Select Committee on Economic Affairs Fourth Report



4.1. This chapter examines the impact of population ageing on participation in both paid and unpaid work. Participation in the formal labour market declines quite sharply after age 55, and the majority of men have exited the labour market before the state pension age of 65. Some exit from the labour market is a result of individual choice, but some of it may be imposed by personal factors such as ill health, or by the actions of employers. We examine the patterns of early exit from the labour force for various sub-groups of the population, and consider the extent to which these patterns may be driven by supply-side or demand-side factors.

Participation rates below pension age

4.2. Participation in paid employment in the United Kingdom declined sharply for men over the age of 50 between the 1960s and the mid 1990s, since when participation rates have increased. Since 1997, the employment rate for men aged between 50 and the state pension age has risen from over 67 per cent to 70 per cent. Participation rates for older women have also increased over this period, from 61 per cent to 66 per cent, but this has been driven by a rise in general employment rates for successive cohorts of women who entered the labour market from the 1960s onwards. [37] Thus male and female age-specific participation rates are converging (Q143).

4.3. In 1999, the average age of withdrawal from the labour force in the United Kingdom was 62.6 years for men and 60.4 years for women. By state pension age (65 years for men and 60 years for women) around two thirds of men and half of women have left the labour market. In total, a third of people aged between 50 and state pension age are outside the labour market - almost 3 million people.[38]

4.4. The pattern of participation at older ages in the United Kingdom is similar to that in other European Union countries. Italy, for instance, has an average age of withdrawal for both men and women which is several years below that of the United Kingdom, whereas in Sweden the average age of withdrawal is one year higher than in the United Kingdom for men, and two years higher for women.[39] Box 2 reports data on trends in labour force participation by age in the United Kingdom, and locates the UK experience in international context.

Box 2

Trends in labour force participation by age

Labour force participation among older men has declined in most industrialised countries since the early 1980s, although the past five years have witnessed a general reversal of this trend. The figure below shows this trend for men aged 60-64. [40]

More striking than the trend over time is the difference in the level of participation. There is virtually no tradition of early withdrawal from the labour force by men in Japan. In the USA, Canada and the United Kingdom more than half of men aged 60-64 remain economically active. In most EU countries, however, early retirement is the norm, with barely a third of men aged 60-64 continuing in employment in Germany and Italy, and fewer than one in five in France.

A more comprehensive indicator of economic activity across all age groups, rather than just a single age range, is provided by the average age of withdrawal from the labour force. This measure reveals the average age at which employment income has to be replaced by pension or other non-wage income. Again the UK age of withdrawal lies above the norm for EU countries, close to that for North America, and below that for Japan.

Average age of withdrawal from the labour force in selected OECD countries in 1999[41]
CountryMen Women
Italy58.8 57.9
Germany60.3 60.1
Netherlands60.4 59.8
Canada62.4 60.8
United Kingdom62.6 60.4
USA64.6 63.4
Japan68.5 64.7

4.5. There is broad support for increasing rates of employment among older workers, for example by the Trades Union Congress (TUC) (Q326), the Confederation of British Industry (CBI) (p 220), the Engineering Employers' Federation (EEF) (Q571) and Age Concern (p 95). It is viewed as being consistent with trends in life expectancy and age-specific health status; the current and anticipated demand from employers for more older workers to balance an anticipated decline in the size of the younger adult workforce; and the desires of many people in their 50s and 60s to remain attached to the labour market, though not necessarily on a full-time basis or in their primary occupation.

4.6. We note with approval that the Government share this desire to increase employment among older workers, which they regard as "essential if we are to address the pensions challenge."[42] Higher participation rates at older ages, and thus longer average working lives, help to address the pensions challenge for individuals by increasing the value of their life-time earnings, and thus increasing the personal resources they have to support themselves in old age. Higher participation rates at older ages also help to address the macroeconomic aspect of the pensions challenge, by increasing the ratio of workers to non-workers in the economy.

