Changing Business Cycle Dynamics in the US: The Role of Women's Employment

I examine the effects of female labor market behavior on the dynamics of aggregate employment and hours in the trend and over the business cycle in the US. I argue that the steep increase in women's labor force participation throughout the 1970s and 1980s, and its flattening out since the 1990s can contribute to explain three puzzling phenomena experienced by the US: the non-stationarity of aggregate hours, the decrease business cycle volatility of labor market variables before the great recession, and the recent jobless recoveries. To develop the analysis, I first construct an aggregate time series for hours by gender similar to the series used in in aggregate business cycle analysis, and provide descriptive empirical evidence supporting the hypothesis proposed in this paper. I then develop and estimate a dynamic stochastic general equilibrium model that allows for gender differences in hours and wages to assess the implications of the changing trend in female participation on the behavior of aggregate variables. I find that female specific shocks explain a substantial fraction of the volatility of aggregate outcomes, both at the business cycle frequency and in the longer run. Using a number of counterfactuals, I show that the trend in female participation and its variation over time can rationalize the three phenomena of interest. These findings have implications for the behavior of aggregate labor market variables in other advanced economies where female labor force participation is experiencing secular changes.