The process funding pension plans is viewed as a dynamic control process. Two performance measures are introduced to evaluate the effectiveness of plan contributions: the cost-induced performance measure (CIPM) and the ratio-induced performance measure (RIPM). A dynamic programming approach is used to determining the optimal contributions with the objective of minimizing the performance measure. The methodology developed is applied to a sample of members of Taiwan's Public Employees Pension Plan (Tai-PERS). We show that RIPM produces more stable results than those using CIPM.