From Tilak´s Substack

By Tilak Doshi

A review of “The Energy Trap: Why the Renewable Energy Transition Can’t Work – and What Can” by Richard Lyon (Swift Press, 2026)

Richard Lyon’s The Energy Trap is a short book with large ambitions: to strip the energy debate back to first principles and make the key policy issues accessible to the intelligent reader. He describes it as “a friendly guide that gives you the minimum set of facts you need, with references so you can check the claims yourself. … You’re perfectly capable of understanding what’s happening if someone takes the trouble to explain it clearly.” The result, as he says in his prologue, can be read in “three cups of tea and half a packet of biscuits”.

The author combines training in electrical engineering, petroleum engineering and energy economics with more than 25 years in senior upstream oil and gas roles in the UK, Norway, Azerbaijan, Congo and Cameroon. Having spent much of my own career in the industry at the downstream end, in corporate planning where the outlook for demand, supply and prices is the bread and butter of the analyst, I read the book with great interest. What insights could an upstream engineer and energy economist who worked on resource assessment, field development and extraction decisions – working at the ‘coal face’ of the supply side, so to speak – bring to corporate planning?

Physics, depletion and escaping the energy trap

The book argues its case in three broad strands, spread across seven short parts in 110 pages of main text. First comes the physics of energy. “If a policy or a technology cannot satisfy the non-negotiable laws that govern how energy is captured, converted, stored and dissipated, it does not matter how elegant the spreadsheet or how passionate the advocacy: it will fail.” Useful work depends not just on the quantity of energy available but on its quality – what Lyon calls the energy gradient, energy density and areal power density – together with the distinction between a fuel and an energy carrier. On these grounds wind and solar – being diffuse and intermittent – cannot substitute for hydrocarbons at industrial scale. The first part of the book’s subtitle, “why the renewable energy transition can’t work”, is explained as the paradox of renewable energy “in which we discover that renewables aren’t renewable, and that building the replacement for fossil fuels requires more fossil fuel than it saves”.

The second part of the argument deals with depletion. Lyon argues that “not all oil is equal” – “the fact that the ‘peak oil was wrong’ narrative has never understood”. Fossil fuels are the “depleting inheritance that no energy policy acknowledges”. According to Lyon, conventional oil has already passed its peak on the curve first described by Hubbert. “Economic oil – the cheap conventional oil that built the modern world – peaked around 2006.” He contrasts Saudi conventional crude, which he estimates costs about $10 a barrel to extract, with more expensive US shale, where new wells need over $60 a barrel to break even. Lyon estimates existing Canadian oil sands operations break even at roughly $30-$40 a barrel. The massive increase in unconventional oil driven primarily by the US shale revolution – which kept global oil production growing – was, in Lyon’s words, made possible because “cheap capital changed the arithmetic. With borrowing costs near zero, companies could drill at a loss and refinance indefinitely.” In his Daily Sceptic interview, Lyon says: “It was having vast amounts of cheap imaginary money thrown at it. … But it is not real oil, it’s uneconomic oil. It’s oil fuelled by imaginary money.”

Given that today’s renewables are unable to provide an alternative, and with depleting fossil fuel resources, we arrive at Lyon’s “energy trap”. Renewable energy – which cannot replace fossil fuels as a sustainable alternative to support modern industrial civilisation – depletes our fossil fuel inheritance even faster. Wind turbines, solar panels, batteries and their components are hugely dependent on fossil fuel inputs for mining, refining and final delivery. This is a problem because we need the fossil fuel pipeline to last as long as possible while we build a truly sustainable energy system.

In Part VII, entitled “Forging a New Realism”, Lyon explains how “escape from the trap is possible”. Three things are required:

The first is that there exists exactly one proven technology capable of moving civilisation up the energy ladder: nuclear power. The second is that even with nuclear, some contraction in total energy use is unavoidable – the physics constrains not just what we build but how fast we can build it. The third is that navigating the contraction requires a blueprint built on what the physics permits, not on what the politics would prefer.

This move towards a sustainable, largely nuclear energy future will involve reducing total energy throughput as “some contraction in total energy use is unavoidable”. He states baldly:

What the physics now demands is a transition up the quality ladder while total throughput contracts [italics in original]. … The honest question is not whether living standards will adjust – they will – but whether the adjustment is a managed descent into a durable, lower-throughput civilisation, or an unmanaged collapse into something much worse.

