Risk Sharing by Insurance Groups

Background. Global insurance market is dominated by insurance groups with the largest 100 insurance groups accounting for more than 85% of assets of the global insurance industry. Once insurers assume the risk from policyholders, it can either be kept on their balance sheet, or be passed on in the intermediation chain to other insurers, reinsurers, and the financial market. However, most of the risk resides within insurance groups while only about 10-15% is transferred outside the groups (OECD, 2018), potentially limiting risk-sharing.The overall objective of the proposed research is to analyze the following interrelated questions: (i)What is and which factors shape the size, structure, and complexity of international insurance groups? How does the group structure affect financial performance and risk-bearing capacity? (ii)What is the mechanism that determines the group’s incentives to share risk internally between group subsidiaries and externally with unaffiliated (re)insurers and the financial market? (iii)How do insurance group’s incentives for internal and external risk transfer affect its ability to insurer large risks, e.g., pandemic, climate, and cyber risks? Rationale. Research on the risk sharing and performance of insurance groups has been scarce. This is surprising as groups are the primary mechanism for risk-sharing in the insurance industry which facilitates the risk transfer equal to $7 trillion, or 8% of global GDP, in 2021. Possible reasons are that the insurance industry operates under a variety of national regulatory regimes and reporting standards, and the process of designing group supervision standards started only recently after the crisis experience in 2008. The specific aim of the research proposal is to carry out two projects. Project A ‘Informed capital’ is a theoretical contribution addressing questions (ii) and (iii). Building on the idea that insurers can learn information about their risks, we develop a pecking order theory of risk financing of insurance groups. Internal group reinsurance is the most informed form of capital followed by external reinsurance and the financial market. However, internal and external reinsurance capacity is limited by the group’s and reinsurance industry’s equity capital, respectively. Further, internal reinsurance imposes a negative externality on insolvency risk of other group subsidiaries. The novel contribution of Project A is that the model enables to reconcile the internal and external risk transfer by insurers. The equilibrium analysis reveals an excessive use of internal reinsurance which constrains aggregate risk sharing, particularly for large and information-sensitive risks like cyber and climate. It also explains how the interaction between different forms of financing is affected by the cross-border capital movement restrictions, the cost of external capital and by operation of the group subsidiaries under multiple regulatory or tax regimes. Project B ‘Global insurance groups’ is an empirical contribution addressing questions (i) and (iii). We build and analyze the novel comprehensive dataset on international insurance groups. Our aim is to assemble a dataset that contains time-series information about the insurance group corporate structure (i.e., the corporate tree where nodes are the subsidiaries and links are the ownership links) as well as financial information about the headquarters and subsidiaries, including the internal and external reinsurance data. The sample includes 100-120 global groups that represent the vast majoring of the global insurance market.Besides uncovering stylized facts about the global insurance market, we perform several empirical analyses to uncover the relationships between insurance group’s structure and its financial performance and risk. We also investigate the effect of the group structure on the performance and risk of the subsidiaries. We analyze the extent of intragroup risk sharing using internal reinsurance and how suboptimal risk sharing is related to corporate structure and complexity. Finally, we assess the impact of operating in multiple regulatory regimes on group’s structure and complexity.Expected results. We expect that our research will deliver the following results documented in two papers publishable in top-tier journals: (1) A comprehensive theory that integrates the internal and external reinsurance and capital market financing of insurance groups in a single model. (2) Stylized facts about the relationship between insurance groups size, global footprint, business focus and complexity. (3) Analysis of the relationship between group’ structure (i.e., size, global footprint, business focus, and complexity) and the financial performance and risk of the headquarters as well as the subsidiaries. (4) Analysis of the efficiency (i.e., risk-bearing capacity maximization) of internal risk sharing in insurance groups and its interrelationship with the nature of insured risks and the external reinsurance utilization. (5) Analysis of the impact of multiple regulatory regimes on group structure and the usage of offshore subsidiaries.