How central bank narratives shape household inflation expectations
The paper, by Eric Tong and Rennae Cherry, develops a text‑based event‑study framework to identify “communication shocks” as households actually experience central bank messages, and shows that these shocks measurably move one‑year‑ahead inflation expectations in the UK, Canada, and US.
- The authors construct communication shocks by comparing the tone of central bank narratives in policy announcements with the tone of pre‑announcement newspaper coverage over the three days before each meeting, treating the gap as a “narrative surprise” analogous to an asset‑price surprise.
- They convert both central bank texts and newspaper articles into numerical measures of policy tone using a dictionary‑based textual analysis, then decompose narrative surprises into two orthogonal components: a stance communication shock (unexpected messages about the policy stance) and an information communication shock (unexpected messages about the central bank’s assessment of economic conditions).
- The framework is applied to monetary policy announcements of the Bank of Canada, Bank of England, and Federal Reserve from the late 1990s/early 2000s to 2024, estimating effects on households’ one‑year‑ahead inflation expectations using smooth local projections while controlling for policy‑rate changes, shadow rates, and yield‑curve‑based conventional monetary policy shocks.
Findings
- Central bank narratives significantly shape media coverage and, through that channel, affect households’ inflation expectations: tighter stance communication shocks reduce one‑year‑ahead inflation expectations, while more expansionary information shocks increase them.
- These effects are stronger when households’ attention to monetary policy and inflation is high, and forecast‑error variance decompositions show that communication shocks account for a non‑trivial share of the variation in expectations across the three countries.
- By contrast, conventional monetary policy shocks identified from high‑frequency asset‑price moves do not generate comparable responses in household expectations, leading the authors to conclude that the view “central bank communication rarely reaches the public” is too stark and that its effects depend critically on how messages are intermediated by the media and perceived by households.
Central bank communications that reach the public
Authors: Eric Tong, Rennae Cherry
From: Bank of England, Reserve Bank of New Zealand
SEP releases and monetary policy surprises
Andrew B. Martinez and Tara M. Sinclair, in this paper, combine a simple theoretical model of information effects with high‑frequency event‑study evidence to show that the Federal Reserve’s Summary of Economic Projections (SEP) releases are a major source of measured monetary policy surprises and that these surprises carry economically meaningful information.
The authors extend Bauer–Swanson and Jarociński–Karadi frameworks to incorporate the intermittent release of central bank forecasts, modelling how SEP releases both add information and anchor private expectations between meetings. Empirically, they use high‑frequency measures of monetary policy surprises around FOMC announcements since 2013, compare surprises at SEP vs non‑SEP meetings, and construct a novel SEP surprise variable from a Bloomberg survey of expected median FOMC projections to isolate the information content of the projections themselves. They then relate SEP surprises to changes in forecasters’ inflation expectations and to macroeconomic outcomes, assessing whether projection releases convey information beyond the policy rate decision.
Findings
- Monetary policy surprises are 2–3 times larger in absolute value at SEP meetings than at non‑SEP meetings, consistent with the model’s prediction that forecast releases amplify measured shocks by revealing additional information and by anchoring expectations between releases.
- The SEP surprise measure explains about 30% of the variation in monetary policy surprises at SEP meetings and largely accounts for the difference between SEP and non‑SEP meetings, indicating that projection releases are central to those shocks.
- SEP surprises lead forecasters to revise inflation expectations and are associated with meaningful changes in macroeconomic variables, showing that the information conveyed in projections and not just the rate decision, plays an important role in policy transmission.
When the Fed Reveals Its Hand: The SEP and Monetary Policy Surprises
Authors: Andrew B. Martinez, Tara M. Sinclair
From: American University, The George Washington University
