The backdrop: the challenge of long-lived societies (Why don’t people work longer?)

Editor’s note: This text corresponds to the first instalment of the article by Professor Nicholas Barr for CENIE, “Why don’t we work longer?”. Throughout October we will be publishing the different parts on a weekly basis, within the framework of the activities organised on the occasion of Financial Education Day 2025.

In the early 2000s I was a member of a British government committee tasked with investigating policies for extending working life. The first draft of the report by officials put considerable emphasis on age discrimination, but I suspected that that was over-simplified and therefore asked the following question: Why might it be rational not to employ older workers? This article offers some answers and policy responses. The opening section sets the context. The following two sections focus on diagnosis, asking in turn why individuals might not work longer or move to part-time work, and why employers might not hire older workers or provide more part-time work. The fourth section discusses the resulting policy directions. The final section concludes. 

The backdrop: What is the problem?

The starting point are the twin roots of population ageing.

Longer lives in England Wales, ‘While [there were] 265 centenarians … in 1950, there were 6,680 of them 50 years later …. In developed countries the number of people celebrating their 100th birthday doubled each decade between 1950 and 1980 …’ (Vaupel and v. Kistowski 2005). 

Christensen at al. (2009, Table 1) estimate that someone aged 20 today in one of eight OECD countries (Canada, Denmark, France, Germany, Italy, Japan, UK and USA) has a better than 50% chance of living to be 100 years old.  That is all wonderful, good news but creates problems paying for pensions.

Lower fertility: the second element is that women are having fewer children. Data collected by the United Nations show a steady decline in fertility in OECD countries from 3.3 children per woman in 1960 to 1.5 in 2023. That, too, can be regarded as good news in that women are empowered to make choices about their fertility, and also because good for the planet. But fewer births mean fewer workers in the future, which, again, creates problems paying for pensions.

Strategic policy directions

In tackling the problem, the best way forward is to increase the growth rate of national output, thus increasing the resources available both for workers and pensioners. Policies include more saving to finance more investment in physical and human capital to make each individual worker more productive (Barr 2021), and policies to increase the labour force.

Policies to increase the labour force include:
• Raising labour-force participation at all ages.
• Importing labour either directly through immigration or indirectly by exporting capital to countries with a younger population; and
• Extending working life.

Achieving the last (the topic of this article) involves later retirement and also more flexible retirement.

Later retirement

If people live longer and retirement age is fixed, the costs of a defined-benefit pension plan will increase, putting sustainability at risk. In a defined-contribution plan, a person has to spread their pension savings over more years, putting adequacy at risk.

An obvious part of the solution is that pension age should rise in a sensible way as life expectancy increases. Suppose that people on average work for 40 years and are retired for 20, i.e. two years of work for each year of retirement. If policy makers wish to preserve that relativity, then pension age should rise by 8 months for each year by which life expectancy increases.

Raising pension age, however, needs to recognise that that richer people generally live longer than poorer people. In the richest parts of the USA life expectancy is around 85 years, in the poorest parts up to 20 years lower. In the UK, the difference for men is 10 years. 

Though there are no easy or quick answers, policies that can help include a mix of: 

• Allowing early access to a reduced pension.
• Allowing early access to an unreduced benefit for workers with more than n years of contributions, hence a lower pension age for people who entered the labour force earlier, who tend to be lower earners. There might also be early access for someone with serious chronic health problems.
• Adjusting pension age in line with changes in the life expectancy of lower-income groups rather than of the population as a whole.
• Having a progressive pension formula such that people with lower earnings receive more pension per Euro of contribution than higher earners.
• Over time improving health care and introducing action on diet and lifestyle to reduce the difference between the life expectancy of different groups.

More flexible retirement

Alongside the argument for later retirement is the separate argument for more flexible retirement. When retirement was invented (for example, tax-financed non-contributory pensions in New Zealand in 1898), someone aged 65 was old and often infirm, and interfered with the productivity of younger workers. The purpose of pensions was to clear out unproductive older workers, so it made sense for retirement to be mandatory and complete. 

Since then, people are living longer and so should work longer. But in addition, countries are richer and so can afford to give people a period of leisure at the end of their working lives. But that means that the purpose of retirement has changed – no longer a device for weeding out dead wood, but a social invention for dividing adult life between work and leisure. Thus, it is right to recognize that individuals vary widely in their preferences and personal circumstances. Many people do not want to retire fully as soon as they are allowed, because of the extra earnings, because of the resulting extra pension, because for many people their job is part of their identity and/or because they continue to enjoy working, either in their current job or another one.

Older workers – desirable as well as necessary

Though an essential element in addressing population ageing, it is mistaken to think of working longer as merely a regrettable necessity. As just indicated, there are gains for many workers in terms of individual well-being as well as financially. Equally, there are benefits to the wider economy, not just through having more workers but also in the skills and experiences that older people can bring to the workplace (Financial Times 2025).

In brief, older workers bring gains of at least three sorts 

• Human interaction: while artificial intelligence will take over many routine tasks, emotional engagement, fostering teamwork, etc. requires people – emotional intelligence is different from and a necessary complement to digital intelligence. Older workers have not lost emotional intelligence – indeed are likely to have gained it over the life course.
• Older workers have experience of earlier technological upheavals, notably the spread of the internet, hence some knowledge of how to navigate the unpredictable impacts of rapid change.
• Diversity: it is now widely understood (see Indeed 2025, Teambuilding.com 2025) that diversity has benefits for the welfare of individual workers, for productivity and for profitability. Age is an important dimension of diversity.

The puzzle 

Good policy design would (a) adjust pensions spending to aggregate resource pressures in part by raising the average retirement age and (b) accommodate diversity across individuals by offering choice over how a person moves from full-time work to full retirement. 

Such choice and flexibility would be good economics, good politics and good social policy even if there were no problem paying for pensions, and so there is widespread agreement that flexible retirement is a good idea. But it has not happened on a large scale. Why?