Pay Equity Committee on May 2nd, 2016

[The Chair, Liberal, Anita Vandenbeld]

I call the meeting to order.

Thank you very much to our witnesses from Statistics Canada. We have Alison Hale, who is the director of the labour statistics division, and Marie Drolet, research economist with the labour statistics division.

I believe that members have a hard copy of the PowerPoint presentation, entitled An Overview of the Gender Pay Gap. If you don't, it's being distributed right now.

Welcome.

Thank you very much.

We will start with a 10-minute presentation.

Will you both be speaking? I will call on Ms. Hale. Thank you.

[Alison Hale, Director, Labour Statistics Division, Statistics Canada]

Good afternoon.

First I want to remind everybody that today is Census Day. We're all very happy at StatsCan today. I want to remind everybody to check your mailboxes. I also want to thank you for the opportunity to present to the Special Committee on Pay Equity. Our presentation today contains data and analysis from Statistics Canada related to the gender pay gap that may inform your discussions.

The fact that men continue to earn more than women is not new. The first objective of today's presentation is not to provide a single definitive estimate of the pay gap but rather to describe the different measures that are commonly used to describe gender pay differences and to demonstrate that measurement and methodology matter. The second goal is to demonstrate how the Canadian gender pay gap has changed over time.

The most widely cited statistics on the gender pay gap are based on annual earnings. Women working full year, full time, earn 74¢ for every dollar earned by men. An alternative measure is based on hourly wages. Women earn on average about 86¢ for every dollar earned by men. Why is there such a large difference in these ratios?

One caveat with the earnings ratio is that it does not accurately account for differences in work volume. In 2015 men working full time worked about 3.3 hours longer than women working full time, so in principle there could be no gender gap in hourly wages while there was one in annual earnings simply because men work more hours than women do.

Another caveat regarding the earnings ratio is that it excludes a large and changing segment of the population. In 2013, roughly 68% of women and 75% of men worked full year, full time. Ratios based on hourly wages overcome these problems and have the added advantage of being job-specific, thereby facilitating comparisons between the wages of men and women.

Slide 3 looks just at the trend since the 1980s in these two gender-based ratios: one using annual earnings, which is the one on the bottom, the solid line, of full-time, full-year workers; and the other using an hourly wage rate of full-time workers. Between 1984 and 1992, the annual earnings ratio increased by 7.5%. After 1992, the series for the annual earnings remained relatively stagnant. This differs from the hourly wage ratio, which increased moderately throughout the period.

When they're trying to describe the gender wage gap, researchers also look at the attributes men and women bring into the workplace. In 2012 men had roughly three years' more work experience than women did. This difference is accounted for by work interruptions, by any restriction in the number of hours worked per week, or by the number of weeks worked per year. Difference in experience, coupled with the fact that wages increase with work experience, accounts for about 11% of the gender pay gap.

It is a well-known fact that educational attainment of both men and women has been rising, but there have been persistent differences between the fields of study chosen by men and women. Since wages differ by field of study, the choice of discipline accounts for about 4% of the overall gender pay gap.

The types of workplace to which men and women belong also differ. The characteristics of the workplace account for more of the gender pay gap than do differences in work characteristics, such as education and experience. Women are concentrated in low-wage workplaces. The results show that women earn about 15% less than men do when the workplace is not taken into account, compared to 8% less when workplace controls were included.

When the wage ratio is adjusted for differences in the observable characteristics of men and women and the workplaces to which they belong, women earn on average over 90¢ for every dollar earned by men.

Despite the long list of factors used in the various studies, a portion of the gender wage pay gap cannot be explained.

There are numerous other factors not accounted for that may contribute to our understanding of the gender pay differences. Take, for example, pre-labour market experiences. Personal choices, expectations of family and friends, or the education system may influence the level of educational attainment, the choice of field of study, whether to participate in the labour force, job selection, work habits, and career progression. Differences in any of these factors can manifest themselves in different labour market behaviours and ultimately in different labour market outcomes, namely wages.

Going to the next slide, and again trying to explain why the gender wage gap narrowed between 1988 and 2008, we see that between 1988 and 2008, women's real wages increased by 12%, with increases occurring in all age groups and at all points across the wage distribution, but the dramatic change occurred among older workers and those at the higher end of the wage distribution. In contrast, the real wages of men increased by 1.3%. Losses were incurred among men in some age groups and at the lower end of the wage distribution.

While women make up half the paid workforce in Canada, it is their changing relative position that matters for wages. Three trends were noted.

One is education. The proportion of women in the labour force with a university degree has nearly doubled from the early 1990s and now surpasses that of men. Most university degrees are now granted to women. The wage gap among the university-educated remained at 16% due to persistent gender differences in the major field of study. Women outnumber men in social sciences and health, while men outnumber women in mathematics and engineering.

