Retirement in the 21st Century
Just as the face of employment is changing, so is the nature of retirement.
The number of people working into later life generally fell for three decades from the 1960s, but then began to climb in the 1990s. What has contributed to these change over the past 50 years?
- looks at the factors contributing to this growth in employment, including changes to the labour market, government policies and pensions, and;
- distinguishes between factors which mainly affected the demand for older workers and those that affected the supply of older workers.
- Employment rates among older men declined dramatically from the late 1960s until the mid-1990s, then started to increase steadily, though remain well below their previous peak. Employment rates among older women have risen since the late 1960s, with only a brief interruption when they declined in the early 1980s.
- The decline in employment among older men to the mid-1990s reflected a combination of factors. On the demand side, the recessions in the early 1980s and 1990s reduced demand temporarily, while shifts in demand away from low-skilled workers and the manufacturing sector disproportionately affected older men. On the supply side, increases in the coverage and generosity of occupational pensions, and widespread use by employers of early retirement incentives, are likely to have led to earlier retirement, while disability benefits were increasingly used as a route to early retirement, particularly when labour demand declined during the 1980s and 1990s recessions.
- Many of these factors affected women too but they were offset by the rising labour market attachment of successive cohorts of women following significant increases in employment protection in the 1970s.
- The more recent rise in employment among older men reflects, to some extent, a reversal of these downward pressures. For example, the late 1990s and early 2000s saw robust economic growth, an apparent decline in employers’ use of early retirement incentives, and reforms tightening access to disability benefits. Improvements in health and reforms increasing the availability of flexible working may also have played a role. However, other factors might have been expected to put downward pressure on employment – such as rising private pension wealth and other wealth (particularly housing). Further empirical analysis is needed to test the importance of these potential explanations.
Employment rates among older men have risen significantly since the mid-1990s, following a long period of decline from the late 1960s onwards. Employment among older women has been rising steadily for decades, with only a brief interruption during the early 1980s recession. This report explores the factors behind these trends, distinguishing between factors which mainly affected the demand for older workers and those that affected the supply of older workers.
The employment rate among older men declined dramatically from the late 1960s until the mid-1990s (as shown in the graph below), with a particularly rapid decline during the early 1980s: more than 89 per cent of men aged 50-64 were in paid work in 1968, compared to just 64 per cent in 1993. This trend went into reverse in the mid-1990s, and the employment rate has risen steadily since, though it still remains well below its previous peak. Employment rates among older women also fell in the early 1980s but otherwise have displayed a secular increase since the late 1960s, which gained speed in the 2000s.
Employment rates from the late 1960s to early 1990s
This study sought to understand the turnaround in employment among older men in the mid-1990s. As a starting point, it is important to understand why employment rates declined in the first place. This study reviewed the extensive literature on this topic.
Between the late 1960s and the mid-1990s a number of factors – affecting both the demand for and supply of older workers – put downward pressure on the labour force participation of older people.
On the demand side, key factors included: the temporary drop in demand during the recessions in the early 1980s and (to a lesser extent) early 1990s; a lasting fall in employer demand for low-skilled workers relative to high-skilled workers due to changes in the nature of economic activity in the UK, which particularly affected older generations who were relatively less skilled; and a shift in industrial composition from manufacturing to services, affecting older men more than women.
On the supply side, this period saw increases in the coverage and generosity of occupational pensions, which are likely to have led to earlier retirement because of the financial incentives provided by such schemes and employers made use of generous early retirement incentives through such schemes. In addition, disability benefits were increasingly used as a route to early retirement, particularly during the periods of weak labour demand in the early 1980s and early 1990s.
For women, these downward pressures were more than offset by substantially higher labour force attachment among successive cohorts of women.
Rising employment since the mid-1990s
Why has employment among older men been rising since the mid-1990s? This study drew on a combination of existing research and new analysis using the English Longitudinal Study of Ageing (ELSA, collected from 2002–03 to 2012–13) to help answer this question.
