Bitcoin's mining difficulty fell 10.09% on June 14 at block 953,568 - from 138.96 trillion to 124.93 trillion. It was the second-largest difficulty drop of 2026 and the 11th largest in Bitcoin's entire history. That kind of adjustment doesn't happen because things are going well. It happens because miners are shutting off machines they can no longer afford to run.
The network's total hashrate has declined roughly 12% in June alone and now sits 23% below its October 2025 peak. JPMorgan estimates that approximately 20% of active miners are currently operating at a loss, with all-in production costs running between $78,000 and $84,300 per Bitcoin. The problem: Bitcoin has been trading around $64,700 for five consecutive months, roughly 19% below what it costs the average miner to produce a single coin.
Publicly listed miners responded the way distressed companies do - they sold. Miners dumped more than 32,000 BTC in Q1 2026, exceeding all of 2025's sales combined in a single quarter. Marathon Digital (MARA) liquidated $1.1 billion from its Bitcoin treasury in just three weeks. Riot Platforms sold 2.5 times more Bitcoin than it actually mined. The mining industry's relationship to Bitcoin has flipped: instead of being net accumulators, the biggest miners have become net sellers, and the sell pressure is showing up in the market.
Why miners are underwater
The economics of Bitcoin mining in mid-2026 are brutally simple. It costs more to mine a Bitcoin than you can sell it for, and that has been true for five consecutive months.
JPMorgan's most recent analysis puts the all-in cost of mining one Bitcoin between $78,000 and $84,300. That figure includes electricity (typically the largest single expense), hardware depreciation on ASIC rigs, facility costs, labor, insurance, and corporate overhead. With Bitcoin trading around $64,700, the average miner is losing somewhere between $13,300 and $19,600 on every coin it produces.
Not every miner pays the same production cost. Operations with access to cheap hydroelectric or stranded natural gas can mine below $60,000 per coin. But the industry average tells the story: the post-halving economics that worked when Bitcoin was above $90,000 in late 2024 and early 2025 broke down when the price retreated and stayed down. The April 2024 halving cut the block reward from 6.25 BTC to 3.125 BTC, meaning miners now earn half as many coins per block while their fixed costs stayed roughly the same.
What's particularly dangerous about the current environment is how long it has lasted. Miners can absorb a few weeks of unprofitable production by drawing down cash reserves or borrowing against equipment. Five months of operating below breakeven forces structural decisions - shutting down older machines, closing facilities, or selling Bitcoin reserves to fund continued operations.
The mining difficulty beta to Bitcoin's price - a measure of how sensitive network participation is to price movements - has risen to 0.62, the highest reading in years. That means the network is now significantly more reactive to price declines than it was in previous cycles. When Bitcoin drops, miners leave faster, and difficulty adjusts more sharply.
The great miner fire sale
The numbers from Q1 2026 read like a liquidation event. Publicly listed mining companies collectively sold more than 32,000 BTC in just three months, exceeding the total volume of miner sales for all of 2025. In previous downturns, miners sold gradually to cover operating expenses. This time, the largest operators dumped reserves in bulk.
Marathon Digital (MARA), the largest publicly listed miner by Bitcoin holdings, sold $1.1 billion worth of Bitcoin from its treasury in a three-week window. That represented a sharp reversal for a company that had spent years building one of the largest corporate Bitcoin treasuries in the industry. Marathon didn't sell because it wanted to - it sold because it needed the cash to cover debt service and keep its facilities running while mining revenue couldn't cover costs.
Riot Platforms went even further on a relative basis, selling 2.5 times more Bitcoin than it mined in a single quarter. For a company whose entire business model is producing Bitcoin, selling more than you mine is the definitional sign of distress. Riot's operations were generating coins, but the coins were worth less than the cost to produce them, so the company reached into its reserves to bridge the gap.
The sell pressure from miners has real market impact. When long-term holders are accumulating, fresh miner selling acts as a structural headwind on price. Every coin a miner dumps on the open market needs a buyer, and at current volumes, miners have been one of the most significant sources of consistent sell pressure in the market.
What the difficulty drop actually means
Bitcoin's difficulty adjustment is a self-correcting mechanism built into the protocol. Every 2,016 blocks - roughly two weeks - the network recalculates how hard it should be to mine a block based on how quickly the previous 2,016 blocks were found. If blocks came in faster than the 10-minute target, difficulty goes up. If blocks were slower, difficulty goes down.
