Overview

The New York Fed’s horizon-varying VECM models how the FOMC’s SEP projection and an OIS market forecast respond to their gap for the same future date. In an illustrative eight-quarter-ahead shock, a 25-basis-point SEP surprise becomes 5 basis points of initial market repricing and about 18 basis points in the model’s median eventual adjustment. Both series can revise; convergence is not a trading rule or calendar forecast.

Learning outcomes

The forecast gap can close through two-sided adjustment: the market may move toward the SEP while the SEP moves toward the market.
In the eight-quarter illustration, 25 basis points becomes 5 initially and roughly 18 in the modeled median response; these are different stages.
Convergence is estimated on average, while low market-rate explanatory power and broad uncertainty limit prediction.

Keywords

FOMC dot plotSEPVECMConvergenceForecast gapOIS

How the Market and the Fed’s Projection Catch Up

The initial five-basis-point market response does not tell the whole story of a 25-basis-point dot-plot surprise. The New York Fed study follows the SEP projection and the market forecast for the same future date across subsequent quarters, asking whether the two estimates remain apart or adjust toward one another. This time dimension supports a more precise judgment: the market can discount the dots at first while still incorporating their information later, and the SEP projection can also move as policymakers update their own assessment.

A forecast gap creates a two-sided adjustment

Imagine fixing one future date and recording two estimates for the federal funds rate at that date. One estimate is the SEP projection; the other is the market forecast, captured by an OIS forward rate. At a particular release, the SEP may sit above the market, or the market may sit above the SEP. The difference is a forecast gap. It is not automatically an error by one side. It is a disagreement about what rate will be appropriate or realized at a future point, given different information and judgments about how conditions will evolve.

The New York Fed study examines what happens to that gap over time with a horizon-varying vector error-correction model, or VECM. The name is more intimidating than the central idea. The model allows changes in the SEP projection and changes in the market forecast to depend partly on their own recent movements and partly on the distance between them. The “error correction” is the tendency for a gap to generate subsequent adjustment rather than persist indefinitely as the target date approaches.

This is a two-sided process. When the SEP projection is above the market forecast, the market tends to move upward toward the SEP, while the SEP tends to move downward toward the market. When the market is above the SEP, the directions reverse. In the reported model, the error-correction coefficients have the signs that correspond to this behavior: the SEP’s coefficient is negative and the market’s is positive when the gap is defined as the SEP minus the market. The point is not that the market always capitulates or that policymakers always follow investors. Both estimates can change.

stateDiagram-v2 [*] --> ForecastGap ForecastGap --> MarketAdjustment: SEP above market ForecastGap --> SEPAdjustment: market above SEP MarketAdjustment --> MutualRevision SEPAdjustment --> MutualRevision MutualRevision --> NearerDate NearerDate --> Convergence: shorter horizons adjust faster Convergence --> UncertainOutcome: realized rate remains uncertain UncertainOutcome --> [*]

The lifecycle has no single owner. The market forecast changes as investors receive new data, reassess risks, and learn more about policy. The SEP projection changes as FOMC participants reassess the outlook and the appropriate policy response. The gap is therefore not a queue in which the market waits for permission from the dots.

Published by DeepMox

FAQ

Why can the SEP projection and market forecast both move toward each other?

Because the VECM lets each series respond to the gap between them as well as to its own recent movements; neither side must capitulate.

What does the eight-quarter FOMC dot plot illustration show?

It shows a 25-basis-point SEP surprise, a 5-basis-point initial market response, and a modeled median eventual adjustment of about 18 basis points. These are stages, not competing forecasts.

Does convergence make the FOMC dot plot a reliable rate forecast?

No. Convergence describes an average tendency for the SEP and market series to reduce their gap, while low explanatory power and broad uncertainty limit prediction.

Previous9 minFOMC Dot Plot: Signal Without Command

Next10 minFOMC Dot Plot: Forecast, Reaction Function, or Promise?

FOMC Dot Plot: How Forecasts Converge

How the Market and the Fed’s Projection Catch Up

A forecast gap creates a two-sided adjustment

FAQ