FPSLREB Decisions

Heard at Ottawa, Ontario,
April 20 to 24, 2015.

I. Complaint before the Public Service Labour Relations and Employment Board

1 On April 23, 2009, the Canadian Human Rights Commission (CHRC) referred this complaint to the Public Service Labour Relations Board pursuant to subsection 396(1) of the Budget Implementation Act, 2009 ("the BIA"), S.C. 2009, c. 2, which came into force on March 12, 2009. Section 396 provides as follows:

396.(1) The following complaints with respect to employees that are before the Canadian Human Rights Commission on the day on which this Act receives royal assent, or that are filed with that Commission during the period beginning on that day and ending on the day on which section 399 comes into force, shall, despite section 44 of the Canadian Human Rights Act, without delay, be referred by the Commission to the Board:

(a) complaints based on section 7 or 10 of the Canadian Human Rights Act, if the complaint is in respect of the employer establishing or maintaining differences in wages between male and female employees; and

(b) complaints based on section 11 of the Canadian Human Rights Act.

(2) The complaints referred to in subsection (1) shall be dealt with by the Board as required by this section.

(3) The Board has, in relation to a complaint referred to it, in addition to the powers conferred on it under the Public Service Labour Relations Act, the power to interpret and apply sections 7, 10 and 11 of the Canadian Human Rights Act, and the Equal Wages Guidelines, 1986, in respect of employees, even after the coming into force of section 399.

(4) The Board shall review the complaint in a summary way and shall refer it to the employer that is the subject of the complaint, or to the employer that is the subject of the complaint and the bargaining agent of the employees who filed the complaint, as the Board considers appropriate, unless it appears to the Board that the complaint is trivial, frivolous or vexatious or was made in bad faith.

(5) If the Board refers a complaint under subsection (4) to an employer, or to an employer and a bargaining agent, it may assist them in resolving any matters relating to the complaint by any means that it considers appropriate.

(6) If the employer, or the employer and the bargaining agent, as the case may be, do not resolve the matters relating to the complaint within 180 days after the complaint is referred to them, or any longer period or periods that may be authorized by the Board, the Board shall schedule a hearing.

(7) The Board shall determine its own procedure but shall give full opportunity to the employer, or the employer and the bargaining agent, as the case may be, to present evidence and make submissions to it.

(8) The Board shall make a decision in writing in respect of the complaint and send a copy of its decision with the reasons for it to the employer, or the employer and the bargaining agent, as the case may be.

(9) The Board has, in relation to complaints referred to in this section, the power to make any order that a member or panel may make under section 53 of the Canadian Human Rights Act, except that no monetary remedy may be granted by the Board in respect of the complaint other than a lump sum payment, and the payment may be only in respect of a period that ends on or before the day on which section 394 comes into force.

2 On March 6, 2009, Karen Hall and Association of Canadian Financial Officers ("the complainant") had filed a complaint with the Canadian Human Rights Commission based on sections 7, 10 and 11 of the Canadian Human Rights Act ("the CHRA"), R.S.C., 1985, c. H-6. The complainant Ms. Hall alleges that she has been discriminated against on the basis of gender in wages and in her ability to access employment benefits.

3 Section 7 of the CHRA provides as follows:

7. It is a discriminatory practice, directly or indirectly,

(a) to refuse to employ or continue to employ any individual, or

(b) in the course of employment, to differentiate adversely in relation to an employee,

on a prohibited ground of discrimination.

4 Section 10 of the CHRA provides as follows:

10. It is a discriminatory practice for an employer, employee organization or employer organization

(a) to establish or pursue a policy or practice, or

(b) to enter into an agreement affecting recruitment, referral, hiring, promotion, training, apprenticeship, transfer or any other matter relating to employment or prospective employment,

that deprives or tends to deprive an individual or class of individuals of any employment opportunities on a prohibited ground of discrimination.

5 Section 11 of the CHRA provides as follows:

11. (1) It is a discriminatory practice for an employer to establish or maintain differences in wages between male and female employees employed in the same establishment who are performing work of equal value.

