cated people aged 55 and older as Besl and Kale
predict for the United States.
Government policy and social and economic
developments appear to be causing the change of
direction in the participation rate of the 55-plus
group in the United States and may have a similar
impact in Canada in the future. These include
less-generous retirement support by private and
public sectors and a higher percentage of older
women who need to be financially independent,
after three decades of increased incidence of mar-
riage breakdown (Besl and Kale, 1996). Other
likely influences on the decision by Americans
about when to leave the workforce mentioned by
Besl and Kale are legislation eliminating manda-
tory retirement and outlawing age discrimination,
fiscal constraints on the growth in Social Security
benefits and the operation of medicare, which is
projected to run out of funds before 2002, and
regulations implemented over time to raise the
normal retirement age for full Social Security
benefits from 65 to 67. All these considerations
lead Besl and Kale to expect higher labour force
participation rates among older adults in the next
century. Labour force projections that assume
that participation rates for those aged 55 and older
will remain at the current low levels may be overly
pessimistic and may underestimate the size of the
future labour force.
Finally, the rising trend in the ratio of the self-
employed to total employment may have a signifi-
cantly positive effect on the participation rate,
since these people tend to retire later than salaried
employees (Gower, 1997).33 The acceleration in
the growth of this ratio in the mid-1990s may,
however, be temporary to the extent that self-em-
ployment was a second-best solution for some of
those who took early retirement as part of the
massive reduction in employment in the public
sector.
A plausible outcome for the 55-plus age group
in Canada would be a relatively strong increase —
as much as four to five percentage points — be-
tween 1996 and 2006. This compares with the
BLS projection for the United States of a rise of 6.5
percentage points for this age group over the same
period (Fullerton, 1997).
Conclusion
This article has attempted to develop a synthe-
sis of various pieces of information about past
structural influences on the participation rates of
Canada and the United States with a view to hy-
pothesizing how much potential underlying
strength there is over the next few years.
Compositional and structural factors appear to
have played a significant role in the decline in the
participation rate in Canada over the 1990s and in
the different performance between the two coun-
tries. Rough estimates of the effect of the shifts in
weights of demographic groups in the working
age population and between students and non-
students appear to account for one to 1.5 percent-
age points of the 2.69 percentage point decline
from 1989 to 1997, or 37 per cent to 56 per cent.
This is not a minor effect and strongly suggests it
was unreasonable to expect the participation rate
to have regained the peak level of 67.5 per cent
during this time. Compositional effects also ap-
pear to account for about 30 per cent of the differ-
ence between in the change in the two countries’
participation rates over the period.
In addition, structural factors such as the con-
vergence of female participation rates to male
rates and a continued trend in early retirements
among 55 and over males have all helped to put
downward pressure on the aggregate participa-
tion rates.
Over the next decade, Canada, as well as the
United States and many other members of the Or-
ganization for Economic Co-operation & Develop-
ment, will experience a rising population share of
older people, who have on average the lowest par-
ticipation rates of the major groups. This change
in composition would result in a decline in the ag-
gregate participation rate without offsetting in-
creases in the participation rates of age- pecific
groups (Chart 5). There is unlikely to be a much
larger increase in the school attendance rate, as in
the case of teens it appears to be approaching a
saturation point. However, because a large pro-
portion of the increase appears to have been
structural, the impact will be permanent.
In the case of the largest group — core-age
workers — the participation rate appears to have
room to increase moderately. As workers enter
the core-age group with higher levels of education
than those leaving it, there may be a better match
Summer 1999
Canadian Business Economics
13