Inégalité des hauts revenus et l écart salarial homme-femme

Transcription

1 Conférence François-Albert-Angers, Congrès SCSE, le 11 mai 2017 Inégalité des hauts revenus et l écart salarial homme-femme Nicole Fortin Vancouver School of Economics 1

2 Conférence François-Albert-Angers, Congrès SCSE, le 11 mai 2017 Inégalité des hauts revenus et l écart salarial homme-femme avec la collaboration de Aneta Bonikowska et Marie Drolet 2

3 3

4 24% U.S. Income Share: Top 1% vs. Bottom 50% 22% Income Share 20% 18% 16% 14% 12% 10% 8% Top 1% Bottom 50% 6% 4% Pre-tax national income Pre-tax national Income Fiscal income Source: Piketty, Saez, Zucman (2016) Appendix Table II-B1 4

5 Source: Alvaredo, Atkinson, Piketty and Saez (2013)

6 Source: Alvaredo, Atkinson, Piketty and Saez (2013)

7 Income Share 24% 22% 20% 18% 16% 14% 12% 10% 8% 6% 4% Canada - Income Shares: Top 1% vs. Bottom 50% Bottom 50% Top 1% Source: CANSIM Table , from LAD data Top 1% - with capital gains Top 1% -After tax Bottom 50% - Before tax Bottom 50% - After tax

8 Part du dernier 1% de revenu 16% 14% 12% 10% 8% 6% Figure 1. Part du dernier centile de revenu (P99-P100) dans les revenus déclarés totaux Canada - Incluant les gains de capital Canada - Excluant les gains de capital Québec - Incluant les gains de capital Québec - Excluant les gains de capital 4% Source: Fortin (2017), CANSIM Tableau , from LAD Data. 8

9 Largest Increases in Top Income Shares in Countries with Long Standing High Share of Women in LF Women's Share of the Labour Force United States Canada United Kingdom France Germany Sweden Source: BLS and Conference Board, International Labor Statistics, adjusted to US concepts, persons aged 15/16 and over

10 In 2015, women hold 8.5 per cent of the highest-paid positions in Canada s top 100 listed companies, according a report by global executive search firm Rosenzweig & Company. That s almost double the 4.6 per cent recorded in 2006, the first year Rosenzweig studied women in executive positions, and up from 7.4 per cent in But in real terms, it represents 45 women in top jobs such as chief executive, chief financial officer or vice-president of a company, including eight female CEOs. 10

11 Call for Action to Improve Gender Diversity in Top Jobs OSC proposes rules for women on boards, but stops short of quotas Barbara Shecter Financial Post January 16, :19 AM ET Companies listed on the Toronto Stock Exchange will have to disclose how many women they have on their boards and in their executive ranks as well as set targets for the future, the Ontario Securities Commission said Thursday. But the proposals fall short of quotas. OSC rebukes firms for lack of action on gender-diversity rules Janet McFarland Wednesday, Jun. 10, :23PM EDT Following mandatory disclosures rules implemented on January 1 st, 2015, many companies have shown bare technical compliance with the new gender-diversity reporting rules and it is simply not good enough, Ontario Securities Commission chair Howard Wetston says. 11

12 Increasing Earnings Inequality in Top Incomes and the Gender Pay Gap Questions of interest: 1) What are the consequences of the under-representation of women in top jobs for the overall gender pay gap? 2) How is it contributing to the slowdown in the convergence of female/male pay? 3) What public policies and firm practices are effective to improve this under-representation? 12

13 Increasing Earnings Inequality in Top Incomes and the Gender Pay Gap Apply the approach used in the analysis of earnings inequality in top incomes (developed by Thomas Piketty, Emmanuel Saez, and co-authors) to characterize top earners Use positional ranks to construct a proxy of vertical segregation at the top of the earnings distribution (Fortin and Lemieux, 1998; Bayer and Charles, 2016) Apply reweighing techniques à la DiNardo, Fortin and Lemieux (1996) [DFL] to construct counterfactual gender pay gaps 13

