Bitcoin treasury companies are losing their financing edge as mNAV premiums fade, forcing firms to choose between buying, debt cuts and AI pivots.
In 2024 and 2025, corporate Bitcoin treasuries all seemed to follow the same playbook. The rules were simple: trade above the value of your coins, sell new shares, buy more, repeat. That playbook required one crucial condition. The company’s shares needed to trade at a premium to its underlying crypto. By June 2026, that premium disappeared for most of the companies that built their story on it.
As of June 22, 2026, BitcoinTreasuries counted 199 public companies holding 1.264 million Bitcoin (BTC), worth about $79 billion at the time of writing.
Source: BitcoinTreasuries
Among those companies, a divide has emerged. Some Bitcoin treasuries can still raise capital against their coins. For others, the balance sheet now drives every decision. As of mid-2026, the first group has become much smaller than the second.
Join us in showcasing the cryptocurrency revolution, one newsletter at a time. Subscribe now to get daily news and market updates right to your inbox, along with our millions of other subscribers (that’s right, millions love us!) — what are you waiting for?
The Scale Of It
The holdings tracked by BitcoinTreasuries are almost entirely Bitcoin, with small slices of Ethereum (ETH), Solana (SOL), XRP (XRP), and BNB (BNB) making up the rest. While a handful of governments hold Bitcoin — mostly from forfeitures — those holdings are outside the scope of this story.
The Mechanism That Built The Model
The number that separates a buyer from a seller in this sector is mNAV, or market net asset value. It compares a company's market value to the dollar value of the crypto on its balance sheet.
A company trading above an mNAV of 1.0 can sell new shares and use the proceeds to buy more crypto. Because each new share sells for more than the crypto it funds, each purchase increases shareholders' crypto-per-share. At an mNAV of less than 1.0, the trade runs in reverse. Selling shares to buy crypto shrinks each share's crypto backing rather than growing it.
That mechanism explains why the treasury trade looked uniform in 2024 but appears fractured now. The premium that made it accretive has compressed across nearly the entire sector, not just for the companies already under pressure.
Strategy Kept Buying After Losing The Edge That Made Buying Pay Off
Strategy's mNAV peaked near 3x to 4x during the 2024 bull run, sat around 2.5x in December 2024, and fell to roughly 1.16x by spring 2026.
The company kept buying anyway, adding 520 BTC on June 22 at $67,068 and bringing its total to 847,363 BTC.
Source: Strategy
On June 10, Investor's Business Daily reported that the stock traded roughly 17% below the breakeven level for effective stock-funded Bitcoin purchases.
Strategy spent four years building a model where its corporate Bitcoin purchases outperformed a simple buy-and-hold strategy. By June, the same purchases risked diluting shareholders. The company is still accumulating Bitcoin. However, Strategy no longer has the same financing advantages it once did.
Metaplanet and Twenty One Respond to the Same Problem Differently
Metaplanet built its name on buying more aggressively than almost anyone else. That strategy worked when its shares traded at a premium to its underlying Bitcoin holdings.
By June 2026, the stock was down 42% for the year and 85% over 12 months. CEO Simon Gerovich said the company would consider buying back shares if mNAV stayed under 1.0.
Source: X
A buyback only makes sense once issuing new shares stops paying for itself. By entertaining stock buybacks, a Bitcoin treasury company is effectively implying the playbook it used in 2024 and 2025 has stopped working.
Source: TradingView
Twenty One has had a rougher version of the same year. Its stock fell from an all-time high of $47 in April 2025 to $5.50 in June 2026.
Controlling shareholder Tether responded with a unique proposal. In April 2026, Tether proposed merging Twenty One with Strike, a Bitcoin financial services company, and Elektron Energy, a Bitcoin mining operation. This would attach Twenty One’s treasury to two businesses that earn revenue independently of Bitcoin’s price.
When a standalone treasury can no longer raise cheap capital against its coins, bolting it onto something with its own cash flow is one of the few moves left.
Staking Gives Ethereum Treasuries More Room, But Not Immunity
BitMine holds 5,672,956 ETH and stakes more than 4.7 million of it. Its mNAV sits near 1.01. That means BitMine’s stock no longer trades at any real premium to its underlying holdings. Still, BitMine’s challenge is milder than Strategy’s or Metaplanet’s. Its staking yield is the reason.
