OPTIMAL RISK DIVERSIFICATION FOR A REINSURANCE COMPANY UNDER A STOCHASTIC CLAIM APPROACH


Creative Commons License

Mert Ö. M., Korn R., Kestel A. S.

European Actuarial Journal Conference 2024, Lisbon, Portekiz, 9 - 11 Eylül 2024, ss.102-103, (Özet Bildiri)

  • Yayın Türü: Bildiri / Özet Bildiri
  • Basıldığı Şehir: Lisbon
  • Basıldığı Ülke: Portekiz
  • Sayfa Sayıları: ss.102-103
  • Orta Doğu Teknik Üniversitesi Adresli: Evet

Özet

Given the critical role insurance markets play in the global economy, the necessity for reinsurance contracts has increased due to factors like population expansion, catastrophic incidents, and political and economic uncertainties. For instance, the extensive devastation caused by natural disasters such as the Maras¸ earthquake disaster in Türkiye may surpass the coverage capacity of insurance companies alone. This underscores the significance of reinsurance policies, particularly in the face of natural disasters such as this. With the surge in the risk of insurance losses resulting from such events, individual reinsurance firms may decide to enter into agreements with multiple insurance companies. Among these agreements, some are recognized as stop-loss contracts. This study aims to explore both (i) contracts with retention and (ii) contracts encompassing both retention and upper limits (cap), which are types of stop-loss contracts. The study focuses on optimizing risk diversification strategies for reinsurance companies across both types of contracts.

The stochastic loss and exposure curve models to be used in the study are among the factors that increase the importance of the research in terms of providing both randomness and time-dependent modeling, and making a fair premium allocation between the parties. This study is motivated by the studies [1], [2], and [3], in which the geometric Brownian motion model and Pareto-Beta jump diffusion model are used in loss modeling and the examination of loss under GBM in terms of drift and diffusion, and the examination of loss under Pareto-Beta jump diffusion in terms of drift and diffusion, and the jump tendency that is caused by extreme-valued losses resulting from events such as natural disasters are aimed.

In this study, we consider the stop-loss contracts involving one reinsurer and two insurers under the stochastic claim approach. For this, we propose the Heston model with an approximated variance process to capture the dynamics of the sum of claim amounts belonging to different stop-loss contracts. Under this assumption, we implement the time influence on the stop-loss contract in the framework of the stochastic model and derive the parties’ time-dependent expected costs and exposure curves. Moreover, the analytical expressions of exposure curves are presented to ascertain the allocation of premiums between the reinsurer and insurer, considering factors such as time, loss, retention, and both retention and cap mechanisms.

Furthermore, in this study, we use partially real data due to difficulties such as confidentiality in accessing real data in terms of actuarial losses. The other part of the data to be used is obtained through simulation on real data, and we examine the compatibility of the proposed model with this data. In this regard, parameter estimates and error analyzes are made under some analytically derived assumptions.

The main aim of the study is to find the optimal risk diversification levels for the stop-loss contracts with one reinsurer and two insurers for both contract types and to represent them as barrier levels which are derived from the equivalence relationship between the expected total costs of the reinsurer and insurers. To achieve this, we formulate the analytical expression for the total cost difference. Obtaining a solution in closed form poses challenges due to the integration of characteristic functions from the Heston model, resulting in complex numbers. Hence, we employ the Newton iteration with an adaptable learning rate to find the optimal barrier levels. We also give numerical illustrations for our findings, and we anticipate that the findings of this study will provide practical insights for practitioners in devising strategies that involve multiple stop-loss contracts and optimizing premium allocation within their optimal framework.