Why Markets Move Before the Fed
Policy transmits when it is priced, not when it is delivered. By the time a cut or hike arrives, the economy has usually been living with it for months — which changes what a meeting is actually for.
Every cycle, a version of the same complaint circulates: the Fed has not even moved yet, and mortgage rates have already jumped — or fallen — by more than any single decision could deliver. The complaint mistakes the system for being broken when it is working exactly as designed. Markets moving before the Fed is not front-running in the pejorative sense, and it is not markets being jumpy. It is the transmission mechanism itself. An investor who waits for the announcement to reposition is not being prudent; they are consuming the news after the economy has already digested it.
The framing we use: policy transmits when it is priced, not when it is delivered. No household borrows at the federal funds rate and no business invests at it. They borrow at term rates, and term rates are built from the expected path of policy — so the moment expectations change, every rate that matters changes with them, months or quarters ahead of any delivered move. The delivered decision is the receipt, not the purchase. Once that inversion of the usual mental model is installed, three questions answer themselves: why decision days are so often anticlimactic, why some are violent anyway, and why the calendar of data releases matters more than the calendar of meetings.
The expected path is the policy
Walk the chain concretely. A lender pricing a long mortgage today must estimate what short rates will average over years — so the mortgage rate embeds the whole expected path of policy, updated continuously. A company deciding on a bond issue faces coupons priced off the same expectations. An equity investor's discount rate keys off long yields built from them. Now let a strong run of inflation data arrive: the market revises the expected path upward, term yields rise, mortgage rates follow within days, issuance windows narrow, multiples compress. Financial conditions have tightened materially — and the Fed has not lifted a finger. When the hike eventually arrives, it changes an overnight rate that touches almost no one directly, while ratifying a tightening the economy has been living with since the data landed. The tightening happened when it was priced. The meeting published it.
Three layers of why
That final gap deserves its own sentence, because it is where this framework stops being descriptive and starts being useful. At any moment there is a priced path — observable directly in rate futures — and a distribution of paths the data could plausibly support. The investable object is the distance between them. Sterling's Embedded Intelligence maintains exactly this comparison as a running discipline: not predicting what the Fed will do, but measuring what the market has already paid for, and flagging when the priced path has drifted far from what incoming data can support in either direction. Sterling's operating rule compresses to one line — the meeting is not the event; the repricing is the event — and repricings are scheduled by the data calendar, not the policy calendar.
The illustration: easing that arrives before the easing
Consider the anatomy of a typical modern easing cycle, drawn as a composite from the historical pattern rather than any single episode. The sequence begins with data: inflation decelerating, labor markets loosening. The market revises the expected path down — first pricing one cut, then several. Term yields fall. Mortgage rates fall with them. Corporate issuance reopens at better coupons. Equity multiples expand as the discount rate eases. All of this occurs while the policy rate has not moved — the economy is receiving genuine, measurable easing, delivered entirely by the repricing of the path. By the time the first cut is announced, a large share of the cycle's total easing effect on financial conditions has often already been delivered, and the announcement itself passes with modest market response — the receipt, not the purchase.
This is also why the folk strategy of buying the first cut has such an uneven historical record, and why its failures are so instructive. The investor executing it is buying something the market sold them months earlier — at the announcement, the relief is largely in the price, and what remains is exposure to whether the priced path was right. When cuts arrive as normalization into a stable economy, the priced path tends to hold and the strategy feels vindicated. When cuts arrive because conditions are deteriorating faster than the path assumed, the subsequent repricing goes the other way — more cuts priced, and asset prices falling anyway, because what forced the cuts was worse than what was paid for. In both branches, the decisive variable was never the announcement. It was the accuracy of the path that had been priced long before it.
The strongest case against this framework
The serious objection: if markets move before the Fed, they move wrongly often enough that the anticipation is noise. The record contains entire priced cutting cycles that never arrived, terminal rates repriced repeatedly, and easing priced on data that promptly reversed. On this view, the priced path is a fickle poll, the decision is the only hard fact, and a disciplined investor should weight delivered policy over market chatter about future policy — the announcement, at least, is true.
We accept the premise — priced paths are frequently wrong — and reject the conclusion, because it confuses two different claims. This framework does not claim the market predicts the Fed correctly; it claims financial conditions follow the priced path, right or wrong. A mispriced path still sets mortgage rates, still prices bond issues, still discounts equities — the economy borrows at the market's error. The delivered decision is a hard fact about one overnight rate; the priced path is the operative fact about every rate that matters. Waiting for delivered policy does not protect an investor from mispricing — it merely ensures they respond to conditions after those conditions have already acted on the economy and their portfolio. The honest refinement is about humility, and it sharpens the framework rather than blunting it: treat the priced path as the operative policy variable, and treat its fallibility as the opportunity set. The question is never whether the market has moved before the Fed — it always has — but whether it has moved to a place the data can defend.
What a meeting is actually for
None of this makes meetings irrelevant — it makes them specific. A meeting is the highest-bandwidth scheduled disclosure of the reaction function: projections, dissents, the press conference's tone under questioning. The step announced is usually the least informative item on the page. The practical reading order on a decision day inverts the popular one: skip past the decision, and ask what changed in the mapping — does the committee now require more evidence to cut, or less? Has the described destination moved? Did the tone toward inflation risk shift? Those revisions propagate into every future meeting simultaneously, which is why they, and not the 25 basis points everyone spent weeks debating, are what move markets on the days meetings actually matter.
How to apply this framework
- Track the priced path as the live policy variable. Futures-implied policy expectations are observable continuously and are what term rates, mortgage rates, and discount rates are actually built from. The delivered rate tells you where policy was; the priced path tells you where conditions already are.
- Treat data releases as the primary policy events. The releases that move the expected path — inflation, employment — are where the discount rate is made. The meeting calendar mostly schedules confirmations; the data calendar schedules the repricings.
- On decision days, read for the reaction function, not the step. The step was priced. What can still surprise is the mapping from data to policy — the evidence threshold, the destination, the tolerance for inflation or unemployment. When markets move violently on a meeting, that mapping is what moved.
- Measure how much is already delivered before extrapolating any move. Before treating a first cut or hike as the start of anything, check how far conditions — term yields, mortgage rates, multiples — have already traveled on the priced path. The gap between what is priced and what the data can defend, not the announcement, is where the durable positioning information lives.