Public vs Private - Is there a sector effect in post-retirement mortality?

In this paper we show there is significant diversity of longevity experience for pension plans within both the public and private sectors in both the UK and Canada.

By interrogating large datasets of pension plan mortality data, we can identify data fields which predict very different longevity outcomes for individuals within pension plans. It is better to use these predictors to set a tailored assumption for a plan than to assume the same experience as the sector average.

We find there is little argument for using separate datasets to model public and private sector post-retirement longevity. Further, once we have included other more descriptive predictors, such as place of residence, occupation and affluence measures, we also find there is little to no benefit of including a public/private sector identifier to capture longevity diversity in either the UK or Canada. We expect similar results to hold in the US and will test this in our next generation US model.