Paper Profitability Context and the Cross-Section of Stock Returns

Empirical asset pricing models implicitly assume that firm characteristics are context-free. However, companies provide rich narrative context that allows investors to put numeric information into perspective. These narratives include discussions of non-quantitative factors such as demand trends, competitive dynamics, and future strategies, all of which impact the interpretation of a firm’s performance. Our study emphasizes the significance of contextual information in asset pricing, specifically focusing on the narratives surrounding profitability figures. Using a large language model, we incorporate contextual information into the measurement of profitability. We find that context-adjusted profitability significantly outperforms conventional profitability measures both in statistical and economic terms. Furthermore, the context-adjusted profitability factor shows superior performance in portfolio tests and helps to resolve the biggest challenge facing the five-factor model (Fama and French, 2015). Our results underscore that incorporating context not only improves investment strategies but also enhances the asset pricing models.

Get the Paper