CapSolve: A Solvency Assessment and Prediction Framework for Workers’ Compensation Captive Insurance Companies

Captive insurance has become a popular way to run workers’ compensation programs in recent years. The assessment and prediction of the solvency for the workers’ compensation captive (WCC) are essential for its stable operation and the protection of the insured employees. However, the publicly available solvency assess-ment methods uniquely designed for WCC are lacking. In this paper, we present CapSolve, the first kind of open-source model for WCC solvency assessment and prediction. The model is developed from the Insurance Regulatory Information System (IRIS) and has several significant advantages over IRIS. For instance, compared with the IRIS deterministic point estimation, this model is probabilistic, which can return richer information such as distribution and Value-at-Risk (VaR) of future solvency. The inputs of the model are the financial statements of a WCC. The Monte Carlo simulation techniques are applied to compute the solvency scores. A Panjer recursion formula is given for aggregate loss distribution computing with retention and payroll exposure changes in consideration. Based on the default or user-defined solvency judgment criteria, it can predict the future solvency scores under different retention levels and perform VaR calculation to estimate the required capital at the current year. This model can be used by actuaries, WCC managers, and regulators to assess and predict the solvency of their WCC companies, conduct stress tests and scenario analysis, as well as to support the decision of the optimized retention selection and regulations.