Measuring and testing for the systemically important financial institutions

• We develop a statistical test to identify systemically important banks.
• The test is based on the systemic risk contribution ΔCoVaR.
• We apply the test on a sample of 26 European banks.
• Accounting for statistical significance implies that very few banks can be ranked.

This paper analyzes ΔCoVaR proposed by Adrian and Brunnermeier (2011) as a tool for identifying/ranking systemically important institutions. We develop a test of significance of ΔCoVaR that allows determining whether or not a financial institution can be classified as being systemically important on the basis of the estimated systemic risk contribution, as well as a test of dominance aimed at testing whether or not, according to ΔCoVaR, one financial institution is more systemically important than another. We provide an empirical application on a sample of 26 large European banks to show the importance of statistical testing when using ΔCoVaR, and more generally also other market-based systemic risk measures, in this context.