14 Increasing Earnings Inequality in Top Incomes and the Gender Pay Gap Appeal to administrative data to capture the highest incomes As in the analysis of top incomes shares, we will use all earnings data from the Canadian Longitudinal Worker Files (LWF, ) Show comparative analyzes for Sweden (LISA, ) and for the United Kingdom (ASHE, ) Supplemented by hourly wage data from public use Canadian (CAN-LFS, ) and UK Labour Force Survey (UK-LFS, ) 14

15 Increasing Earnings Inequality in Top Incomes and the Gender Pay Gap When residual inequality experienced stupendous increases in the 1980s, Blau and Kahn (1997) coined the term swimming upstream to characterize women s pursuit of pay equality in the face of countervailing currents. Have recent increases in top incomes lead to similar effects, therefore accounting for the slower progress in the gender pay and growing unexplained (by traditional factors) share? To the extent that some of the increases in top incomes are associated with excesses in rent seeking rather than innovation (Aghion s Carlos Sim vs Steve Jobs), curtailing those excesses would slow the countervailing currents 15

16 Canadian Data Longitudinal Worker File (LWF) LWF is a 10% random sample of all Canadian workers Years: integrates data from the T1 and T4 files of Canada (CRA) and the LEAP (Statistics Canada) Annual earnings from all jobs, include bonuses, honorariums, etc. Selected if > half of minimum wage earnings equivalent Select workers age 25 to 64 Labour Force Survey (LFS) Public Use Monthly survey on approximately 100,000 individuals rotating 6- months panel sample design Years: Hourly wage of employees from main job Selected if > half the minimum wage Select workers age 25 to 64 16

17 Canadian Data Longitudinal Worker File (LWF) No self-employment income No labour supply information Top coded at P99.99 $2,000,000 in 1983 to $10,000,000 in 2000 Available covariates: union coverage, age, industry CPI adjusted to 2010$CAN Labour Force Survey (LFS) Public Use No self-employment income Number of weeks worked unavailable Top-coding (P99.9) from $95/hour in 1997 $125/hour in 2015 At 2080 (=52wk*40hrs) hrs/year, from $200,000 to $260,000 Available covariates: age, union, education, occupation, industry, firm size, etc. 17

18 Trends 1) Evolution of female/male labour force participation a) Extensive margin (LFP) b) Intensive margin (hours of work) 2) Evolution of female/male average wage and earnings ratios 3) Evolution of female shares across top percentiles of the overall distribution of wage and earnings 4) Counterfactuals with alternative simulations 18

19 Steep Growth in Women s Labour Force Participation* Followed by a Leveling-Off Participation Rate Labour Force Participation Rate - Ages 25 to Canadian Men Canadian Women Quebec Men Quebec Women Source: Fortin (2017), LFS Public use files, *Labour force participants include employed (at work or on-leave) and unemployed individuals 19

20 Decline in LFP after the Great Recession in the US Source: Blau and Kahn (2016) 20

21 The Women s Liberation Movement of the 1960s and The Pill Goldin and Katz (2002) and Bailey (2006) point out to important changes in women s LFP occurring in the 1960 s Women born after the mid-1950s had access to reliable contraception More likely to pursue higher education and enter life-long careers Accompanied by a decline in traditional gender roles attitudes which stabilized in the mid-1990s in the U.S. (Fortin, 2015) Before married women were more likely `secondary workers who entered the labour market when kids were in school Mulligan and Rubinstein (2013) argue that the closing of the gender pay gap is largely due to changing selection of women into the labour market 21

22 Generational Effects in the Growth of Women s LFP 1.0 Women's Labour Force Participation by Synthetic Birth Cohort Participation Rate < All Source: Fortin, Drolet and Bonikowska (2016), LFS public use files, ages 25 to 64 year 22

23 Generational Effects in the Growth of Women s LFP 1.0 Women's Labour Force Participation by Synthetic Birth Cohort Participation Rate All Source: Fortin, Drolet and Bonikowska (2016), LFS public use files, ages 25 to 64 year 23

