Crypto funds posted $1.67 billion weekly outflow as bitcoin led redemptions

CoinShares reported on June 1, 2026 that digital-asset investment products recorded $1.67 billion of net outflows in its latest weekly measurement, extending withdrawals to a third consecutive week. The firm called it the second-largest weekly outflow of 2026, behind only the week associated with January 23.

The result mattered because it measured investor activity through regulated and listed products rather than cryptocurrency transfers between anonymous wallets. It showed a broad retreat from exchange-traded exposure at a moment when bitcoin and ether products had become important institutional access routes. It did not, however, measure every crypto fund, every exchange or capital moving directly on-chain.

Bitcoin carried most of the withdrawals

CoinShares attributed $1.438 billion of the weekly outflow to bitcoin investment products, the largest weekly bitcoin-product withdrawal it had recorded in 2026 through June 1. Ether products accounted for another $257 million. The United States contributed $1.63 billion of regional outflows, while CoinShares reported smaller withdrawals of $25.7 million from Germany, $6.6 million from Sweden and $4.5 million from Hong Kong.

Those figures are provider estimates for CoinShares’ tracked universe. The dated report described the measurement as weekly but did not publish an exact start time, cutoff time or time zone in its accessible text. “Weekly” here therefore means CoinShares’ reporting interval immediately preceding its June 1 publication, not a Coinburn reconstruction of seven calendar days.

The components also should not be mechanically added into a new total. Asset categories and country categories are different views of the same product flows, and the published figures are rounded.

A three-week institutional retreat

CoinShares put cumulative outflows over the three consecutive negative weeks at $4.21 billion. It said assets under management fell to $141 billion from $148 billion in the prior weekly report, the lowest level since early April 2026.

That $7 billion change in assets under management was not presented as redemptions alone. Fund assets move with both net subscriptions and the market value of the underlying holdings, so the difference between $148 billion and $141 billion cannot be treated as a separate $7 billion investor withdrawal.

Breadth narrowed as well. CoinShares said only five assets registered meaningful inflows above $1 million, down from nine in the preceding report. It identified XRP at $20.3 million, Hyperliquid at $10.8 million and Near at $7.6 million as the notable positive exceptions. The contrast suggested selective demand rather than an equal withdrawal from every tracked asset.

What the report could and could not show

CoinShares interpreted the three-week run as Iran-related risk aversion overwhelming any supportive effect from progress on the U.S. CLARITY Act. That was the research firm’s contemporaneous explanation, not a controlled causal finding. Fund-flow data can establish estimated creations and redemptions within a covered product universe; it cannot by itself prove why investors acted or how much a particular headline moved cryptocurrency prices.

Contemporaneous reports from CoinDesk and The Block independently carried the same $1.67 billion weekly total, $4.21 billion three-week total and $1.438 billion bitcoin-product figure, all attributed to CoinShares. Their agreement corroborates what the June 1 report said, but it does not create a separate underlying dataset.

The event-day reading

As of June 1, 2026, the defensible conclusion was narrow but consequential: regulated digital-asset products were experiencing one of their heaviest withdrawal weeks of the year, and bitcoin products dominated the retreat. The record did not establish a permanent rejection of crypto exposure, a completed market cycle or a forecast for subsequent prices.

The next checks were whether a fourth weekly outflow followed, whether assets under management stabilized, and whether issuer-level or jurisdiction-level data reconciled with CoinShares’ aggregate. Those outcomes were not knowable from the June 1 publication and are not projected backward into this reconstruction.