4.7. The potential contribution that an increase in participation rates could make to the way in which population ageing will affect our economy and society was emphasised by Mr John Grieve Smith and Mr Phil Mullan. They noted that labour force participation rates are more variable and volatile than the age structure of a population, and are much more amenable to policy intervention than is the age structure. As an example, they pointed out that between 1979 and 2000 the number of people over state pension age rose by 1.3 million, but that the increase in the "working-age" population who were not in work - the unemployed, the otherwise inactive, and those on Government training schemes - was 2.2 million. Thus over this period labour market changes substantially outweighed demographic forces in their upward effect on the economic dependency ratio between workers and pensioners.[43]

4.8. John Grieve Smith and Phil Mullan suggested that an increase in participation rates in the United Kingdom is perfectly feasible in the light of experience in other EU countries. They noted that if over the next 30 years the United Kingdom raised its age- and sex-specific participation rates to the current levels in Sweden or Denmark, this would generate an extra 2 million workers, and would serve to keep the ratio of workers to total population in the United Kingdom stable at its 1999 level. It might also be noted that if those EU countries with particularly low participation rates, such as Italy, were to raise their rates to the current UK level, this would make a substantial contribution to the financing of old age support in these countries.

4.9. The CBI made a similar point by reference to the United Kingdom's recent labour market history. It suggested that raising the participation rate for the 50-65 age group from its current level to the United Kingdom average for all people of working age of 79 per cent is a realistic objective, given the fact that the employment rate for men aged 50-65 was 84 per cent as recently as 1979, whereas today it stands at 70 per cent. [44]

4.10. Given the Government's explicit commitment to extend opportunities for older workers, and the widespread support for this ambition, it is surprising that enhanced employment opportunities for older workers do not appear to be fully reflected in some official forecasts of economic activity rates. Mr Grieve Smith and Mr Mullan noted that projections of age-specific activity rates in 2030, as used by the Government Actuary's Department in the 1999 Quinquennial Review of the National Insurance Fund posit a fall in rates for men aged 55-59 from 74.6 per cent in 1999 to 71.1 per cent, and from 53.8 per cent for men aged 60-64 to 49.2 per cent. For women aged 55-59, the projected activity rate rises from 1999 to 2030, but only marginally from 53.1 per cent to 54.5 per cent, which seems surprisingly small given the cohort effects on female participation rates noted in paragraph 4.2 above.

4.11. The Government Actuary's Department agreed that the 1999 Quinquennial Review did not include any sensitivity analysis considering the effect on the main projection of assuming alternative age- and sex-specific economic activity rates.[45] We welcome the fact that the 2000 Quinquennial Review includes alternative calculations based on higher activity rates for older workers.[46] This work includes a full sensitivity analysis, which we hope becomes the norm.

4.12. We conclude that official projections of future age-specific labour force participation rates are inconsistent with the Government's stated objectives and policy intentions of increasing the employment rates of older people.

4.13. We recognise that such projections are calculated on the basis of there being no change in policy. However, when significant new Government policy objectives are introduced, we believe that it is important that official labour force projections take account of the possible effects of those objectives.

4.14. We therefore recommend that the Government pay full attention to the need for their labour force projections and policy objectives to be consistent and compatible with each other, bearing in mind that each interacts with the other.

Participation rates above state pension age

4.15. The Government's ambition to raise labour force participation rates for people aged between 50 and state pension age is matched by a desire to facilitate greater choice and employment opportunity for workers above state pension age. The 2002 pensions Green Paper included proposals to bring forward the date at which more generous rewards are offered to people who choose to defer the receipt of their state pension beyond the normal state pension age, and to allow members of occupational pension schemes to continue working for the sponsoring employer while drawing their occupational pension.[47]

4.16. However, the Government have rejected the idea of raising participation rates beyond age 65 by means of an increase in the state pension age. They have argued that such a move would disproportionately affect lower-income people who rely more on state benefits in retirement, and who, on average, have lower life-expectancy than the population average (Q49). Furthermore, most people have already left the labour market before they reach state pension age, and Mr Peter Tompkins (PricewaterhouseCoopers) noted that an increase in the state pension age to 67 might therefore do little other than increase the number of low-paid older persons who are dependent on non-pension means-tested benefits (Q259).

4.17. Opinion among other witnesses on the idea of raising the state pension age was divided. We agree with Professor Blake, who noted that "it seems very odd, as we live longer and in greater health, that we have this fixed retirement age of 65 for men, soon to be 65 for women" (Q195). He suggested that it would be appropriate to raise the state pension age in response to these improved mortality rates, but to do so with a considerable delay of 20 to 30 years in order to allow individuals to adjust their expectations and their work and savings plans (Q196). The CBI, on the other hand, felt that there is no pressing need to raise the state pension age, and that effort should be directed instead towards increasing participation rates for people in the 50-65 age range.[48] Baroness Greengross suggested that the state pension age should be kept under review in the light of active, fitter and longer life (Q104). Ms Jeannie Drake (Commissioner, Equal Opportunities Commissioner) pointed out that an increase in the state pension age might well lead to many older people re-presenting themselves to the state for means-tested benefits rather than pensions, because of the difficulties they face in securing employment (Q502).