24 Continued Gender Convergence? According to the Mincer-Polachek hypothesis (1974), gender differences in experience and labour force attachment are the key determinants of the gender wage gap. Blau and Kahn (2016) found that declining gender differences in experience in the United States accounted for % of wage convergence between men and women over the period. Going forward, Goldin (2014) suggested that the impact of work force interruptions for family responsibilities should be understood in the context of temporal flexibility (or the lack thereof) in impacting the gender wage gap. 24

25 Less Convergence in Gender Gap in Hours 1.00 Gender Ratio in Average Total Weekly Hours by Synthetic Birth Cohort < All Source: Fortin, Drolet and Bonikowska (2016), LFS data, ages 25 to 64 year, employed with positive hours of work, usual hours from all jobs 25

26 Gender Gap in Hours and Increasing Earnings Inequality in Top Incomes Kuhn and Lozano (2008) had shown increases in long hours of work (>48 hours a week) among highly educated highly-paid older men was greatest in detailed occupations and industries with larger increases in residual wage inequality. Pointing to some high penalty for flexibility in some high wage occupations, Goldin (2014) further conjectures that rewards to working long hours are an obstacle for the gender gap in pay to vanish Cortes and Pan (2015) find that highly competitive jobs (O*NET characteristics) also have long hours Cortes and Pan (2016) find that across countries long hours lowers the share of married women in corresponding occupations 26

27 Trends 1) Evolution of female/male labour force participation a) Extensive margin (LFP) : substantial convergence b) Intensive margin (hours of work): less convergence 2) Evolution of female/male raw wage and earnings ratios: Continuing progress? 3) Evolution of female shares across top percentiles of the overall distribution of wage and earnings 4) Counterfactuals with alternative simulations 27

28 What is the ratio of women s to men s earnings on average in Canada? Source: Drolet (2011) 25 to 54 year olds, various data sources. 28

29 What is the ratio of women s to men s earnings on average in Canada? Hourly Wage ratio 85% is the preferred measure to consider whether employers treat women fairly and should be used in statements such as women earn 85 cents out of every $1 men earn Annual Earnings ratio 65% mixes the number of hours worked with how much is earned by hour provides a better measure of the welfare of women More women work part-time, many women working full-time full-year work less hours a week than men (clerical vs. industrial workers) But the All Annual Earnings measure is the only one available for the very top income groups 29

30 1.00 Generational Effects in the Gender Pay Gap Gender Gap in Hourly Wages by Synthetic Birth Cohort All Source: Fortin, Drolet and Bonikowska (2016), LFS data, ages 25 to 64 year, hourly wage on the main job 30

31 Generational Effects in the Gender Pay Gap 1.00 Gender Gap in Annual Earnings by Synthetic Birth Cohort All Source: Fortin, Drolet and Bonikowska (2016), LWF data, ages 25 to 64 year, 3-year moving average annual earnings from all jobs 31

32 Standard Decomposition of the Gender Pay Gap The Oaxaca-Blinder decomposition starts with gender-specific OLS regressions of individual characteristics on (log) wages: YY gg = XX gg ββ gg + εε gg, g = m, f Constructs a counterfactual wage such as what would be the average wage of women if they had the same characteristics as men YY ff mm = XX mm ββ ff = qqqqqqqqqqqqqqqqqqqq mm pppppppppp ff Divides the average gender pay gap into explained and unexplained part YY mm YY ff = (YY ff mm YY ff ) + ( YY mm YY ff mm ) = ( XX mm XX ff )ββ ff + XX mm (ββ mm ββ ff ) explained unexplained 32

33 Gender Pay Gap Largely Unexplained by Human Capital Variables For the United States, Blau and Kahn (2016) using human capital variables, including actual experience from the PSID, find a notable decline in the unexplained gap from log points in 1980 to log points in But as a share of the gender gap in both years, the unexplained portion is actually a larger share of gap in 2010 (85%) than in 1980 (71%). For Canada, Baker and Drolet (2010) also report some progress in the unexplained gap from log points in 1981 to log points in But this represents an increase, from 1981 (61%) to 2008 (85%), in the share of gap that is unexplained by education, occupation and industry. 33