BitMine has faced similar headwinds before. Its mNAV first dropped below 1.0 in late 2025. In response, the company began paying a small annual dividend instead of leaning further on share sales.
SharpLink, the next-largest ETH treasury at 868,699 ETH, has staked nearly all of its position for the same reason. Staking buys these companies time, giving them an advantage compared to pure Bitcoin treasuries. Still, it doesn't exempt them from market pressures.
Read More: Will Bitcoin Survive Quantum Computing? Inside the Race Toward Q-Day
Miners Are Solving a Different Problem Entirely
Riot and MARA get folded into the same conversation as Strategy or Metaplanet, but they're running a different business. They are operating companies with power bills and equipment loans that don't pause for a treasury narrative. Selling part of what they mine resembles traditional cash management, rather than directional trading.
MARA sold 15,133 BTC in March 2026, raised about $1.1 billion, and used it to repurchase convertible notes, holding 36,303 BTC afterward.
Riot sold 3,778 BTC in the first quarter, raised $289.5 million, and held 15,680 BTC at quarter's end. Meanwhile, it reported $33.2 million in data center revenue tied to a 50-megawatt contract with AMD.
The data center buildout running through both companies connects directly to the Bitcoin sales. Mining sites and AI compute sites need the same input: cheap power at scale. A miner shifting capacity toward AI work is redeploying physical infrastructure toward a group of customers offering steadier revenue than Bitcoin mining.
Read More: Bitcoin Miners Are Pivoting to AI: What Does This Mean for BTC?
The Companies That Actually Left
Only a small group of companies sold everything. These are the ones that can be said to have "dumped."
Meitu sold its full position, about 31,000 ETH and 940 BTC, by December 2024, for roughly $180 million combined, and put most of the proceeds toward a special dividend.
Genius Group sold its last 84 BTC to repay $8.5 million in debt. Bitdeer's BTC balance fell to 31 BTC by its March 2026 update, while the company grew its AI cloud and infrastructure revenue in its place.
Empery Digital sits between the exits and the reductions. It sold 370 BTC since late March 2026 for about $24.7 million and used the proceeds to repay a term loan. It also released roughly 1,800 BTC that had been held as collateral, keeping 2,989 BTC on hand. That's debt management with a byproduct of lower leverage.
Sort all of these by motive, and the pattern holds up.
MARA, Genius Group, and Empery sold to handle debt. Meitu sold to fund a dividend and cover business costs. MARA, Riot, and Bitdeer shifted capital or infrastructure toward AI work.
Strategy's 32 BTC sale in June was bookkeeping. Almost none of these decisions reflect a view on where Bitcoin's price is headed. They reflect debts coming due and commitments already made.
Don't Count What Isn't Theirs
Coinbase holds 16,492 BTC on its own balance sheet. Block holds 9,032 BTC. Both of those figures exclude crypto the companies hold on behalf of customers.
Source: CoinMarketCap
What the Split Actually Comes Down To
The divide running through this sector is about capital access, not belief in crypto. Strategy and Metaplanet kept buying after their stocks fell below the value of their own coins. The result was that buying continued without the financing edge that justified it in the first place.
Riot and MARA sold Bitcoin for reasons tied to debt and infrastructure spending, unrelated to their outlook on its price. The companies that exited entirely solved a specific balance sheet problem rather than changing their mind about crypto.
This article contains links to third-party websites or other content for information purposes only (“Third-Party Sites”). The Third-Party Sites are not under the control of CoinMarketCap, and CoinMarketCap is not responsible for the content of any Third-Party Site, including without limitation any link contained in a Third-Party Site, or any changes or updates to a Third-Party Site. CoinMarketCap is providing these links to you only as a convenience, and the inclusion of any link does not imply endorsement, approval or recommendation by CoinMarketCap of the site or any association with its operators. This article is intended to be used and must be used for informational purposes only. It is important to do your own research and analysis before making any material decisions related to any of the products or services described. This article is not intended as, and shall not be construed as, financial advice. The views and opinions expressed in this article are the author’s [company’s] own and do not necessarily reflect those of CoinMarketCap.