Looking at job tenure, we see that there has been an increase in the average job tenure among women, so that there is no longer a difference when compared to men. Also, women used to be much more likely to be in a job that had just started, with a tenure time of less than three months in the 1990s. That is no longer the case.

Finally, structural changes in the Canadian economy help explain why the wage ratio narrowed between 1988 and 2008. The shift away from manufacturing jobs and a decline in unionization had a disproportionately larger impact on men's wages. Women increased their representation in high-wage occupations, such as senior management positions, throughout the period. However, the wage gap in these occupations continued to be larger than the average. This is not surprising, since increasing representation is first apparent at the lower entry-level positions within these high-wage occupations.

Other Canadian research has shown that women continue to choose a more narrow range of occupations than men and that women still represent a large portion of workers employed in the 20 poorest-paid occupations. Female-dominated occupations, such as health and education, had relatively small initial pay gaps at the beginning of this period and changed very little over time. These changes in marketable characteristics accounted for about 60% of the narrowing of the wage gap between 1988 and 2008.

Going to slide 7, I'll turn to examining the changing relationship between gender wage gap and age.

In any given year, wage gaps are larger among older workers than younger workers. For example, in 2008 the wage gap was 19.3% among those aged 50 to 54, versus 9.9% among those aged 25 to 29. Also, in year-to-year differences, wage gaps in all age groups are becoming smaller. Among persons aged 35 to 39 the wage gap was 23.2% in 1988, whereas it had shrunk to 16.3% in 2008.

In following a birth cohort over time, there is little evidence of a widening of the wage gap as workers age. For example, in 1988 the wage gap was 20.6% among those aged 30 to 34. By 2008, when this group was aged 50 to 54, the wage gap for this cohort was 19.3%, which is not much of a change.

Generally, the wage gap early in their career is a good indicator of the gap throughout a generation's working life.

[The Chair, Liberal, Anita Vandenbeld]

If I may, I'll ask you to finish up quickly.

[Alison Hale]

Yes, I'm on the summary page, so it works out well for once.

The goal of this presentation was to highlight that gender wage inequality is complex and that it requires analysis from a number of different perspectives. The narrowing of the wage gap since the late 1980s is due to compositional changes, to changes in how the labour market compensates workers, and to the fact that wages of men and women no longer diverge as they age.

There are two things to keep in mind when looking at the gender pay gap.

First, measurement matters in terms of both level and trend. In the current context, hourly wage ratios are higher than annual earnings ratios, with the earnings ratio showing little change over time, while the wage ratios continue to increase.

Second, methodology also matters. The raw or unadjusted pay gaps do not simultaneously control for a variety of factors other than gender that can influence wages. As shown, the gender pay gap narrows considerably to over 90¢ after controlling for these effects.

I hope you've found this interesting. Also included in the package are some supplementary slides that provide more information.

[The Chair, Liberal, Anita Vandenbeld]

Thank you very much for that excellent information.

We'll now go into rounds of questioning. We'll start with Ms. Dzerowicz, who has seven minutes.

[Julie Dzerowicz, Liberal, Davenport, ON]

Yes, seven minutes are more than enough. Thank you.

First, I want to begin by saying happy Census Day.

I want to also thank you so much for the excellent presentation. There's so much to wrap my mind around in a short period of time.

There's one thing I'm grappling with as we're going through and trying to think about proactive pay equity legislation, and this is at the federal level. The thing I'm trying to wrap my head around is we have a number of women who are not paid equally to men at the federal level, but they don't have an equal comparator in terms of an automatic male group. It's not as though we have men and women in a group, and all these men are paid more, and we want women to be paid the same.

What advice, or what information, do you think we should be looking at? Say there are 50 or 60 categories of women for which we're trying to see how we can create equal pay. What are the statistics, or what's the information you would advise us to look at when we're trying to create legislation that will create an equalizing force? That's my first question.

I'm going to ask the second question. We have two provinces that have put in pay equity legislation. I don't know if there's anything you can draw from them in terms of what they have done that you think has been particularly helpful or if you think there's some additional information you might want to look at. I don't know if there's anything you can comment on in terms of what they have looked at.

Those are my first two questions.

[Alison Hale]

To answer the second question, I'm not aware of any studies that have looked at the two provinces that have put in pay equity to see if there are differences.

[Marie Drolet, Research Economist, Labour Statistics Division, Statistics Canada]

On how to equalize wages across occupations, for a number of these studies that we have, we have controls for experience and job tenure. I'm not sure if the occupations you are addressing have that information associated with them. We do know that wages increase with experience. Wages do increase with occupation. We find that in a lot of our studies, if you control for the workplace, the wage gap is half as large within a workplace than it is economy-wide. Getting some information based on that may help your discussion.

