Evolution of the Female Labour Force Participation Rate in Canada, 1976-1994: a Cohort Analysis

The participation rate of women aged 25-64 rose greatly in the 1970s and 1980s, but has stagnated in the 1990s. In principle, this development could reflect either the poor growth performance of the economy this decade or the completion of the integration of women into the labour force. In the fourth article of this symposium, Paul Beaudry and Thomas Lemieux use a cohort analysis to shed light on the explanation of this stagnation in female labour force participation. Using data from the Survey of Consumer Finances for the 1976-94 period, the authors track the participation rates over time of representative groups of women who entered the labour force at different points in time. They decompose a cohort’s participation rate into three effects: a macroeconomic effect common across cohorts linked to factors such as recessions and employment insurance generosity; an age or life-cycle effect; and a cohort-specific effect which shows the differences between cohorts for a given age and macroeconomic effect. The authors find that the cohort effects are likely the dominant factor in explaining the recent stagnation of female participation, just as it explained the large increases in the 1970s and 1980s. The recession of the early 1990s, which according to the authors reduced the female participation rate by 1 percentage point, merely amplified the stagnation phenomenon. As the cohort effects stabilize with the narrowing of the gap between male and female participation rates, the stagnation would have occurred, albeit later in the 1990s, even if more favourable macroeconomic conditions had prevailed. The authors conclude that there is still room for a 2-3 percentage point increase in the participation rate of women 25-64, but the magnitude of the increases of the 1970s and 1980s is not possible as the cohort effects that prevailed then no longer exist. The authors stress that their results are dependent on the amount of flexibility used to capture the cohort effect so that the age profile and its slope can trace both the rise and the flattening of the participation rate by age. They point out that over time participation behaviour of women 25-64 is converging toward that of men, namely, high and flat participation profiles to at least age 55. They also note that the much smaller increase in the female participation rate in the United States in the 1990s relative to the 1970s and 1980s despite the robust U.S. labour market supports their findings as the cohort effects were also levelling out south of the border.