As AI democratizes sophisticated investment strategies, improved understanding of agent-client relationships becomes key for industry professionals who will need to add more value profitably. We believe this is better revealed than in prior agent research by more realistically modeling manager-client coalitions as vehicles for increasing their joint utility. Our model optimizes portfolio asset allocations based on weighting each party's expected utility as their own risk-adjusted financial return with minimum requirements.
We model manager decisions as partly driven by fee structures and the need to maintain and grow future business — factors often dominating conventional portfolio optimization. Our model incorporates both current fee income and expected future fees based on perceived client reactions to performance.