What the Fed’s Dots Actually Represent
The FOMC dot plot is easier to misread when its basic objects are left implicit. This chapter identifies what a dot represents, how its horizon and median are constructed, and how researchers turn the gap between a released projection and prerelease expectations into a measurable surprise. The payoff is practical: once the object and comparator are clear, a market reaction can be interpreted as pass-through from new information rather than as a verdict on whether the Federal Reserve kept a promise.
The dots are projections for a date, not a countdown
The federal funds rate is the FOMC’s principal policy instrument. It is the overnight rate at which depository institutions lend reserve balances to one another. Although the transaction is short-term, changes in the rate affect a wider network of financial conditions: other short- and medium-term interest rates, the foreign-exchange value of the dollar, asset prices, and the spending and investment choices of households and firms.
The dot plot belongs to the Summary of Economic Projections, or SEP. At quarterly intervals, FOMC participants are asked to report projections for key macroeconomic variables and for the federal funds rate at several future dates, assuming what they consider appropriate monetary policy. The published plot shows the individual rate projections for the end of the current year and for future calendar-year horizons. In March and June, the SEP includes three rate horizons; in September and December, it includes four.
Those labels describe calendar endpoints, not identical forecasting distances. A projection for the end of the current year is closer to the date of publication than a projection for the following year, and the time remaining changes from one quarterly release to the next. That matters when a reader compares a dot with a market rate: the comparison must refer to the same future date. Otherwise, a difference may reflect the fact that one number concerns a nearer point on the policy path rather than a disagreement about the same point.
The horizon also changes how much uncertainty is naturally embedded in the projection. A near-term rate assessment is made after fewer unknowns have time to intervene; a projection several years ahead depends on more assumptions about inflation, output, financial conditions, and the policy response. This does not make the distant dot useless. It does mean that its meaning should be read as a conditional view over a longer chain of possible developments, not as a more distant appointment on the FOMC calendar.
A dot is therefore a participant’s assessment for a particular date under a particular outlook. It is not a countdown to a scheduled move at the next meeting. The distinction is easy to lose because a chart turns a set of conditional judgments into points on a clean vertical scale. But the underlying question is not “what rate has the Committee already decided to deliver?” It is closer to “given the information and assumptions available now, what policy rate would be appropriate at that future point?”
The displayed median is another source of confusion. The median is the middle value in the distribution of participants’ projections. It is useful as a compact summary, but it does not turn the projections into a binding decision or a single guaranteed institutional forecast. A median can move even when several participants remain on the same side of it, and it can stay unchanged while the distribution around it changes. The plotted middle is informative, but it cannot carry every assumption, disagreement, or contingency that produced the individual points.
For that reason, the median should be treated as a summary statistic with a specific job, not as a complete account of the Committee’s thinking. It provides a reference for the central tendency of the projections. It does not tell a reader how likely each alternative is, whether the participants share the same reasoning, or which developments would move the middle of the distribution. The surrounding economic projections and public explanations are needed to supply that context.
That compression is not a flaw in the chart. Public communication needs a form that can be read quickly, and the median provides a common reference point. The problem begins when visual precision is mistaken for certainty about the future. A dot at a particular rate and horizon does not, by itself, state the probability of that outcome, the full range of alternative outcomes, or the exact conditions that would make participants revise it. Those questions belong to the broader economic outlook and to the FOMC’s reaction function—the way policy responds to changing conditions.
The distinction becomes clearer if the three relevant rates are placed side by side. First is the SEP projection: a conditional assessment of the future federal funds rate. Second is the realized federal funds rate: the rate the Committee actually sets as future meetings arrive. Third is the market expectation: the rate path investors infer from the SEP, economic data, statements, speeches, and other information. The three can point in the same direction while remaining different objects.
This separation prevents a common category error. A market can agree that the median projection is the FOMC participants’ central assessment and still price a different future rate. That difference may reflect a different view of the economic outlook, a different estimate of how policy will respond to new data, or uncertainty about how much weight the Committee will place on the projection. Conversely, a market rate can move closer to the median without implying that investors believe the FOMC is bound to deliver it. The market is updating its distribution of possible outcomes, not signing onto an institutional contract.
The distinction is especially important when a path changes at several horizons. A lower median for the end of the current year might coexist with little change farther out, suggesting a different timing of adjustment rather than a permanent change in the longer-run stance. A movement at a distant horizon may carry a different degree of uncertainty from a movement at a near horizon. Reading the dots responsibly therefore requires asking which date changed, what was already expected at that date, and which assumptions make the projection coherent.
The realized rate is the outcome that arrives after the economy has supplied more information. Inflation may be higher or lower than expected; demand may weaken; financial conditions may change; or the transmission of a given policy rate may be stronger or weaker than policymakers thought. If those conditions change, a later decision can differ from an earlier projection without turning the earlier projection into a broken contract. A forecast is conditional precisely because its relevance depends on the conditions that support it.
The market expectation is also not a neutral copy of the median. Investors must decide how much weight to place on the projection and how much uncertainty to attach to it. They may expect the FOMC to respond differently from what the median currently implies, or they may think the economic outlook will change before the projected date. Market prices can therefore sit above or below the dots while still reflecting careful attention to them.
