Jeff Currie of the Carlyle Group, a name that carries weight in commodity circles, recently stated that the global oil market is entering a structural shortage. For most, this is a macro talking point. For Bitcoin miners, it is a direct audit of their cost assumptions.
Context Currie is not a fringe voice. He spent decades at Goldman Sachs as their global head of commodities research, and his calls on oil and metals have shaped institutional positioning. When he says structural shortage, he means the gap between supply and demand is not cyclical—it is a function of chronic underinvestment in new production, driven by ESG pressures and the energy transition rhetoric. The IEA’s latest World Energy Outlook already flagged that upstream oil investment would need to rise from $499 billion in 2023 to over $600 billion annually by 2030 to meet demand. We are nowhere near that.
For cryptocurrency miners, the link is indirect but lethal. Oil prices do not directly set electricity rates, but they set the marginal cost of generation in many regions. In the United States, natural gas prices are tightly correlated with oil, and gas-fired peaker plants often set the clearing price for wholesale electricity. In Texas, where a significant portion of Bitcoin’s hash rate operates on the ERCOT grid, winter storms and gas supply constraints have already sent spot prices into the hundreds of dollars per megawatt-hour. A sustained oil price above $100 per barrel would push that baseline up.
Core: Dissecting the Cost Structure Let me be specific. According to data from the Cambridge Centre for Alternative Finance, the average global electricity cost for Bitcoin mining in Q4 2024 was approximately $0.046 per kWh. But that average masks a bimodal distribution. Miners with fixed-price power purchase agreements (PPAs) secured in 2020–2021 pay as low as $0.02–$0.03 per kWh. Those relying on spot market pricing—especially in regions like Kazakhstan, Iran, and parts of the US—face rates that have already risen 30–40% since 2022.
If oil prices remain elevated, the transmission mechanism works through three channels: 1. Direct generation costs – Diesel and natural gas generators become more expensive to run, reducing the profitability of off-grid mining operations. About 15% of the global hash rate is estimated to be powered by gas flaring or dedicated gas generators, especially in the Permian Basin. Those operations are currently profitable because they burn otherwise wasted gas. But if the price of that gas increases (due to higher oil-linked gas prices or stricter flaring regulations), their margin evaporates. 2. Grid electricity price pass-through – In deregulated markets like Texas and New York, ancillary service costs and bid prices from gas plants will rise. ERCOT’s real-time energy prices for June 2025 are already up 22% year-over-year. Miners who hedge via fixed-price swaps may insulate themselves, but the broader trend favors higher average power costs. 3. Opportunity cost of capital – Oil-driven inflation could keep the Federal Reserve’s interest rates higher for longer, increasing the cost of debt for mining expansion. Marathon Digital’s recent bond issuance carried a 7.8% coupon. That is not sustainable if hashprice remains below $50/PH/s.
To quantify: at a hash price of $55 per petahash per day (current as of March 2025), a miner with 10 EH/s and an all-in cost of $0.04 per kWh breaks even around $48 per PH/s. If electricity costs rise to $0.06 per kWh, the break-even hashprice jumps to $62 per PH/s. That is a 25% increase in the minimum viable price. Given that Bitcoin price is largely uncorrelated with oil, the risk of margin compression is real and asymmetric.
Contrarian: What the bulls are not wrong about I have to address the counterarguments, because dismissing them would be intellectually dishonest. First, the percentage of global mining that relies on oil-linked electricity is small. The largest mining hubs—Sichuan (hydro), Norway (hydro), Upstate New York (nuclear/hydro), and Iceland (geothermal)—are not directly exposed to oil prices. Even Texas, despite its gas-heavy grid, has seen a surge in solar and battery storage that could moderate peak pricing. Second, miners are becoming more efficient. The newest generation of ASICs (e.g., Antminer S21 Pro) consume 16 J/TH, compared to 30 J/TH just three years ago. Efficiency gains may offset cost increases.
Third, Currie’s thesis may be wrong. The shale revolution in the US has repeatedly defied predictions of peak oil. OPEC+ has spare capacity of roughly 4 million barrels per day. If prices rise too high, Saudi Arabia can flood the market. Supply constraints are a narrative that can be punctured by a single policy change.
Takeaway The question is not whether oil will reach $120. The question is whether miners have stress-tested their models against a sustained rise in energy costs. Most have not. Their investor decks show hash price curves that assume constant electricity prices. That is a vulnerability in plain sight. Trust is a vulnerability vector, and the code here—the energy market fundamentals—is speaking louder than any whitepaper. Miners who ignore it will find their margins audited by a force they cannot fork.