GPT Deciphering Fedspeak: Quantifying Dissent Among Hawks and Doves

TL;DR

The study tackles how FOMC sentiment toward inflation is reflected across transcripts and statements, arguing that transcripts reveal dissent obscured in final statements. Using GPT-4, the authors construct a gold-label framework for hawk/dove sentiment and derive logit-based scores to compare documents, finding that transcripts contain substantially more dissent than statements. A key metric, $ heta^{(L)} = \log\left( (\text{Hawk}+0.5)/(\text{Dove}+0.5) \right)$, captures relative hawkishness and aligns with manual annotations, though extremes are better represented when ingesting entire statements rather than sentence-by-sentence averages. The results imply that relying solely on statements underestimates dissent and that LLM-driven analysis of multiple document types can improve understanding of FOMC signaling and its implications for policy and markets. Future work could extend analyses to minutes and refine prompts for reproducibility and finer-grained interpretation of dissent signals. $

Abstract

Markets and policymakers around the world hang on the consequential monetary policy decisions made by the Federal Open Market Committee (FOMC). Publicly available textual documentation of their meetings provides insight into members' attitudes about the economy. We use GPT-4 to quantify dissent among members on the topic of inflation. We find that transcripts and minutes reflect the diversity of member views about the macroeconomic outlook in a way that is lost or omitted from the public statements. In fact, diverging opinions that shed light upon the committee's "true" attitudes are almost entirely omitted from the final statements. Hence, we argue that forecasting FOMC sentiment based solely on statements will not sufficiently reflect dissent among the hawks and doves.

Figures (8)

• Figure 1: The transcripts (left) contains opinions and disagreements but statements (right) are concise. We analyze both datasets using GPT-4 prompting.
• Figure 2: Text taken from an FOMC meeting on December 11, 2001. A dovish statement does not reflect the hawkish sentiment of Speaker 1. GPT-4 can quantify dissent lost from transcripts to statements.
• Figure 3: Our analysis of statements finds that averaging at the sentence-level (gray) loses information since the average sentence is Neutral. Ingesting the overall statement (blue) better mirrors the manual gold label (gold).
• Figure 4: Our analysis of transcripts with GPT-4 at the speaker level shows that meetings consistently have dissenting opinions. Similarly to the previous figure, 2001 to 2004 is dovish (blue), while 2004 to 2006 is hawkish (orange).
• Figure 5: Discussion of inflation (Red and Blue topics) is more contentious than other topics and the average of all data (dotted line).