Skip to main content
Article | Pensions Briefing

What could the mortality impact of COVID-19 be for pension scheme liabilities?

The importance of looking beyond the headline death-toll figures

Retirement|Pension Board and Trustee Consulting|Pensions Corporate Consulting|Pensions Risk Solutions|Pensions Technology
COVID 19 Coronavirus

By Stephen Caine | April 28, 2020

A severe pandemic outcome may have a limited immediate impact on pension scheme mortality, but can the longer-term implications be more material? Stephen Caine explores in more detail.

Attributing someone’s death to a cause is not a perfect science, especially during a pandemic. Some deaths, particularly outside hospitals, may not have been correctly identified as COVID-19-related (with both under- and over-reporting), and elderly people often die of multiple health issues. Tragically, some people are likely to have died from non-COVID-19 illnesses through the additional stress on the health service and/or because they avoided GPs/hospitals. It is therefore important to look at total deaths occurring, rather than just those that are directly attributed to coronavirus.

Analysis carried out by the Actuarial Profession suggests that the total number of deaths in England and Wales by 27 April may have exceeded those expected by this stage of a normal year by between 38,000 and 45,000. This is around twice as high as the 19,544 who had died in hospital by that date after testing positive for COVID-19. The number of additional deaths is of course expected to increase.

What’s the potential impact on defined benefit pension scheme liabilities?

There are two routes by which the pandemic could affect the size of the pension payments that fall due to be paid in future. First, there will be an immediate effect, where scheme members sadly die as a direct or indirect result of the pandemic. Second, the pandemic’s legacy might affect future mortality rates.

  • Whilst it looks as though we may have already surpassed this, if the number of additional deaths in the population could have been held down to 25,000, this would have represented a circa 4% increase in the number of deaths in 2020. We estimate this might reduce the average pension scheme’s liabilities by under 0.1%, if post-pandemic mortality rates were not affected.
  • 250,000 deaths (an outcome more akin to the Spanish Flu of 1918/19) would represent about a 40% increase in deaths – unprecedented in recent times. However, this may only reduce the average pension scheme’s liabilities by 0.5%.

It might be surprising to find that even a severe pandemic scenario with a material death toll is not expected to have a significant immediate impact on pension scheme liabilities. This is because it is a one-off shock that disproportionately affects the elderly. It cuts short the stream of pension payments to these individuals by a few years on average, while leaving projected payments to surviving members as they were.

Payments to the much larger number of members who should not die during the pandemic might, however, also be affected if mortality rates in the aftermath of the virus differ from those that would have been experienced. Such changes could be negative, either because the virus turns out to have long-term health implications for those who recover or because of how public wellbeing and reduced government resources following the economic shutdown affect people’s health. For instance, if life expectancy improvements stalled for another decade, similar to what has been seen in the UK since 2011, the value of the future pension payments that an average scheme is committed to meeting could be 1%-2% less than currently expected (in addition to any reduction from the immediate impact of deaths during the pandemic). These trends would, however, take years to be realised and we could find that the opposite happens – healthier habits influenced by the pandemic could improve life expectancy.

If life expectancy improvements stalled for another decade, the value of the future pension payments that an average scheme is committed to meeting could be 1%-2% less than currently expected.”

Stephen Caine | Senior Mortality Consultant

Right now, predicting the outcome for a given pension scheme is very challenging and will depend on how a scheme’s membership is affected, which will vary with the age- and sex-profile of its members (the immediate impact will be bigger for schemes with older and more male memberships) and potentially where they live. The profile of deaths in the population will also continue to evolve. However, it seems that, the short-term liability impact of even a very bad scenario will not be huge for most schemes and could be eclipsed by the indirect long-term mortality impact as well by as the effects of financial market movements.


Stephen Caine
Senior Mortality Consultant

Contact Us