The most important comparison is not always between the new dot plot and the previous dot plot. It is between the new median and what participants in the market expected immediately before the release. A projection can be high in absolute terms but unsurprising if investors had already anticipated it. Conversely, a modest change from the previous SEP can be a large surprise if market expectations moved in the opposite direction during the intervening weeks.
How a projection becomes a measurable surprise
The New York Fed study of market reactions formalizes this comparison. For each projection horizon, the authors subtract a prerelease survey expectation for the median SEP federal funds rate from the median that the FOMC actually publishes. The result is a surprise at a matched horizon. It isolates the part of the released projection that was not already represented in the selected market expectation.
The study uses two surveys for that prerelease benchmark. The New York Fed’s Survey of Market Expectations has asked market participants about median SEP federal funds rate projections since 2015. A Bloomberg survey has asked similar questions since 2017. These surveys are not measures of every person’s belief about interest rates. They are structured observations of what selected professional respondents expected before the release. Their value for this research question is that they establish a comparator before the information arrives.
A simple hypothetical makes the arithmetic clear. If respondents expected the median projection for a future date to be 3.75 percent and the released median was 4.00 percent, the surprise would be positive 25 basis points. The calculation does not say that the future policy rate will be 4.00 percent. It says that the published median was 25 basis points above the prerelease survey benchmark. The next question is how the market price for that same future date changes when it receives the information.
To observe market expectations, the study uses overnight index swap, or OIS, forward rates at daily and quarterly frequencies. An OIS forward rate is a market-based price for an interest rate over a future period, so it provides a way to track the path investors are pricing rather than only the current overnight rate. The authors match the market horizon to the SEP horizon as closely as possible. That matching matters: comparing a projection for the end of a calendar year with a market rate for a materially different date would mix a policy surprise with a maturity mismatch.
An OIS forward is best understood here as a proxy for a market-implied expectation, not as a direct survey response. Its price is formed by participants who are also weighing uncertainty and the risks attached to holding or financing positions. The study uses the movement in that price to measure how market expectations adjust around the release; it does not claim that the forward rate is a pure, error-free reading of what every investor believes. The instrument is valuable because it is continuously priced and can be aligned with the future date represented by the projection.
For narrower event windows, the study uses additional interest-rate derivatives. Futures cover the shortest horizon, while Eurodollar futures are used before 2021 and Secured Overnight Financing Rate futures during 2021–23 for other horizons. Beginning in September 2023, intraday OIS forward rates are used at all available horizons because they can match the SEP dates more exactly. These details serve one purpose: to measure the market’s response at the date and horizon represented by the dot, not to create an inventory of financial products.
The timing of the observation is as important as the instrument. Daily data measure the change around the release day, but that change can include the press conference and other news arriving during the day. Intraday data narrow the window to 30 minutes around the SEP release, which makes it less likely that unrelated developments drive the measured movement. Even that narrow window is not a pure dot-plot experiment: the FOMC statement is released at the same time as the SEP, so the market reaction can reflect both pieces of communication.
The study also controls for the movement in market expectations between the date when the New York Fed survey was collected and the day before the SEP release. That interval matters because the survey is not taken at the last possible second. Economic data, speeches, or other market developments may cause expectations to move after respondents answer. Without the control, a measured difference could partly reflect information that arrived before the release rather than the surprise in the SEP itself.
The same logic explains why the study treats the survey and the market instrument as complementary rather than interchangeable. The survey supplies an explicit pre-release expectation for the median. The OIS or futures price supplies a continuous market measure that can move as new information arrives and can be observed immediately around publication. Comparing them makes the unexpected part of the release visible, while the timing control helps keep earlier information from being mislabeled as new information.
This is the basic sequence:
The sequence separates three events that are often collapsed into one. Participants form conditional projections. Survey respondents report what they expect those projections to be. The public release then reveals the median, alongside the FOMC statement, and market instruments record the adjustment. A “surprise” belongs to the comparison between the second and third events; a market reaction belongs to what happens after the third.
This design also shows why the study’s result should not be generalized carelessly. The survey benchmark represents professional respondents, not the entire public. OIS forwards and futures are market prices, not direct questionnaires to households and firms. The event window captures a bundle of contemporaneous communication, especially in the intraday specification. And intraday observations are unavailable for projection horizons beyond three years. These are not technical footnotes detached from the conclusion. They define what the evidence can identify.
They also clarify why matched-horizon measurement is more informative than simply asking whether the latest dot is higher or lower than before. The market is constantly updating. A release matters to the extent that it changes what investors thought was likely at a particular future date. By comparing the released median with a prerelease benchmark and then tracking the corresponding market price, the study can ask how much of the new information passes through immediately.
That question is deliberately narrower than “is the dot plot accurate?” Accuracy would require comparing a projection with a later outcome, and even then the result would depend on how the intervening state of the economy changed. The event study instead asks whether markets react systematically to unexpected information in the projection. The answer to that question comes next, but the measurement logic already sets the interpretive boundary: the dots are conditional projections, and their surprises are defined relative to expectations that existed before publication.