What the nickname refers to

The Federal Open Market Committee (FOMC) publishes the Summary of Economic Projections (SEP) alongside several of its meetings each year. Reporters call it the dot plot because one page of the document is a chart: each dot marks where one member of the Committee expects the federal funds rate to sit at the end of a given year. That chart draws the headlines, but it sits beside projections for growth, unemployment and inflation, and a written account of how the Committee sees the risks around them.

Why it stops short of a forecast

Each dot is one policymaker's own projection, made under their own view of how the economy will move and what policy would be appropriate in response. There is no vote on the dots, and no single number the Committee agrees to stand behind in public. Reading the chart as a promise about where rates are headed asks it to do a job it was never built for. Reading it as a record of what each member currently believes, open to revision at the next projection, matches what the document actually is.

The glossary entry for the Summary of Economic Projections sets out where the document sits among the FOMC's other releases, and The Forecast Scorecard follows one year's projection across each time it was revised, set against what happened.

Each committee member submits a number for where they expect the federal funds rate to sit at the end of a given year, and once those numbers are collected, three different summaries can be made of them. They tell you different things, and mixing them up is the easiest way to misread the Summary of Economic Projections (SEP), the document these figures come from.

The median

The median is the middle value once every participant's projection for that year is lined up from lowest to highest. If nineteen members submit a figure, the median is the tenth. It is not an average and it is not the committee's collective view in any formal sense: it is simply the projection that sits in the middle of the pack, and it moves only when enough individual projections move past it.

The central tendency

The central tendency is a narrower range built by dropping the three highest and three lowest projections and reporting the span of what remains. It is meant to show where most of the committee sits, once the most extreme views on either side are set aside. A central tendency can narrow or widen from one publication to the next even if the median does not move at all, because the change is happening at the edges of the group.

The range

The range is the full spread, from the single lowest projection submitted to the single highest, with nothing excluded. It is the widest of the three figures by construction, and it is the one most sensitive to a single member taking an unusual view. A wide range does not mean the committee is divided down the middle; it can mean eighteen members agree closely and one does not.

Why the median can move against the other two

Because the median depends only on the middle value, it can hold steady while the central tendency and the range both shift, if the movement happens among members above or below the midpoint. Equally, the median can move even when the range stays the same width, if projections shuffle position without any new extreme being added. Reading only the median tells you where the middle sits; reading it alongside the central tendency and the range tells you whether that middle is representative of the group or is being held in place while opinion moves around it.

The mechanics of where these figures are published and how to track one across successive editions are covered on The Forecast Scorecard, which sets a chosen year's median against later revisions and against what actually happened.

Why a chart and not a table

Each participant on the Federal Open Market Committee submits a projection for growth, unemployment, inflation, and the appropriate path for the policy rate, but the Summary of Economic Projections (SEP) never lists whose figure belongs to whom. Publishing them as a table, with a row for each seat, would let a reader match a name to a number and treat that individual's view as a standing position. Instead, every projection appears as an unlabelled dot on a chart, one column of dots for each year, so the reader can see how widely the group's individual assessments spread without being able to trace any single dot back to one policymaker.

That anonymity is part of why the document is read as a distribution. A cluster of dots close together says the participants largely agree on where the rate should be by a given year; a wide scatter says they do not. Either way, the chart shows disagreement or agreement among individuals.

What the FOMC says the projections are not

The material published alongside the SEP states plainly that these are not a committee forecast and not a commitment to any path for the federal funds target range. Each projection is one participant's own assessment of appropriate monetary policy, made on that person's individual reading of how growth, employment and inflation are likely to unfold. Nobody's name is attached to a dot, and no vote binds the Committee to the rate any dot shows.

Reading the chart as a promise is the mistake the FOMC's own language is written to head off. The guide to reading a central bank's own words covers how to spot the language that separates a projection from a commitment across the documents a central bank publishes.

The Federal Reserve Board publishes the Summary of Economic Projections (SEP) four times a year, alongside the March, June, September and December meetings of the Federal Open Market Committee. Each edition sits in the materials for that meeting on the Federal Reserve Board's own site, released at the same time as the post-meeting statement, so the projection tables can be matched to the exact date they were made.

Reading the footnotes

The footnotes beneath the projection tables are not incidental. They record which participants' figures were included in a given calculation, since a member who has left or joined the Committee between meetings changes who is being averaged, and they note when a projection assumes a particular path for policy. A number read without its footnote can look like a shift in the Committee's thinking when it is really a shift in who was counted.

Tracking one year across publications

Because the SEP is published four times a year, a single target year, say the projection for a given calendar year's inflation or growth, appears in several successive editions before that year arrives. Lining up the March, June, September and December figures for the same year shows whether the median moved and by how much, and whether it moved steadily or jumped between two adjacent meetings. Reading them in isolation loses that movement; reading them in sequence is what shows whether the Committee's view was firming up or changing direction.

The Forecast Scorecard does this lining-up for a chosen year, plotting the FOMC's median projection as it stood at each publication date against what the data later showed. Run the same exercise on the raw tables and the figures should match, since both are drawn from the Federal Reserve Board's own published record.