An education in the basics of energy

Much of the first part – the physics – will be familiar to energy economists who have long regarded much of the energy-transition literature as, in Mark Mills’s phrase, “an exercise in magical thinking”. For example, a kilogram of diesel holds roughly 44 times the energy of a kilogram of the best lithium-ion battery. To carry the same energy, a battery needs far more weight, materials and capital than a tank of diesel. Equally important is Lyon’s reminder that electricity is a carrier, not a source. It accounts for only about a fifth of global final energy consumption; the rest is supplied largely by the direct use of fossil fuels for industrial heat, transport and as industrial feedstock. Steel, cement, ammonia and plastics cannot simply be electrified away.

Lyon describes the stakes in energy policy. Around half the world’s population is fed by crops grown with synthetic nitrogen fertiliser derived from fossil fuels. if we were to stop oil and gas production today, as some climate activists demand, then within a decade, “hundreds of millions of people would starve to death.”. He uses the fertiliser example to convey how much depends on hydrocarbons, but his depletion argument is about oil. Fertiliser depends mainly on gas and coal: over 70% of the ammonia on which it is based is made from natural gas, and most of the rest is from coal, not oil. By his own account, gas reserves stand at about 50 years of current consumption, and he calls gas “a significant buffer”.

On nuclear power, Lyon rightly urges Western governments to strip out regulatory accretions – “multi-year licensing iterations for each new reactor design, dose limits set far below natural background radiation, public inquiries that can outlast a project’s planning consent” – while keeping “the genuinely load-bearing safety standards”. The prize is large. South Korea builds reactors for around $2,200 per kilowatt of capacity, against close to $10,000 for Vogtle in Georgia and more still for Britain’s Hinkley Point C. Hinkley’s cost is now put at up to £35 billion. This amounts to a gap of roughly four to six times in construction cost per kilowatt. Lost construction experience and non-standard designs matter, but regulatory burdens explain a large part of the difference.

Nuclear plays a big role in Lyon’s future sustainable energy supply system:

Nuclear is the next rung on the energy ladder. It is the only energy source available to humanity that is denser, more concentrated and more reliable than fossil fuels. … The fossil inheritance offers a few decades. Uranium, on current reactor technology and at current scale, offers a century… though scaled up to replace fossil energy entirely, that horizon shortens, perhaps to decades.

Breeder reactors using uranium-238 and thorium reactors have not been commercialised at scale but “a serious nuclear programme must include commercialising these cycles”. The small modular, molten salt and high-temperature gas reactors Lyon mentions remain largely at the demonstration or first-of-a-kind stage – outside China and Russia almost entirely so. Nuclear generates electricity, not feedstock, and there are many hard-to-electrify areas in modern industrial economies, as Lyon is aware.

New large reactors remain among the costliest new-build options in Lazard’s annual cost comparisons, and their economics depend on running at near-full output; they cannot be ramped up and down as cheaply as combined-cycle gas turbines.

“Economic oil” and the limits of EROI

The nub of Lyon’s distinctive argument is that oil has already peaked – not in total output, which continues to rise, but in what he calls “economic oil”. In his telling, the cheap conventional crude that built the modern world peaked around 2006, as the International Energy Agency’s World Energy Outlook 2010 acknowledged. The remainder – Canadian oil sands and American shale – is “uneconomic”, kept alive by cheap credit. “It is not real oil,” he told his interviewer, “It’s uneconomic oil. It’s oil fuelled by imaginary money.”

Since the entire “energy trap” thesis rests on this distinction, it deserves careful scrutiny. Lyon’s key metric is energy return on energy invested (EROI): over 30:1 for Saudi conventional crude, he says, against far less for Canadian oil sands (about 4:1). To an economist, however, EROI has little analytical meaning for investment decisions, because it treats energy as the only factor of production. Delivering a barrel of oil also requires labour, capital, land, materials, time and know-how; two projects with identical EROI can have very different costs.

For Saudi Aramco and a private Canadian oil sands operator alike, the relevant corporate finance test is whether the return on capital committed exceeds its cost. If it does, the project creates value; if it does not, it destroys value – something neither a government nor private shareholders will tolerate for long. A low EROI may simply mean that producers are applying capital and technique to a lower-grade resource that takes more energy to extract – a normal, value-creating response to relative prices, as in Canada’s oil sands or US shale. It is the simple ROI (return on investment), not EROI, that matters in corporate finance. At Brent prices of over $100 a barrel in October 2026, existing Canadian oil sands operations (with breakeven prices of $30-$40 a barrel), and new US shale wells (breakeven at around $60-$70) are well in the money. In 2025, Brent averaged $69 a barrel, down from $81 in 2024. For enterprises in US shale and Canadian oil sands operating at favourable, lower cost locations, this is hardly uneconomic.