(2) In assessing the value of work performed by employees employed in the same establishment, the criterion to be applied is the composite of the skill, effort and responsibility required in the performance of the work and the conditions under which the work is performed.

(3) Separate establishments established or maintained by an employer solely or principally for the purpose of establishing or maintaining differences in wages between male and female employees shall be deemed for the purposes of this section to be the same establishment.

(4) Notwithstanding subsection (1), it is not a discriminatory practice to pay to male and female employees different wages if the difference is based on a factor prescribed by guidelines, issued by the Canadian Human Rights Commission pursuant to subsection 27(2), to be a reasonable factor that justifies the difference.

(5) For greater certainty, sex does not constitute a reasonable factor justifying a difference in wages.

(6) An employer shall not reduce wages in order to eliminate a discriminatory practice described in this section.

(7) For the purposes of this section, "wages" means any form of remuneration payable for work performed by an individual and includes

(a) salaries, commissions, vacation pay, dismissal wages and bonuses;

(b) reasonable value for board, rent, housing and lodging;

(c) payments in kind;

(d) employer contributions to pension funds or plans, long-term disability plans and all forms of health insurance plans; and

(e) any other advantage received directly or indirectly from the individual's employer.

6 On June 10, 2009, the Public Service Labour Relations Board reviewed the complaint pursuant to subsection 396(4) of the BIA and referred it to the Treasury Board, the employer that is the subject of the complaint, and the bargaining agent (the complainant Association of Canadian Financial Officers). Since the parties did not resolve the matter within 180 days of the referral, the PSLRB scheduled a hearing into the matter, pursuant to subsection 396(6) of the BIA.

7 On November 1, 2014, the Public Service Labour Relations and Employment Board Act (S.C. 2013, c. 40, s. 365) was proclaimed into force (SI/2014-84), creating the Public Service Labour Relations and Employment Board ("the new Board") to replace the former Public Service Labour Relations Board ("the former Board") as well as the former Public Service Staffing Tribunal. On the same day, the consequential and transitional amendments contained in sections 366 to 466 of the Economic Action Plan 2013 Act, No. 2 (S.C. 2013, c. 40) also came into force (SI/2014-84.) Pursuant to section 441 of the Economic Action Plan 2013 Act, No. 2, the new Board replaced the former Board for the purpose of section 396 of the BIA.

II. Summary of the evidence

8 The parties submitted an extensive agreed statement of facts, which was marked as Exhibit 2, and a joint book of documents (Exhibit 3). In the agreed statement of facts, they set out the background to the complaint, a summary of the history of pay equity in the Canadian public service, a history of the Financial Officer (FI) group, the qualifications and the nature of the group's work, and a snapshot of its current composition.

9 The parties have also agreed that the appropriate test for determining whether there is a gender-based wage gap between a female-predominant group and a male-predominant group is that set out in both the majority and the dissenting opinions in Public Service Alliance of Canada v. Canada Post Corporation, 2010 FCA 56 ("the Canada Post test"). The matter was appealed to the Supreme Court of Canada, which agreed with the dissenting reasons of the lower court (Public Service Alliance of Canada v. Canada Post Corporation, 2011 SCC 57). According to the Canada Post test, the following elements must be proven, on a balance of probabilities, to establish a prima facie case of discrimination under section 11 of the CHRA:

1. The complainant occupational group is predominantly of one sex, and the comparator occupational group is predominantly of the other sex.
2. The female-dominated occupational group and the male-dominated occupational group being compared are composed of employees employed in the same establishment.
3. The value of the work being compared between the two occupational groups has been assessed reliably on the basis of the composite of the skill, effort and responsibility required in the performance of the work and the conditions under which the work is performed. The resulting assessment establishes that the work being compared is of equal value.
4. A comparison made of the wages being paid to the employees of the two occupational groups for work of equal value demonstrates that there is a difference in wages between the two, the predominantly female occupational group being paid a lesser wage that the predominantly male occupational group. This difference is commonly called a "wage gap."

