Abstract
"The recent global financial crisis of 2007-2009 is widely regarded as the worst since the great depression and
threatened the global financial system with a total collapse. Healthy banks are important for every economy.
Financial distress has a negative impact on the prosperity of a country and is prone to spread beyond the banking
sector. Hence, the development of an adequate early warning system for bank failures is essential. This thesis
distinguishes itself from the vast literature of bankruptcy, bank failure and bank exit prediction models by
introducing novel categorical parameters inspired by Switzerland’s banking landscape. This thesis evaluates data
from 274 banks in Switzerland over the period from 2007 to 2017 using a Generalised Linear Model (GLM) with
logit link function and a Multinomial Logistic Regression (MNL) to evaluate determinants of corporate
restructuring and financial distress. For model comparison, it presents a robustness test via a Bayesian framework
with Markov Chain Monte Carlo-methods featuring a basic Gaussian random-walk Metropolis-Hastings
algorithm. The findings suggest that total assets and net interest margin impact bank exit and mergers and
acquisitions (M&A) categories. Furthermore, the study reveals that cost-to-income has a positive relationship to
bank exit. Therefore, both net interest margin and cost-to-income are the main key performance indicators in the
field of retail banking and wealth management. Furthermore, the findings suggest that the two macroeconomic
covariates, gross domestic product and unemployment rate, are unrelated to bank exit. Specifically, for the Swiss
area the results indicate that banks operating in the Zurich cluster have a higher exit likelihood and become a M&A
target more frequently, than banks operating in the Geneva area."