In last weeks Newsletter I talked about a new form of pension. Setting state retirement age based on Remaining Life Expectancy (RLE). This was consistent with the Four Ages model. It had been adopted in a number of countries including Denmark. Their model assumes that the pension age will track life expectancy. It will provide 14.5 years of pension.(Newsletter #255 The Fourth Age) It appears that the Danish Government is now considering revising the model.

Alternative Approaches To Setting a Retirement Age.

Governments have used different approaches based on the objectives they chose. Setting pensionable age based on a fixed number of years clearly favours government affordability. Another “affordability” model is based on share of GDP. The cost of providing the pension is computed as a percentage of GDP. In the UK today it is 5.1%. That number is growing in countries around the world. Forecasts are constructed of pension costs and GDP. The state pensionable age is set to maintain a target share of GDP (say 5%).

An alternative approach uses the dependency ratio. The number of people claiming state pension relative to the number of working people. In almost all country pensions are paid out of current taxation. Each generation pays for the pension of their “parents”. This model sets retirement age to maintain a fixed dependency ratio. This seems fairer to the younger generation. If not as life expectancy increases the costs to the working population would go up.

In the UK the retirement age is currently set to maintain a fixed share of life with a pension. Retirement age is set based on a recipient having a pension for 31% of their lives. It is worth pointing out that this has not always been the case. When my father was born in 1918 the state retirement age was 70. His life expectancy at birth was 58. He died at 75, having claimed a pension at 65. With a 31% target he should have retired at 52.

This model is in answer to inter- generational fairness. A model that uses a fixed number of years means that future generations would have a smaller share of their life with a pension. This is the problem that Denmark is wrestling with.

The Communication Problem

There is an additional issue when setting retirement age. There has to be a long lead- time to allow people to adjust their plans. Whatever changes are made they take many years to implement fairly. This complicates the decision. It makes forecasting a real issue. It puts an emphasis on the ease of communication of the rules and their logic.

There is still a debate in the UK about a decision to equalize pensionable age for men and women. In 1995 the Government decreed that women born before 1950 would retain a pensionable age of 60. Those born after 1955 would have a pensionable age of 65. This change was to be made gradually and not to be completed until 2018. Women are still campaigning for compensation because of a claimed lack of communication.

All Proposals Have Issues.

There are common issues. The forecasting of cohort life expectancy is an imprecise art. If the life expectancy turns out to be wrong, it can influence affordability. It also impacts individual fairness. Forecasting GDP is as, if not more, difficult.

The Age Model assumes implicitly that the third age of life extends. When used in the context of pensions, it assumes that until the onset of disability and frailty people are capable of working. Using the existing Danish model, citizens will not have a pension, so how else are they to live? If “70 is the New 60” then it assumes that at 70 individuals are perfectly capable of working. This takes no account of the nature of their work. A manual labourer at 60 may not be able to carry on working. It also assumes that the jobs are available for older people to work.

All models respond to increases in life expectancy. However using different models produces different trajectories for retirement age and costs. All show the need to increase the UK state retirement age. All show a massive increase in the costs. ( The same logic applies in most developed economies).

Setting a 30% “share of life” target means that a retirement age of 70 is needed by 2062. Setting a fixed “20 years of pension” means 70 arrives around 2066. Fixing pensions at 5% of GDP means that retirement at 70 would occur in 2033. Some of these dates may seem a long time away but will impact generations that have already been born.

The current costs of the UK state pension are around £115Bn. That could rise to £382Bn by 2070 with a “share of life” of 30%. A “twenty years of pension” target comes out about the same number. It can go lower using a cap on share of GDP or a tight dependency ratio.

It Is Not Fair In Any Case

A State Pension is a blunt instrument. It takes no account of the life of individual citizens. Life expectancy is not homogeneous. There is a 12-year difference between those living in the most and least deprived areas of the UK. There are regional differences that go beyond deprivation. The UK state pension is universal.

Individuals pay into their pensions in the UK through “National Insurance”. They pay directly. They also pay indirectly through contributions from their employers. The return on their money paid depends on how much they take out as pension over the rest of their life. That depends on life expectancy and the state retirement age.

There is no computable, rational answer. It depends on your objectives. They depend on the politicians!