10 Counsel for the complainant provided the evidence of four witnesses over three full days of hearing, including Ms. Hall, Milton Isaacs, Paul Durber and Alan Sunter. The evidence of Mr. Durber, who conducted the job evaluation study that formed the basis of Mr. Sunter's conclusion that a wage gap existed between the combined female-predominant FI-1 and FI-2 groups and the comparator male group as outlined in the complaint, took two of those days.

11 Mr. Isaacs, the current ACFO president, testified concerning the many changes he has seen in the nature of accounting processes in the federal government during his FI career, which commenced in 1980. The accounting processes for which he was responsible had evolved from cash accounting to the much more complicated and involved accrual accounting. This changed the depth and breadth of FI competencies as did the introduction of an enterprise management information system, which forced the FIs to better understand the programs in which they were embedded. This changing work is not captured in the classification standard as it did not exist when the standard was written. The current standard does not reflect the complexity of the work currently being done.

12 Mr. Isaacs referred in his evidence to the Treasury Board's effort, beginning in 1995, to develop a new classification system (the Universal Classification Standard (UCS)), a project that was later abandoned. Mr. Isaacs noted that during this process, the employer was surprised to learn of the degree and complexity of the FIs' work. He pointed out that in an effort to become revenue neutral, the bargaining agent made many compromises that ended up diluting the FI work description. According to Mr. Isaacs, when the UCS was finished, the UCS committee could see that the FIs were analogous to "fighter pilots" but that on paper, they were "piper pilots." This was a movement from the original standard but far from the fighter-pilot level. The old FI standard was equivalent to "driving a car."

13 Mr. Isaacs has been ACFO President since 2005. In 2007, the ACFO put together a committee to examine the discrepancies in the FI's work descriptions and classification. Classification and pay level changes are linked through the collective bargaining process. The employer was experiencing recruitment issues, particularly at the FI-3 and FI-4 levels. As a result, a transitional allowance was implemented for these levels. The lower levels were treated differently as there was no recruitment problem there.

14 Mr. Isaacs concluded his testimony by stating that FI employees' work has become more complex, has more influence and requires more professional judgement at all levels, which are not "traditionally considered to be female characteristics". The ACFO wants to have the increased complexity recognized in the salaries of all levels. To that end, the ACFO is working with the employer to update the classification system and to find new benchmarks.

15 The complainant Ms. Hall testified that as an FI-1 and FI-2, she was involved in financial forecasting, financial management, budgeting, invoicing contract partners and providing guidance to her clients. In her current FI-3 position, she has these duties plus supervisory responsibilities. The differences between FI-1 and FI-2 were the client profile and budgetary responsibilities. There was no change in the work when the employer changed the qualification standard to include a recognized accounting designation.

16 Ms. Hall led the team implementing the Royal Canadian Mounted Police pay equity settlement in 2007. She presented information to public service employees on the implementation plan. Four different payments were made, with calculations required for each one. During this time, she came to realize that there was an element of female predominance within the FI population. It appeared to her that the FIs were paid less than other professional groups that were male dominated. With the increasing responsibilities, the gap broadened. With each increase in the duties, the FIs got further behind because of the compounding effect of the payments.

17 As an ACFO director, Ms. Hall was assigned to the bargaining agent's Pay Equity Committee. She participated in focus groups with the FIs and received education about pay equity. After the Program Administration group members received their pay equity settlement, she decided to file her complaint in the hope that the FI classification system would change. Attempts to change the system through collective bargaining had been unsuccessful, so she filed the complaint on March 8, 2009, in hopes that a more formal process would result in a classification system that would value her work fairly.

18 Mr. Durber is well known in the circles of pay equity disputes in the public service, having been an expert witness in the hearings of many complaints before the CHRC and this Board. The parties agreed that he be qualified as an expert witness in pay equity, including the estimation of gender-based wage differentials within the context of the federal public service (Exhibit 5).

19 During his testimony, Mr. Durber explained the results from his original study and his reply study in response to concerns raised by the employer's experts (Exhibit 4, tabs C (1) and C (2)). He also submitted Exhibit 11, which is an updated data report in which he had corrected certain anomalies.