34 Source: Blau and Kahn (2016) 34

35 Gender Pay Gap Largely Unexplained by Human Capital Variables Baker and Drolet (2010) explain that in many dimensions, such as education, women increasingly have an advantage over men. But because women s wages have not seen commensurate increases, these are countervailing factors to explain the gap. They argue that most significant exception to this is the industrial distribution of employment in which men maintain a significant advantage. 35

36 Industry Composition Largest Single Explanatory Factor* in the Private Sector Yet, Schirle (2015) finds that in most provinces more than half of the gap is unexplained Source: Schirle (2015), LFS 2014, hourly wages of private sector full-time employees, ages * with the exception of Manitoba 36

37 Trends 1) Evolution of female/male labour force participation a) Extensive margin (LFP) b) Intensive margin (hours of work) 2) Evolution of female/male average wage and earnings ratios: Slower progress in recent years, share of the gap unexplained has increased, industry may remain a potent explanatory variable 3) Evolution of female shares across top percentiles of the overall distribution of wage and earnings 4) Counterfactuals with alternative simulations 37

38 Soaring Top Incomes in the United States Share of total income for each group 25% 20% 15% 10% 5% 0% Decomposing Top 10% into 3 Groups, Top 1% (incomes above $392,000 in 2013) Top 5-1% (incomes between $165,500 and $392,000) Top 10-5% (incomes between $116,500 and $165,500) Source: Piketty and Saez, 2003 updated to Series based on pre-tax cash market income including realized capital gains and excluding government transfers

39 Gender Gap in Top Incomes Follow Guvenen, Kaplan, and Song (2014) in using the thresholds of the wage and earnings distribution for men and women combined Depart from the traditional literature on the glass ceiling which compares the pay gap at percentiles of the gender-specific distributions Depart from most of the literature which uses the logarithm of wages or earnings in order to emphasize the top end Allow for the construction of counterfactuals to study the underrepresentation of women in top income groups 39

40 Thresholds of Top Incomes Men and Women Combined Longitudinal Worker File (LWF) Annual Earnings for all jobs 1) Top 0.1% > $662,860 2) Top 1% > $206,785 3) Top 10% > $92,000 Labour Force Survey (LFS) Hourly wages on the main job 1) Top 0.1% > $66 ( $128,705) 2) Top 1% >$53 ($116,922) 3) Top 10% >$35 ($80,352) at 2080 hours 40

41 Larger Increases for Top Earners in Canada 500, , , , ,000 Canadian Annual Average Earnings ($CAN 2010) 6%/yr 2,300,000 2,100,000 1,900,000 1,700,000 1,500, , , , ,000 50, %/yr 1.25%/yr 0.6%/yr Bottom 90% Next 9% Next 0.9% Top 0.1% (right axis) 1,300,000 1,100, , , , ,000 Source: LWF , years old, Annual earnings from all jobs 41

42 Some Increases in Lower Groupings in Sweden Swedish Annual Average Earnings (SEK 2010) SEK 10, %/yr 5.8%/yr 2.3%/yr 1.8%/yr SEK 100, Bottom 90% Next 9% Next 0.9% Top 0.1% (right axis) Source: Fortin, Bell, and Boehm (2017), LISA data. 42

43 Top Income Groups Hit by the Financial Crisis in the UK United Kingdom Average Annual Earnings ( 2010) 200, , , %/yr 900, , , , , %/yr 75,000 50, %/yr 25, %/yr Bottom 90% Next 9% Next 0.9% Top 0.1% (right axis) Source: Fortin, Bell, and Boehm (2017), ASHE data. 600, , , , ,000 43

44 Slower Convergence in Share of Women among Top Earners in Canada 0.60 Share of Women in Selected Percentiles of Annual Earnings All Bottom 90% Next 9% Next 0.9% Top 0.1% Source: Fortin, Drolet and Bonikowska (2016), LWF , years old, Annual earnings from all jobs 44

