Working out how long you will live

Russell Hutchinson and his team review life expectancy calculators and find there is an opportunity awaiting in this space.

Tuesday, May 9th 2023, 8:15AM

by Russell Hutchinson

In June last year we had one of our data science people, Ed Foster, take a look at life expectancy calculators. These calculators can be useful tools for conversations about both life and health risks and investment risks. This is because it can be awkward to our loved ones if we die ‘too soon’ and awkward for us if we don’t have enough saved to survive on for a long time in retirement – a problem often described as ‘living too long’, although few would agree with the term.

Our life expectancy is a topic which piques the interest of us. Insurers and many wellbeing schemes offer calculators to enable clients to estimate what age we can hope to live to as a way of building engagement around the fact of mortality and the question of what we can do to improve our health.

There are numerous calculators in play, all requiring varying degrees of user input and level of detail. We reviewed a dozen of them, but will bring to your attention just three: one offered by Fisher Funds, two options from Statistics New Zealand, plus we added the real-age calculator, which does nearly the same job, thrown in to the set as well.

All the calculators vary considerably in the number and type of questions asked. For example, the Stats NZ calculator essentially provides an average simply for current age and gender for a New Zealander, which we felt was a good baseline from which to work.

It gave Ed 88 years.

Fisher Funds was a similarly simple calculator – not venturing into the dangerous shoals of considering diet, exercise, and medical history, to attempt a more accurate number. Curiously, Statistics New Zealand life tables use a simple look up approach and no specialised questions and arrive at slightly lower numbers (which we felt were more accurate).

Many off-shore calculators ask more questions and arrive at slightly lower answers. Although interesting because of the data they consider, we won’t worry about them here because they are using different baseline mortality.

The United States, for example, where there are many such calculators has lower overall life expectancy due to poor health insurance coverage, high levels of gun violence, and the spreading opioid epidemic. But from that analysis we were able to identify that, broadly, more questions tended to yield better results in terms of accuracy.

Of particular interest is the AIA Vitality life calculation which gives us a biological age to set against our chronological age. This is similar to a life expectancy calculator but brings the point back to the present – telling you whether you are ‘old’ or ‘young’ for your chronological age.

We found that answering “I don’t know” to several questions led to worse results, when we’ve been able to supply the data, they improve. This suggests the model assumes the worst if you cannot answer these details. There are also limited family history questions with the onus on the individual and indeed the present.

With AIA’s calculator being offered as a ‘real-age’ type of tool within the AIA Vitality program, which requires a subscription, there is probably a gap in the market for an insurer, or come to think of it, an adviser, to offer a calculator of their own.

I would like to see one to build engagement, by offering good, New Zealand-focused, free, public calculator.

Tags: Opinion