Why Institutional Bitcoin Holdings Dropped 10% in Three Months—and What's Replacing Them

The sale marked a historic inflection point. For the first time, the company that had built its brand on relentless accumulation began liquidating to self-fund obligations . It was the third discrete BTC sale of 2026 and the sixth consecutive week without a purchase . The sale followed an $8.33 billion Q2 operating loss, largely driven by impairment on its digital-asset holdings . Strategy's holdings fell to 842,138 BTC, down from 843,775 .

2. MARA Holdings: From HODL to Active Treasury Management

MARA Holdings, the largest publicly listed Bitcoin miner by BTC held, sold 20,880 BTC in Q1 2026 for roughly $1.5 billion to fund operations and debt repurchases, then sold another 2,213 BTC in Q2 at an average price of $73,078 . It did not buy a single Bitcoin in Q2 .

The company's BTC holdings fell 29% year-over-year to 35,577 BTC . It reported a $611.3 million net loss and a 27% revenue decline for Q2 2026 . MARA had formally rewritten its treasury policy in March 2026, authorizing the sale of accumulated stockpile Bitcoin for the first time—a stark departure from its earlier HODL approach .

3. Bhutan: A Sovereign Treasury in Quiet Liquidation

The Royal Government of Bhutan has liquidated roughly 70% of its Bitcoin portfolio since October 2024, cutting reserves from approximately 13,000 BTC to around 3,774 BTC . On August 7, 2026, Bhutan moved another 434.87 BTC (~$28 million), resuming outflows after a month-long pause .

Total 2026 outflows exceeded $233 million, with sales funding the Gelephu Mindfulness City project and national infrastructure . Mining inflows have slowed sharply after the 2024 halving, leaving the kingdom as a "pure seller" with no fresh supply coming in .

4. The Collapse of the Equity-Premium Feedback Loop

The broader institutional reduction was amplified by a roughly 50% Bitcoin price decline—from an all-time high near $126,000 down to about $64,000 . That price collapse broke the equity-premium feedback loop that had fueled Strategy's accumulation: the practice of issuing stock at a premium to its net-asset value and using the cash to buy more BTC. CryptoQuant explicitly warned that the corporate treasury trade is "breaking" as the model that amplified demand collapses .

The Counter-Narrative: $626 Million in ETF Inflows

While direct corporate treasuries were selling, spot Bitcoin ETFs saw a structural inflow. In early August 2026, net inflows reached $626 million, led by BlackRock's IBIT .

This bifurcation—corporations selling, ETF investors buying—explains how the broader institutional category could decline 10% even as fresh capital entered the ETF wrapper. Institutional demand is not retreating from Bitcoin. It is rotating from self-custody corporate treasuries toward regulated ETF vehicles that offer liquidity, custody, and balance-sheet convenience.

What This Means for Bitcoin's Institutional Thesis

The three-month sell-off represents the most severe structural test for the original corporate-treasury model since its inception. The equity-premium arbitrage that made Strategy's approach viable depended on a rising Bitcoin price. When that price fell 50%, the model cracked.

But the simultaneous inflow into spot ETFs suggests a more nuanced picture: institutions are not abandoning Bitcoin exposure; they are re-evaluating the wrapper. ETFs offer regulatory clarity, lower counterparty risk, and easier portfolio management. Whether that ultimately supports a larger institutional base over the long term or simply relocates the same demand is the open question.