45 Similar Trends in Female Shares in Sweden and the UK Share of Women in Selected Percentiles of Annual Earnings - Sweden Share of Women in Selected Percentiles of Annual Earnings United Kingdom All Bottom 90% Next 9% Next 0.9% Top 0.1% All Bottom 90% Next 9% Next 0.9% Top 0.1% 45

46 Gender Differences in Hourly Wage Distributions Density A Top 10% Top 1% Top 0.1% Women Men Density B Top 10% Top 1% Top 0.1% Women Men Hourly Wage ($2010) Hourly Wage ($2010) Source: Fortin, Drolet and Bonikowska (2016), LFS , years old, Hourly wage from the main job 46

47 Gender Differences Top 10% in Hourly Top 1% Top Wage 0.1% Distributions Density A Top 10% Top 1% Top 0.1% 3:2 Women Men 4:1 Density :1 B Top 10% Top 1% Top 0.1% Women Men Hourly Wage ($2010) Hourly Wage ($2010) Source: Fortin, Drolet and Bonikowska (2016) Computation, LFS , years old, Hourly wage from the main job 47

48 Gender Differences in Hourly Wage Distributions - UK A D Density Top 10% Top 1% Top 0.1% Women Men Density Top 10% Top 1% Top 0.1% Women Men Hourly Wage ( 2010) Hourly Wage ( 2010) Source: UK-LFS, years old, Hourly wage from the main job 48

49 Under-representation of women in top jobs makes for a less favorable overall gender pay ratio 1.10 Female-Male Average Hourly Wages Ratios by Selected Percentiles - Canada All Bottom 90% Next 9% Next 0.9% Top 0.1% Median Source: Fortin (2017), LFS , years old, Hourly wages from the main job 49

50 Under-representation of women in top jobs slows progress in the overall gender pay ratio 1.10 Female-Male Earnings Ratios by Earnings Percentile - Canada All Bottom 90% Next 9% Next 0.9% Top 0.1% Source: Fortin, Drolet and Bonikowska (2016), LWF , years old, Annual earnings from all jobs 50

51 Similar Differences in Ratios in Sweden and the UK No Upward Trend in Gender Earnings Ratio in Top 0.1% Female-Male Earnings Ratios by Earnings Percentile - Sweden Female-Male Earnings Ratios by Earnings Percentile - United Kingdom All Bottom 90% Next 9% Next 0.9% Top 0.1% All Bottom 90% Next 9% Next 0.9% Top 0.1% Source: Fortin, Bell, and Boehm (2017), workers ages 25 to 64, LISA data for Sweden, ASHE data for the UK. 51

52 Trends 1) Evolution of female/male labour force participation a) Extensive margin (LFP) b) Intensive margin (hours of work) 2) Evolution of female/male average wage and earnings ratios: 3) Evolution of female shares across top percentiles of the overall distribution of wage and earnings 4) Counterfactuals with alternative simulations a) Using male shares in selected percentile earnings b) Estimating standard O-B with selected percentile of hourly wages 52

53 Counterfactual Gender Pay Gaps and Reweighting Kline (2011) shows that the counterfactual (letting DD ii = 1 denote male), μμ 0 1 = EE[XX ii DD ii = 1] ββ 0 can be computed from an OB regression μμ 0 1 = EE[XX ii DD ii = 1] EE[XX ii XX ii DD ii = 0] 1 EE[XX ii XX ii DD ii = 0] Or using reweighting à la DFL μμ 0 1 = EE[ww(XX ii )YY ii DD ii = 0] where ww(xx ii ) P( XX ii DD ii =1) P( XX ii DD ii =0) = 1 ππ ππ ee(xx ii ) (1 ee(xx ii )) with ππ P(DD ii = 1) and ee(xx ii )= P( DD ii = 1 XX ii ), under the assumptions of common support ee(xx ii )< 1 and conditional independence (YY 1 ii, YY 0 ii ) DD ii XX ii 53

