<scp>AI</scp> ‐Generated Price‐Setting Insights from Firms' Earnings Calls*

We introduce new firm‐level indices covering input costs, demand and final prices based on firms' earnings calls. These are constructed using a powerful large language model (LLM). We show that our new LLM‐based indices are superior to those based on detailed lists of keywords. The new indices have a leading relationship with conceptually similar indices from popular business surveys, as well as related official statistics, such as consumer price inflation, demonstrating that they are relevant for assessing current economic conditions. Next, we use our indices to examine various aspects of firms' price‐setting behaviour. The reduced‐form associations we estimate indicate that price‐setting behaviour depends on the source of the shocks firms face (demand or cost‐driven), the direction of the shock (with firms reacting more to cost increases relative to decreases) and which industries are most affected. This underscores the importance of continuing to develop rich multisector models of the economy to better understand firms' reactions to different types of shocks.