Valuation of diversified banks: New evidence

This paper reconsiders the effect of diversification on bank valuation. Our objective is to provide new evidence based on a unified estimation framework that places particular emphasis on separating the effects of diversification (specialised banks vs. diversified banks) from those of bank type (investment banks vs. commercial banks). Consistent with prior studies, we find a significant diversification discount at the end of the 1990s. Our main finding is that it decreases over time and practically vanishes after the financial crisis. We do not find support for the hypothesis that the diversification effect is influenced by geographical or regulatory factors. The valuation impact of bank characteristics varies over time, particularly in the financial crisis, but this structural break does not explain the observed decrease of the diversification discount. We show that the pre-crisis discount is considerably smaller in a robust regression, which in part is driven by banks with a large share of non-interest income.