Mortality Improvement – Increasing Again?
As Canadian defined benefit plan sponsors are aware, it wasn’t that long ago that the mortality tables and mortality improvement rates were updated. In 2014 the Canadian Institute of Actuaries released a new series of mortality tables (the CPM mortality tables), and a new mortality improvement scale (the CPM-B improvement scale). Transitioning to these new tables and improvement scale was a bit painful, as they generally increased a plan’s liability by about 5%.
This past September, the Canadian Institute of Actuaries issued a Task Force Report on Mortality Improvement. One of the main purposes of the report was to construct a new mortality improvement scale – not just for work on pension and post-retirement benefit plans, but for life insurance, social security, and other actuarial work too.
The new mortality improvement scale that was produced by the Task Force is called the MI-2017 improvement scale. While there are some noteworthy differences between the new MI-2017 improvement scale and the CPM-B improvement scale that came out in 2014, a key message for plan sponsors is that adopting this new improvement scale would increase a plan’s liability slightly (in most cases, less than 1%).
The CPM-B and MI-2017 improvement scales are quite similar in that they are two-dimensional scales, varying by age and calendar year. In particular, at every age the improvement scales start off using recent mortality improvement experience, and transition to an assumed ultimate mortality improvement rate over a period of about 20 years.
One small difference is that the MI-2017 improvement scale reflects more recent mortality improvement experience by reflecting experience up to 2015, whereas the CPM-B improvement scale only reflects improvement experience up to 2007.
However, a major difference is the ultimate mortality improvement rate used in these scales – with the MI-2017 improvement scale using an ultimate improvement rate of 1.0% for most ages, and the CPM-B improvement scale using an ultimate improvement rate of 0.8% for most ages. This ultimate improvement rate is a very subjective assumption, as no one knows how future mortality trends will progress. Furthermore, the increase in the ultimate improvement rate from 0.8% to 1.0% is a key reason for the increase in the liabilities caused by transition from CPM-B to the MI-2017 improvement scale.
Implications for Plan Sponsors
As a generalization, it is our expectation that plan sponsors and actuaries will move to adopt the new MI-2017 improvement scale over the next year or so.
As many of you know, there are different measures of a defined benefit plan’s liability: the benefit obligations used for financial reporting purposes, as well as the liabilities used for funding purposes (i.e. going-concern and hypothetical wind-up).
For the benefit obligations used for financial reporting purposes, management is responsible for the selection of the assumptions. Given that the MI-2017 improvement scale was only recently released, that the impact of the change from the CPM-B improvement scale is not terribly significant, and the fact that a key reason for the increase is due to the change in the rather subjective ultimate improvement rate, it is our expectation that many plan sponsors may not adopt this assumption in their upcoming fiscal year 2017 year-end financial statements. Nevertheless, plan sponsors should be aware of this development, and will want to consider ‘if and when’ to adopt the MI-2017 improvement scale, and potentially discuss their approach with their auditor.
For going-concern funding valuations, the MI-2017 improvement scale may be reflected as the best-estimate assumption as early as 2018.
However, for hypothetical wind-up valuations, we note that the MI-2017 improvement scale has not yet been adopted for the commuted value transfer value standard. As such, this new improvement scale will not have an immediate impact on hypothetical wind-up valuations. However, I would expect the Actuarial Standards Board to make this change in the not too distant future.