Optimal Reinsurance Strategies Under Nigerian Regulatory and Market Constraints: A Stochastic Optimisation Approach

This study develops a stochastic optimisation framework for reinsurance strategy selection under the specific regulatory and market constraints facing Nigerian insurance companies, contributing an original actuarial decision-making model to the reinsurance optimisation literature. The classical results of reinsurance optimisation theory, including Borch's theorem for mutually optimal quota share treaties and Arrow's theorem for excess-of-loss optimality, are derived under assumptions of complete markets and unconstrained reinsurance access that are violated in the Nigerian context. Nigerian insurers face regulatory constraints on maximum net retention, limited domestic reinsurance capacity, foreign exchange costs of offshore cession, counterparty credit risk on non-rated reinsurers, and NAICOM first-right-of-refusal requirements for Nigeria Re. This study modifies the stochastic optimisation framework to incorporate these institutional constraints and derives optimal reinsurance structures under alternative objective functions including minimum ruin probability, maximum expected utility of terminal surplus, and minimum cost subject to solvency constraints. The framework is applied to portfolio data from five Nigerian non-life insurers, with Monte Carlo simulation of aggregate claims and reinsurance recovery outcomes. Findings establish that constrained-optimal reinsurance structures differ substantially from their unconstrained theoretical counterparts, with Nigeria Re first-refusal constraints creating portfolio rebalancing costs averaging 8 percent of reinsurance premium. The study contributes an original constrained reinsurance optimisation model for Nigeria and recommends that NAICOM review the first-refusal requirement to reduce its adverse impact on ceding company risk management efficiency.

Keywords: reinsurance optimisation, stochastic control, Nigeria Re, solvency, actuarial decision theory.