The Continuous Mortality Investigation (CMI) has published a consultation seeking views on proposed changes to its Mortality Projections Model.

Its paper is consulting on proposed methods and core parameters for the next version of the model, following a review that sought to address concerns around the sustainability of the initial approaches taken following the Covid-19 pandemic.

Alongside the usual changes to calibration data, the CMI proposed making four changes as to how it calibrates the model to historical data.

The two primary changes aim to enable the CMI to better reflect mortality trends, with the first being an extension to the existing age-period-cohort improvement (ACPI) model to have three period terms.

This would allow the model to reflect different mortality trends for different age groups, rather than one in previous versions of the model.

The second was changes to model the impact of the Covid-19 pandemic on mortality by adding a ‘fitted overlay’ to the APCI model, rather than using weights to project underlying mortality improvements.

It will use the name ‘age-period-cohort-overlay improvement’ (APCOI) model for the new version of the APCI model that includes the overlay.

The other two changes were not the focus of the review, but the CMI deemed them as necessary so the APCOI model with three period terms and fitted overlay “produces reasonable and reliable results”.

These were applying existing cohort constraints at low and high ages after fitting the APCOI model, rather than as part of the fitting process, and changing the algorithm used for the APCOI model to be quicker and more robust.

The consultation is open to users of the model until 25 March 2025. The model is used by nearly all defined benefit scheme trustees for assessing funding positions, and by sponsors to report on their schemes’ financial positions.

Commenting on the consultation launch, CMI Mortality Projections Committee chair, Cobus Daneel, said: “The Covid-19 pandemic was challenging for all mortality projection models.

"It was uncertain how mortality rates would react beyond the initial shock and the CMI took a pragmatic approach, avoiding material revisions in projected life expectancies until a clearer picture emerged.

"We excluded data from the peak of the pandemic, as these rates were unlikely to be a good guide to the future, and smoothed through data for other years, focussing on a plausible level of mortality rates in the medium term.

“That approach worked well initially, but required subjective judgements regarding how much emphasis should be placed on recent data.

“Now that mortality at pensioner ages is closer to pre-pandemic levels, we have an opportunity to improve the CMI Model, providing a better fit to recent data across the full age range during and after the pandemic.

“We recognise that different users of the model have different views. The CMI Model provides a framework for them to adjust parameters to reflect their own portfolios and views on the impact of the pandemic, and to communicate those to others.”

LCP partner and longevity expert, Ben Rees, added: “We welcome these proposed changes to the model.

“In our view, it is a positive step forward to reflect both the contrasting trends in mortality we have seen between younger and older generations, and to incorporate an explicit and transparent allowance for the impact of the pandemic and its aftermath.”

---