When I’m 84

9 January 2019

If the current savings shortfall were to be eradicated without any changes to saving behaviours, public disbursements or stronger investment returns, the average retirement age would need to increase from 63 to 68 across advanced economies. In the UK, it would need to increase from 63 to 70 (Franklin & Hochlaf 2017). Is this possible and would it make life even tougher for future generations?

Although just 21% of 65- to 69-year-olds in the UK participate in the workforce, in the US 32% do (figure 7). In South Korea, the participation rate is around 47% and in Iceland it’s 52%, so clearly institutions and culture can play a huge role in facilitating longer working lives (OECD data).

Certainly more and more people recognise the need to work later in life. In a worldwide poll by Aegon, 57% of workers envisioned a delayed retirement or a prolonged transition where they continue working at least part-time. In the most recent British Social Attitudes Survey, more than a third of 18- to 24-year-olds and roughly a fifth of 25- to 34-year-olds expected to retire in their 70s. Only 10% of people aged 70 to 74 work today.

Willing… but are they able?

On the face of it, there appears plenty of scope for older people to participate more in the labour market if they had the incentive to do so. Around 80% of British men in their mid-50s are employed, but this falls sharply to just 35% of men in their mid-60s. For women, there’s a greater fall, from 75% to 25%. The average age of leaving the labour market has risen over the past two decades, and the employment rate among over 50s has increased, but it is still lower today for men than it was before 1970.

Three quarters of workers in their 50s would like to still be in work in their early 60s (DWP 2017). That said, many may be willing but unable to work. There are almost one million unemployed 50- to 64-year-olds that are willing, or would like, to work, but a recent study suggests that 26% of them are ready to contribute. There is a clear need for proper employment support to be put in place (Franklin & Hochlaf 2017).

Retirement decisions are influenced considerably by the state pension age, even though compulsory retirement is now illegal, so increasing it should help. Among retirees over 65, 57% cite ‘reached state pension age’ as the key reason for leaving work when they did. Research by the Institute for Fiscal Studies backs up the theory that raising the SPA should induce behavioural change. The employment rate among 60-year-old women rose by 7.3 percentage points when the female SPA rose from 60 to 61; it also had the unexpected effect of boosting the employment rate of male partners by 4.2 percentage points (Cribb et al 2014).

Figure 7: Working in later life

^{Labour force participation rates are increasing for older people around the world}

_{^{Source: Datastream and Rathbones.}}

Does longer life = better life?

Yet we cannot escape the fact that working lives are lengthening by less than total life expectancy, and this is in part due to ill health. So will the increasing need or will to work be matched by the ability to work?

EU data tells us that, while life expectancy is still going up, the number of healthy years expected after 65 hasn’t risen for over a decade. Obesity rates have doubled in the US, England and France since 1990 and show no signs of slowing, and this impacts working lifespans more than total lifespans. A study from the Oxford Institute of Population Ageing found that obesity reduces life expectancy by 1.5 years but reduces healthy life by 6. Figure 8 shows that the ‘disability-free’ life expectancy has actually fallen across the UK, especially among women, who are now expected to have fewer years free from physical or mental impairment than men.

Of course, with the right policies, some individuals could be supported to overcome health barriers and remain in, or return to, the workplace. And, indeed, over the past 10 years the increases in older male employment rates have coincided with a decrease in the proportion reporting they are economically inactive due to being sick or disabled.

Caring responsibilities could also limit the ability for older people to work. Evidence suggests that caring for over 10 hours a week has a substantial negative effect on employment: only 56% of all 50- to 64-year-olds who spend over 10 hours per week providing informal care are in employment compared with 74% of males and 64% with no caring responsibility (DWP 2017).

Carers UK estimates that there will be a 40% increase in the number of carers needed by 2037, totalling nine million carers (Carers UK 2015). In London the economic contribution through informal care is estimated at around £14,600 per older caregiver annually, amounting to a total of £4.6 billion (Barrett et al 2013). If workers haven’t saved enough to retire, they are unlikely to be able to afford private carers for too long either.

Employment practices must facilitate more flexible working. Without them, the increasing need for carers will be extremely difficult to manage and this would have a serious impact on the workforce. It would also put future households ‘too poor to retire’ in an impossible situation. Currently, only a third of employers have a formal, written policy or an informal, verbal policy in place to support carers in their workplace (Thomson 2018).

If older workers prolong their working lives on a part-time or flexible basis, the increase in older-age employment in terms of hours worked may be much less significant than the increase in headcount would suggest.

In sum, today’s younger generations are likely to work for longer out of financial necessity rather than willingness, and there already appears to be an increasing willingness to work among older generations. There are questions over the ability of older people to work, especially considering the need for carers. Health is also a concern, but although healthy life expectancy has stagnated, it is already much longer than the average working life.

The ‘lump of labour’ fallacy

So if people remain in work for longer, will that mean fewer jobs for younger generations?

In short, no. Although it sounds plausible in theory, there are many empirical studies that refute the idea that the old take jobs from the young. It has been a hot topic among researchers since the financial crisis when youth unemployment rose alarmingly in many countries. The weight of evidence strongly suggests that the greater employment of older people actually leads to better outcomes for the young. The patterns are consistent across gender and levels of education. This isn’t about the ‘price’ of young workers adjusting either: there is no negative relationship between older workers’ employment and young workers’ wages (Munnell and Wu 2012).

Research findings are consistent across the world. Using data spanning almost 50 years, from 22 advanced economies, an OECD study found that old and young workers are complements to one another rather than substitutes (Kalwij et al 2010).

