News and Networks: Using Text Analytics to Assess Bank Networks During COVID-19 Crisis

Abstract

We study the "interconnectedness" of stress-tested banks by analyzing their financial news coverage. Following a text-to-network approach and using the COVID-19
pandemic as an external shock, we examine how bank networks behave during stress periods. We propose a measure of systemic risk that employs text-based eigenvector centrality, a relative measure of influence in a network. We show that this measure complements other traditional systemic risk measures, delivers more stable systemic risk rankings, and explains movements in financial variables. Our results showcase the value of soft information in the context of systemic risk.

Keywords: Banks, Networks, Systemic Risk, Text Analysis, COVID-19

JEL Classification: G1, D85, L14, G21, G28, G32