US spot bitcoin ETFs finished July 2026 with $172.4 million in net inflows, according to SoSoValue. That is the smallest monthly figure since the products launched in January 2024. For comparison, July 2025 alone drew more than $6 billion.
The number that gets quoted is the record low. The more useful framing is that July was positive, and it broke a two-month streak of heavy redemptions.
The two months before it were far worse
| Month | Net flow |
|---|---|
| May 2026 | −$2.43 billion |
| June 2026 | −$4.51 billion |
| July 2026 | +$172.4 million |
June's $4.51 billion of redemptions was the largest single-month withdrawal in the history of these products. Across May and June, roughly $6.94 billion left. Against that, a positive July is a stabilization, not a collapse.
So the accurate description is not "investors are fleeing bitcoin ETFs." It is that demand went flat. The products stopped bleeding and also stopped attracting anything meaningful.
It deteriorated into month end
The monthly total hides a July that got worse as it went. Two large outflow days mid-month, $225 million on July 23 and $240 million on July 24, erased most of the gains built earlier. The final trading day of the month saw a $265.4 million net outflow, the largest single-day withdrawal since July 13, which turned the closing week into a $61.53 million outflow after three consecutive weekly inflows.
A month that ends on its worst day tells you the marginal buyer left before the calendar did. Whether that continues is the thing worth watching in August, and it is why the record-low figure matters more as a trend marker than as a data point.
Where the money went instead
Capital did not leave the asset class. It moved.
| Product | Position |
|---|---|
| XRP spot ETFs | ~$1.5 billion cumulative, on a four-month inflow streak that added over $300 million. Took $27.29 million in July. |
| Solana spot ETFs | ~$1.13 billion cumulative. Took $14.62 million in July. |
| Hyperliquid (HYPE) | $195.16 million net inflows over the trailing 30 days. |
| Ethereum spot ETFs | $342.85 million in July, roughly double bitcoin's total. |
XRP is running the longest active inflow streak in the category. Ethereum, a far larger and more established product set than any of the altcoin funds, out-raised bitcoin by about two to one in the same month.
This is not an altcoin rally
The rotation is narrow, and one number makes that clear.
Dogecoin ETFs recorded flows on just two days in the entire month of July, and finished with roughly $526,000 of net outflows. A product tied to one of the most recognizable names in crypto attracted essentially nothing.
So the money is not spreading indiscriminately down the risk curve. It is going to a specific list: XRP, Solana, Ethereum, and Hyperliquid. Everything else in the wrapper is being ignored. That is selection, not enthusiasm, and it produces a two-tier market where a handful of altcoin ETFs accumulate real assets while the rest exist as listings with no flow behind them.
The structural reason bitcoin is losing this contest
Here is the part that most flow commentary skips, and it is not about sentiment.
Several of the products taking in money pay a yield. Bitcoin ETFs cannot.
Solana ETFs distribute staking rewards in the range of 6 to 7 percent. BlackRock's staking ether product distributes in the range of 3 to 4 percent. A spot bitcoin ETF distributes nothing, because bitcoin has no native staking mechanism to distribute from. Its holders pay a sponsor fee and receive price exposure, and that is the entire product.
For an allocator deciding where a marginal dollar of crypto exposure goes, that is a real difference. A Solana ETF is price exposure plus an income stream. A bitcoin ETF is price exposure minus a fee. When bitcoin is trending hard, nobody cares. When it spends a quarter going sideways, the comparison starts to matter.
That yield is not free, and the mechanics deserve scrutiny before anyone treats 6 to 7 percent as the number they will receive. Staking products withhold a share of rewards before they reach the shareholder, and that withholding can exceed the headline sponsor fee. We went through that arithmetic in our piece on the Morgan Stanley staking trusts, where the product with the lowest advertised fee turns out to be the more expensive one to hold on Solana.
But the direction of travel is clear. As more assets get wrapped in ETFs and more of those wrappers stake, bitcoin becomes the only major crypto asset whose ETF structurally cannot compete on income. That is a durable disadvantage in flat markets and it has nothing to do with anyone's opinion of bitcoin.
August has already answered two of these
Three questions followed from July's data. Two resolved within days of the month closing, and both cut against reading July as the start of a trend.
Would August turn negative? No, sharply the opposite. Spot bitcoin ETFs have taken in $381 million in the opening days of August with no single outflow day, which SoSoValue's data makes the strongest start to any month since April. August 3 alone drew $170.09 million, or 2,657 bitcoin. A handful of August sessions have out-raised the whole of July.
Would ether join the outflow side? Yes. Ether funds shed $11.42 million on August 3, roughly 5,805 ether, reversing the prior session. Ethereum dominance rallied 11 percent through July and has not carried into August.
Would the altcoin list widen? Unclear, and the evidence is thin. Solana, Chainlink, BNB, Avalanche, Polkadot, Dogecoin and Litecoin all posted positive daily flows on August 3, which is a wider set than July's four names. But dollar totals were not published for most of them, and XRP took only $1.15 million that session. Positive is not the same as material.
So the rotation described above is a July phenomenon that has already partly reversed. That does not invalidate the structural argument, because a fee-versus-income comparison plays out over quarters rather than days, and one strong week of bitcoin inflows is not evidence that yield stopped mattering.
It is a direct warning against treating one month of flow data as a trend. That is the most common mistake in flow commentary, and it is why this piece opened with the two months that preceded July rather than July alone.
Securing the unseen
Flow data is one of the few genuinely public windows into what large allocators are doing, and it is routinely reported in a way that inverts its meaning. "Bitcoin ETFs post record low inflows" and "bitcoin ETFs break a two-month outflow streak" describe the same $172.4 million, and each is accurate.
The practical habit is to read the monthly figure, the daily path inside it, and the comparison set together. July's total was positive. Its final day was the worst in three weeks. Its nearest competitor raised twice as much at a fraction of the size. None of those three facts is the story on its own.
Learn more at 0xhades.io/research