20 Mr. Durber testified that to be successful under section 11 of the CHRA in a pay equity complaint, the work at issue must be done by both men and women. He explained the Equal Wages Guidelines, 1986 (SOR/86-1082); the regulations applicable to pay equity under the CHRA require that a female-dominant group must be identified to trigger a complaint. Then a male-dominant comparator group must be identified. In order to assess predominance, the composition of the groups by gender is examined at the point one year before the complaint was made. In this case, Mr. Durber combined the FI-1 and FI-2 levels to constitute a female-predominant group. The comparator groups he examined were those set out in Ms. Hall's complaint (Commerce Officer (CO), Computer Systems (CS), Engineering (EN-ENG), Meteorology (MT)). In the course of his evaluations, he determined that the CO group should have been excluded from the comparator group as it was gender-neutral.

21 In his explanation of his process, Mr. Durber explained his choice of the "New Zealand Equitable Job Evaluation Plan" (Exhibit C (1), annex A; "the New Zealand plan") as an appropriate tool for his evaluation of the jobs in question. He also explained his evaluation process, which was a paper exercise that he conducted alone in which he reviewed a sample of 110 work descriptions. The New Zealand plan, although abandoned by the New Zealand government in its beta-testing phase, was appropriate in Mr. Durber's opinion for the purposes of the pay equity study in this case. He amended the plan for use in this case and did not use a questionnaire or interview incumbents in the positions reviewed as set out in the New Zealand plan's "Equitable Job Evaluation User's Guide" (Exhibit 15); nor did he apply the unit sizing factor to his evaluations of factor 6 (Resources) and factor 7 (Organizational Outcomes), even though this would have increased the overall scores allocated by up to a maximum of 1.8 times the score for these factors, depending on the size of the organization. Despite the exigencies of the user's guide to the New Zealand plan, which Mr. Durber chose not to follow, the quality of the job information he obtained, according to him, met the requirements of section 11 of the CHRA.

22 On cross-examination, Mr. Durber admitted that he noted none of these changes to the plan and that he did not explain his rationale for the changes in his report or reply report.

23 Mr. Durber testified that his conclusions in the original report, as set out in his reply report, were reliable once clerical errors were corrected and that they clearly indicated that a gender-based wage gap existed. The weightings that he assigned to the various factors were reasonable. The corrections between the original report and the reply report reduced the gender-based wage gap, but it was still apparent. In his mind, he had successfully valued the work of the complainant and comparator groups by examining a proper sample of job information using a gender-neutral plan. Based on his results, in 25 observations, the female-predominant FI-1 and FI-2 groups were paid less than their comparators. He had sufficient observations in which the male comparators were paid at a higher rate than members of the complainant group with the same job evaluation score that he was able to draw a wage line.

24 The respondent had Mr. Durber's report evaluated by its own experts, who identified what he referred to as clerical errors in the data set upon which he based his conclusions, which were fixed and submitted in his reply report. Further revisions to his data were made subsequent to the reply report and were submitted as Exhibit 11. As a result, the mean varied slightly between the two reports. The range of scores for the FI-1 positions was 375 to 500, while the mean for the male comparators was 400 to 500. The same pattern was identified with the FI-2 positions. Therefore, Mr. Durber concluded that on average, the FIs and the comparators had work of equal value sufficient to discharge part 3 of the Canada Post test.

25 According to Mr. Durber, sections 7 and 10 of the CHRA deal with systemic discrimination as a result of policies and procedures that may affect pay based on, among other things, pay structure. The FI standard, used to rank FI jobs in the public service, is outdated and does not comply with the four factors identified in the CHRA. The standard denies members of the female-predominant groups (FI-1 and FI-2) access to the higher levels of the pay grid (FI-3 and FI-4).

26 In the course of a very extensive cross-examination, Mr. Durber admitted that his primary concern was the value of the FI group positions. He admitted that he used jobs whose ratings were outside the range in his calculations. In addition, he admitted that he had included evaluations of the gender-neutral CO group as comparators. In at least one case in which he disagreed with the employer's classification of a position, he changed it to what, in his opinion, it should have been.

