It is alleged that the Canada Post Corporation as Employer, has violated Section 11 of the Canadian Human Rights Act by paying employees in the male-dominated Postal Operations Group more than employees in the female-dominated Clerical and Regulatory Group for work of equal value. The wage rates of the male-dominated Postal Operations Group exceed those of the female-dominated Clerical and Regulatory Group by as much as 58.9 per cent for work of equal value. It is alleged that sex composition of the two groups has resulted in wage discrimination against the Clerical and Regulatory Group, contrary to Section 11.

Corrective Action:

That all employees within the CR Group employed by Canada Post Corporation receive wages, as defined in paragraph 11(6) [now section 11(7)] of the Canadian Human Rights Act, equal to the wages of employees within the PO Group performing work of equal value.
That this corrective action be made retroactive to October 16, 1981.

B. The Investigation Stage

[2] Once a complaint is received by the Canadian Human Rights Commission (the Commission), the general procedures followed are those enunciated in the Canadian Human Rights Act¹ (the Act).

[3] When the Commission receives a complaint for filing under the Act, it may designate an investigator to conduct a preliminary review of the nature and details of the complaint. All parties to the complaint are involved in this review. The respondent's defences to the allegations in the complaint are an integral part of the review.

[4] When this review is completed, the Commission has the authority to determine that a valid defence has been submitted by the respondent, and that the complaint, therefore, cannot be substantiated. Alternatively, the Commission has the authority to appoint a conciliator for the purpose of attempting to bring about a settlement of the complaint. A third option is for the Commission to refer the complaint to the Canadian Human Rights Tribunal for an inquiry involving all parties, including the Commission as a representative of the public interest.

[5] In the case of a complaint brought under section 11 of the Act, the Commission's authority to conduct its investigation includes authority to gather pertinent job fact data. The Commission may request information from the respondent, such as lists of employees, job descriptions, and related job data including input from supervisory and management personnel and employee interviews. Even on-job-site observations may be requested.

[6] The receipt of job fact data is crucial to the Commission's consideration of the complaint and its final recommendation based on the facts it has before it. The value of the work of the male and female employees cited in the complaint needs to be established and compared, as do the wages of the male and female employees. The evaluation process must include consideration of the four factors specified in subsection 11(2) of the Act, namely: skill, effort, responsibility, and working conditions.

[7] The Commission's usual practice is to undertake the work evaluation process using a job evaluation committee, and the employer's own existing evaluation plan, provided it is suitable for a complaint brought under section 11 of the Act. Failing this, the Commission must find an alternative job evaluation plan which is unbiased, gender neutral, and appropriate for the task.

[8] The Commission must investigate, as well, the appropriateness of the comparator chosen and named in the complaint.

[9] Ultimately, the Commission must make its preliminary comparison of the job values and wages between the complainant and the comparator groups named in the complaint. An Investigation Report, based on the Commission findings, will then be drafted. Once the parties involved have vetted it, a Final Investigation Report, with its recommendations, will be presented to the Commissioners of the Commission who will make the final decision regarding the Commission's involvement with the complaint.

[10] In the case of the Complaint before this Tribunal, the Investigation Stage was prolonged. There were a number of reasons for this.

[11] As early as 1982, even before the Complaint was filed, the complainant, the Public Service Alliance of Canada (the Alliance), and the respondent, Canada Post Corporation (Canada Post), had agreed to work jointly on the development of a job evaluation plan, known as System One.

[12] Therefore, during 1984 and most of 1985, the Commission did not pursue its investigation of the Complaint actively. Instead, it made periodic checks on the state of the joint development of System One. Differences of opinion between the Alliance and Canada Post, including the withdrawal, at one point, of the Alliance from active participation in the evaluation development process, led to many delays and limited progress in developing the joint System One plan. Finally, the Commission decided to re-activate its investigation in October 1985.

[13] From late 1985 and through 1986, the Commission was active in developing a Job Fact Sheet, a questionnaire intended for use in gathering current job data for the complainant positions - jobs in the Clerical and Regulatory Group (CR's), a group noted in the Complaint as "female-dominated", and for the named comparator positions - jobs in the Postal Operations Group (PO's), noted in the Complaint as "male-dominated".

[14] During this time, Canada Post expressed serious concern to the Commission about the design and content of the proposed Job Fact Sheet. Additionally, Canada Post expressed to the Commission its reservations about the investigation process in general. The Commission had indicated that the Job Fact Sheet was to be answered by a sample of the CR group first. It was intended, eventually, to be a prime job data-gathering tool for the Commission's investigation.

[15] At the same time that it was expressing its reservations about the Commission's investigation process, Canada Post did answer the Commission's requests for job data information by providing employee printouts and other information. It cautioned that job descriptions and organization charts which were required as attachments to the Job Fact Sheet would often be out-of-date. The Alliance advised the Commission that the job descriptions should be union-approved.

[16] By December 1986, a sample of CR employees at Canada Post had completed the Job Fact Sheet, and had been interviewed by Commission staff, using an Interview Guide created by the Commission to clarify answers given on the Job Fact Sheet. Additionally, during the interview process, relevant supervisory staff had been interviewed to clarify answers given by the incumbents sampled.

[17] From April to September 1987, a number of Commission staff evaluated the sample of 194 CR positions using the data collected in 1986. System One was the basis for these evaluations, although it was an uncompleted plan, and the Alliance had advised against its use for evaluation purposes. These evaluations were eventually set aside, and not used in the final investigation process.

[18] Protracted correspondence, meetings and discussions ensued from late 1987 through to mid-1991 between the Commission and Canada Post concerning the sampling of, and job data collecting from, the PO comparator group. The Commission was unsuccessful in seeking the co-operation of the relevant comparator group unions to collect this information. Moreover, Canada Post questioned the size of the proposed sample of the PO comparator positions, and declined to have the Job Fact Sheet completed by PO employees on company time.

[19] Meanwhile, the Alliance was increasingly concerned with the limited progress in the Commission's investigation of the Complaint. The Commission had threatened, on at least two occasions, to invoke section 58 of the Act, to obtain, from Canada Post, information it required to continue its investigation. Meetings involving senior managers from the Commission and from Canada Post were subsequently held, leading to the development, by the Commission, of a preliminary set of 10 "generic" PO job specifications.

[20] Eventually, the Commission was able to finalize its 10 "generic" PO job specifications based upon data obtained from Canada Post. This took place from July to October 1991. Although Canada Post indicated that the creation of these "generic" jobs excluded several PO jobs, there never was a resolution to this difficulty. Intervening events, such as a union strike in August 1991, extended the investigation time even more. The Commission moved forward, pushed by the concerns of the Alliance which were made evident by its threat to bring an application for mandamus under the Act to compel the Commission to complete its investigation. Its staff commenced the evaluation of CR and PO Benchmark positions, after which the 10 "generic" PO jobs were to be evaluated and the original sample of 194 CR positions was to be re-evaluated.

[21] In the midst of this activity, the Commission's senior investigator was temporarily re-assigned from his position as head of this investigation to address other priorities. To complete the work expeditiously, the PO Supervisory positions were dropped from the Complaint, and the CR sample was reduced from 194 to 93 positions. A consultant was added to Commission staff for the evaluation process which was using, as its evaluation tool, the XYZ Hay Job Evaluation Plan, an off-the-shelf plan. System One could not be used as it had never been accepted by the union, and it was never meant to be used to compare jobs represented by unions other than the Alliance. Moreover, Canada Post had also advised the Commission that System One would not be suitable for evaluating PO jobs.

[22] The Commission completed its CR and PO job evaluations and its investigation work in November 1991. There was no briefing session with Canada Post before the draft Investigation Report was released to the parties on December 16, 1991, along with a request to submit any comments by January 6, 1992. Comments were submitted by both parties by late January 1992; the Commission's Final Investigation Report, dated January 24, 1992, did not incorporate any of them. The Final Report concluded that there was a demonstrable wage difference when comparing wages and job values in the male and female-dominated groups named in the Complaint. The Report recommended referral of the Complaint to the Canadian Human Rights Panel (now known as the Canadian Human Rights Tribunal).

[23] The Commissioners considered the Final Investigation Report and, having regard to all the circumstances of the Complaint, decided, on March 16, 1992, to institute an inquiry into the Complaint by means of a referral to the Canadian Human Rights Tribunal which would assign the matter to a specific Tribunal panel for a hearing.

[24] The Tribunal panel was established on May 11, 1992, a Pre-hearing Conference was held September 21, 1992, and hearings and deliberations got underway on November 25, 1992. The written and oral submissions were completed on August 27, 2003, although written submissions concerning the Decision of the Federal Court of Appeal in the `Airlines Case' [Canadian Human Rights Commission v. Air Canada, Canadian Airlines International Limited and Canadian Union of Public Employees (Airline Division), [2004] F.C.J. No. 483] were submitted in mid-August 2004. In June 2004, the original Chair of this Tribunal, Benjamin Schecter, resigned.

C. Population

Complainant and Comparator Groups

[25] The Commission's Final Investigation Report, dated January 24, 1992, indicates that the total population (with the break-down by job category of each of the complainant and comparator

groups) was as follows (presumably as of a particular date during the Investigation Stage, although no effective date is mentioned in the Report):

Complainant Group (Clerical and Regulatory Group)

CR 2 260

CR 3 950

CR 4 950

CR 5 150

Total Clerical and Regulatory Group 2,3l0

Comparator Group (Postal Operations Group)

Internal Mail Processing and Complementary Postal Service Sub-group

PO INT 2 1,283

INT 3 2

INT 4 18,020

INT 5 1,205

20,510

External Mail Collection and Delivery Services Sub-group

PO EXT 1 17,549

EXT 2 2,224

EXT 3 48

19,821

Supervisory Sub-group

PO SUP 1 549

SUP 2 1,343

SUP 3 427

SUP 4 331

SUP 5 96

SUP 6 22

2,768

Total Postal Operations Group 43,099

[26] By way of comparison, the total population levels of the complainant and comparator groups as presented in the documentation (undated) supporting the August 24, 1983 Complaint are as follows:

Clerical and Regulatory Group (Complainant) - CR's 2,316

Postal Operations Group (Comparator) - PO's

PO INT 25,056

PO EXT 21,661

PO SUP 4,195

PO Total 50,912

D. Setting and Context, 1981 through 1991

[27] To assist in understanding this lengthy and complex case, the Tribunal considers it important that the historical setting and context be identified. In particular, what was going on in the "world" in which all three parties were operating during the crucial years 1981 through 1991?

[28] The Canadian Human Rights Act was enacted on July 14, 1977 and proclaimed in force on March 1, 1978. Section 11 of the Act took effect on March 1, 1978. When this Complaint was filed with the Commission on August 24, 1983, a number of other individual and group complaints alleging discrimination under section 11 of the Act had already been brought by the Alliance, and other public and private sector unions. Tribunals were appointed to hear some of the cases but the majority were settled after negotiation, using Commission facilitators, and with the consent of the Commission.

[29] The Act, a quasi-constitutional human rights statute, enunciates general principles concerning the prohibition of discrimination on particular grounds. It established a Canadian Human Rights Commission which was given the authority to be actively involved in the evolution of the Act through its handling of complaints, and its development and issuance of Guidelines, under subsection 27(2). Additionally, the Commission was required to undertake or to support research programs relating to its duties, and to foster public understanding and recognition of the purposes of the Act, while discouraging and reducing the various discriminatory practices the Act addresses. All of this, undoubtedly, placed challenging demands on the Commission and its staff during this early period. At the same time, this was a period of increasingly tight fiscal management at both the provincial and federal levels of government.

[30] Collective bargaining was introduced to the Canadian Public Service in March 1967 under the aegis of the Public Service Staff Relations Act² (PSSRA) which provided that the government and the Public Service Commission had to promulgate and declare occupational job categories in groups, as a preliminary to formal unionization of government employees. Each job category had to be defined by listing the groups of employees making up that category. Employees at the Post Office, which was a Department of the Canadian government at the time, were included in the same categories and groups as employees in other government departments, except for employees who were directly involved in the handling of mail. This unique group bore the title, "mail handlers" and included postal clerks, letter carriers, mail dispatchers, supervisory mail handlers, and several other functions involved in the sortation and delivery of mail.

[31] The daily movement of massive volumes of different types of mail in a country the size of Canada, with its different time zones and variety of climatic conditions, requires a vast, well-coordinated operational network. Inevitably, such a network includes thousands of corporate or contracted people and thousands of postal outlets in both urban and rural areas, in addition to many mail-processing facilities across the country. The state of employee relations is obviously a vital element in operating such a complex network successfully. Prior to the enactment of the PSSRA in 1967 and the subsequent certification of various unions to represent particular occupational groups of employees within the then Post Office Department, employees tended to be represented, informally, by staff associations. The earliest of such postal associations is believed to have been formed in 1889.

[32] In the 1960's and 1970's, the Post Office Department experienced one of its most unsettled periods of labour relations. While this was a period when the postal code system was introduced (1971) and mechanized mail processing technology was evolving, it was also a period of many management-employee disputes leading to several major strikes.

[33] The Post Office Department was succeeded by Canada Post Corporation with the proclamation of the Canada Post Corporation Act³ on October 16, 1981. One of the objects of the new Corporation, specified in the enabling legislation, was "...the need to conduct its operations on a self-sustaining financial basis while providing a standard of service that will meet the needs of the people of Canada...".⁴ Creation of the Crown Corporation appeared to have the support of all national political parties and most organized labour, business and consumer organizations. There also seemed to be a consensus that one of a number of desirable objectives for the new Corporation would be the reform of its collective bargaining structure in the interests of achieving labour peace.

[34] Upon becoming a Crown Corporation, the bargaining units certified under the PSSRA were deemed to be bargaining units under the Canada Labour Code⁵, and the bargaining agents representing these bargaining units were to remain in place, presumably to provide a transitional period of relative stability and an opportunity for the new Corporation to reorganize. This did, however, pre-empt an early start to the reform of the collective bargaining process which was further delayed by the passage, in 1982, of the federal `6 and 5' cost control legislation. The Canada Labour Relations Board (CLRB) issued a policy statement in February 1984 calling for an overall review of the bargaining unit structure of the Corporation at an appropriate time in the future. This review finally got underway in May 1985 when the Corporation filed its application with the CLRB for study of the appropriateness of all of its then existing bargaining units.

[35] The 1985 CLRB study took the form of a Bargaining Unit Review Process (BURP) with the first phase of hearings concluding in December 1987; CLRB's first decision was released on February 10, 1988. The CLRB heard from eight unions involving twenty-six bargaining units (representing about 58,000 employees), and ordered that they all be consolidated into four bargaining unions and four bargaining units. The four unions are as follows:

Canadian Postmasters and Assistants Association (CPAA)

Canadian Union of Postal Workers (CUPW), comprising the Letter Carriers Union of Canada (LCUC), the International Brotherhood of Electrical Workers (IBEW), the General Labour and Trades Group, the General Services Group, and the original CUPW

Public Service Alliance of Canada (PSAC), representing administrative, technical and professional employees, involving the combination of 15 separate units into one collective bargaining unit

Association of Postal Officials of Canada (APOC), representing operational supervisory employees but excluding lead hands and first-line managers

[36] It was not until 1988 that the bargaining unit consolidation occurred and the 1989-1992 round of labour negotiations was the first held with representatives of the consolidated units - some eight years after achieving Crown Corporation status. Understandably, while the BURP study was on-going, negotiations continued between the Corporation and the original 26 bargaining units. In fact, there were active negotiations during this period with LCUC, CUPW, CPAA, APOC and PSAC, some overlapping with each other and some with special mediation assistance. Despite vigorous negotiations, three strikes occurred in the 1980's, one of which involved PSAC.

[37] The 1989-1992 round of negotiations between the Corporation and CUPW were particularly challenging for all parties, leading to unsuccessful mediation, rotating strikes and Parliamentary back-to-work legislation in 1991. Agreements were concluded during this same period with the other three unions - CPAA, APOC and PSAC - without work stoppages.

II. LEGISLATIVE BACKGROUND

A. Nature of Human Rights Legislation

[38] Human rights legislation is a child of the 1970's. Although at the beginning of the twentieth century, there had been demands, often by women, for equality rights, it would be decades before legislation, both provincial and federal, addressed discrimination in general.

[39] Discrimination in the area of work was addressed after the First World War when the International Labour Organization was founded in 1919. At about that time, the Canadian government legislated a minimum wage for women.

[40] The Universal Declaration of Human Rights⁶ was proclaimed by the General Assembly of the United Nations in December 1948. It was viewed, at the time, as the first step in the formulation of an "international bill of human rights" that would have legal as well as moral force. Article 23 of this Declaration reads in part that "[e]veryone, without any discrimination, has the right to equal pay for equal work".

[41] By 1951, the principle of equal pay for work of equal value was articulated by the International Labour Organization in its C100 Equal Remuneration Convention⁷. This Convention was ratified by Canada in 1972 and signalled Canada's commitment to the active pursuit of the human rights of workers, including the principle of "equal pay for work of equal value". This commitment was reaffirmed when Canada ratified in 1976 the United Nations International Covenant on Economic, Social and Cultural Rights⁸, the International Covenant on Civil and Political Rights⁹, and the Optional Protocol to the International Covenant on Civil and Political Rights¹⁰. These United Nations Covenants made a reality of the dream for an "International Bill of Human Rights".

[42] Article 7 of the International Covenant on Economic, Social and Cultural Rights recognizes the right of everyone to "[f]air wages and equal remuneration for work of equal value without distinction of any kind, in particular women being guaranteed conditions of work not inferior to those enjoyed by men, with equal pay for equal work".¹¹

[43] In 1970, Canada established a Royal Commission to inquire into the Status of Women. The Royal Commission's Report focused on continuing discrimination involving women in the workplace.¹² Female participation in the Canadian workforce had continued to grow over the decades, increasing during the 20-year period 1960 - 1979 by the same percentage as it had taken sixty years to achieve between 1901 and 1961.

[44] Canada's commitment to eliminate discrimination in the workplace was enlarged to include a broader definition of human rights by the promulgation of the Canadian Human Rights Act in 1978 and, in 1981, by Canada's signing of the United Nations Convention on the Elimination of all Forms of Discrimination against Women.¹³

[45] The general goals of human rights legislation are the prevention of discrimination and the promotion of public education to eliminate discrimination. These goals are based on society's belief in equality rights for its members. After the fact, they are an attempt to make victims of discrimination "whole" either through consensual or mandated resolution. Dickson, C.J. noted in Canada (Human Rights Commission) v. Taylor, [1990] 3 S.C.R. 892 that the general purpose of the Canadian Human Rights Act, as set out in section 2, is "...the promotion of equal opportunity unhindered by discriminatory practices...".¹⁴

[46] A legislative protection of human rights demands statutory interpretation which is broad and purposive, which is made in "...a manner consistent with its overarching goals...".¹⁵ In other words, an interpretation of human rights legislation must advance the purpose of that legislation to educate the public and to eradicate discrimination. To do this, the interpretation should give the legislation a generous reading, avoiding a narrow, overly technical analysis. Such an interpretation will construe the rights in the legislation broadly and liberally, while interpreting the legislation's restrictions and exceptions in a stricter manner.

[47] The Supreme Court of Canada noted in Winnipeg School Division No. 1 v. Craton, [1985] 2 S.C.R. 150 that:

Human rights legislation is of a special nature and declares public policy regarding matters of general concern. It is not constitutional in nature in the sense that it may not be altered, amended, or repealed by the Legislature. It is, however, of such a nature that it may not be altered, amended, or repealed, nor may exceptions be created to its provisions, save by clear legislative pronouncement.¹⁶

[48] This characterization of the Canadian Human Rights Act as quasi-constitutional demands a thoughtful and modern approach to its interpretation. The following commentary, taken from

E.A. Dreidger, Construction of Statutes¹⁷and Ruth Sullivan, Dreidger on the Construction of Statutes¹⁸ indicates the modern, contextual approach to statutory interpretation:

...the words of an Act are to be read in their entire context and in their grammatical and ordinary sense, harmoniously with the scheme of the Act, and the intention of Parliament.¹⁹

There is only one rule in modern interpretation, namely, courts are obliged to determine the meaning of legislation in its total context, having regard to the purpose of the legislation, the consequences of proposed interpretations, the presumptions and special rules of interpretation, as well as admissible external aids. In other words, the courts must consider and take into account all relevant and admissible indicators of legislative meaning.²⁰

The meaning of words in legislation depends not only on their immediate context but also on a larger context which includes the Act as a whole and the statute as a whole. The presumptions of coherence and consistency apply not only to Acts dealing with the same subject but also, albeit with lesser force, to the entire body of statute law produced by a legislature...Therefore, other things being equal, interpretations that minimize the possibility of conflict or incoherence among different enactments are preferred.²¹

[49] In addition to these commentaries, the Supreme Court has underlined the need to use the Interpretation Act, as did Iacobucci, J. when he indicated that:

I also rely upon s. 10 of the Interpretation Act, R.S.O. 1980, c. 219, which provides that every Act `shall be deemed to be remedial' and directs that every Act shall `receive such fair, large and liberal construction and interpretation as will best ensure the attainment of the object of the Act according to its true intent, meaning and spirit'.²²

[50] In addition to a consideration of the nature of human rights legislation, and the consequent principles of statute interpretation when dealing with such special legislation, this Tribunal must also, during its decision-making process, consider the history of the Canadian Human Rights Act. That history is examined in the next several paragraphs.

B. History of the Canadian Human Rights Act

[51] As already noted, the Canadian Human Rights Act was enacted in 1977, and proclaimed in force in early 1978. Even though over 25 years have passed, equality rights remain the subject of litigation and discussion. Mme Justice L'Heureux-Dube, speaking after receiving an Honourary LL.D. from the Law Society of Upper Canada in 2002, noted that:

The isms and phobias - racism, sexism, homophobia, and the malevolent rest - are all fountainheads of discrimination and harassment. They have no place in this era of human rights. ...Equality will be the battle of the millennium. At times, equality's standard bearers will feel like they are standing alone and will be harshly criticized for their positions. But, for those who do what is right, affirmation and solidarity come in due course. For it is my firm belief that justice without equality is no justice at all.²³

[52] Section 2 of the Canadian Human Rights Act addresses the goal of equality. It notes that the purpose of the Act is to:

...give effect, within the purview of matters coming within the legislative authority of Parliament, to the principle that all individuals should have an opportunity equal with other individuals to make for themselves the lives that they are able and wish to have and to have their needs accommodated, consistent with their duties and obligations as members of society, without being hindered in or prevented from doing so by discriminatory practices based on race, national or ethnic origin, colour, religion, age, sex, sexual orientation, marital status, family status, disability or conviction for an offence for which a pardon has been granted.

[53] The Minister of Justice at the time, the Honourable Ron Basford, indicated in May 1977 during the Parliamentary debates which preceded the enactment of the Act, and more particularly, section 11 of the Act, that:

There will no doubt be some problems ...[with] the concept [of equal pay for work of equal value]. The federal government has adopted a different approach: that we should legislate the principle and, through the Commission and through its efforts at setting out guidelines, solve those problems...as to how that is to be implemented and how it is to be brought about.²⁴

[54] In other words, section 11 of the Act is an enunciation of a principle, without legislating stringent rules indicating how that principle is to be effected. The Honourable Ron Basford stated that this section of the proposed Act was fashioned to address the specific problem of the occupational segregation of women, with its accompanying historical lower wage rates which were based on the undervaluing of women's work in the marketplace. The need to address this problem had been one of the underlying reasons for the International Conventions of the mid-twentieth century, and was a key recommendation of the Report of the Royal Commission on the Status of Women.²⁵

[55] From these International Conventions and the Royal Commission Report, the broad concept arose of basing wages on the value of work being done. Section 11 of the Act deals with the principle that there should be no discrimination in wage rates based on sex. The basis for the wage should be the value of the work being done.

