Bitcoin (BTC) traded at $63,999 on July 25, 2026, roughly 18% below the $78,000 all-in production cost that JPMorgan attributes to the public mining sector. The bank’s analysts, led by managing director Nikolaos Panigirtzoglou, say mining economics have “worsened” in 2026, with about 20% of miners now unprofitable and the price sitting below production cost for five consecutive months (The Block).
The number that matters is not the margin, though. It is that publicly traded miners sold more than 32,000 BTC during the first quarter of 2026 to fund operating expenses — more than their combined sales across the whole of 2025. That is the mechanism by which a below-cost price actually transmits into the market, and it has been running for two quarters.
Why the hashrate has not capitulated
The standard model says sustained below-cost pricing forces marginal miners offline, hashrate falls, difficulty adjusts downward, and the survivors’ unit economics repair. Five months in, that has not happened at scale. Network hashrate reached an all-time high in early 2026, crossing 1 zettahash per second, and JPMorgan’s own observation is that hashrate and difficulty are now reacting more sensitively to price swings, with more operators clustered near break-even.
Two things explain the delay. Miners with sunk hardware and contracted power keep running while revenue exceeds marginal electricity cost, even when it does not cover depreciation — and the $78,000 figure is an all-in number including hardware depreciation and overhead, not a cash cost. A miner can be deeply unprofitable on an accounting basis and rationally keep hashing. Second, treasury sales substitute for shutdown. Selling 32,000 BTC is what allows a fleet to keep running through a period the model says it should have exited.
That substitution has a limit, and it is a balance-sheet limit rather than a price one. Treasury depletion is finite in a way that operating losses are not, which is why the capitulation signal to watch is the sales rate rather than the hashrate.
The reflexive problem with treasury funding
Miner treasury sales are supply hitting a market that is already absorbing negative flows elsewhere. Bitcoin exchange-traded funds have not been a reliable offset this year — we covered the rotation in Bitcoin ETFs bleed $3.4bn as Solana and BNB wrappers build AUM, and the more recent rebound in IBIT flipping to a $209m inflow has been episodic rather than structural.
The reflexivity is straightforward and unhelpful: price below cost forces treasury sales, treasury sales add supply, added supply weighs on price, and the gap to production cost widens. Nothing in that loop resolves without either a price recovery driven by demand from outside the mining sector, or enough shutdowns to reset difficulty. Neither has arrived.
What this means for the sector
For miners, the operative question through Q3 is refinancing rather than efficiency. A fleet that has already sold two quarters of treasury has less balance sheet to sell and faces hardware replacement cycles it cannot defer indefinitely. Expect consolidation, power-contract renegotiation, and continued pivots toward artificial-intelligence and high-performance-computing hosting, which is now the sector’s main non-Bitcoin revenue line.
For allocators, published hashrate has become a poor health indicator. It measures deployed capacity, not solvency, and the two have decoupled. The more informative disclosures are quarterly BTC sales as a proportion of production, cash cost per coin as distinct from all-in cost, and the maturity profile of power contracts. Very few miners present all three clearly.
For the wider market, a persistent below-cost regime is a structural bear signal that historically has not resolved quickly. The price path we set out in Bitcoin to $58,000 by Q3 2026: the ETF flow-gap case assumed exactly this kind of supply pressure without a matching institutional bid.
What to watch
Three markers. First, Q2 miner disclosures showing whether the 32,000 BTC quarterly sales rate accelerated — an increase would confirm treasuries are being drawn down faster than production replaces them. Second, a sustained downward difficulty adjustment, which would be the first genuine evidence of capacity leaving rather than being refinanced. Third, whether the $78,000 estimate itself falls: JPMorgan derives it from electricity, hardware depreciation and overhead across public miners, so a wave of cheap secondhand hardware or renegotiated power would lower the bar without any price recovery.
Until one of those moves, the sector is running a slow balance-sheet drain rather than a fast capitulation — which is less dramatic, harder to trade, and considerably more difficult to reverse.
Karthik Subramanian
Karthik Subramanian is a founder, writer, and technology consultant with nine years in the crypto ecosystem. He covers token economics, L1/L2 infrastructure, DeFi protocols, wallets/custody, and the bridge between crypto and forex—broker technology, liquidity, and macro drivers. Karthik’s writing focuses on clear, practical frameworks that help professionals evaluate new products and on-chain innovation alongside FX market realities.