Averages, margins and Ricardian rent

The author says that there is an “oil price above which the economy contracts”. He writes:

The economy can absorb expensive, difficult oil only as long as a massive base of cheap, easy oil dilutes the total cost. A $120 barrel is viable if the volume-weighted average is $60.

Economic decisions, however, are not made on weighted average cost; they are made at the margin. Every buyer in the world market pays the going price of crude, adjusted for quality (gravity and sulphur content) and location. The finance minister of an oil-importing country cares about the balance-of-payments impact of the marginal barrel, not the average cost of extraction across Saudi Arabia and Alberta. The primacy of the margin over the average is the lesson of the diamond-water paradox, which the marginal revolution resolved.

On the supply side, both the Saudi and the Canadian producer sell at the best price buyers will offer. Buyers will judge the value of the crude by the main refined products it yields – diesel, gasoline, heavy fuel oil, LPG and bitumen. The Saudis, with extraction costs of around $10 a barrel or less, earn a large spread. The marginal producer in US shale or Canadian oil sands earns just enough to pay all factors of production – landowners, labour, lenders and shareholders – their going rates. The difference is what economists since David Ricardo have called economic rent, accruing to the low-cost, inframarginal producer. A shrinking share of cheap oil would raise both prices and rents, and higher prices call forth more marginal supply.

A market that corrects itself

Lyon then describes an amplifying spiral when “average cost exceeds what the economy can bear”: growth stalls; demand and prices fall; marginal oil stops, “not gradually, but suddenly”; production “lurches down”, and the economy, “starved of the energy that production was supplying – contracts again, dragging demand and price lower still and killing more marginal supply”.

This amplifying spiral described by Lyon is at odds with received microeconomic theory of market adjustment responses. When demand has fallen for whatever reason (leaving aside the role of average cost in that fall), the economy does not need the supply that is withdrawn. Shutting in high-cost output is the adjustment, not the disease. As the traders’ saying goes, ‘the best cure for low prices is low prices’. Lyon’s account of the adjustment dynamics inverts the Le Chatelier principle, which Paul Samuelson – the doyen of modern economics – carried over from thermodynamics into economics: a system in equilibrium, when perturbed, generates responses that partly offset the disturbance. Nothing in first-principles physics causes markets to exhibit runaway rather than dampening feedback.

The historical record points the same way. The oil shocks of the 1970s caused global recessions, but they also brought forth new supply from the North Sea and Alaska, substitution away from oil towards natural gas and coal in power generation, more efficient cars and machinery, and conservation. Oil prices collapsed in 1986, after the 2008 run-up, over 2014-16 and again in 2020 with the Covid lockdowns. Each time high-cost output was cut, prices recovered and supply returned. World oil supply averaged some 106 million barrels a day in 2025, on IEA estimates. None of this resembles a doom loop of a perpetually crashing economy.

Shale and “imaginary money”

If quantitative easing made shale oil “imaginary”, the same standard should presumably apply to everything produced during the era of near-zero interest rates: factories, software, housing and, not least, wind and solar farms. Why should monetary conditions render shale uniquely uneconomic?

The claim that shale never made money is based on several missteps. It typically pools the heavy losses of price-crash years with profitable ones, overlooks that early spending was investment for future output, and treats inflated acreage bonuses – a transfer of rent to landowners – as if they were production costs. Companies did overpay in the early land rush and did lose money when prices collapsed in 2014-15. Critics of peak-oil analysis such as Michael Lynch have made the broader point: losses during a price collapse are not the same as an inability of the industry to make a return for shareholders.

The decisive test came after the monetary tap was turned off. The Federal Reserve began raising rates in March 2022 and held them at 5.25-5.5% from July 2023. Yet American crude production rose to a record 13.6 million barrels a day in 2025, up from 5.0 million in 2008. Shale is best understood as a new class of resource opened up by horizontal drilling and hydraulic fracturing: technology, not credit, turned a geological curiosity into a commercial province.

Lyon’s view of “imaginary money” raises its own questions. Rather than the monetarist approach – Milton Friedman’s k-percent rule of steady money growth in line with economic growth – he would anchor money to energy: “Print more money than the energy supply can back and the currency loses value. It is that simple.” What this money supply that “the energy supply can back” should be in practice is not spelled out.

