South Africa's Generational Economy

Abstract

While countries around the world have experienced unprecedented shifts in their population

age structures over the last 70 years, it has only really been over the last 20 years that research

into the impact of the structure of the population on the economy has gained momentum.

Analytically, it is the recognition that engagement in the economy and the resulting economic

ows between individuals vary with age that underpins this impact: children consume more

than they produce; prime working-age cohorts produce more than they consume, transferring

surpluses to others or saving them for old age; and the elderly use transfers from others, asset

income or dissaving to _nance their consumption given low levels of labour income.

These economic ows at each age are quanti_ed by the National Transfer Accounts (NTA)

methodology, providing a view of the so-called generational economy and allowing us to see how

di_erent age groups produce, consume, share, and save resources at a given point in time. NTAs

are constructed from a variety of data sources|including household survey data, administrative

data, national accounts data and population data|to be consistent with National Accounts and

can be thought of as age-disaggregations of various national accounting aggregates. NTAs have

been constructed for a growing number of countries around the world, with South Africa one

of the _rst countries on the African continent to have constructed NTAs.

This thesis utilises the NTA methodology to construct partial or full accounts for South

Africa for _ve years between 1995 and 2015. These accounts are used to analyse three aspects

of the generational economy or economic lifecycle. First, the estimates are used to estimate

the magnitude of South Africa's demographic dividend, the potential economic bene_ts that

arise due to the changing population age structure as the working-age population grows relative

to the total population. Second, the NTA estimates for 2015 are disaggregated by race to

assess di_erences in the economic lifecycle across these groups, which are used as proxies for

socioeconomic status, and to assess the implications of high levels of inequality on the estimates

themselves and on projections of the pro_les into the future. Finally, by incorporating time use

data to estimate time allocations to unpaid housework and care activities for men and women

across the lifecycle, the exclusion of non-market services from the national accounts production

boundary, and therefore from NTA, is addressed, making it possible to properly assess the full

economic contributions of men and women across the life course.