Analysis of Bitcoin ETF Fund Fluctuations: The Institutional Allocation Logic from Outflows in Q1 to Recovery in Q2

In April 2026, U.S. spot Bitcoin ETFs recorded approximately $2.44 billion in net inflows, delivering the strongest monthly performance since October 2025. After five prior consecutive months of persistent outflows, capital finally appears to have turned around. Market sentiment improved accordingly, and discussions about “institutions returning” have picked up again.

However, in a trading day in mid-May, ETFs saw an estimated single-day net outflow of about $635 million—like a bucket of cold water thrown on the repair expectations that had just started to catch fire.

This set of data outlines the core characteristics of the current Bitcoin ETF capital flows: not a V-shaped rebound, and not a free fall, but a tug-of-war that alternates between repair and wavering. Stepping back from the pessimism of Q1 and looking at the end of May, the question that really deserves to be asked is no longer “Why did it bleed in Q1,” but “What exactly does the April inflow mean”—is it the beginning of a cyclical repair, or a tactical breather amid a large-scale exodus?

Starting from Q1 capital outflows as the analytical baseline, this article connects the latest capital flow dynamics from April to May, and dissects the true strength of Bitcoin ETF capital recovery from three dimensions: divergence in institutional behavior, the macro environment, and market narratives. The “Bitcoin ETF” discussed in this article refers to spot Bitcoin ETF products listed in the U.S. market.

From Record Inflows to a Tug-of-War in Recovery: The Three-Stage Evolution of Capital Flows

Looking back at Bitcoin ETF capital flows from 2024 to the present, they can be clearly divided into three stages. The current market is in the internal tug-of-war of the third stage—this is the key time anchor for understanding the logic behind institutional behavior today.

First Stage (January 2024 to October 2025): Explosive Growth Period. Since the SEC approved the first batch of spot Bitcoin ETFs, institutional capital has poured in at an unprecedented pace. According to SoSoValue data, cumulative net inflows reached a peak of about $61.19 billion in October 2025, while Bitcoin prices hit an all-time high of $126,193. ETFs were widely viewed as “enduring buy pressure,” and the institutional narrative reached its high point during this period.

Second Stage (November 2025 to February 2026): Sustained Outflows Period. Capital flows reversed direction. In November 2025, monthly outflows were about $3.5 billion; in December, about $1.1 billion; in January 2026, about $1.61 billion; and in February, outflows continued at about $206 million. Over four months, total net outflows were about $6.38 billion. The main selling pressure came from the leading products IBIT and FBTC, with large hedge funds and some endowments strategically cutting exposure as the driving force behind the exodus.

Third Stage (March 2026 to Present): Recovery and Tug-of-War Period. March recorded net inflows of $1.32 billion, ending four months of outflows; in April, net inflows further reached about $2.44 billion, the strongest single month in nearly half a year. However, the recovery is not solid: based on publicly available data (May 4, 2026), cumulative net inflows are about $58.72 billion, still about $2.5 billion below the peak; the single-day outflow of about $635 million in mid-May directly exposes the fragility of the recovery.

The core fact revealed by these three stages is this: ETF capital flows have shifted from the “trend-based inflows” of 2024 to a “repeated tug-of-war between recovery and withdrawal.” This shift itself is more worth attention than the outflow numbers in a single quarter.

Data Breakdown: Who Is Coming Back, and Who Is Still Watching

Structural Differentiation in Capital Recovery

Breaking down April’s $2.44 billion net inflow reveals many issues through its structural characteristics.

Top products are pulling back capital again, but concentration keeps rising. BlackRock’s IBIT attracted about 70% of total U.S. Bitcoin ETF inflows in April—about $1.71 billion. As of May 23, 2026, IBIT’s assets under management were about $61.1 billion, accounting for roughly 62% of the total U.S. spot Bitcoin ETF market. The direction of capital returning is highly concentrated—not a broad rise across the entire ETF landscape, but top products absorbing liquidity one-way. This means the recovery is not a widespread revival of confidence, but capital concentrating toward the most liquid, most stable-branded products amid uncertainty.