[56] As the Canadian commitment to International Conventions, and to the recommendations of its own federally-appointed Royal Commission, was addressed by section 11 of the Act, its purpose must be seen in that historical light.

[57] Accordingly, section 11, although complaint-driven, as is the Act in general, may be interpreted as Parliament's means of addressing systemic discrimination based upon sex, in employment.

[58] Although the principle of "equal pay for work of equal value" is the basis for section 11, the Act does not articulate how the principle is to be implemented. While section 11 spells out for the complainant the criteria to be used to assess value of work - that is, the composite of the four factors of skill, effort, responsibility, and working conditions - the evaluation process to be employed is not articulated.

[59] The Commission is given broad authority to deal with the intricacies of section 11, such as the ability to issue binding guidelines concerning certain concepts in the section. This guideline-making power creates what can be described as statutory rules to guide the interpretation of section 11, analogous to the creation of regulations for other legislation.

[60] The Complaint before this Tribunal demands an interpretation of all aspects of section 11. It is believed to be the first complaint based on section 11 of the Act, referred to the Canadian Human Rights Tribunal, to require such a comprehensive review.

III. THE FUNDAMENTAL ISSUES

[61] There are four fundamental issues to be addressed as the Tribunal examines this Complaint. These are identified below, and will be examined in detail in Sections IV, V and VI of this Decision.

A. Independence and Impartiality of the Tribunal

[62] Is the Tribunal an institutionally independent and impartial quasi-judicial body? In particular, does the Act create a reasonable apprehension of institutional bias in the Tribunal because it gives the Commission power to issue Equal Wages Guidelines²⁶ (the Guidelines), which are binding on the Commission, a party before the Tribunal, and binding on the Tribunal?

B. Retroactivity and Validity of the Guidelines

[63] Can a statute be applied retroactively or retrospectively? Can a delegated power to issue subordinate legislation, such as the Guidelines, be exercised retroactively or retrospectively?

[64] What is the test for the validity of subordinate legislation? Are subsection 8(2) and sections 11 to 15 of the 1986 Guidelines²⁷ valid?

C. Proof by Presumption

[65] Evans, J. noted that:

[S]ubsection 11(1) can ... be seen to have tackled the problem of proof by enacting a presumption that, when men and women are paid different wages for work of equal value, that difference is based on sex, unless it can be attributed to a factor identified by the Commission in a guideline as constituting a reasonable justification for it.²⁸

[66] Although all parties in this Complaint accept the above statement by Mr. Justice Evans, the question arises: Is this presumption a presumption rebuttable by factors other than those identified in the Guidelines?

**D. Prima Facie Case**

[67] Has the complainant established a prima facie case of discrimination, based on section 11 of the Act?

[68] A prima facie case has been defined as follows:

...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.²⁹

[69] The standard of proof to determine whether such a prima facie case has been established by the complainant is the civil standard, a balance of probabilities. Once a prima facie case has been established by the complainant, the evidentiary burden shifts to the respondent to show a justification for the discrimination, using the balance of probabilities as the standard of proof.

IV. EXAMINATION OF THE FUNDAMENTAL ISSUES

A. Independence and Impartiality of the Tribunal

[70] Canada Post first raised this issue in May 1998 when its newly retained counsel advised the Tribunal that a Motion concerning "the judicial and institutional independence" of the Tribunal was being considered.

[71] The issue of the Tribunal's independence and impartiality had arisen after the release, on March 23, 1998, of a decision by McGillis, J. in the Federal Court (Trial Division).³⁰

[72] That decision was the culmination of a judicial review of a decision made by the tribunal appointed to hear a complaint, brought under section 11 of the Act by some Bell Canada employees and their unions. After hearing argument on Bell Canada's Motion requesting that the tribunal find itself unable to proceed due to an apprehension of bias, the tribunal ruled that it was "...an independent quasi-judicial body, institutionally capable of providing a fair hearing in accordance with the principles of natural justice".³¹

[73] The judicial review application was allowed, the Federal Court concluding that the legislative scheme of the Act, at the time, did not provide tribunal members with sufficient security of tenure or financial security to allow the tribunal to operate with independence and impartiality. Of concern was a link between the tribunal members' remuneration and the Commission, which would be a party before most tribunals. Additionally, McGillis, J. indicated that there were concerns about the Commission's ability to make binding Guidelines. McGillis, J. commented, in obiter dicta, that "...the manner in which, in the opinion of the Commission, any provision of this Act applies in a particular case..." should cause concern for the ability of the tribunal to act independently and impartially.³² Such a problem could, suggested McGillis, J., be corrected by making the Guidelines non-binding on the tribunal.

[74] Therefore, all proceedings in the Bell Canada complaint were quashed and an Order was rendered that no further proceedings in the complaint take place until legislative changes were made to address the relevant problems.

[75] Canada Post brought its Motion in June 1998, requesting the following:

an Order or ruling by the Tribunal that it is not an independent or impartial tribunal capable of providing a fair hearing in accordance with the principles of fundamental justice guaranteed by section 2(e) of the Bill of Rights, R.S.C. 1985, because, inter alia, it is bound by the Guidelines in interpreting section 11 of the Canadian Human Rights Act issued by the Canadian Human Rights Commission, a party in interest before it;
an Order or ruling by the Tribunal that it is not an independent or impartial tribunal capable of providing a fair hearing in accordance with the principles of fundamental justice guaranteed by section 2(e) of the Bill of Rights because the remuneration of the Tribunal's members is fixed by the Commission and, from the beginning of the Tribunal hearings until January 1, 1997, was provided to the members by cheques issued by the Commission;
in the alternative, an Order by the Tribunal referring the questions raised above to the Federal Court under section 18.3 of the Federal Court Act R.S.C. 1985 c. F-7 and Rules 320 and 323 of the Federal Court Rules, 1998.

[76] The Motion was argued in August/September 1998. The Tribunal issued its decision on October 21, 1998, dismissing Canada Post's Motion, as follows:

With regard to the issues of financial security in a tribunal and the security of tenure of tribunal members, the Tribunal concludes that there is no question that waiver is available as an objection to an allegation of a reasonable apprehension of bias, as demonstrated by authorities cited.... One fact is indisputable. At no time during the last six years, did Mr. Juriansz, counsel for the Respondent, raise the issues of security of tenure or financial security of the Tribunal... The Tribunal concludes that, on a balance of probabilities, the Respondent would have had the knowledge to object, in a timely fashion, to the jurisdiction of this Tribunal based upon a reasonable apprehension of bias arising from these two issues. Therefore, because such an objection was never made, for whatever reason, the Respondent must be deemed to have impliedly waived its right to challenge the independence or impartiality of the Tribunal on the basis of the said two issues.³³

[77] Concerning Canada Post's argument that the binding nature of the Guidelines, created by the Commission, a party before the Tribunal, produced a situation where the Tribunal could not

provide a fair hearing in accordance with the principles of fundamental justice, the Tribunal ruled as follows:

With respect to the binding nature of the Guidelines pursuant to section 27(3) of the Act, the Tribunal finds that from the outset of the hearing there was an undisputed understanding amongst the parties of the day to address this issue in final submissions, after hearing evidence of all the parties.

Another way of looking at this particular matter is that the exercising of the Respondent's right to object to the issue of the binding nature of the Guidelines on the grounds of invalidating the Tribunal's independence and impartiality, was postponed, by consent, from the start of the proceedings to closing argument, because all parties agreed it was wise that the Tribunal hear evidence first so the Tribunal would understand what the Guidelines were intended to accomplish.

The Respondent's right to object has, therefore, not been rescinded-- it has been reserved and remains in place to be exercised "at the end of the day". There is no question of waiver here. Nothing has been waived with respect to the Guidelines issue - just an understanding and concurrence openly and fairly arrived at, to address that issue later on.³⁴

[78] Accordingly, the Tribunal continued to hear evidence.

[79] On June 30, 1998, a number of amendments to the Act came into effect, including the following:

Subsection 27(2) provides for the Commission to issue a guideline binding on the Commission and a tribunal only "in a class of cases described in the guideline" rather than "in a particular case or in a class of cases".

Subsection 48.2(2) recognizes that a tribunal member whose appointment expires "... may, with the approval of the Chairperson, conclude any inquiry that the member has begun".

Subsection 48.6(1) provides that tribunal members shall be paid "... such remuneration as may be fixed by the Governor in Council" rather than "... as may be prescribed by by-law of the Commission".³⁵

[80] After the Act had been amended in June 1998, the Vice-Chairman of the Canadian Human Rights Tribunal decided to proceed with the tribunal hearing of the Bell Canada complaint. Bell Canada, however, maintained its position that, even with the amendment to subsection 27(2) of the Act, the tribunal was precluded from making an independent judgement in any class of cases in which binding Guidelines were issued by the Commission, a party in interest before the tribunal. It argued, in a judicial review application of the April 1999 decision³⁶ to proceed with the Bell Canada hearing, that the binding nature of the Guidelines leads to an inevitable perception of bias and lack of institutional independence. Tremblay-Lamer, J. of the Federal Court (Trial Division) agreed. In a decision rendered on November 2, 2000, the binding nature of the Guidelines issued by the Commission was found to be incompatible with the guarantees of institutional independence and impartiality necessary to the tribunal's decision-making powers.³⁷

[81] The Federal Court of Appeal decision of May 24, 2001 reversed the decision of Tremblay-Lamer, J.³⁸ This Appeal decision was upheld by the Supreme Court of Canada, which issued its decision on June 26, 2003.³⁹ It found that subsections 27(2) and (3) of the Act, as amended, relating to the issuance of binding Guidelines, were not inconsistent with section 2(e) of the Canadian Bill of Rights, S.C. 1960, c. 44, which requires that parties be given a "fair hearing in accordance with the principles of fundamental justice". Neither were the subsections inconsistent with the constitutional principle of adjudicative independence. Therefore, those subsections of the Act were found to be operable and applicable.

[82] The Supreme Court of Canada addressed Bell Canada's specific argument that the binding nature of the Guidelines creates a perception that a tribunal, hearing a complaint, lacks independence and impartiality. Further, Bell Canada had argued that guidelines, created by a party before it, and binding on the tribunal, would create an apprehension of bias. The Supreme Court noted the following:

As the Commission has readily acknowledged, the guideline power is constrained. The Commission, like other bodies to whom the power to make subordinate legislation has been delegated, cannot exceed the power that has been given to it and is subject to strict judicial review ... The Tribunal can, and indeed must, refuse to apply guidelines that it finds to be ultra vires the Commission as contrary to the Commission's enabling legislation, the Act, the Canadian Charter of Rights and Freedoms and the Canadian Bill of Rights. The Tribunal's power to `decide all questions of law or fact necessary to determining the matter' under s. 50(2) of the Act is clearly a general power to consider questions of law; including questions pertaining to the Charter and the Canadian Bill of Rights ... No invalid law binds the Tribunal. Moreover, the Commission's guidelines, like all subordinate legislation, are subject to the presumption against retroactivity. Since the Act does not contain explicit language indicating an intent to dispense with this presumption, no guideline can apply retroactively. This is a significant bar to attempting to influence a case that is currently being prosecuted before the Tribunal by promulgating a new guideline. Finally, any party before the Tribunal could challenge a guideline on the basis that it was issued by the Commission in bad faith or for an improper purpose; and no guideline can purport to override the requirements of procedural fairness that govern the Tribunal.⁴⁰

Parliament's choice was obviously that the Commission should exercise a delegated legislative function. Like all powers to make subordinate legislation, the Commission's guideline power under ss. 27(2) and 27(3) is strictly constrained. We fail to see, then, that the guideline power under the Act would lead an informed person, viewing the matter realistically and practically and having thought the matter through, to apprehend a `real likelihood of bias'.⁴¹

[83] Thus, the Supreme Court of Canada has answered the argument about the operation of the Guidelines and their impact upon the impartiality and independence of this Tribunal. That argument was left "to the end of the day" in the Tribunal's decision on Canada Post's 1998 Motion. The Supreme Court's lengthy discussion of the Commission's guideline-making power under the Act is as applicable to the power given to the Commission when the Act was first enacted as it is today.

[84] In its oral submission on the Supreme Court's decision in the Bell Canada case, Canada Post maintained its stance that the Supreme Court decision did not address tribunals that were constituted and operating prior to the enactment of the 1998 amendments to the Act. Canada Post cited the opening paragraph of the Supreme Court decision in Bell Canada which identified the issue before the Court as being whether the Tribunal lacked independence and impartiality because of the power of the Commission to issue guidelines "...concerning a `class of cases'..." which would be binding on the tribunal.

[85] The Supreme Court was therefore, according to Canada Post's submission, addressing post-1998 tribunals. It was not until the 1998 amendments that the Commission's guideline-making power was confined to a `class of cases'. Prior to the 1998 amendments, former subsection 27(2) of the Act authorized the issuance of guidelines in respect of `a particular case' as well as a `class of cases'.

[86] Canada Post's argument was that the current Tribunal, having been established in 1992, was not encompassed by the Supreme Court's Bell Canada decision.

[87] The Commission's position on this matter was that it had been specifically dealt with by the Federal Court of Appeal in the unanimous decision in Northwest Territories v. Public Service Alliance of Canada, [2001] F.C.J. No. 791. At paragraph 41 of that decision, the Federal Court of Appeal noted, as follows:

The appellant [Government of the Northwest Territories] contends that the amended provision still compromises the independence and impartiality of the Human Rights Tribunal Panel assigned to hearing the complaint against it. The appellant assumed, rightly so in my view, that subsection 27(3) as it now exists in its more restricted form due to the amendment to subsection 27(2) applies to the hearing of the complaint against it ... It is reasonable in the circumstances to infer that Parliament intended the new but more limited subsections 27(2) and (3) to continue to apply to inquiries in respect of a class of cases, such as this one, commenced before the amendment and continued thereafter, especially as the 1998 amendment was remedial and aimed at suppressing a possible violation of the requirements of natural justice. The combined effect of the amendment and the transitional provision was, on the one hand, to restrict the CHRC's power to issue binding guidelines to classes of cases and, on the other hand, to allow the guidelines already issued in respect of a class of cases to be binding on the three members of the Human Rights Tribunal Panel completing the inquiry in this case.

[88] The Court of Appeal, in the Commission's view, concluded that the amended section 27 applied to the proceedings of the Northwest Territories case despite the fact that its tribunal had been appointed prior to the 1998 amendments. The panel continued under the transitional provisions of the Act while not interfering with the application of the amended Act. Also, no binding guidelines specific to the appellant's case had been issued by the Commission.

[89] The Commission argued that, since the Northwest Territories case was governed by the same pre-1998 provisions as this Tribunal, the Court of Appeal's decision, which is uncontradicted by the Supreme Court's decision in Bell Canada, is equally applicable and, indeed, binding on this Tribunal.

[90] As with the Northwest Territories case, this Tribunal continued under the Act's transitional provisions and was bound by the section 27 amendment of 1998. Finally, no case specific guideline had been issued by the Commission.

[91] The Tribunal finds the Commission's submission to be more persuasive, and agrees that the Federal Court of Appeal decision in the Northwest Territories case is relevant and binding on its deliberations.

[92] This Tribunal, for all the reasons noted in this Section, finds that it is, itself, an independent and impartial quasi-judicial body, capable of providing a fair hearing in accordance with the principles of fundamental justice.

B. Retroactivity and Validity of the Guidelines

(i) Background

[93] In addition to being a complete answer to Canada Post's argument concerning the binding nature of the Guidelines in relation to the independence and impartiality of the Tribunal, the Supreme Court of Canada has also, in its decision of June 2003⁴², addressed the issue of the retroactivity and validity of the Guidelines.

[94] As noted in paragraph [82], above, although the Guidelines are described as "binding" they are binding on the Tribunal only if they are not invalid, for "no invalid law binds the Tribunal". The Tribunal may find that the Guidelines have been drafted by the Commission in such a way that they "exceed the power that has been given to it...[and are therefore,] ultra vires the Commission as contrary to the Commission's enabling legislation, the Act, the Canadian Charter of Rights and Freedoms, and the Canadian Bill of Rights".

[95] Additionally, the Supreme Court addressed the "retroactivity" of the Guidelines. It noted that:

...the Commission's guidelines, like all subordinate legislation, are subject to the presumption against retroactivity. Since the Act does not contain explicit language indicating an intent to dispense with this presumption, no guideline can apply retroactively. This is a significant bar to attempting to influence a case that is currently being prosecuted before the Tribunal by promulgating a new guideline.⁴³

[96] As the Supreme Court also noted, "...any party before the Tribunal could challenge a guideline on the basis that it was issued by the Commission in bad faith or for an improper purpose...".⁴⁴

[97] Before this Tribunal, Canada Post argued, based upon its interpretation of the presumption against retroactivity, that the Guidelines which must be used for this Complaint are those which were in force at the time the Complaint was filed with the Commission in 1983. Therefore, the argument is that only the 1978 Guidelines (amended in 1982)⁴⁵ should be of interest to this Tribunal in its decision-making process.

[98] Additionally, Canada Post argued that, if the Tribunal rejects its submissions concerning retroactivity and accepts the 1986 Guidelines as pertinent to this Complaint, some of those 1986 Guidelines should be found to be invalid. Canada Post challenges the 1986 Guidelines, subsection 8(2), and sections 11 through 15.

[99] There has been no challenge to a guideline based upon an argument that the guideline was promulgated in bad faith or for an improper purpose. Guidelines are promulgated only after the Commission has received input from various interest groups such as federally regulated companies, government agencies, and government departments. In this case, amongst those interested and actively involved in giving advice to the Commission before the promulgation of the 1986 Guidelines, was Canada Post.

(ii) How is the Concept of "retroactivity" pertinent to this Complaint?

a) Submissions of the Parties

[100] All parties agree, in submissions concerning retroactivity, that, as Canada Post articulated in its submissions, "retroactivity is a question of what law applies at a particular point in time".⁴⁶

[101] As Canada Post noted in its submissions concerning retroactivity, administrative law academics, like Sullivan and Dreidger, have written volumes on the retroactive application of the law. In the words of Canada Post's counsel, "a retroactive application of a law changes the past effects of a past situation, a situation giving rise to the effects is past and the effects are past".⁴⁷

[102] Canada Post submissions continued, noting that:

a retrospective application of the law change[s] the future effects of past situations. The situation with which we are concerned is already past but the effects haven't all past (sic). Some are in the future, and if the law can change them it is a retrospective application ... An immediate application of the law changes the future effects of an ongoing situation ... the law applies as of the day it comes into force. So, anything that is happening after that, the law applies ... the prospective application of law, where the law that comes into force can only apply to situations and effects that arise after ...What about the situations that had already started before it came into effect?... The old law survives, the law that has been repealed, the Guideline that has been revoked, applies, but only for the limited purpose of governing the situations until they are over ... So, even though the Act says that Guidelines are revoked when new Guidelines are issued ... the concept of survival overrides that and let's (sic) the old law apply if it is necessary to do so because the new law is only prospective.⁴⁸

[103] These submissions by Canada Post outline the different applications of the law based on the timing of what Canada Post has called "situations" and the necessity that the law applicable to those "situations" be used. In its argument, Canada Post emphasized the prejudice which would accrue to any respondent who was unable to know, with specificity, what the complaint against him or her was. Without that knowledge, Canada Post argued, a respondent would be deprived of an ability to make a full answer to the complaint. The necessity for fairness to all is the foundation for the presumption against retroactivity. In general, Canada Post argued, the "rules of the game" must be known before the game is played; that is to say, a respondent must know what law is applicable at the time the respondent is served with a complaint, unless there is specific language in the legislation which allows for a change, "mid-game".

[104] This argument anticipated the Supreme Court of Canada's review of retroactivity in the Bell Canada decision of June 2003. That decision specifically pointed to the lack of inclusion in the Canadian Human Rights Act of an intention that the Guidelines be applied retroactively. Therefore, the Guidelines cannot be applied retroactively.

[105] If the Guidelines cannot be applied retroactively, what is the "situation" which pinpoints the time when a specific guideline is to be applied? During its submissions on this topic, Canada Post presented the hypothetical example of a contractual summer employment arrangement involving an hourly minimum wage rate which, through legislation in mid-summer, changed. In such a case, the wage rate changes when the legislation is promulgated, notwithstanding the contractual arrangement. The new wage rate is not retroactive to the beginning of the contract. This example involves specificity. There is a contract. There is a specific legislated change as of a specific date. There is no grey area in the example. An allegation of discrimination is not part of the equation.

[106] Canada Post argued that, similarly, there is no grey area in the Complaint before this Tribunal. The date the Complaint was brought to the Commission should be the date which seals the law applicable to the Complaint. Canada Post argued that a respondent must know what law is applicable at the time the respondent faces a complaint. According to Canada Post, this is important because, from the time a complaint is made, a respondent must know what the rules are in order to articulate its position. During the investigation of a complaint, the respondent's position will be influenced by those "rules". If there is a change in the "rules" after a complaint is brought, Canada Post argued, the respondent will be prejudiced.

[107] Canada Post made a further argument that employing the 1986 Guidelines would interfere with its vested right to rely on the defences it had under the 1978 Guidelines when the Complaint was first filed. In particular, Canada Post cited Gustavson Drilling (1964) Ltd. v. Minister of National Revenue⁴⁹ to support the position that regardless of whether legislation is retroactive or even retrospective, it is presumed that there is no intention to interfere with vested rights (unless the legislature intends otherwise).

[108] The protection of vested rights, argued Canada Post, is reinforced by the federal Interpretation Act where the term `enactment' includes a statute or a regulation. Section 43(c) of the Interpretation Act reads as follows:

43. Where an enactment is repealed in whole or in part, the repeal does not...

(c) affect any right, privilege, obligation or liability acquired, accrued, accruing, or incurred under the enactment so repealed...

[109] Canada Post maintained that its vested rights would be infringed if the 1986 Guidelines were applicable because they impose on the Tribunal rules for interpreting section 11 of the Act which differ from the 1978 Guidelines rules in ways that are important to Canada Post's defence. Canada Post cited a number of such differences which it believed would produce an unfair result.

[110] As already noted, all parties agreed that there is a presumption against retroactive application of legislation unless otherwise provided in the enabling statute. The Commission also agreed that, in the context of subordinate legislation, there is a legal restriction against such application rather than a presumption.

[111] Further, the Commission agreed with Canada Post that the definition of retroactive application is the application of a new law to past facts. The Commission, however, stressed that the facts must be completed.

[112] In referring to the temporal application of law, the Commission's submissions drew on Professor Sullivan's writings regarding the need to situate facts in time:

Legislation clearly is retroactive if it applies to facts all of which have ended before it comes into force. Legislation clearly is prospective if it applies to facts all of which began after its coming into force. But what of on-going facts, facts in progress? These are either continuing facts, begun but not ended when the legislation comes into force, or successive facts, some occurring before and some after the commencement. The application of legislation to on-going facts is not retroactive because ... there is no attempt to reach into the past and alter the law or the rights of persons as of an earlier date...[emphasis added]⁵⁰

[113] It is therefore, in the Commission's view, important to identify the particular set of facts that is relevant to the case concerned. In the case before the Tribunal, it is a question of what facts were in play when the 1986 Guidelines came into force. The Commission argued that the facts at that time were clearly "on-going" because the Complaint addresses alleged systemic wage discrimination.