The 10.09% drop on June 14 tells you that a significant amount of hashrate left the network in the two-week window preceding the adjustment. Miners with high electricity costs, older-generation ASICs, or thin margins simply turned off their machines because running them was losing money. The difficulty algorithm noticed that blocks were taking longer than 10 minutes to find and dialed back the difficulty accordingly.
For the miners who survive, this is actually good news. Lower difficulty means each unit of hashrate earns approximately 11% more Bitcoin than it did before the adjustment. If you were mining at a razor-thin margin before June 14, the difficulty drop just gave you breathing room. Your electricity costs didn't change, your hardware didn't improve, but the protocol is now giving you more coins per hash.
That's the elegance of the system. When unprofitable miners capitulate and leave, the remaining miners benefit. The difficulty drops, their share of the block reward increases, and the economics improve until equilibrium is restored. It's a built-in survival mechanism that has played out in every Bitcoin downturn since the network launched.
But 10.09% drops don't happen often. This was the 11th-largest downward adjustment in Bitcoin's history, and the second time in 2026 that difficulty fell sharply. The frequency of large negative adjustments is a real-time indicator of industry stress.
The AI pivot isn't working either
When Bitcoin mining margins compressed after the April 2024 halving, the industry's Plan B was supposed to be artificial intelligence. Mining companies had data centers, they had power contracts, they had cooling infrastructure - why not lease that capacity to AI companies that needed massive compute resources?
The theory was sound. The execution has been a different story. A VanEck report published in June 2026 found that miners have delivered only about 25% of the AI and high-performance computing (HPC) capacity they leased to clients. The rest remains unbuilt, delayed, or stuck in permitting and engineering backlogs. Converting a Bitcoin mining facility to handle AI workloads isn't as simple as swapping out ASICs for GPUs - it requires different power density profiles, cooling systems, networking infrastructure, and often entirely different building specifications.
The funding gap is staggering. VanEck estimates a $50 billion near-term shortfall in the capital required for miners to build out their contracted AI capacity, with long-term needs reaching $221 billion. That's capital the mining industry doesn't have, especially when its core business is generating losses. The companies that need AI revenue most urgently are the same companies least able to fund the infrastructure to deliver it.
This matters beyond the mining industry because the Bitcoin-to-AI pivot narrative drove a significant portion of mining stock valuations in 2025. Investors priced in AI optionality as a hedge against Bitcoin price risk. If that optionality turns out to be worth less than expected - because the conversion costs are too high, timelines are too long, or the competitive landscape for AI data centers is dominated by better-capitalized players like Microsoft and Amazon - then mining stocks need to be repriced to reflect a pure-play Bitcoin business operating below breakeven.
Who survives and who doesn't
Not every mining company is responding to this crisis the same way, and the divergent strategies tell you a lot about who management thinks is going to win.
Marathon Digital (MARA) has the largest Bitcoin treasury among publicly listed miners but chose to draw it down aggressively. The $1.1 billion in sales over three weeks wasn't a strategic repositioning - it was a cash crunch. Marathon is betting that surviving the downturn is worth more than holding Bitcoin at a loss, and it has the treasury depth to execute that bet. But every coin sold is a coin Marathon doesn't benefit from if prices recover.
Riot Platforms is in a more precarious position. Selling 2.5 times more Bitcoin than you mine in a quarter means you're consuming reserves faster than you're replenishing them. If Bitcoin's price doesn't recover meaningfully in the next two to three quarters, Riot faces hard choices about facility closures and workforce reductions.
CleanSpark is the outlier, and it's making the opposite bet. While its competitors are selling and cutting, CleanSpark added 585 megawatts of new mining capacity during the downturn. That's a massive expansion at a time when most of the industry is contracting. CleanSpark's thesis is straightforward: Bitcoin will eventually trade above production cost again, and when it does, the miners who expanded during the downturn will capture an outsized share of the network's hashrate and block rewards. If they're right, they win big. If they're wrong, they've invested hundreds of millions into capacity that loses money.
The historical pattern favors the contrarians. In every previous mining downturn - 2018, 2022, and the post-halving squeeze of late 2024 - the companies that expanded during stress and survived to the recovery captured significant market share from the companies that didn't make it.
