THE RELATIONSHIP BETWEEN STOCK RETURN SKEWNESS AND BANK FEATURES

Silvia Bressan,Alex Weissensteiner

doi:10.1142/s2282717x1850010x

Silvia Bressan, Alex Weissensteiner

Open Access

https://doi.org/10.1142/s2282717x1850010x

Copy DOI

Abstract

This paper studies to what extent bank-specific characteristics relate to stock return skewness. The main finding is that stock return skewness decreases significantly in bank size, measured in terms of total assets, i.e stocks of large banks are less skewed than those of small banks. This result holds for backward-looking skewness computed using the past stock returns, as well as for forward-looking skewness extracted from stock options. We interpret the empirical evidence by arguing that bank size increases the likelihood to have severe losses, to the point that investors expect to be compensated by receiving higher expected returns.

Highlights

The sampleWe source the data for the analysis from Orbis Bank Focus and from Datastream. We focus on companies from the United States classied by Orbis Bank Focus as \ ̄nancialrms." We keep only therms \listed" and \active," obtaining the annual accounting-based information for the years 2008 to 2016
Introduction and Connection to the LiteratureSkewness of stock returns has been studied quite extensively in the literature
We argue that increasing subordinated debt would provide banks with an incentive to be more risk aggressive, whereas deposits would discipline d Among the several papers where the bank size is approximated in the same way as we do in our analysis, see for example Demirgüç-Kunt & Huizinga (2013)

Summary

The sample

We source the data for the analysis from Orbis Bank Focus and from Datastream. We focus on companies from the United States classied by Orbis Bank Focus as \ ̄nancialrms." We keep only therms \listed" and \active," obtaining the annual accounting-based information for the years 2008 to 2016. To tackle this issue, we estimate the following linear regression model using the selection method of backward stepwise selection (Wooldridge 2012), in order to explain the skewness of bank j at time t in relation to bank-specic features: Skewnessj;t 1⁄4 0 þ Â Â Bank featuresj;t þ t þ j;t; ð2:6Þ where skewness is, alternatively, ^; ^ðMVÞ, and ^ðoptÞ. Standard errors in parenthesis. *p < 0:05, **p < 0:01 and ***p < 0:001

Empirical Evidence

Findings

Conclusions