The institutions that fled most aggressively earlier have not returned at scale. During Q1, hedge fund Brevan Howard reduced its IBIT holdings by about 86%, and Harvard University endowment cut 43%. Whether these institutions replenished positions in April is not confirmed by publicly available 13F filings. Logically, unless there is a fundamental change in macro signals, strategic position reductions are unlikely to be fully reversed within just one to two months. A more likely scenario is that April’s inflows came more from new allocation-oriented capital or existing Bitcoin holders who had not previously participated in the outflow, rather than from the repurchase of those who sold earlier.

Trading activity volatility trends downward. According to Cointelegraph data, Bitcoin ETF monthly trading volume in Q1 showed a choppy pattern: about $87 billion in January, rising to about $93 billion in February, then falling back to about $79 billion in March. The overall downward trend in trading activity sent warning signals earlier than the net outflow data—institutions not only reduced holdings at certain stages, but also lowered trading frequency. April’s trading volume rebounded somewhat, but it remains below the level of Q4 2025 overall, so the breadth of participation in the recovery still needs a question mark.

Cross-Comparison with the Same Period in 2024/2025

Even after factoring in April’s inflows, the overall pace of capital flows in 2026 is still clearly lagging. According to JPMorgan’s analysis, total digital asset net flows in Q1 2026 were about $11 billion—only one-third of the same period in 2025. If the timeline is extended: according to public data, full-year 2025 net inflows were about $21.35 billion to $22 billion; from the start of 2026 through May, net inflows were only about $1.47 billion, forming a clear deceleration curve. The decelerating growth trend—about $35.24 billion in 2024 → about $21.5 billion to $22 billion in 2025 → about $1.47 billion from early 2026 to date—suggests that the recovery has not yet returned to a growth track.

Authenticity Check for the Recovery: Three Analytical Frameworks

Against the backdrop of the current ETF capital tug-of-war, the market has very different interpretations. Below, we examine the true quality of April’s inflow from three angles.

Framework One: Cyclical Recovery Theory (Optimistic, but Dependent on Preconditions)

This framework holds that the recovery in March and April proves that allocation demand was merely temporarily suppressed by macro uncertainty. Once the Federal Reserve releases rate-cut signals in the second half of the year, capital inflows should accelerate, and cumulative net inflows could potentially return to the $61.19 billion peak. In earlier observations, Bloomberg ETF analyst Eric Balchunas pointed out that Bitcoin ETF investors showed holding resilience beyond traditional assets during roughly 40% drawdowns—this provides partial support for the framework.

The quality of the recovery depends on the Federal Reserve’s policy path. If there is only one rate cut for the entire year, or if rates remain unchanged, the premise behind the recovery logic would be greatly weakened. The large single-day outflows that reappeared in May have already formed the first stress test for the recovery thesis.

Framework Two: Structural Slowing Theory (Neutral to Cautious, Strong Data Support)

This framework argues that the decelerating growth curve shows that ETF penetration is approaching a stage-level ceiling. The “low-hanging fruit” after ETF approval in 2024—meaning institutions that previously could not allocate to Bitcoin due to compliance barriers—have largely already entered. Future incremental gains will be more difficult to obtain, and capital flows will likely remain “pulse-like” rather than “trend-like” over the long term; the April inflow is merely an expression of a pulse in this context.

Confirmation of the new investor structure. If subsequent 13F filings show that April’s inflows mainly came from allocation-oriented rather than trading-oriented institutions, this framework would be weakened; conversely, if the returning capital is dominated by trading-oriented funds that move quickly in and out, then the structural slowdown view becomes more convincing.

