy study looking into producing jet fuel from seawater concluded it would cost around $7 per gallon using Navy nuclear energy and small ship-board production plants. That was too high to be competitive (and the production volume would have been too small), but cheaper nuclear energy combined with large scale processing and refining would likely reduce that cost substantially.
Okay, Joe. I was suggesting that the enormous difference in energy use varies as much within any nation as between nations — which suggests that we can live with less energy, if we were to so choose. Your point appears to be that we’ve not been choosing otherwise. This is obvious. My point is only that we could live with vastly less energy if we chose to do so.
I don’t think that would require draconian domination on the part of government “leadership”. It would, however, require a profound shift in our political and social lives — and economic lives. But it’s doable.
It’s merely unlikely — and mainly because (I believe) we’re caught up in cultural, social and political inertia. And most people are ignorant of these matters, and don’t seem to care much, anyway. THAT’s our problem.
“This also means that, barring the development of some new type of energy
source, society will have to adapt to a much lower energy future.”
Or how about new ways of deriving energy from very old energy sources? Seems like some of the new kinds of nuclear being developed could work for that. The energy potential there is far greater than it ever was for hydrocarbons. And if we actually want to combat long term global warming, we’ll need to remove hundreds of billions of tonnes of CO2–which would require a much higher energy future.
Fracking postponed rationing.
Just curious. Is it possible for the developed world to make changes in energy use (say with our 1.6 billion people) such that it can offset the demands of the developing world (6.5 billion) even if their demands are a fraction of ours? It would appear that even if their demand was 1/5 of ours, they still would be equaling ours. Keeping in mind their population is growing by 80 million a year and their desire for more energy is also rapidly growing as well, we are not the sole source of this problem of climate change and resource depletion—along with everything else. I work hard at being energy aware and have cut back, but do not feel I can offset the impact of their massive growth rate. On a side note, the sheer trampling of human feet in the over-crowded nations is devastatingly taking out all forms of the natural world. India, one third the size of the US, has 1.4 billion people. Help me..