by Mish Shedlock, Mish Talk:
Bitcoin miners have better things to do than mine Bitcoin.
The Great Repricing
Michael Green AKA Professor Plum on X, has a fascinating post on the future of Bitcoin.
Please consider The Great Re-Pricing: When the “Digital Gold” Bill Came Due
In April 2025, I asked a simple question in my note, “What is Gold?”:
“If the $1.7 trillion now sitting in bitcoin had flowed into bullion instead, what would the yellow metal trade at?”
TRUTH LIVES on at https://sgtreport.tv/
My models suggested that bitcoin acted as a “shadow discount” on Gold, depressing its price by 30–50% by diverting monetary flows into a faster, digital horse. I argued that if those flows ever reversed—if the “Passive Bid” for bitcoin evaporated—Gold would violently re-rate to where it belonged.
Eight months later, we have our answer. In 2025, bitcoin and Gold flows became negatively correlated.
Now, this is not a STRONG relationship, and I’m not remotely going to suggest that the only source of gold flows is coming from bitcoin. But it IS happening. And as I noted last week, it’s an indication of what I believe is the most significant asset rotation of the second half of the 2020s — not a “rotation” from Tech to Value, but a rotation from Energy Consumers to Energy Conservers/Producers.
The bitcoin “Store of Value” trade has fundamentally broken in two. Unlike gold, which is mined with energy, but then remains “gold” regardless of how much mining energy is expended, bitcoin requires continual energy expenditure to maintain the bitcoin network. The mining stops and bitcoin stops; the mining slows, and the bitcoin network’s “safety” and performance degrade.
The “Jaws of Death” (bitcoin’s insolvency)
To understand why Gold is soaring, you first have to understand why bitcoin is bleeding.
In the “Post-Capitalist” era of zero interest rates and surplus energy, we believed we could solve financial problems with code. We ignored the Second Law of Thermodynamics: entropy. Maintaining a digital ledger requires a constant injection of ordered energy.
This chart is the receipt for that entropy [Lead Chart]
I have noted this relationship in many discussions of bitcoin. It’s important to remember what “bitcoin” actually is — the token issued to miners (accountants) for maintaining the network. If the price of bitcoin falls, miners receive less compensation for doing the work.
- The Divergence: Since October 2025, bitcoin’s price has crashed ~27%.
- The Stickiness: The network Hashrate (the cost to maintain the bitcoin network) has only dropped ~5-8%.
The miners are currently doing the same amount of work for 27% less revenue. They have billions in sunk capex, and as long as bitcoin remains above the marginal cash cost of mining (~$85K), they will keep mining. This, in itself, is nothing new. It has always been the case that mining has periods of unprofitability. With an “all in” (including depreciation cost) of roughly $130K, the average public miner is now deeply unprofitable and selling everything they mine to generate cash. All else equal, the hashrate must fall, and miners must move to lower cost regions as equipment depreciates and existing power purchase agreements (PPAs) roll off into new, higher pricing.
The difference this time is that bitcoin miners are no longer using “surplus” energy (debates about renewable sources have notably disappeared), and are now bidding against Microsoft and Amazon for “Machine Food” (electricity). This is not a “dip”; it is a squeeze. The marginal miner is underwater, burning treasury (selling bitcoin) to keep the lights on, hoping for a bailout that isn’t coming. While bitcoin survived previous drawdowns because “true believer” miners had no competing use for their power connections, today AI datacenters pay 3-4x the revenue per kilowatt as bitcoin mining — and the miners are switching.
The resulting decline in hash rate won’t slow the supply of new bitcoin as the network difficulty adjustment will fall. But if the price remains elevated and the hashrate falls, the network becomes increasingly vulnerable to attacks from the remaining miners. Chinese concentration in 2021 has given way to US concentration. And US concentration, in a region where financial crimes are at least nominally punishable (e.g. FTX), is now giving way to a migration to Africa — where miners are perhaps “less incentivized” towards good behavior. If you really believe that the global store of value will be maintained by African hydro, and that once in place, those energy connections do not face the same AI arbitrage by adding high-speed data connections, I’m not really sure how to help you.
And all of this is occurring as the entire bitcoin network faces rising security risks from the emergence of quantum computing and another “halving” in 2028 (which will reduce revenue per hash by 50% unless the bitcoin price increases by 100%). A major capex boom to replace existing mining equipment while miners are barely hanging on and increasingly drawn to AI substitution ahead of a 50% revenue reduction in two years? Unlikely (but admittedly possible).
The “Passive Bid” Has Diminished
At the same time that network risks are rising, the demand side is diminishing. Bitcoin is not temporarily flow-driven. It is necessarily flow-driven. In the absence of endogenous cash generation, price discovery collapses to a reflexive equilibrium governed by marginal buyer demand and balance-sheet constraints.
For two years, the ETF complex provided a mindless, price-agnostic bid for bitcoin. That tap has slowed radically, and now bitcoin must find a new untapped bid. My hunch is that 2026 will start with “value” buyers, rebalancers, and tax-loss harvesting from 2025 returning to bitcoin ETFs and driving prices higher for a time.
