Risk and premium allocation in decentralized finance insurance

Yang, Jiajie

Permalink

https://hdl.handle.net/2142/129267 Copy

Description

Title

Risk and premium allocation in decentralized finance insurance

Author(s)

Yang, Jiajie

Issue Date

2025-04-29

Director of Research (if dissertation) or Advisor (if thesis)

Wei, Wei

Doctoral Committee Chair(s)

Feng, Runhuan

Committee Member(s)

• Quan, Zhiyu
• Jing, Xiaochen

Department of Study

Mathematics

Discipline

Mathematics

Degree Granting Institution

University of Illinois Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Decentralized insurance
• Peer-to-peer insurance
• Distributed insurance
• Risk allocation
• Capital allocation.

Abstract

In this thesis, we mainly consider two practical models in decentralized finance insurance. In the first part, we will consider the risk allocation in peer-to-peer insurance. Using the criterion of minimizing total variance, we propose an optimal risk-sharing framework and demonstrate that the linear risk-sharing strategy should be applied to the residual risks rather than the original risks, which is contrary to common practice in the literature. This approach provides better variance reduction and stronger robustness. We further extend the model by introducing constraints that capture desirable principles such as fairness and participation incentivization. In the second part, we will consider the risk and premium allocation in distributed insurance. Unlike peer-to-peer insurance, distributed insurance enables individual investors to serve as insurers by sharing both risk and premium. We introduce a flexible loss allocation that generalizes the model that is currently being used in practice. It also unifies the existing allocation methods in traditional insurance, including proportional allocation and tranche insurance. This flexibility provides investors with multiple investment options, which increases market accessibility. We then design a premium allocation that incorporate the investment incentivization. The investors can be demonstrated to have better investment performance when investing more capital. Additionally, we develop a multi-period model in distributed insurance, which addresses the problem of lacking this model in practice. This model accounts for claim development and multi-period investment durations. We show that this system not only maintains fairness but also encourages longer-term investment, which aligns with the practical needs of a growing distributed insurance market.

Graduation Semester

2025-05

Type of Resource

Thesis

Handle URL

https://hdl.handle.net/2142/129267

Copyright and License Information

Copyright 2025 Jiajie Yang

Owning Collections