27 Exhibits 19, 20 and 21 which identified many inconsistencies and errors in Mr. Durber's report were put to him on cross-examination. At one point, the hearing adjourned early for the day, and Mr. Durber was provided with copies of these exhibits to review and was asked to prepare his responses to the deficiencies in his data that they identified. He was then to advise the panel of the Board the following day as to his findings.

28 Some of these inconsistencies, omissions or errors noted by counsel for the respondent were attributed to clerical errors by Mr. Durber. Included were unpopulated graphs and missing regression lines. Others were mistakes such as those set out in Exhibit 19, which lists five different groups of identical generic job descriptions that had been given different ratings. Mr. Durber was keen to correct them and volunteered to submit more revised data. In all, he admitted to making errors in the total points awarded to 18 of 110 evaluations. An additional 65 errors or discrepancies were noted in Exhibits 20 and 21, of which Mr. Durber accepted 20 as mistakes and stated that the remaining 45 were accurate statements but that the conclusions drawn by counsel for the respondent were misleading as he was convinced that his understanding of the work was solid and the valuations correct. He repeatedly stated in cross-examination that the errors should not have occurred regardless of whether the response was to agree in full with the employer's counsel or to agree but state that her conclusions were misleading as the correction to the data would not have been substantial.

29 On the second day of Mr. Durber's testimony, before the hearing adjourned early and while the review of Exhibits 19 and 20 was ongoing, he was asked if he had "sore thumbed" his data. This is the process by which scores are examined to see if any stick out like "sore thumbs", which would call for further investigation. Mr. Durber's response to the question was that he had not because he did not view the process as relevant.

30 The next day, in light of the number of errors being examined, Mr. Durber insisted that he did in fact conduct a "sore thumb" exercise on the data and that he had merely missed the clerical errors. He became emotional, highly defensive and non-responsive in his testimony concerning the data discrepancies. At one point, he admitted that approximately 40% of the total point scores in his report were wrong. He acknowledged that numerous typographical and reference issues were identified, which should have been caught before his report or any of the more recent reiterations of his data had been released. However, he testified that these errors would have minimal impact on his overall results. In total, 41 job evaluation scores could be wrong based on inconsistencies and on his amendment to the tool used in those evaluations. At the conclusion of his testimony, Mr. Durber offered to correct the data yet again and to provide an updated version to the Board. Counsel for the respondent strenuously objected to his offer and the Board refused it.

31 Mr. Durber testified as to the effect of a violation of section 11 of the CHRA on a complaint under sections 7 and 10. Discrimination under these sections results from policies and procedures that may affect pay. Such a system is used for job ranking, upon which salaries are negotiated through collective bargaining. In this case, the current job structure does not measure the financial management work and does not comply with the four factors set out in the Equal Wages Guidelines, 1986. Mr. Durber testified that the current FI standard overlooks many features of female work, although he provided no precision as to these deficiencies. In his opinion, the complaint is founded as it denies female FIs access to higher levels of pay based on his evaluation of the male comparators. The denial to access to equal pay has a systemic cause, which lies in the FI standard itself and is evident from the results of his study.

32 Mr. Sunter testified as an expert in statistics, including in the estimation of gender-based differentials within the context of the federal public service. He conducted the statistical analysis of Mr. Durber's data upon which the wage gap was calculated. The purpose of his analysis was to confirm Mr. Durber's conclusion with more precision. Based on this data, he conducted his regression analysis and concluded that there was a wage gap of approximately $4557 at the FI-1 level and $2878 at the FI-2 level. Mr. Durber provided Mr. Sunter all the data that had been relied on; Mr. Sunter assumed it was accurate. He stated that in the course of his analysis, he identified certain errors in Mr. Durber's data, which slightly reduced the wage gaps identified in Mr. Durber's conclusions. Mr. Durber's data was the sole basis of his statistical analysis, even though he noted errors in its content. On cross-examination, Mr. Sunter admitted that not all the data had been provided to him as certain evaluations were missing and that at least one job evaluation was included twice. He was also unaware that Mr. Durber had not applied the augmenting formula for factors 6 and 7 of the New Zealand plan as required in its handbook.

