Population Research and Policy Review 18: 261–277, 1999.
© 1999 Kluwer Academic Publishers. Printed in the Netherlands.
Age structure and economic policy:
The case of saving and growth
Department of Economics, Uppsala University, Uppsala, Sweden
Abstract. The age structure of the population affects aggregate saving, which affects growth
through investment. Growth in turn is inﬂuenced by other age structure effects and feeds back
into aggregate saving by well known life cycle mechanisms. Some of these feedbacks are
generally ignored in empirical work. Especially the age structure effect on macroeconomic
variables is a commonly overlooked, yet easily accessible factor useful for prediction, policy
evaluation and design. The connection between age structure, savings and growth in the OECD
from 1950 to 1990 illustrates how policy analysis that ignores the macroeconomic effects
and feedbacks from age structure changes is liable to lead to faulty and potentially costly
conclusions about policy issues.
Keywords: Demography, Europe, Growth, Panel study, Saving
The age distribution of the population is a useful tool for understanding the
economy, not a deplorable imperfection to control for. Panel regressions using
age structure, growth and saving in the OECD countries 1950–1990 show the
patterns expected from life cycle and human capital models. Although this
result probably reﬂects aggregation properties rather than individual behavior,
it may nevertheless be useful for forecasting macroeconomic performance.
The efﬁcacy and relevance of economic policy depend on the quantita-
tive effects of changes in policy parameters. By necessity inferences have
to be made from the historical experience of actual economies. The pre-
dicted effects are, therefore, contingent on institutional facts, technological
and cultural background, demographic structure etc. It is clearly impossible
to allow for the potential changes in every such background variable. The age
structure can, however, be fairly accurately projected several years into the
future. Individual economic behavior changes profoundly over the life cycle.
Consequently, variations in the age composition of the population change
aggregate economic behavior.
The design of pension systems, health care, education and so forth, is un-
avoidably concerned with the age distribution of the population. However, the