Stories that (are) Move(d by) Markets: A Causal Exploration of Market Shocks and Semantic Shifts across Different Partisan Groups

2025-02-20

AI Summary

This study investigates the causal relationship between financial market shocks and shifts in public semantic representations, with particular attention to how partisan alignment moderates both the predictive power of textual signals for market volatility and heterogeneous market responses. Methodologically, it introduces the first causal inference framework linking semantic drift in text to anomalous market movements, integrating LiNGAM-based causal discovery, dynamic word embeddings (TimeLM), cross-media textual time-series modeling, and grouped heterogeneous causal effect estimation. Key contributions are: (1) robust evidence that semantic space displacement significantly predicts market deviations (p < 0.01); (2) identification of a 37% gap in predictive accuracy attributable to partisan orientation; and (3) demonstration that during exogenous crises—such as the COVID-19 pandemic—the explanatory power of textual signals increases 2.8-fold, validating text-derived semantic dynamics as a potent, externally valid predictor of macroeconomic fluctuations.

Technology Category

• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: The sharing economy
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Macroeconomic fluctuations and the narratives that shape them form a mutually reinforcing cycle: public discourse can spur behavioural changes leading to economic shifts, which then result in changes in the stories that propagate. We show that shifts in semantic embedding space can be causally linked to financial market shocks -- deviations from the expected market behaviour. Furthermore, we show how partisanship can influence the predictive power of text for market fluctuations and shape reactions to those same shocks. We also provide some evidence that text-based signals are particularly salient during unexpected events such as COVID-19, highlighting the value of language data as an exogenous variable in economic forecasting. Our findings underscore the bidirectional relationship between news outlets and market shocks, offering a novel empirical approach to studying their effect on each other.

Problem

• Market shocks impact semantic shifts.
• Partisanship influences market fluctuation predictions.
• Language data aids economic forecasting.

Innovation

• Causal link semantic shifts
• Partisanship influences market predictions
• Text signals during unexpected events