IMF-使用DSGE模型中的冲击分解评估历史事件（英）-2025.3_39页_2mb

Working Papers Summary

Title: Evaluating Historical Episodes Using Shock Decompositions in the DSGE Model
Authors: Zamid Aligishiev, Michael Ben-Gad, and Joseph Pearlman

Introduction and Main Contribution:
The paper addresses the challenge of interpreting historical episodes in Dynamic Stochastic General Equilibrium (DSGE) models. While standard methods for shock decomposition (DC2) often fail to isolate the effects of shocks within a specific sub-period due to persistence in prior shocks, and alternative methods (DC3) overlook initial conditions, the authors propose a preferred method (DC1) that precisely isolates the impact of shocks within the analyzed sub-period. This method helps avoid biased interpretations driven by historical factors not under focus during the episode.

Key Findings and Implications:

• Comparative Analysis of Decomposition Methods:
  The study compares three methods (DC1, DC2, and DC3) using an extended DSGE model. Each method produces different interpretations of the same historical events, particularly during periods such as post-World War II growth and the 2008 financial crisis. For example:

  • During 1964-1966 U.S. growth, DC1 attributes the expansion primarily to cumulative effects from prior shocks, while DC2 emphasizes fiscal policy shocks.
  • In the 2008 financial crisis, DC1 identifies interest rate spreads and preferences shocks as key drivers, whereas DC2 significantly underestimates the persistent effects of prior shocks.

• Economic Model Extensions:
  The authors extend the DU model by including:

  • Additional data series (interest rates, debt, labor market variables).
  • A preference shock parameterized with persistent negative effects during the 2001 recession and COVID-19 pandemic.
  • Four interest rate wedges to isolate financial frictions.
    These extensions enhance the model’s ability to explain persistent trends in U.S. macroeconomic history.

Policy Relevance:
The paper underscores the importance of selecting appropriate shock decomposition methods for accurate historical analysis. Poor methodological choices can obscure or distort the effects of monetary and fiscal policies during critical periods. The authors advocate for the use of DC1 when analyzing sub-periods, especially those involving recessions or policy shifts.

recommendations:

• Researchers should prioritize DC1 to isolate period-specific shocks.
• Central banks and policy institutions should adopt DC1 methodologies to ensure clarity in policy evaluations.
• Governments may use the updated model to assess stimuli effectiveness, particularly during crises, by focusing on exogenous factors like preference and policy rate shocks.

Conclusion:
The paper corrects existing gaps in DSGE shock decompositions by providing a refined methodology. This improves historical episode analysis and supports better-informed policy recommendations during economic transitions.

Academia Contribution:

• Offers a systematic comparison of shock decomposition methods.
• Explores the role of preference shocks during recessions and fiscal policy adjustments.
• Enhances DSGE applicability for real-world economic narrative by addressing prior shock persistence.