Which Voices Move Markets? Speaker Identity and the Cross-Section of Post-Earnings Returns

2026-04-14

AI Summary

This study investigates the heterogeneous impact of different speakers—such as analysts and CFOs—during earnings conference calls on post-call cross-sectional stock returns. Leveraging transcripts from S&P 500 firms between 2015 and 2025, the paper introduces a novel role-weighted sentiment measure constructed using FinBERT, which empirically estimates speaker-specific sentiment influence weights. Compared to conventional dictionary-based approaches like Loughran-McDonald, this method demonstrates significantly superior out-of-sample predictive performance, achieving a Spearman information coefficient of 0.142 and generating a monthly long-short portfolio alpha of 2.03% (t = 6.49). These results remain robust after controlling for standardized unexpected earnings (SUE) and the Fama-French five-factor model, underscoring the critical role of speaker heterogeneity in financial text sentiment analysis.

Technology Category

• Natural Language Processing: Sentiment Analysis, Stylistic Analysis, and Argument Mining
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Web Mining and Content Analysis: Sentiment analysis and opinion mining
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

We utilize FinBERT, a domain-specific transformer model, to parse 6.5 million sentences from 16,428 S&P 500 quarterly earnings call transcripts (2015-2025) and demonstrate that post-earnings stock returns are not equally affected by all speakers in a conference call. Our section-weighted sentiment, with empirically derived speaker weights (Analyst 49%, CFO 30%, Executive 16%, Other 5%), achieves an out-of-sample Spearman IC of 0.142 versus 0.115 in-sample, generates monthly long-short alpha of 2.03% unexplained by the Fama-French five-factor model (t = 6.49), and remains significant after controlling for standardized unexpected earnings (SUE). FinBERT section-weighted sentiment entirely subsumes the Loughran-McDonald dictionary approach (FinBERT t = 5.90; LM t = 0.86 in the combined specification). Signal decay analysis and cumulative abnormal return charts confirm gradual price adjustment consistent with sluggish assimilation of soft information. All results undergo rigorous out-of-sample validation with an explicit temporal split, yielding improved rather than deteriorated predictive power.

Problem

• speaker identity
• post-earnings returns
• earnings call transcripts
• sentiment analysis
• market reaction

Innovation

• FinBERT
• speaker-weighted sentiment
• earnings call transcripts
• cross-sectional returns
• out-of-sample validation