FOMC In Silico: A Multi-Agent System for Monetary Policy Decision Modeling

Sophia Kazinnik () and Tara Sinclair

No 2025-005, Working Papers from The George Washington University, Department of Economics, H. O. Stekler Research Program on Forecasting

Abstract: We develop a multi-agent framework for modeling the Federal Open Market Committee (FOMC) decision making process. The framework combines two approaches: an LLM-based simulation and a Monte Carlo implementation of a generalized Bayesian voting model. Both begin from identical prior beliefs about the appropriate interest rate for each committee member, formed using real-time data and member profiles. In a simulation replicating the July 2025 FOMC meeting, both tracks deliver rates near the 4.25-4.50% range's upper end (4.42% LLM, 4.38% MC). Political pressure scenario increases dissent and dispersion: the LLM track averages 4.38% and shows dissent in 88% of meetings; the MC track averages 4.39% and shows dissent in 61% of meetings. A negative jobs revision scenario moves outcomes lower: LLM at 4.30% (dissent in 74% of meeting), and MC at 4.32% (dissent in 62% of meeting), with final decisions remaining inside the 4.25-4.50% range. The framework isolates small, scenario-dependent wedges between behavioral and rational baselines, offering an in silico environment for counterfactual evaluation in monetary policy.

Keywords: Generative AI; Multi-Agent Systems; Large Language Models, Federal Open Market Committee; Monetary Policy; Simulations (search for similar items in EconPapers)
JEL-codes: C63 C73 D83 E52 E58 (search for similar items in EconPapers)
Pages: 52 pages
Date: 2025-08
New Economics Papers: this item is included in nep-mac and nep-mon
References: Add references at CitEc
Citations:

Downloads: (external link)
https://www2.gwu.edu/~forcpgm/2025-005.pdf First version, 2025 (application/pdf)

Related works:
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:gwc:wpaper:2025-005

Access Statistics for this paper

More papers in Working Papers from The George Washington University, Department of Economics, H. O. Stekler Research Program on Forecasting Contact information at EDIRC.
Bibliographic data for series maintained by GW Economics Department ().