Evidence on the dynamics of unemployment by gender

Abstract

We present empirical evidence regarding unemployment dynamics for women and men in eight OECD countries. Unit-root tests are used to examine the unemployment dynamics of women and men. Failure to reject the unit-root hypothesis is consistent with unemployment hysteresis. Rejection of the unit-root hypothesis indicates that unemployment dynamics are best explained by the natural rate of unemployment or the structuralist view. We find evidence of gender differences in unemployment dynamics in Canada, Germany and the US, but not in other countries. While there are some differences in the extent of persistence across gender and across countries, the degree of persistence for both female and male unemployment rates is fairly low in all countries. Our results, therefore, contrast with substantial empirical evidence of high levels of unemployment persistence in European countries.

Notes

¹There are a few studies that look at differences in unemployment persistence across labour force groups. For instance, Tolvi (2003) finds less unemployment persistence for women than for men in Finland.

²It is well accepted among economists that demographic characteristics influence unemployment persistence over the business cycle. For example, Darby et al. (1985) find that the influx of women into the labour force in the 1970s increased the natural rate of unemployment by about 1% in the United States.

³The effect of these factors on the dynamics of female and male unemployment rates is discussed in more detail in Section III.

⁴We note that although a gender unemployment gap still exists in several OECD countries, there is evidence of convergence between the female and male unemployment rates in a number of OECD countries. See Queneau and Sen (2006) for further details.

⁵See Sen (2003) for further details regarding the choice of the form of break. Not withstanding the choice of the form of break, we also used the unit-root test designed for the trend-break stationary alternative that allows for only a break in the intercept (the Crash model). The empirical results, available upon request, do not alter the main findings of this article.

⁶It is relatively well accepted among economists that labour market institutions such as the unemployment insurance system, the level of employment protection and labour taxes cause unemployment persistence. For instance, Nickell et al . (2005) demonstrate that changes in labour market institutions account for about 55% of the increase in European unemployment over the period 1960 to 1995.