Framework Three: Supply Diversion Theory (Valid in the Medium to Long Term, Hard to Quantify in the Short Term)

This framework emphasizes the surge in the supply of crypto ETF products. In Q1 2026, about 26 new single-asset crypto ETF applications or listings were already submitted or launched, covering Ethereum, Solana, XRP, and more. Asset choice diversification is beneficial for the industry in the medium to long term, but in the short term it diverts capital that could have been concentrated in Bitcoin ETFs. Bitcoin ETFs are no longer the only institutional-grade tool for allocating to crypto—this is a variable that did not exist in 2024.

Whether ETF capital flowing into Ethereum and other assets creates a clear “see-saw effect” with Bitcoin ETFs is not yet verified due to insufficient data. However, the existence of the factor itself cannot be ignored.

Industry Impact: From a Unidirectional Engine to a Bidirectional Channel

A Fundamental Shift in the Role of ETFs

In 2024, Bitcoin ETFs were a unidirectional engine for market gains: every $1 inflow meant the fund had to buy Bitcoin in the open market. The sustained outflows from November 2025 to February 2026 proved that the reverse mechanism is equally effective—every $1 outflow means the fund must sell Bitcoin in the market. The ETF’s “structural buy” characteristic only holds when there are net inflows; during redemption cycles, it turns into structural sell pressure of equal strength.

This mechanism’s real-world validation has profound implications for institutional allocation logic. In a market with relatively limited liquidity, large ETF redemptions can create price impacts far greater than those in the stock market. This may push more institutions in the future to adopt a more cautious cadence—building positions in batches, setting stricter stop-loss discipline, or hedging ETF position risks through derivatives.

A Positive Side Product of Capital Exodus

It is worth noting that Q1 capital outflows did not trigger a full-scale market crash. BTC holdings on centralized exchanges decreased continuously from over 3.2 million in 2023 to fewer than 2.7 million in March 2026, indicating that ETF outflows do not equal all holders selling. The market is shifting from trading-held to long-term custody, which objectively reduces supply pressure on the float.

At the same time, Morgan Stanley expanded crypto trading services through its E-Trade platform in 2026, offering BTC, ETH, and SOL trading to millions of retail brokerage clients. This shows that even if ETFs face short-term headwinds, traditional financial institutions are still moving forward with building infrastructure channels for crypto assets—their assessment of long-term demand has not changed due to one quarter’s capital flow.

Conclusion: Survival Rules for a Tug-of-War Period

As of May 25, 2026, according to Gate market data, Bitcoin is trading at $77,148.1, up 11.76% over the past 30 days but down 22.08% over the past year. ETF capital has undergone a dramatic switch from continuous outflows to phased recovery, and the volatility in May is a reminder to everyone: the recovery is not yet complete.

Looking back at Q1 from the end of May, the truly important takeaway is not the number “$500 million outflow” itself—relative to total AUM of about $86.9 billion at the end of Q1, it does not constitute a systemic threat. The important takeaways lie in the confirmation of three structural signals:

First, ETFs are bidirectional channels. They can carry massive inflows as well as concentrate selling pressure for release. Any assumption based on “ETFs equal perpetual buy pressure” needs to be revisited.

Second, institutions are not one whole. The fragmented picture in Q1—some institutions cutting positions sharply, while others add holdings against the trend—continues into the capital tug-of-war from April to May. Future directions of investment decisions will depend on how these different entities assess the macro outlook, rather than a vague “institutional consensus.”

Third, the authenticity test of the recovery is still ongoing. The strong April rebound was an important vote of confidence, but high concentration, the downward drift in trading activity volatility, and the reappearance of outflows in mid-May all indicate that the capital recovery has not yet passed a stress test. The real test is not the strength of a rebound in a single month, but whether, when the next macro shock arrives, the returning capital chooses to stay or exit.

For market participants, the most important thing to avoid in a tug-of-war period is letting emotion be over-driven by single-month data. April’s rebound should not be interpreted as a signal of a trend reversal, and May’s volatility does not need to be seen as evidence that the recovery has ended. During the transition from trend-driven to event-driven ETF capital flows, the direction of data is more valuable than the absolute value of the numbers, and confirming that direction requires not just one month, but the accumulation of evidence across a whole quarter.