Capturing international influences in U.S. monetary policy through a NLP approach

Laurent Ferrara and Nicolas de Roux ()
Additional contact information
Nicolas de Roux: EconomiX - EconomiX - UPN - Université Paris Nanterre - CNRS - Centre National de la Recherche Scientifique

Working Papers from HAL

Abstract: Officially, the U.S. Federal Reserve has a statutory dual domestic mandate of price stability and full employment, but, in this paper, we question the role of the international environment in shaping Fed monetary policy decisions. In this respect, we use minutes of the Federal Open Market Committee (FOMC) and construct indexes of the attention paid by U.S. monetary policymakers to the international economic and financial situation. These indexes are built by applying natural language processing (NLP) techniques ranging from word count to built-from-scratch machine learning models, to OpenAI's GPT models. By integrating those text-based indicators into a Taylor rule, we derive various quantitative measures of the external influences on Fed decisions. Our results show that when there is a focus on international topics within the FOMC, the Fed's monetary policy generally tends to be more accommodative than expected by a standard Taylor rule. This result is robust to various alternatives that includes a time-varying neutral interest rate or a shadow central bank interest rate.

Keywords: Monetary policy; Federal Reserve; FOMC minutes; International environment; Natural Language Processing; Machine Learning (search for similar items in EconPapers)
Date: 2025
Note: View the original document on HAL open archive server: https://hal.science/hal-05072535v1
References: Add references at CitEc
Citations:

Downloads: (external link)
https://hal.science/hal-05072535v1/document (application/pdf)

Related works:
Working Paper: Capturing international influences in U.S. monetary policy through a NLP approach (2025)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hal:wpaper:hal-05072535

Access Statistics for this paper

More papers in Working Papers from HAL
Bibliographic data for series maintained by CCSD ().