Life Expectancy Hasn’t Fallen Even If The Effect Is Worse On The Poor

ONS tells us that life expectancy has fallen, is falling, this not necessarily being something that’s entirely true. There being good reason to believe that what has actually happened is that we overestimated life expectancy before, the numbers now coming back into realignment with reality.

So, this might not be true at all:

[perfectpullquote align=”full” bordertop=”false” cite=”” link=”” color=”” class=”” size=””] The life expectancy of girls living in the poorest areas of England has fallen by almost 100 days in the past five years, official figures show. The Office of National Statistics (ONS) yesterday released data showing that girls born in deprived areas in 2017 are expected to die more than three months earlier than those born in the exact same areas in 2012. Researchers said that the data revealed “a significant widening in the inequality of life expectancy”. Women in the most deprived areas in England can expect to live for 78.7 years, while women in the least deprived areas can live for 86.2 years – a gap of 7.5 years. In total, the data shows that women born in the most deprived areas of England had the biggest drop in their life expectancy. Between 2012 to 2014 and 2015 to 2017, the life expectancy at birth of women living in the most deprived areas in England fell by 98 days. In comparison, the life expectancy of women living in the least deprived areas increased by 84 days. [/perfectpullquote]

Worth noting that at least part of this is entirely wrong. For we don’t measure and don’t estimate life expectancy based upon place of birth. We do it upon place of death. It’s death certificates we count and we don’t match them up with birth ones. Yes, there’s an importance to this, it’s not mere pendantry. For there’s such a thing as migration. And people who become poor tend to go live in poor areas, people who become rich leave them. And one reason why someone might become poor is ill health.

Which is at least part of the cause of the inequality in life spans across such differing economic areas. Self sorting of those likely to die younger into those poorer areas. I’m inclined to say that that’s all of it but I’ll only insist that it’s some of it.

But to the larger point, the decline. There’s good reason – OK, reasonable reason – to believe that it’s entirely an artefact. That the fall in lifespan is just because we over measured it before:

[perfectpullquote align=”full” bordertop=”false” cite=”” link=”” color=”” class=”” size=””] No, the improvement in mortality is NOT KNOWN because all we do know is that the mortality rates for older ages reported from 2001-2010 are *wrong* – they were based on a calculated number of people alive at different ages and understated the mortality rates by the same %age as the numbers assumed to be alive were overstated. All the “missing people” in the 2011 census are the numbers overstated in the 2001 census plus net emigration of the elderly in the following decade.
All the wonderful improvement in 2001-10 was a mirage due to errors in the 2001 census (you know: the one that said that Westminster had fewer inhabitants, men women and children, in total than it had heads of households paying Council Tax). As the numbers alive gradually shrank, the fixed error became gradually more important so there appeared to be a progressive improvement in mortality thanks to New Labour’s wonderful policies.

Gordon Brown had the cheek to blame Actuaries for underestimating mortality improvements when they had been right and the government-produced figures were wrong. [/perfectpullquote]

Yes, that’s just a comment on a blog making those assertions. But the writer of it has significant experience in the field. And it is true that if we over estimate lifespans in one decade then as people actually die we’ll have to be lowering our estimates. Further, if the problem is the Census then that also explains the differential changes across socioeconomic areas. For, as Polly Toynbee was so insistent about at the time, it’ll be the poor areas which are most miscounted by the Census.