[114] Drawing on Professor Sullivan's writings again, the Commission noted:

Such an application [to on-going facts] may affect existing rights and interests, but it is not retroactive. Legislation that applies to on-going facts is said to have "immediate effect"...⁵¹

[115] The Commission emphasized the on-going nature of systemic discrimination by referring to a decision of Mr. Justice Hugessen:

Systemic discrimination is a continuing phenomenon ... By its very nature, it extends over time.⁵²

[116] Consequently, the Commission concluded that when the 1986 Guidelines came into effect, they applied immediately and generally to all on-going facts - that is, facts that started in the past and continued to the present or future. The facts involved in an allegation of systemic wage discrimination would be such on-going facts.

[117] The Commission argued, additionally, that the 1986 Guidelines did not apply new legal consequences to past facts, and did not change the past legal consequences of past facts. It was, therefore, not a retroactive application of the 1986 Guidelines. Rather, the 1986 Guidelines codified the evolving Commission practice concerning the interpretation of section 11 of the Act.

[118] With respect to Canada Post's position that its vested rights would be infringed by the use of the 1986 Guidelines, the Commission argued that the concept of vested rights is not easily applied in the field of human rights adjudication. The Commission noted that the only cases cited by Canada Post in support of its position related to relevant facts that were in the past and were found in a torts context. This is in stark contrast with the Complaint which deals with on-going facts in a human rights context.

[119] The Commission also argued that it is difficult to visualize how the 1986 Guidelines could interfere with pre-existing rights or impose new obligations on Canada Post, because the Guidelines simply interpret and give precision to rights and obligations that pre-existed their enactment. They do not in any way lead to changes in the law.

[120] Moreover, the Commission asserted, the 1986 Guidelines did not remove any defences previously recognized by the 1978 Guidelines on which Canada Post might have wished to rely. In fact, the 1986 Guidelines added to the list of `reasonable factors' found in the 1978 Guidelines but Canada Post had not relied on any `reasonable factor' defences. Therefore, the change in the 1986 Guidelines concerning `reasonable factors' was not something which adversely affected Canada Post's position.

[121] The Commission observed that the presumption against interference with vested rights normally involves ambiguity in the interpretation of statutes or regulations. The Commission's submission was that there is no ambiguity in this case, since the regulation-making power in section 27 of the Act makes it clear that guidelines apply immediately:

27(3) A guideline issued under subsection (2) is, until it is revoked or modified, binding on the Commission and any member or panel assigned under...(emphasis added)

[122] Finally, the Commission made the point that even if there were ambiguity, the presumption against interference with vested rights protects only those rights that had vested at the time of legislative or regulatory amendment. Not only is the legal scope of vested rights important, but also of import are the public policy issues that arise from the presumption.

[123] In 2001, Marceau, J. noted in Veale v. Law Society of Alberta, that "[t]here is no concrete definition of what constitutes a `vested right', primarily because it is difficult to generalize across the cases and as each case must be studied individually".⁵³ In speaking of the review of retrospective legislation, Marceau, J. indicated that judges are often faced with a policy conundrum - "...whether to apply the new and improved law for the greater good even though this may be unfair to some, or to delay the application of that law in respect of some because of the injustice they would suffer".⁵⁴

[124] The Commission argued that the 1986 Guidelines benefit the greater good by bringing much needed procedural detail to the interpretation of section 11 of the Act, while causing no injustice to Canada Post.

[125] For all these reasons, the Commission concluded that the presumption against interference with vested rights does not apply in this instance.

b) Tribunal's Analysis

[126] Canada Post is clear in stating its position that neither a statute nor subordinate legislation can be applied retroactively, and the date this Complaint was brought to the Commission should be the date which seals the applicable law. Canada Post also argued that the 1986 Guidelines would infringe on its vested right to rely on defences it had under the 1978 Guidelines which were in effect when the Complaint was filed with the Commission in 1983. Hence, the 1978 Guidelines should prevail.

[127] Put another way, Canada Post argued that its submissions were in accord with the Supreme Court of Canada's view, as noted in the Bell Canada decision, that the Guidelines can properly influence the outcome of future cases where no-one, including the Commission, can anticipate whose particular interests the Guidelines will favour. Canada Post argued that the Guidelines could improperly influence the outcome of a case where their particular impact is already known and their application is controlled by the Commission's timing of referring that case to a tribunal. Hence, the Guideline-making power cannot be interpreted to permit the Commission to apply a Guideline to a complaint it is already investigating when the Guideline is issued.

[128] Thus, Canada Post argued that the Supreme Court's reasoning in the Bell Canada case supports Canada Post's position that the 1986 Guidelines should not apply to the Complaint because to do so would permit the Commission to influence, improperly, its outcome. The Commission was already investigating the Complaint when the 1986 Guidelines were introduced. By virtue of its decision to issue the 1986 Guidelines before referring the Complaint to the Tribunal, the Commission, in Canada Post's submission, controlled what the Tribunal was bound to apply to the Complaint, knowing its likely impact on the outcome of the Complaint.

[129] Interestingly, while Canada Post, in its submissions (p. 14-15), stated that the Supreme Court decision "...strongly supports Canada Post's position that the 1986 Guidelines should not apply to this Complaint at all because to do so would permit the Commission to improperly influence its outcome", Canada Post did not cite any examples of such improper influence or even any hints of such improper influence by the Commission. It is presented simply as a possible threat of impropriety, a suggestion of a creation of possible bias or impartiality, without substantiation.

[130] While agreeing with Canada Post that there is a presumption against retroactive application of legislation, and a legal restriction with respect to subordinate legislation, the Commission has argued, in line with Professor Sullivan's thesis, that the Complaint before this Tribunal deals with on-going facts. These facts relate to an allegation of on-going sexual discrimination in wages, as described in section 11 of the Act.

[131] The Commission has maintained that the application of legislation, including subordinate legislation, to on-going facts is not retroactive because there is no attempt to alter past law or the rights of persons as of an earlier date. The Commission's position is that legislation or regulations that apply to on-going facts have immediate, not retroactive, effect.

[132] With respect to Canada Post's point that its vested rights would be infringed if the 1986 Guidelines were applicable, the Commission dismissed this concern in the absence of any ambiguity in interpreting the statute and the Guidelines. The Commission was also of the view that Canada Post failed to demonstrate that the 1986 Guidelines removed any defences previously recognized by the 1978 Guidelines on which Canada Post might have wished to rely, at least with respect to the critical `reasonable factors'.

[133] The Complaint before this Tribunal involves an allegation of sexual discrimination in wages, as described in section 11 of the Act. That allegation is one of systemic discrimination. Section 11 of the Act was drafted using, as its primary basis, the International Labour Organization's 1951 Convention 100 (ratified by Canada in 1972) as well as recommendations from the Report of the Royal Commission on the Status of Women.⁵⁵ These historical documents addressed the issue of systemic discrimination against women in the area of wages, with the most basic recommendation being that all wages be based on the value of the work being performed.

[134] Systemic discrimination has been defined by the Supreme Court of Canada [C.N.R. v. Canada (Human Rights Commission), [1987] 1 S.C.R. 1114 at 1139] as follows:

...systemic discrimination in an employment context is discrimination that results from the simple operation of established procedures of recruitment, hiring and promotion, none of which is necessarily designed to promote discrimination. The discrimination is then reinforced by the very exclusion of the disadvantaged group because the exclusion fosters the belief, both within and outside the group, that the exclusion is the result of "natural" forces, for example, that women "just can't do the job" (see the Abella Report, pp. 9-10). To combat systemic discrimination, it is essential to create a climate in which both negative practices and negative attitudes can be challenged and discouraged.

[135] The discrimination being alleged in the Complaint is, therefore, ongoing, by definition.

[136] In addressing the issue of retroactivity, both Canada Post and the Commission have made reference to the Gustavson Drilling case and to Professor Sullivan's volume on Driedger on the Construction of Statutes.⁵⁶ Some of these references have already been identified above, but, given the complexity of the subject, the Tribunal considers it appropriate to examine Professor Sullivan's relevant writings in greater depth.

[137] Professor Sullivan states that the terms `retroactive' and `retrospective' are often used interchangeably but the growing trend is to define `retroactive' legislation as legislation that applies to past facts and changes the past legal consequences of completed transactions. `Retrospective' legislation changes the future consequences of completed transactions by imposing new liabilities or obligations.⁵⁷ The Tribunal accepts use of the term `retroactive' as defined by Professor Sullivan.

[138] Professor Sullivan indicates that the 1977 Gustavson Drilling decision confirmed that it is strongly presumed that legislation is not intended to have a retroactive application to facts that occurred before the legislation came into force. To apply this principle, Professor Sullivan writes that it is necessary to identify which facts are relevant to the legislation and to situate them in time relative to its effective date.

[139] The relevant facts are the `fact-situation' of the case concerned, which, with respect to the case before this Tribunal, includes alleged systemic wage discrimination.

[140] Situating the facts in time involves, in Professor Sullivan's model, determining whether the fact-situation is ephemeral, continuing or successive. She defines these choices as follows:

Ephemeral fact situations consist of facts that begin and end within a short period of time, such as actions or events. The facts are complete and become part of the past as soon as the action or event ends; the legal consequences attaching to the fact-situation are fixed as of that moment.

(...)

Continuing fact situations consist of one or more facts that endure over a period of time...A continuing fact can be any state of affairs or status or relationship that is capable of persisting over time ... Where no limit in time is stipulated, a continuing fact situation continues and does not become part of the past until the fact-situation itself - the state of affairs or condition or relationship - comes to an end.

(...)

Successive fact situations consist of facts, whether ephemeral or continuing, that occur at separate times ... A fact-pattern, defined in terms of successive facts, is not complete and does not become part of the past until the final fact in the series, whether ephemeral or continuing, comes to an end.⁵⁸

[141] Professor Sullivan goes on to say that once the fact-situation has been identified - and, in this case, the Tribunal considers it to be a continuing fact-situation - the test set out in the legislation must be applied to the relevant facts. An application is not retroactive unless all the relevant facts were past when the provision came into force. With respect to a state of affairs such as the on-going systemic wage discrimination alleged in this Complaint, the provision (in this case, the 1986 Guidelines) is not retroactive unless the state of affairs has ended before commencement of the provision. Clearly, the position of the Commission, supported by the Alliance, is that the alleged systemic discrimination state of affairs did not end when the 1986 Guidelines became effective.

[142] The application of legislation, whether statutory or subordinate, to on-going facts or facts-in-progress, is not, according to Professor Sullivan, retroactive because "...to use the language of Dickson, J. in the Gustavson Drilling case, there is no attempt to reach into the past and alter the law or the rights of persons as of an earlier date".⁵⁹

[143] Professor Sullivan continues:

Legislation that applies to on-going facts is said to have `immediate effect'. Its application is both immediate and general: `immediate' in the sense that the new rule operates from the moment of commencement, displacing whatever rule was formerly applicable to the relevant facts, and `general' in the sense that the new rule applies to all relevant facts, on-going as well as new.⁶⁰

[144] Although Canada Post submitted that to use the 1986 Guidelines to interpret section 11 of the Act for a complaint that originated in 1983 would amount to applying those Guidelines retroactively, the Tribunal finds that one is not dealing with the retroactivity of the 1986 Guidelines in this case. One is dealing with what Professor Sullivan has called a continuing "state of affairs" fact-situation. When the 1986 Guidelines came into effect they applied immediately and generally to all the on-going facts that started in the past and continued to the then-present and to the future. This included all facts involved in the alleged systemic wage discrimination.

[145] Therefore, the Tribunal concludes that the 1986 Guidelines are not being applied retroactively in this case, but are addressing an on-going, and continuing, fact-situation without being unfair or prejudicial to Canada Post.

[146] It is appropriate to address the Commission argument, made after the Supreme Court of Canada's decision of June 2003 in the Bell Canada case. The Commission submitted that the relevant point in time for determining what law applies to a complaint is the date of its referral to a tribunal. This point in time was described by the Commission as "the point of crystallization".

[147] The Commission stated that, once referral has been effected, new guidelines issued by the Commission during the life of a tribunal would not apply to the referred complaint. To do otherwise would constitute retroactive application of those new guidelines which is clearly unacceptable.

[148] The Commission further argued that the Supreme Court of Canada, in paragraph 47 of its Bell Canada decision "...appears to accept the position taken by the Commission before the Court that the referral date is the relevant cut-off point".⁶¹ That paragraph acknowledges that the Guidelines, like all subordinate legislation, are subject to the presumption against retroactivity. The Supreme Court indicated that the presumption "...is a significant bar to attempting to influence a case that is currently being prosecuted before the Tribunal by promulgating a new guideline".⁶²

[149] Canada Post's arguments on this same Supreme Court decision underlined that the Supreme Court, while stating the principle that no guideline can apply retroactively, did not declare that the date of the referral to a tribunal is the point in time for determining what law applies to a complaint. The Court, according to Canada Post's argument, simply cited an example of a hypothetical case being considered before a tribunal and indicated that retroactivity could not apply as it would be improper, in such an example, to allow the Commission to influence the outcome of the case by means of the promulgation of a new guideline. Under such circumstances, the Commission would be a party before the tribunal and also the drafting agency for the new guideline which would be, according to the Act, binding on the tribunal.

[150] The Tribunal finds Canada Post's argument to be the more persuasive one. The Tribunal does not consider the Supreme Court's decision in the Bell Canada case to have endorsed the date of a complaint's referral to the Canadian Human Rights Tribunal as the relevant cut-off point for determining what law applies to a complaint. Rather, the Supreme Court has cited but one obvious example to illustrate that the Commission's guideline-making power is constrained and cannot be applied retroactively. Moreover, the example underlines the Supreme Court's comment that a party is always at liberty to question the propriety of the Commission's guideline-making power, based on an argument that the guideline was made in bad faith or for an improper purpose.

[151] Canada Post has also argued that neither the Commission nor the Alliance can fairly or legally rely on the 1986 Guidelines in addressing the Complaint because that reliance would interfere with Canada Post's vested right to rely on defences available to it as of the date the Complaint was filed in 1983.

[152] Canada Post is arguing that the 1986 Guidelines impose on the Tribunal rules for interpreting section 11 of the Act which differ, in ways important to Canada Post's defences, from the rules that prevailed in the 1978 Guidelines. One of three examples of such differences mentioned by Canada Post was the ability, in the 1978 Guidelines, to include, when evaluating jobs, the value of overtime or shift work premiums. Since the 1986 Guidelines prohibit this inclusion, Canada Post argued that the latest Guidelines have removed a right of defence which was vested for Canada Post as of the date of the filing of the Complaint. This produces, according to Canada Post, an unfair result.

[153] As already noted, the Commission, in its submissions on vested rights, referred to the 2001 decision of Marceau, J. in which he stated that "[t]here is no concrete definition of what constitutes a `vested right', primarily because it is difficult to generalize across the cases and as each case must be studied individually...".⁶³ He also indicated that judges are often faced with a policy conundrum in addressing vested rights and may have to rule on the basis of the "greater good".⁶⁴

[154] To rule on the basis of the "greater good" introduces another dimension to the analysis. For example, are there features of the 1986 Guidelines that better benefit the "greater good" than the features of the 1978 Guidelines? Is this achievable without imposing unfairness on any of the parties?

[155] Professor Sullivan states in her examination of vested rights the following:

The key to weighing the presumption against interference with vested rights is the degree of unfairness the interference would create in particular cases. Where the curtailment or abolition of a right seems particularly arbitrary or unfair, the courts require cogent evidence that the legislature contemplated and desired this result. Where the interference is less troubling, the presumption is easily rebutted.⁶⁵

[156] The Tribunal has, therefore, asked itself: Was the promulgation of the 1986 Guidelines unfair to Canada Post, given the 1983 date of the Complaint? Does the promulgation constitute an infringement of Canada Post's vested rights?

[157] The Tribunal considers the period of 1983 to 1986 to be a part of the continuum that constitutes the life of this case. These three initial years should not be viewed in isolation but should be seen in the context of the continuing fact-situation that existed at the time the 1986 Guidelines came into force.

[158] By 1986, although little had been accomplished amongst the parties in the investigation of the Complaint, all parties had kept one another apprised of work being done affecting the Complaint. For example, work continued by Canada Post and the Alliance in developing System One as a tool for evaluating the positions held by clerical staff at Canada Post. The Commission was informed of this work.

[159] Furthermore, Canada Post and the Alliance were actively involved during this period in the Commission's attempts to retrieve data for its job evaluation process. In fact, interviews of sample CR incumbents had commenced just prior to the 1986 Guidelines becoming effective in November of that year.

[160] The Tribunal has already established that the 1986 Guidelines are not retroactive and make no attempt to alter past law or the rights of anyone as of an earlier date. Rather, the 1986 Guidelines apply to the on-going fact-situation with immediate effect.

[161] The 1986 Guidelines had come into effect on November 18, 1986, long before the Commission referred this Complaint, on March 16, 1992, to the Canadian Human Rights Tribunal for a hearing. The Commission had played a role in the discussions amongst the parties as the Complaint moved through the Investigation Stage. Many of the matters discussed by the parties before 1986 involved issues which later became part of the 1986 Guidelines, such as occupational groups and methods of job evaluation, including assessment of value.

[162] There was, therefore, an understanding, by all concerned, of the Complaint as originally drafted. Although the 1986 Guidelines represent a significant change from the 1978 Guidelines, their introduction did little more than codify some of the Commission's procedures with which all parties had been dealing from the date of the Complaint. The wording of the Complaint, itself, exemplifies the historical nature of these procedures, as it speaks of female and male-dominated occupational groups, and the wages paid to employees within these groups. These procedures are not a part of the Act, nor were they a part of the 1978 Guidelines. They are, however, a part of the 1986 Guidelines.

[163] Real unfairness or prejudice would arise, as the Supreme Court indicated, if guidelines which were pertinent to a complaint already sent to be heard by a tribunal were promulgated after its referral to that tribunal. Even in complaints under section 11 of the Act, the Commission could, by promulgation of guidelines during the life of a tribunal, influence its outcome. That is not what happened in this case.

[164] With respect to Canada Post's example of the 1986 Guidelines' exclusion of overtime or shift work premiums from the value of work being an infringement of its vested rights, the Tribunal prefers the Commission's submission. The Commission indicated in submissions that this is an example of a neutral policy "trade-off". The complainant does not include the overtime or shift work premium in the value of wages, while the employer does not include overtime or shift work in its job point value. It is not an example of the removal of a Canada Post right of defence.

[165] In terms of the "greater good" argument, the Tribunal accepts that the Commission's promulgation of the 1986 Guidelines was an attempt to bring much needed clarification to the interpretation of section 11 of the Act, without injustice to any party. The creation of the Guidelines was completed after many years of consultation with companies and organizations, including Canada Post itself. The Tribunal accepts that the Commission's decision to create new guidelines in 1986 was for the benefit of the "greater good".

[166] Therefore, the Tribunal fails to understand how the introduction of the 1986 Guidelines after the presentation of the Complaint to the Canadian Human Rights Commission has been unfair or prejudicial to Canada Post, an infringement on its vested rights, or an improper influence upon the outcome of the Complaint before this Tribunal.

[167] Accordingly, the Tribunal concludes that the 1986 Guidelines are applicable to the issues to be addressed in the current Complaint. The question of the retroactivity of these Guidelines is not applicable to this Complaint, brought under section 11 of the Act. The facts involved are on-going, or continuing, and, as such, do not give rise to a concern about retroactivity. Additionally, the Tribunal finds that there is no infringement of Canada Post's vested rights because of the applicability of the 1986 Guidelines.

**(iii) Are subsection 8(2) and sections 11-15 of the 1986 Guidelines Valid?**

a) Submissions of the Parties

[168] All parties agreed that the Guidelines are subordinate legislation, created under the power given to the Canadian Human Rights Commission by section 27 of the Act, and as such, must not be in conflict with the Act. There is, however, a presumption that subordinate legislation is valid. When a party challenges subordinate legislation, the onus is on that party to convince the decision-maker that the subordinate legislation being challenged is invalid. The question to be answered by the decision-maker is a question of law.

[169] In this Complaint, Canada Post has challenged subsection 8(2) and sections 11 to 15 inclusive, of the 1986 Guidelines, based on its argument that a simple reading of the Act, giving straight-forward meaning to the words of the Act, and section 11 in particular, creates an inconsistency with the words and meaning in the challenged sections of the 1986 Guidelines. It is this lack of cohesion between the words and meaning of section 11 of the Act, as interpreted by Canada Post, and the words and meaning of those sections of the 1986 Guidelines, the subordinate legislation, which creates, according to Canada Post, a situation where the Commission has not exercised its power under section 27 of the Act in a reasonable manner, and thus caused those sections of the 1986 Guidelines to be invalid.

[170] The parties' submissions dealt with what should be the acceptable approach to determine validity of guidelines. Once argument was heard concerning the test for validity, further submissions were made by each party concerning its position on the issue of the validity of sections of the Guidelines impugned by Canada Post.

[171] Canada Post and the Commission both referred to the Oldman River case⁶⁶, a decision of the Supreme Court of Canada which addressed statute interpretation, specifically in the context of a situation where there were two federal statutes and a subordinate Guidelines Order involved.

[172] Quoting from the work of Professor Ruth Sullivan, Canada Post urged the Tribunal to separate the enabling legislation and its subordinate regulations (or in this case, the Guidelines) before determining the validity of the latter. It noted Sullivan's words, as follows:

Statutes are paramount over regulations... The presumption of coherence applies to regulations as well as to statutes. It is presumed that regulatory provisions are meant to work together, not only with their own enabling legislation but with other Acts and other regulations as well.⁶⁷

[173] Canada Post submitted that, when testing the validity of challenged subordinate legislation, the Guidelines, the Tribunal must first construe the enabling legislation, and then assess the validity of the impugned sections of the Guidelines. Canada Post considered that this methodology is different from that espoused by the Commission's arguments.

[174] The Commission's argument concerning statute interpretation in the face of a challenge to subordinate legislation also drew on the Oldman River case. It cited the following to underline its submissions concerning the test to be made for the validity of the impugned sections of the Guidelines:

The basic principles of law are not in doubt. Just as subordinate legislation cannot conflict with its parent legislation ... so too it cannot conflict with other Acts of Parliament, ... unless a statute so authorizes... Ordinarily, then, an Act of Parliament must prevail over inconsistent or conflicting subordinate legislation. However, as a matter of construction a court will, where possible, prefer an interpretation that permits reconciliation of the two. `Inconsistency' in this context refers to a situation where two legislative enactments cannot stand together.⁶⁸

[175] The Commission argued that, in line with the Oldman River case and with the writings of Professor Sullivan, the presumption of coherence presumes that regulatory provisions are meant to work together with their parent legislation as well as with other Acts and regulations.

[176] The Commission submission pointed to what it perceived to be Canada Post's argument that this presumption of cohesion disappears once there has been a challenge to the validity of subordinate legislation. The Commission argued that were this to be the case, there would no longer be a recognition of the importance of seeking reconciliation of differences as was underlined in the Oldman River case.

