Interconnectedness among insurers and reinsurers at a global level is not well-understood and may pose a significant risk to the sector, with implications for the macro-economy. Models of the complex interactions among re-insurers are with other participants in the financial system and the real economy are at a very early stage of development and the market remains opaque both to regulators and market participants. To begin to assess the stability of the global reinsurance market, and the risk of contagion amplifying potential losses, the authors create several plausible networks between global reinsurers, each consistent with reinsurer financial statement data, and some involving results from the theory of random matrices. These networks are stress-tested under a series of severe but plausible catastrophic loss scenarios. This analysis contributes to the literature by: (1) applying a network model approach common in the banking literature to the insurance industry (2) assessing the interconnections among reinsurers through potential claims rather than premiums, and (3) investigating the most opaque part of the global insurance market, namely counterparty arrangements among global reinsurers (retrocession). The authors find that contagion in the global retrocession market is plausible and that the size of the potential market is sensitive to: (A) the distribution of risk among counterparties (B) the trigger for financial distress (C) the time horizon for claims resolution and (D) the degree of loss-netting. The findings suggest that further study of industry practices in these four areas would improve our ability to assess risk in the insurance sector and promote financial stability.
Joint work with Jane Voll, Bank of Canada; Darrell Leadbetter, OSFI; Bin Lu, Department of Applied Mathematics, Western University; Nadia Kotb, Department of Statistical & Actuarial Sciences, Western University.
Bienvenue à tous!