The Buyer of Last Resort: How Computational Demand Could End Britain's Energy Waste
Britain discarded 8.3 TWh of wind in 2024 and flares North Sea gas it cannot sell. Why Bitcoin mining, specifically, is the only industrial load that can monetise stranded energy.
Part 1 of 2: The Problem
In 2024, Britain paid wind farms £393 million to switch off and discarded 8.3 terawatt-hours of their output in the process, 91% more wasted wind than the year before. In the same waters, North Sea operators burned 658 million cubic metres of gas straight into the atmosphere because no buyer existed. One waste is paid for. The other is simply destroyed. Both are energy with nowhere to go, and the surrounding costs run into billions a year while barely registering in public debate about energy policy.
This article examines a proposition that will make environmentalists uncomfortable: Bitcoin mining, specifically its proof-of-work protocol, is the only existing technology capable of monetising energy that would otherwise be wasted. Not cryptocurrency generally. Bitcoin’s particular technical characteristics make it unlike any other industrial electricity consumer on earth. The broader implications for Britain’s energy independence follow in Part 2. First, the waste itself.
Paying Wind Farms to Do Nothing
The direct payments to wind farms are the visible tip of a much larger bill. National Grid’s Electricity System Operator (now NESO) reports that total balancing costs reached £2.7 billion in 2024/25, of which £1.7 billion was the cost of managing transmission constraints, up 64% year on year. The mechanism is simple to describe and expensive to operate.
Wind farms in Scotland generate electricity. Demand sits in South East England. Transmission lines between the two cannot carry the full load during high wind. So the grid operator pays Scottish wind farms to reduce output while simultaneously paying gas plants in the south to ramp up. Consumers fund both. That combined turn-down-and-turn-up cost is what the £1.7 billion constraint bill measures: it is not a single cheque to wind farms, but the price of a grid that cannot move its own electricity from where it is made to where it is needed.
Written evidence to Parliament from SSE puts the cause plainly: “the cost of grid constraints is a growing concern which is feeding negative and misleading coverage about wind farms, even though the root cause is a lack of strategic planning which led to underinvestment in grid infrastructure over the last decade.”
That framing is accurate. The root cause is infrastructure. The result is clean electricity that was viable to build, viable to generate, and impossible to deliver.
How Much Gets Wasted
The volume figures are stark. The Renewable Energy Foundation’s analysis of grid data found 8.3 TWh of wind generation discarded in 2024, with 98% of it Scottish and a single wind farm (Seagreen) accounting for 40% of all constrained volume. NESO’s own figures put wind curtailment at 13% of the wind that would otherwise have reached the grid.
Wind generation discarded through curtailment, 2010–2024
- 2019 £139m paid
- 2023 £310m paid
- 2024 £393m paid
Modelling by FTI Consulting for Ofgem shows the trajectory: under Britain’s current national pricing, around 18 TWh of wind is curtailed in 2025, rising to almost 70 TWh by 2035 as more capacity is added without the wires to carry it. NESO’s system impact modelling adds that “constraint costs in a high wind year could be significantly higher than those in a low wind year.” The waste is weather-driven as well as infrastructure-driven, and it grows.
Successful renewable deployment, in other words, increases waste. More wind capacity in Scotland means more frequent periods when generation exceeds the wires’ ability to carry it south. More constraint payments. More curtailed clean electricity. The reward for building more turbines is a larger bill for not using them. NESO forecasts the annual balancing cost could peak near £8 billion by 2030 if the grid build does not catch up.
How Europe Compares
Britain is not unique here, but the scale relative to system size is striking. A 2025 report by E3G and EMBER for the Beyond Fossil Fuels coalition found renewables curtailment cost €7.2 billion across just seven European countries in 2024, and observed that “high levels of curtailment of renewable energy demonstrates the need for more investment in clean flexibility and improved grid balancing.” It is a coalition advocacy report, so its framing serves a clean-flexibility argument. But the underlying figures are drawn from national grid operators, and the UK line (8,300 GWh of wind, 10% curtailment) matches the British data exactly.