33 On cross-examination, he was asked to review Exhibits 19, 20 and 21 and to comment on the impact these errors would have on his conclusions. It was put to Mr. Durber that if the errors are removed, there is a question of the reliability of the conclusions but that if the errors remain, the question of bias is raised. Mr. Sunter could not testify as to the impact these errors would have on his report because he did not know their magnitude. If the errors were within the range of 7% to 10%, Mr. Sunter would not be concerned. The larger errors were a cause for concern, and the data should be corrected and the analysis redone.

34 At the close of the evidence, counsel for the respondent made a motion for non-suit while clearly indicating that if it were unsuccessful, it intended to call evidence. The basis of this motion was that the complainant has not presented a prima facie case sufficient to discharge part 3 of the Canada Post test and to establish that the value of the work compared between the two occupational groups has been assessed reliably. Arguments were then heard on the motion.

III. Summary of the arguments

A. For the respondent

35 The test for a non-suit is whether the complainant has met the onus of establishing a prima facie case. In this matter, the complainant has not established that on a prima facie basis the job values of the complainant and comparator groups have been determined reliably. Specifically, Mr. Durber's report does not meet the requirement of part 3 of the Canada Post test. The respondent has clearly established that Mr. Durber's data is unreliable and that any conclusions drawn from it are unreliable.

36 The PSLRB considered its authority to consider an application for non-suit in Bremsak v. Professional Institute of the Public Service of Canada et al., 2013 PSLRB 22, aff'd on other grounds 2014 FCA 11. The panel in that case concluded as follows at paragraph 91:

91 … The power to hear a non-suit application is discretionary and incidental to the powers of a panel of the Board to determine its practices and procedures. It is not a procedure frequently or readily granted, but it can be opportune to allow a party to argue that there is no case to meet… .

37 This would be an appropriate case for the Board to exercise this discretion, given the cost to all parties of proceeding in light of the complainant's clear failure to prove on a prima facie basis that the jobs were evaluated reliably.

38 The Federal Court, in Filgueira v. Garfield Container Transport Inc., 2006 FC 785 at para. 24, considered as follows what constituted a prima facie case, citing Ontario Human Rights Commission v. Simpsons-Sears, [1985] 2 S.C.R. 536, in the context of a human rights complaint:

…

The complainant in proceedings before human rights tribunals must show a prima facie case of discrimination. A prima facie case in this context is one which covers the allegations made and which, if they are believed, is complete and sufficient to justify a verdict in the complainant's favour in the absence of an answer from the respondent-employer.

…

39 In determining whether a prima facie case has been established, the evidence to be considered is not merely that elicited on direct examination. Cross-examination must also be considered. The evidence of Mr. Durber as a whole was not reliable and provided no credible sense of the value of the work examined. He was the job-evaluation expert who was relied upon to establish the job values Mr. Sunter used in his statistical analysis. Considering the totality of Mr. Durber's evidence and the inconsistencies and ongoing corrections to the job scores, there is no sense of when and if his scores will be reliable.

40 Before he was cross-examined, Mr. Durber provided three iterations of his scores. He had three opportunities to get them right. Between his initial report and his reply report, he corrected 18 total scores. In Exhibit 11, he made an additional nine corrections. After Exhibit 11, another 66 inconsistencies in his scores were identified, all of which might have been actual errors that impacted the total scores. He admitted that 41 of these errors were in point scores affecting job totals.

41 Of the errors identified in Exhibits 21, 31 of the 41 errors related to male comparator jobs. Twenty-five of them suggested that the male comparators were underrated, making these jobs overpaid vis-à-vis female work of the same value. Of the 10 female jobs included in the 41 jobs identified, 6 errors could have resulted in overvaluing the female job, making them underpaid vis-à-vis their make comparators.

