Stochastic Models for Pensionable Service

This paper investigates the effects of vesting rules and termination rates on the accumulation of pensionable service. First, a theoretical framework is developed, through stochastic models in continuous time, to characterize the cumulative pensionable service with reference to its moments, under an arbitrary age-service vesting rule. In the most general case, the rate of employment termination is allowed to be a function of both age and time spent in an employment. Next, a discrete computational model is presented for the determination of the distribution of pensionable service and of related measures, using age and service-dependent termination rates as input. This model is applied to Canadian Public Service and the results are discussed. Also considered are the employer's perspective and passage from pensionable service to pension benefits.