The banking sector, stress and financial crisis: symmetric and asymmetric analysis

ABSTRACT

This study investigates the impact of economic and financial stress on US banking sector returns during periods of crisis and tranquility. It further examines symmetric and asymmetric effects. The study applies GARCH (1, 1) methodology and describes stock returns based on the Fama–French–Carhart extended capital asset-pricing four systematic factors model. The results indicate that during the entire study period (from 10 January 2003 to 29 September 2017), US banking sector returns responded negatively to stress-induced changes, and investors were more sensitive to stress increases (negative news) than stress declines (positive news), especially during the financial crisis. These results support the view that stress shocks constitute a systematic asset price risk to the US banking sector. Investors and policymakers should both consider these shocks when modelling asset prices and evaluating banks’ stability.

KEYWORDS:

• Economic stress
• financial stress
• banking sector
• asymmetric effect
• financial crisis

JEL CLASSIFICATION:

• G12
• G21

Acknowledgments

The authors would like to thank anonymous referee for his constructive comments on a previous version of this paper.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Other studies, notably Adrian, Friedman, and Muir (2015); Gandhi and Lustig (2015); Carmichael and Alain Coën (2018), emphasize on the importance of considering Fama–French SMB (small minus big) and HML (high minus low) factors when pricing bank stocks.

² The distributions of US banking sector stock returns data series are characterized by skewness, leptokurtic behaviour, and non-normality (see Table A2). Tests not reported here reveal that the return series have inherent linear and non-linear dependency as well as volatility clustering, while the estimation of the GARCH (1, 1) model with Student’s t distribution is adequate. Results are available upon request.:

³ To avoid multicollinearity, in Equation (1), we use orthogonalized weekly excess market returns.

⁴ The financial crisis period is based on the official timelines of the Federal Reserve Board of St. Louis (Dimitriou, Kenourgios, and Simos 2013).

⁵ The results are available from the authors upon request.

⁶ The results are available from the authors upon request.