City Research Online

Abstract

Customers of pension plans often rely on financial advice to make investment decisions. This paper proposes a pension plan design that improves both optimization and communication of retirement savings. Unlike conventional approaches, we advocate setting only an upper limit on desired income in the payout phase, while allowing the allocation to risky assets to remain unconstrained. Simulation results show that this design stabilizes both the average and volatility of pension payouts without imposing inefficient restrictions on early investment choices. To enhance client understanding, we introduce a set of decision criteria for retirement drawdowns that supports simple, intuitive interaction and builds trust. Our approach is fully implementable in AI-driven wealth management systems, enabling fintech applications to guide clients effectively and assisting human advisors in providing better, data-driven insights. The proposed framework balances risk-adjusted returns and client comprehension, offering a practical, evidence-based solution for improving long-term retirement outcomes.

Publication Type:	Article
Additional Information:	This is an Accepted Manuscript of an article to be published by Taylor & Francis in North American Actuarial Journal, available at: www.tandfonline.com/journals/UAAJ
Publisher Keywords:	financial innovation, fintech, income drawdown, long-term saving, pension, artificial intelligence in wealth management
Subjects:	H Social Sciences > HC Economic History and Conditions H Social Sciences > HF Commerce H Social Sciences > HN Social history and conditions. Social problems. Social reform Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Departments:	Bayes Business School Bayes Business School > Faculty of Finance
SWORD Depositor:	Symplectic Administrator

Export

Downloads