Public companies hold 1.26 million BTC — over 6% of Bitcoin's fixed 21-million supply — valued at approximately $79 billion. That figure grew by 110,000 BTC in Q2 2026 alone, 1.8 times the total acquired in the prior two quarters combined, and 2.1 times what miners produced in the same period. Bu...
"We will probably sell some bitcoin to pay a dividend just to inoculate the market and send the message that we did it." — Michael Saylor, Executive Chairman, Strategy Inc., Q1 2026 Earnings Call
Public companies hold 1.26 million BTC — over 6% of Bitcoin's fixed 21-million supply — valued at approximately $79 billion. That figure grew by 110,000 BTC in Q2 2026 alone, 1.8 times the total acquired in the prior two quarters combined, and 2.1 times what miners produced in the same period.
But the aggregate number masks a structural fracture. The corporate Bitcoin treasury sector has split into two camps: firms that are doubling down on accumulation, and firms that are liquidating holdings to service debt, fund dividends, or pivot into artificial intelligence infrastructure. Strategy Inc., the largest holder at 843,775 BTC, sold 3,588 coins in early July to cover preferred stock dividends — its second sale in weeks after breaking a years-long "never sell" pledge. Meanwhile, Strive added 21 BTC on July 17, Metaplanet bought 2,823 BTC in Q2, and American Bitcoin Corp pushed its stack to 8,000 BTC on July 6.
The divergence is not random. It tracks directly to a single metric: the market-to-net-asset-value ratio (mNAV). When it sits above 1.0, the financial engineering works. When it drops below, the model breaks. Strategy's enterprise mNAV fell below 1.0 in late June 2026 — the first time the market valued the company at less than the Bitcoin on its balance sheet. That threshold separates the accumulators from the liquidators.
Approximately 198 public companies held Bitcoin on their balance sheets as of early July 2026, according to BitcoinTreasuries.net. Combined holdings stood at 1.268 million BTC, worth roughly $79 billion at a BTC price of approximately $64,200.
Q2 2026 saw the largest quarterly corporate Bitcoin acquisition on record: 110,000 BTC purchased by public companies, per data compiled by CryptoBriefing. That figure exceeded the prior two quarters combined by a factor of 1.8x. It also outstripped new Bitcoin mining production by 2.1x, meaning corporate treasuries absorbed more than double the coins miners added to circulating supply.
The top five public holders account for the vast majority of corporate BTC:
| Company | BTC Holdings | Approx. Value (USD) | |---|---|---| | Strategy (MSTR) | 843,775 | $54.2B | | Twenty One Capital (XXI) | 43,514 | $2.8B | | Metaplanet (3350.T) | 43,000 | $2.8B | | Strive (ASST) | 19,921 | $1.3B | | American Bitcoin Corp | 8,000 | $514M |
Strategy alone holds 66.5% of all corporate Bitcoin. This concentration creates an asymmetric dependency: the health of the corporate treasury sector is, in practice, the health of one company.
A subset of treasury firms continued buying through mid-July 2026, undeterred by BTC trading roughly 57% below its November 2024 highs.
Metaplanet added 2,823 BTC in Q2 2026 for $225 million (35.886 billion yen) at an average price of $78,872 per coin, lifting total holdings from 40,177 to 43,000 BTC. The Tokyo-listed firm reported $10.95 million in operating revenue from its bitcoin options trading segment during the quarter. This income stream — using options strategies to generate yield on held BTC — partially subsidizes further accumulation.
Strive purchased 21 BTC between July 13 and July 17 at an average price of $63,221, bringing holdings to 19,921 BTC. The company has continued small, consistent purchases throughout 2026.
American Bitcoin Corp increased its position to 8,000 BTC with a purchase recorded on July 6. The firm ranks as the 16th-largest public corporate holder, with a stack valued at approximately $495.7 million.
Twenty One Capital, formed through a combination of Cantor Equity Partners, Tether, and SoftBank, holds 43,514 BTC — a position unchanged since the company's formation. The Austin-based firm was structured from inception as a Bitcoin treasury vehicle and has not made additional purchases or sales.
On the opposite end, several firms sold BTC in 2026 — some partially, others entirely.
Strategy sold 3,588 BTC for approximately $216 million during the week ending July 5, 2026, at an average price of roughly $60,000 per coin. Proceeds funded scheduled dividend payments on four series of preferred stock: STRC (monthly), STRF, STRE, and STRD (quarterly). This followed an initial sale of 32 BTC for $2.5 million between May 26 and May 31, which triggered a 31% decline in the MSTR share price over the following month. The company's aggregate cost basis across its 843,775 BTC stands at $63.69 billion, or $75,476 per coin — roughly 17.5% above the spot price as of July 20.
Bitdeer Technologies reduced its Bitcoin holdings to zero as of February 20, 2026, liquidating its entire treasury to fund a pivot into AI data center infrastructure.
Core Scientific sold 1,992 BTC worth $175 million in March 2026 to finance operational transitions toward high-performance computing.
