We empirically investigate how bank internationalization, organizational complexity, and geographical complexity stemming from foreign-affiliate type and geographic dispersion affect parent bank stability and profitability. We base our analysis on unique, hand-collected data for the worldwide locations of subsidiaries and branches of EU banks. Our results show that internationalization benefits bank stability by reducing default risk, and it is significantly associated with lower earnings volatility but poorer profitability. With regard to foreign organizational complexity, banks with both foreign subsidiaries and foreign branches are more stable than banks with foreign branches exclusively, which are more stable than banks with only foreign subsidiaries. Nevertheless, higher geographic complexity is associated with lower default risk, higher volatility in earnings, and higher profitability. Further investigations on the sovereign debt crisis and bank size indicate that the sovereign debt crisis in 2011 amplified the relationship and our findings mainly hold for small banks.

In this paper, we explore whether economic uncertainty differently affects the default risk of Islamic and conventional banks. Using a sample of 568 banks from 20 countries between 2009 and 2018, we use the World Uncertainty Index (WUI) by Ahir et al. (2018) to conduct a study based on a comparable measure across countries. Our findings indicate that economic uncertainty increases the default risk of conventional banks but does not affect Islamic banks’ default risk. To understand why, we explore the influence of religiosity, institutional factors, and bank-level heterogeneity. We observe that Islamic banks’ default risk is not significantly affected by uncertainty in all types of countries, but such a difference with conventional banks mainly holds for banks with higher income diversification, larger size, and that are publicly traded. Moreover, our findings show that conventional banks suffer more from uncertainty in terms of stability in countries with higher religiosity and with a higher share of profit-loss sharing (PLS) contracts. Our results are robust to alternative estimation techniques to deal with endogeneity and to alternative variable measurements.