Over the past few weeks there have been some publications in the field of mortality that make for interesting reading. In this blog, I am going to focus on the Continuous Mortality Investigation (CMI) producing their latest mortality projections – which, quite surprisingly, showed that mortality rates were higher in 2015 than 2014.
In figures, because that’s what we actuaries like, 2015 mortality improvements are estimated to be around 2.3% p.a. lower for 18-102 year olds and around 3.2% p.a. lower for 65-102 year olds.
So, the 2015 figures alone show a slight reversal in the continual improvement in mortality seen over recent years, and highlight the slowdown in the rate of mortality improvements. We have seen this in previous versions of the CMI mortality projection model, with new models producing lower life expectancies than the previous iteration being the norm in recent years. Looking at the four year period from 2011 to 2015, average annual mortality improvements using the CMI model are estimated to be around 0.3% p.a. for the 18-102 age group and 0.1% p.a. for the 65-102 age group. Therefore, average life expectancy is estimated to have increased by around 3-4 months over the whole of the period between 2011 and 2015. In comparison, in the period from 2000 to 2011, life expectancy increased by around 3 months each year. This emphasises the potential slowdown in mortality improvements shown by the CMI model.
For pension schemes (including trustees and sponsoring employers), this could potentially be welcome news, in a purely financial sense of course. Roughly, an increase of one year in life expectancy results in a 3-4% increase in liabilities. Therefore, any slowdown in improvements would help reduce liability figures.
However, as always there are notes of caution to be sounded. The higher than expected mortality rates seen in the period since 2011 (in particular the mortality rates seen in 2015) may be due to short term variations rather than being reflective of a longer term trend. For example, the slowdown looks to be more pronounced for the older age group, indicating that variations in the effect of illnesses to which the elderly are particularly prone may be having an effect on the annual mortality rates – an effect that may not be borne out over the longer term.
The implications for the CMI model are as yet unknown. The first version of the next iteration of the model will be published in March 2017. It will be worth waiting for that to gain a further insight in to the trend in mortality improvements. In the meantime, it would be advisable not to place potentially undue emphasis on the recent changes in mortality.
So that’s dealt with trustees and sponsoring employers. What then about members of a pension scheme? In the “good old days” when Defined Benefit (“DB”) pension schemes ruled the roost, there wasn’t much for the pension scheme member to worry about. The pension received at retirement was defined, a proportion of the pension increased in line with inflation and was paid until the member died. However, with the increased prevalence of Defined Contribution (“DC”) pension schemes (either in the workplace or personal pension schemes), the member now has to think a bit more about their life expectancy and their financial planning in retirement. If an individual was to take their DC pension pot all as cash, there is the potential that this amount would be funding a shorter retirement than previously. If the member was to buy an annuity or other retirement product, such providers may decide to allow for this effect in their pricing (although that remains to be seen). In my next blog, I will look further into how mortality affects members of DC schemes, and how it affects planning for retirement.