Research Programme

Flexible ageing: new ways to measure and explore the diverse experience of population ageing in Scotland, using the Scottish Longitudinal Study

ESRC Secondary Data Analysis Initiative logo

John MacInnes, Jeroen Spijker

The Project is funded under the ESRC Secondary Data Analysis Initiative

Project summary


Elderly health and mortality have been steadily improving in Scotland since the 1960s despite the persistence of large regional and local differences, and a life expectancy that remains below other Western European countries. 'Population ageing' will intensify in Scotland as fertility levels are likely to remain below replacement level, while life expectancy will continue to increase. By contrast, the speed of individual ageing or senescence will continue to slow down as older people become more active and health conscious, delaying the onset of chronic disease and reducing mortality. Population ageing will therefore have substantial social consequences both for individual wellbeing and for the sustainability and effectiveness of pension, health and social care systems. However, assessing these consequences needs a better understanding of the relationship between individual and population ageing, since the expansion of the older population will go hand in hand with changes in its composition and characteristics.

The project has three objectives:

1. To reassess the concepts of population ageing and the size and composition of the 'old' population by constructing alternative measurements of ageing based on (a) years of remaining healthy life expectancy, and (b) proportion of total life expectancy remaining, and to compare these with conventional measures, such as the old-age dependency ratio.
2. On the basis of this reassessment to construct a taxonomy of diverse ageing experiences (e.g. through health status, current and past socio-economic status, living arrangements and civil status) and compare these across birth cohorts and by occupational social class.
3. To review what implications the results have for forecasts of future dependency ratios and the evolution of demand for health and social care.


This project takes a sociodemographic approach to assess the challenges and opportunities of ageing populations in different contexts. Using data from the Scottish Longitudinal Survey, new measures of who comprises the elderly population based on years of remaining life expectancy (as predicted by risk factors such as gender, occupation, economic activity, civil status, subjective rated health and area of residence) rather than years since birth are constructed. Age is treated in terms of years left until death rather than calendar age. The latter are routinely used in the social and political analysis (e.g. in calculation of dependency ratios) because of the straightforward availability of data and their relevance to eligibility criteria set by public policy for e.g. pension entitlement and other social benefits. However, these commonly-used measures do not consider the impact of the underlying driver of population ageing in the first place: improvements in health which reduce morbidity and mortality.

These alternative definitions of the elderly based on remaining life expectancy are used to construct estimates of the size and composition of the elderly, and their ratio to the actually economically active population in Scotland, using SLS and register data. This is then compared with 'traditional' population pyramids and dependency ratios for the period 1975 - 2025.


The usual approaches to definitions of 'old age', in terms of fixed age categories, such as taking the male retirement age of 65 as a cut-off point, or defining the old age dependency ratio (OADR) by expressing the size of the population aged 65+ as a proportion of the working age (16 or 20 up to the State Pension Age) population at a given point in time, have been challenged in two ways by this study:

First, the size of the employed workforce is driven by many factors other than age so that its relationship to the population age structure is rather dynamic. For example, the employment rate of women has increased dramatically over the last 50 years as the ‘male breadwinner’ employment system has weakened, while the skill demands of a high technology economy has increased the age of entry to the labour market. Less than one half of men and women wait until the State Pension Age to leave the labour market, with the result that most adult ‘dependents’ (in the sense of not being in employment) are now below pension age. However, a growing minority of older people are now working beyond the State Pension Age, and given changes in public policy, attitudes to age discrimination, and employers’ human resource management policies, the size of this minority might be expected to increase.

Second, this research makes the case for using Remaining Life Expectancy (RLE) based measures of the size of the elderly population. Instead of taking a fixed age in years as the boundary used to define old age, this measure takes account of falling older age mortality by counting back from expected age at death. The distribution and average of this time can be calculated using life table data, taken from such sources as the Human Mortality Database, and census data linked to vital event registration. Thus, if we define the elderly as those within a certain amount of time of expected death, the boundary between the elderly and others will gradually move upwards as life expectancy increases. This is important because when mortality is falling substantially (as it is now in the UK and many other countries) this rapid change in ‘years left’ means that many behaviours and attitudes, including health related behaviours and attitudes, saving behaviour and consumption of health and social care services may correlate as strongly or more strongly with RLE (‘years left’) than with age (‘years lived’).


Project activities

Date Activity Description
30 April-2 May 2015 “Time-to-Death Patterns in Markers of Age and Dependency” PAA Annual Meeting 2015, San Diego Poster presented by Jeroen Spijker
3-5 December 2014 “Incorporating time-to-death (TTD) in health-based population ageing measurements” and “Time-to-death patterns in markers of age and dependency.” New Measures of Age and Ageing, Vienna Papers co-presented by Jeroen Spijker and John MacInnes
1-3 September 2014 “The myth of old age” BSG 43rd Annual Conference, University of Southampton Paper co-presented by Jeroen Spijker and John MacInnes
7 July 2014 “Population ageing: the time-bomb that isn’t?” Westminster Palace, London Address made by John MacInnes to the Parliamentary University Group
25-28 June 2014 “How should population ageing be measured?” and “Decomposing and recomposing the population pyramid by remaining years of life” EPC Conference, Budapest Papers co-presented by Jeroen Spijker and John MacInnes
1-3 March 2014 “Population Aging: How Should It Be Measured?” and “Decomposing and Recomposing the Population Pyramid by Remaining Years of Life.” PAA Annual Meeting 2014, Boston Papers co-presented by Jeroen Spijker and John MacInnes
29 January 2014 “Flexible ageing: new ways to measure and explore the diverse experience of population aging in Scotland, using the Scottish Longitudinal Study”  Data analysis for effective policy for older people (ESRC Age UK Showcase Event), Tavis House, London Presentation by Jeroen Spijker and John MacInnes
















Spijker, J. and MacInnes, J. (2014) Hoe grijs is Nederland eigenlijk? [How grey is The Netherlands in reality?] DEMOS, bulletin over bevolking en samenleving, 30 (4), 1-4

Spijker J., MacInnes J. (2013) Population ageing: the timebomb that isn’t? British Medical Journal, 347, 20-22.

Spijker J., MacInnes J. (2013) Population ageing in Scotland: Time for a re-think? Scottish Affairs, 85, 53-74.


Media activities

Website articles:

"Defusing the Population "Time Bomb" on - 13 February 2014.       

"The politics of population ageing"  on, written by John MacInnes and Jeroen Spijker – 6 February 2014           

"Getting older doesn't make you more conservativeon (UK Edition), written by John MacInnes and Jeroen Spijker – 11 January 2014.

"Hard evidence: can we afford an ageing population?" on – 9 January 2014.

"Nurse warning after rise in proportion aged at least 50" on – 3 January 2014.

"Part I: Dementia: the "epidemic" of metaphorson – 1 January 2014.


You can also browse all CPC media outputs and population-related articles from CPC members on our! page.