Bitcoin ETF inflows are frequently interpreted as a thermometer of institutional confidence in Bitcoin. However, a recent report by IOSG Ventures argues that on a week-to-week basis, ETF flows are primarily driven by a hidden arbitrage trade known as cash-and-carry, not by conviction. Arbitrageurs buy the ETF while shorting CME futures to hedge price risk, making them indistinguishable from true believers in the data.
The study finds that approximately half of the weekly flow variance can be explained by new short positions added by leveraged funds (hedge funds), with a correlation coefficient of 0.70. Bitcoin's weekly price returns show no statistically significant predictive power over ETF flows. Weekly capital does not chase price performance; it moves in lockstep with a hedged interest-rate trade.
How Arbitrage Contaminates the Headline Numbers
Cash-and-carry arbitrage is a classic strategy: when futures trade at a premium to spot (e.g., spot at $100, three-month futures at $103), a trader can buy the spot (via ETF) and short the futures, locking in the $3 spread. Regardless of Bitcoin's price movement, the profit is fixed (annualized as the basis). As long as the basis exceeds the risk-free rate on T-bills, the trade is attractive.
The problem: buying the ETF registers as an inflow, identical to a long-term holder. But the arbitrageur leaves a second footprint: a short position on CME futures of equal value. True believers leave one footprint; arbitrageurs leave two. The CFTC's weekly Commitments of Traders report for CME Bitcoin futures discloses the positions of leveraged funds, making it possible to track this activity.
What the Data Shows: Flows Follow Futures, Not Price
Since the ETF launch, weekly additions to leveraged funds' short positions and ETF inflows have moved almost one-to-one. Regression analysis shows that about 50% of weekly flow variation is explained by changes in short positions (correlation 0.70). Bitcoin's own price return explains virtually nothing. As a weekly signal, ETF demand is largely arbitrage activity.
However, arbitrage dominates the variance, not the level. Of the cumulative ~$55 billion net inflows into Bitcoin ETFs, the arbitrage component currently accounts for only about $10 billion. The rest is steady directional buying of roughly $400 million per week, which compounded over two years forms the bulk of the “mountain.” In asset terms, the hedged portion peaked at ~14% of ETF assets in 2024 but has now fallen to 4%–5%.
A more accurate statement: ETF flows overstate the volatility of conviction, not its level. The weekly ups and downs are largely “rented” – arbitrage capital that comes and goes – while the truly settled assets are mostly “owned.”
This Trade Is Unwinding
Leveraged funds' short positions grew from ~$3 billion at launch to a peak of ~$14 billion by end of 2024, then steadily declined to ~$4.5 billion. In June, shorts halved from ~$6.4 billion to $4.3 billion, while ETFs recorded daily outflows of $300–500 million. Looked at superficially, this appears like panic. But with futures data, it is a routine liquidation of a trade that is no longer profitable – a different story behind the same outflow numbers.
When the basis compresses to near risk-free rates, the trade stops being worthwhile. Weekly demand weakens exactly when the basis compresses. True believers don't care about futures basis; this weekly demand clearly does.
Ethereum ETFs: The Same Trade, But the Numbers Don't Add Up
Applying the same test to Ethereum ETFs reveals weaker patterns: lower correlation with futures shorts and almost no steady directional buying. The reason: holding spot ETH instead of futures means forgoing staking yields of ~3%–4% per year. After deducting that, the ETH basis is often negative, making arbitrage unprofitable. Ethereum ETFs therefore lack both strong conviction buying and robust arbitrage support; they are smaller and noisier than their Bitcoin counterparts.
How to Read ETF Flows Going Forward
The key takeaway is not a price call but a framework. When the basis is rich, expect seemingly strong “institutional demand” – much of it hedged. Don't mistake that strength for conviction. When the basis compresses, expect inflows and shorts to fade together – don't mistake the resulting outflows for a market verdict on Bitcoin. The two numbers to watch are: the annualized basis relative to T-bill yields, and the net short position of leveraged funds in the weekly CME report. They will tell you how much of the next “demand” headline is real.
IOSG notes that this analysis is not about predicting prices but about correctly interpreting flow data. The real directional buying is genuine and patient, and today constitutes the vast majority of assets. The “rented” portion has been going home for two years.