We have evidence running the other way too. The Job Release Scheme of 1977—88 which facilitated early retirement in the name of raising youth employment prospects had no positive impact. Similar schemes to increase youth employment in Germany, Denmark and France were also found to be unsuccessful (Banks et al 2008).

The idea that there is a finite number of jobs to go round is referred to as the ‘lump of labour’ fallacy by economists. It has been around since the 1850s; it has been referred to as fallacy since the 1890s (at the time, counteracting concerns about women entering the workforce). Employment is not a zero-sum game, rather it will expand if there are more workers contributing to the economy. Older workers tend to be more productive than the young, and are paid more accordingly. More productive, higher paid workers beget more productive capacity and more consumption and that begets more jobs elsewhere.

It is for this reason that countless studies suggest raising the participation rates of older workers also raises GDP. For example, research by the National Institute for Economic and Social Research (NIESR) shows that adding one year to everyone’s working life in the UK could increase GDP by 1% a year, equivalent to £18 billion in 2015 when the study was carried out. Of course, if older workers are working for longer in order to save more, as we predict, then aggregate demand may increase by a smaller magnitude than previously estimated, and the number of new jobs created may not be quite as large.

The impact is likely still positive. It is possible that, if older workers’ productivity falls below that of younger workers, the lump of labour theory may not be so fallacious. But this is rather farfetched. Although productivity gains decelerate with age, and there is even some evidence that it decreases in the last years of working lives (Dostie 2011), older workers are still considerably more productive than the young (Eichhorst et al 2014). Importantly, wage increases also decelerate with age (although a little less so than productivity).

Figure 8: Live long and prosper

^{Disability-free life expectancy has fallen for some groups over the past few years.}

^{_{Source: Datastream and Rathbones. Data represents years expectancy was measured.}}

The self-fulfilling prophecy of productivity

Furthermore, age-declining productivity may be a self-fulfilling prophecy. If a worker anticipates early retirement, that worker will be less eager to invest in training to prevent productivity from deteriorating. If an employer expects a worker to retire early, that employer won’t have an incentive to invest in them either. In the UK, only 15% of 60- to 69-year-olds either want or expect any workplace training (Thomson 2018). But patterns of training and productivity may shift with workers’ attitudes towards a later retirement.

Importantly, the number of people in their 40s seems to be the key for innovation and change. The median age of those responsible for innovation in the US has been remarkably stable at around 48, whereas the median age of managers who adopt their ideas is lower at around 40 (Feyrer 2008). Those aged 40 are not likely to be crowded out by older workers, and we should breathe a sigh of relief that this cohort is not set to peak in the UK until 2033 and globally until 2096.

On the other hand, older workers have a significant amount of specialised, even firm-specific, knowledge that they find difficult to transfer to new roles or new companies. When good, older workers are let go during a downturn, the older among them undergo a permanent loss of human capital and the economy a permanent loss of productivity (Fujita & Fujiwara 2014). In the UK, the long-term unemployment rate for those aged 50 years and over is higher than for both younger and middle-aged groups (DWP 2017). Therefore, higher old-age participation may lead to a higher ‘natural’ rate of unemployment, and it could mean that employment takes longer to recover after a recession.

Finally, any lingering concerns must be set against current demographic projections. The UK population of 20- to 29-year-olds is set to decrease by 500,000 over the next 10 years. In Europe, the contraction is even more severe. Put another way, if we hold current age-specific rates of labour force participation in the UK constant, the total workforce will increase by just 0.1% a year over the next 10 years. To create more jobs, we need more workers.

Conclusion: a longer, harder road ahead

The research and data drawn upon in this paper clearly suggest that younger generations are likely to be less comfortable in retirement than their predecessors. They are more responsible for their welfare and must bear risks previously shouldered by defined benefit pension plan providers. Yet they don’t seem prepared. To make matters worse governments aren’t ready to fund large increases in the number of state pension beneficiaries either.

While it goes too far to pronounce younger generations ‘too poor to retire’, ‘too poor to retire at the same age as their parents’ seems fair.

Over the long run, we envisage some combination of higher taxation, longer working lives and higher rates of saving. The long-term growth rate of household consumption will suffer as a result. Aggregate consumption growth will slow as future retirees will have less adequate resources than their predecessors. Indeed, the median UK pensioner’s income is already higher than the median worker’s.

Whether individuals save more today in order to meet adequate rates of income in retirement, or continue as they are and cut back dramatically later on is difficult to say and in no small way dependent on government policy. In the UK, for example, auto-enrolment has gone a long way to improving the prospects of millennials, but it is not enough.

As resistant as your audience may be, we encourage our readers to discuss pensions with younger friends and family members.

Investors need to adjust down their projections for growth in the market for consumer goods and services in developed markets, if they have not done so already. Retailers are already challenged by technological disruption but concerns about the overall growth of household consumption are rarely heard.

On the other hand, the opportunity for financial services companies is considerable. Estimates of the additional savings it would take to close the ‘savings gap’ are many times even the largest investment manager’s assets under management (WEF 2015). Those that find innovative ways to engage young savers stand to do especially well.

There is also an opportunity for firms offering new technology and business services to facilitate the flexible working that would enable participation in the workforce later in life.

Rising life expectancy, decreasing home ownership, unfunded government spending, lower investment returns and inadequate private saving are making it less and less likely that younger generations will retire when their parents did. If the current savings shortfall were to be eradicated without any changes to saving behaviours, public disbursements or stronger investment returns, the average retirement age would need to increase from 63 to 68 across advanced economies. In the UK, it would need to increase from 63 to 70 (Franklin & Hochlaf 2017). Is this possible and would it make life even tougher for future generations?