[177] The Commission argued that it is immaterial whether one takes its approach of reading the enabling legislation and the subordinate legislation together, or Canada Post's approach of first construing the enabling legislation and then addressing the subordinate legislation. The important part of the exercise is to test whether there is a consistency and a cohesion between the two levels of legislation.

[178] To begin such an interpretive exercise, section 11 and subsections 27(2), (3) and (4) of the Act and the challenged subsection 8(2) and sections 11, 12, 13, 14, and 15 of the 1986 Guidelines, read as follows:

Canadian Human Rights Act

Equal wages

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.

Assessment of value of work

11(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.

Separate establishments

11(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.

Different wages based on prescribed reasonable factors

11(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.

Idem

11(5) For greater certainty, sex does not constitute a reasonable factor justifying a difference in wages.

No reduction of wages

11(6) An employer shall not reduce wages in order to eliminate a discriminatory practice described in this section.

Definition of "wages"

11(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.

Powers, duties and functions [of the Commission]

Guidelines

27(2) The Commission may, on application or on its own initiative, by order, issue a guideline setting out the extent to which and the manner in which, in the opinion of the Commission, any provision of this Act applies in a class of cases described in the guideline.

Guideline binding

27(3) A guideline issued under subsection (2) is, until it is revoked or modified, binding on the Commission and any member or panel assigned under subsection 49(2) with respect to the resolution of a complaint under Part III regarding a case falling within the description contained in the guideline.

Publication

27(4) Each guideline issued under subsection (2) shall be published in Part II of the Canada Gazette.

Equal Wages Guidelines, 1986

Assessment of Value - Working Conditions

8(2) For the purposes of subsection 11(2) of the Act, the requirement to work overtime or to work shifts is not to be considered in assessing working conditions where a wage, in excess of the basic wage, is paid for that overtime or shift work.

Complaints by Individuals

11(1) Where a complaint alleging a difference in wages is filed by or on behalf of an individual who is a member of an identifiable occupational group, the composition of the group according to sex is a factor in determining whether the practice complained of is discriminatory on the ground of sex.

(2) In the case of a complaint by an individual, where at least two other employees of the establishment perform work of equal value, the weighted average wage paid to those employees shall be used to calculate the adjustment to the complainant's wages.

Complaints by Groups

12 Where a complaint alleging different wages is filed by or on behalf of an identifiable occupational group, the group must be predominantly of one sex and the group to which the comparison is made must be predominantly of the other sex.

13 For the purposes of section 12, an occupational group is composed predominantly of one sex where the number of members of that sex constituted, for the year immediately preceding the day on which the complaint is filed, at least

(a) 70 per cent of the occupational group, if the group has less than 100 members;

(b) 60 per cent of the occupational group, if the group has from 100 to 500 members; and

(c) 55 per cent of the occupational group, if the group has more than 500 members.

14 Where a comparison is made between the occupational group that filed a complaint alleging a difference in wages and other occupational groups, those other groups are deemed to be one group.

15(1) Where a complaint alleging a difference in wages between an occupational group and any other occupational group is filed and a direct comparison of the value of the work performed and the wages received by employees of the occupational group cannot be made, for the purposes of section 11 of the Act, the work performed and the wages received by the employees of each occupational group may be compared indirectly.

15(2) For the purposes of comparing wages received by employees of the occupational groups referred to in subsection (1), the wage curve of the other occupational group referred to in that subsection shall be used to establish the difference in wages, if any, between the employees of the occupational group on behalf of which the complaint is made and the other occupational group.

[179] Canada Post submitted that its interpretation of section 11 of the Act is purposive, broad and liberal while, at the same time, follows the modern approach to statute interpretation. The latter demands that the decision-maker read the words of the statute "in their entire context and in their grammatical and ordinary sense, harmoniously with the scheme of the Act, and the intention of Parliament".⁶⁹

[180] In subsection 11(1), the words which have elicited a marked difference of interpretation amongst the parties are "male and female employees". Canada Post submitted that this phrase, seminal to the whole of section 11, means what it says. It is talking about discrimination between employees who are either men or women, and not between groups of people made up of men and of women. The discrimination is based on sex, and manifests itself in a difference in wages paid to the men and women involved.

[181] Canada Post submitted that the focus of concern in section 11 of the Act is sex discrimination between individual men and women as seen in the difference in wages paid by employers to those men and those women for the work they do. If this interpretation is accepted, sections 11 through 15 of the Guidelines must be invalid because they do not relate to section 11 of the Act in any way. Subsection 11(1), so interpreted, does not speak of occupational groups and of the need to determine the gender percentage of those groups in order to classify them as either "male-dominated" or "female-dominated". Therefore, the concerns addressed by sections 11 through 15 of the Guidelines would not be applicable at all. Those sections would represent the promulgation of guidelines by the Commission which are not consistent with the underlying statute. Therefore, they would be invalid.

[182] Canada Post's ordinary meaning interpretation of subsection 11(1) might read as follows, according to its submissions:

It's a discriminatory practice ... something that is prohibited, for an employer to establish and maintain, that is, to exercise some creation or some power to create or continue, differences, that is higher and lower wages ... between two people, two classes of people, male employees and female employees...between employees who are men and employees who are women...employed in the same establishment ... [which] means subject to a common wage and personnel policy ... There is only a difference of wages that's prohibited if the employees, the male and female employees in the same establishment, are performing work, their work, their individual work of equal value...⁷⁰

[183] Canada Post argued that a complaint, based upon section 11 so interpreted, could be made by any individual, man or woman, or by any group of men or women, without the constraint of artificial barriers against persons who are employed in occupational groups whose work is classified as gender neutral or "male". The complaint mechanism would become more accessible to all employees. Therefore, the general purpose of the Act, to eliminate discrimination based upon, inter alia, sex, would be advanced. The restriction, argued Canada Post, which is created by the Commission's interpretation of section 11, especially in its promulgation of sections 11 through 15 of the Guidelines, would be removed.

[184] Of more import, however, according to Canada Post, is the Commission's apparent transforming of the plain language of subsection 11(1) of the Act into a completely different approach to the concept of "equal pay for work of equal value" through the use of the Guidelines. Instead of dealing with discrimination based on gender in the arena of wages, the Commission, according to Canada Post, has interpreted section 11 to be focused on discrimination based on the undervaluation of women's work in segregated occupational groups. In other words, the Commission has decided that section 11 of the Act addresses the concept of "pay equity". Once the Commission decided to deal with section 11 in that manner, it had to define the occupational groups.

[185] The basis of Canada Post's submissions concerning the interpretation of section 11 of the Act is that the section is not about "pay equity". The "work of equal value" which must be compared in order to prove the discriminatory practice being denounced by section 11 is the work of each of the men and each of the women employees involved in the complaint. The section does not address the work of occupational groups made up of men and women who are doing "women's work" or "men's work".

[186] Canada Post submitted that its interpretation of section 11 is a natural progression, historically, from previous legislation which addressed discrimination against working women. The first such legislation, early in the twentieth century, was a minimum wage for women employees. That was followed, decades later, by legislation denouncing the practice of paying women lower wages for work which was found to be either the same, or substantially similar, to work being done by men. Although this natural progression could lead, eventually, to the concept of "pay equity", Canada Post's argument is that section 11 cannot be interpreted as a movement on the continuum to that point.

[187] Canada Post's submission was that section 11 of the Act cannot be characterized as addressing the concept of "pay equity". Canada Post emphasized, in its final argument, that provincial legislation concerning "pay equity" is specific in nature. There is usually a separate provincial Act which is entitled a "Pay Equity Act". The concept is not incorporated into provincial human rights legislation because, generally, it is not complaint-driven but rather is a mandated concept carrying specific methodologies and rules for its implementation.

[188] Canada Post argued that the process of dealing with "pay equity" derives its ideas from academic studies and literature which has evolved concerning this abstract concept. The methodology is based upon job classes which are predominantly female or predominantly male because the purpose of "pay equity" studies, and eventually, "pay equity" legislation, is to address the inequities which have evolved in employment due to occupational segregation and the undervaluing of predominantly female occupations. Canada Post's argument continued, however, to stress that the Canadian Human Rights Act is concerned with the difference in wages between men and women based on gender discrimination not the broad concept of "pay equity". The Act is, according to Canada Post, concerned about protecting individuals from disadvantage or discrimination resulting from fundamental individual characteristics.

[189] Canada Post's submission was that the Act has been promulgated for the benefit of individuals in Canadian society. The purpose of the Act, as set out in section 2, underlines that this human rights legislation was created so that "...all individuals should have an opportunity equal with other individuals to make for themselves the lives that they are able and wish to have...". The grounds for discrimination which are listed in the Act are characteristics of individuals. Section 3 of the Act specifically indicates that the list of grounds is "[f]or all purposes of this Act". As one reads further in the Act, each section follows the next to speak of "individuals" or "employees".

[190] This commonality is broken by section 11 of the Act which does not speak of an individual employee or a group of employees, but addresses discrimination involving "male and female employees". It is this change in language which alerts us, argued Canada Post, to the fact that this particular section addresses systemic discrimination, for it indicates Parliament's concern about whether there is systemic discrimination, whether there is a pattern within the overall establishment of setting differences in wages between men and women performing work of equal value.

[191] Canada Post underlined in its submissions that it accepts that section 11 addresses systemic discrimination, but argued that this acceptance does not mean that it accepts that complaints of systemic discrimination can only be brought by groups. Indeed, it submitted that individual complaints can be made based on allegations of systemic discrimination

[192] As counsel for Canada Post submitted:

...if it's systemic discrimination, and if it's men and women employed in the establishment, you take the wages of the women employed in the establishment and you take the wages of the men employed in the establishment, and you compare the work of men and women performing work of equal value and their wages. If there is a pattern of wage differences, then there is a violation.⁷¹

[193] This comparison cannot, according to Canada Post, be equated with the evaluation process which is either mandated or is commonly followed when one is dealing with "pay equity" issues. That process most often involves the comparison of the value of work of groups of persons who are doing either "women's work" or "men's work".

[194] The Commission's submission was that there is binding precedent from Evans, J., then of the Federal Court (Trial Division), in the Treasury Board case⁷², that the Commission's interpretation that section 11 does address the issue of "pay equity", as evidenced by the promulgation of the 1986 Guidelines, is correct. In answer, Canada Post argued that the acceptance by Evans, J. of the Commission's interpretation (and the use of the 1986 Guidelines) was based on the fact that all parties involved in that case accepted that interpretation. As there was no challenge to the Commission's interpretation of section 11, the presumption that the Guidelines were valid was never challenged. Canada Post noted that any comment made by Evans, J. must be accepted by this Tribunal as merely that - comment which can be useful to the Tribunal as it crafts its decision concerning the issue, but not binding on the Tribunal as precedent.

[195] Additionally, Evans, J. made extensive comment on the viability of Guideline 14, which addresses occupational groups, in the context of section 11 of the Act. The interpretation of that particular guideline was the main issue to be decided by the tribunal which heard the Treasury Board case and by the Federal Court which reviewed its decision. The comments of Hugessen, J. in the Department of National Defence case⁷³ were alluded to by Evans, J. and, therefore, should be, according to the Commission arguments, of import to this Tribunal's interpretation of section 11 of the Act.

[196] The Commission submitted that the Supreme Court of Canada, in its June 2003 decision in Bell Canada, found that the Commission has the power, conferred upon it by section 27 of the Act, to create guidelines which are analogous to Regulations. That decision, argued the Commission, has created the presumption that the Guidelines are valid.

[197] As is a challenge to promulgated Regulations, a challenge to the validity of the Guidelines is difficult. Courts prefer to accept that the subordinate legislation can be reconciled with its enabling legislation. If it were to accept Canada Post's interpretation, the Commission argued, the Tribunal would have to find that there is an operational conflict between the Act and the Guidelines, and that there is no ability to reconcile the conflict.

[198] The Commission argued that Canada Post's choice to begin its "ordinary meaning" argument with the words "male and female employees" avoids the true meaning of section 11. The Commission based its submissions on the historical evolution of the concept of "equal pay for work of equal value", as well as comments made by the Courts. After so doing, it submitted that the Tribunal should accept the Commission's interpretation of section 11 of the Act as Parliament's enunciation of the principle of "pay equity". The Commission further submitted that Parliament addressed the difficulty of dealing with the abstract concept of "pay equity" by giving the Commission the tools to make that principle operational. Those tools include the Commission's ability to promulgate guidelines, pursuant to section 27 of the Act.

[199] Therefore, the Commission promulgated its guidelines to accord with the purpose of "pay equity" legislation. The Commission submitted that the purpose of such legislation is to ameliorate the occupational segregation of women and the discriminatory payment of lesser wages to those segregated groups for work which is equal in value to work done by groups mainly composed of men. The main focus of section 11 of the Act, according to the Commission, should be the work which is being done, and its value, not the gender of the incumbents who are doing the work.

[200] Based on this broad and liberal interpretation, argued the Commission, sections 12 through 15 of the 1986 Guidelines represent the methodology which must be used to make the principle enunciated in section 11 of the Act a workable theory. Those sections of the Guidelines are absolutely connected to section 11 of the Act, and make the two areas of the legislation work in tandem, to create a cohesive whole which is the basis for the evaluation work which must be completed to establish whether a complaint can be substantiated.

[201] The Commission argued that, notwithstanding Canada Post's adamant submissions that its interpretation of section 11 of the Act is broad and purposive, and is actually more liberal in its ability to encompass any complainant(s), the reality and the natural conclusions which the Canada Post interpretation would create are narrow and restrictive. Its interpretation does not address the broad concept of "pay equity" which is what the legislation was intended to address from its beginnings.

[202] According to the Commission's submissions, Canada Post's interpretation would restrict the evaluation process to a singular methodology. Only the job-to-job approach could be used. Only an examination of the whole "system" could be made in the evaluation process, even if the complainant were a single individual. Although there was a concession by Canada Post that representative sampling could be done at the evaluation stage, the Commission argued that Canada Post's interpretation of section 11 of the Act would create a cumbersome methodology which would, in fact, be regressive in nature.

b) Tribunal's Analysis

[203] All parties have quoted from Sullivan's Driedger on the Construction of Statutes, in its many iterations. The Tribunal agrees that this work is seminal when one is dealing with statute interpretation. Of note are the commentaries on what Driedger styled "the modern rule" of interpretation, as follows:

Today there is only one principle or approach, namely, the words of an Act are to be read in their entire context, and in their grammatical and ordinary sense harmoniously with the scheme of the Act, the object of the Act, and the intention of Parliament.⁷⁴

There is only one rule in modern interpretation, namely, courts are obliged to determine the meaning of the legislation in its total context, having regard to the purpose of the legislation, the consequences of proposed interpretations, the presumptions and special rules of interpretation, as well as admissible external aids. In other words, the courts must consider and take into account all relevant and admissible indicators of legislative meaning. After taking these into account, the court will then adopt an interpretation that is appropriate. An appropriate interpretation is one that can be justified in terms of (a) its plausibility, that is, its compliance with the legislative text; (b) its efficacy, that is, its promotion of the legislative purpose; and (c) its acceptability, that is, the outcome is reasonable and just.⁷⁵

[204] Additionally, the Interpretation Act must be considered. It indicates that:

Every enactment is deemed remedial, and shall be given such fair, large and liberal construction and interpretation as best ensures the attainment of its objects.⁷⁶

[205] Based on these interpretive principles, the Tribunal finds that the Commission's submissions concerning the interpretation of section 11 of the Act represent the more appropriate approach to this section. The Tribunal has read the section within the context of the Act. Additionally, it has considered the interpretations of section 11 which have been presented by tribunals and Courts in the past. The Tribunal has before it expert evidence which addressed the historical evolution of the concept of "pay equity". This evidence, combined with comments made by Member of Parliament, the Honourable Ron Basford, during the discussions preceding the promulgation of the Act, reinforces, in the Tribunal's view, the finding that section 11 of the Act is intended to address the issue of "pay equity".

[206] The Tribunal accepts that this interpretation is compatible with the purpose of the legislation, its context, and its legislative history. The purpose of the Act is set out in section 2, as follows:

The purpose of this Act is to extend the laws in Canada to give effect, within the purview of matters coming within the legislative authority of Parliament, to the principle that all individuals should have an opportunity equal with other individuals to make for themselves the lives that they are able and wish to have and to have their needs accommodated, consistent with their duties and obligations as members of society, without being hindered in or prevented from doing so by discriminatory practices based on race, national or ethnic origin, colour, religion, age, sex, sexual orientation, marital status, family status, disability or conviction for an offence for which a pardon has been granted.

[207] Canada Post's submission that this purpose underlines the Act's scope as legislation which targets discrimination solely against individuals on specific grounds cannot be accepted. It is the Tribunal's view that this is a restrictive interpretation of legislation which is clearly meant to address the broad issue of discrimination against all persons. Acts of Parliament must be interpreted using large and liberal construction which will result in fairness to all. The elimination of groups made up of both male and female persons from the protection of the Act would result in a narrowing of the purpose of the Act. The Tribunal rejects this interpretation, and accepts that section 11 of the Act, in addressing discrimination in the area of "pay equity", conforms with the general purpose of the Act.

[208] One of Canada Post's submissions was based upon what it characterized as an historical continuum of legislation. This continuum included other work-related legislation which has addressed the very real problem of a difference in wages paid to female and male workers, such as the Canada Labour Code. As noted already, successive governments have attempted, from the turn of the nineteenth/twentieth centuries, to deal with the problem of differences in wages paid to male and female workers. From Canada Post's perspective, the passage of section 11 in the Act represents stage 4 in a continuum which began with minimum wage legislation, and moved on to "equal pay for the same work" done by males and females, slightly modified to become "equal pay for substantially similar work" and, according to their argument, would naturally evolve to become "equal pay for work of equal value" done by male and female workers. Eventually, in Canada Post's submission, "pay equity", a concept further along on the continuum, and somewhat distant from the first four concepts in its methodology and its focus, might become an issue to be addressed as stage 5. Its main argument, however, was that the concept of "pay equity" is not currently a part of the Act.

[209] Canada Post conceded that the focus on occupational groups, deemed predominantly female or predominantly male, which can be compared using various methodologies involving such statistical means as regression analysis, is a legitimate characteristic of a "pay equity" study. Its submission was, however, that section 11 of the Act is not about "pay equity".

[210] "Pay equity" legislation is, according to Canada Post, something entirely dissimilar from a denunciation of a difference in wages between men and women for work of equal value. While Canada Post agrees that "pay equity" has its focus on the problem of the occupational segregation of women, and the related problem of the undervaluation of women's work, it believes this interpretation represents a leap in conceptual thinking from what it argues are the clear words of section 11. As such, Canada Post implies that "pay equity" cannot be what Parliament meant to address when it created section 11 of the Act.

[211] The Tribunal rejects this argument that "pay equity", as a concept, is beyond the scope of the Act. The concept has already been accepted as the interpretive basis for section 11 of the Act. In one of the first cases to discuss section 11, Public Service Alliance of Canada v. Canada (Department of National Defence), the Federal Court of Appeal noted that:

[t]he case concerns pay equity...the appellant, as bargaining agent for the employees concerned, alleged that the respondent employer was not paying certain female employees wages equal to those paid to certain male employees performing work of equal value ... in contravention of Sections 7 and 11 of the Canadian Human Rights Act.⁷⁷

[212] Thus, from the very introduction to the judgement in that case, Hugessen, J. characterized section 11 of the Act as a section which had been created, specifically, to address the problem of "pay equity".

[213] He quoted at length from Dickson, C.J. who "[i]n the seminal case of Action Travail des Femmes v. Canadian National Railway Co." spoke for the Supreme Court of Canada.⁷⁸ In that case, Dickson, C.J. quoted the Abella Report, to conclude that:

...systemic discrimination in an employment context is discrimination that results from the simple operation of established procedures of recruitment, hiring and promotion, none of which is necessarily designed to promote discrimination. The discrimination is then reinforced by the very exclusion of the disadvantaged group because the exclusion fosters the belief, both within and outside the group, that the exclusion is the result of `natural' forces, for example, that women `just can't do the job' ... It is compounded by the attitudes of managers and co-workers who accept stereotyped visions of the skills and `proper role' of the affected group, visions which lead to the firmly held conviction that members of that group are incapable of doing a particular job, even when that conclusion is objectively false.⁷⁹ (emphasis added)

[214] Hugessen, J. then quoted from the Human Rights Tribunal decision in Public Service Alliance of Canada v. Treasury Board which described systemic discrimination as

[emphasizing] the most subtle forms of discrimination ... [and recognizing] that long-standing social and cultural mores carry within them value assumptions that contribute to discrimination in ways that are substantially or entirely hidden and unconscious. Thus, the historical experience which has tended to undervalue the work of women may be perpetuated through assumptions that certain types of work historically performed by women are inherently less valuable than certain types of work historically performed by men.⁸⁰

[215] Clearly, the language of the Treasury Board tribunal is the language of "pay equity".

[216] From this position, Hugessen, J. indicated at paragraph 15 that "[I]t is arguable, indeed, that the type of discrimination which pay equity is designed to counteract is always systemic". He went on to quote from Nan Weiner and Morley Gunderson, Pay Equity Issues: Options and Experiences (Toronto: Butterworths, 1990) as follows

...pay equity is designed to address a kind of systemic discrimination. Systemic discrimination is found in employment systems. It is the unintended byproduct of seemingly neutral policies and practices. However, these policies and practices may well result in an adverse or disparate impact on one group vis-a-vis another (e.g., on women versus men) ... Pay equity requires changes to pay systems to ensure that women's jobs are not undervalued. (emphasis added)

[217] Again, Hugessen, J. underlined his decision that section 11 of the Act is dealing with the concept of "pay equity". He clearly indicated, by quoting from Weiner and Gunderson, that this concept deals with women's jobs which, historically, have been undervalued and must be addressed to change that systemic discrimination. It is the jobs which are of primary importance, not the gender of the incumbents.

[218] This point, that the basis for equal value legislation in Canada was the perceived need to address occupational segregation and the undervaluing of women's work, was emphasized to the Tribunal in this Complaint by Professor Pat Armstrong who was accepted as an expert in women's work, women's wages, and the sociological aspects of equal pay legislation. She stressed that, historically in Canada, there has been a segregation of jobs into female and male dominated areas.

[219] Dr. Armstrong noted that one of the federal government's responses to the Royal Commission on the Status of Women (1970) was the promulgation of the Canadian Human Rights Act, 1978, including section 11 of the Act to address systemic discrimination in wages, due to job segregation. This historical background, therefore, must be taken into consideration when interpreting section 11 of the Act. The purpose of the section must include the need to address the undervaluation of women's work, as seen in the segregation of that work into occupational groups dominated by women.

[220] Section 11 of the Act, she indicated, is about "pay equity" and, as such, must deal with male and female-dominated job classes in order to address the occupational segregation which that concept targets. In addition to addressing jobs and gender, "pay equity" must discuss these issues through the prism of occupational segregation.

[221] This expert evidence echoes the statement of the Honourable Ron Basford, Minister of Justice during Parliamentary debates in 1977 which preceded passage of the Act. As noted in paragraphs [53] and [54] of this Decision, the Honourable Ron Basford anticipated problems with the concept of equal pay for work of equal value as presented in section 11 of the Act. He noted, however, that the government's approach was to legislate the principle, and give to the Commission the task of solving any problems involved in the implementation of that principle. He went on to indicate that the underlying problem was occupational segregation of women, and their historical lower wages, caused by an undervaluation of women's work.