The difference is timing. Most European nations with high renewable penetration invested in transmission and flexibility earlier in their transitions. Britain chose to build generators first and wires second. The mismatch is growing.
The Gas Nobody Wants
While wind curtailment occasionally makes headlines, North Sea gas flaring receives almost none, despite representing direct greenhouse gas emissions from destroyed energy.
The North Sea Transition Authority’s 2024 emissions report records a 49% reduction in flaring between 2018 and 2023, to 691 million cubic metres, then the lowest level on record. The 2025 report shows flaring volume falling a further 4.8% in 2024, to 658 mcm, now 51% below the 2018 baseline. That is genuine progress. It is also not enough: flaring still accounts for roughly 17% of upstream oil and gas greenhouse-gas emissions.
When oil is extracted, natural gas comes up with it. In fields where gas infrastructure does not exist or is not economical to build, operators burn the gas rather than venting it raw. Flaring converts methane to CO₂, which is far preferable to releasing it. Fossil methane’s global warming potential over 100 years is around 30 — precisely 29.8 ± 11, per IPCC AR6 Working Group 1, Chapter 7, Table 7.15, and roughly 80 over a 20-year horizon. Burning it is much better than releasing it. But it is still wasted energy producing direct emissions.
Why Operators Light the Match
The economics are brutal. Capturing and selling associated gas requires separation infrastructure on the platform, compression to meet pipeline specifications, a physical pipeline connection to existing networks, and gas prices that justify the capital outlay. For older installations, smaller fields, or remote locations, those costs exceed the value of the gas. Flaring becomes the rational choice. Energy is destroyed because saving it costs more than losing it.
This is the textbook definition of stranded energy: a resource that exists, that has value in principle, and that no buyer will pay to collect. Hold that thought — because the same description applies to the curtailed wind off Scotland, and to one industrial process that can consume both.
Stranded Energy: The Common Thread
Curtailed wind and flared gas look different operationally. Economically, they share a defining feature: energy exists at a location where no buyer will pay for it.
The conventional solutions all face constraints that limit their usefulness.
Transmission infrastructure could carry Scottish wind power south, but new power lines cost billions and take years. Planning permission alone can exceed a decade. By the time new transmission is operational, generation capacity may have shifted again.
Battery storage can time-shift electricity by hours, occasionally days. Seasonal storage (banking summer wind for winter demand) requires terawatt-hours of capacity. Current battery technology is orders of magnitude short of that at any remotely economic price.
Industrial demand response from aluminium smelters, chemical plants, or conventional data centres offers limited flexibility. These operations cannot stop and restart without damaging equipment or losing production value. Most run continuously at steady loads because that is what their economics require.
Interconnector exports provide an outlet when European prices are favourable, but capacity is limited and Continental wholesale prices tend to correlate with British ones. When British wind is plentiful and cheap, so is European wind.
What is needed is a buyer that can locate anywhere, tolerate frequent interruptions without loss, operate on no fixed schedule, settle economically in real time, and require no transmission infrastructure beyond the generation site.
No conventional industrial process meets all five criteria.
What Makes Bitcoin Mining Different
Bitcoin mining is the only existing industrial load with this specific combination of properties. That claim requires substantiation, not assertion.
Interruption tolerance. A Bitcoin mining operation switched off mid-computation loses nothing except the opportunity cost of that period. Each hash attempt is statistically independent of every prior attempt: the protocol models mining as a sequence of independent trials, as set out in the Bitcoin whitepaper (Nakamoto, 2008). Stop for ten seconds or ten hours; resume where electricity is available. No data corruption, no restart cost, no work-in-progress destroyed. Contrast this with a data centre (interrupting server operations corrupts data and breaks service agreements), an AI training cluster (stopping mid-run wastes all computation since the last checkpoint), or a manufacturing line (halting damages product and requires costly restarts).
