Skip to main content

The Ghost of Christmas yet to come!

This month saw the release of the latest national life tables from the Office for National Statistics (or ONS). Blaise Smith, Partner reviews and…

  1. This month saw the release of the latest national life tables from the Office for National Statistics (or ONS). They show period life expectancy in the UK (and its constituent parts) for the years 2016/18.
  2. The previous ONS life tables published in September 2018 were for the period 2014/16.
  3. It is important to understand that each period shows data from three consecutive years, the latest being 2016, 2017 and 2018.
  4. The previous tables showed data for the years 2014, 2015 and 2016.
  5. It will be noted that there is an overlap of data in that the year 2016 appears in both data sets.
  6. The ONS statistical bulletin therefore offers a comparison between 2016/18 and 2013/15 as they are the latest studies with no overlap of data.
  7. The publication of the 2014/16 data in September 2018 received widespread media attention. This was because for the first time in history, there had been no reported improvement in life expectancy at birth in the UK, remaining as it did at 79.2 years for a male and 82.9 years for a female.
  8. The question last year was whether the stalling in life expectancy was just a statistical blip in a trend of continuous improvement or the start of a new long term trend of life expectancy flattening out or even reducing. So what do this year’s tables tell us about that?
  9. The headline figures (based upon the latest historical data without any projection for future improvement) are 79.3 years for a male and 82.9 years for a female. So in broad terms, it is unchanged.
  10. The ONS statistical bulletin compares 2016/18 data with 2013/15 and concludes that a +0.2% improvement has dropped to +0.1%, indicating that the slowdown reported last year is continuing. Both sets of tables are historical data about people who have died.
  11. In addition, the ONS also provides so-called “cohort” tables which are based on assumptions as to future improvements in mortality. In Dickensian terms, the dead represent the ghosts of Christmas past and the projected figures are for the ghosts of Christmas yet to come.
  12. As regular readers will know, the courts assess damages for future loss by reference to actuarial tables prepared by the Government Actuary’s Department (the “Ogden Tables”). The Ogden Tables are in their 7th edition published in 2012 with an 8th on their way. They in turn are based upon the ONS life tables. Ogden Tables 1-26 currently adopt the “cohort” ONS tables from 2008 which assume a continuous increase in life expectancy for the general population of the UK.
  13. It follows from the latest historical data that earlier cohort (or projected) life expectancy tables can no longer be considered a reliable indicator of future trends. The latest ONS release shows that the 2008 cohort tables, adopted by the Ogden working party in 2012, were too optimistic.
  14. The 2016/18 ONS period tables show that the figures released last year can no longer be dismissed as just a “blip”. As the ONS have themselves commented, the slowdown which has been observed since 2011 is continuing with much lower increases than were previously anticipated.
  15. We have a court system for calculating future loss based upon historical data and projected life expectancy. When the life expectancy data changes, so must the assumptions for projecting future mortality.
  16. The 2018 ONS cohort tables project life expectancy at birth for a new born male as 87.6 and for a female 90.2 years. The corresponding figures in the current edition of the Ogden tables, published in 2012 and based upon the 2008 cohort tables, are 88.96 and 92.57 years respectively.
  17. The ONS also provide period tables based upon data from single years, but say they are not as statistically reliable as for a three-year period. The latest single year tables show male life expectancy at birth at 79.24 in 2018 compared with 79.28 in 2017 and 79.25 in 2014. This suggests there has been no observable change over a five-year period.
  18. One reason for adopting data for a three-year period instead is to avoid such things as the effect of a flu outbreak. However, the year of overlap between the latest two three-year periods was the year 2016. That year saw a high take up of flu vaccines and only a moderate level of influenza activity in the UK (see Public Health England: “Surveillance of influenza and other respiratory viruses in the UK, Winter 2016/17”). It is difficult to see how 2016 can be blamed as a “blip” which infects studies for both 2014/16 and 2016/18. 
  19. The 2018 ONS “cohort” UK life tables assume a target mortality improvement rate of 1.2% pa from 2043 (i.e. after 25 years). There may be all sorts of reasons why that assumption might still be correct, but one of them is not the current observed trend in the age at which people are dying.
  20. The ONS rationalised its approach to its principal projections in 2018 on the basis that long term trends should take precedence over recent changes. It did, however, offer a “low variant” of 0.1% pa improvement for both sexes as an alternative to both its principal and high projections. Perhaps that should now be regarded as the more correct approach (giving projected life expectancy at birth of 80.8 for a male and 84 for a female) until such time as the period data changes again.
  21. It is beyond the scope of this note to consider the reasons for the slowdown in improved life expectancy. It is no doubt a multifactorial issue. It might include the tailing off of improvements from changed smoking habits coinciding with rising levels of obesity (see “Prevalence of obesity ages 18+,2010-14, World Health Organisation,2017”). Social and wealth factors must also play a part as indicated by the ONS breakdown of its own data showing that gains in life expectancy in the South of England have been matched by corresponding losses in the North.
  22. As well as illustrating that there are huge regional inequalities within the UK, the latest data also shows that whilst we are living a tiny bit longer, we are surviving with extra years of poor health which are greater than any gains in the years of good health. Perhaps, after all, it is not the years in your life which count as much as the life in your years.


  • The stalling of the trend of continuous improvement in life expectancy in the UK, first observed from 2011, has now been confirmed by the latest ONS release. It shows that life expectancy has continued flattening out over the five-year period from 2014 to 2018. This can no longer be dismissed as just a statistical blip.
  • Projected mortality tables will need reviewing once again as will the Ogden Tables.

And finally

The outlook for a Tiny Tim Cratchit born in 1841 (two years before publication of A Christmas Carol) would have been 40.2 years at birth. Today it has more than doubled to 80.8 or 87.6 years depending on which variant of projected life expectancy we choose to adopt. But Tiny Tim “bore a little crutch, and had his limbs supported by an iron frame”. Whether those living with disabilities will enjoy the same improvements in life expectancy as the rest of the population is a story for another day (“Long-term survival after traumatic spinal cord injury: a 70-year British Study” (Savic and others, 2017).

If you have any questions or would like to more know about our update, please contact Blaise Smith, Partner, on 0113 213 4056, or

Share on Twitter