54 Counterfactual Gender Pay Gaps and Reweighting The sample analogues are: ππ = NN 1 NN and 1 ππ ππ = NN 0 NN 1 If XX ii is a j-category variable, ee(xx iijj ) = NN 1jj ee(xx and ii ) NN jj (1 ee(xx ii )) = NN 1jj, NN 0jj So that reweighing observations requires only the ratio of shares in each j- category: ww(xx iijj )= NN 0 NN 1jj = SS 1jj NN 1 NN 0jj SS 0jj where SS 1jj is the share of group 1 in category j With conditional means, the overall mean is YY 0 = ii SS 0jj YY 0jj, so that YY oo 1 = jj SS 1jj ii SS 0jj SS 0jj YY 0jj = jj SS 1jj ii YY 0jj 54

55 Table 1. Average Professorial Salaries at UBC in 2010 Gender Rank Numbers % of All % of Average Female/ Gender women Salary Male Ratio Gap Men All Women All Men Full Women Full Men Associate Women Associate Men Assistant Women Assistant If the proportion of women across professorial ranks was identical to men, the overall counterfactual average female salary would be: 51.8/ / / = , and the overall ratio would be /134955(*100)=95% The salary gap explained by rank is = More that 53% of the gap is accounted for by the gender differences in the proportion of faculty members across rank.

56 Table 2. Oaxaca-Blinder Decomposition of Average Professorial Salaries at UBC in 2010 Variables: Model 1 % of gap Model 2 % of gap Raw Gender Salary Differentials *** *** Accounted for by differences in characteristics Professorial Rank *** 53.28% *** 46.38% CRC, DUP * 3.81% Years in Rank ** 8.23% Departmental Dummies ** 21.58% Total Explained *** 53.28% *** 80.01% Total Unexplained *** 46.72% *** 19.99% Note: Using female coefficients. *** p<0.01, ** p<0.05, * p<0.1 See UBC (2011) for alternative specifications. The more complete specification accounts for 80% of the gap, 46% of which from vertical segregation and 22% from horizontal segregation. This leaves an unexplained gender gap of 2.2% of average professorial salary 56

57 If the shares of women in percentiles grouping* were the same as men s, the gap in annual earnings would be 20 point lower Female-Male Annual Earnings Ratios - Canada = 58% = 45% Simulated Ratio Actual Ratio Source: Fortin, Drolet and Bonikowska (2016), LWF , years old, Annual earnings from all jobs *percentiles grouping: bottom 90%, next 9%, next 0.9%, top 0.1% 57

58 If the shares of women in percentiles grouping* were the same as men s, the gap would be 50% lower Female-Male Annual Earnings Ratios - Sweden Female-Male Annual Earnings Ratios United Kingdom = 45% 0.14 = 53% = 48% 0.18 = 53% Actual ratio Simulated ratio Actual ratio Simulated ratio Source: Fortin, Bell, and Boehm (2017), workers ages 25 to 64, LISA data for Sweden, ASHE data for the UK. 58

59 If the shares of women in percentiles grouping* were the same as men s, the gap would be 6-9 points lower Counterfactual Hourly Wage Gender Ratio - Canada = 41% 0.07 = 45% Actual Ratio Simulated Ratio Source: Fortin, Drolet and Bonikowska (2016), LFS , years old, Hourly wages on the main job *percentiles grouping: bottom 90%, next 9%, next 0.9%, top 0.1% 59

60 Against traditional factors in O-B decomposition, centile groupings remain dominant and growing over time 120% 100% 80% 60% 40% 20% 0% -20% 36% 44% 36% 48% 17% 37% Sweden - LISA UK- LFS Canada - LFS Total Unexplained Selected Centiles Occupation Industry Part-time, Tenure Education Demographics Region Source: Fortin, Bell, and Boehm (2017), workers ages 25 to 64, LISA data for Sweden, LFS data for Canada and the UK. Entries are male/female differences in the explanatory variables multiplied by the corresponding female coefficients as a share of the gender pay gap. 61

61 In provinces with Pay Equity, the explanatory power of occupations, like education, has gone negative 130% 110% 90% 70% 50% 30% 10% -10% -30% 30% 29% 34% 37% 28% 34% 4% 11% -6% -2% -5% % Quebec Ontario ROC Source: Fortin(2017), workers ages 25 to 64, LFS data for Canada. Entries are male/female differences in the explanatory variables multiplied by the corresponding female coefficients as a share of the gender pay gap. Total Unexplained Selected Centiles Occupation Industry Part-time, Tenure Education Demographics Province 62

