Thank you, Sylvie.
The fact that men continue to earn more than women is not new. My goal here is not to provide a single, definitive answer to what the pay gap is, but rather to present the different measures that are commonly used to describe pay differentials and to show that measurement and methodology matter.
Turning to the first slide, the most commonly cited statistic on the gender pay gap is based on annual earnings. Here, women working full-year, full-time earn 72¢ for every dollar earned by men. This is recent data from 2008 that is publicly available on CANSIM.
An alternative measure is based on hourly wages. Here we see that women on average earn 84¢ for every dollar earned by men.
Why the large difference in the ratios? Well, a caveat of the earnings ratio is that it does not accurately account for differences in work volume. In 2007 full-time men worked on average four hours longer than full-time women. So you can have a gap in earnings simply because men and women worked a different number of hours.
Another caveat of the earnings ratio is that it excludes a significant portion of the workforce. Roughly 65% of women work full-year, full-time, compared with 75% of men.
The ratio based on hourly wages tends to overcome these two problems and has an added advantage of being job-specific, so it makes comparisons between men and women in their jobs much easier to do.
On the next slide we see the trends in the compensation rate ratio. The earnings ratio is a consistent time series starting in 1976, and the wage ratio provides a consistent time series beginning in 1997. Between 1976 and 1992 we see that the earnings ratio increased by about 11%, and after 1992 the earnings ratio is roughly constant.
This differs somewhat from the wage series. Beginning in 1997, we see that the wage ratio increased by 3.2% between 1997 and 2009.
Let us turn to the next slide. What are the factors that explain the gap? In describing wage differences, researchers tend to look at the attributes men and women bring to the labour market. This approach—that is, assuming that wages are tied to the individual worker—is really due to the type of data that's available to researchers; that is, large-scale household survey data. That's what has dominated the empirical literature.
Here we look at differences in work experience. We find that men on average have four years' longer work experience than women, and this difference in work experience accounts for differences in work interruptions, restrictions on number of hours worked per week or number of weeks worked per year. The fact that wages increase with experience, coupled with the fact that men and women have a different amount of work experience, explains about 11% of the wage gap.
It's also a well-known fact that both men and women have been increasing their educational attainment in recent years; however, they still choose traditional disciplines. Wages differ by field of study, and men and women are in different disciplines; this accounts for about 4% of the wage gap.
The next bullet on this slide looks at men and women belonging to different types of workplaces. This is an analysis that comes from a linked employee-employer data set. Here we learn that characteristics of the workplace explain more of the gender pay gap than the characteristics of the worker.
So when we think about the wage ratio and we adjust for things that we think matter, such as experience, education, and where men and women work, we see that on average women earn a little more than 90¢ for every dollar earned by men.