42 When faced with Exhibit 19, outlining the five groups of generic work descriptions evaluated differently, Mr. Durber tried to make corrections to his data while testifying. At the end of his testimony, he offered to submit another iteration of his data. A reasonable person cannot conclude given the inconsistencies and errors in Mr. Durber's data and the number of iterations of the data he issued that the job values have been reliably evaluated. Consequently, the complainant has not met the third part of the Canada Post test. Given that this is a fatal flaw to the establishment of gender discrimination with respect to wages, the complaint should be dismissed.

B. For the complainant

43 The complainant accepted that this Board has the authority to determine if the complainant discharged its prima facie burden of proof. This is a regular civil burden and not the criminal burden of proof beyond a reasonable doubt. The third part of the Canada Post test requires that the value of the work be assessed reliably for a conclusion to be made that the complainant group and the comparator group perform work of equal value. Perfection is not required. If the Board is convinced that 50% plus 1 of the assessments were conducted reliably, the burden of proof has been met.

44 Given that 110 jobs were evaluated against 12 factors each, it is highly probable that errors occurred. The question is whether these errors pushed the pay line in one direction or the other. Mr. Sunter testified that he expected a range in scores of 7% to 10% for any one classification. From a statistical perspective, these are acceptable variances. When questioned concerning Exhibit 21, Mr. Sunter stated that some of the errors caused him no concern from a statistical perspective while the size of others did cause him concern and would require further examination.

45 In determining whether there is sufficient evidence to satisfy part 3 of the Canada Post test, the combination of evidence from Mr. Durber and Mr. Sunter must be considered. The complainant relied on the wage gap established by Mr. Sunter. Part of the consideration of whether the evidence was sufficient must be drawn from Mr. Durber's testimony.

IV. Reasons

46 Both parties agreed that if I conclude that Mr. Durber's evaluation of the complainant and comparator jobs does not meet the test of reliability as set out in part 3 of the Canada Post test, it is fatal to the success of this complaint. Mr. Sunter's calculations and conclusions depended on the reliability of the data provided to him for analysis.

47 Based on the evidence presented, I have serious doubts about the reliability of Mr. Durber's analysis. His decision not to follow the handbook to the New Zealand plan, which outlines the appropriate use of the evaluation tool and clearly recommends that the evaluation be conducted by a group, that questionnaires be designed and used, and that interviews be undertaken when necessary, and his choice to ignore the augmentation rates for factors 6 and 7, when applicable, are of particular concern. Of further concern is that he made these changes to the plan and that he did not note them in either of his reports.

48 Furthermore, between the date of his initial report and the date of his testimony, Mr. Durber released three iterations of his data, all of which were different from that provided to Mr. Sunter. In addition, he conceded that Exhibits 19, 20 and 21 reflected further errors or omissions in his data, which should not have occurred and for which he had little or no explanation.

49 Counsel for the complainant suggested that I assess Mr. Durber's evidence in order to conclude whether the job evaluation study he conducted was reliable. I found that Mr. Durber spoke with great authority on direct examination but when faced with the sheer volume of discrepancies in his data, he became emotional, aggressive, uncooperative and unresponsive. He did admit to a significant number of errors, which he dismissed as either clerical errors or as things he just missed. He also admitted to other errors, which had a direct impact on the total scores of some jobs. He had no real explanation for why he would rate identical generic work descriptions differently other than to say, when pressed, an error had been made. Nor did he provide any reasons for unilaterally reclassifying at least one position based on his determination that it was improperly classified. While initially claiming that he did not see the need for a "sore thumb" of his data, he later insisted that he had conducted such an exercise.

50 Despite Mr. Durber's limited explanations, I cannot ignore the multitude of errors and inconsistencies in his data. The significance of this number, in combination with his choice to use only parts of a tool he chose to evaluate the work in question, causes me concern. I have considered Mr. Sunter's testimony on the impact these errors might have had on his regression analysis and his statements that some had a significant impact and would require a review of his analysis.

51 If I accept Mr. Durber's testimony without hesitation, including the accuracy of his data, I may be able to conclude that the complainant has established a prima facie case of gender bias. However, given the sheer number of errors and anomalies identified in his data, it appears that his approach to collecting, evaluating, recording and validating his data was less than reliable. His responses to any challenge to his data also go directly to the reliability of his data but are not the sole basis for my conclusion that the complainant has not established that the requirements of part 3 of the Canada Post test have been met.