62 Trends in Gender Ratios: Quebec vs. Other Provinces Female-Male Hourly Wage Ratio (25 to 64 Year Old) Female-Male Hourly Wage Ratio (25 to 64 Year Old) Hourly Wage Ratio Hourly Wage Ratio Year Quebec Ontario 95% CI 95% CI Year Quebec Synthetic Provinces 95% CI 95% CI Note: Synthetic provinces weights: PEI, NS, NB, MA, AL 64

63 Impact of Under-Representation in Top Jobs Overall in Canada (except for Quebec), Sweden, and the UK, the underrepresentation of women in top jobs accounts for a predominant and growing share of the gender pay gap. Even against industry and tenure, it is the most significantly explanatory factor In Quebec and Ontario, provinces that implemented pay equity, the unexplained portion of the gender gap has been growing over time as the explanatory power of occupations has gone negative. Women in Quebec in senior management (a male-dominated job among others), are not compensated in a commensurate way 65

64 Bottom-Line Looking back at the transformation of women s work in Canada over the 20 th century, Fortin and Huberman (2002) had argued that the decline in vertical segregation had contributed more to the improvement of women s labour market outcomes than changes in horizontal segregation. With increasing earnings inequality in top incomes, further improvements in vertical segregation, relatively more women in top jobs will be likely be even more important for further decline in the gender pay gap in the 21 st century But unlike in the 20 th century, further educational attainment alone will not yield those changes! 66

65 Gender Pay Differentials and Pay Policies Gender pay differentials within occupation within firm Equal Pay for Equal Work In Canada and U.S. complained based system Gender pay differentials within establishment Equal Pay Label Available in Switzerland, Luxembourg, and Germany starting in 2010 Excel regression-aided program (e.g. Logib-CH and Logib-D) help companies find whether their firm-level adjusted female penalty is less than the desired 5% with p-value 0.05, accounting for age, education, training, and other job characteristics to gain label which gives them a competitive edge and profile to attract the best employees. 67

66 Public Policy and Gender Pay Differentials Gender pay differentials across comparable female-dominated occupations and male-dominated, resulting from horizontal segregation, are the focus of Pay Equity policies, implemented in the private sector of Canada s two most populous provinces: Ontario (1996) and Quebec (2001) Gender pay differentials arising from the potential obstacles that women face climbing (or not) the job ladder (vertical segregation) Employment Equity, enacted in the Federal jurisdiction in principle could address disparities across the job ladder. 68

67 Higher Representation of Women in Tops Jobs! What to Do? In recent years, many countries have pushed for more general gender equality in decisionmaking with bolder moves. Both in the political sphere and on corporate boards. Many have implemented female quotas in legislative assemblies and on the board of directors of firms on public stock exchanges. 69

68 Women s Quotas for Corporate Boards? Women on Boards and Employment Share Women on Boards (%) ID ID MY MY MY NONO NO SE SE FI FI SE SE FI FI PH DKFR FR NL DK NZ AU US DE PH US DENL DK NL DK PL PL CA NZ NZ US GB CA THTH CH ATAT GB DE HU CZCH IE PH CH THES GB AT BE AUBE GR IE GR IT PL ES DE IE GR SG SG BE CZ CH BE AT ESGB HK FR HK FR NL HU GR HU RU RU ES PT RU IT IT IT PT PT JP JP JP Female Employment Share (%) No Regulation or Quota Pre-Quota Post-Quota Pre-Regulation Post-Regulation Fitted values Years: 2006, 2009, 2011, 2013, 2014 Source: Dizik, 2015 Source: Fortin, Bell, and Boehm (2017) 70