[222] Given this background, as well as expert evidence, the Tribunal accepts that section 11 of the Act addresses the concept of "pay equity" which translates into "equal pay for work of equal value" between male and female workers.

[223] The principle of "pay equity" between men's work and women's work which has equal value demands a methodology which has evolved as the concept has evolved. The methodology to be used to address the concept is not part of section 11. As the Supreme Court underlined in its June 2003 decision, the task of fleshing out the operation of section 11 of the Act has been given to the Commission. The promulgation of the 1986 Guidelines is the direct result of the task mandated to that body.

[224] Each section of the Guidelines which Canada Post has challenged, based on its interpretation of the Act, addresses the concept of "pay equity". Indeed, Canada Post conceded in its submissions concerning the interpretation of section 11 of the Act that the Guidelines would be coherent and logical if one were dealing with "pay equity" in section 11. Their position was that section 11 does not make the conceptual leap to "pay equity" but rather must be interpreted using a straight-forward "simple meaning" approach.

[225] The Tribunal finds that "pay equity" is the concept which section 11 was created to address. The words of the text allow for the plausibility of this interpretation, as commentators have used the terms "equal pay for work of equal value", "comparable worth", and "pay equity" almost interchangeably. This interpretation of section 11 is efficacious, as it promotes the legislative purpose as enunciated by the Minister of Justice immediately prior to the promulgation of the Act, as well as the intended purpose of the Act, section 2, read in a broad and liberal manner. Given the finding that this interpretation is plausible and efficacious, it is accepted as a reasonable and just interpretation which addresses the purpose of the Act both specifically in the section itself, and within the context of the whole philosophy of the Act.

[226] Accordingly, the Tribunal concludes that "pay equity" has been accepted as the interpretative basis for section 11 of the Act, which addresses the undervaluation of work performed by women in occupational groups dominated by women. Examination of male and female job classes, therefore, becomes an important aspect of any "pay equity" study and the Commission's 1986 Guidelines, particularly those sections challenged by Canada Post, provide assistance in making that possible.

[227] The test to be applied by the Tribunal in determining the validity of the particular sections of the Guidelines impugned by Canada Post is whether or not they are consistent with the meaning of section 11 of the Act. Canada Post has argued that they are inconsistent with section 11.

[228] The Tribunal reiterates that a proper interpretation of section 11 recognizes that the section was created to address the concept of "pay equity", as described above. The Commission was entrusted, pursuant to the Act, to implement the concept and was required to make it operational by means of promulgating certain guidelines.

[229] The Tribunal, therefore, concludes that the impugned sections 11 through 15 of the 1986 Guidelines, in providing guidance to interpret the "pay equity"-based section 11 of the Act, are vital to that interpretation. They also provide a cohesion and a wholeness to the legislation and are consistent with the meaning of section 11 of the Act and are, accordingly, valid and operable.

[230] Additionally, subsection 8(2) of the 1986 Guidelines addresses specifically the methodology to be used when dealing with a particular aspect of the working conditions factor set forth in subsection 11(2) of the Act. As such, it, too, is necessary to the fleshing out of the principles of the Act, and is consistent with the meaning of section 11.

[231] In reaching these conclusions, the Tribunal has addressed the submissions of both Canada Post and the Commission concerning how best to test for the validity of the subordinate legislation. In effect, the Tribunal has endorsed the approach that one first interprets the enabling legislation, and then, based on that interpretation, determines whether the impugned subordinate legislation is consistent with its enabling statute.

[232] Finally, the Tribunal considers it relevant to refer to Mr. Justice Evans' decision in the Treasury Board case, in which he indicated the following concerning the validity of an impugned guideline:

In view of the breadth of the statutory language of subsection 27(2), and of the attributes of the body to which the discretion has been conferred, a provision in any guidelines issues will only be held to be invalid if it is clearly incompatible with the terms of the grant of statutory power, when construed in light of the purposes of the Act...⁸¹

[233] Using the language of Mr. Justice Evans, the Tribunal finds that subsection 8(2) and sections 11 through 15 of the 1986 Guidelines, challenged by Canada Post, are not incompatible with their enabling legislation when construed in light of the purposes of the Act. Indeed, the impugned Guidelines are necessary to the smooth operation of the Act and are found to be valid.

C. Proof by Presumption

[234] The question to be addressed is whether or not the proof by presumption referred to by Evans, J. in the Treasury Board decision, is a rebuttable presumption. All parties in this Complaint have agreed that a presumption, by its very nature of being a presumption, can be rebutted. The real question is what constitutes an acceptable rebuttal under the circumstances of this Complaint? Can this presumption, for example, be rebutted by "reasonable factors" other than those identified in the Guidelines?

[235] Evans, J. noted that wage differences between men and women performing work of equal value that are attributable to prescribed "reasonable factors" other than sex, are exempt from the reach of section 11 of the Act. He stated that:

Accordingly, once a complainant has established a difference in the wages paid to male and female employees performing work of equal value, a breach of section 11 is thereby established, subject only to the employer's demonstrating that the difference is attributable to one of the `reasonable factors' prescribed in Section 16 of the Guidelines.⁸²

[236] Evans, J. concluded, at paragraph 152, that:

Subsection 11(1) can thus be seen to have tackled the problem of proof by enacting a presumption that, when men and women are paid different wages for work of equal value that difference is based on sex, unless it can be attributed to a factor identified by the Commission in a guideline as constituting a reasonable justification for it.

[237] In addition, Evans, J. stated that:

...the nature of systemic discrimination often makes it difficult to prove that the disadvantaged position in the workplace of many members of particular groups is based on the attributes associated with the groups to which they belong. This is because, as Dickson, C.J. observed ... systemic discrimination `results from the simple operation of established procedures of recruitment, hiring and promotion, none of which is necessarily designed to promote discrimination'. Accordingly, an employer's wage policies and practices may be based on such deep-rooted social norms and assumptions about the value of the work performed by women that it would be extremely difficult to establish in a forensic setting that, if women were paid less than men performing work of equal value, that difference was based on sex.⁸³

[238] The Commission and the Alliance argued in their submissions that the only way to rebut the aforementioned proof by presumption is to rely on a "reasonable factor" identified in the Guidelines.

[239] Canada Post argued that, while relying on a "reasonable factor" in the Guidelines is certainly one way to rebut the presumption, that option is not the only one.

[240] Canada Post elaborated that it would be inconsistent with the purpose of the Act to allow only the Commission to displace the presumption of subsection 11(1) by means of its specified "reasonable factors". It argued that the purpose of the Act is to address discrimination on various grounds and section 11 does not include all of the grounds. It concerns discrimination based only on sex.

[241] Accordingly, Canada Post added, the respondent or employer should be able to lead evidence to show that the reason for a wage difference, while not being a "reasonable factor", may be due to some cause other than sex. In other words, the list of "reasonable factors" cannot be close-ended but rather must be open-ended, thus providing an additional line of defence to rebut the presumption.

[242] While the evidence must be persuasive and the burden of proof lies clearly with the respondent or employer, in Canada Post's view, it should have the opportunity to rebut the presumption by leading such evidence.

[243] Essentially, Canada Post argued that if the employer or respondent put evidence before the Tribunal which showed, on a balance of probabilities, that the wage gap was not the result of sex discrimination, then that would constitute a rebuttal of the presumption. It is Canada Post's position that Evans, J. has not foreclosed this argument.

[244] The Commission argued that Evans, J. was very clear about what constitutes a rebuttable presumption under subsection 11(1), namely, that only evidence of the presence of "reasonable factors" described in section 16 of the 1986 Guidelines can rebut the presumption that, once a difference in wages between male and female employees performing work of equal value is established, on a balance of probabilities, discrimination based on the ground of sex is also established.

[245] Moreover, the Commission cited paragraph 48 of the Supreme Court of Canada decision, dated June 26, 2003, in Bell Canada (Supra note 39). This citation refers to the role of the Commission when it issues guidelines specifically dealing with the "reasonable factors" noted in subsections 11(4) and 27(2) of the Act to justify gender wage differences, as follows:

This provision clearly contemplates guidelines adding precision to the Act, without in any way trumping or overriding the Act itself.

[246] In the Commission's submission, an open-ended list of "reasonable factors" would not serve the purpose of adding precision to the Act. Nor would it serve the principle of narrowly construing defences in human rights cases generally.

[247] Finally, it is helpful to consider the testimony of expert witness for the Commission, Professor Pat Armstrong, concerning systemic discrimination as a concept. The witness was responding to a question from Canada Post's counsel, in cross-examination, relating to the Province of Ontario's human rights legislation:

Systemic discrimination is presumed ... systemic discrimination refers to discrimination that arises from a variety of factors, not a single factor, like a single employer behaving inappropriately. Equal pay for work of equal value is based on a certain kind of discrimination. That is what is recognized as systemic discrimination, which is why guilt is not the issue, or, as Morley Gunderson says in his work for the Abella Commission, why it is not a question of even looking for root causes. It is an issue of trying to make pay more equal between male - and female - dominated work. So, it is not a question of discrimination in the general sense, but in the specific sense of systemic discrimination.⁸⁴

[248] The Tribunal accepts that section 11 of the Act is addressing, primarily, a particular discriminatory practice commonly known as systemic discrimination. This type of discrimination has often arisen, historically, from recruiting and hiring policies and practices that have inherently, but not necessarily intentionally, resulted in female employees being paid less than male employees for work of comparable value. The concept of "equal pay for work of equal value" is, therefore, an attempt to address systemic discrimination by measuring the value of work performed by men and women.

[249] The Tribunal notes that Evans, J. has ruled in his decision of October 19, 1999 in the Treasury Board case, that subsection 11(1) effectively enacts a presumption that:

...when men and women are paid different wages for work of equal value that difference is based on sex, unless it can be attributed to a factor identified by the Commission in a guideline as constituting a reasonable justification for it.⁸⁵

[250] The Tribunal also notes that, while all parties have agreed that a presumption, by definition, is rebuttable, there is not unanimity on what constitutes an acceptable rebuttal under the circumstances of the Complaint.

[251] Evans, J. clearly states that the presumption under subsection 11(1) can be rebutted by "reasonable factors" established by the Commission under subsections 11(4) and 27(2) of the Act. On the other hand, Canada Post has argued that rebuttal should not be limited to the "reasonable factors" included in the Guidelines, but should be "open-ended".

[252] The Tribunal notes that the aforementioned Supreme Court of Canada decision supports the view that the legislative intent was to add precision to the Act in terms of the guideline-making power which, in the Tribunal's opinion, is compatible with taking a "close-ended" approach to the establishment of "reasonable factors". Moreover, a close-ended list of "reasonable factors" would, in the Tribunal's view, also be compatible with the principle of narrowly construing defences in human rights cases.

[253] Accordingly, the Tribunal concludes that the presumption enacted by subsection 11 (1) of the Act, while being a rebuttable presumption, is one that can be rebutted only by "reasonable factors" identified, from time to time, by the Commission, pursuant to subsections 11(4) and 27(2) of the Act.

**V. PRIMA FACIE CASE**

A. Background and Elements of a Prima Facie Case for a Complaint brought under Section 11 of the Act

[254] Because of the systemic nature of the discrimination alleged in the Complaint before the Tribunal, the Complaint is addressed using the current Act, as amended in 1998. This is evident from the discussion of retroactivity and the validity of the 1986 Guidelines noted in Section IV, B of this Decision. Therefore, the Tribunal must look at each element of section 11, as it currently reads. Each element in section 11 of the Act must be substantiated, on a balance of probabilities, in order to substantiate the Complaint.

[255] Section 11 proscribes sexual discrimination in the determination of wages. Subsection 11(1) provides that 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. Subsection 11(2) provides the criterion for assessing the value of the work being compared. The value assigned to the work must be based on the composite of the skill, effort, responsibility and working conditions involved in performing the work. Additionally, subsection 11(4) allows an exemption to employers from a finding of discrimination because of special circumstances which are described as "reasonable factors".

[256] As noted in Section IV, C of this Decision, section 11 contains a built-in presumption of discrimination based upon sex, one of the prohibited discriminatory factors noted in the Act, when a difference in wages has been found to exist between male and female employees, employed in the same establishment, performing work of equal value. This presumption is subject to the constraint of "reasonable factors", presented in subsection 11(4) and expanded in definition by the Guidelines.

[257] As noted in Section IV, B of this Decision, the 1986 Guidelines are necessary to any discussion of section 11 of the Act, as they illuminate the principle of "pay equity" which is the basis for the section. Therefore, when addressing section 11 in the context of the Complaint before this Tribunal, each of the following elements must be proven, on a balance of probabilities. The elements are taken from section 11 of the Act and from the guidance which is offered concerning the particularizing of the section through guidelines promulgated by the Commission pursuant to its mandate under section 27 of the Act.

(1) The complainant occupational group is predominantly of one sex and the comparator occupational group is predominantly of the other sex. In this Complaint, that means the complainant CR's must be predominantly female and the comparator PO's must be predominantly male.

(2) The female-dominated occupational group and the male-dominated occupational group being compared are composed of employees who are 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 than the predominantly male occupational group. This wage difference is commonly called a "wage gap".

B. Does the Complainant group and the Comparator group represent, respectively, a predominantly female occupational group and a predominantly male occupational group, suitable for comparison of work, under the Act?

[258] The history of the groups known as CR's and PO's begins when the Treasury Board classification system was created for federal government departments in the 1960's. That system is still generally maintained in the federal sphere to date. The Post Office Department within the federal government was the precursor to the Crown Corporation, Canada Post. When, by federal statute, the Crown Corporation, Canada Post, was established in 1981, the federal government classification standards and the wage scales attributed to the government classification levels were maintained for those government employees who became employees of Canada Post. This was accomplished pursuant to the transition rules in the statute which created Canada Post as a Crown Corporation.

[259] This Complaint was presented to the Commission by the CR occupational group, employed by the newly-created Canada Post. The CR group identified itself in the Complaint as "female-dominated". The group was made up of workers who had been classified as "Clerical and Regulatory" when they were employed in the Post Office Department. This Treasury Board classification was used for all Clerical and Regulatory workers employed throughout the federal government. When the CR's in the Complaint became part of the Crown Corporation, Canada Post, their CR classification was maintained. There was, however, an undertaking between the Alliance, the union representing the CR's and certain other occupational groups, and Canada Post that negotiations to re-evaluate the CR and other positions would eventually take place. This undertaking was the basis for the work which Canada Post management and the Alliance engaged in when they attempted to create the "System One" evaluation scheme.

[260] The complainant group chose, as its comparator for the Complaint, the "male-dominated" Postal Operations group, the PO's. The PO's had been, like the CR's, employees of the federal government when the Post Office was a government department and had retained their status as PO's when they became employees of the newly-created Crown Corporation. These employees, represented by the Canadian Union of Postal Workers (CUPW), had originally been members of a unique federal group called "mail handlers".

[261] As noted in paragraphs [25] and [26], above, on January 24, 1992, the date of the Commission's Final Investigation Report of this Complaint, with its recommendation that the Complaint be referred to the Tribunal, there were 2,310 CR positions, separated into levels of CR-2 (260 positions), CR-3 (950 positions), CR-4 (950 positions), and CR-5 (150 positions). There were 43,099 PO positions, separated into PO-INT positions (with four levels), PO-EXT positions (with three levels) and PO-SUP positions (with 6 levels). Although the actual effective date of these numbers has not been identified in the Commission's Final Report, it is assumed they represent the populations of the two groups as of the date of the Final Report, or close to that date.

[262] In the year the Complaint was filed, 1983, the number of CR positions was virtually the same (2,316) as in early 1992 although the number of PO positions was larger by almost 8,000 positions (50,912).

[263] Section 11 of the Act addresses work and wages in the context of "pay equity". Historically, "pay equity" has attempted to address the gender-based segregation of work, and the wages which flow from this segregation. Traditionally, the wages paid for work generally performed by women have been less than those paid for work generally performed by men. Because section 11 does not provide a definition for what constitutes a predominantly gender-based occupational group, the Tribunal must seek clarification from sections 12 and 13 of the 1986 Guidelines. The Commission used its powers under section 27 of the Act to produce this practical guideline for group complaints. Sections 12 and 13 of the 1986 Guidelines permit a comparison between "occupational groups" as long as those groupings represent work being done predominantly by males and predominantly by females.

[264] Section 13 of the 1986 Guidelines identifies several formulas for determining when an occupational group is considered to be predominantly of one sex. For example, an occupational group numbering more than 500 is deemed to be composed predominantly of one sex if at least 55% of its members are of that sex. In the Complaint before the Tribunal, each group, as a whole, was made up of more than 500 members both at the time the Complaint was filed, and at the time it was referred to the Tribunal.

[265] The complainant group had indicated to the Commission, and expressed the belief in the wording of the Complaint itself, that it was a female-dominated group. The group chosen as a comparator was presented by the complainant as a male-dominated group. In 1983, over 80% of the CR group was comprised of female employees and just over 75% of the PO group was comprised of male employees.⁸⁶ At the time of referral of the Complaint to the Tribunal in 1992, the CR group remained predominantly female, with a percentage factor of over 83% female, and the PO group (which was now made up of only the PO-INT and the PO-EXT subgroups, the PO-SUP subgroup having been removed by the Commission during its investigation) remained predominantly male, with a percentage factor of just above 71% male.⁸⁷

[266] The Alliance and the Commission argue that these percentages are sufficient to classify the complainant group as being comprised of employees predominantly of the female sex, and the comparator group as being comprised of employees predominantly of the male sex.

[267] The submissions of both the Alliance and the Commission concerning the gender predominance of the two groups are based on their interpretation of the 1986 Guidelines. Additionally, they argue that the Guidelines are, unless found to be ultra vires (which is not the case in this Complaint), binding on the Tribunal, according to the Act.

[268] According to the 1986 Guidelines, argue the Alliance and the Commission, when an occupational group exceeds 500 in number, only 55% of the persons in the group need be of one gender to deem that group to be doing work of persons of that gender. Therefore, in this Complaint, the groups are deemed to be doing either work generally performed by women (the CR's) or work generally performed by men (the PO's) based on the fact that they are groups larger than 500 in total, and the percentage of either female or male members of each group makes its work representative of either female or male work. The argument is that this is a simple arithmetical computation which, once made, is one factor in choosing a complainant and comparator. It is the factor, however, which satisfies the element of section 11 (clarified by the Guidelines) which demands that, when one is dealing with a group complaint, the complainant be a predominantly female group and the comparator, a predominantly male group.

[269] Canada Post argues that the percentages are illusory. Its submission is that the Postal Operations group cannot be viewed as a melded group. The PO group is, and traditionally has been, according to Canada Post, a group which aspires to the principle of "straight-line" wage rates. Canada Post's argument stresses that, during the history of the Complaint, the PO-4 level has always been the largest single element of the subgroup, PO-INT. It is the PO-4 level which is, according to Canada Post, most representative of the PO occupational group as a whole, and the classification category where the most PO jobs are found. Indeed, Canada Post argues that the PO-4 level of the Postal Operations group has never been anything but essentially neutral in its gender make-up and should be more properly regarded as representative of the entire PO group. In 1983, 53% of employees classified at the PO-4 level were male and 47%, female. In 1992, the figures were 50.6% male and 49.4% female. If the Postal Operations group were defined in the manner of the PO-4 level, Canada Post submits that, as the comparator, it would not fit within the definition of a "predominantly male" comparator group pursuant to the Guidelines.

[270] Canada Post's argument is that to take the Postal Operations group as a whole is to ignore the historical trend by which the number of PO-4 level employees is becoming increasingly the most critical and representative category of Postal Operations workers. In fact, employees classified at the PO-4 level within the Internal Mail Processing and Complementary Postal Service Subgroup represented just over 83% of its Subgroup total in 1983, and 88% in 1992. On the other hand, as a percentage of the entire Postal Operations group, PO-4 level employees represented 41% in 1983 and almost 42% in 1992.

[271] The Tribunal does not accept this argument. The federal government job classification scheme is predicated upon the concept of groups of employees, bound together by occupational job categories. Within these groupings, the concept of levels is connected to wage differentials. Historically, these levels, with their wage differentials, were based on factors such as seniority, management's view of the importance of the work performed at each level, and the requisite training and skills necessary. That a union at Canada Post, representing many or all of the Postal Operations group may have decided to attempt to create a situation where the classification levels are essentially unrelated to wage differentials cannot change the historical concept that is the basis for the groups and levels themselves. It is this concept that is important to the designation of "occupational group" in sections 12 and 13 of the 1986 Guidelines, and to the issue of "pay equity" in section 11 of the Act.

[272] Therefore, the Tribunal accepts that the complainant occupational group, the CR's, and the comparator group, the PO's, are representative, respectively, of a female-dominated group and a male-dominated group because each is over 500 in number, and because each contains at least 55% of female employees (the complainant CR's) and male employees (the comparator PO's). This conclusion is based upon the 1986 Guidelines which indicate the importance of the size of each group, and the necessary percentage of either males or females in each occupational group of a specified size which will deem the group to be either male-dominant or female-dominant.

[273] The Tribunal is bound by the Act to follow the Guidelines which address the specifics of the Complaint before it, a "pay equity" complaint under section 11 of the Act, dealing with occupational groups.

[274] Canada Post submits that, even if the groups are gender appropriate, the Alliance's choice of the Postal Operations group as its comparator was made because of that group's position, at the time, as being highly paid. Such a choice, in the "pay equity" context, would, in Canada Post's submission, be "cherry picking" and, therefore, not appropriate.

[275] Mr. Norman Willis, a witness for Canada Post who was accepted by the Tribunal as an expert in pay equity and in job evaluation, was one of a number of witnesses who explained the concept of "cherry picking".

[276] He explained that, in a "pay equity" group complaint, the complainant group chooses its comparator group. "Cherry picking" in "pay equity" situations envisions a scenario where the complainant group chooses a comparator group which, while often small in members, represents the most highly paid of a number of available comparator groups. Although wages, understandably, is one natural aspect of the choice, as the "pay equity" complaint always involves an allegation of payment of less wages to the complainant when compared with the chosen comparator, choosing a group based solely on its characteristic of having high wages compared with the complainant group is not acceptable as a starting point for a legitimate "pay equity" comparison. It would skew the results of evaluation and comparison, in favour of the complainant. Allowing a "cherry picked" comparator would create upheaval within an establishment, as subsequent comparisons would be inevitable between the original complainant and other workers.

[277] During his explanation of "cherry picking", Mr. Willis expressed the opinion that the Complaint before the Tribunal was tainted from the beginning because of the complainant's "cherry picking" of the comparator, based on the relatively high wages paid to employees in the Postal Operations occupational group. When confronted with the fact that the membership of the PO group was by no means a small group, but rather represented approximately 80% of all Canada Post employees, he agreed that this choice would have been a "very big cherry".⁸⁸

[278] On behalf of the Commission, Mr. Paul Durber, Director of the Pay Equity Directorate at the Commission, and accepted by the Tribunal as an expert in pay equity, indicated in his evidence that the Postal Operations group, as a whole, was approved by the Commission as a suitable comparator group, as it was part of the employer's occupational groupings. At the beginning of the Commission investigation, the PO group also appeared to offer a certain ease of evaluation and comparison because of the general homogeneous nature of the various jobs in each of the PO-INT and PO-EXT subgroups.

