Every quarter the Federal Reserve publishes a Summary of Economic Projections, and every quarter market participants misread it. The SEP is not a policy commitment. It is a snapshot of nineteen individuals' conditional forecasts, aggregated into a median that acquires an authority its authors never intended. With the March projections now in circulation alongside the June update, the more useful editorial question is not what the dots say, but how to read them without being led astray.
The SEP bundles four objects: projections for real GDP growth, the unemployment rate, PCE inflation (the Fed's preferred price index, based on personal consumption expenditures), and the appropriate path of the federal funds rate. The last of these is the dot plot — one anonymous dot per FOMC participant, per year, showing where each thinks policy should sit if the economy evolves as they expect. As the New York Times noted in its explainer, the correct way to read the release is to focus on the median path, the dispersion around it, revisions from the previous SEP, and the underlying assumptions each participant is making.
Why the median misleads
The median dot is the number that moves markets, and it is the number most likely to be wrong. It compresses a distribution of conditional forecasts — each participant's best guess given their own view of productivity, the neutral rate, and the labour market — into a single scalar. When dispersion is wide, the median is a fiction that no single participant actually holds. The more informative signal sits in the tails: how many dots cluster above the median, how far the most hawkish participant sits from the most dovish, and whether that gap is widening or narrowing between meetings. A tight cluster implies committee consensus and a credible path. A bimodal distribution implies an argument the Chair has not yet resolved.
The March and June releases together offer a natural experiment in revision. The Federal Reserve Board's own framing of the June 16–17 release emphasised that policy decisions will remain data-dependent — language that exists precisely to inoculate the Committee against the accusation that the dots are a promise. Between March and June, the meaningful editorial content is the delta: which participants moved, in which direction, and whether the shift was driven by realised inflation, revised estimates of the neutral rate, or a reassessment of labour-market slack. A median that moves fifty basis points with no change in the underlying macro forecasts implies a shift in the Committee's reaction function, which is a far more consequential signal than a shift in the point forecast.
The dot plot is not a forward curve. Treating it as one is the single most common analytical error in Fed-watching.
The dossier here is thin on quantified forecasts — the March release text was not surfaced in usable form, and no external analyst in the cluster has staked a probability on the next move. That absence is itself instructive. The operationalisable claims a serious reader should build from an SEP are narrow and technical: the probability that the year-end median funds rate is revised down at the next meeting; the probability that dispersion widens; the probability that the longer-run dot — the Committee's implicit estimate of the neutral rate — drifts higher. These are the resolvable questions. The seductive one — what will the Fed do next — is precisely the one the SEP is designed not to answer.
Briefings are synthesised by the Ledger Desk from multiple sources cited in the sidebar. They are distinct from Articles, which are written by named contributors and carry a tracked Calibration Index. The Desk does not currently carry a Brier score; this is a deliberate choice for the v0.1 editorial layer and will be revisited.