I don't think I have anything else to add to job tenure and experience.

[Julie Dzerowicz, Liberal, Davenport, ON]

I'll go to a couple of more things. One thing we're told—and I think I also hear it when I go to door to door—is that the world of work is changing. We're getting a lot more contract work and precarious work. How is it in the statistics we're seeing that this impacts the wage gap between men and women? Do you have any information around that?

[Alison Hale]

It's definitely true that the world of work is changing. Our monthly labour force survey does collect information, so you can look at the wages of people in those different types of jobs, based on the characteristics of the job, to see if there is a difference. We haven't seen anything noticeable by gender in that area.

[Marie Drolet]

The gender wage gap by part-time status is small because men and women find themselves in very different part-time jobs. There's more heterogeneity within part-time jobs for women compared with men. Men are more likely to be in retail sales part-time jobs, whereas women can take on a variety of different part-time tasks. The labour force survey has information on self-employment, but I think it only has annual earnings associated with that, and not an hourly wage rate.

[Julie Dzerowicz, Liberal, Davenport, ON]

Do I have any time left?

[The Chair, Liberal, Anita Vandenbeld]

Three minutes.

[Julie Dzerowicz, Liberal, Davenport, ON]

One of the many interesting things that you've mentioned is that while more and more women are going to university and becoming more educated, more women, you said, still choose 20 of the poorest occupations. Can you give us a few examples of that and tell us why that might be the case?

[Marie Drolet]

In terms of university education, we still see a gender difference in major field of study. Wages vary by major field of study, so you're going to see a gender wage gap with that.

Some of the Canadian research has shown that among university graduates, the gender gap is small when you first graduate but tends to get larger five years later. For example, in the 1995 cohort of graduates, the wage gap was about 6% or 7%. Five years after the fact it was more like 13% or 14%, so you see that it increases over time.

In the 12 lowest-paid occupations, we still see that women tend to choose a more narrow range of occupations than men. I think two-thirds of women are in education, health, office support, and sales and service occupations. In terms of the 20 poorest-paid occupations, sales support and office and clerical workers are the lowest paid.

[Julie Dzerowicz, Liberal, Davenport, ON]

Thank you.

I'm passing it over to Sonia.

[Sonia Sidhu, Liberal, Brampton South, ON]

Thanks to the panel for being with us.

Ms. Hale, you bring up some very interesting sector-by-sector analysis on the wage gap. Where would you direct us as Parliament to look in terms of work already done in this regard, whether in terms of sector or regional analysis?

[Alison Hale]

You'll notice in the presentation that there are some references to papers already done by Marie, who has looked at things across Canada.

Marie, are you able to answer?

[Marie Drolet]

I'm not sure about the regional analysis. I haven't personally looked at that. One of my studies looked at the importance of the workplace and workplace characteristics. That could address what you're aiming for with sector- or industry-type analysis. In that particular analysis I looked at non-profit firms, different workplace practices, and how that may impact the gender wage gap.

I don't have the reference here.

[Alison Hale]

We can send you some of the papers that Marie has written on some of that.

[The Chair, Liberal, Anita Vandenbeld]

Send those to the clerk, and then we'll distribute them to the committee.

That's the end of your time, so we are now going to Ms. Gladu for seven minutes.

[Marilyn Gladu, Conservative, Sarnia—Lambton, ON]

Excellent.

I want to welcome you and thank you for being here.

As a statistical specialist for a global business for a few years, I want you to know I love StatsCan.

I'm going to start with a question that has to do with overall compensation as opposed to the hourly wage. Do you have any information on how much of that extra gap that occurs is due to bonuses and other forms of compensation?

[Alison Hale]

The information we were providing today would include overtime but not bonuses. We don't have that information.

[Marilyn Gladu, Conservative, Sarnia—Lambton, ON]

I see that as an area where there is an opportunity to discriminate in a sneaky way, if I could say so. I think that will be important.

In the work that you talked about, you said that there were workplace characteristics that contribute to the devaluing of a woman's work. Can you talk a little bit more about the workplace characteristics that are evaluated and which ones specifically you think are undervalued?

[Marie Drolet]

In that particular study, the statistics that Alison cited in saying that the wage gap is half as large within a workplace as it is economy-wide used information from our workplace and employee survey, and that was circa 1999 or so. In that particular one we used workplace size and industry as control variables. There are other workplace characteristics. I'm just trying to remember what they all were. I didn't go over that particular study because it was a bit older. There was a non-profit flag. I do think there was something to do with different compensation, such as bonuses and what have you, but that was mostly when I was able to control for the workplace.

Most of these studies cited here always believe that wages are tied to the individual worker and not necessarily to the workplace. That was the first Canadian study that brought in the workplace characteristics.