[279] According to the evidence of Mr. Chris Jones, the union representative for the complainant group, one reason the comparator group was chosen was because of similarities in the duties and responsibilities of certain CR and PO jobs. A most obvious example was a CR job entitled `customer service clerk' and a PO job entitled `wicket clerk'. Although each job appeared to call for almost identical work, at the time of the Complaint each was paid differently. The superior wage of the wicket clerk and other PO jobs/positions made the apparently predominantly male Postal Operations group an obvious choice as comparator for the complainant. Additionally, the PO group represented, in absolute numbers, the majority of postal employees.

[280] Mr. Jones indicated that although the PO wages were thought to be generally higher than those of the CR's, the sheer size of the Postal Operations group was significant as a reason for its choice. As the largest group of Canada Post employees, representing by far the majority of the employer's total number of employees, the PO's were a natural choice of comparator for the CR's. The fact, too, that some of the work being performed by employees in both the complainant and comparator groups was similar in terms of skill, effort, responsibility, and working conditions underlined, for the complainant, the appropriateness of its choice of comparator.

[281] The Tribunal accepts that the largest occupational group within the organization, a group representing about 80% of the total Canada Post employee population, was an appropriate group to choose as a comparator. It appeared to be a predominantly male occupational group according to the Guidelines. The additional knowledge that certain members of the PO group were performing work which, in some instances at least, was similar to the work being performed by the complainant group added to the appropriateness of the choice.

[282] Additionally, the evidence indicates that there were few other comparators which could have been chosen. At the time of the issuance of the Complaint, the General Labour and Trades, and the General Services occupational groups - both apparently male-dominated, according to the Guidelines - represented a small percentage of Canada Post employees. Moreover, there is no evidence that the work being performed by members of these groups was observed to be similar to that of any members of the CR complainant group.

[283] Accordingly, the Tribunal finds that the complainant, a predominantly female occupational group, and the comparator, a predominantly male occupational group, are appropriately designated under section 11 of the Act and the 1986 Guidelines as representative groups for comparison of work generally performed by women and work generally performed by men. Therefore, the first element necessary to the establishment of a prima facie case under section 11 of the Act has been met.

C. Are the Complainant and the Comparator groups employed in the same `establishment'?

[284] Subsection 11(1) of the Act reads as follows:

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. (emphasis added)

[285] Subsection 11(3) of the Act states:

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.

[286] Section 10 of the 1986 Guidelines reads as follows:

Employees of an Establishment

For the purpose of section 11 of the Act, employees of an establishment include, notwithstanding any collective agreement applicable to any employees of the establishment, all employees of the employer subject to a common personnel and wage policy, whether or not such policy is administered centrally. (emphasis added)

[287] The French language version of section 10 of the 1986 Guidelines reads as follows:

Employés d'un établissement

Pour l'application de l'article 11 de la Loi, les employés d'un établissement comprennent, indépendamment des conventions collectives, tous les employés au service de l'employeur qui sont visés par la même politique en matière de personnel et de salaires, que celle-ci soit ou non administrée par un service central. (emphasis added)

(i) Evolution of the Definition of Establishment

[288] The Commission established the Task Force on Equal Pay in November 1977 "to study the implications of administering section 11 of the Act". Among other matters, the Task Force addressed the question of defining the word `establishment' as used in subsection 11(1).

[289] The Task Force's report, entitled "Equal Pay for Work of Equal Value" and dated March 1978, recommended that `establishment' be defined along the following lines, and be included in a guideline:

`Establishment' means all buildings, works or other places of business of an employer within the limits of the larger of a municipality, municipal district, metropolitan region, county or the national capital region. (Recommendation to be completed)⁸⁹

[290] The Task Force noted that this definition was incomplete and would require further consideration. In this connection, it observed that "the introduction of the word `establishment' in section 11 was deemed to be an attempt to introduce the factor of regional differences in wage levels as a legitimate reason for differences in wages between employees".⁹⁰

[291] Mr. Paul Durber testified that, in his opinion, the above-noted geographic definition of `establishment' probably did not find its way into the September 1978 Guidelines because the Task Force had highlighted some conflicting views on the issue. In addition, he mentioned the need for the Commission to gain more experience in implementing section 11 before enshrining the definition in the Guidelines.⁹¹ There was, therefore, no definition of `establishment' in the 1978 Guidelines.

[292] In September 1984, the Commission issued the Interpretation Guide for section 11 of the Act, entitled "Equal Pay for Work of Equal Value".⁹² This Guide was intended to assist employers and employees to understand how the Commission would assess complaints by providing definitions of certain terms used in the Act and elaborating on the `reasonable factors' included in the 1978 Guidelines.

[293] The 1984 Interpretation Guide defined `establishment' as follows:

An establishment refers to all buildings, works or other installations of an employer's business that are located within the limits of a municipality, a municipal district, a metropolitan area, a county or the national capital region, whichever is the largest, or such larger geographic limits that may be established by the employer or jointly by the employer and the union.

[294] Mr. Durber testified that despite the 1984 Interpretation Guide's general support of the geographic definition of `establishment', the Commission did not apply it consistently as the Commission also frequently used an alternative definition based on functional lines, particularly for cases involving the Federal Government and national organizations which it considered to be single nation-wide `establishments'.

[295] The Chief Commissioner, in a letter dated March 19, 1985, to about 60 public and private sector employers, including Canada Post, sought their views on a number of proposed definitions and guidelines, including the definition of `establishment'. The request was aimed at removing much of the uncertainty experienced by some employers in implementing their own "pay equity" programs.

[296] The Chief Commissioner's letter indicated that it was proposed to define `establishment' more broadly and on a different basis than the one used in the Commission's Interpretation Guide. The proposal was that a functional definition would replace the Guide's geographic definition. Specifically, the Commission proposed that:

Employees of an employer shall be considered to be in the same establishment when they are subject to a common set of personnel and compensation policies, regulations and procedures; and when these policies, regulations and procedures are developed and controlled centrally even though their administration may be delegated to small units of organization.

[297] Canada Post's Vice-President, Personnel, responded to the Chief Commissioner's letter on June 3, 1985, specifically addressing the proposed definition of `establishment' as follows:

The Commission's proposed definition of `establishment' is also a source of concern. While the Commission has clear authority under the provisions of Section 22 of the Canadian Human Rights Act to provide guidelines, it is our understanding that those guidelines must conform with the Act taken as a whole. To move from a geographically-based to a functionally-based determination of establishment, we suggest, would be inconsistent with the Act and at odds with other statutory and judicial interpretations of the expression which frequently speak in terms of location rather than function.

[298] Consultations between the Commission and the various employers, including Canada Post, led to certain changes in the proposed definitions and guidelines. Ultimately, the 1978 Guidelines (as modified in 1982) were replaced by the November 1986 Guidelines which for the first time included a definition of `establishment' (section 10). This definition was functionally-based along the lines of the one identified in paragraph [296] above.

[299] At the same time, a new `reasonable factor' was added to the then-existing list (section 16), recognizing that a difference in wages between male and female employees performing work of equal value in the same establishment is justified by "regional rates of wages, where the wage scale that applies to the employees provides for different rates of wages for the same job depending on the defined geographic area of the workplace".

[300] Accordingly, the Commission had formally moved, by late 1986, to a functionally-based definition of `establishment' from its earlier, inconsistently applied, policy of employing a regionally-based definition. The reasons for the Commission's shift are best explained by Mr. Durber's following response:

Q. Could you remind us of the rationale for shifting to the functional basis?

A. Yes. My view is that it was to allow a broader, let us say, more liberal interpretation and application of section 11.⁹³

[301] The Commission indicated that during the Investigation Stage of this Complaint (1984-1991), the assumption that the complainant and comparator groups were in one `establishment' was uncontested. Canada Post did not raise the issue of the definition of establishment within the context of the Complaint during this period, although it was involved in discussions with the Commission on this very topic during the drafting of the 1986 Guidelines.

[302] Mr. Durber testified that he recalled learning from the investigator of the Complaint "sometime in 1991" that Canada Post's lead contact had indicated that the Corporation was "thinking about whether establishment might not be an issue". Mr. Durber was unaware of any formal indication from Canada Post that the definition of `establishment' would be argued as part of the Respondent's challenge to the Complaint.⁹⁴

[303] After a lengthy cross-examination, including questions on the Commission's work to define `establishment', Mr. Durber was asked by Commission counsel if his original opinion concerning the meaning of `establishment' had changed. He replied that, in the context of this Complaint, he

... continue[d] ... to see there being one establishment, a good deal of commonality at the level of management accountability and otherwise, bringing those groups into one establishment, as meant by section 10 of the Equal Wages Guidelines, and, thus permitting continued comparison of job value as between clerks and postal operations people.⁹⁵

[304] Both the Commission and the Alliance acknowledged that counsel for Canada Post did raise questions about the meaning of the term `establishment' in the context of section 11 of the Act and section 10 of the 1986 Guidelines during his opening remarks before this Tribunal, in February 1993, as follows:

The next issue is the one that we see as being of pivotal importance, and that is the question of establishment ... So, the legislation certainly contemplates different establishments within one employer. The big question, which has never been considered, and as far as we are aware, never been argued, is: What is an establishment?⁹⁶

[305] Canada Post's counsel subsequently addressed the matter of `establishment' specifically in the context of the Guidelines, noting that "the 1978 Guidelines contain no definition of `establishment'. The 1986 Guidelines contain a definition of `establishment'".⁹⁷

[306] He then stated what Canada Post's position would be, as follows:

Our position will be that the other side must come up with a definition of `establishment' which includes the CR's and the PO-Internal, PO-External and PO-Sups in the same establishment, to the exclusion of other groups. If the definition of `establishment' excludes other workers who work in the operational area at Canada Post, how does it include the CR's?⁹⁸

**(ii) The `Airlines Case'**

[307] All parties agree that this Tribunal is bound by the decision of the Federal Court of Appeal, dated March 18, 2004, which addressed the issue of the definition of `establishment' in the context of section 11 of the Act, and section 10 of the 1986 Guidelines. This decision reversed the decision of the tribunal and the Federal Court (Trial Division) in a "pay equity" complaint brought by flight attendants (predominantly female) at Air Canada and Canadian Airlines, who were represented by one union. They named, as their comparator groups, pilots and maintenance/technical workers (predominantly male) who were represented by two other unions. A fundamental issue dealt with by the tribunal as a preliminary matter was whether the complainant and the comparators were employed in the same establishment.

[308] The tribunal found that the complainant, represented by the Canadian Union of Public Employees (Airline Division), failed to demonstrate "any semblance of essential common wage and personnel policies across the bargaining units"⁹⁹ and concluded that the three employee groups were not in the same `establishment' for the purposes of a section 11 complaint. The Federal Court (Trial Division) upheld the decision of the tribunal.¹⁰⁰

[309] The Federal Court of Appeal unanimously reversed this finding, deciding that the functionally-based definition of `establishment' in section 10 of the 1986 Guidelines would, in most cases, place all employees of an employer in the same establishment even though some employees might be represented by different unions.¹⁰¹ The complainant and the chosen comparators in the `Airlines Case' were, accordingly, found by the Court to be in the same establishment for the purposes of section 11 of the Act.

[310] The Federal Court of Appeal highlighted the importance of interpreting human rights legislation broadly, liberally and purposively in the context of the words and purpose of the statute concerned. The Court stressed the necessity to interpret the Act and the 1986 Guidelines in a purposive manner, always being aware of the quasi-constitutional nature of the Act, and its aim to eliminate discrimination. The Court noted that in complaints brought under section 11 of the Act, the "broad purpose" of the section - "to preclude wage discrimination on account of gender"¹⁰² and "the more particular purpose ... the promotion of pay equity"¹⁰³ - must guide the interpretation of the words of the section and the Guidelines promulgated by the Commission.

[311] The Court indicated that the test for the interpretation of the word `establishment' in the context of these purposes was whether there was "evidence that the employer treats the employee groups as being part of a single, integrated business. If there is such evidence, the employees are in the same establishment".¹⁰⁴ Evans, J., in his concurring reasons, stated that "...employees of the same employer will normally be subject to `a common personnel and wage policy' when they are employed in the same business entity".¹⁰⁵

[312] In other words, the definition of `establishment', as noted in section 10 of the 1986 Guidelines, was accepted by the Court as necessitating evidence of common personnel and wage policies which would be general in nature. There would be no need to examine the minute details of different collective agreements negotiated by unions which represent the groups being compared. The Court agreed that "the definition of establishment should not be based on the myriad of details found in collective agreements".¹⁰⁶

[313] Mr. Justice Evans, in his concurring reasons, also indicated that "[t]he terms of collective agreements that apply to complainants and other employees with whom they wish to be compared for pay equity purposes are irrelevant to determining whether the complainants and the comparators are employed in the same establishment within the meaning of section 11 of the ... Act ... and section 10 of the Equal Wages Guidelines..."¹⁰⁷

(iii) Impact on the Current Case of the Federal Court of Appeal Decision in the `Airlines Case'

[314] Although, as noted above, all parties to the Complaint before this Tribunal acknowledged that the Tribunal is bound by the decision of the Federal Court of Appeal in the `Airlines Case', Canada Post argued, in its written submissions, that the decision was also important to the Tribunal because it underscored several of Canada Post's previous arguments.

[315] Three of Canada Post's arguments deserve particular reference. The first related to the question of the essential objective of section 11 of the Act and the presumption of discrimination based upon sex, found in that section. The second dealt with an accurate and fair determination of the value of `wages' for the purposes of a section 11 inquiry. The third concerned the relevance of collective bargaining strength in a "pay equity" study.

[316] As noted in paragraph [310], clear guidance on the question of the essential objective of section 11 has been provided by the Federal Court of Appeal in the `Airlines Case' decision. With respect to the issue of the presumption of discrimination, the Tribunal has already addressed this matter in Section IV, C of this Decision.

[317] Canada Post's second argument relating to the definition of `wages' is considered in the context of Section VIII, entitled Non-Wage Forms of Compensation, which follows in this Decision.

[318] Canada Post, in its third argument, has reasoned that the Federal Court of Appeal decision recognizes that bargaining strength is not only relevant to the `Airlines Case' but also constitutes an important part of any inquiry into sex-based wage discrimination.

[319] The Tribunal finds that the Federal Court of Appeal decision confines its consideration of bargaining strength to

...the factors which the Tribunal is to use in determining whether employees receive equal wages and perform work of equal value. To the degree that the evidence of differing bargaining strength is evidence pertaining to these factors, it is relevant and will be considered by the Tribunal at the substantive phase of the analysis.¹⁰⁸ (emphasis added)

[320] The factors are identified by the Federal Court of Appeal as being those set out in section 11 of the Act and in the Guidelines. They are, therefore, by definition, limited to the interpretation of section 11 and to the "reasonable factors" identified in section 16 of the 1986 Guidelines.

[321] The Tribunal does not find that the decision of the Federal Court of Appeal in the `Airlines Case' sanctioned an `open-ended' approach to the "reasonable factors", allowing the admittance of additional factors such as bargaining strength to those already provided for in the Guidelines. The Federal Court of Appeal clearly indicates that to the degree that this evidence of differing bargaining strength is evidence pertaining to the factors set out in section 11 of the Act and in the Guidelines, that evidence will be considered at the substantive phase of a tribunal's analysis.

[322] Has `bargaining strength' been presented in this Complaint as evidence pertaining to the factors listed in section 16 of the 1986 Guidelines as `reasonable factors' to justify an employer establishing or maintaining a difference in wages between male and female employees performing work of equal value in the same establishment? Canada Post has submitted that the differences in the collective bargaining philosophies of the complainant and comparator groups must be considered by the Tribunal. It has also argued that the historically gender-neutral, and numerically large, PO-4 sub-group, an active representative in the collective bargaining process for the PO group, must be considered because of its "straight-line" wages philosophy.

[323] The Commission and the Alliance have submitted that union bargaining strength has never been designated as a "reasonable factor" in the Guidelines. Canada Post's arguments, therefore, should not be considered unless there is some evidence which links those arguments to the "reasonable factors". Very early in the hearing, counsel for the Commission addressed this very point, as follows:

Another point that is not disputed is the strength of unions or union bargaining strength or whatever. It's not in the Guidelines. It wasn't in the Guidelines in 1978, 1982, or 1986. It has never been in the Guidelines...¹⁰⁹

[324] The Tribunal is not aware of any bargaining strength evidence specifically pertaining to the factors set out in section 11 of the Act and in the Guidelines having been submitted, at any time, by the parties in this Complaint.

[325] The original Complaint did not address directly the issue of `establishment'. The Complaint was drafted by the Complainant to indicate that the employer, Canada Post, had allegedly violated section 11 of the Act "by paying employees in the male-dominated Postal Operations Group more than employees in the female-dominated Clerical and Regulatory Group for work of equal value". Further, it was alleged that the sex composition of the two groups was the basis for the difference in wages, and, thus, the Complaint alleged discrimination based on sex.

[326] Therefore, as noted by Mr. Durber in his evidence about the Commission's interpretation of `establishment' when it was dealing with complaints brought by groups working for the Federal Government and other large nation-wide corporations, it is apparent that the Complainant and the Commission assumed, from the inception of the Complaint, that `establishment' and `the employer' were synonymous.

[327] For the reasons given by the Federal Court of Appeal in the `Airlines Case' decision, this assumption would appear to be the correct one. Although there will be times when an employer has more than one establishment within its purview, in most cases the employer and the term `establishment', in the context of the Act, will be one and the same when the employer treats its employee groups as being part of an integrated business entity with a commonality of personnel and wage policies.

[328] In this Complaint, considerable evidence was presented, usually through Canada Post witnesses, that Canada Post operated as an integrated business entity with, generally, overall personnel and wage policies. Several examples of such evidence are considered below.

[329] In May 1997, Elisabeth Kriegler, President and CEO of Elisabeth Kriegler and Associates, an organization of Change Management Consultants, appeared before this Tribunal. She had been called by Canada Post as a general witness. She had served at the Vice-Presidential level in several corporate functional areas in Canada Post during the period 1983 to 1992. She then occupied the office of Senior Vice-President - Administration from 1992 to 1995. From 1995 to early 1997, she was President of Canada Post Systems Management Ltd., a company which owned the intellectual property of a number of management systems and processes developed, over the years, by Canada Post and marketed, internationally, through licensing arrangements.

[330] Ms. Kriegler emphasized that operating a postal system is probably one of the most complex logistics businesses in the world involving, in Canada Post's case, not only its own employees but also many thousands of others under contract. "All of them are an intricate part of this integrated network, all of which must operate in concert and according to standards and in harmony..."¹¹⁰

[331] As pointed out in paragraph [33], in carrying out its objects, the new Crown Corporation created in October 1981 was to have regard to "...the need to conduct its operations on a self-sustaining financial basis while providing a standard of service that will meet the needs of the people of Canada...". Ms. Kriegler indicated that this called for the efficient and effective collection, processing and delivery of mail within a financially competitive framework which, in turn, necessitated the development and introduction of a comprehensive series of operational, financial, human resources, marketing and management systems.

[332] Ms. Kriegler reported that the operations of Canada Post "...are the heart and soul of this Corporation, how without it, it is not a Corporation, it is not a business...".¹¹¹ The establishment of the National Control Centre in Ottawa, in the mid-1980s, reflected the crucial need for a centralized operations control and monitoring system.

[333] Members of Canada Post's senior management meet daily in the National Control Centre to review operational problems which are fed into headquarters from divisional control centres across the country. This encourages operating people to make decisions as and where problems arise. The operating network is supported by a series of systems that track and trace the movement of mail throughout Canada.

[334] The Tribunal members had the opportunity to visit the National Control Centre and saw it in action with its inward and outward flow of information visually displayed, in colour, on screen, against the backdrop of a giant map of Canada.

[335] The role and impact of Canada Post's operational functions including the National Control Centre are, perhaps, best summed up by Ms. Kriegler's following statements made before the Tribunal in May/June 1997:

...and in fact today, notwithstanding decentralization and empowerment ... the control of the operation is totally central today and must always remain so because the minute you let that loose, the network starts falling apart. So it is centrally controlled and that is the role of the Control Centre and that is why the President and the Chairman and the Chief Financial Officer and the Marketing Senior VP and the Operating Senior VP and all their senior people sit at that table every single morning. That is the central control.¹¹²

All the employees know that they are a component or a part of that larger integrated system.¹¹³

[336] Certainly, the nature of the operations of Canada Post, and particularly the role of the National Control Centre, offer clear evidence of Canada Post functioning as a single integrated business and treating all of its employee groups as essential components of that entity.

[337] Ms. Kriegler also demonstrated that this was not limited to the area of operations. The development of management functional areas of responsibility following the creation of Canada Post as a Crown Corporation went well beyond the critical area of operations. Ms. Kriegler reported that the various supporting management functions were gradually brought in from the different agencies of government and developed under the wing of Canada Post with its own staff.

[338] Functions such as finance, personnel and staff relations, and labour relations were transferred shortly after Crown Corporation status. Purchasing, pay and benefits, property management and legal affairs were transferred at later dates. Canada Post staff had to be built up, and policies, standards and procedures had to be developed in each of the functional areas.

[339] The 1980's also saw a growing emphasis on marketing and meeting the customers' needs. Retail outlets and franchising arrangements evolved. Even a Research and Development Centre emerged where new Canada Post products, services and equipment were conceived, developed and tested.

[340] Organizationally, there was, according to Ms. Kriegler, an Executive Vice-President of Personnel and Labour Relations in 1983. By 1992, Ms. Kriegler, herself, assumed policy responsibility as Senior Vice-President - Administration, for human resources, labour relations, personnel, legal affairs, and several other functions.

[341] Ms. Kriegler identified certain situations where a common corporate approach was taken by Canada Post in the personnel policy field or in areas closely related to personnel policy, or, as it is currently more frequently called, Human Resources policy. One was the creation of the Canada Post Learning Institute which established a common centralized training budget by drawing particular training program funds from individual operating units. A principal objective was to coordinate the development and use of training programs for employees from across the entire organization to derive a more effective return for both employees and the employer.

[342] Another example was with respect to labour relations and collective bargaining strategies which, inevitably, touch on wage policy. Between bargaining sessions, Canada Post coordinated, at the senior management level, the development of goals and strategies it would like to achieve with its various unions. Ms. Kriegler, in her capacity as Senior Vice-President - Administration, would take such proposed goals and strategies to the Management Committee for consideration.

[343] There was also evidence presented by other Canada Post witnesses who appeared before this Tribunal which indicated central corporate direction in areas involving Compensation and Benefits, labour relations, employee training and human resources policy. Examples are explored below.

[344] Mr. Harry Phillips, Director - Safety, Ergonomics and Industrial Hygiene at Canada Post headquarters, testified in August 1997. He spoke about the Corporate Manual System which came into effect in 1989 for the purpose of providing appropriate corporate direction in consolidating all functional procedures. As examples, he mentioned procedures dealing with functional areas such as operations, engineering, human resources (personnel), and safety hazards.

