This thesis delves into the nuanced realm of monetary policy communication, focusing on its strategic importance in the actions of central banks, particularly the U.S. Federal Reserve and the Bank of Korea. Recognizing the critical role of communicati...
This thesis delves into the nuanced realm of monetary policy communication, focusing on its strategic importance in the actions of central banks, particularly the U.S. Federal Reserve and the Bank of Korea. Recognizing the critical role of communication in enhancing the efficacy of monetary policy, this study explores how central banks manage the timing, audience, and content of their communications. These communications encompass diverse aspects such as policy objectives, decisions, economic forecasts, and future directions, reflecting both regular and irregular modes of policy dissemination. A pioneering aspect of this research is the application of Natural Language Processing (NLP) techniques, specifically BERT (Bidirectional Encoder Representations from Transformers), to analyze the impact of these communications on key financial indicators in the South Korean economy. This study seeks to fill a significant gap in existing literature by providing a comprehensive examination of the effects of U.S. central bank communications on South Korea's economic outlook. Two central hypotheses are proposed. The first suggests that the sentiments expressed in U.S. central bank communications have a positive influence on South Korea's economic prospects, while the second focuses on the impact of specific keywords and themes in these communications on South Korea's economy. The research utilizes FinBERT for sentiment analysis and BERTopic for topic modeling to assess the influence of these communications on crucial financial instruments in South Korea, particularly the term spread between CD91 and Government Bond 3-year rates. The thesis adopts an interdisciplinary approach, blending linguistic analysis with economic theory, to interpret and analyze the complex language of central bank communications. This methodology aligns with recent research emphasizing the significance of language in economic decision-making. The findings not only contribute to academic discussions in economics and finance but also offer practical insights for policymakers and market analysts. This comprehensive approach enhances our understanding of the global monetary policy's impact on local economic conditions, addressing the intricate interdependencies within the global financial system. The structure of the thesis includes an exploration of the effects of monetary policy and central bank communications on economic indicators, a detailed methodology including data collection and preprocessing, and an analysis of the findings using advanced text mining techniques and economic theory. The thesis concludes by interpreting these findings, underscoring the pivotal role of central bank communications in shaping economic outcomes in a globally interconnected financial landscape. Keywords include FOMC, Monetary Policy Communication, BERTopic Topic Modeling, Fin-Bert, and Sentiment Analysis