Life expectancies are among the most material assumptions used to assess pension scheme funding requirements. This means that for sponsors of schemes at or close to full funding, obtaining clarity in this area is likely to be vital to their pension strategies in the years ahead.
In the early phases of the Covid-19 pandemic, many schemes took the view that there was insufficient longitudinal data to inform quantitative estimates of the impact of Covid-19 on their populations. It is notable that to date very few defined benefit schemes have moved to reflect the impact of Covid-19 on their memberships.
This view is beginning to evolve as it is becoming evident from national statistics that all-cause “excess deaths” continued during the latter half of 2021 while Covid-19 deaths were very low. The figure below using national data from the Office for Health Improvement and Disparities illustrates the differences in the patterns of excess deaths since 1 April 2021, by which time much of the UK adult population had been vaccinated and Covid-19 deaths were at low levels. Middle-aged groups have experienced 12% more deaths that would have been expected in absence of the pandemic compared to 1% fewer deaths than would have been expected in the oldest (85+ years) age group.
All-cause excess deaths (%) for adults in England since 1 April 2021 to January 2022 by age group
Source: Office for Health Improvement and Disparities
The oldest age groups were impacted worst during the first year of the pandemic where direct Covid-19 deaths were highest. It is plausible that excess deaths since 1 April 2021 reflect the ‘indirect impacts’ of the pandemic whereby the strain that the pandemic has put on non-Covid-19 services, alongside changes in health-seeking behaviour of individuals. For example, we have seen nearly 1,000 more weekly deaths in people’s private homes across England since the pandemic began, with heart disease and stroke amongst the leading causes which could suggest patients unable to access the highest quality of care when they need it most. Alongside this displacement of death there have been delays to cancer diagnosis and commencement of treatment affecting more than 50,000 patients to date which will impact future longevity.
As a result many sponsors are beginning to assess the impact that Covid-19 has had on “best estimate” liabilities in their corporate accounts and subsequently reflecting this in wider funding measures. Given these trends and the uncertainty of whether these indirect impacts of the pandemic may persist and how they may impact different sections of the population differently, longevity is perhaps the area of pension scheme experience that currently shows the greatest level of uncertainty – and so the greatest need for contingency planning.
For example the chart below illustrates a range of potential scenarios that at various times during the past two years might have been considered plausible.
Covid-19 pandemic scenarios to inform longevity assumptions
Based on current evidence, LCP’s latest analysis suggests that the likely medium to longer-term impact of the pandemic could lead to downwards adjustments in the rate of future improvements that could result in falls in liabilities by up 1-2% compared to pre pandemic levels. For sponsors this is important as, without making proactive steps to validate their scheme’s own situation, trustees have little incentive to rapidly update the potentially significant margins for prudence that may now sit in their life expectancy assumptions.
For sponsors of the largest schemes, their membership data is likely to be a strong basis for meaningful experience analysis to add value and draw inferences about the range of potential lasting impacts of the Covid-19 pandemic on future trends to inform into future funding settlements. For example, for FTSE100 schemes, LCP estimates that a 1.5% adjustment to liabilities might correspond to around a £1bn p.a. reduction in annual contribution requirements.
However, given the range of future scenarios for the lasting impacts of the pandemic at this early stage, achieving this level of savings may require further development of contingent funding mechanisms, for example so that additional contributions are paid if longevity improvements revert to previous trends. Given the levels of materiality and uncertainty involved, we expect to see rapid development in this area in upcoming valuations.