Location flexibility. A mining operation needs electricity, an internet connection, and cooling. The bandwidth requirement is trivial: the protocol settles one block roughly every ten minutes, a few megabytes of data, well within satellite range. No proximity to customers, no supply-chain logistics, no skilled workforce permanently on-site, no planning permission for complex industrial processes. Place miners adjacent to a Scottish wind farm, on an offshore platform, at a landfill methane-capture site, beside a wastewater plant generating biogas. The economics work wherever stranded energy exists. Aluminium smelters need ports. Chemical plants need supply chains. Data centres need fibre and urban proximity. Bitcoin miners need a plug and a sky.
Real-time economic settlement. Mining produces an instantly settleable, globally tradeable digital asset. Electricity goes in; Bitcoin comes out; Bitcoin sells on exchanges with daily spot volume well into the tens of billions of dollars; fiat pays the electricity bill. The loop closes in minutes. Compare hydrogen production (requires storage, transport, buyer negotiations), synthetic fuels (requires refining, distribution, customers), or direct air capture (requires long-term carbon-credit contracts). No intermediaries. No minimum volumes. No contract negotiations. Revenue from the first hour of operation.
Each property alone is useful. Together, they describe something that does not exist elsewhere in industrial electricity consumption.
Why Aggregate Numbers Miss the Point
Discussion of Bitcoin’s energy consumption almost always reaches for the country comparison: “Bitcoin uses as much energy as Argentina.” These comparisons are accurate at the aggregate level. They are also analytically useless for evaluating what happens when a specific mining operation connects to a specific energy source.
The distinction that matters is marginal impact.
A Bitcoin miner running on coal-fired electricity in a supply-constrained grid increases emissions and competes with residential demand. Negative marginal impact. A Bitcoin miner consuming Scottish wind that would otherwise be curtailed reduces waste, lowers constraint payments, and improves the economics of the wind farm that generated it. Positive marginal impact. A Bitcoin miner burning North Sea associated gas that would otherwise be flared does something stronger still, and here the thermodynamics are precise.
A 2023 MIT CEEPR analysis by Stoll and colleagues measured the combustion efficiency of gas burned in a mining genset against gas burned in a routine flare. A flare destroys about 91.1% of the methane that passes through it; the rest escapes unburned, and unburned methane is the high-GWP problem. A mining genset, running the same gas through a controlled engine, combusts 99.9% of it. Running stranded gas through a miner instead of a flare cuts the total CO₂-equivalent emissions of that gas by around 25%. The same paper estimates that mining could address roughly 4% of US oil-and-gas methane through routine flaring alone, rising toward 63% if flare-outage events are included.
This is the part of the argument that should give honest environmentalists pause. Mining flared gas is not merely “less bad than wasting it.” On the methane that would otherwise be flared, it is measurably better for the climate than the flare — while also turning destroyed energy into revenue.
The aggregate figure tells you Bitcoin uses a lot of energy. The marginal analysis tells you whether a particular mining operation makes energy systems better or worse. Both can be true at once. The academic literature now supports the marginal framing directly: alongside the MIT work, a 2023 NBER paper by Papp, Almond and Zhang measured the real emissions response of mining to local grid conditions and found cases where the marginal impact runs net-negative. Identical computational loads can help or harm depending entirely on energy source and grid context. The policy question is whether regulation should steer mining toward locations where it absorbs waste and provides grid services, rather than pretending a decentralised protocol can be banned out of existence.
Britain’s Industrial Electricity Price Problem
Britain manages to combine substantial renewable generation, significant energy waste, and the highest industrial electricity prices in the developed world. This deserves more attention than it receives.
