Bitcoin mining margins have stabilized roughly one year into the post-halving cycle, settling at hash-price levels that are 20 to 25 percent below the pre-halving baseline but materially above the worst-case forecasts that circulated in mid-2025. The mining industry that survived the squeeze looks structurally different from the one that entered it.
The Squeeze, Briefly
The April 2024 halving cut the per-block subsidy from 6.25 BTC to 3.125 BTC, mechanically halving the daily revenue available to miners holding constant network share. The conventional pre-halving narrative held that the price recovery in the months following would offset the subsidy compression and preserve aggregate miner economics. That narrative was partially correct: BTC did appreciate through 2024 and into 2025.
But network hash rate also grew through the same period, as deferred capacity additions came online and as institutional miners with cheaper capital pushed deployment forward. The combination meant that the per-terahash revenue available to miners — hash price — fell faster than per-BTC revenue rose. By late 2024, hash price had compressed to levels that were uneconomic for the higher-cost segment of the industry.
The shakeout that followed lasted roughly nine months and removed meaningful capacity from the network. The capacity that came offline was not random.
Who Got Squeezed Out
The miners that closed or scaled down through the squeeze had several common characteristics. They tended to operate at electricity costs above 6 cents per kWh on a fully-loaded basis. They tended to operate older-generation ASIC fleets — primarily Antminer S19 series and equivalents — that had become structurally uneconomic at post-halving hash prices. And they tended to be financed with debt structures that required ongoing refinancing, which became progressively harder to obtain as their break-even economics deteriorated.
A meaningful portion of the displaced capacity was concentrated in specific U.S. regions where the combination of higher electricity costs and older fleet vintages created a particularly difficult margin profile. Texas operations on shale-power arrangements were the most visible cohort — the hedged-electricity arrangements that had looked attractive in 2022 and 2023 became liabilities when natural gas pricing stabilized at higher levels through 2025.
Who Consolidated
The capacity that the displaced miners exited was largely absorbed by a handful of public mining operators. Marathon Digital, Riot Platforms, CleanSpark, and Iris Energy each meaningfully expanded their share of total network hash rate over the trailing 18 months. The expansion was funded primarily through equity issuance and convertible debt, and it has produced an industry structure in which the top five public miners now control a materially larger share of network hash than they did pre-halving.
The economics of the consolidation have been favorable for the survivors. Acquired capacity came at distressed prices, the surviving miners benefited from network difficulty resetting lower as displaced capacity exited, and the combination of cheaper acquisition cost and improved hash price economics produced solid first-quarter 2026 cash-flow numbers across the public-miner cohort.
The smaller private miners that survived the squeeze did so by operating at electricity costs in the 4-cent range or below, by running newer-generation ASICs, or by having patient balance-sheet financing that did not require continual refinancing. The survivor pool is smaller than it was, but the survivors have meaningfully better unit economics than the average miner did pre-halving.
The AI Compute Pivot
A factor that complicates clean post-halving comparisons: a non-trivial portion of mining capacity that came offline did not actually leave the data-center business. Several operators repurposed shell sites to AI compute hosting, leveraging the existing power infrastructure and cooling for GPU deployments rather than ASICs.
The conversion economics are specific to the site. A purpose-built ASIC mining facility is not a drop-in fit for high-density GPU workloads, and the retrofitting cost is meaningful. But for facilities with appropriate power capacity and cooling design, the pivot has been a credible alternative to either continued mining at compressed margins or full closure.
The result is that some "mining capacity exits" through the post-halving period are better understood as data-center repurposing, with the underlying real estate and power contracts remaining productive but in a different end use. This is not visible in network hash rate data; it is visible in the operating disclosures of several of the affected operators.
What Hash Price Looks Like Going Forward
The current hash price level is the equilibrium that has emerged from the post-halving rebalancing. Whether it persists depends on three variables.
First, BTC price. A continued price uptrend mechanically supports hash price; a sustained price drawdown would compress margins again, particularly for the surviving mid-tier operators that are not at the bottom of the cost curve.
Second, network hash rate growth. New ASIC generations from Bitmain and the Chinese manufacturers continue to deploy, and the question is whether deployment growth outpaces or lags the BTC-price-driven revenue growth. The current trajectory suggests rough balance, but the dynamics are not stable.
Third, the next halving. The 2028 halving will repeat the subsidy compression and trigger a similar dynamic. The industry that enters that halving will be smaller and more concentrated than the one that entered 2024, and its sensitivity to compressed margins will be different. Whether the survivors of this cycle prepare materially better for the next one is a question the industry has not yet answered.
For now, the post-halving stabilization is real, the survivors are profitable, and the consolidation has produced a more concentrated industry structure. That is a meaningful change from the pre-halving baseline even if the headline hash rate numbers have continued to grind higher.
