New figures out from ONS means that the newspapers have an opportunity to point to the inequality of lifespan in Britain:

A baby born in the North West of England will live on average two years less than a child born in the South East, new Government figures have revealed. The figures give a snapshot of life in Britain today and reveal the divisions between life-expectancy rates for people living in different areas of Britain.

The figures themselves are correct, average life expectancy does vary around the country and across socio-economic classes. But the interpretation put on them is not correct. For no one is measuring the life expectancy of someone born in a particular place. They are measuring the age of death of people in that specific place. The error can be seen in this second story inspired by the same ONS numbers:

Eastbourne has become the first place in the country to boast a population with an average age of more than 70. The Meads district of the famously genteel East Sussex town was identified by the Office for National Statistics as having the oldest residents in England and Wales. Named by officials as Eastbourne 012B, the well-heeled area has a population with an average age of 71.1, compared with the national average of 39.7.

We do not believe that all people born in Eastbourne have exceptionally long lives. Quite the contrary, we believe that people who live long enough to move when they retire go to Eastbourne. Which of course means that people who live long enough to move when they retire must move away from some other part of the country, lowering the observed age at death in those places.

A goodly part of the inequality of lifespans is simply that people move around. And we have parts of the country where older people preferentially move to. Thus, inevitably, we end up with concentrations of the long lived in some paces and a relative paucity of them in others.

Try this for an extreme example of the same phenomenon. Measure the average age of death in a children's hospice against that in a home for those with senile dementia. We will see a very great difference in average life span, a vast inequality. One that will tell us absolutely nothing at all about the average life expectancy of those people when they were born nor of the people who were born in the same area.

Geographic measurement of lifespan inequality reflects where people die and at what age, not the potential lifespans of those born in those places: for people do move around.