This analysis measures geopolitical risk sentiment through earnings calls, highlighting the perception of risks and opportunities across industries.
ConstructionTo measure how industries perceive geopolitical risk, we construct an industry-specific geopolitical risk sentiment index grounded in firm communications through earnings calls. Building on Niepmann and Shen (2025) and Shen (2025), we use the NL Analytics platformâa commercial implementation of the natural language processing methodology developed by Hassan et al. (2019)âto process over 240,000 earnings call transcripts from approximately 7,000 U.S.-domiciled firms, spanning the period 2002:Q1 to 2024:Q3. This platform enables us to apply our own natural language process-based criteria within its framework. We use the geopolitical risk dictionary from Caldara and Iacoviello (2022) and identify GPR-relevant sentences, classifying them as negative or positive based on their proximity to sentiment words from the Loughran and McDonald (2011) dictionary. By quantifying both positive and negative sentiment, we can distinguish between instances in which geopolitical developments are viewed as risks versus opportunitiesâa nuance that pure mention-count methods cannot capture.
No takes yet. Share an insight, caveat, or question.
Culver et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: