Projections Life Table AG2022

Since the AG2014 publication a stochastic model is used, so that in addition to the well-known ‘best estimate’ mortality probabilities to set provisions and contribution levels, one can also
allow for the uncertainty in mortality probabilities. Over the course of ten years the
model was improved every two years and before you now lies the publication of
Projections Life Table AG2022 (in short AG2022). Life expectations has often made the
headlines in 2020 and 2021 because of COVID-19. This has prompted the AG Mortality
Research Committee (Commissie Sterfte Onderzoek or ‘CSO’) in 2021 to issue an
intermediate publication on the possible effects of COVID-19 on life expectancy. The
most prominent conclusions were that much is still uncertain and that we can only read
into the tea leaves to predict the long-term effects of COVID-19. Even now there is only a
limited picture of these. There are however developments that give hope, such as a high
vaccination coverage and a milder omicron variant.

CSO and the Projection Life Tables Working Group have investigated and considered a
number of potential COVID-19 scenarios. Over two years COVID-19 dominated the news:
many contaminations, many hospitalisations, postponed care and many more deaths.
The current reality raises the question what the long-term effect of COVID-19 will be and
how to include this effect in the mortality probability forecast and long-term life
expectancy. In this publication we will elaborate extensively on the COVID-19 modelling
approach and the view of the committee.

This publication gives an update of the projections table. The CSO and working group not
only considered the potential effects of COVID-19. Other research was also done,
improving the model in some areas. The committee and working group have heeded the
questions from within the profession that were raised at earlier publication events and
deep dive sessions. Not every line of research has led to a model adjustment. It is
however true to say that a minor adjustment in the model or a model parameter could
lead to substantial changes in provisions and contribution levels.

As ever, the AG Boards extends its gratitude to the committee and the working group for
their unremitting commitment to achieve the best possible life expectancy forecast.
Clearly, given the exceptional circumstances in the past two years, the AG is happy with
the thoroughness of the research and the timely publication of this projections table.