IMF-家庭预期有助于预测通货膨胀吗？（英）-2023.10-32页_1mb

Summary

I. Introduction and Motivation

• Rising global inflation since 2021 has increased interest in different measures of inflation expectations.
• This paper analyzes whether changes in the distribution of household inflation expectations contain information about future inflation.
• Uses microdata from the US, UK, Germany, and Canada. Focuses specifically on US data to explore predictive power econometrically, examining moments (mean, variance, skewness) beyond only the median.

II. Recent Shifts in Household Expectations Distributions

• Observed significant shifts in inflation expectations:
  • Distributions shifted rightward ("fat tailing": increased frequency of high inflation beliefs).
  • Changes in variance (increased disagreement among households) and skewness (fatter right tail).
  • Many households expect significantly higher inflation than pre-2020 levels (< 10 percent or more).
• These shifts correlate with actual high inflation.

III. Econometric Analysis of Predictive Power (US Data)

• Methodology: Regressed one-year-ahead inflation on:
  • Current inflation.
  • Moments of the household inflation expectation distribution (median, variance, skewness).
  • Moments of professional forecasts and market-based expectations.
  • Robustness tests included multiple models (different breakdowns, time series frequencies, functional principal components).
• Key Findings:
  • Distributional characteristics beyond the median add statistically significant predictive power for one-year-ahead CPI inflation.
  • Variance: Decrease predicts higher inflation (negative coefficient). Sign changes with regime (positive as inflation rises, negative after a peak, e.g., pre-1980 in US).
  • Skewness: Decrease predicts higher inflation (negative coefficient). The sign remains robust.
  • This predictive power is especially significant during high-inflation periods and stronger when models account for inflation regimes.
  • Distributional characteristics add value beyond metrics from professional forecasts and market expectations.

IV. Conclusion

• Novel evidence suggests household inflation expectations contain information useful for predicting future inflation, complementary to professional forecasts and market indicators.
• Policymakers should monitor household expectations closely.
• Acknowledges work as a first step needing further exploration (e.g., other countries, impact on consumption/wage setting).
• Concludes the paper by calling for deeper analysis of open questions regarding the role of household expectations.