Population ageing in Asia and the Pacific: Dependency metrics for policy

https://doi.org/10.1016/j.jeoa.2016.05.002Get rights and content

Abstract

This paper explores the effects of population ageing in Asia. Standard demographic dependency ratios leave out much of economic significance from demographic projections. We examine, adapt and estimate various measures of dependency, indicative of health, long term care, labour market, economic wellbeing, and fiscal sustainability, and relate them to one another. A range of projection scenarios is used to illustrate the effects of social, institutional and policy changes across several Asia–Pacific economies chosen to represent Asia’s demographic, social and economic diversity.

Introduction

The often cited concern about population ageing relates to the increase in the number of older people compared to workers. A standard approach to quantifying this relationship is the age dependency ratio, typically defined as the number of people aged 65 and over as a proportion of those aged 15–64. The ratio is readily comparable across time and jurisdiction; but its simplicity rests on a number of assumptions about what it means to be ‘dependent’. Changing patterns of human capital accumulation and labour productivity, labour force participation and work intensity by age and sex, health status and disability, public spending by age, among other socio-economic shifts, alter the actual levels of dependency across generations, and the implied social, economic, and fiscal stress.

Systematically taking these factors into account would furnish policy makers and households with a better understanding of how demographic change affects the economy and transfer systems, allowing for improved policy and economic decisions. Initial attempts to capture some of these effects have been undertaken for various regions, notably in Europe (for example Sanderson and Scherbov, 2005, Spijker and MacInnes, 2013, Prskawetz and Sambt, 2014, Loichinger et al., 2014, Zamaro et al., 2008). As noted by Holzman (2013, p120) “…one should not be misled by the traditional measurement and projections of population ageing: they need to be reviewed and revised, particularly with regard to the onset of what is considered old”.

In this paper we build on some of the above studies, presenting a number of methods of calculating age dependency and illustrating these with projections for countries in Asia. We estimate three dependency measures that relate to health, two that relate to the wider economy and one that takes account of dependence on public budgets. As a result, each measure can be thought to have a specific application (e.g. measuring dependency related to health, long term care, or fiscal policy). In presenting each one, we relax the simplifying assumptions (e.g. age cut-off, actual rather than implied labour force status) and therefore refine the concept of dependency into one that is arguably more realistic. For example, in the final sections we make use of National Transfer Accounts, which are especially interesting in Asia because differences in development translate to contrasting economic lifecycles between the citizens of different countries.

The analysis includes nine Asia–Pacific countries with different levels of ageing and development (Table 1). Some, such as India, Indonesia, and the Philippines are relatively young and will remain so for some time. Korea and Taiwan are ageing at an unprecedented rate, in contrast to Australia which has higher rates of fertility and migration. China and Thailand are still middle-income countries but will soon resemble western countries in their demographic structure. Finally, while Japan still has some ageing ahead of it, it already serves as an example of what other countries may look like in the medium-term (though with notably different social support systems).

The rest of the paper is organised as follows. Section ‘Measuring dependency’ outlines the different concepts of dependency, which are then presented for each of the countries in section 3–8. For each measure we offer scenarios that indicate the scale of potential institutional or policy changes. For example, what would be the effect if the average 50-year-old was employed at the rate of a 45-year-old, if the labour force experienced higher productivity growth, or if health and long term care spending in India and Indonesia matched the per capita spending seen in in Japan? In section ‘Discussion and conclusions’ we discuss the overall policy implications and conclude.

Section snippets

Measuring dependency

Historically, the starting point for any discussion of demographic change (e.g., as early as Dublin and Lotka 1949, or OECD 1988) has appealed to dependency ratio (or its inverse, the support ratio): The number of ‘dependants’ per 100 ‘supporting’ population. In the numerator, the total dependant population usually consists of children below age 15 (or in education, sometimes using age 20 as the cut off) and the elderly over the age of 65 (or, in developing countries, over the age of 60). The

Dependency based on remaining life expectancy

In redefining dependency, we can start by considering old age as a proportion or period of time before death, thereby anchoring the concept to changes in life expectancy.

We calculate Sanderson and Scherbov’s OADRRLE15, which compares the population within 15 years of remaining life expectancy with the population aged over 15 but with more than 15 years of remaining life expectancy (see Fig. 1, panel B).

The projections rely on updated age-based population and period life expectancy data from UN

Dependency based on unhealthy life expectancy

Disability criteria provide another basis for setting the threshold age for dependency. Increasingly, data are available to estimate the age at which people become severely disabled, on average. The following measure uses data on healthy and unhealthy life expectancy at older ages (assumed to be concentrated at the end of life) to decide on this threshold. The measure can be thought of as a Very Old Age Dependency (VOADR), indicative of the relative resource changes necessary to cope with the

Dependency based on burden of disease

Having varied the threshold for dependency based on life expectancy and morbidity we next relax the assumption that there exists a binary, single-age distinction. We do so by looking at actual health measures at each age, essentially using health as a proxy for the level need and support.

