The Covid-19 pandemic severely disrupted the global economy. This paper introduces the CoRisk-Index, a real-time economic indicator measuring corporate risk perceptions related to Covid-19. Using data mining techniques on US company reports (10-K filings), the index combines the frequency of 'corona' words and the average text negativity of corona-related sentences. The CoRisk-Index correlates with US unemployment rates across industries and market volatility, even preempting stock market losses. Topic modeling and natural language processing provide granular insights into industry-specific concerns. The index bridges the gap between volatile stock market data and longer-term macroeconomic figures, aiding researchers and policymakers in assessing and responding to economic risks.

Fabian Stephany, Leonie Neuhäuser, Niklas Stoehr, Philipp Darius, Ole Teutloff, Fabian Braesemann