The Office for National Statistics reports that in 2023 the UK had the highest industrial electricity prices of the 24 countries reporting to the International Energy Agency: about four times the level paid in the United States and Canada, almost 50% above France and Germany, and 46% above the IEA median. UK non-domestic electricity averaged 25.97 pence per kWh in late 2024, having peaked above 28 p/kWh in 2023. The pattern is electricity-specific: UK industrial gas, by contrast, sat slightly below the IEA median, which makes the electricity premium harder to dismiss as a general energy-cost story.
A country that wastes clean electricity while charging the developed world’s highest prices for the electricity it does deliver is not a country with an energy strategy. It is a country with an energy market that has developed pathologies nobody is incentivised to cure.
The Price-Waste Paradox
The explanation lies in market design. British wholesale electricity prices are set by marginal cost: the most expensive generator needed to meet demand sets the price for all electricity in that settlement period. When gas plants sit on the margin (frequently), gas prices determine system-wide electricity costs, including for wind that produces at near-zero marginal cost.
Industrial users then pay wholesale prices plus transmission charges, distribution charges, and system-operation costs. These non-commodity charges are substantial and rising, driven partly by the growing expense of managing a grid with high renewable penetration and inadequate flexibility.
Meanwhile, constraint payments transfer money from consumers to generators without delivering a single watt of useful electricity. The costs are socialised across all bills, raising the effective price while reducing useful energy delivered.
British industry pays the developed world’s highest electricity prices. British wind farms get paid to switch off. Both statements are true simultaneously, and the market sees no contradiction.
Why Britain’s Miner Went Abroad
The consequence is predictable. Argo Blockchain, founded and headquartered in London, once the flagship British name in the industry, never ran its machines on British electrons. It built them in West Texas and Quebec, where power was a fraction of the UK price. In December 2022 it sold its flagship Texas facility to Galaxy Digital to stave off insolvency, and in December 2025 it delisted from the London Stock Exchange entirely, retaining only its US listing as part of a restructuring that diluted existing shareholders to a fraction of the company. Britain’s one homegrown listed miner ran abroad, struggled, and left.
The counter-example is on the other side of the Atlantic. Riot Platforms mines in Texas on power that costs a fraction of UK industrial rates, and in August 2023 alone earned $31.7 million by curtailing during demand peaks and selling its flexibility back to the grid. That single month exceeded all the grid credits it had earned in the whole of 2022. The difference is not entrepreneurial talent. It is the price of electricity and a market that treats flexible demand as an asset.
Britain builds the wind farms. Texas gets the flexible demand and the grid-balancing benefits. The irony writes itself.
The Texas Model
Texas faces grid challenges similar to Britain’s: high renewable penetration, transmission constraints, price volatility. Its electricity market handles them differently.
The ERCOT market’s independent monitor reports hundreds of hours each year when prices sit at or below $0 per MWh, with 32% of real-time generation capacity offered at or below zero (primarily wind and solar), and the windiest West zone seeing the most sub-zero pricing of all. Large flexible loads, including Bitcoin miners, participate in several grid mechanisms: consuming surplus electricity (and being paid to do so during negative-pricing events), providing guaranteed curtailment during peak demand in exchange for capacity payments, and offering rapid demand adjustments for frequency regulation.
The Texas grid treats flexible demand as a resource, not an inconvenience. Bitcoin miners profit from surplus electricity that would otherwise be wasted. The grid gets a shock absorber. Consumers benefit from reduced volatility. Britain has none of these mechanisms functioning at scale for mining, despite facing transmission-constraint costs that, in absolute terms, dwarf the ones Texas manages.
Who Benefits From the Status Quo
Every source cited in this article operates within incentive structures that shape how it presents data and its appetite for change.
Grid operators (NESO) are budget-funded and judged on reliability, not cost efficiency; constraint payments are an operational tool, not a crisis demanding intervention. Wind farm operators receive constraint payments at rates comparable to generation revenue: they are economically neutral whether spinning or curtailed, so feel no urgency. Traditional gas and nuclear generators benefit from high wholesale prices when they sit on the margin, and transmission constraints preserve the regional price differentials that favour generation near demand. The Bitcoin mining industry has an obvious interest in cheap electricity, but most operators have already relocated to where markets allow it. Environmental NGOs generally oppose Bitcoin on aggregate-consumption grounds and rarely engage with marginal-impact arguments. And DESNZ and the Treasury face pressure to lower energy costs but lack tools that do not involve slow, expensive infrastructure or politically contentious market restructuring.
