The information in systemic risk rankings

A principal components-based methodology to combine systemic risk rankings of financial institutions with the aim of achieving a robust combined ranking.

Since the 2008–2009 global financial crisis, many new approaches have been developed to quantify and rank ‘systemic risk’ contributions of firms in the financial sector. We argue that the presence of multiple alternative and arguably noisy measures of financial sector firms’ systemic risk contributions raises two sets of important questions. First, is there a straightforward way to combine currently available systemic risk rankings to amplify the signal and reduce the noise due to model risk and estimation uncertainty? And if so, is the combined ranking sufficiently robust for policy purposes and targeted banking supervision? Second, does a robust measure of systemic importance correlate with financial institutions’ cost of debt finance in a way that suggests a public sector guarantee for the most systemically important institutions, as suggested by other studies?

In our study we provide a principal components-based methodology to combine systemic risk rankings of financial institutions with the aim of achieving a robust combined rank- ing. The combined ranking is based on six risk ranking methodologies and disentangles their common (signal) and idiosyncratic parts. In order to extract the common informa- tion content of different input rankings, we perform an iterated cross-sectional aggregation of rankings based on a fairly straightforward factor model. The combined ranking is con- structed from the eigenvalue that explains most of the variance of the observed data subject to a normalization.

We apply our general framework to the European Union financial sector, studying N = 113 firms during T = 139 months from March 2002 to September 2013. As a result, our sample contains most of the build-up phase of financial instability in Europe during the expansion years of 2003-07, the materialization of systemic risks during the global financial crisis between 2008-09, and the most acute phase of the euro area sovereign debt crisis from 2010-12.

Our analysis focuses on three main empirical findings. First, we demonstrate that the cross-sectional consistency between the different rankings is far from perfect. The respective mean rank correlations are positive, but routinely fall below 0.5. The poor association is not due to a few outliers, but is symptomatic of the fact that different ranking methodologies actually order financial firms differently in the cross-section. Second, when studying the time series of the results from our cross-sectional principal components analysis, we find a rising discrepancy during 2006–2007 between the loadings of price-based systemic risk rankings (such as VaR, ?CoVaR, and MES) versus systemic risk rankings that also incorporate book values (such as Leverage and SRisk). This appears to signal a dislocation between market prices and fundamentals already in the onset to the 2008 financial crisis. Finally, we investigate the relationship between systemic importance and credit default swap (CDS) prices. In this regard we demonstrate that the systemic importance of financial firms tends to vary negatively with their CDS premia, provided that the sovereign is financially healthy. 

logo experttube

Video's op het gebied van Audit & Control, Actuariaat & Risk Management, Juridisch & Fiscale Zaken, Pensioenen, Schade & Hypotheken, Compliance en Investment Management.

Bekijk ons volledige overzicht op

logo CareerTube

Videoplatform met werkenbij video's van toonaangevende organisaties in de financiële wereld. Met een focus op de finance specialisatie zorgt de koppeling met de 17 (niche) vacaturesites van CareerGuide direct voor een relevant bereik.

Bekijk ons volledige overzicht op