Signals in the storm: the impact of government announcements and COVID-19 developments on financial markets

Abstract The COVID-19 pandemic presented an unprecedented challenge to global health systems and financial markets. This study specifically investigates the dynamic interaction between the communication of the UK’s Government, the progression of the COVID-19 pandemic, and stock market trends in three distinct phases of the pandemic: initial outbreak, peak crisis, and vaccination rollout. Using a combination of machine learning techniques, particularly BERT-based sentiment analysis, correlation, and Granger causality analyses, we dissect the relationship between public announcement sentiment, pandemic metrics (cases and deaths), and the returns and prices of the stock markets. Our findings delineate a pattern of market sensitivity to the tone and content of government communications and the severity of the pandemic. The study not only underscores the critical role of government communications in managing expectations during crises, but also illustrates the effects of pandemic developments on the stock market.