Discovery, reserves and above-ground factors

The book’s quantitative claim is that discoveries peaked in the 1960s, that cumulative discovery has converged on an ultimately recoverable resource of about two trillion barrels of conventional crude, and that the peak is therefore “a matter of arithmetic”. But, as Lynch showed in ‘The New Pessimism about Petroleum Resources’, estimates of field size tend to grow over time, and the fall in discoveries after the 1970s owed much to Middle Eastern governments nationalising foreign operations and cutting exploration while sitting on large surpluses. Discovery is not set by geology alone. ‘Above-ground’ factors such as fiscal regimes, sanctions, war and access to capital decide whether resources are explored and developed. M. A. Adelman – described by MIT as “a giant in the field of energy economics” – argued that reserves are an inventory replenished by investment, not a fixed stock being drawn down. The IEA’s 2025 finding that nearly 90% of upstream investment has been dedicated to offsetting production decline describes a capital-intensive industry replacing depreciating assets. It is not, by itself, evidence of imminent exhaustion.

First principles, Malthus and Julian Simon

Lyon recognises the catastrophic consequences of abandoning fossil fuels abruptly, as climate activists demand. Yet his remedies – conserve every barrel, “electrify everything that you can possibly electrify”, as he said in his podcast interview of his book, and pursue “a deliberate, planned reduction in total energy throughput” with energy “allocated by policy that understands the difference” – amount to planned demand reduction. The motive differs from the climate establishment’s – depletion, not emissions – but the prescription of planned demand reduction is much the same. He would redirect “the subsidies, the planning privileges and the levies loaded onto household bills” to nuclear, geothermal and insulation. Subsidies, in other words, are acceptable when directed at the right technologies. But this is precisely the climate-industrial complex’s argument for subsidising renewables.

Who decides what is ‘the right kind’ of energy? In Lyon’s argument, navigating a managed decline in total energy consumption “requires a blueprint built on what the physics permits, not on what the politics would prefer”. The test on what to invest in for this managed decline is simple: “Does the investment permanently reduce civilisation’s dependence on hydrocarbons, or does it merely shift that dependence onto a parallel system that still requires hydrocarbons to build and maintain?” Nuclear, renewables like geothermal, or eliminating demand – say, through better building design – pass the test. Wind and solar most decidedly fail.

Thomas Malthus argued that population growth would outrun increases in food production. The Club of Rome’s The Limits to Growth (1972) used computer models whose results were driven by similar assumptions of finite resources and exponential growth. The Hubbert curve, which the author treats as proven and already in effect for conventional oil, takes the recoverable resource as fixed in quantity, and so does not anticipate improving technology.

The alternative tradition is that of Julian Simon, author of The Ultimate Resource. The relevant question is not how much oil lies underground, but what human ingenuity can discover, develop and substitute, given prices and institutions. Simon famously won his wager with Paul Ehrlich over the prices of five metals between 1980 and 1990. This is not to argue that we can never run out of any particular resource; that is a misreading of his thesis. The Simon Abundance Index (SAI) of Gale Pooley and Marian Tupy extends his logic, using hours of work as the common measuring rod of scarcity rather than real or nominal prices, which do not capture rising incomes. The index scales the result for population growth. Measured in hours of work – the relevant measure of scarcity in terms of human welfare – the average time price of 50 basic commodities, crude oil, coal and natural gas among them, fell by 70.9% between 1980 and 2025, even as world population rose by 85%. On the index’s methodology, an hour of work bought 244% more of the basket in 2025 than in 1980.

The index’s data show that crude oil tracked the basket closely: its time price fell by 69.8% over the same period, so an hour’s work buys 3.3 times as much oil as it did in 1980. I have made a similar case about the brave new world of ample oil, and argued in Forbes that the oil majors’ asset write-downs in 2020 were not the end of the oil age.

Julian Simon’s insights are consistent with the approach of Friedrich Hayek, whose essay ‘The Use of Knowledge in Society’ explained why free-market prices, rather than planners, convey the dispersed information about scarcity on which sound decisions depend. Human flourishing has come not from technocratic control over consumption but from the freedom to produce, innovate, invest and exchange in voluntary contracts with other individuals.

In the blurbs on the back cover, Matt Ridley – author of the bestseller The Rational Optimist – calls it “The best book on energy you will ever read. Superbly written, thrillingly argued and utterly terrifying.” It is indeed written in succinct, accessible prose, guiding readers to think about energy economics and the physical requirements of modern civilisation. The “utterly terrifying” aspect of the book comes from treating present constraints as fixed limits rather than as problems that markets, technology and human creativity have repeatedly transformed for the better. In that phrase, the ghost of Malthus rises once more.

This article was first published in the Daily Sceptic https://dailysceptic.org/2026/10/14/are-fossil-fuels-running-out/

Dr Tilak K. Doshi is the Daily Sceptic’s Energy Editor. He is an economist, a member of the CO2 Coalition and a former (cancelled) contributor to Forbes. Follow him on Substack and X.