Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Mind the Shift: Decoding Monetary Policy Stance from FOMC Statements with Large Language Models

Yixuan Tang, Yi Yang

Abstract

Federal Open Market Committee (FOMC) statements are a major source of monetary-policy information, and even subtle changes in their wording can move global financial markets. A central task is therefore to measure the hawkish--dovish stance conveyed in these texts. Existing approaches typically treat stance detection as a standard classification problem, labeling each statement in isolation. However, the interpretation of monetary-policy communication is inherently relative: market reactions depend not only on the tone of a statement, but also on how that tone shifts across meetings. We introduce Delta-Consistent Scoring (DCS), an annotation-free framework that maps frozen large language model (LLM) representations to continuous stance scores by jointly modeling absolute stance and relative inter-meeting shifts. Rather than relying on manual hawkish--dovish labels, DCS uses consecutive meetings as a source of self-supervision. It learns an absolute stance score for each statement and a relative shift score between consecutive statements. A delta-consistency objective encourages changes in absolute scores to align with the relative shifts. This allows DCS to recover a temporally coherent stance trajectory without manual labels. Across four LLM backbones, DCS consistently outperforms supervised probes and LLM-as-judge baselines, achieving up to 71.1% accuracy on sentence-level hawkish--dovish classification. The resulting meeting-level scores are also economically meaningful: they correlate strongly with inflation indicators and are significantly associated with Treasury yield movements. Overall, the results suggest that LLM representations encode monetary-policy signals that can be recovered through relative temporal structure.

Mind the Shift: Decoding Monetary Policy Stance from FOMC Statements with Large Language Models

Abstract

Federal Open Market Committee (FOMC) statements are a major source of monetary-policy information, and even subtle changes in their wording can move global financial markets. A central task is therefore to measure the hawkish--dovish stance conveyed in these texts. Existing approaches typically treat stance detection as a standard classification problem, labeling each statement in isolation. However, the interpretation of monetary-policy communication is inherently relative: market reactions depend not only on the tone of a statement, but also on how that tone shifts across meetings. We introduce Delta-Consistent Scoring (DCS), an annotation-free framework that maps frozen large language model (LLM) representations to continuous stance scores by jointly modeling absolute stance and relative inter-meeting shifts. Rather than relying on manual hawkish--dovish labels, DCS uses consecutive meetings as a source of self-supervision. It learns an absolute stance score for each statement and a relative shift score between consecutive statements. A delta-consistency objective encourages changes in absolute scores to align with the relative shifts. This allows DCS to recover a temporally coherent stance trajectory without manual labels. Across four LLM backbones, DCS consistently outperforms supervised probes and LLM-as-judge baselines, achieving up to 71.1% accuracy on sentence-level hawkish--dovish classification. The resulting meeting-level scores are also economically meaningful: they correlate strongly with inflation indicators and are significantly associated with Treasury yield movements. Overall, the results suggest that LLM representations encode monetary-policy signals that can be recovered through relative temporal structure.
Paper Structure (44 sections, 7 equations, 4 figures, 11 tables)

This paper contains 44 sections, 7 equations, 4 figures, 11 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Measuring monetary-policy stance from text.
  4. Self-supervised probing of latent representations.
  5. Method
  6. Problem Formulation
  7. Contextual Feature Extraction
  8. Dual-Axis Projection
  9. Delta-Consistent Objective
  10. Post-Hoc Directional Anchoring
  11. Experiments
  12. Data
  13. FOMC statements.
  14. Sentence-level benchmark.
  15. Macroeconomic and market data.
  16. ...and 29 more sections

Figures (4)

  • Figure 1: Monetary-policy stance is inherently relative. Statement $t$ ("a single rate hike ahead") is hawkish in isolation ($s_t$ above neutral), yet signals a dovish shift relative to Statement $t{-}1$ ("multiple rate hikes ahead").
  • Figure 2: Overview of Delta-Consistent Scoring (DCS). Given two consecutive FOMC statements, a frozen LLM produces absolute and relative representations. A dual-axis projection module maps these representations to an absolute stance score for each statement and estimates the relative shift between them. DCS then aligns score differences with estimated shifts through a delta-consistency objective, turning temporal ordering into a source of self-supervision without requiring human stance labels.
  • Figure 3: Federal Funds Rate (blue, left axis) and DCS stance scores (orange, right axis) across four policy periods. Shaded regions mark the four evaluation periods: P1 (2003--2008, conventional policy), P2 (2008--2015, near-zero rates), P3 (2015--2020, policy normalization), and P4 (2020--2025, pandemic and tightening cycle).
  • Figure 4: Spearman $\rho$ between DCS stance scores and year-over-year CPI and PPI changes across layers of DeepSeek-R1-14B. The final layer (L48, highlighted) achieves the highest correlations on both indicators.