Frequently asked questions
What happened to Bitcoin mining difficulty in June 2026?
Bitcoin mining difficulty dropped 10.09% on June 14, 2026, at block height 953,568. The adjustment brought difficulty from 138.96 trillion down to 124.93 trillion, making it the second-largest difficulty decline of 2026 and the 11th-largest negative adjustment in Bitcoin's entire history. The drop was triggered by miners shutting off machines that were no longer profitable to operate.
Why are Bitcoin miners unprofitable in 2026?
The core issue is that Bitcoin's price has been trading around $64,700 for five consecutive months while the all-in cost to mine one Bitcoin runs between $78,000 and $84,300, according to JPMorgan. That gap means the average miner loses $13,300 to $19,600 on every coin it produces. The April 2024 halving cut the block reward in half, doubling the effective cost per coin, and Bitcoin's price hasn't recovered enough to compensate.
How much Bitcoin did miners sell in 2026?
Publicly listed Bitcoin miners sold more than 32,000 BTC in Q1 2026 alone, exceeding the total volume of miner sales for all of 2025 combined. Marathon Digital (MARA) sold $1.1 billion from its Bitcoin treasury in a three-week window, and Riot Platforms sold 2.5 times more Bitcoin than it mined during the quarter.
What does a difficulty drop mean for Bitcoin?
When Bitcoin's mining difficulty drops, it means a significant number of miners have stopped operating and left the network. The protocol automatically adjusts difficulty every 2,016 blocks (roughly two weeks) to maintain a target block time of 10 minutes. Lower difficulty benefits remaining miners by giving them approximately 11% more BTC per unit of hashrate, making their operations more profitable without any change in costs.
What is the Bitcoin mining difficulty beta?
The mining difficulty beta measures how sensitive network hashrate is to changes in Bitcoin's price. As of June 2026, the beta has risen to 0.62, meaning the network is significantly more reactive to price declines than in previous cycles. A higher beta indicates that more miners are operating near their breakeven point and will shut off machines quickly when prices fall.
How much does it cost to mine one Bitcoin in 2026?
JPMorgan estimates the all-in cost of producing one Bitcoin at approximately $78,000 to $84,300 as of June 2026. This includes electricity, ASIC hardware depreciation, facility costs, labor, insurance, and corporate overhead. Individual miners' costs vary significantly - operations with access to cheap hydroelectric power can mine below $60,000, while those paying retail electricity rates face costs well above the average.
Is CleanSpark expanding during the mining downturn?
Yes. While most miners are selling Bitcoin reserves and cutting costs, CleanSpark added 585 megawatts of new mining capacity during the downturn. CleanSpark is betting that Bitcoin will trade above production cost again and that miners who expand during stress will capture outsized market share when the recovery arrives. It's a contrarian strategy that has historically rewarded survivors of previous mining downturns.
Are Bitcoin miners pivoting to AI successfully?
Not yet. According to a VanEck report from June 2026, miners have delivered only about 25% of the AI and high-performance computing capacity they leased to clients. The remaining capacity is stuck in permitting, engineering, or construction delays. The industry faces a $50 billion near-term funding gap and $221 billion in long-term capital needs to build out contracted AI infrastructure, capital that distressed mining companies simply don't have.
How much hashrate has Bitcoin lost in 2026?
Bitcoin's total network hashrate dropped approximately 12% in June 2026 alone and sits roughly 23% below the all-time peak reached in October 2025. The decline reflects miners shutting down unprofitable operations as Bitcoin trades well below production cost, particularly operators running older-generation ASIC hardware or paying above-average electricity rates.
Will the Bitcoin difficulty drop help miners recover?
The difficulty cut gives surviving miners immediate relief - approximately 11% more Bitcoin per unit of hashrate deployed. That narrows the gap between revenue and production costs, potentially bringing some borderline-profitable operations back above breakeven. However, a single difficulty adjustment isn't enough to fix the structural problem. If Bitcoin's price remains near $64,700 while production costs stay above $78,000, even the reduced difficulty won't make mining profitable for most operators. A sustained price recovery is what the industry ultimately needs.
Sean Ristau | @SeanRistau | 21Rates / The Daily Stack