22
Michael Fertig and Christoph M. Schmidt
Abel 2001) support this view, but they also demonstrate that many and rather
strong assumptions have to hold for this conclusion to be reliable. However, if
such a relationship between the age structure of the population and asset re-
turns exists, this might have an impact on the rate of return of investment and,
therefore, via capital accumulation on growth, wages, and (un-) employment
rates.
The existing empirical evidence on the relationship between the population
age structure and asset prices is rather mixed. Poterba (2001) analyzes the as-
sociation between the age structure of the population and the returns to stocks
and bonds in the US. Using age-wealth profiles from repeated cross-sections
of the Survey of Consumer Finances for the period 1983-1995, he demon-
strates that wealth rises sharply when households are in their thirties and for-
ties but the decline is much less pronounced as soon as they enter retirement.
Furthermore, this paper also analyzes time series data on the population struc-
ture and real returns on different financial assets for Canada, the UK and the
US. It does not find any robust evidence on the response of asset returns to
changes in the population age structure. Therefore, one should be hesitant to
predict large changes in asset values due to demographic change.
Hence, it is rather unlikely that the relative decline in labor supply due to pop-
ulation ageing will be compensated by a higher capital intensity in the produc-
tion process (Borsch-Supan 2003). Thus the growth in per-capita income will
suffer, even though domestic capital may flow increasingly to more labor-
abundant investment opportunities abroad. Therefore, to keep growth rates of
aggregate output at their current levels, a higher human capital accumulation
of younger cohorts is indispensable. Population ageing might act as an addi-
tional incentive for younger cohorts to invest in human capital, if the relative
shift in labor supply results in a rise of young workers’ wages, since for this age
group the returns to education will increase. All other things equal, this should
lead to higher human capital accumulation by the young. However, it is also
conceivable that a shrinking labor force which reduces labor market competi-
tion might counterbalance this effect, so that the net impact of population age-
ing on the human capital acquisition of younger cohorts is ambiguous. Em-
pirically, this phenomenon is quite well researched for the US (e.g. Connelly
1986; Connelly, Gottschalk 1995; Stapleton, Young 1988). These studies dem-
onstrate that due to the decline in the private returns to education in response
to larger relative cohort sizes, the human capital accumulation of relatively
large cohorts decreases.
Fertig (2003) analyzes this relationship for the case of Germany. Utilizing data
for birth cohorts entering the German labor market during the 1980s and
1990s the author demonstrates that relative own cohort sizes display a nega-
tive impact on different indicators (highest schooling and highest professional