How to Read the FOMC Minutes: A Bitcoin Investor’s Field Guide

Three weeks after every Federal Open Market Committee meeting, at 2:00 p.m. Eastern, the Federal Reserve publishes the minutes of that meeting. The next release is Wednesday, August 19, 2026, covering the July 28–29 meeting at which the Committee held its target range at 3.50–3.75% on a 9–3 vote, with Beth Hammack, Neel Kashkari and Lorie Logan dissenting in favour of a quarter-point increase.

Most retail investors skip the minutes, on the reasonable-sounding grounds that everything in them is three weeks old. That instinct is wrong, and understanding why is the whole point of this guide. The statement is a negotiated committee document written to say as little as possible. The press conference is one person’s framing, delivered under adversarial questioning. The minutes are the only Fed publication that tells you the shape of the disagreement — how many people held each view, how strongly, and what evidence would move them. Markets that cannot price the next meeting are not short of numbers. They are short of information about the distribution of opinion inside the room, and that is exactly what this document contains.

This guide sits alongside our companion pieces on reading the FOMC statement and press conference and the decision-day playbook. Those cover the day itself. This one covers the document that arrives three weeks later and, more often than people expect, moves the market harder.

Trap zero: participants are not members

Start here, because almost every bad take on a minutes release comes from missing this distinction. The FOMC has nineteen participants — the seven Board governors plus all twelve Reserve Bank presidents — but only twelve voting members at any time: the governors, the New York Fed president, and four other presidents on an annual rotation.

The minutes use both words, deliberately and consistently. When a paragraph says “participants”, it is describing the views of all nineteen people in the room, voters and non-voters alike. When it says “members”, it is describing the twelve who actually cast the vote, and you will almost always find that word in the section describing the policy action itself.

Why this matters more than anything else in the document: the recorded vote is a lower bound on dissent, not a measure of it. A July meeting that recorded three dissenting members may reveal, in the participants’ section, that six or seven participants favoured an increase — the extras being non-voting presidents whose disagreement never appeared in the vote tally. The rotation changes in January. A hawkish bloc that is invisible in 2026’s votes can become the majority in 2027 without a single person changing their mind. That is the single most valuable forward-looking fact the minutes contain, and it is available nowhere else.

The quantifier ladder: the Fed’s hidden vote count

The minutes never say “eleven participants thought X.” They use a vocabulary of quantifiers that has been remarkably stable for decades, and which Fed watchers treat as an approximate scale. It is not officially defined, and the Fed has never published a key, but the conventional reading runs roughly as follows.

These are market conventions, not official Federal Reserve definitions. Use them as an ordering, not as a headcount.

The trap inside the ladder is “several” versus “some”. In everyday English most people read “some” as vaguer and larger. In Fedspeak the order is reversed: several outranks some. A sentence reading “several participants judged that a further increase would likely be appropriate” is materially more hawkish than the same sentence with “some”, and traders reading quickly get this backwards constantly.

The real signal, though, is not the level of the quantifier. It is the change from the previous set of minutes. “A few” becoming “several” on the same proposition, across two consecutive meetings, is a coalition growing. That is a directional fact about the next decision. Keep the previous minutes open in a second tab and diff the paragraphs; this is the highest-yield ten minutes available to a retail reader on release day, and it is why professionals run text-comparison scripts the instant the PDF drops.

The document, in the order it is written

The minutes follow a fixed structure. Knowing it lets you skip straight to what matters instead of reading twelve thousand words front to back.

• 1. Developments in financial markets and open market operations. The desk report. Usually procedural, but this is where balance-sheet mechanics, reserve conditions and money-market plumbing get discussed. Skim it unless there is a funding-stress story running.
  2. Staff review of the economic and financial situation. What the Board staff observed. Backward-looking. Skim.
  3. Staff economic outlook. The staff forecast — institutionally distinct from the participants’ own projections and frequently different from them. Worth reading when the staff mention a recession probability or revise the outlook meaningfully.
  4. Participants’ views on current conditions and the economic outlook.This is the section. All nineteen views, aggregated by quantifier, typically the longest part of the document. If you read only one section, read this one.
  5. Committee policy action. The twelve voting members, the rationale for the decision taken, and the language they agreed for the statement. Watch for discussion of what would justify a different decision next time.
  6. Voting record and dissent statements. Who voted which way and, where provided, the dissenters’ stated reasons in their own framing.

The five things to read first, in market order

• 1. The quantifier attached to the hawkish or dovish proposition. Find the sentence in the participants’ section that describes support for the action the market is not currently pricing, and read its quantifier. On Wednesday that is the sentence describing how many participants favoured an increase in July.
  2. Any conditional sentence containing “would” or “if”. Phrases of the form “participants noted that if inflation were to...” are the Committee telling you its reaction function. These are the most forward-looking sentences in the document and are systematically under-read.
  3. Whether the risks language is symmetric. Look for whether participants describe risks to inflation and to employment as balanced, or whether one side is described as having grown. Asymmetry is a tell about the next move.
  4. The staff outlook revision. If the staff changed their view of growth, unemployment or inflation between meetings, the direction of that revision anchors everything the participants then say.
  5. Anything about the balance sheet. Least glamorous, most under-priced. Runoff pace, reserve-scarcity discussion and repo-facility usage move liquidity conditions, and liquidity conditions move bitcoin with a lag that most crypto commentary ignores entirely.

The transmission chain: minutes to bitcoin

The minutes do not touch bitcoin directly. They move it through a chain, and knowing the chain tells you which move is real and which is noise.