52 While a witness's demeanour while under examination is a consideration when determining credibility, it cannot be the sole basis upon which to assess credibility. As stated as follows at paragraph 233 of Robitaille v. Deputy Head (Department of Transport), 2010 PSLRB 70 (referring to Faryna v. Chorny, [1952] 2 D.L.R. 354):

233. To assess a witness' credibility, the person hearing the evidence must not solely rely on the impression left by the witness but must base the assessment on an examination of how the testimony given fits into the evidence as a whole, taking into account other testimony, the facts established, a reasonable probability of events and the assessor's experience in human relations… .

53 My conclusions are not based solely on the impression left by Mr. Durber. It is a combination of his total testimony over 21⁄2 days of hearing, the exhibits, his changing testimony and his emotional reaction to an in-depth examination of his data that leads me to the conclusion that the outcome of his study is not reliable and that there is reason to doubt the credibility of his data. My findings are also based on the explanation provided by Mr. Sunter concerning the effect of inaccurate data on his calculations and the possibility of the injection of bias in the study.

54 Counsel for the complainant argued that to establish a prima facie case in this matter, it need only be demonstrated that 50% plus 1 of the results are accurate. It is clear that 31 of 41 male comparator jobs had errors in the reported data and that 25 of 31 of these results suggest that the male comparator jobs were undervalued, making them appear as if they were overpaid. Six of 10 evaluations of female jobs contained errors that made them appear overvalued and underpaid. Five groups of identical generic job descriptions, comprising a total of 13 work descriptions, had different total scores within each of the 5 groups. Sixty-six inconsistencies were noted in the work description analysis results, which may equate to 66 actual errors. Up to 41 of these errors, Mr. Durber admitted, would alter the total scores for that work description. Other additional errors include the 19 positions that Mr. Durber identified as being wrong and his inclusion of the CO work descriptions when he admitted that they should not have been as it is a gender-neutral job group.

55 I recognize that Mr. Durber evaluated 110 positions and 1200 factors in his study. Using counsel for the complainant's suggestion that as long as 601 are correct, the complainant has met the burden of proof, is not a realistic approach to determining the reliability of Mr. Durber's data. The employer argued that a reasonable person looking at the sheer number of errors and inconsistencies in Mr. Durber's data would conclude that the work of the female-predominant group and the male comparators was not reliably evaluated, and I concur.

56 Mr. Durber's evidence in support of the part of the complaint claiming violations of sections 7 and 10 of the CHRA was also based on his data that he believed indicated that the female-predominant FI-1 and FI-2 combined group was otherwise subjected to systemic discrimination in pay. This claim is also not supported by his data. The errors that lead me to conclude that his data is unreliable for pay equity purposes under section 11 of the CHRA make any conclusion of systemic bias against women in the combined FI-1 and FI-2 groups equally unreliable. There was no other evidence on which to support a conclusion that sections 7 and 10 of the CHRA were violated.

57 As noted by the adjudicator in Bremsak at paragraph 91, the power to hear a non-suit application is discretionary and incidental to the powers of a panel of the Board. It is not a usual practice of the Board; nor is it readily granted. In this case, as counsel for both parties agreed, this would be an appropriate situation in which to exercise the discretionary power to consider the employer's motion for non-suit, given that an additional 10 days of hearing were scheduled, which would have caused great expense to the parties. The parties also concurred in their joint request that a ruling be issued from the bench the same day in order to avoid any further incurring of expenses incidental to the hearing of this matter.

58 For all of these reasons, I find that the job evaluation study conducted by Mr. Durber is on the balance of probabilities unreliable. As this is a requirement of part 3 of the Canada Post test, the complainant has not established on a prima facie basis that the value of the work being compared between the two occupational groups was assessed reliably.

59 For all of the above reasons, the Board makes the following order:

V. Order

60 The complaint is dismissed.

June 16, 2015.

Margaret T.A. Shannon,
a panel of the Public Service Labour Relations and Employment Board