U.S. banks’ lending, financial stability, and text-based sentiment analysis

Article  -  Thursday, 7 April 2022

Agoraki M.E. , Aslanidis, N., Kouretas, G.: "U.S. banks’ lending, financial stability, and text-based sentiment analysis", Journal of Economic Behavior and Organization

We examine the impact of investor sentiment on bank credit and financial stability. We also investigate how loan growth may affect bank stability. We use a large panel data set of U.S. commercial banks over the period 1999Q1-2015Q4, using bank-level data. Investor sentiment is proxied by two novel but alternative measures based on textual analysis. First, we employ the sentiment measure constructed by García (2013) based on the fraction of positive and negative words in two columns of financial news from the New York Times. Second, we employ the text-based measure of uncertainty constructed by Manela and Moreira (2017) called News Implied Volatility, which uses front-page articles of the Wall Street Journal. The results show that banks' lending falls when investor sentiment is low, while this effect is more pronounced when banks hold a higher level of credit risk. These effects are more pronounced during recessions with loan growth also responding negatively to the anxiety of investors throughout recessions. Finally, we show that the Great Financial Crisis had a negative effect on investor sentiment leading to a decline in U.S. lending behavior and an increase in the U.S. banking sector instability.

Up