The Continuous Mortality Investigation (CMI) has launched a consultation on how its next model should incorporate mortality data for 2020 following the COVID-19 outbreak.
Owned by the Institute and Faculty of Actuaries (IFoA), the CMI typically releases an updated version of its Mortality Projections Model each year, which are critical for valuing pension scheme and insurance company liabilities.
The next version, CMI_2020, is expected by the end of March 2021, however, it is feared that mortality rates recorded during the COVID-19 crisis could skew life expectancy expectations in a misleading way.
“The exceptional mortality experience during the coronavirus pandemic means that a version of CMI_2020 that takes account of mortality data for 2020 in the usual way would be likely to show substantial falls in life expectancy, which we think would be in excess of what most users of the model would consider reasonable,” the CMI said.
In response, the CMI is consulting on two proposed changes to the methodology for CMI_2020:
- To address the exceptional data for 2020, it has proposed to modify the calibration process for the model so that users can place more or less weight on data for individual years. For the core version of the model, it currently proposes to place no weight on data for 2020, and full weight on other years, but will review this in light of mortality experience for the rest of 2020.
- Unrelated to the experience of 2020, it has proposed to amend the age range of the dataset used to calibrate the model. The intention of this change is to avoid unrealistically low initial mortality improvements at high ages, caused by limitations in how the model copes with large historical shifts in mortality improvements that have varied significantly by age.
The CMI currently estimates that there have been around 60,600 more deaths in the UK since the start of the COVID-19 pandemic than there were during the corresponding period for 2019.
However, professional services firm Aon has cautioned against an overreaction, and suggested that future mortality may actually be lighter than anticipated.
The company explained how individuals who died in 2020 may have otherwise died in the next few years, while increased attention to healthcare and social care as a result of the pandemic could actually improve long-term mortality rates.
Tim Gordon, head of demographic horizons at Aon's Risk Settlement Group, said: "The mortality experienced in 2020 is unlikely to be indicative of longer-term trends. For this reason, we think the CMI’s proposal to place no weight on 2020 data is a sensible ‘least bad’ option.
“COVID-19 is going to cause real problems for all mortality models that project trends, but it’s important that insurers, reinsurers and pension schemes do not overreact to this short-term disruption to their model updates.
“Longevity risk has not gone away. Pension schemes and insurance companies trying to reduce their exposure to longevity risk have the same access to the necessary tools, such as longevity swaps and bulk annuities, as they had before the pandemic. And the risk settlement market pricing remains competitive.”
The deadline for responses to the CMI's consultation is 1 November 2020, with the results expected by mid-December.
Image credit: iStock
Author: Chris Seekings