Which of Australia's baby boomers expect to delay their retirement? An occupational overview.

• Author: Jackson, Natalie
• Position: Contributed Article

Abstract

This paper examines the retirement intentions of Australian Baby Boomers by occupation. Workers from 14 of 35 occupations expect to retire earlier than the national average of 64.3 years, with 'white collar' workers more likely to be among the early retiring, and 'blue collar' workers among the late. Early-retiring intentions will be reinforced by short gaps between preferred and expected retirement age, relatively high levels of financial security and formal discussion about retirement, yet lower than average levels of desire for transition-to-retirement arrangements; late-retiring intentions will be reinforced by more or less the opposite. Early-retiring occupations are also the largest and have the potential to cost the economy 1.26 million person-years of working life. The findings indicate that policy interventions should be targeted, that interventions should include non-economic inducements and reforms, including efforts to ensure a healthier longer work life, and that Baby Boomer retirement needs to be 'managed' at an institutional level.

Introduction

A handful of recent studies on the retirement intentions of Australia's Baby Boomers (in Australia, born 1946-65) have each come to a similar conclusion: most Boomers are open to the idea of later retirement, with many expecting to work well into their sixties. This is good news for the Australian Government, which was concerned that Boomers would follow their immediate predecessors and retire 'early'. As recently as 1997, average age of retirement was 58 for males and 41 for females (ABS 1998). By 2005, for those aged 45+ years and in the labour force, it had risen to 62.3 years (63 for males and 61 for females--ABS 2006). As reported below, our overall finding is of a median expected retirement age of 64.3 years for workers aged 40-59 in 2006 (64.7 for males and 64.0 for females). Like ABS, we also find that expected retirement age is greater for those closer to retirement, with a median age of 65.6 years for those currently aged 55-59, compared with 64 years for those aged 40-49 years.

While the news is good, intentions to delay retirement--at least as they were prior to the global economic 'meltdown'--are by no means distributed evenly through the population, and this is particularly so when considered by occupation (e.g. Management Advisory Committee 2003 on public servants; O'Brien-Pallas, Duffield and Alksnis 2004 on nurses; Schofield and Beard 2005; Schofield, Page, Lyell and Walker 2006 on doctors and nurses in rural/city locations; NSW Government 2006 on public servants). This paper puts earlier findings--most of which have pertained to an individual occupation or industry--into an internally consistent, comparative context, by presenting findings from the Australian Survey of Retirement Attitudes and Motivations (ASRAM), which collected data by occupation in 2006. Additionally, while the changed economic situation means that some of these attitudes and intentions may have since altered, knowing how they differed by occupation in 2006 presents a serendipitous baseline against which future changes can be assessed.

Australian Survey of Retirement Attitudes and Motivations

The ASRAM is a nationally representative survey of 2501 Australian workers aged 40-59 years in 2006--essentially the Baby Boom generation. (2) Its major aim was to assess the extent to which older workers' retirement intentions and motivations are in line with, and likely to interact with, new government retirement policies designed to encourage later retirement. The fact that it does so by occupation provides a unique opportunity to compare these intentions and motivations across a range of different contexts.

The survey was conducted by telephone between July and November 2006 and sampled respondents who were currently participating in the paid workforce. The age group 40-59 years was selected as the most appropriate to survey as all were likely to have at least begun to contemplate their own retirement. The active labour force status sample criteria made contacting respondents a lengthy task, with only six per cent of phone calls making contact with a survey-eligible individual. Despite this low contact rate, the eventual response rate was high. Only 13 per cent (379) of eventual contacts who were survey-eligible refused to participate, giving a response rate of 87 percent.

