Central bank communication during the war: the case of the National Bank of Ukraine

How does war change communications between state and society? We evaluate the language of the National Bank of Ukraine (NBU), as presented in post-decision announcements and tweets from September 2019 to March 2024. We distinguish between pre-war and wartime samples. Using a variety of natural language processing methods, we extract four variables describing the messages of the NBU (sentiment, polarity, subjectivity, and saturation of war language). Additionally, we use the BERTopic modeling technique to extract, identify, and quantify the underlying themes in both types of corpora. Nonparametric tests confirm the differences in the distributions of variables before and after the invasion. The NBU changes the tone and subjectivity of both types of communication after the invasion. The results are robust to different methods of analysis, but are more ambiguous for Twitter/X. To the best of our knowledge, we present the first comprehensive case study of wartime language by a central bank.