4.18. Ms Alison O'Connell (Director, Pensions Policy Institute) suggested that the Government's appeal to social class differences in mortality rates as a justification for maintaining the current state pension age was unconvincing. She noted that the three-year gap in life expectancy between people in the manual and non-manual groups was less than the five-year gap between men and women, or the seven-year gap between smokers and non-smokers, yet there is no suggestion that state pension age should vary to take account of these inequalities (Q903).

4.19. It is clear that opinions differ about the desirability of driving up participation rates beyond age 65 through an increase in the state pension age. The Government's approach of increasing choice for current workers who wish to work beyond 65 is appropriate for today. However, the Government's decision to reject the case now for a phased increase in the state pension age may not be appropriate for the future. Even on the basis of the Government Actuary's conservative estimate that mortality improvements will tail off by mid-century (paragraph 2.18 above), it is expected that average life expectancy at age 65 will rise by an additional 3 years for men and an additional 2.7 years for women. By 2050, life expectancy at 65 is projected to be 19 years for men and 21.7 years for women, compared with 11.5 years for men and 13.3 years for women in 1928 when the state pension age was set at 65.

4.20. We believe that it would be prudent to link this expected extension in length of life with new expectations that individuals should normally work beyond age 65. Although state pension age does not directly affect employment beyond 65, as people do not have to stop work to collect the pension, nevertheless, as Ms O'Connell pointed out, "it has a huge totemic significance … a signal from the Government that [state pension age] is to be raised in the future is a very powerful message to say that people are now expected to work longer" (Q901).

4.21. We recommend that the Government should formally review the state pension age every five years in the light of trends in life expectancy beyond age 65, and in the light of their own desire to increase participation rates at older ages. Our own judgment is that the state pension age should be raised.

Diversity in patterns of participation: regional effects

4.22. Average age-specific participation rates conceal considerable diversity in employment among sub-groups of the population.

4.23. Evidence on the regional diversity of labour market participation among people aged 50 and above was provided by Christina Beatty and Stephen Fothergill (Centre for Regional Economic and Social Research, Sheffield Hallam University). Their research shows that the average United Kingdom employment rate among 50-64 year old men is 70 per cent, but that this varies between a high of 78 per cent in the South East to a low of 55 per cent in the North East. For women the average United Kingdom participation rate of 55 per cent varies ranges from 62 per cent in the South East to 45 per cent in Wales.

4.24. The share of this same 50-64 age group who are out of employment and claiming sickness benefit varies inversely with the employment rate. In large parts of southern and eastern England, the proportion of older men and women claiming sickness-related benefit is under 10 per cent. On the other hand, in most of northern industrial England, in Wales and in much of central Scotland, the proportion is typically 20-30 per cent. The highest claimant rate in May 2002 was in Merthyr Tydfil (South Wales), where 51 per cent of men aged 50-64 were out of employment and claiming sickness-related benefits.[49]

4.25. This evidence suggests that low rates of labour force participation among older workers are not inevitable, but are strongly influenced by local labour market conditions. In many parts of the South-East there is practically full employment among people aged 50-64 who wish to work. This suggests that the Government's ambition of raising the overall participation rate for the 50-64 age group is unlikely to be achieved unless the large regional disparities in older-age participation rates are reduced through a levelling-up towards the rate in the South East. As older workers are less geographically mobile than younger workers (QQ842-4), this levelling-up will require an increase in older-age employment rates in regions of low participation.

4.26. We note that a number of regional initiatives exist, via Action Teams for Jobs and Employment Zones, to help older people move back into employment.[50] However, we believe the scale of the problem may justify a more aggressive policy response.

4.27. We recommend that the Government incorporate explicit targets with respect to older age employment in their Regional Economic Strategies and in the priorities they set for Regional Development Agencies.

Diversity in patterns of participation: the impact of ethnicity

4.28. In so far as older members of the black and minority ethnic communities reside in areas which have generally low participation rates for persons over 50, they will experience a regional employment disadvantage. However, the impact of ethnicity exists independently of this regional effect.