Comparison of sample attributes with census data is difficult on some counts due to the labour force-active nature of the sample (Appendix A). Females are somewhat over represented by comparison with their share of the labour force, as are those aged 55-59 and Professionals. Survey respondents with Bachelor Degree or Higher qualifications are also over represented by comparison with the general population aged 40-59, as are the Australian-born and single persons. These disparities are likely to be a result of the sample's labour force-active status. The employment status and overall occupational distribution of the sample is remarkably similar to the national distribution, as is the large proportion (almost three-quarters) married or partnered. Each age group is also well represented. The sample's socio-demographic characteristics are, therefore, acceptably compatible with those from the general population of older Australian workers.

We undertake the analysis at the major and sub-major occupational levels according to the Australian Standard Classification of Occupations (ABS 1997). (3) This generates data for nine occupational groups at the one-digit level, and 35 at two-digit level. In places we also refer to occupations at the three-digit level (where in 2001, 115 occupations were counted--see Appendix B); however space constraints prevented undertaking the entire analysis at that level. We focus on six survey items: expected retirement age; preferred retirement age; reasons for later retirement than preferred; source of retirement income; level of discussion on retirement with employer/supervisor; and the nature of any transition to retirement preferences. We link these items with data on occupational size and age structure from the 2001 Census. (4) We base the analysis on frequency distributions and their correlations (via the Pearson's Correlation Coefficient) as a means of gaining insight into the data. (5) To minimise potential bias from the differing age structures of each occupation we present retirement age in terms of medians rather than means.

We acknowledge that, because of space constraints, the analysis is heavy on description and light on reference to supporting literature. We also note that number of respondents in many occupations is too small to undertake this analysis by gender.

Expected Age at Retirement Vis-a-vis Occupational Distribution

Table 1 presents data on expected age at retirement in 2006 by major and sub-major occupational code. Also given is the difference in years between the expected retirement age of each occupation and the national average, and the number of all persons aged 40-59 years employed in each occupation in 2001. Multiplying the former by the latter gives an indication of the number of person years potentially lost or gained by retirement intentions.

The median expected retirement age of Australia's Baby Boomers in 2006 was 64.3 years. At the one digit level, Associate Professionals and Professionals and have the two youngest expected retirement ages (63.8 and 64.0 years respectively), and Intermediate Production and Transport Workers, and Labourers and Related Workers, the two oldest (65.8 and 65.5 years).

Associate Professionals and Professionals are also two of the three largest occupational groups. They account for one-third of all employed 40-59 years olds, and for 14 per cent of the total employed workforce. The second-largest occupation, Intermediate Clerical and Sales Workers, is the third earliest-retiring (median age 64.1 years), and accounts for a further 16 per cent of 40-59 year old workers. Together these three groups account for half of all employed 40-59 year olds and over 20 per cent of the total employed workforce. By contrast, Intermediate Production and Transport Workers, and Labourers and Related Workers aged 40-59 years together account for just 17 per cent of all workers at those ages and seven per cent of the total employed workforce.

A general picture thus emerges in which the lower the expected retirement age, the larger the occupation. This is clearly illustrated in Figure 1, which plots the size of each occupation as a percentage of the total employed 40-59 year old population by expected retirement age, arranged from lowest to highest. Included in the early-retiring group are also Managers and Administrators, and Advanced Clerical and Service Workers, both occupations anticipating a median retirement age of 64.2 years.

When converted to person-years potentially lost or gained from expected retirement age, the early retiring Associate Professionals and Professionals occupations generate--at the one-digit level--a combined loss of 436,718 years, while the later retiring Labourers and Related Workers, and Intermediate Production and Transport Workers, potentially contributes 818,918 person years by staying on (Table 2).

As always, there are some caveats to these simple calculations. For example, when the data are aggregated at the two-digit level, the potential loss from the Professional and Associate Professional occupations drops significantly, to just 99,814 person years, because the single largest occupation among this group, Business and Information Professionals, has an expected retirement age that is exactly the same as the average. For similar compositional reasons, the potential gain from the later retiring Labourers and Related Workers, and Intermediate Production and Transport Workers, reduces to 705,653 person years. It is thus more appropriate to consider the overall impact of expected...