We construct a Health Dependency Ratio by applying severity weighed prevalence of disease and injury in 2010 to the projected age structure between 2010 and 2100. This separates out the old age population into

Dependency based on labour market status

Health status based dependency is helpful for thinking about health related demands as well as ability to work and take on caring responsibilities. To the extent that we want to measure market based production it makes sense to look at employment. Taking account of employment status and population structure also moves us a step closer to considering potential levels of GDP growth and the demographic dividend (when employment-to-population rates are particularly high due to large relative

Dependency based on economic consumption and production

Employment status only provides information about the rates of work, instead of quantifying the intensity or returns from work and the potentially different rates of productivity of different groups. It also ignores the levels of economic need of the dependent population. Here we move to a more realistic analysis of economic dependency by taking these into account.

We follow the NTA approach of calculating economic support ratios (Lee and Mason, 2011b, UN, 2013), though we use its inverse. This

Dependency based on transfers between government and individuals

To understand the extent to which the age structure affects government budgets, it is instructive to look at how demography affects public expenditure on the one hand and taxes and charges on the other. The patterns will not necessarily reflect demographic dependency ratios, not only because of the arbitrary age cut-offs of such ratios but also because as societies age and develop their readiness to privately fund spending on children transitions to publicly financing old age.

We follow Miller

Discussion and conclusions

This paper contributes to an ongoing debate about quantifying the effects of population ageing. It sought to gain insights into the complexities of demographic transition in the world’s most demographically and economically dynamic region – Asia, by extending standard dependency ratios to capture health, labour force and public economic factors. Several Asia–Pacific economies have been chosen for analysis, with the aim of reflecting Asia’s demographic, social and economic diversity.

A range of

References (47)

  • J. Salomon et al.

    Common values in assessing health outcomes from disease and injury: disability weights measurement study for the Global Burden of Disease Study 2010

    Lancet

    (2012)
  • ABS [Australian Bureau of Statistics] (2011) ‘The 2011 Census’,...
  • ABS [Australian Bureau of Statistics] (2012) ‘Cat. 4430.0 Disability aging and carers: Summary of findings 2009’,...
  • C. An et al.

    Intergenerational resources allocation in the Republic of Korea

  • J. Campbell et al.

    How different countries allocate long-term care resources to users: a comparative snapshot

  • Chomik, R., MacLennan, M., 2014. ‘Aged Care in Australia, part 1’, CEPAR Research Brief, ARC Centre for Population...
  • R. Chomik et al.

    Population ageing and social security in Asia

    Asian Econ. Policy Rev.

    (2015)
  • Cutler, D., K. Ghosh, M. Landrum (2013) ‘Evidence for significant compression of morbidity in the elderly US...
  • D. Cutler et al.

    An aging society: opportunity or challenge?

    Brookings Pap. Econ. Act.

    (1990)
  • L. Dublin et al.

    Length of Life, A study of the life table

    (1949)
  • EC [European Commission]

    The 2012 Ageing Report: Economic and budgetary projections for the 27 EU Member States (2010–2060)

    (2012)
  • EC [European Commission]

    The 2015 Ageing Report: Underlying Assumptions and Projection Methodologies

    (2014)
  • J. Flaherty et al.

    China: the aging giant

    J. Am. Geriatr. Soc.

    (2007)
  • J. Fries

    Natural death, and the compression of morbidity

    N. Engl. J. Med.

    (1980)
  • Gruber, J., Wise D. (1998) ‘Social Security and Retirement: An International Comparison’, The American Economic Review,...
  • R. Holzman

    An optimistic perspective on population ageing and old-age financial protection

    Malaysian J. Econ. Stud.

    (2013)
  • IHME [Institute for Health Metrics Evaluation] (2012) ‘Global Burden of Disease data (2010)’,...
  • IHME [Institute for Health Metrics Evaluation] (2013) ‘Global Burden of Disease data (2013)’,...
  • ILO [International Labor Organisation’ (2015) ‘ILOSTAT: Annual Indicators’,...
  • R.D. Lee

    Population age structure, intergenerational transfer, and wealth: a new approach, with applications to the United States

    J. Hum. Resour.

    (1994)
  • Lee, R. and Mason A. [principal authors and eds.] (2011a) ‘Population Aging and the Generational Economy: A Global...
  • Lee, R., Mason A. (2013) ‘Reformulating the Support Ratio to Reflect Asset Income and Transfers’,...
  • R. Lee et al.

    National Transfer Accounts: An Overview, Policy in Focus

    (2014)
  • Cited by (21)

    View all citing articles on Scopus

    This paper was initially prepared for the Demographic Dividend and Population Aging in Asia conference held in Honolulu on October 28–29 2015. We acknowledge financial support from the ARC Centre of Excellence in Population Ageing Research (CEPAR).

    View full text