Essays on Corporate Communications: Risk Measures, Analyst Dynamics, and Reporting Consistency

This thesis employs natural language processing (NLP) techniques, state-of-the-art large language models (LLMs), and artificial intelligence (AI) to extract critical information from earnings conference calls. The thesis introduces three novel text-based measures designed to investigate firms’ exposure to financial reporting risk, analyst engagement during earnings conference calls, and inconsistencies in financial performance across disclosure channels. The thesis offers valuable insights into the impact of these disclosures on capital market outcomes and financial reporting quality.