美联储-2019年以来美联储工作人员通胀预测错误回顾（英）-2025.8_31页_2mb

Retrospective on Federal Reserve Staff Inflation Forecast Errors Since 2019

This paper examines the Federal Reserve Board Staff's inflation forecast errors from 2019 onward, with a focus on the COVID-19 pandemic period. The analysis shows that staff consistently underestimated both the size and persistence of inflation surges, particularly following the economic reopening in early 2021. Key reasons include supply-demand imbalances, model specification errors, and limited historical data for comparison. Staff responses included incorporating new data indicators, adjusting inflation models, and accounting for nonlinear effects, improving forecast accuracy but leaving significant uncertainty. Inflation was primarily driven by supply shocks and rapid demand recovery, with lessons highlighting the challenge of forecasting in volatile environments.

Key Findings:

• Staff forecast errors peaked from 2021 to 2022, due to miscalculations of supply disruptions and labor market dynamics.
• Improvements include better measurement of market imbalances and updated Phillips curve frameworks.
• Uncertainty remains regarding the quantitative impact of various drivers, and applicability to future scenarios.