Minutes → rate expectations → short-end yields and the dollar → the price of holding a zero-yield asset → bitcoin. A hawkish surprise raises the implied path of policy, lifts two-year yields, firms the dollar, and raises the opportunity cost of holding an asset that pays no coupon — which is bearish bitcoin at the margin. A dovish surprise runs the chain in reverse. The link from step three to step four is where the transmission is weakest and where the ETF era has genuinely changed things: flows into US spot bitcoin ETFs can overwhelm the rates signal for days at a time, which is why this desk grades macro reactions at the close rather than in the first thirty minutes.

Three practical consequences. First, the 2:00 p.m. candle is not information — it is a headline-scanning algorithm reacting to keyword density before anyone has read the participants’ section. The considered move typically arrives between 2:20 and 3:00 p.m. ET. Second, if bitcoin moves hard and two-year yields and the dollar do not, the chain is broken and the bitcoin move is about something else. Third, always check the reaction in rates markets before concluding the minutes mattered at all.

Four traps

• • The staleness trap. The minutes are three weeks old, and the world moves. Between July 29 and August 19 the US saw a negative payroll print, an in-line CPI, a retail sales miss, a New York manufacturing index at a four-year high and Brent crude back above $90. Anything in the minutes that is conditional on data has already been partly overtaken. Read the minutes for the reaction function, not the conclusion — the conclusion is expired, the reaction function is not.
  • The dissent-is-news trap. The dissents were published in the statement three weeks ago. They are not new information and reporting them as though they are is the most common minutes-day error. What is new is how close the non-dissenters were.
  • The cherry-pick trap. A twelve-thousand-word consensus document contains a sentence supporting almost any thesis. Anyone quoting one sentence without its quantifier is selling you something. Demand the quantifier.
  • The redaction trap. The minutes are edited and approved by the Committee itself before release. They are an accurate account, not a transcript. Full transcripts are published with a five-year lag, and they routinely reveal that the minutes smoothed sharper disagreements. Treat the minutes as the most candid timely document, not as the candid one.

Worked example: what to look for on Wednesday, August 19

The July 28–29 meeting held rates at 3.50–3.75% on a 9–3 vote, the three dissents all favouring a hike. That configuration — a hold delivered over hawkish dissent — makes the participants’ section unusually informative, because the interesting question is not why three people dissented but how many of the other sixteen were close.

• • Read for the hawkish quantifier. If the minutes say “several participants judged that an increase in the target range would have been appropriate at this meeting”, that is meaningfully more hawkish than the 9–3 vote implied and should lift September odds. “A few” would roughly match the three recorded dissents and tell you nothing new. This is exactly the proposition this desk has written down as marker X1.
  • Read for the conditional. Any sentence beginning “participants agreed that further increases would be warranted if...” is the reaction function. Match its condition against what has happened since July 29 — and note that both the energy and goods-inflation impulses have strengthened in the interim.
  • Read the staff outlook on inflation. If the staff marked up their inflation path in July, the August data have since gone the same way, and the September debate starts from a worse place than the July vote suggested.
  • Then check two-year yields and the dollar at 3:00 p.m., before forming any view on what it means for bitcoin.

Context for the stakes: September hike odds are currently scattered across sources rather than converged — one futures-implied read near 65%, another near 32.6%, prediction markets lower still. When the providers disagree by that margin, the market genuinely does not know, and a document that reveals the internal distribution of opinion has unusual power to move it. The next FOMC decision is September 16.

Five rules

• 1. Read the participants’ section first and everything else only if you have time.
  2. Never quote a sentence from the minutes without its quantifier; the quantifier is the content.
  3. Diff against the previous minutes. The change in language is the signal; the level is context.
  4. Ignore the 2:00 p.m. candle. Grade the reaction at the close, and only after checking that rates and the dollar moved too.
  5. Read the minutes for the reaction function, not the conclusion. The conclusion is three weeks stale by construction; the reaction function is what governs the next meeting.

What are the FOMC minutes?

They are the detailed account of a Federal Open Market Committee meeting, published three weeks after the meeting at 2:00 p.m. ET. They describe the staff economic review, the views of all nineteen participants, the policy action taken by the twelve voting members and the voting record. They are an approved account, not a transcript; full transcripts are released with a five-year lag.

What is the difference between participants and members in the FOMC minutes?

Participants means all nineteen people in the room — seven Board governors and twelve Reserve Bank presidents. Members means the twelve who hold a vote at that time. Because non-voting participants’ views appear in the minutes but not in the vote tally, the recorded vote is a lower bound on disagreement, and the participants’ section can reveal a much larger hawkish or dovish bloc than the vote suggested.

What do “a few”, “some” and “several” mean in the Fed minutes?

By market convention, roughly: a couple is two, a few is three to four, some is four to six, several is five to eight, many is eight to twelve, and most is a clear majority. Note that several is conventionally read as larger than some, which is the reverse of ordinary English usage. These are conventions rather than official Federal Reserve definitions.

When are the next FOMC minutes released in 2026?

The minutes of the July 28–29, 2026 meeting are published at 2:00 p.m. ET on Wednesday, August 19, 2026. The next policy decision follows on September 16, 2026.

Do the FOMC minutes move bitcoin?

Indirectly. The chain runs from the minutes to rate expectations, then to short-end yields and the dollar, then to the opportunity cost of holding a zero-yield asset, and only then to bitcoin. If bitcoin moves on a minutes release while two-year yields and the dollar do not, the move is being driven by something other than the Fed — frequently ETF flows, which in the ETF era can overwhelm the rates signal for days at a time.

Disclaimer: This article is for informational purposes only and does not constitute investment advice. Cryptocurrencies and crypto-linked equities are volatile and you can lose money. Do your own research and consult a licensed financial advisor before making investment decisions.