Three new working papers on the transmission of monetary policy through households in the US, Japan, and EU converge on the same conclusion, which contradicts the standard transmission mechanisms emphasized in New Keynesian models. The three papers use different identification strategies. Grigoli et al. use randomized information treatments conducted in the U.S., Takeshi Niizeki uses a combination of high-frequency event study and information experiments from Japan, while De Grauwe & Ji use observational CES data from the EU with correlational analysis.
The U.S. case
The paper by Francesco Grigoli, Damiano Sandri, Yuriy Gorodnichenko, and Olivier Coibion surveys over 25,000 U.S. households, combining hypothetical policy scenarios with eight randomized information treatments (each conveying factual data on a different macroeconomic variable) to separately identify how households revise expectations and how each type of expectation passes through to spending and portfolio decisions.
Findings:
- Households do reduce spending in response to a perceived Fed tightening, particularly on durables, but the effect is asymmetric: rate hikes reduce spending, while rate cuts have no statistically significant effect.
- The dominant transmission channel is inflation expectations, not intertemporal substitution. Households expect rate hikes to raise inflation consistent with a cost-channel interpretation i.e. higher policy rates → higher borrowing costs for firms→ higher firm costs → passed through to consumer prices. It is this expectation that drives most of the consumption decline.
- The income and real interest rate channels are largely inactive: households do not perceive a meaningful pass-through from policy rates to wages, so the standard New Keynesian mechanisms contribute little to the estimated response.
- Portfolio reallocations follow the same logic: households shift away from stocks and toward bank deposits after a perceived tightening, driven primarily by revised inflation expectations rather than expected asset returns.
- Financial literacy is a first-order moderator: the entire spending effect is concentrated among households with basic Fed knowledge and financial literacy; for all others, pass-through is statistically indistinguishable from zero.
- A precautionary labor supply channel emerges: both rate hikes and cuts prompt households to seek additional income, suggesting any perceived policy change signals greater uncertainty rather than a clear directional shift.
Monetary Policy According to Households: Perceptions, Reactions and Channels
Authors: Francesco Grigoli, Damiano Sandri, Yuriy Gorodnichenko, Olivier Coibion
From: Georgetown University, BIS, UC Berkeley, UT Austin
The Japanese case
Takeshi Niizeki in this paper tracks all 16 Bank of Japan monetary policy announcements between April 2023 and March 2025, including the historic exit from negative interest rate policy in March 2024 and subsequent rate hikes to 0.25% and 0.5%. He uses two identification strategies: a high-frequency design surveying approximately 750 households two days before and after each press conference, and information provision experiments (RCTs) immediately following three of the largest tightening announcements.
Findings:
- Households exhibit widespread rational inattention: despite measurable increases in news exposure after press conferences, inflation and interest rate expectations show no statistically significant revision across all 16 announcements, including the historic end of unconventional monetary policy.
- The dominant transmission channel is interest income, not the cost channel: the primary reason Japanese households expect tightening to be positive for their consumption is the anticipated rise in deposit income (cited by 67% of respondents), a channel essentially absent in the U.S. and euro area contexts.
- Treatment effects are not persistent: the expectation revisions disappear almost entirely within two weeks in most cases, reflecting deeply entrenched low-inflation norms after decades of near-zero rates.
- Only 24% of households knew the current BoJ policy rate and only 10% had any anticipation of upcoming announcements, underscoring the depth of inattention.
Monetary Policy Announcements and Household Expectations: Evidence from Two Identification Strategies
Author: Takeshi Niizeki
From: Economic and Social Research Institute
The EU case
In this paper, Paul De Grauwe and Yuemei Ji use ECB Consumer Expectations Survey (CES) monthly data from 11 euro area countries covering the full 2020–25 inflation cycle. The methodology is observational using VAR regressions with Granger causality tests, autocorrelation analysis, and correlation analysis and it is applied to actual, perceived, and expected inflation series.
Findings:
- Perceptions lag reality by about 4 months: inflation perceptions react to changes in actual inflation with substantial inertia, producing a cyclical bias: negative during the surge phase (households underestimate inflation) and strongly positive during the decline phase (households overestimate inflation).
- Expectations are more stable than perceptions and more tightly correlated with decision-relevant variables (income, spending, unemployment expectations), making them the more policy-relevant variable for understanding household behavior.
- The transmission channel is indirect and operates through wage formation: inertia in inflation expectations feeds directly into nominal wage negotiations, producing a strong negative correlation between inflation and real wages over the cycle and not through intertemporal substitution or borrowing costs.
- The puzzling consumption link: high inflation expectations correlate positively with higher expected household spending, suggesting households anticipate front-loading consumption before prices rise further, rather than cutting back.
- ECB interest rate hikes did not defeat inflation, energy price declines did. The rate hikes mattered primarily as a credibility signal, preserving trust in the ECB and keeping long-term expectations anchored, not through the standard demand-suppression channel.
Inflation Perceptions and Expectations: Inertia, Biases and Policy Implications
Authors: Paul DE GRAUWE, Yuemei JI
From: LSE, UCL
