Ageing: the future is affordable

An ageing population is great news for individuals. We are living longer, healthier lives and enjoying greater prosperity with each year that passes – and just consider the alternative.

Yet there is a paradox. Even as we reap the benefits of longevity and vitality, we are becoming more anxious about the social and economic effects of ageing upon society. Demographics has turned from a peripheral issue into a major source of concern in Britain and just about every other Western country. A host of topics, including pension resources, falling fertility, projections of increased life expectancy, and the perceived health and long-term care needs of older people, have recently focused attention upon the consequences of our ageing societies. What unites the bulk of contributions to these discussions is the notion that an ageing future is something we should be worried about. As a recent editorial about retirement in The Economist put it: ‘Alas, these golden years come at an unaffordable price.’ (1)

We are told we need to confront some pretty big questions. Can society cope with having so many more old people? Can we really afford our future? In our fatalistic times, these questions have become rhetorical. The self-evident answer, it is assumed, is that we should expect fairly miserable times ahead, and we should abandon any lingering vestiges of high hopes that we may have been harbouring for the future.

But just because the mood of social pessimism is so ubiquitous does not mean we should simply accept it. If we address these questions seriously we find that we can cope with, and afford, an ageing society – and that it is something we should welcome as a challenge, rather than see as a problem from which we should shrink.

• There’s nothing new about ageing

While the pessimists fret about the problems to come as if ageing is some new challenge, they ignore the fact that this has been a remarkably constant feature of industrialised countries since the latter part of the nineteenth century, and societies have had no difficulties ‘coping’ with it in the past.

The underlying reasons have changed a little. In previous eras, falling fertility (much bigger than the falls we have seen recently) and fewer early deaths were the dominant influences, resulting in smaller cohorts of younger people, more of whom lived to be old. Additional factors today are the particular cohort effect of the ageing ‘baby boom’ generation, born in the 1950s and early 1960s, and greater life expectancy for the elderly, which is linked to wider health improvements since the Second World War. However, over the past century as a whole, the overall trend in rising life expectancy and in the increasing average age of the population has been pretty constant; and the young and the old alike have gained enormously in prosperity. Ageing has clearly not been unaffordable.

For most of this time, nobody made much of a deal about rising life expectancy. Potential demographic problems in the developed world were discussed only on rare occasions, and were never the cause of widespread public concern. At the end of the Second World War, for example, a British Royal Commission pondered the potentially negative effects of falling fertility. But the commission’s findings were not even discussed in full by Parliament, and its concerns were lost amid the return of political and economic stability and growth by the 1950s. The subsequent unanticipated baby boom, of course, was a salutary refutation of the commission’s demographic determinism.

Population ageing was a barely visible backdrop to the steady and welcome progress of society over most of the twentieth century. It was normal to expect your children’s generation to live longer than your own, just as you expected them to be better off. Now we take for granted that family life is no longer frequently interrupted by the death of the young. One hundred years ago in Britain, one in four children died before the age of five; today the figure is less than one in 100. One hundred years ago, less than one third of us got to be 65; today five out of six of us can expect to.

One consequence of these changes is that individuals have become more sensitive to death. The fact that premature death is more unusual for the middle aged, and especially for young people, makes it more of a jolt when people are taken from us early. But that is a tiny price to pay for the joyful gain of significantly greater longevity for most of us. It is a sign of the warped character of our contemporary climate of fear and anxiety that we have to remind ourselves that ageing is overwhelmingly a cause for social celebration.

• The parts don’t add up to a problem

We should note that all the specific subjects that have recently elevated the profile of ageing – pensions, health, fertility, longevity, and so on – are not inexorably problematic. While some of these issues are important topics for consideration and discussion, the problems that they may throw up are not primarily demographic in content, and they are open to practical ‘affordable’ solutions regardless of the expected demographic trends.

Without question, pensions have risen to be the biggest source of demographic concern in Britain and across most of Europe. There is some good reason for concern, but the demographic element is exaggerated. On present policies and trends, both state and private schemes are going to give us much less, both in amount and possibly duration, than many of us had been anticipating. But while demographics is invariably given as the ‘objective’ justification for these cutbacks, this is not the main reason.

State pension schemes have become caught up by the wider agenda for narrowing public spending and shifting the balance between the public and the private sectors. Private pensions have certainly been buffeted by the ups and downs of the stock market, where most of these funds are invested. However, the real problems derived from a series of company and government actions taken during the 1990s, cushioned by the belief that equities would boom forever – such as the taxing of pension funds since 1997, and companies taking contribution holidays and/or using the funds to finance redundancies under the euphemism of early retirement. Now the anxious custodians of these funds seem to see only one way forward – reducing the liabilities by finding ways to pay out less (see Old misers, by Phil Mullan).

