Bitcoin rallied after a Fed hike. Who bought it?

Bitcoin climbed past $87,000 after a rate increase that should have made risk harder to own. The public record identifies several sources of demand, but it cannot put a name to every buyer or turn fund inflows into a complete account of the rally.

Bitcoin’s rally after the Federal Reserve raised interest rates has an answer that can be measured, though not a single buyer who can be named. Money returned to US spot Bitcoin exchange traded funds. A public company resumed purchases. Traders caught on the wrong side of the rise had to close positions. By September 21, Bitcoin had moved above $87,000, then surrendered part of the gain two days later.

The timing matters. The Federal Open Market Committee raised its target range by a quarter point to 3.75% to 4.00% on September 16. Its September projections put the median year end policy rate at 4.1% in both 2026 and 2027. In June, those medians were 3.8% and 3.6%. The Fed did not promise a cut.

A hike had been expected by many traders. Expectations alone, however, cannot account for the subsequent purchases. The useful question is narrower: which observable channels brought demand into Bitcoin after the decision, and which claims about the buyers go beyond the evidence?

The funds first lost $746 million

Farside Investors’ daily fund table records $450.4 million of net withdrawals from US spot Bitcoin ETFs on September 15 and another $295.9 million on the day of the Fed decision. Add them: $746.3 million left over two trading sessions. The decision was hardly greeted by an immediate flood into the funds.

The direction changed on September 17. Funds took in $159.5 million, followed by $433.0 million on September 18, $999.0 million on September 21 and $714.7 million on September 22. SoSoValue’s September 23 figures, reported by crypto.news, add $346.98 million for a fifth consecutive positive session.

Together, the five figures total $2,653.18 million. Subtract the $746.3 million of withdrawals from September 15 and 16 and the eight trading days show a net $1,906.88 million entering the funds. The latter is an accounting window, not a measure of money that bought Bitcoin at the rally’s exact hour. It is the net of two sharply different periods, and reporting only the five positive sessions would conceal the withdrawals immediately before the turn.

There is a data trap here. An earlier snapshot of Farside’s September 23 table showed only $32.4 million, with several issuers’ entries still blank. The later SoSoValue reading included $166.29 million for BlackRock’s IBIT and $143.24 million for Fidelity’s FBTC, among other reported fund flows. Treating a blank as zero would understate that day’s total by more than $300 million. These are dated snapshots, and totals can change when issuers report.

The apparent buyers changed between Monday and Wednesday

Monday’s $999.0 million was spread among several funds. Farside recorded $381.4 million for BlackRock’s IBIT, $289.1 million for ARK 21Shares’ ARKB and $238.8 million for Fidelity’s FBTC. Combined, those three accounted for $909.3 million, or about 91% of that day’s total. This identifies fund vehicles, not the institutions or people placing orders through them.

Tuesday’s flow was $714.7 million in Farside’s later table. IBIT received $350.3 million and FBTC $257.4 million. Those two accounted for about 85% of the daily total. By Wednesday, Bitcoin had retreated toward $84,000 even as SoSoValue reported another $346.98 million into the funds.

That final pairing is as important as the Monday surge. A positive ETF print does not guarantee a positive Bitcoin session. Fund subscriptions are one channel of net demand; sellers on exchanges, derivatives positions and the timing of fund hedging all affect the price. The five day fund streak establishes persistent subscriptions. It cannot prove those subscriptions alone pushed Bitcoin through $87,000.

The concentration figures matter for the same reason. On Monday, IBIT, ARKB and FBTC supplied more than nine tenths of net additions; that does not mean three asset managers independently decided to buy Bitcoin with their own balance sheets. A fund can receive orders from brokerage customers, registered advisers and institutions. The issuer reports the fund level flow. Its table does not sort the orders by investor type. Calling all $999 million institutional buying would add a claim the data do not contain.

Monday and Wednesday also illustrate why two kinds of accounting should stay separate. A fund flow is a net change in assets associated with subscriptions and redemptions. A traded fund share can change hands repeatedly between investors in the secondary market without producing the same amount of new fund creation. Price, exchange turnover and ETF net flow are different measurements. A headline can be accurate on one and misleading on another.

ETF inflows are net creations or subscriptions valued in dollars, not a public register of every underlying investor or a second by second ledger of coins bought. Market makers and authorized participants can bridge a fund trade and its underlying hedge at different times. Nor does a dollar flow translate into an exact Bitcoin quantity without choosing a price and knowing when the exposure was acquired. Converting $999 million into BTC at the day’s closing quote would produce an illustration, not an audited purchase count.

A corporate purchase is real, but its clock is different

Strategy provides a named buyer. Its September 21 Form 8-K reports the purchase of 950 BTC for $75.7 million, including fees and expenses, at an average $79,670 per coin. The transactions occurred between September 14 and September 20. The filing gives the period, not the individual trades’ timestamps.

That distinction rules out an easy claim. Strategy’s announcement landed on September 21, when Bitcoin rallied, but the filing does not show that Strategy bought Bitcoin during Monday’s surge. Its purchase may have happened before, after or across the September 16 rate decision. It used existing USD Cash and said it issued no shares through its at the market program during the period.

Strive’s September 21 filing identifies another corporate buyer: 1,355 BTC at an average price of roughly $79,475 during September 14 to 18. Its disclosure has the same dating problem for anyone trying to explain a particular candle. The combined reported 2,305 BTC shows corporate accumulation over overlapping periods; it does not measure corporate buying on September 21.

Strategy’s 950 BTC can be compared with its own past and future disclosures. It cannot simply be added to $2.65 billion of ETF inflows and called total market demand. The windows overlap, the units differ and other buyers and sellers are missing.

