Firm-Level Risk and Systemic Risk Analysis in the Insurance Sector during the Subprime Mortgage Crisis: CDS VS. Non-CDS-Based Risk Indictors

Abstract

Although extensive literature has investigated the systemic relevance of the insurance sector during the financial crisis of 2007-2009, mixed results have been presented. This thesis considers firm-level risk assessments in contrast to systemic risk evaluations to examine whether insurers are riskier than banks and non-financial institutions (NFIs). The role of CDS data in capturing risks has been compared with that of non-CDS data as well in this research. By applying CDS and non-CDS data of worldwide banks, insurers and NFIs into the two groups of risk methodologies, this thesis shows that insurers contribute more to and are affected more by credit risk than the other two sectors. This study also finds that credit risk links among companies are stronger than other types of risk connections. What’ more, CDS provides earlier risk warning signals than non-CDS information. In addition, firm-level risk has relatively weak non-linear correlation with systemic risk, i.e. firm risk is not able to reflect some information that is only contained in systemic risk. Finally, the systemic vulnerability analysis of distress dependence matrix (DiDe SV) and SRISK, involving both equity and balance sheet information, are superior to any other risk measures adopted in this thesis in terms of its predictive ability of the subprime mortgage crisis.

Date of Award	3 Apr 2019
Original language	English
Awarding Institution	University of Bath
Supervisor	Simone Giansante (Supervisor) & Ian Tonks (Supervisor)