The additional healthcare needs of an ageing population seem like another self-evident financial burden of an older population. But this is an unnecessarily bleak perspective. Although on average, healthcare costs rise with age, the financial consequences of this trend could be powerfully offset by the way all people, young and old, are getting healthier. And we are likely to be healthier still over the next 50 years, as a result of growing up with the better nutrition and living conditions of the past 50 years.

Because the trend is towards the compression of illness into the later years of life, a growing share of a person’s need for health treatment is being concentrated into their final months. Relatively more of the resources devoted to treating and looking after ill people is now consumed in the final period leading up to death, than it is in getting us back on our feet in earlier parts of our lives, and this trend looks like continuing. Therefore it is the cost of dying, rather than the cost of living, that is becoming the pertinent issue for health expenditure. However, there is evidence that the cost of dying decreases with age – and however old we are when we die, we only die once. Therefore the aggregate for society of this more important ‘cost of dying’ is likely to fall with an older population. All these health trends at least redress, and potentially could absolutely reduce, the total real cost of treating morbidity over the course of a person’s life.

Unfortunately, it is impossible to illustrate this with health spending figures, because the morbidity features of an older population are inevitably distorted by the society-wide trends in health – in particular, the obsession with health and the phenomenon of the ‘worried well’, which persist despite the fact that many objective measures show us to be healthier than ever. This health paradox is doubtless going to get worse as the focus of government health policy continues to shift from curing ill health to campaigns designed to promote ‘healthy living’ (see A New Year prescription, by Dr Michael Fitzpatrick). Dealing with real morbidity, or even providing an improved heath service with faster and better treatment, are no longer the main determinants of health spending levels: the financial effects of medicalising more and more social issues tends to swamp everything else, including the health costs of ageing.

Falling fertility has also become a major source of demographic concern. It is well recognised that in most industrialised countries birth rates are falling, and are now frequently below replacement levels. If they stayed at this level then populations would get smaller as well as older (2). But the valid statistical contribution that the falling birth-rate does make to population ageing in itself proves nothing about the affordability of ageing.

The demographic scaremongering about the falling birth-rate distracts attention from the more constructive discussions society could be having about fertility levels. For example, what does falling fertility tell us about wider social trends and the greater difficulties people seem to have in making long-term personal commitments? Why can’t we learn the lessons about the need for decent and extensive social childcare facilities when we contrast low-fertility, low-female-employment Italy, with high-fertility, high-female-employment Sweden? While panic focuses simply on the imbalance between pensioners and babies, these more useful debates are all but ignored.

At the other end of life, projections that we can expect to live longer have become a focus for demographic anxieties. At the end of 2003, the regular biennial review of the UK Government Actuary’s Department made a sharp upward revision to forecasts of life expectancy for the elderly over the next 40 years. By 2041, a man aged 65 will expect to live to about 86, up from about 84 in the department’s previous forecast. This increase derives partly from the relatively better health of the average baby boomer, and it follows the established assumptions of many demographers and gerontologists.

This is surely good news…unless you were looking forward to dying. Yet the revision was immediately reported as another example of the problem of ageing. ‘Longer life forecasts raise fears over costs’ headlined the Financial Times on 7 January 2004, while the Institute and Faculty of Actuaries was reported as warning that ‘increased life expectancy poses as great a threat to the solvency of pension schemes as falling stock markets’ (3). In reality, however, the fact that we will live longer does not outstrip the other problems for pension schemes that have been created by policy. Nor does this real demographic fact tell us anything about the basic affordability of social ageing – let alone about the affordability of pensions. Let’s look at the facts.

• Who’s supporting whom?

All assertions and presumptions about the potential unaffordability of future populations ultimately come down to just one statistic – the changing support ratio. Also known as its inverse, the dependency ratio, this measures the relationship between the numbers of working and retired people. Whatever other assumptions you may make about pension reforms, about the healthcare and long-term care costs of a greater number of older people, or about shifts in fertility rates and longevity, the scaremongers’ argument is that none of this can impact significantly on this one measure. This is their trump card. A large part of the population who will be around in 40 or even 50 years time has been born – or, as they wittily say, ‘cannot be unborn’ – so the parameters of what will happen to the population structure are broadly fixed.

