Menu Close

Hard Evidence: can we afford an ageing population?

Old people aren’t museum relics. Mr Push

Stories about “population ageing” often have a number of things in common – it is bad, it is new, and it will overwhelm us all. The major fear is a burden of cost and caring that more older people will create.

“Population ageing” has been taking place for almost two centuries in the UK and kicked in about half a century ago in developing countries. It starts whenever decent sanitation systems, diet and public health measures start to reduce very high levels of infant mortality.

The fastest increases in life expectancy actually occurred in the UK between about 1850 and 1950, and contrary to the idea that population ageing particularly affects rich countries, it is proceeding faster in developing nations as life expectancies catch up with the levels achieved in the west.

Population ageing, then, is nothing new. And we have been well able to “afford” ageing so far because so many of its effects are positive. Imagine you make widgets. You can choose between (a) machines that are rather more expensive, but much more versatile, and highly reliable for 40 years or more, and (b) cheaper machines which often fail before they’ve even been delivered, and have a rather unpredictable lifespan: some last 20 years, others soldier on until 50 or even 60, although much after 40 they are pretty worn out.

Machines are not people, but we only need to reflect on the social and economic cost of high fertility and mortality to see why population ageing is so positive. Lower mortality is a tremendous economic bonus for the simple reason that it reduces the amount of social effort devoted to reproduction.

False notions of dependency

So long sucker. Amsterdamised

We’ve got to watch how “old” is defined. Putting aside the argument that getting old is a good thing – we all want to live longer don’t we? – and that we’re actually getting younger (if you count the years left we actually have to live), over time we can expect people of successive generations to be healthier and fitter and have longer to live at any age than their predecessors.

Sweden was one of the first countries where the majority of a generation (that born in 1818) lived to 50 years old. Spain was almost a century behind (in 1903). An age in years that we might think of as “old” now may not be in 50 years time.

Most old people are not “dependent”. For the first time, a million people aged 65 or over are still in paid work. Indeed if we define dependency as “not in paid work” then there are more dependents of “working age” in the UK than there are people over state pension age who don’t work.

A great many older people report their health as good. And at the time of the 2011 census there were about 300,000 people aged over 64 in care homes (including public and private, with or without nursing) – just over half a percent of the population of England and Wales. Many are affluent and support younger family members. Grandparents are the most important source of childcare after parents themselves; more important than either public or private childcare.

Increase in older population providing unpaid care. ONS

We confuse old with poor

Those who are poor are rarely poor because they are old: they are poor now because they were poor when they were younger, unable either to accumulate assets or pension rights to draw on in later life.

We can get a sense of trends in the potential burden of dependency by looking at the ratio of the numbers of those in employment (and so who pay tax or social insurance) and the numbers of the population who have an average life expectancy of 15 years or less – what we call the Real Elderly Dependency Ratio.

Real Elderly Dependency Ratio and Old Age Dependency Ratio Selected Countries 1950 to 2010. Author’s calculations. Human Mortality Database

As the figure above shows, this ratio has been falling in the UK and the US for 30 years or more. The only affluent country where this ratio has been steadily rising is Japan. And the reason for this is not the longevity of the Japanese but Japan’s snail-like progress on gender equality (so that very few mothers work) and its refusal to accept immigration (which keeps the population younger by drawing in workers from developing countries with higher fertility rates).

The commonly used Old Age Dependency Ratio (also shown in the figure above), which simply compares the number of people aged over 65 with numbers aged 16 to 64, has been rising steadily, but tells us nothing about either “dependency” or how many people are actually working. All it tells us is something we already know: that fertility and mortality are falling.

We can’t afford pensions!

A little over a century ago, annual working hours in the UK were double what they are now (at around 3,000 hours per year). Working lives were also longer: boys and girls might leave school at 12 or 13, whereas now employment rates don’t peak till the mid-20s as students leave higher education, and most retire from work before reaching the state pension age.

Over that century, trend growth in productivity has nevertheless delivered phenomenal increases in living standards. There is no economic reason, apart from the failure of political will and imagination, why continued economic growth cannot continue to fund decreases in the relative length of working lives just as it funded the shortening of the working day, week and year. And the progress towards greater gender equality in employment and the dismantling of the old “male breadwinner” system has actually increased the length of many women’s working lives.

Only the affluent will retire early

There is a yet more pernicious aspect to the state pension age rise. Only those who have not accumulated the assets necessary to choose retirement will be forced to work on.

The affluent will continue to be able to afford to retire early. Worst of all, it is those most likely to end up working up to the state pension age who will be least likely to survive to enjoy a long retirement. The poor, those in manual jobs, or living in areas of social deprivation, have life expectancies five to ten years below their more privileged peers.

None of this means that older people shouldn’t be encouraged to work longer – if that is what they choose to do. However, to suggest that an economy as productive as that of the UK “cannot afford” to let its least affluent members leave work till they are 68 or 70 is grotesque. This is not about a policy driven by economic or demographic pressure but the slow political asphyxiation of the welfare state.


Definitions vary across time and place (at what point does the general weakening of cognitive function that accompanies “normal” ageing cross the threshold to dementia?) but there is a close connection to age and “early onset” dementia is very rare.

As life expectancy increases we could expect many more older people to live with, and die with, dementia. If it is mostly mild and compatible with independent living it has few implications. If it is severe it has the potential to rapidly increase demand for social and healthcare.

Dementia charities have done an excellent job of summoning up bleak scenarios in order to persuade governments to fund more research. The G8 recently held the Dementia Summit in London where leading nations have committed to developing a cure or treatment for dementia by 2025, while our own government’s Department of Health has since 2009 its own national dementia strategy that aims to “achieve better awareness of dementia, early diagnosis and high quality treatment”. However, these aren’t necessarily sober estimates. For example, a recent test on the 880,000 estimated 20 years ago to have dementia by now suggested the figure was closer to 670,000 – a fall of nearly a quarter.

What we really do not know, because the evidence isn’t good enough, is how the relationship between age and dementia prevalence is changing. The dream scenario is that longer lives also mean later dementia, the nightmare one is that longer lives come with a fixed relationship between age and dementia, so that a rapidly increasing proportion of the extra years in longer lives are spent with the condition. We do know of factors that appear to delay the age at dementia onset, including more mental, physical, or social activity: something vibrant communities can provide. We know of no good evidence for the nightmare scenario, and with political will, we can surely avoid it.

Hard Evidence is a series of articles in which academics use research evidence to tackle the trickiest public policy questions.

Want to write?

Write an article and join a growing community of more than 175,100 academics and researchers from 4,816 institutions.

Register now