Riot Platforms sold approximately $200 million in Bitcoin to fund operations and AI expansion.
K Wave Media redirected $485 million of its remaining equity purchase facility — originally structured for Bitcoin accumulation — toward AI infrastructure investments under an amended agreement with Anson Funds. The company subsequently sold its remaining 88 BTC on July 1 to repay $6 million in debt, bringing holdings to zero.
Genius Group, an AI-focused education company that once held 440 BTC, exited its position entirely, selling its remaining 84 BTC to repay $8.5 million in outstanding debt.
The divergence between accumulators and liquidators maps to a single variable: the market-to-net-asset-value premium.
Bitcoin treasury firms historically traded at mNAV ratios well above 1.0. A company holding $1 billion in BTC might trade at an enterprise value of $2 billion, reflecting the market's willingness to pay a premium for leveraged Bitcoin exposure packaged in equity form. That premium enabled a self-reinforcing cycle: issue shares or convertible debt at favorable terms, buy more Bitcoin, increase holdings per share, attract more capital.
By late June 2026, Strategy's enterprise mNAV dropped below 1.0 for the first time. With the stock at approximately $82 — down roughly 85% from its November 2024 high — Strategy's enterprise value sat at about $50.4 billion, while its Bitcoin holdings were worth approximately $51.1 billion. The market valued the company at less than the sum of its Bitcoin.
At sub-1.0 mNAV, the financial engineering reverses. New equity issuance becomes dilutive because shares are being sold at prices below the per-share Bitcoin value. Convertible debt, a tool Strategy used extensively to scale its position, becomes prohibitively expensive or unavailable. The self-reinforcing accumulation loop stopped functioning.
MSTR recovered to $100.77 on July 2, but the structural damage to the thesis remains. According to NYDIG's Greg Cipolaro, the mNAV metric itself is "woefully deficient" for failing to reflect balance-sheet risks embedded in convertible notes and preferred equity obligations.
A distinct but related trend: Bitcoin miners are selling treasury holdings and converting infrastructure to AI workloads.
By early March 2026, public miners had sold over 15,000 BTC, according to CoinDesk reporting. The sellers include Riot Platforms, MARA Holdings, Bitdeer, Hut 8, TeraWulf, and Bitfarms — firms that historically mined and held Bitcoin as a core strategy.
The economic rationale is straightforward. AI data center contracts offer contracted revenue with margins reportedly above 85%, compared with volatile mining economics subject to hash rate competition, energy costs, and Bitcoin price fluctuations. Industry projections cited by Intellectia.ai suggest AI and high-performance computing could account for 70% of revenue for transformed miners by year-end 2026, representing a potential $40 billion revenue opportunity.
This miner-to-AI pipeline has a secondary effect on Bitcoin markets: it removes a category of structural holders. Miners who once accumulated mined BTC on their balance sheets now sell production immediately — or liquidate existing treasuries — to fund infrastructure capex.
Corporate treasuries now control 6.03% of Bitcoin's total 21-million supply. Including government holdings and exchange reserves, the proportion of Bitcoin supply held by institutional and quasi-institutional actors has grown significantly.
The concentration in a single entity — Strategy holding 66.5% of all corporate BTC — creates tail risk. Any forced liquidation event, whether driven by margin calls on convertible debt, regulatory action, or a liquidity crisis requiring rapid asset sales, would introduce supply volume that markets may struggle to absorb at current prices.
Strategy's aggregate cost basis of $75,476 per BTC sits approximately $11,000 above current spot prices. The company is, on a mark-to-market basis, underwater on its aggregate position. While this does not trigger an immediate liquidity event, it narrows the margin of safety and increases sensitivity to further price declines.
The corporate Bitcoin treasury thesis has entered a new phase. The aggregate data — record quarterly purchases, holdings exceeding 6% of total supply — suggests accelerating institutional adoption. The disaggregated data tells a different story: a sector splitting along mNAV lines, with well-capitalized or operationally profitable accumulators on one side and overleveraged or strategically pivoting liquidators on the other.
The accumulators benefit from either operational cash flow (Metaplanet's options income), structural positioning (Twenty One Capital's founding endowment), or disciplined incremental buying (Strive's small weekly purchases). The liquidators share a common feature: their original thesis depended on equity premium financing that no longer exists at sub-1.0 mNAV levels.
The miner-to-AI migration adds a compounding factor. As miners convert from holders to sellers, Bitcoin loses a class of natural accumulators with direct access to newly minted supply. The net effect is a shift in market structure: corporate demand increasingly flows through a smaller number of larger, more leveraged entities, while the base of diversified holders narrows.
Whether the remaining accumulators can sustain buying at sufficient volume to offset liquidation pressure — while Bitcoin trades 57% below cycle highs and Strategy's cost basis sits 17.5% above spot — will determine whether corporate treasury holdings represent a structural demand floor or a concentrated source of future sell pressure.