MovePithWHY PRICES MOVE

What a bitcoin ETF is, and how it actually affects the price

Spot ETFs are one of the most significant structural changes crypto has seen in years, and the source of its most consistently misread data. This article covers the mechanism: how shares come into existence, when spot is actually purchased, and what “net inflows of $X million” does and doesn't tell you.

L4 EventsLu ZhiyuanUpdated 1287 words / 6 min read

What it is, precisely

An exchange-traded fund is a fund whose shares trade on a conventional stock exchange. A spot bitcoin ETF has a simple design goal: let an investor get exposure to bitcoin's price by buying a ticker inside the brokerage account they already have, without opening a crypto exchange account, managing private keys, or dealing with custody and withdrawals.

Two facts on the timeline: US spot bitcoin ETFs were approved and began trading in January 2024; spot ether ETFs began trading in July 2024. Before that the US market had futures-based products and trust structures, which work differently.

For our purposes, ETFs belong to layer 4 in the four-layer framework: they change the channel through which money reaches the market and who the participants are, not today's balance of buyers and sellers.

Creation and redemption: when spot is actually bought

This is the part most coverage skips, and the part you need.

ETF shares aren't conjured from nothing. They're created and redeemed in the primary market by authorised participants — typically large financial institutions:

  1. Demand appears

    Buying in the secondary market pushes the ETF's market price above its net asset value.

  2. An authorised participant creates new shares

    To capture that gap, the participant creates a basket of new shares with the fund, delivering the corresponding assets or cash under the fund's rules. This is the step where spot actually gets bought.

  3. The new shares reach the secondary market

    They're sold on the exchange, the premium narrows, and the ETF price returns toward net asset value.

  4. Redemption is the same process in reverse

    With persistent selling and the market price below net asset value, participants redeem shares and the corresponding spot is sold.

So there is a layer between “somebody bought the ETF” and “somebody bought bitcoin”: when buyers and sellers trade with each other in the secondary market, the fund's holdings don't change at all. Only creation and redemption moves the underlying. It's also why daily ETF trading volume can vastly exceed that day's creations and redemptions.

Three traps in the “net flows” number

Trap 1: it isn't “new money”

Net flow measures creations minus redemptions. But the money behind a creation might be genuinely new capital, existing crypto exposure relocating (someone sells their own spot and holds through the ETF instead), or an institution rebalancing internally. Only the first is new money, and the three are indistinguishable in the data.

Trap 2: methodology and settlement lag

Different sources count different things: some by creation date, some by settlement date; some cover only certain products, some include the continuing redemptions of trust-structure vehicles. The same day's “net flow” can differ substantially between two websites. That's not fabrication, it's methodology. When you see one of these numbers, ask what's measured, over what window, from which source — the three questions from why “sentiment improved” explains nothing.

Trap 3: the causal arrow often points the other way

Coverage says “ETF inflows drove the price up”. But creation is frequently a response to a price rise — secondary-market buying pushes the share price above net asset value, and that gap is what triggers the participant to create. Price and flows are often two measurements of the same event rather than cause and effect. Reading co-movement as causation is the most widespread error in this category of reporting.

A check you can run yourself

When you see “large ETF inflows/outflows” tied to a day's price action, verify two things: which day the flow data refers to (it's often disclosed with a day's lag), and when during the day the price actually moved. If the bulk of the move happened before the flow data could have been known, the causal chain fails on chronology alone.

What ETFs genuinely changed

Set aside the daily noise; three structural effects matter far more than any single day's flow figure:

  • Who participates changed. Capital that previously couldn't hold crypto directly because of custody or compliance constraints — some institutions, retirement accounts, adviser-managed portfolios — gained a compliant route. That's a slow but persistent shift.
  • Concentration and custody changed. A meaningful quantity of spot is now held by a small number of custodians. That reduces individual custody risk and introduces a new concentration question — a neutral fact, not a verdict.
  • Part of price discovery moved. Conventional trading hours carry more weight than they used to, which increases linkage with traditional markets — and may make macro transmission more direct.

Why approval day often looks “wrong”

ETF approval is a long-anticipated event: the filings, the delays and the deadlines are all public, and the market prices the probability repeatedly before the outcome. So the common pattern is that the price reflected the expectation over the preceding weeks, and the approval itself becomes the moment some participants take profits.

That isn't the folk wisdom about “buy the rumour, sell the news” being mystically true — it's the ordinary consequence of expectation pricing. Full mechanism in why the price already moved before you saw the news.

Premium and discount: a reading you can check yourself

The gap between an ETF's market price and its net asset value — the premium or discount — is the most direct window into this machinery, and it's public.

In normal conditions the gap is small, because as soon as it widens enough to cover costs, authorised participants create or redeem and arbitrage it away. That is the mechanism working. Conversely, when the gap stays wide, it usually indicates friction in the creation and redemption channel itself: strained liquidity in the underlying, or an operational constraint somewhere in the chain.

That gives you a concrete cross-check. When you see a report of “massive inflows or outflows”, look at the premium or discount at the same time. If the premium never widened while the report claims enormous creations, there's a contradiction to explain — most likely secondary-market volume being mistaken for creations and redemptions.

Common questions

What's the difference between buying the ETF and buying coins?

Mechanically: the ETF trades during the stock exchange's hours (spot crypto trades continuously), usually charges a management fee, cannot be withdrawn to your own wallet, and the underlying is held by a custodian. Which is preferable depends on your account structure, tax position and custody preferences. This site makes no recommendation and does not assess suitability.

Do consecutive days of net inflows mean the price will rise?

We don't make that inference. Mechanically, flows and price influence each other within the same window and the direction is hard to separate; and the net-flow measure isn't equivalent to new capital in the first place. Using it as a standalone signal has no foundation.

Can I see the ETF's bitcoin on-chain?

Some issuers disclose custody addresses or publish holdings periodically, and verifiability varies by product. But an address shows you balance changes, not intent — translating address activity directly into market behaviour is a common over-read; see are whales really dumping.