[345] Mr. Ron Featherstone testified in December 1998, as Manager of Collection and Delivery for the Northern Zone in Vancouver. He indicated that one of his responsibilities was to establish "...long term objectives that complement the Corporation's Operating Principles and its Corporate Objectives...". Under cross-examination by Commission counsel, Mr. Featherstone agreed that it is his understanding that corporate principles and corporate objectives are intended to guide all employees of the Corporation, and would apply to both CR and PO employees and their respective work.¹¹⁴

[346] In April 1999, Ms. Joanne Hronowski, a Payroll Officer for the Prairie Region (who had, on occasion, served as Acting Manager - Pay and Benefits) testified that several manuals which had originally been issued by the Department of Supply and Services, guided Payroll and Benefits officers in their work. Updates to the manuals, communiqués and informative circulars about items such as particular kinds of benefits, were received from Head Office. "...So most of the stuff was vetted at the Head Office level and then came through us".¹¹⁵ She also testified that the then-current payroll system used in her region was a national one. "It is driven totally by Head Office".¹¹⁶

[347] Mr. Charles Reece, a long-time employee of Canada Post and, most recently, Manager for the Revenue Verification Unit of the Gateway Bulk Mail Facility at Mississauga, Ontario, when testifying in April 1999, about his facility's capacity to train supervisors and staff, said as follows:

And at times we even had people from Head Office come down. Usually when something new was being introduced and it was something that we didn't know anything about, they would come down and do formalized training with us.¹¹⁷

[348] In May 1999, Mr. Frank Pasacreta, Vice-President of Operations for the B.C. Maritime Employers Association, testified before the Tribunal. He had held this position since 1987. Prior to that, he had been Manager of Labour Relations for Canada Post in Vancouver, responsible for the Pacific Region from 1984 to1987. Mr. Pasacreta was asked by Commission counsel if, during his tenure in the Pacific Region, he or his staff had been involved in collective bargaining negotiations for employees in his Region. His response was as follows:

...primarily the people who did the bargaining were the folks at head office. Some of us sat in on some of the sessions. I sat in on a few myself, but the primary responsibility was a head office responsibility.¹¹⁸

[349] Ms. Karin Vogt, Compensation and Benefits Officer, at the management level, from Burnaby, B.C., testified in September 1999 that Superannuation, Procedures and other manuals constitute an important source of information for those working in her area of responsibility. The use of computer-based record systems is also critical. She confirmed that manual up-dates and communiqués came from Head Office. She also indicated that training sessions were sometimes handled by headquarters. Specifically, she noted that "...last year we had someone come out from head office and she went over the disability insurance plan".¹¹⁹

[350] Mr. Brian Wilson testified in May 1999. A long-time employee of Canada Post, he retired in 1995. His last position was Manager - Employee Relations, Central Region. During cross-examination by Commission counsel, Mr. Wilson confirmed that it was his understanding that a series of personnel policy directives existed in Canada Post, most of which would have been

issued by the corporate Human Resources group in headquarters. These are applicable to all employees throughout the Corporation, and include the following:

Official Languages Policy
Employee Assistance Program (counselling and referral service)
Sensitive Information Policy
Religious Observance, Sick Leave and Pregnancy, Modified Duties, Sign Language
Human Rights and Employment Equity, including Partnering with Women, Sexual Harassment, People with Disabilities, Visible Minorities

[351] The Tribunal concludes that the above-noted evidence demonstrates that Canada Post, during the time frame of this Complaint, had become an increasingly well integrated business entity with considerable corporate level policy direction. The corporate policy direction extended to the various regional operations of Canada Post, encompassing its many employees across the country. Indeed, much of the evidence suggests a very good mutual working relationship between the regions and Head Office.

[352] Human Resources (or personnel) policy direction clearly emanated from corporate headquarters and addressed all employees as members of the integrated business. Equally, labour relations, including collective bargaining negotiations which include wage policy considerations, received corporate direction, and even direct involvement, from Head Office.

[353] Therefore, the Tribunal finds that all employees of Canada Post have been, as applicable, subject to the various common corporate policy directives issued by the Corporation, including those respecting personnel and wage policies. As a result, the Tribunal finds that, for the purposes of section 11 of the Act, the employee groups representing the complainant and the comparator are employed in the same establishment.

[354] Accordingly, the second element necessary to the establishment of a prima facie case under section 11 of the Act has been met.

D. Does the comparison of the work of the Complainant group and the Comparator group establish that the work being compared is equal in value?

Are the jobs/positions data and the process comparing the work of the Complainant and the Comparator groups reliable?

(i) Background

[355] All three parties have recognized the importance of undertaking job evaluations with reliable job information and with a reliable job evaluation plan. Additionally, the plan and the process chosen must be suitable for "pay equity" purposes. This is not in dispute. What has to be determined, however, is the extent of the reliability of the job information and of the methodology employed in the evaluation of the jobs/positions involved in this particular Complaint. To be able to come to a reasonable conclusion concerning the value of the work performed by the complainant and the comparator occupational groups, the evaluation process as a whole must be reliable, on a balance of probabilities.

[356] The Commission presented, as evidence, a booklet entitled "Implementing Pay Equity in the Federal Jurisdiction".¹²⁰ Originally, this had been a paper written by staff within the `Pay Equity Directorate' of the Canadian Human Rights Commission. It was published as a booklet by the Commission in March 1992. At the time of publication, the Commission had been involved in a number of "pay equity" complaints, one of them the Complaint before this Tribunal.

[357] The introduction to the booklet states that it had been prepared in response to requests from employers and unions. It was meant to be advice about how "to pursue effective pay equity programs" under the Act and its accompanying Guidelines. The introduction states further that the points made in the booklet "are based on the experience of Commission staff working in pay equity, as well as comments received from employees and unions on earlier drafts of this paper".

[358] While this booklet did not exist during the Investigation Stage (1984-1992) of this Complaint, much of the thinking expressed therein was evolving within the Commission during that period. It is, therefore, helpful to refer to it to provide background to the Commission's thoughts, by 1992, about job evaluation plans, their administration, and the collection of job data/information. The following paragraphs are excerpts from various sections of the booklet which are considered pertinent.

Job Evaluation Plans

Job evaluation plans are the key to determining what constitutes "work of equal value". They do not eliminate subjectivity from the process of valuing work, but they do make the process systematic and so ensure that values are applied in a way that is consistent. Without such a systematic examination of job values, it is easy to perpetuate prevailing stereotypes about the worth of different occupations - stereotypes that generally work to the disadvantage of jobs done by women.

Job evaluation plans may be developed especially for an organization or they may be based on a standard plan purchased from a consulting firm. These latter are "off-the-shelf" and have established criteria for evaluations, while others use computers to generate criteria based on data gathered within the organization. All plans eventually rely on a set of standard factors and weightings against which different jobs are rated.

In order to be an acceptable instrument for implementing pay equity, a plan must meet a number of tests:

it must include the four value criteria set out in the Act and elaborated upon in the Guidelines: skill, effort, responsibility and working conditions;
it must measure value in a way that allows ready comparison between jobs - usually this means adopting a point-factor rating system - other types of systems may be acceptable in certain circumstances, for example, paired comparisons in small organizations; and
it must be free of gender bias: Gender bias refers to any factor or behaviour which, even unintentionally, unfairly favours one sex over the other. In the context of pay equity studies, gender bias can affect both the design of job evaluation plans and their application.

Because pay equity is premised on the assumption that the worth of different positions across an organization should be compared, use of a single plan to evaluate all jobs is essential.

With respect to the plans themselves:

factor definitions should be generic, relying as little as possible on illustrative job descriptions that could produce bias - this helps minimize raters' tendency to stereotype tasks or functions as male or female;
the factors used must incorporate all significant elements of all the work being evaluated, including those aspects of female-dominated jobs traditionally overlooked in job evaluation;
weightings given to factors typical of predominantly male work and predominantly female work should be equitable; and
computerized plans should be programmed such that important elements of traditionally female work are not left out of the automated development of factors and weighting. Typically, this requires especially careful design of questionnaires to ensure that key information is not missed.

With respect to the administration of plans:

women and men should have similar representation on all committees;
participants should be drawn from all levels of the organization; and
it must be made clear to participants that during the pay equity process, all are equal - those from the lower ranks of an organization should feel comfortable expressing their own views and challenging the opinions of others.

Collection of Job Information

With respect to the collection of job information:

job descriptions should not be used on their own or treated as the primary source of data, since they often replicate prevailing stereotypes and are not always an up-to-date, accurate reflection of work done;
instead, sources of information which allow the incumbent himself or herself to outline work duties should be employed - in most cases, this involves use of a questionnaire;
the questionnaire must be carefully designed and tested, possibly through a pilot study to ensure that it captures all significant aspects of male- and female-dominated jobs and is appropriate to the structure of the job evaluation plan;
it must be made clear to all involved that the questionnaire should reflect actual work being done, not theoretical duties;
supervisors should be given an opportunity to review completed questionnaires and add any comments or reservations on an attached sheet; and
where questionnaires do not seem to offer sufficient information, they may be followed up with face-to-face, structured interviews between evaluators and incumbents.

Questionnaires can be open, closed or mixed, depending on the requirements of the plan, the preferences of those running the study, and the size of the organization. In smaller organizations, it may not be possible to carry out the testing needed to develop a reliable closed questionnaire. However, open questionnaires must be used with care. Efforts must be made to ensure that men and women use similar language to explain their work. Thus, when open questions are chosen, instructions should be included which encourage all those filling out the questionnaire to use accurate terms in describing their job functions. Examples showing how different sorts of duties could best be described may be appropriate.

Joint Employer-Employee Cooperation

Although the legal onus for ensuring pay equity is on the employer, the Commission believes that pay equity programs are most successful when based on full cooperation between employer and employees ... Both sides must contribute to the process. Employers provide the funding for studies and any necessary adjustments, as well as the informed perspective of managers on what different jobs entail. Employees provide the key job information and support for any changes to prevailing relativities that may result. Both employers and employees have input into the definition of job worth as reflected in the job evaluation plan and its application.

Joint studies generally begin with an agreement between the employer and bargaining agent(s) which outlines the objectives of the initiative and its basic structure. A joint steering committee may be set up to choose an appropriate job evaluation plan, perform benchmark evaluations, establish evaluation committees, work out other details of the study and guide it through to a successful conclusion. Most of the actual evaluations are carried out by one or more evaluation committees, which, we suggest, should include a comparable number of women and men from all levels of the organization. Any wage adjustments found to be necessary as a result of the evaluations must be agreed to by both sides.

[359] The Tribunal finds that the aforementioned points described in the Commission's booklet constitute a general guide and benchmark model for collecting reliable information and for processing that information in a manner that should, given an acceptable job evaluation plan and competent evaluators, result in the determination of reliable values of the work being assessed and compared in a "pay equity" study.

[360] These points made by the Commission are the very points made by such experts as Dr. Pat Armstrong, accepted by the Tribunal as an expert in women's work, women's wages, and the sociology of equal pay legislation, who presented evidence to the Tribunal concerning, amongst other things, the history and development of the concept of "pay equity", and the methodologies used to implement that concept.

[361] It should be noted, however, that the points made in the Commission's booklet are predicated upon the assumption that the "pay equity" process will be one which involves an employer and its employees in a working partnership. "Pay equity" will be the common goal of that partnership. The booklet, and its advice concerning "pay equity" studies, does not envision a process which is evolving in a litigious context.

(ii) Issues

[362] Consequently, the issues which will be addressed are as follows:

What job evaluation system, or plan, was used to undertake the evaluation of the CR and PO jobs/positions, and how reliable was it?
What process was used and how reliable was it in analyzing the collected job data/information for purposes of assigning values to the CR and PO jobs/positions considered?
What job data/information was collected, and from what sources, and how reliable was it?
What were the resulting values attributed to the various CR and PO jobs/positions, and how reliable were they?

[363] These issues can best be considered by distinguishing between two periods of time:

First, the duration of the Investigation Stage of the Complaint - 1984 to 1992, when the Commission was coordinating the collection of job data and performing job evaluations, and

Second, the period after the establishment of the Tribunal in 1992, when the Alliance had engaged a three-person team of professional job evaluators to review the job data already collected by the Commission and augment it where possible, and to undertake independent evaluations of the jobs of the complainant and comparator occupational groups. The professional team was active in mid-1993 and late 1994, and re-visited their work in July 1997 and June 2000.

E. Review of Job Information Collected and Methodology Used: Investigation Stage

[364] A joint employer/employee follow-up study of the Complaint before this Tribunal was not undertaken. There was, therefore, no opportunity to establish a joint steering committee to co-ordinate the selection of a job evaluation plan, the gathering of job data/information, and the evaluation of jobs. Of necessity, the Commission took the lead, through its Investigation Stage, in coordinating matters relating to complaint follow-up with both the Complainant and the Respondent.

[365] While a joint employer/employee study would have been the most suitable way of addressing the Complaint, the Tribunal finds that the lack of such a study, for whatever reasons, need not pre-empt addressing the Complaint by other approaches. The Complaint was made pursuant to section 11 of the Act; the Commission's mandate is to investigate each complaint made to it under the Act. When the parties cannot be moved to a negotiated settlement, the Commission's job is to investigate a complaint to the best of its ability. For example, while the job evaluation plan used to determine the value of particular job work should, normally, be a plan already in use by the employer, in its absence an off-the-shelf evaluation plan would be acceptable if it is free of gender-bias and capable of generating a reliable result.

(i) Commission's 1987 Job Evaluations

[366] In this Complaint, as already noted in paragraph [17] the first job evaluations were conducted by Commission staff in 1987 based on data collected in 1986. Those evaluations involved what the Commission described as a random sample of 194 CR positions. The sampled CR positions were evaluated using Canada Post's System One job evaluation plan. No PO positions were evaluated in 1987.

[367] System One was a plan that had been categorized as having "Hay-like factors". It did, however, contain some variations from the standard "Hay Method" evaluation plan, most especially with respect to the working conditions factor. As the Commission investigated the Complaint, System One was still under joint development by Canada Post and the Alliance. It was intended for use by employees represented by Alliance bargaining units throughout Canada Post. As Canada Post pointed out, System One would, therefore, not be suitable for the eventual evaluation of PO jobs, since their incumbents were represented by other bargaining units. Moreover, the Alliance advised against its use even for CR job evaluation purposes at this incomplete stage of development.

[368] The principal sources of job information for the evaluation of the 194 CR positions in 1987 were successive lists of employee print-outs furnished by Canada Post and a Job Fact Sheet - a detailed Questionnaire - that had been designed by the Commission for completion by employees sampled from those lists. Upon its completion, with the requisite attachment of the relevant job description and organization chart by each employee respondent, the Job Fact Sheet was to be signed off by the appropriate supervisor and Division Manager. The questionnaire, or Job Fact Sheet, was assembled in the first half of 1986. During the summer of that year, it was completed by the CR employees who had been randomly chosen to be a representative sample of CR positions.

[369] The Commission had decided that such a sampling of CR incumbents would be necessary since a full census of the total CR population of about 2,300 would be unmanageable in terms of time and money. A stratified random sample was developed in 1986 by a senior official of the Commission. It consisted, initially, of 246 names of CR incumbents plus 33 "alternates" for a total of 279. Subsequently, some names were dropped and others were added. The actual number involved is difficult to verify from the available documentation. The final proposed CR sample may have been as high as 355, including "alternates". What is clear, however, is that the Commission received 194 completed and useable Job Fact Sheets from CR incumbents which became the basis for the 1987 CR evaluation.

[370] Meanwhile, an Interview Guide had been developed by the Commission with input from the Alliance and Canada Post. Its purpose was to guide the Commission's investigator during follow-up interviews which were to be conducted with the incumbents, to clarify answers given on the Job Fact Sheet. Space was provided on the form for the investigator to record comments made by the incumbent and the accompanying supervisor. The form was not seen by incumbents. It was intended that, like the Job Fact Sheet, the Interview Guide would be used with both CR and PO employees. In fact, however, both were used only to elicit information from the CRs. All interviews were completed by December 1986.

[371] The Job Fact Sheet was used as the Commission's primary source of job information and the other materials served as secondary and tertiary sources.

How were the job evaluations of the sample of 194 CR positions actually conducted in 1987?

[372] An `evaluation team' comprised of two Commission officers was established to conduct the evaluations. These evaluations were done from April to September 1987. The team was supplemented by one of three additional officers, assigned progressively based upon which officer had interviewed the CR incumbent of the position being evaluated. The team, of mixed gender, used the System One plan, drawing data from the 1986 Job Fact Sheets, job descriptions, organization charts, and the Interview Guides.

[373] Mr. Paul Durber, Director - Pay Equity of the Commission, indicated in his evidence before this Tribunal in June 1993, that he and the senior investigator of the Complaint had decided, about mid-1991, to discard the 1987 CR evaluations and to subject the 1986 Job Fact Sheet information to another job evaluation assessment. Most of the Commission's officers who had served on the 1987 evaluation committee were no longer on staff, so a new group was struck. Hence, the 1987 CR evaluation results were not used in the Commission's final investigation process.

(ii) Commission's 1991 Job Evaluations

[374] A new set of job evaluations was undertaken by the Commission staff in 1991 for use in its Final Investigation Stage. These involved 93 CR positions (reduced by the Commission from the original 194) and 10 `generic' PO jobs. The possible use of the evolving System One job evaluation plan was considered by the Commission but discarded in favour of the off-the-shelf Hay XYZ Job Evaluation Plan.

How were the job evaluations of the sample of 93 CR positions and 10 "generic" PO jobs actually conducted in 1991?

[375] The Commission evaluation work was now to be a comparison between the sampled CR positions and 10 `generic' PO jobs, the creation of which had resulted, essentially, from the inability of the Commission and Canada Post to agree on sample sizes and data collection instruments for the comparator PO positions. The Commission had consulted Statistics Canada and received its recommendation concerning stratified random sampling of the PO community, comprising internal, external and supervisory sub-groups, and had planned to proceed using the Job Fact Sheet questionnaire. Canada Post, however, would not allow PO incumbents to complete the Job Fact Sheet on company time. The union representing the PO's would not allow their membership to participate in "after hours" unpaid work. As a result, the Commission opted to use the information made available to them by Canada Post, and, using that information, created a grouping of `generic' PO job categories - covering both internal and external operational functions, but excluding the PO supervisors. The 10 `generic' jobs were, therefore, not actual positions but represented the ten mostly homogeneous jobs done by PO incumbents.

[376] The creation of the 10 `generic' PO jobs involved the dropping, by the Commission, of the PO supervisors subgroup (PO-SUP). This was a significant action as the PO-SUP subgroup represented 6 different levels of supervision with a large number of job titles. Many of the titles occurred at more than one level, making it difficult to reconcile them into `job specifications' without a sampling of incumbents and use of a Job Fact Sheet, or questionnaire. With the Commission's decision to move to the 10 `generic' PO jobs, it was considered too onerous and delaying to sort out the PO-SUP situation with Canada Post. While the Alliance was consulted, Canada Post was only advised of this decision. One important result was an inconsistency with the CR sample. That sample included supervisors at the CR-5 level.

[377] Commission staff received a short period of training in the use of the standard XYZ Hay Plan from a senior manager at Hay Canada, Mr. Roger Childerhose. After the training period, the Commission's senior investigator and one other senior officer began evaluating, individually, 16 positions which they pronounced to be Benchmark positions (10 CR positions and 6 `generic' PO jobs). This new evaluation work began in July 1991. They then jointly reviewed and "sore-thumbed" all 16, and periodically consulted with the Commission's Director - Pay Equity, Mr. Paul Durber. The two officers continued to re-evaluate CR-2's and CR-3's, intending to re-evaluate all of the original sample of 194 CR's and to evaluate the 10 PO "generics". The senior investigator was called off the job to handle other priorities, and the second officer continued on her own. She was subsequently joined by another officer plus an outside consultant. All three then rated batches of CR and `generic' PO jobs individually, followed by periodic informal joint review and "sore-thumbing". This "team" was of mixed gender.

[378] By September 1991, partway through the re-evaluations of the CR's, the officer-in-charge was asked to reduce the original sample of 194 CR's to a more workable number. After studying this situation, she proposed a revised level of 93 which was accepted by the Commission as the new sample.

[379] The Commission's evaluators used the off-the-shelf Hay XYZ Evaluation Plan for both the 93 CR and the 10 PO positions/jobs. It was this Plan which was the basis for their training session with the Hay organization. The sources of job information for the CR's was essentially the same as used in the 1987 evaluations - the relevant 1986 Job Fact Sheets and supporting job descriptions, organization charts and Interview Guides. For the `generic' PO jobs, data was drawn from `job specifications', which the Commission had compiled from information provided by Canada Post management, in 1990 and 1991, as well as from job descriptions and job profiles, also furnished by Canada Post.

[380] As with the 1987 evaluations, the Commission used the completed Job Fact Sheet as its primary source of job information for the 93 CR evaluations. For the PO evaluations, the `job specifications' were regarded as the primary source.

[381] It was, therefore, upon these evaluations of 93 CR positions and 10 `generic' PO jobs, which were completed by November 1991, that the Commission based its investigation findings. These, in turn, led to the conclusions of the Commission's Final Investigation Report of January 1992, including the recommendation that the Complaint be sent to the Canadian Human Rights Tribunal for a hearing.

F. Review Of Job Information Collected And Methodology Used: Tribunal Stage

(i) The Professional Team

[382] Early in 1993, with the Tribunal's proceedings well underway, the Alliance engaged a three-person team of professional job evaluators (hereinafter called the `Professional Team') to provide an expert review of the Commission's 1991 evaluations of 93 CR positions and 10 `generic' PO jobs, and to undertake independent evaluations. The Professional Team was comprised of the following persons:

Dr. Bernard Ingster has engaged in a consulting practice in human resources matters since 1967, including job classification and evaluation. During an early part of his career, he served as Director of Services, Hay Associates, Philadelphia, and between 1971 and 1977 he had an independent affiliation with Hay while working with clients. Since 1977, Dr. Ingster has operated as an independent consultant in fields such as organization and job structure analysis and design, compensation systems, performance assessment practices and job evaluation plan development. His clients have ranged from industrial companies to public health facilities to educational institutions and law firms. Dr. Ingster earned his doctorate at Rutgers University after acquiring degrees at LaSalle College (Philadelphia) and Temple University.

Dr. Martin G. Wolf obtained his Bachelor of Science and Master of Science degrees in 1958 and 1959 respectively from the University of North Texas, majoring in Psychology. He earned his Ph.D. in 1964 from Case Western Reserve University, Cleveland, primarily in clinical psychology with a minor in industrial psychology. Dr. Wolf began his career with the IBM Corporation in the field of personnel administration and then spent time in the late 1960's with a management consulting firm dealing with the improvement of employee training programs and human resources policies and procedures. He subsequently spent time as a Management Psychologist analysing knowledge, skills and abilities requirements of positions. In the early 1970's he became self employed as a management consultant in Cleveland specializing in sales training programs, executive searches and computerized tracking systems. Dr. Wolf joined Hay Management Consultants in 1974 serving at their Pittsburgh facility until 1981, and then at their Philadelphia site until 1989. During his Hay career he worked extensively with a variety of clients in developing job evaluation, performance enhancement and compensation systems. His last position at Hay was as Corporate Director, Technology Development. Dr. Wolf founded his own consulting operation in 1989 - MAS Management Advisory Services Inc. and was still active in that organization when he appeared before the Tribunal. Areas of emphasis for his company have been the development and implementation of computer-supported job evaluation, salary administration and performance enhancement systems as well as conducting change management projects. Dr. Wolf appeared before the Tribunal as spokesperson for the Professional Team and testified that he had spent 30 years in job evaluation including about 20 years working with the Hay process. He had served as a "correlator" at Hay - a "keeper of the flame" role in maintaining the integrity of the Hay system. He estimated that he had evaluated "slightly upwards of 10,000" jobs using the Hay process, including office clerical, and payroll systems jobs, and, in his early days, some blue collar jobs. He was qualified by the Tribunal as an expert in Hay-based job evaluation and Hay-based compensation.

