Tara Sinclair on Building a Synthetic FOMC Through AI - Macro Musings with David Beckworth Recap

Podcast: Macro Musings with David Beckworth

Published: 2025-11-10

Duration: 52 min

Guests: Tara Sinclair

Summary

Tara Sinclair discusses her innovative paper on using AI to simulate FOMC meetings, exploring how this technology can illuminate monetary policy decision-making processes.

What Happened

Tara Sinclair joins David Beckworth to discuss her paper 'FOMC and Silico,' which explores using large language models (LLMs) to simulate Federal Open Market Committee (FOMC) meetings. This approach allows for the examination of decision-making processes within the FOMC, including the dynamics of persuasion, institutional norms, and career incentives. Sinclair highlights the challenge posed by missing economic data, especially during critical economic turning points, and discusses the limitations of relying solely on private sector data. She argues for the importance of government-provided data for accurate economic analysis and policy-making.

Sinclair explains the two models used in her research: a baseline Bayesian voting model and an LLM deliberation model, which incorporate more human-like interactions among FOMC members. The use of AI in this context is positioned as a potential tool for experimental economics in macro, enabling the testing of different scenarios and decision-making processes. Sinclair and Beckworth delve into the potential applications of this research, including stress-testing central bank governance and exploring the impact of different chair candidates on FOMC dynamics.

The conversation touches on the potential for AI to revolutionize macroeconomic research by providing a sandbox for testing various hypotheses in monetary policy. The discussion also covers the possible future applications of AI in forecasting and the implications for economic data availability. Sinclair and Beckworth discuss how AI could potentially aid in understanding the effects of political pressure on FOMC decisions and the importance of maintaining diverse representation within the committee.

Sinclair's insights suggest that while AI can offer valuable tools for simulation and forecasting, it remains crucial to check these outputs against human judgment to ensure accuracy and relevance. The episode concludes with a discussion on the broader implications of AI for the economics profession, emphasizing the need for economists to engage with this technology to remain at the forefront of research and policy-making.

Key Insights