Abstract
We examine the extent of persistence in the gender unemployment gap of eight OECD countries: Australia, Canada, Finland, France, Germany, Italy, Japan and the United States using data over the period 1965 to 2002. Although we are unable to reject the unit root null hypothesis for all countries with the exception of Finland and Italy, the half-life measure suggests that the extent of persistence is relatively low for all countries.
Acknowledgement
The author Amit Sen acknowledges support from a Faculty Development Grant at Xavier University.
Notes
1 Following Sen (Citation2003), we use the mixed model specification of the trend-break alternative that allows for a simultaneous break in the intercept and slope of the trend function.
2 In practice, the value of the lag-truncation parameter (k*) is unknown, and so a data-dependent method for choosing the appropriate value of k* is used. In this paper, we use Perron and Vogelsang's (Citation1992) k(t-sig) method for selecting the lag-truncation parameter. First, we specify an upper bound ‘kmax’ for the lag-truncation parameter. The chosen value of the lag-truncation parameter (k*) is determined according to the following ‘general-to-specific’ procedure: the last lag in an autoregression of order k* is significant, but the last lag in an autoregression of order greater than k* is insignificant. The significance of the coefficient is assessed using the 10% critical values based on a standard Normal distribution.
3 Owing to the possibility that the mixed model statistic may not reject the unit root null hypothesis due to low power, we also calculated the conventional augmented Dickey-Fuller (ADF) unit root statistics. We are unable to reject the unit root null hypothesis with the ADF statistics for all countries except Finland. We do not report these results in the paper to save space, but they are available from the author upon request.