These figures cannot lie, we are told – often by the same people who are happy to express their sceptical bent on other issues by quoting Mark Twain’s aphorism on ‘lies, damned lies and statistics’. Over the next 40 years we can expect the numbers of people over the state pension age to rise significantly – in Britain’s case by around 5.5million, about half as many again as there are today – and the age balance will shift in an upward direction.

The support ratio is defined as the number of working-age people for each elderly dependant, with ‘working age’ usually taken as everyone aged between 16 and the state pension age, and ‘elderly dependants’ as anyone above this age. In Britain, on current population projections, this ratio falls from about 3.4 today to a trough of about 2.4 in 40 years time. The exact ratios forecast vary between Western countries, but the shorthand average often used is a fall in the ratio from four to two – meaning that each pensioner will be supported by half the number of ‘workers’ as pensioners are today. It is this shift that leads many to assume trouble ahead. As Frances Cairncross, senior editor at The Economist, writes: ‘In a decade’s time, many countries thus start to face a huge problem: how to support a vastly larger population of old folk.’ (4)

Adair Turner, chair of the government-established Independent Pensions Commission, provides an example of how policymakers view this apparent problem, and how they propose to fix it. With regard to pension contributions, Turner explains that the forecast for falling support ratios ‘faces a society with essentially just three choices’. He summarises these as: 1) increased retirement ages; 2) poorer pensioners; and 3) bigger worker contributions (5).

The same support ratio statistic is used to substantiate ‘affordability problems’ with regards to the health and long-term care needs of older people. With relatively more old people requiring support from relatively fewer younger people, it seems inevitable that the living standards of one or other, or both, of these groups must fall.

The formal logic of this seems sound. Consider a collection of units, X, that provides a support fund out of its resources for a collection of units, Y. If Y gets bigger relative to X, then either some units (people) from Y move over to join X (increased retirement ages); or each unit (person) of Y gets less support from the fund (poorer pensioners); or X collectively, and therefore each unit (person) of X, has to contribute more to the fund (bigger worker contributions).

However, this formal argument can be misleading in both fact and implication, if it loses sight of a number of other factors. Namely:

• If not every unit of X is an active contributor now, and this number can vary in the future, then it is impossible to draw any conclusions either about how much extra today’s contributors will need to pay, or about the need for their numbers to be supplemented from Y. Furthermore, if active members of X are in fact already supporting a much bigger group than Y, then we would need to know about what is happening to this bigger group before drawing any conclusions.




• If each contributor in X has more resources in the future, say double compared to now, then contributions to the support fund can rise absolutely, allowing each member of Y to receive more support, while each active member of X still retains more resources for themselves than today.




• If, in reality, some members of Y are also contributors to the support fund, the original premise breaks down, so any conclusions require more data.




• If members of X also contribute to other funds, no conclusions can be drawn about the total demands on active contributors without knowing what is happening to the size of these other funds.

The real world of ageing throws up exactly these issues. The trend of the statistical ratio might not ‘lie’ – but it has so little bearing on the real world that any conclusions drawn from it are grossly misleading.

This ‘support ratio’ ignores the fact that a large number of people of ‘working age’ do not work. In Britain this totals about nine million people (including those in education, the unemployed and the otherwise economically inactive), and in most advanced countries it represents between a quarter and a third of the working-age population. This makes an enormous difference to what we can call the real, or economic, support ratio – which would compare active workers and non-workers, making the reasonable simplification that every non-worker – young, middle-aged or old – requires the economic support of those who are working.

Most people who will quote a support ratio today of between three and four people of working age per pensioner could probably also give reasonable estimates for the size of the working and total UK populations – 27million working out of a total 59million. If asked, they could work out that the real economic support ratio today is, in fact, less than one (27:32, or 0.84 to be more precise).

People don’t make the connection between this commonsense real economic support ratio of less than unity, and the conventional ageing figure of three to four, because we have become so predisposed to assuming that ageing is a problem. The conventional support ratio is not produced to ‘prove’ that the ageing problem exists – rather it is used to illustrate a problem that is assumed to be plain ‘common sense’. Again, this shows that being anxious about the future is not a provable (or disprovable) perspective – it has become our conditioned state of mind.

The real economic support figure of less than one is pretty unimportant in itself. What is relevant to demolishing the ‘unaffordability’ scenario is what happens to it in the future. Official, and characteristically pessimistic, projections are that in about 2030 the absolute employment level will only be similar to present levels, while at the same time the total population will rise to peak at around 65million. This would represent a shift in the economic support ratio to 0.71, an increasing support requirement per worker of less than one fifth. This compares to that dramatic near doubling headlined by the conventional support statistics.