Some buyers were closing losing bets

The other identifiable class of buyer did not necessarily want to own Bitcoin for months. A trader short a Bitcoin perpetual or futures contract must buy back exposure to exit. If price rises quickly, liquidation can force that purchase. Those buy orders can add fuel to the move that made the short untenable.

Nansen senior research analyst Nicolai Sondergaard described the rally as a combination of ETF demand and short covering. A September 23 crypto.news account quoted CoinMarketCap research lead Alice Liu saying covering, rather than new buying, drove much of the rise. These are analysts’ interpretations, not a trader by trader audit.

One check comes after the peak. A CryptoQuant analyst’s exchange data, cited by crypto.news, showed Binance Bitcoin open interest falling from about $5.4 billion to $4.9 billion between September 21 and 23. Bitcoin had declined by then. Falling open interest shows positions closed, but it does not, on its own, separate shorts closed during the rally from longs unwound during the retreat. The $500 million change is a change in the dollar value of outstanding positions, not $500 million of confirmed short buying.

No comprehensive public tape identifies the beneficial owner behind each ETF order and each derivatives close. A forced short buy and a patient fund subscription may both lift demand, but they imply different things about what happens when price stops climbing.

The strongest case for lasting demand has a limit

There is a serious argument that the inflows represent more than a squeeze. Five consecutive positive fund sessions, worth $2.65 billion, spanned the climb and continued on September 23 after the pullback. The purchases were not confined to one fund. IBIT, FBTC and ARKB all drew substantial cash on September 21. Corporate filings show at least two companies buying during the surrounding week. Those are observable commitments, regardless of the Fed’s posture.

The opposing interpretation has a different strength. Bitcoin’s move to about $87,300 did not hold; price was back near $84,000 by September 23. The ETF print records subscriptions during a day, while the price reflects all orders at their execution times. If new fund cash keeps arriving as Bitcoin falls, it means sellers are meeting it. That does not make the fund demand imaginary. It means its price impact cannot be read directly off its dollar total.

Recent crypto.news coverage of the ETF streak reported the pullback alongside continuing inflows and falling Binance open interest. Those facts support a mixed account: underlying fund demand was present, while leverage amplified and then retreated from the move. The public numbers do not tell us the exact share attributable to each.

For a cleaner test, compare the next complete fund reports with the path of open interest after the rally. If subscriptions remain large while open interest stops contracting, the evidence for demand beyond traders closing shorts strengthens. If subscriptions dry up and the price keeps slipping, Monday’s surge looks more dependent on temporary buying. Neither pattern proves causation on its own, because the public series are aggregated across markets and reported at different frequencies.

There is a second timing problem. Bitcoin trades around the clock; US listed fund shares trade during US market hours. The biggest crypto move can happen before an ETF session opens or after it closes. A daily ETF total cannot be laid over a 24 hour Bitcoin candle as though both cover identical hours. Any precise account of Monday would require intraday spot order flow, fund creation timing and derivatives transactions on a common clock. The available public daily series fall short of that standard.

The Fed did not become a buyer’s signal

The Fed’s September statement said inflation remained elevated and economic activity was expanding at a solid pace. Its 2026 and 2027 median rate projections of 4.1% are each higher than in June. The September projection table is a set of participants’ assessments of appropriate policy, not a binding schedule of decisions, but it offers little support for a claim that a near term rate cut drove the immediate rally.

A risk asset can rise after a hawkish decision if the surprise was already priced, other yields fall, or buyers in its own market outweigh macro pressure. Those are possible mechanisms, not proof that one explains this week. The observed fund reversal begins September 17, one day after the announcement. By September 21 it was large enough to see without a macro theory: $999 million in reported net subscriptions.

The more revealing development came when price and subscriptions parted ways on September 23. That is where the original question becomes testable. If fund creations keep coming while short positioning stays less crowded and Bitcoin holds its gains, the evidence for sustained cash demand improves. If creations turn negative and the price loses the levels recovered after the decision, the short squeeze explanation gains weight. Neither result can assign every past trade to an individual investor.

What the public record can actually name

The records identify funds, companies and position types. They do not name the ultimate owner of IBIT shares purchased on September 21. They do not date Strategy’s 950 BTC to Monday. They do not show that every liquidation produced a spot Bitcoin purchase on an exchange.

What they do show is enough to reject two simple versions of the story. Bitcoin did not rally because investors immediately celebrated the September 16 hike: ETF funds lost a combined $746.3 million on September 15 and 16. Nor was the rebound just a chart artifact with no recorded cash demand: about $2.65 billion entered spot funds over the next five sessions, on the figures available September 24.

A public filing adds 950 BTC of Strategy purchases during September 14 to 20, while Strive reports another 1,355 BTC during September 14 to 18. Short covering plausibly accelerated the price move, but no audited decomposition of its contribution has been published. The answer is a set of buyers, operating on different clocks.

What to watch

Daily ETF creations: Check the complete issuer table after every fund has reported; a blank cell is not a zero.

Fund concentration: A positive total spread across IBIT, FBTC and ARKB differs from one driven by a single product.

Open interest with price: Rising price and falling open interest can fit short covering; falling price and falling open interest can reflect long unwinds.

Corporate filings: Read transaction windows in each 8-K before assigning a purchase to a specific trading day.

Price against flows: Compare Bitcoin’s daily close with that day’s fund subscriptions. The September 23 divergence deserves more attention than an inflow headline alone.

FAQ

Disclaimer: This article is for information and educational purposes only and does not constitute financial or investment advice. Figures reflect regulatory filings and reporting available at the time of writing and change with each disclosure. Nothing here is a recommendation to buy, sell, or hold any security or asset. Always do your own research. Information is accurate as of September 24, 2026.