On the demand side, the period between the mid-1990s and 2007 was one of robust economic growth, and this is likely to have raised employment rates among older men.
On the supply side, the picture is more mixed: while a number of factors are likely to have increased the supply of older workers, others would have been expected to reduce it.
Factors which are likely to have increased employment rates include:
- Reductions in employers’ provision of generous ‘early retirement’ options in defined benefit pensions because of the rising cost of these, which was driven by regulatory reforms, rising life expectancy, lower inflation and lower equity returns.
- Continued improvements in the health of older people.
- Reforms in the mid-1990s and during the 2000s that reduced the on-flow to and increased the off-flow from disability benefits.
- The increase in the female state pension age (which began in April 2010) has also increased employment rates of women in their early 60s.
At the same time, other influences would have been expected to put downward pressure on employment:
- Pension coverage (and wealth held in private pensions) has continued to increase among those approaching retirement.
- Large increases in net wealth among older people, driven largely by rising house prices, would (in theory) be expected to reduce labour supply, though in practice a range of empirical studies have suggested that this ‘wealth effect’ on retirement decisions may be small.
Finally, a number of new policies are likely to have increased both the demand for and supply of older workers. Anti-age discrimination policies have become increasingly stringent: a voluntary code of practice was adopted in 1999, followed by a law in 2006 banning age discrimination in all recruitment, promotion and training, including prohibiting mandatory retirement before age 65 (extended to all ages in 2011). Reforms have also been introduced to encourage flexible working, which may benefit older workers.
While there are a number of plausible explanations for rising employment among older people, many of these have not been subjected to rigorous empirical analysis. Priorities for further research should include assessing the impact of recent disability reforms on employment, documenting the apparent shift away from early retirement packages, and assessing the effect of anti age-discrimination legislation.
The gradual nature of retirement
The process of ‘retirement’ looks very different for different people. For some, leaving work is a fairly rapid and permanent transition, whereas for others it is more gradual, involving periods spent working part time, and/or periods of self-employment.
This study used the detailed longitudinal information available from ELSA to look at individuals’ transitions out of work over the 10-year period from 2002–03 to 2012–13, focussing on three aspects of this transition: reducing hours, exiting work and returning to work.
Those who are in poor health are significantly less likely to be in work at age 50 and, even among those who are working at that age, they are likely to exit more quickly than healthier individuals. There are also differences in the rate at which people exit employment by pension status: of those who were initially working and aged 50–59, 52 per cent of those with only a defined contribution pension were still working 10 years later, compared to 43 per cent of those with no private pension, and 35 per cent of those with only a defined benefit pension.
A significant minority of people who are initially full-time employees experience a period of part-time work or self employment before they fully ‘retire’: of those aged 50-59 and working full time in 2002–03, over the following 10 years just over a quarter experienced a spell working part time (as an employee or self-employed), 5 per cent had a period of self-employment (either full- or part-time), and 7 per cent spent time unemployed. Reducing hours of work seems to have been more common among those who change jobs (either with the same employer or changing employer) than among those who remain in the same job. For more highly-educated people, switching to self-employment in particular is associated with a large and significant cut in average hours of work.
Employment rates among older people have changed significantly over the past 50 years. This study sought to improve our understanding of the factors behind these trends, with a particular focus on the turnaround in male employment since the mid-1990s. The analysis distinguished between factors that acted mainly on the demand for older workers and those that mainly affected the supply of older workers.
A number of the trends that have contributed to rising labour force participation over the last decade are expected to continue, while the impact of other factors (such as the shift from defined benefit to defined contribution pensions and legislation governing flexible working) are yet to be fully felt. Therefore, we might expect to see a continued increase in labour force participation of older people over the next decade.
About the project
This study was part of a programme of work examining the outlook for poverty and living standards in older age carried out by researchers at the Institute for Fiscal Studies, and funded by the Joseph Rowntree Foundation. It involved a review of the relevant literature, and analysis of the Family Expenditure Survey, Labour Force Survey and the English Longitudinal Study of Ageing.