4.29. Unemployment rates among black and minority ethnic populations are higher for all age groups than among the white population. However, this disadvantage appears to be compounded by age. In the late 1990s, the unemployment rate for whites was 5 per cent for both 35-44 year olds and for people aged between 45 and state pension age. For blacks it was 12 per cent among 35-44 year olds, and 16 per cent among those between 45 and state pension age, and for members of the Pakistani/Bangladeshi communities it was 13 per cent among the 35-44 age group, but 26 per cent among persons aged between 45 and state pension age.[51]

4.30. Ms Maureen Fraser (Commission for Racial Equality) told us that, although direct information is scarce, because of a dearth of representative survey data on the black and minority ethnic population, it appears that many black and minority ethnic persons aged between 45 and state pension age experience both an "ethnic penalty" and an "age penalty" in their interaction with the labour market (Q527).

4.31. We note that the Government have developed policies to improve both the employability and the participation rates of ethnic minorities, as outlined in the Cabinet Office Strategy Unit report on Ethnic Minorities and the Labour Market.[52] However, much of the policy energy is to be directed at labour market entrants and younger workers.

4.32. We recommend that, as part of the Government's evolving strategy for improving the position of ethnic minorities in the labour market, particular attention should be paid to the "double disadvantage" of age and ethnicity encountered by members of ethnic minorities aged over 50.

Diversity in patterns of participation: the impact of gender

4.33. There have been very significant changes over time in the participation rates of women; these changes have affected successive cohorts of women, and have been closely related to women's experience of childbearing. As is the case with many other important social adjustments, the rise in female participation rates has occurred gradually over a number of decades without substantial social disruption, and this reflects the extent to which British society can adapt successfully over time to new social norms.

4.34. Professor Heather Joshi (Centre for Longitudinal Studies, Institute of Education, University of London) provided us with evidence about the labour market experience of successive cohorts of British women born between 1910 and 1970 (and thus reaching state pension age between 1970 and 2035). If one excludes absence from the labour market for reasons of childcare, women in the 1910 cohort worked for less than two thirds the amount of time as men. For the 1958 cohort (that is, those entering the labour market in the mid-1970s) and subsequent cohorts, however, women have worked to virtually the same degree as men, excepting time out of the labour market for birth and childcare. [53]

4.35. An even greater change over time relates to the length of the employment break associated with birth. For the 1910 cohort, the average mother returned to work 13 years after the birth of her child, but for the 1970 birth cohort half of mothers had returned to work within a year of birth. This represents a revolution in the way in which motherhood interacts with employment

4.36. There is, however, a growing divergence over time in the experience of mothers according to their level of qualification. Mothers with higher qualification essentially now pursue continuous careers, whilst mothers with no qualifications typically take employment breaks of 4 to 5 years after the birth of a child. This diversity in female employment patterns has a significant impact on women's economic resources in old age, an issue discussed in chapter 8 below.

Explanations of early retirement: supply and demand

4.37. The reasons for the diversity in age-specific participation rates are more difficult to discern than is the diversity itself. It is clear that both supply-side factors (the skills, capacities, motivations and preferences of workers) and demand-side factors (the needs and preferences of employers) are important, but the relative weight that should be given to each is disputed. Different witnesses emphasised different causes. (The issues of individual capacity and skills and of employer preferences are discussed below in chapters 5 and 6 respectively.)

4.38. The CBI emphasised the disincentives to employment that are created by some elements of the benefits system, noting particularly the fact that recipients of Incapacity Benefit (IB) have received little encouragement from the Benefits Agency and the Employment Service to seek work; that the higher rate of IB compared to Jobseekers Allowance provides a strong incentive to move on to IB; and that once on IB it is difficult to move off it, since older IB recipients would typically take home only about 25 to 35 per cent more from work than from IB.[54]

4.39. The TUC, on the other hand, viewed high rates of receipt of Incapacity Benefit more as a consequence of restricted employment opportunities for older manual workers in those part of the country that have experienced significant restructuring of their historic industries (Q347). However, the TUC also noted that many European countries, even those with low overall levels of unemployment, experience large numbers of older people on long-term benefits, so the level of aggregate demand in the labour market cannot be the only explanation of high levels of long-term benefit receipt (Q347). We give further consideration to the impact of health on employability in chapter 5 below.