69 Higher Representation of Women in Tops Jobs in Canada? Short of calling for gender quotas, the Canadian Securities Administrators of seven provinces and territories (CSA, 2015) implemented comply-orexplain female representation rules on January 1, 2015 (Shecter, 2014; McFarland, 2015). These rules require companies listed on their stock exchanges to disclose how many women they have on their boards and in their executive ranks. But many companies have shown bare `technical compliance with the reporting rules introduced last year and it is "simply not good enough," says Howard Wetston, the Ontario Securities Commission chair. 71

70 Higher Representation of Women in Tops Jobs! Do quotas help? Yes, for Women on Boards but no evidence of trickle down in country-fixed effects models Source: Fortin, Bell, and Boehm (2017) Dependent Variable: Table 4: Impact of Board Quotas and Disclosures Rules Women on Boards Women in Senior Management Mean Explanatory Variables Quotas (1.172) (1.245) (1.045) (0.943) Disclosure Rules (0.952) (1.092) (0.687) (0.685) Relative Female Employment Rate (20.38) (21.23) (37.81) (37.79) Log GDP per capita (PPP) (4.264) (6.690) (3.614) (3.199) R-square No. of observations OECD only No Yes No Yes No.of countries Note: Dependent variables are the share of women on corporate boards from BoardEx data (European PWN, 2008) from 2006 to 2009, from GMI data (Gladman and Lamb, 2013) from 2009 to 2014, and the share of women in senior management from ILO (2014). The data on the relative female employment rate, computed as the ratio of female employment rate to the total employment rate, is from the World Bank. Estimates from country fixed-effects models with robust standard errors clustered at the country level. * p < 0.01, ** p < 0.05, * p <

71 Higher Representation of Women in Tops Jobs! Any Trickle-Down Effects? Mixed results on relative female wages Flabbi et al. (2013) studying Italian firms also find mixed results linked to substantial sorting of female managers across establishments. small and less productive establishments that invest less, pay their employees lower wages, but are more female-friendly are more likely to be led by women. Bertrand, Black, Jensen, and Lleras-Muney (2014) show that the Norwegian quotas increased representation of women among top 5 highest earners, but had no effect at other points in the distribution or on the gender pay gap. 73

72 Women Fail to Move from Bottom 90% to Next 9% in Early Career (age 30) in Canadian LWF Share of Men and Women Moving from Bottom 90% to Next 9% - Recent Synthetic Cohorts _84 84_85 85_86 86_87 87_88 88_89 89_90 90_91 91_92 92_93 93_94 94_95 95_96 96_97 97_98 98_99 99_00 00_01 01_02 02_03 03_04 04_05 05_06 06_07 07_08 08_09 09_10 Men 1960 Men 1970 Men 1980 Men All Women 1960 Women 1970 Source: Fortin, Drolet and Bonikowska (2016) Computation, LWF , years old, Annual earnings from all jobs

73 Mommy Track or Opting Out? Recent studies of the impact of children on mothers earnings have found large and persistent negative effects on labour market outcomes o USA: Wilde, Batchelder, and Ellwood (2010) o Germany: Fitzenberger, Sommerfeld, and Steffes (2013), Adda, Dustmann, and Stevens (2015) o Portugal: Card, Cardoso and Kline (2016) o Spain: Fernández-Kranz, Lacuesta, and Rodríguez-Planas (2013) o Italy: Del Bono and Vuri (2011) o Austria: Pamminger, Weber, and Winter-Ebmer (2014) o Sweden: Angelov, Johansson, and Lindahl (2016), Karimi, Hotz, and Johansson (2016) o Denmark: Kleven, Landais, and Søgaard (2016) Following childbirth, mothers often move to part-time work or a more flexible schedule, to family-friendly, less profitable, and lower paying firms, and are less likely to be promoted

74 Family Friendly Public Policy and Gender Pay Differentials The remedial family-friendly public policies and firm practices fall into two camps: 1) those that allow women to take time off from work (maternity and child care leave, flexible work hours) to attend to their family responsibilities can lead to a possible reduction in labour attachment 2) those that assist women in finding substitutes for their child care responsibilities (on-site child care, universal childcare, etc.) most likely to foster gender equality in the workplace keeping in mind that gender biases in the workplace also need to be addressed 76

75 Restez à l écoute! Merci! 77