New CMI model will lead to the first significant increase in pensioner life expectancies

In this blog, we unpack the new CMI_2024 model, set out what has changed and what it will mean for you. 

As part of the most in-depth review for several years, the CMI has published its latest Mortality Projections Model, CMI_2024. This release is significant as the model is used by almost all trustees of defined benefit pension schemes for assessing the funding position of their schemes, and for sponsors to report their schemes’ financial position on their balance sheets. The model is also a key reference point in the pension risk transfer market. 

Many trustees and corporates have become accustomed to seeing successive CMI models reduce life expectancies – and indeed, this has been the case for the previous 10 years. However, for the first time, the new model may lead to a significant increase in liabilities for many schemes if they choose to use the core model “out-of-the-box”.  

The core version of the model results in a notable increase in life expectancies at age 65 of around 3 months for males, and a modest increase of 2 weeks for females. 

How long members live in practice will be influenced by real-world drivers rather than the results of an actuarial model. For example, it remains to be seen how quickly the NHS will recover from its current position and the extent to which developments, such as anti-obesity medications, will improve future mortality. We encourage trustees and corporates to consider whether the changes in life expectancies produced by the core CMI model are appropriate for their circumstances. Those who independently form an alternate view on future mortality can take advantage of the CMI model’s flexibility to express that view. 

Key changes in the CMI_2024 mortality model 

The new version includes two key changes from previous versions of the model, intended to ensure the model better reflects the real world: 

• The model now directly incorporates the impact of the pandemic on mortality rates, whereas previous versions essentially smoothed through the distortion produced by the pandemic to project underlying mortality.  
• The model now projects different mortality trends for different age groups. This better reflects recent trends where mortality has improved more rapidly for pensioners than for people of working age.  

Whilst the key principles of these changes have remained consistent since the consultation in February, the CMI have made a couple of additional technical changes to the model, causing a further increase in life expectancies for those who use the core model. 

Impact of CMI_2024 on pension scheme insurance pricing 

For many pension schemes, the most significant impact of a new CMI model is if it leads to changes in the affordability of insurance. 

It is worth saying that the insurers and reinsurers active in the pension risk transfer market typically do not rely on the core CMI model to set their view of the future. Instead, they independently derive a view, taking into consideration factors such as how the NHS may rebound or how specific causes of death may progress. 

That said, every day is a school day! The CMI consultation gave all market participants the opportunity to see the views of their peers on different aspects of how the CMI model operates. It is not inconceivable that some of these ideas could be brought into insurers’ own models. The impact that this causes will depend on the specific changes being incorporated. 

We believe that any movement from the insurance market is likely to be more muted than the changes in the core model. This means that, for schemes that move with the core version of the new model, it is likely that insurance will appear relatively better value. 

How insurers may respond to CMI_2024 

Like many insurers, at LCP we form our own view of the future, leveraging the insights of our Health Analytics colleagues and our actuarial expertise. From here, we calibrate the CMI model to reflect that view.  

For example, I recently wrote a blog about how our view of old-age mortality projections differs considerably from the core CMI model. 

What CMI_2024 means for schemes intending to retain longevity risk and run on? 

Longevity risk is very different to other risks posed by DB pension schemes. The actual risk of members dying at a different rate than assumed pans out over many decades. 

The more immediate concern is the risk that new data or learnings cause actuaries to reassess what the appropriate value of liabilities should be, leading to their clients taking different decisions.

CMI_2024 is an excellent example of this risk materialising. The new core model suggests that life expectancies are now around 1% higher than they were under the previous core model. As such, overnight, trustees and corporates that simply adopt the core CMI model may find that their funding level has fallen by around 1%. 

One way of mitigating this risk (and so reducing your dependence on the core CMI model) would be to form your own view of the future, based on advice that reflects your own scheme’s characteristics. Please speak to one of LCP’s mortality experts, or one of your usual contacts, if you would like to explore this option and reduce your dependence on future changes to the CMI model.