20
Michael Fertig and Christoph M. Schmidt
early retirement. Women, in particular, have displayed a much more complex
labor supply behavior than men in all advanced economies, evidenced by a
large literature on the labor supply of married women (form a seminal source
see Killingsworth, Heckman 1986). In times of high unemployment home pro-
duction is becoming relatively attractive, and reduced labor market participa-
tion might be the safety valve to mitigate the consequences of generational
crowding.
These arguments call for a concentration of the empirical analysis on employ-
ment rates instead of unemployment rates. This is exactly the route followed
by our own empirical study discussed in the next section. It seems clear,
though, that several confounding developments, most importantly the rising
secular trend towards higher labor force participation associated with the
higher education and lower fertility of more recent generations of women will
need to be disentangled empirically from the effects of generational crowding.
These complications call for pursing a cross-country perspective, and for bas-
ing the analysis on micro data, since this permits taking educational attain-
ment into account.
In essence, if rigidities prevent the wage structure from adjusting completely
to variations in relative factor scarcities, the consequences will arise in terms of
employment and unemployment. Previous literature placed its attention on
age-specific unemployment rates. The empirical evidence on the relationship
between cohort sizes and unemployment is very mixed. Korenman/Neumark
(2000) examine the effect of changes in the population age structure on youth
employment and unemployment rates utilizing time-series data for 11 Euro-
pean countries as well as Australia, Canada, Japan and the US for a period cov-
ering 1970-1994. Their results suggest that large youth cohorts lead to in-
creases in the youth unemployment rate, but, if, any modest effects on the em-
ployment rates of youths. The authors also report weak evidence that labor
market institutions that decrease flexibility lead to more pronounced re-
sponses of youth (un-) employment rates to changes in the population age
structure.
Shimer (2001) uses data from all states in the US for the period from 1978 to
1996 to investigate the reaction of the (youth) unemployment and labor mar-
ket participation rate to changes in the population age structure. The results of
his instrumental variable approach suggest a rather large negative effect of an
increase in the relative size of younger cohorts on the unemployment rate of
younger and older workers, i.e. the larger the relative size of younger cohorts
the lower are the unemployment rates of both age groups. Furthermore, the
author reports an increase in labor market participation rates for younger and,
although less pronounced, for older workers in response to an increase of the
relative size of younger cohorts.