None of these actors has a strong incentive to pursue computational demand as a solution to energy waste. The opportunity sits in the gap between institutional interests — which is precisely where opportunities tend to sit in British energy policy.
Why Energy Independence Is at Stake
(This section previews Part 2.)
Britain’s energy import dependency, covered in previous hostile.eco analysis, creates strategic vulnerability. Any mechanism that extracts more economic value from domestic energy resources (wind or flared gas) reduces that vulnerability at the margin.
Bitcoin mining offers a use for energy that Britain currently pays to waste (constraint payments), burns to atmosphere (flared gas), or fails to monetise at all (generation that cannot reach demand). Part 2 examines whether that theoretical potential can become operational reality, and how the same root cause, the developed world’s most expensive industrial power, is now driving away the AI data centres that could finance Britain’s nuclear future.
Counterarguments and Complexity
Five objections deserve honest engagement.
“Bitcoin is environmentally catastrophic regardless of energy source.” The Cambridge Bitcoin Electricity Consumption Index put Bitcoin’s 2024 consumption at around 138 TWh, comparable to a mid-sized country. The aggregate figure is large. It is also irrelevant to the policy question, which is not “should Bitcoin exist?” but “given that it does exist and cannot be banned by any single jurisdiction, should British policy push mining toward locations where it absorbs waste?” Prohibition drives mining to less-regulated grids. Engagement directs it to useful locations.
“Better solutions exist for curtailment.” Proponents of batteries, hydrogen, and industrial demand response argue these deserve priority. None currently delivers the complete package: instant interruption tolerance, arbitrary location, real-time settlement. Batteries require massive capital for seasonal storage that does not yet exist economically. Hydrogen needs downstream infrastructure and customers that have not materialised. Other industrial demand response cannot halt without production losses. Bitcoin mining may not be the optimal long-term solution. It may be the only one available now, while the optimal ones mature.
“Flared gas should be captured for conventional use.” Correct, where economically viable. Where gas-capture infrastructure exists or can be built at reasonable cost, conventional sales beat mining. The Bitcoin argument applies to truly stranded gas: volumes too small for pipeline infrastructure, fields too remote, end-of-life installations where no operator will invest capital they cannot recover. These are a subset of total flaring, but a persistent one.
“Mining creates e-waste and hardware obsolescence.” Bitcoin mining equipment becomes obsolete as network difficulty rises and more efficient hardware ships. Older ASICs are typically scrapped rather than recycled, and the 18-to-24-month obsolescence cycle exacerbates the problem. The industry has no clear answer to equipment-lifecycle management. This is a legitimate environmental cost that honest advocates of mining-as-grid-service must acknowledge.
“Bitcoin has no intrinsic value.” This philosophical objection questions whether any energy consumption for digital tokens is justified. The empirical reality is that Bitcoin has a global market valuation, settles billions in daily transactions, and provides financial access in jurisdictions with unstable currencies. Whether that constitutes sufficient social utility is a values question. For the purposes of this article, Bitcoin’s market value is simply what makes it capable of monetising stranded energy. If nobody valued it, miners would not buy the electricity.
What Follows in Part 2
The second article examines how Bitcoin mining could integrate with UK renewable and gas-flaring projects, the regulatory changes required, and the strategic case for energy-backed digital assets, then turns to the larger prize. The same root cause that drives miners abroad, the developed world’s most expensive industrial electricity, is now driving away the hyperscale AI data centres whose long-term power contracts could finance a new generation of British nuclear. Stranded energy and patient capital are two faces of one argument.
