The corporate Bitcoin treasury model — the strategy of issuing equity and debt to accumulate BTC on public company balance sheets — is undergoing its most severe stress test since inception. With Bitcoin trading near $59,300 as of June 25, approximately 80% of public companies holding BTC as a tr...
"If we were to fund all of our dividends exclusively by selling bitcoin over the next year, we would buy 20 bitcoin for every one we sold. So it's no different than buying 20 bitcoin and selling no bitcoin." — Michael Saylor, Executive Chairman, Strategy Inc., Q1 2026 Earnings Call
The corporate Bitcoin treasury model — the strategy of issuing equity and debt to accumulate BTC on public company balance sheets — is undergoing its most severe stress test since inception. With Bitcoin trading near $59,300 as of June 25, approximately 80% of public companies holding BTC as a treasury asset are sitting on unrealized losses. The June 2026 drawdown has erased an estimated $62 billion in combined market capitalization from these firms.
Strategy Inc. (NASDAQ: MSTR), the largest corporate holder with 847,363 BTC, carries approximately $12.55 billion in paper losses against an average cost basis of $75,646 per coin. Its shares closed at $96.31 on June 25 — down 78.37% from July 2025 highs and trading at roughly 0.81x the net asset value of its bitcoin holdings. The company's STRC perpetual preferred stock trades at $80.84, a 19% discount to its $100 par value. The model that once commanded premium valuations now trades at a discount.
This report examines the structural mechanics behind the unwind, which companies are selling and why, and what the data implies for the viability of leveraged corporate Bitcoin treasury strategies going forward.
The corporate Bitcoin treasury strategy rests on a simple feedback loop: issue equity or convertible debt at a premium to bitcoin NAV, use the proceeds to buy more BTC, and let the resulting BTC-per-share accretion justify continued premium pricing. When BTC appreciates, the loop is self-reinforcing. When BTC depreciates, it reverses.
Bitcoin peaked above $108,000 in December 2024 and has since declined approximately 45%, settling near $59,300 by late June 2026. Spot Bitcoin ETFs have recorded net outflows of approximately 62,794 BTC over a three-week period, the second-largest outflow streak on record. The U.S. Personal Consumption Expenditures report released June 25 showed inflation running hotter than forecast, compressing rate-cut expectations and sending risk assets lower.
The capital rotation away from crypto and into AI equities has compounded the problem. According to CryptoBriefing, Bitcoin realized losses averaged $1.35 billion per day over the past week. Multiple crypto firms — Kraken, Ledger, Consensys, and Grayscale — paused U.S. public listing plans, citing unfavorable conditions.
Strategy Inc. holds 847,363 BTC acquired at a total cost of $64.1 billion, yielding an average cost basis of approximately $75,646. At $59,300, the portfolio carries an unrealized loss of roughly $13.9 billion.
On June 1, Strategy disclosed it had sold 32 BTC between May 26 and May 31 for approximately $2.5 million — its first net bitcoin disposition since December 2022. The purpose: funding the dividend on its STRC perpetual preferred stock, which carries an 11.50% annual rate after seven consecutive increases since its July 2025 launch at 9%.
MSTR shares fell more than 5% on the announcement. The stock closed at $96.31 on June 25, against a 52-week high of $457.22 and a 52-week low of $92.28. The company disclosed $2.21 billion in total cash reserves, enough to fund STRC dividends for approximately 10 months.
The pace of accumulation has slowed markedly. In the week ending June 21, Strategy purchased 520 BTC for $34.9 million at an average price of $67,068 — roughly two-thirds less than the prior week's 1,587 BTC. The "never sell" narrative now carries, in Saylor's own words, a deliberate asterisk: he told analysts on the May 5 earnings call that the company would "probably sell some Bitcoin to fund a dividend just to inoculate the market."
Strategy's 32-BTC sale was symbolically significant but operationally immaterial — less than 0.004% of holdings. The more consequential selling is occurring at smaller treasury firms facing acute balance sheet pressure.
Riot Platforms sold 3,778 BTC in Q1 2026 — more than double its 1,473 BTC production — generating $289.5 million. Proceeds funded a strategic pivot into AI and high-performance computing (HPC), including a 10-year capacity agreement with AMD for 25 MW (expandable to 200 MW) at its Corsicana, Texas facility. An earlier December 2025 sale of 2,201 BTC at $88,870 average netted $200 million for the same purpose.
Genius Group liquidated its entire bitcoin position, selling its last 84 BTC to repay $8.5 million in debt. The Singapore-based firm had held up to 440 BTC as recently as March 2025.
Empery Digital sold 370 BTC at an average price of $66,632, generating $24.7 million to fully repay its term loan. The sale also freed approximately 1,800 BTC that had been posted as collateral.