Employment rates tend to be much more influential on the economic support ratio than the pace of ageing – a point occasionally acknowledged by the Government Actuary’s Department but never highlighted (6). And employment rates are much more variable, and at least potentially subject to our control, than the age structure. Accordingly they could easily rise from those pessimistic official projections.

Forty years ago the male employment rate in Britain was about 95 per cent. Now it has fallen to the high 70s. To reverse that trend by about two thirds, back to a rate in the high 80s over the next 40 years, would be enough on its own to stabilise the economic dependency ratio at the current level. If this happened, the entire ‘unaffordability’ thesis would disappear. Or, to put it another way, if Britain raised its employment rate to the level currently enjoyed by Sweden or Denmark, that would also do it.

The effective average age of retirement is currently below the state pension age of 65, standing at about 62.5 for men. Surveys show that a good proportion of the three million people between the age of 50 and the state pension age say that they would like to work if they could. With this being the case, it’s not difficult to envisage where the extra supply of workers could come from, if only the job opportunities were available.

• Increasing wealth

Another fundamental fault with the support ratio – and this applies just as much to the economic support ratio – is that it ignores rising productivity. By the end of the normal 40-year horizon, when ageing peaks, the average worker will be producing double the wealth compared to today. And this presumes quite modest productivity growth of only about 1.75 per cent a year – slightly below recent experience and well below the long-term historical average. This will roughly translate into double the gross personal income of today – meaning 50 per cent more old people in a country that is wealthier by 100 per cent.

If the economic support ratio were roughly the same as today, every worker and non-worker could, on average, potentially enjoy living standards that were twice as high. Even if it fell by a fifth – the figure assumed by pessimistic official projections – everyone could still be about 80 per cent better off. Do we really think that this is something to be anxious about?

Ignoring the effects of this productivity growth is akin to money illusion in periods of inflation. It is like getting anxious about how to afford things that are now 10 per cent more expensive, and forgetting that your income has grown by 20 per cent over the same time. Moreover, this ‘forgetfulness’ about economic growth expresses something that is genuinely worrying about a lot of discourse today – a loss of historical perspective, and particularly the understanding of how things can change, usually for the better, over time. Worrying about population ageing over the next 40 to 50 years, while abstracting from other things that are bound to change over the same period, would be like the irrational fears of a new student. They might look at an old exam paper and panic about how to answer it but only if they ignore that they should be gaining a lot of knowledge in the intervening period.

The rising productivity that has already occurred provides the main reason why industrialised countries have not been troubled by a falling economic support ratio over the past century (and by even more dramatic falls in the elderly support ratio from about 14:1 in 1900). Not only have they not been stretched financially by demographic trends – they have been able to afford to establish significant support mechanisms for dependants, through the huge extension of public welfare spending.

In Britain this spending (bringing together education, health and social security) cost less than three per cent of gross domestic product (GDP) in 1900. By the 1950s this had risen to about 12 per cent, and is now about 22 per cent. ‘Affordability’ has not been at issue because of the four-fold rise in productivity over that period, providing at the same time vastly improved living conditions for all of us. It has not been a question of working out how to ‘afford’ ageing, but of using the gains of greater prosperity in what (until recently) were widely recognised as progressive and civilised ways of organising for society’s collective benefit.

It is from this perspective that we can view the official projections, which are for an additional four to eight per cent of GDP to be spent on older people in the advanced economies over the next 50 years (7). Leaving aside all the reasons why this looks like a high estimate, let’s say it costs this amount because of society deciding to provide a much better deal for older people. Over 50 years that would mean a rising tax share of GDP of just over 0.1 per cent a year, or in Britain about £1-1.6billion per year in today’s money out of a GDP of over £1000billion.

Yet as a result of rising productivity and prosperity, our discretionary
spending on ‘non-essentials’, beyond food, clothing, housing and transport costs, such as hobbies, extra vacations and eating out, is growing by
considerably more than that each year. This is because the cost of ‘essentials’ takes up less of our income as we become richer. What is the case against devoting a small part of that absolute growth in ‘non-essential’ income in the future to providing handsomely for the elderly? The size of the economic cake is not fixed. As society becomes more prosperous we can all enjoy spending more (on essentials and non-essentials), even if taxation had to rise a little at the same time.