4.40. Some early withdrawal from the labour force is entirely voluntary. The Department for Work and Pensions (DWP) has noted that there is no definitive evidence on the causes of inactivity amongst older workers, or on the extent to which early withdrawal from the labour market is voluntary. Among inactive people aged 50+, about 20 per cent say they have retired early, and they may have done so voluntarily, since many of these individuals are drawing occupational or personal pensions.[55]

4.41. People with good pension coverage are more likely to be able to self-finance their early retirement, but on the other hand, some employers may use company pension schemes to shed older workers. The CBI noted that this had been a common practice in the 1980s, when many companies made use of surpluses in their pension funds to restructure their labour force by offering generous early retirement pensions to their older workers (Q606).

4.42. A further issue which needs to be considered in the evaluation of the relative role of supply-side and demand-side influences on participation rates at older ages is the extent to which individuals are prepared to be flexible with respect to wages. Professor Richard Disney and the CBI both noted that one reason for lower participation rates at higher age might be because some older workers are not prepared to accept employment at lower wages than younger workers, even if their skills and productivity are lower than those of younger workers.[56]

4.43. Regardless of the causes of early retirement, there is evidence that individual choice about when to withdraw from work is important for well-being. According to evidence from Professor Peter Warr (Institute of Work Psychology, University of Sheffield) up to age 60/65 the happiness of employed and of early-retired people is, on average, identical if they have chosen their status. Early retirement is more desired by those who dislike their job, have poor health, or who will have an adequate income after retirement.[57]

4.44. It is also clear that many individuals who have left the labour market continue to perform socially and economically valuable roles. They may provide unpaid child care or elder care for family members (thereby enabling others to be economically active) or participate in the voluntary sector. According to evidence provided by Dr Justin Davis Smith (Institute for Volunteering Research), in 1997 40 per cent of people aged 55-64, 45 per cent of those aged 65-74, and 35 per cent of those aged 75+ participated in formal voluntary activity. These rates are lower than for prime-age adults (57 per cent for persons aged 45-54), but they display a much less rapid rate of decline by age than do employment rates.[58]

4.45. The evidence we have received suggests that there are many potential reasons for the non-participation of older people, including:

  • a preference for leisure rather than work on the part of individuals who are in receipt of an adequate income from private or public resources
  • an imposition on older persons by employers who use various institutional mechanisms to induce them to retire
  • poor health
  • a reluctance of older workers to accept a realistic wage for their labour services
  • a fundamental lack of demand in the local labour market

4.46. We note, however, that there is no agreed ranking of the importance of these various causes of non-participation. Both the Department for Work and Pensions and other researchers have identified the problem of a lack of definitive evidence on the causes of labour force inactivity among older persons. If the Government are to make progress towards their goal of increasing participation rates for people aged between 50 and the state pension age, we believe that they will need better information on the causes of non-participation.

4.47. We recommend that the Government commission further research and analysis into economic inactivity among persons aged 50+, in order to produce a clear basis of evidence on which to build policies to promote higher rates of labour force participation among older persons.

37   DWP, Simplicity, security and choice (2002), p 98 Back

38   DWP, volume II, p 2 Back

39   Ibid Back

40   OECD, Labour Force Statistics 1982-2002 (Paris, 2003) Back

41   DWP, volume II, p 2 Back

42   DWP, Simplicity, security and choice (2002), p 93 Back

43   John Grieve Smith and Phil Mullan, volume II, p 38 Back

44   Confederation of British Industry, volume II, p 219 Back

45   GAD, volume II, p 446 Back

46   GAD, Quinquennial Review of the National Insurance Fund 2000 (2003), p 61 (Cm 6008) Back

47   DWP, Simplicity, security and choice, pp 101, 105 Back

48   CBI, volume II, p 219 Back

49   Christina Beatty and Stephen Fothergill, volume II, p 429 Back

50   DWP, Simplicity, security and choice (2002), p 96 Back

51   PRIAE, volume II, p 200 Back

52   Cabinet Office Strategy Unit, Ethnic Minorities and the Labour Market (2003) Back

53   Professor Heather Joshi, volume II, pp 137-8 Back

54   CBI, volume II, p 220 Back

55   DWP, volume II, p 3 Back

56   Professor Disney, volume II, p.299; CBI, volume II, p 220 Back

57   Professor Warr, volume II, p 486 Back

58  Institute for Volunteering Research, volume II, p 447 Back

previous page contents next page

House of Lords home page Parliament home page House of Commons home page search page enquiries index

© Parliamentary copyright 2004