According to CoinDesk, bitcoin treasury sell-offs have accelerated since April 2026, with even sovereign holders like Bhutan reducing positions. The pattern is consistent: debt-funded BTC acquisitions become untenable when the asset declines and capital markets close.
The viability of the treasury model depends on equity trading at a premium to bitcoin NAV. That premium justified dilutive issuance: if MSTR trades at 2x NAV, issuing shares to buy BTC is accretive to BTC per share. When the premium compresses below 1x, the model mechanically reverses — issuance becomes dilutive, and the firm loses its primary funding channel.
Of the 18 largest digital asset treasury vehicles tracked by 21Shares, 13 now trade at a discount to the market value of their crypto holdings. The deepest discounts belong to Metaplanet, Upexi, and Helius.
Metaplanet holds 35,102 BTC at an average acquisition price of approximately $107,607, implying a total cost around $3.78 billion. With bitcoin NAV at roughly $2.32 billion, the firm sits on unrealized losses of approximately $1.45 billion, or negative 38.5%.
Trump Media & Technology Group holds 11,542 BTC at an average cost of $118,529 — approximately $462 million underwater at current prices.
Tesla is a notable outlier. Its 11,509 BTC were acquired at an average cost of approximately $33,539, leaving it roughly $517 million in the green — though it has not added to its position.
Investor Michael Burry has described the dynamic facing treasury companies as a "reflexive unwind": falling BTC prices compress equity premiums, close the issuance window, and convert the model from accumulate-forever to sell-to-survive. The concern is that forced selling by treasury companies adds supply to an already declining market, further depressing prices and deepening losses — a negative feedback loop.
The data supports this concern directionally. According to CryptoQuant, Strategy is approximately $10.6 billion underwater on coins purchased across 2024 to 2026 specifically. Its equity premium to bitcoin NAV has compressed from well above 2x to approximately 0.81x. The issuance window is not fully closed — Strategy continues small purchases — but the scale has contracted sharply.
The preferred stock market adds another layer of fragility. STRC trading at $80.84 against $100 par, at an 11.50% coupon, implies the market prices meaningful risk that the dividend or principal could face impairment. If Strategy cannot issue equity or debt on favorable terms, and bitcoin does not recover, dividend funding eventually requires either larger bitcoin sales or further cash depletion. The $2.21 billion cash buffer provides roughly 10 months of runway — a finite timeline.
For smaller firms without Strategy's scale or brand, the math is worse. Companies like Genius Group and Empery Digital have already liquidated entirely or partially. Riot Platforms is pivoting its business model to AI infrastructure, effectively exiting the pure-play treasury strategy.
CoinMarketCap data shows that by late 2025, more than 200 public companies collectively held an estimated $150 billion in digital assets. Many purchased near cycle highs. According to CryptoTimes, 77.4% of public companies holding BTC in their corporate treasuries are currently below their cost basis, with 65.6% more than 20% underwater.
Not all treasury strategies are equivalent. The firms with the most defensible positions share common traits: low cost basis, no leverage against holdings, and alternative revenue streams.
Tesla acquired its BTC between 2021 and 2022 at roughly $33,539 per coin. Even at $59,300, it maintains substantial unrealized gains. It has no debt secured by bitcoin and generates revenue from its core automotive and energy businesses.
Strategy, despite its underwater position, retains the largest absolute holding and $2.21 billion in cash. Saylor's stated policy — to buy 10 to 20 BTC for every one sold — suggests the company views small sales as a financing tool, not an exit strategy. Whether this stance holds under sustained price pressure remains an open question.
Mining firms that retained bitcoin face mixed outcomes. Marathon Digital (MARA) holds approximately 38,689 BTC but has reduced its stack. CleanSpark holds roughly 13,099 BTC. For miners, the calculus has shifted: hashprice margin compression has made mining increasingly unprofitable for firms lacking access to ultra-low-cost energy, and several — led by Riot — are converting infrastructure to AI and HPC workloads where revenue per megawatt is higher.
The corporate Bitcoin treasury model was designed for a rising market. In a declining one, every structural advantage — leverage, equity premium, convertible issuance — reverses sign. The data from June 2026 shows this reversal in progress: compressed premiums, underwater cost bases, and a growing list of firms selling BTC to service debt or fund strategic pivots.
This does not necessarily mean the model is permanently impaired. Bitcoin has recovered from drawdowns of similar magnitude before, and firms with low cost bases or strong cash positions can survive extended downturns. But the evidence is clear that the "infinite money glitch" characterization that fueled treasury company formation in 2024 and 2025 was a function of a specific market regime — one that no longer holds.
The relevant question is not whether the model works in bull markets (it does, by definition) but whether the firms deploying it can survive bear markets without forced liquidation. For Strategy, with $2.21 billion in cash and a stated commitment to net accumulation, the answer is plausibly yes for now. For the dozens of smaller imitators with higher cost bases, more leverage, and fewer revenue alternatives, the answer is increasingly no.