• The need for positive policies

Unfortunately, even some of the most constructive and sensible contributions to the ageing discussion hesitate at the prospect of justifying higher spending to this end (8). But if providing a decent healthcare service for all, long-term care for those in need, and a reformed and generous state pension system requires a higher taxation take over the next 40 to 50 years then this should be said openly. Leaders should argue the political case for it in the same way that, in the past, some argued for collectively funding social improvement. Rising prosperity in the future makes the financial case easy to make – a failure of political leadership seems to be the issue holding society back, instead pandering to presumptions about ‘unaffordability’.

Also contrary to the conventional assumptions of dependency, we should recognise that even the average non-working pensioner is a giver as well as a taker. Pensioners pay taxes. Almost one in 10 people above state pension age in Britain are in paid work and pay income tax, and this proportion is likely to grow as our healthy life expectancy increases. On top of this, another 30 per cent currently pay income tax from unearned income – on income from pensions and savings. Again, this percentage can be expected to grow, as at least some future pensioners are likely to have built up higher savings than earlier generations. On top of all this, pensioners pay the same VAT, council tax, petrol and car tax, duty on drinking and smoking, and, until they are 75, TV licence fee, as do the rest of the population.

Even though the average total tax contribution from pensioners will be well below that for pre-pension-age workers, the scaremongers ignore completely that the more pensioners there are, the greater will be this total contribution from the elderly to national taxation.

And we should recognise that the contribution workers make to providing incomes and services to those above pension age is only a part of what their taxation is used for. On average across the European Union (EU), about 12 per cent of GDP goes on the elderly in the form of public healthcare, long-term care and pensions. However public spending overall reaches well over three times this amount – between 40 and 45 per cent. Those obsessed with projecting how this 12 per cent spent on the elderly will rise over the next 40 to 50 years rarely project how the other 30 per cent might change, and may even fall.

There is certainly no ageing-related reason why the public costs of education, health spending for the rest of the population, other social security and personal services spending, plus all the non-welfare spending on transport, public administration, policing, the armed forces, etc, should continue to rise as a percentage of GDP. If anything, demography would tend to imply the opposite – fewer children to educate and fewer non-employed on benefits if the working age population does decline. So there are likely to be some automatic counter-trends reducing government spending.

If, on the other hand, this wasn’t to happen and governments in the future were to argue for and secure higher real spending on these other areas, representing a constant or even increasing share of GDP, then spending more on greater numbers of older people seems at least as legitimate a role for public provision as the others. Indeed, it seems a great deal more legitimate than some.

So, factoring in the impact of rising productivity and the likely scenario for the real economic dependency ratio over the next 50 years, reinforced by the effects of other more realistic trends that we have noted, means that there is no basis for maintaining that ageing is unaffordable. On the contrary – we should be looking forward to greater prosperity for people of all ages. The genuine problem we face, as in so many other areas today, is not financial or economic but is our negative and cynical view of the future. Our real problem is that ageing is perceived to be a problem. Actions taken because of the fear of this older future are already having deleterious effects for lots of older people. If these irrational anxieties are allowed to fester and this direction of policy continues, it will be to all our detriments.

Phil Mullan is the author of The Imaginary Time Bomb: Why an Ageing Population Is Not a Social Problem, IB Tauris, 2000 (buy this book from Amazon (UK) or Amazon (USA))

(1) ‘Work, work, work’, The Economist, 27 March 2004, p16

(2) See Tory shadow secretary for work and pensions David Willetts’ pamphlet ‘Old Europe? Demographic change and pension reform’, Centre for European Reform, September 2003

(3) ‘Increasing life expectancy “a big threat to pensions”‘, Financial Times, 15 March 2004; Research group concludes that life expectancy in retirement will increase sharply, Actuarial Profession, 15 March 2004

(4) ‘A survey of retirement’, The Economist, 27 March 2004, p4

(5) ‘Demographics, economics and social choice (.doc 353KB), Adair Turner, London School of Economics, 6 November 2003

(6) ‘Research shows that changes in workforce participation rates have in the past been a more important factor than demographic trends in influencing real (economic) dependency’, Chris Shaw (GAD), ‘UK population trends in the 21st century’, Population Trends 103, Spring 2001, Office for National Statistics

(7) ‘Budgetary challenges posed by ageing populations’, European Union Economic Policy Committee, 24 October 2001

(8) See for example the recent report on ageing from the House of Lords Economic Affairs Committee, which nonetheless makes many useful recommendations on this issue – Aspects of the Economics of an Ageing Population, HL Paper 179, January 2004