For now, five points stand:
Britain discarded 8.3 TWh of wind in 2024 and paid £393 million directly for the privilege, inside a £2.7 billion balancing bill. North Sea operators flared 658 million cubic metres of gas because capturing it was uneconomical. Bitcoin mining has technical characteristics found in no other industrial load that make it suitable for monetising stranded energy and, on flared gas, combusts methane more completely than the flare it replaces. Britain’s one homegrown listed miner ran its machines abroad and has now left, because UK industrial electricity is the most expensive in the developed world. And the difference between poorly-sited mining and well-sited mining is the difference between environmental harm and environmental benefit.
The question is not whether Bitcoin uses a lot of energy. The question is whether Britain can afford to keep wasting its own.
Data Sources & References
UK Government & Regulators
- NESO 2025 Annual Balancing Costs Report: £2.7bn total balancing, £1.7bn transmission constraints (2024/25), ~£8bn 2030 forecast
- NESO System Impact Assessment Report (November 2024): constraint-cost variability by weather year
- Ofgem / FTI Consulting: Assessment of locational pricing options (October 2023): wind curtailment under national/zonal/nodal designs (18 TWh in 2025 → ~70 TWh by 2035)
- SSE written evidence to the Energy Security and Net Zero Committee: grid constraints, transmission under-investment as root cause
- NSTA Emissions Monitoring Report 2024: 49% flaring reduction 2018–2023; 691 mcm flared 2023
- NSTA Emissions Monitoring Report 2025: 658 mcm flared 2024 (−4.8%, 51% below 2018)
- ONS: The impact of higher energy costs on UK businesses, 2021–2024: UK highest industrial electricity of 24 IEA countries; ~4× US, +46% above IEA median
- DESNZ International industrial energy prices: underlying price comparison dataset
Peer-Reviewed & Academic Research
- Stoll, Klaassen, Gallersdörfer & Neumüller (2023), “Climate Impacts of Bitcoin Mining in the U.S.,” MIT CEEPR Working Paper 2023-11: mining genset 99.9% vs flare 91.1% combustion efficiency; ~25% emissions reduction; 4–63% of US oil-and-gas methane addressable (working paper, not peer-reviewed)
- Papp, Almond & Zhang (2023), “Bitcoin and Carbon Dioxide Emissions,” NBER Working Paper 31745: marginal-emissions response of mining to local grid conditions
- IPCC AR6 WG1 Chapter 7, Table 7.15: fossil methane GWP100 = 29.8 ± 11
- Bitcoin whitepaper (Nakamoto, 2008): proof-of-work protocol, ~10-minute block target
Industry & Market Data
- Renewable Energy Foundation: Discarded wind energy increases by 91% in 2024: 8.3 TWh discarded, £393m direct cost, Seagreen 40% of constraints (BMRS-derived)
- E3G & EMBER for Beyond Fossil Fuels (2025): How Europe’s grid operators are preparing for the energy transition: €7.2bn curtailment across seven countries (coalition report; figures from national TSOs)
- Cambridge Bitcoin Electricity Consumption Index: ~138 TWh network consumption (2024)
- CoinGecko: exchange spot volume: daily spot liquidity
- ERCOT Independent Market Monitor (Potomac Economics), 2024 State of the Market Report: hundreds of hours at/below $0/MWh; 32% of capacity offered ≤$0
- EIA: Texas state energy profile: ~42.3 GW installed wind capacity (end-2024)
- Argo Blockchain: Helios facility sale to Galaxy Digital (2022) and LSE delisting / restructuring update (2025)
- Riot Platforms / SEC filing: power-curtailment credits (August 2023): $31.7m in a single month from ERCOT demand response, a monthly record exceeding all such credits in 2022
This is Part 1 of a two-part series. Part 2 (“The Patient Capital”) examines how AI hyperscale demand could finance Britain’s nuclear future, and why the same root cause strands both Bitcoin miners and data centres offshore.