Communication of ECB Governing Council members: Impact on intraday financial markets from media messages

The goal of this paper is to determine the heterogeneity in media messages across the ECB Governing Council members, employing the GPT language model, and assess its impact on financial markets with high-frequency data since 2014. Our findings uncover trends in sentiments and topics in line with economic developments and similar studies. We highlight a marked divergence between hawkish and dovish members. Using regression analysis, we demonstrate that the media messages after the communication of Governing Council members significantly influence various euro area asset classes. These effects are material across different periods. Our study also highlights specific topics that exert a more pronounced influence on market dynamics. Hawkish communication is observed to have a more substantial effect on sovereign bond yields. Furthermore, media messages of Executive Board members and the Governors of Germany, France and Austria are found to have the most significant overall impact on euro area markets.