Information on life expectancy predictions is essential to pension modelling as a guide to the duration that pensions will be claimed. Recently, in many OECD countries life expectancy data has been used to determine future retirement ages and its consideration in any analysis is worthwhile.
Information is available at 2015‑20 and is calculated at two different points in time, namely, at birth and age 65. The latter is obviously conditional upon surviving to age 65 but gives a greater indication of the duration of pension receipt. In addition to this life expectancy estimates for 2060‑65 are also provided, again dependent on survival to age 65 initially. These are particularly relevant when considering the ageing of the population.
The graphs below for all three indicators show that the life expectancy in the non-OECD Asian economies is well below that of the OECD countries.
The first graph covering life expectancy at birth indicates that the average life expectancy for men and is below 70 years of age in India, Indonesia, Pakistan and the Philippines. Considering that the normal pension age in the Philippines is 65 years then this shows that the average duration for pension receipt is actually only one year for men. In contrast life expectancy in Sri Lanka for women is 29 years higher than the normal retirement age of 50.
The general trend is that the non-OECD economies represented are all at the lower end of the scale with male life expectancy generally under 73 years, compared to the OECD average of 78 years, and it is over 81 years for Australia and Italy, with Hong Kong, China and Singapore also above 81. It is also noticeable that there is quite a difference in the gender gaps of life expectancy, with women as expected always being higher. However women are only expected to live 2.1 years longer in Pakistan, whereas the difference in Viet Nam is 9.2 years. For the OECD as a whole the difference is 5.2 years with most countries having at least a four-year difference.
The second graph takes the analysis one step further as survival to age 65 is now assumed, which is the maximum normal retirement age for all but a handful of economies. Therefore the findings at this age level provide more accurate estimates of average duration for pension claimants than the previous indicators. The difference in life expectancy estimates between the highest and the lowest has reduced to six years for men and ten years for women. Although the non-OECD economies are again at the lower end the results for men are all within four years of the OECD average.
The second part of this graph is the forecast information based on 2060‑65 estimates from the United Nations population database. This clearly shows that the trend in the future will be for the life expectancy gap to decrease for all non-OECD economies in comparison to the OECD average. Whilst life expectancy for the OECD countries continues to increase, it is at a much slower rate than the Asian economies. With retirement ages currently being well below 65 in many of these economies the pressures on the pension system will only increase.
Although life expectancy results are a useful tool in pension analysis they have limited use when used alone. They can provide estimates of average duration of pension receipt for those that actually begin claiming a pension, but they do not assist with providing any information about the actual numbers involved. For this population projections are required and this will now be covered in the following section.