A Risk Based approach for the Solvency Capital requirement for Health Plans

TL;DR

This paper introduces a risk-based method using a three-part regression model and Monte Carlo simulation to estimate the Solvency Capital Requirement for health plans, improving efficiency and accuracy.

Contribution

It proposes an original three-part GLM approach combined with Monte Carlo simulation to assess health plan risks more efficiently than traditional methods.

Findings

Reduces the number of regression models needed

Provides probability distribution of health plan costs

Enables estimation of various risk measures

Abstract

The study deals with the assessment of risk measures for Health Plans in order to assess the Solvency Capital Requirement. For the estimation of the individual health care expenditure for several episode types, we suggest an original approach based on a three-part regression model. We propose three Generalized Linear Models (GLM) to assess claim counts, the allocation of each claim to a specific episode and the severity average expenditures respectively. One of the main practical advantages of our proposal is the reduction of the regression models compared to a traditional approach, where several two-part models for each episode types are requested. As most health plans require co-payments or co-insurance, considering at this stage the non-linearity condition of the reimbursement function, we adopt a Montecarlo simulation to assess the health plan costs. The simulation approach provides the probability distribution of the Net Asset Value of the Health Plan and the estimate of several risk measures.