Ms. Judith Davidson-Palmer, President, EEO Associates, consultants on equity issues, including organizational development and change, and pay and employment equity issues. From 1982 to 1985, Ms. Davidson-Palmer served as National Director - Management and Organization Development, Canada Post Corporation, at its Head Office. She is a graduate of Mount Allison University and obtained her Master of Arts in Psychology from Queen's University.

[383] The Alliance's evidence indicated that, before the Professional Team came together, Dr. Ingster had spent a week with Alliance representatives reviewing the job content materials that the Commission had used for its evaluations. This included copies of completed Job Fact Sheets, position descriptions and Interview Guides. Dr. Ingster concluded that the available documentation could be used for job evaluation purposes by a committee of professionals.

[384] Dr. Ingster was asked to provide "an expert review of the evaluations of 93 Clerical and Regulatory positions and 10 PO jobs that had been developed by the Canadian Human Rights Commission".¹²¹ The Professional Team, as a whole, was then asked to "apply the Hay Method to the job content in accordance with the `best practices' of senior level Hay consultants considered to be expert in the use of the process".¹²²

(ii) Phases 1 and 2

[385] The Team undertook its task in two phases, as follows:

Phase 1 involved the re-evaluation of the Commission's 1991 sample of 93 CR positions and of the 10 `generic' PO jobs; this was tackled in May/June 1993.

Phase 2 involved the evaluation of a further 101 CR positions which was undertaken in November/December 1994. This number represented the remaining balance from the Commission's original 1987 sample of 194 (194 less 93). Subsequently, 4 positions for which the Professional Team felt there was inadequate data were deleted for a revised total of 97 additional CR positions, and a grand total of 190 CR positions evaluated.

[386] The Professional Team called the job evaluation methodology it employed in undertaking its evaluations the Hay factor comparison approach or "the classic Hay Standard". Dr. Ingster indicated that this was an application of the Hay Method "in strict accord with its factor comparison origins".¹²³ Dr. Wolf defined this as the approach that was originally designed by the Hay organization where one assesses the content of each job against the structure of factors provided in the Hay Plan as know-how, problem solving, accountability and working conditions. One compares progressively each job, factor-by-factor, to the next and subsequent jobs. This methodology has also been referred to as a job-to-job comparison of total job content on a factor-by-factor basis. Since the Team considered the working conditions factor to be the least developed of the Hay Plan factors, it created a more elaborate working conditions guide chart for use in this set of evaluations.

[387] What were the principal sources of job information for the Professional Team's evaluations in each of Phase 1 and Phase 2? The Alliance supplied the Professional Team with the following materials from which to draw information:

Phase 1- 93 CR's (May/June 1993)

the relevant 1986 completed Job Fact Sheets
job descriptions attached to the 1986 Job Fact Sheets
organization charts attached to the 1986 Job Fact Sheets
the relevant 1986 completed Interview Guides
the Commission's Rationale Statements from its 1991 evaluations, which was usually a single-sheet summary listing of the principal duties and features of each position evaluated, factor by factor, and of the evaluators' ratings and scores, and reasoning behind them, also by factor
the Professional Team also had access to their own notes, created during telephone interviews which the Team had made in May 1993 with CR incumbents.

Phase 1 - 10 PO `Generics' (May/June 1993)

`job specifications' compiled by the Commission based on data obtained from Canada Post in 1990 and 1991
job descriptions obtained by the Commission from Canada Post in 1990 and 1991
profiles describing characteristics of a number of PO jobs obtained by the Commission from Canada Post
behavioural dimensions obtained by the Commission from Canada Post
the Commission's Rationale Statements from its 1991 evaluations
a variety of Canada Post manuals, handbooks, forms, and training materials.

Phase 2 - 101 CR's (November/December 1994)

the relevant 1986 completed Job Fact Sheets
job descriptions attached to the 1986 Job Fact Sheets
organization charts attached to the 1986 Job Fact Sheets
the relevant 1986 completed Interview Guides
the Professional Team also used their evaluations of the 93 CR positions in Phase 1 to serve as Reference Positions in evaluating the 101 (eventually 97) CR positions in Phase 2, because of the overlap in position content
the Team also had access to their own notes, created during telephone interviews with incumbents, made by the Team in September 1994.

[388] The primary source of job information for the CR's was the position descriptions. These were accepted as received. The primary source of job information for the `generic' PO jobs was the `job specification' created by the Commission for each of the 10 `generic' PO jobs.

(iii) How were the Job Evaluations conducted by the Professional Team?

[389] Phase 1: The Professional Team began by meeting in Ottawa in May 1993 to conduct telephone interviews with the 93 CR incumbents of Phase 1. All three Team members participated in the interviews by conference call, with one member taking the lead in conducting and preparing notes on the interview. Dr. Ingster had allocated the list of 93 incumbents three-ways, so each Team member took the lead for one-third of the calls.

[390] A major purpose of these telephone interviews was to seek additional information about the work environment of the position occupied by each interviewee. This was done because, in the opinion of the Team, the working conditions factor was the least well-documented aspect of the 1986 Job Fact Sheet and other materials the Team had at hand. A second reason was to enable Team members to ask questions they might have from their earlier scanning of the job materials. Interviewees had been alerted to the calls and had before them copies of the relevant Job Fact Sheet and supporting documents. Each was being asked about his or her position as it existed in 1986.

[391] All CR incumbents reached by telephone were interviewed. For some who could not be reached, employees who had occupied the same position in the past, or employees currently occupying a related position might stand-in for the actual 1986 incumbent. Occasionally, the appropriate supervisor might respond. Actual numbers of such stand-ins are not known. Of the total of 93 possible telephone interviews, however, it is known that 59 were completed.

[392] Following the Phase 1 telephone interviews, the Professional Team met, in May/June 1993, in Philadelphia, to undertake the evaluation of the 93 CR positions and the 10 PO `generic' jobs. Dr. Ingster served as chair. The Team set a target of evaluating between 10 and 11 jobs per day over a 10-day period. Each member of the Team had received the job information materials earlier.

[393] The assembled members began by arranging the job information in ascending order of the total job evaluation points that had been assigned by the Commission's evaluation team in their evaluations of 93 CR positions and 10 `generic' PO jobs in 1991; this information came from the Commission's Rationale Statement. An identifying number from 1 to 103 was then assigned to this resulting order of positions/jobs.

[394] Normally, each job was discussed before starting the evaluation process. This was done to clarify any questions Team members might have had or to highlight a particular aspect of the job concerned. The Team then proceeded to assess the job content and to assign a rating for each Hay factor and, progressively, to compare those ratings on a job-to-job basis. In most cases, there was eventually unanimity of the three members in reaching agreement on individual factor ratings. As a minimum, a consensus of two could prevail but never by just a simple vote, only after discussion.

[395] At the start of each day, the members reviewed their decisions and ratings of the previous day, referring to their respective notes. This was to ensure accuracy in their joint recording of the previous day's work. Dr. Wolf did not make notes, but inputted the factor and sub-factor ratings into the computer, during the rating process.

[396] Phase 2: The three Team members attempted to conduct telephone interviews with the 97 CR incumbents of Phase 2. This was done in a similar manner to that described above for Phase 1. These interview contacts were attempted from Ottawa in September 1994. Of the total of 97 possible telephone interviews, 55 were completed.

[397] Following the Phase 2 telephone interviews, the Team met in November/December 1994, in Philadelphia, to undertake the evaluation of the 97 CR positions. Again, Dr. Ingster served as chair. The Team set a target to evaluate about 10 jobs per day over a 10-day period; it actually took 9 days. The same evaluation process as described above for Phase 1 was followed by the Team, with one difference - the full set of 93 evaluated positions of Phase 1 was used as Reference Positions. A `Reference Position' has important job content characteristics which are useful for comparison with an unevaluated position. Given the similarity in positions content between the Phase 1 and Phase 2 sampled CR positions, the Phase 1 positions made suitable Reference Positions for the Phase 2 evaluations.

[398] It is usual practice for those completing job evaluations to prepare, for record purposes, a statement of the reasoning that went into the Evaluation Plan used, the process followed, and the ratings reached, factor-by-factor. In the case of the Commission, this was achieved through its "Rationale Statement". In the case of the Professional Team, it was by means of an "Audit Trail". An Audit Trail was included in the Professional Team's Report to the Alliance covering its evaluations in Phases 1 and 2.

(iv) Two Additional Reviews

[399] The Professional Team was asked to participate in two additional exercises with possible impact on its earlier evaluations (Phases 1 and 2). The first, which occurred in June 1997, was to review a number of newly found job documents. These documents had been misplaced when the Commission moved its office space; they were found in the spring of 1997. The found documents included items such as several previously missing job descriptions, more legible photocopies, and clarification of French translations. The documentation affected 89 of the 190 CR positions which had been evaluated in Phases 1 and 2.

[400] The question to be answered by the Team was whether the newly found material would have had any effect on the earlier evaluations had it been available when those evaluations were being done. It was addressed by only one Team member, Dr. Wolf. His conclusion was that, with one possible minor exception, nothing significant had been added to the original job information. The new material simply served to confirm the Team's previous evaluation results.

[401] The second exercise was completed after the Alliance, in June 2000, requested that the Professional Team review the evidence of a number of Canada Post witnesses who had appeared before the Tribunal since Dr. Wolf's last testimony. This involved about 4,000 pages of written material, including transcripts for about 70 days of testimony and cross-examination, and supporting documents, manuals and related material regarding job content. These documents dealt with the PO jobs, primarily. The Alliance had requested this review to determine what impact the additional evidence presented by Canada Post might have had on the Team's 1993 and 1994 evaluations of CR positions and PO jobs.

[402] The Professional Team's Report concerning the Respondent's voluminous evidence was entered as Reply evidence. Before the group convened to consider the Respondent's evidence, Dr. Wolf read through the transcripts of testimony from all 70 days and through the supporting material. The Team gave him the task to assess the new material's potential usefulness for job evaluation purposes. He screened on the basis of three criteria - relevance, appropriateness and duplication. This resulted in some 36 days of testimony evidence and associated material on which he directed his two colleagues to focus. The three Team members then met and jointly conducted their review over a 5-day period.

[403] After a careful study of the lengthy screened material, it became clear to the Team that the actual work performed by an incumbent of any one of the 10 `generic' PO jobs could vary widely depending on location. The Team determined that the evidence demonstrated the following:

"the use of a single generic description for each of these 10 jobs results in a document that probably describes accurately few, if any, of the many incumbents of these multi-faceted jobs", and
"all of the 10 PO jobs appear to reflect an amalgam of sub-jobs, some of which might fall at different Hay evaluation levels, based on the actual task mix at various locations".¹²⁴

[404] Faced with this finding, the Team indicated that, in fairness to the job evaluation process, it chose to give the benefit of the doubt to the PO jobs and to assess each based on what appeared to be the highest level of tasks commonly performed by an incumbent of that job. The Team examined all 10 `generic' jobs and compared them against similarly levelled CR positions using the Hay factor comparison approach. Members asked themselves whether, as a result of the additional evidence, they now had a different understanding of the jobs than what they had originally.

[405] The Team found that much of the new information provided by Canada Post's witnesses was not relevant to job evaluation. In particular, the Team concluded that none of its original CR position evaluations were affected by the additional evidence. In fact, Dr. Wolf testified that he did not revisit, during the June 2000 exercise, the 1993/1994 CR evaluations, having accepted them as a given.¹²⁵ The Team did, however, acknowledge that the new evidence confirmed that the range of variation in individual incumbent duties for the 10 `generic' PO jobs was much greater than the Team had originally understood it to be.

[406] Few changes in total evaluation scores resulted from this review. Of the 10 PO jobs, five were completely unchanged and three had changes of three points or less. Two jobs, however, changed significantly, one of which increased and one of which decreased in total evaluation value.

(i) The Standard of Reliability

[407] Having focused on how the job information was collected, processed and used for evaluation purposes, it is now appropriate to consider the reliability of the job information, the methodology and the evaluations performed by both the Commission and the Professional Team.

[408] What standard of reliability should the Tribunal use? While all three parties in this Complaint have agreed that they are not seeking perfection, per se, it is necessary to determine what is an acceptable reliability standard in the context of this particular "pay equity" situation.

[409] The decision of the tribunal in the Treasury Board case, which rules out any absolute standard of correctness, is of assistance in this regard:

What is apparent from these comments and from the nature of the subject is that equal pay for work of equal value is a goal to be striven for which cannot be measured precisely and which ought not to be subjected to any absolute standard of correctness. Moreover, gender neutrality in an absolute sense is probably unattainable in an imperfect world and one should therefore be satisfied with reasonably accurate results based on what is, according to one's good sense, a fair and equitable resolution of any discriminatory differentiation between wages paid to males and wages paid to females for doing work of equal value.¹²⁶

[410] Also, Mr. Justice Evans' decision of October 1999 supports a flexible case-by-case approach to the determination of how the concept of equal pay for work of equal value is to be effected, as follows:

In short, the correct interpretation of section 11 in my opinion is that Parliament intended to confer on the agencies created to administer the Act a margin of appreciation in determining on a case-by-case basis, and with the assistance of technical expertise available, how the statutorily endorsed principle of equal pay for work of equal value is to be given effect in any given employment setting.¹²⁷

[411] Finally, Mr. Justice Hugessen's decision of June 1996 in the Department of National Defence case, reiterates the civil burden of proof required of a complainant as being the balance of probabilities which is "...a long way from certainty...":

The burden which a complainant before a Human Rights Tribunal must carry cannot, in my opinion, be placed any higher than the ordinary civil burden of the balance of probabilities. That is a long way from certainty and simply means that the complainant must show that his position is more likely than not.¹²⁸

[412] These rulings support a call for a standard of reasonableness, there being no such thing as absolute reliability. The application of such a standard will depend very much on the context of the situation under examination. The issue is, then, given all the circumstances of the case before this Tribunal, is it more likely than not that the job information, from its various sources, the evaluation system and the process employed, and the resulting evaluations are, despite any weaknesses, sufficiently adequate to enable a fair and reasonable conclusion to be reached, as to whether or not, under section 11 of the Act, there were differences in wages for work of equal value, between the complainant and comparator employees concerned?

[413] Focusing specifically on the job information and data used in this case, a further means of determining whether they are reasonably reliable is to test them against a generally accepted practice of the job evaluation industry. That is the industry's objective of seeking, to the extent possible, accuracy, consistency and completeness in job information being used for job evaluation purposes. Accuracy calls for the data to be correct. Consistency recognizes the need for the same kind of information and generally the same level of detail and quality across all jobs being evaluated. Completeness relates to ensuring that important information about a job is not missed and that the collected data is compatible with the job evaluation plan being used.

[414] A very important factor affecting the Tribunal's judgement about the reasonable reliability of the job information collected, the methodology employed and the evaluation of jobs/positions, is the evidence provided by a number of expert witnesses. Equally important, is the fact that several of these expert witnesses significantly disagreed with, and even sometimes contradicted, each other.

[415] Under these circumstances, the Tribunal found it helpful to examine the evidence of these particular expert witnesses in a very systematic manner to ensure consistency and fairness of treatment.

[416] In compiling the elements that should comprise the systematic approach, the Tribunal was influenced by two recent Court decisions.

[417] The first is a reference, in a Federal Court of Appeal decision dated April 5, 2004, to a discussion of the notion of witness credibility in the reasons offered by O'Halloran, J.A. in Faryna v. Chorny, [1952] 2 D.L.R. 354 at 356-357 (B.C.C.A.):

...the real test of the truth of the story of a witness in such a case must be its harmony with the preponderance of the probabilities which a practical and informed person would readily recognize as reasonable in that place and in those conditions.¹²⁹

[418] The second is a reference in a Federal Court decision dated April 14, 2004, to a discussion of the duties and responsibilities of an expert witness in the reasons offered by Cresswell, J. in National Justice Compania Riviera S.A. v. Prudential Assurance Co. Ltd. ("the Ikarian Reefer"), [1993] 2 Lloyd's Rep. 68 at 81:

...expert evidence presented to the Court should be, and should be seen to be, an independent product of the expert uninfluenced as to form or content by the exigencies of litigation. The expert witness should provide independent assistance to the Court by way of objective unbiased opinion in relation to matters within his or her expertise.¹³⁰

[419] The systematic approach that the Tribunal applied in examining the evidence of the expert witnesses concerned was as follows:

What is the expert qualified to give evidence about, and what Party is calling the expert witness?
What is the expert's mandate?
What is the extent of the expert's knowledge and experience, and his or her standing in the field of expertise concerned?
How did the expert fulfill his or her mandate?
What conclusions did the expert reach?
How did the expert present his or her conclusions to the Tribunal?
What weight does the Tribunal give to the conclusions of the expert?

[420] It is, therefore, time to test for this standard of reasonableness of reliability with the job evaluation undertakings of the Commission and the Professional Team.

(ii) Commission's 1987 Job Evaluations

[421] While the 1987 CR evaluation results were not used in the Commission's Final Investigation process, it is, nonetheless, particularly pertinent to test the reliability of the 1986 instruments and resulting job information, since much of that information and data were employed in the succeeding evaluations.

[422] The Job Fact Sheet or Questionnaire, while completed in 1986 by only a sampling of CR incumbents, was originally planned for use by both CR and PO employees. It was clearly intended to be the most important source of up-to-date job information. In fact, the Commission chose it as its primary source for its 1987 CR evaluations. Yet, according to Canada Post, the design and content of the Job Fact Sheet were seriously flawed. Canada Post indicated both in its submissions to the Tribunal and to the Commission, at the time, that the Job Fact Sheet was self-evaluative in design, a feature considered quite unacceptable for job evaluation purposes. When it made its objections to the Commission, Canada Post proposed major changes and even offered its own design but the Commission declined both options, and chose to carry on with its own format. Moreover, the Job Fact Sheet was designed to relate to the System One evaluation scheme. This evaluation scheme was neither fully developed nor was its use entirely acceptable, at the time, to the Alliance and to Canada Post.

[423] The Job Fact Sheet did not meet the Commission's subsequent model which stated that a Questionnaire "...must be carefully designed and tested, possibly through a pilot study...".¹³¹ The Job Fact Sheet was designed and developed by a senior Commission employee under pressure of time and human resources constraints, and without professional assistance. The formulation of pertinent questions and their order and style of presentation in a crucial survey document to be presented to employees working in a large, busy and dynamic organization with a long history of sensitive labour relations, such as Canada Post, required appropriate professional expertise. The Job Fact Sheet was not the product of such expertise.

[424] Certainly, the Job Fact Sheet was neither designed nor tested by an independent, professional body. All four persons who `tested' the questionnaire were clerical/secretarial workers. No PO-type workers were involved. The results of the test were checked by the lead investigator, who found that the four involved in the testing of the questionnaire had frequently over-rated their respective jobs on a number of factors. This led to some changes in the document but the revised final version was never re-tested.

[425] Two other instruments, considered important job information sources, were job descriptions and organization charts, both of which were to be attached to the Job Fact Sheet by each CR incumbent who completed it. As Canada Post had warned, many of the job descriptions were out-of-date, some dated well before 1986. Indeed, a number were missing, some were "unofficial", and others bore no signature of approval. Additionally, the Alliance was not satisfied that all the job descriptions had been union-endorsed, an accepted right of the union concerned. An almost similar situation prevailed with regard to the organization charts. They were of varying ages, and some were missing.

[426] The Interview Guide, intended to assist the Commission's investigator in subsequently conducting interviews with employees and to record their responses, elicited reservations about its design and use from both the Alliance and Canada Post. These concerns were never fully resolved. It was, like the Job Fact Sheet, designed around the System One evaluation plan.

[427] Canada Post proposed several major changes in the format of the Interview Guide. The Commission chose not to make the proposed changes.

[428] The Alliance, through one of its representatives, had observed the use of the Interview Guide in four of the initial run of interviews and had, as a result, proposed several modifications to improve it. Only a few of these modifications were accepted by the Commission. This decision to accept changes to the Interview Guide was reached after Commission staff had begun to use the original version. Hence, the changes in the document were made in the midst of its use which is not a recommended action in job evaluation circles.

[429] Furthermore, the Commission's lead investigator testified that the Commission normally used an Interview Guide selectively with employee respondents where it was necessary to clear up inconsistencies or other difficulties with responses provided in particular Job Fact Sheets. In this case, the Commission, for reasons unknown to the lead investigator who was not in its employ at this time in the evaluation process, chose to attempt to interview the entire CR sample.

[430] Another factor which caused some consternation was the origin of the CR sample size. A Commission senior staff member had developed a random stratified sample without prior professional input. His sample calculations went through a number of configurations with extra "alternates", but only 194 acceptable Job Fact Sheet responses were received. This became the number of CR positions that were eventually evaluated. Statistics Canada, who was consulted subsequently by the Commission, recommended that the sample size be augmented. This suggestion was not implemented by the Commission.

[431] Finally, the process followed by the Commission Evaluation Team was somewhat unusual. Three of the Team's members rotated as evaluators. There was little evidence presented concerning the extent of the job evaluation training given to these Commission evaluators, or their individual experience as job evaluators.

(iii) Commission's 1991 Job Evaluations

[432] As already noted, the Commission relied on its 1991 evaluations of 93 CR positions and 10 `generic' PO jobs as the basis for reaching its Final Investigation conclusions, and recommendations. The Commission had stated that it would be using the Hay System with its four-factor approach and its XYZ charts to establish value. The Commission was of the view that this Hay version was "...quite capable of measuring `blue collar' and office-environment jobs".¹³²

[433] Neither Canada Post nor the Alliance were in full harmony with the Commission's use, in 1991, of the Hay Job Evaluation Plan. Canada Post, in particular, felt that the Hay Plan was intended, primarily, for "white collar" and management-type work.

[434] Interestingly, Dr. Wolf, in answering a question from counsel for the Alliance, referred to the evaluation methodology used by the Commission, as follows:

Essentially, what the Human Rights Commission evaluators did was to create their own evaluation methodology ... they explicitly said that they were not using anything like traditional Hay methodology. They said they were using an `equal value' approach to job evaluation ... I don't know what you would call what they did, but it was not the Hay process.¹³³

[435] There was also a certain incompatibility in the use of the Hay Plan with a Job Fact Sheet that had been designed to relate to the System One plan. However, System One, incomplete as it might have been, was said to be closely aligned to the Hay Plan. It had four factors for assessing the value of work, as does the Hay Plan. Both plans were considered by the Alliance to be weaker for PO employees than they should be with respect to the working conditions factor.

[436] As with the 1987 evaluations, the Job Fact Sheet was used by the Commission as its primary source of job data for the 1991 CR evaluations. In contrast, the primary source for the PO evaluations was the `job specifications' compiled by the Commission. The resulting comparison between complainant CR positions occupied by particular incumbents with the comparator `generic' PO jobs can be faulted as not being a proper comparison, given that the generic jobs were not actual positions, with incumbents. This is one example of the inconsistent treatment which the job information imposed upon any evaluation team attempting to evaluate these particular positions/jobs in a "pay equity" context.

[437] The reduction in the CR sample from 194 in 1987 to 93 in 1991 seemed to be prompted by the need to speed up the evaluation process. Calculation of the revised sample of 93 was handled by one of the Commission's evaluators, an obviously well intentioned employee, but not a sampling expert. There is even a suspicion that the sample of 93 may be unfairly weighted. This suspicion arises because all 10 CR benchmarks which were developed at the start of the 1991 evaluations, and all CR positions in Canada Post's Head Office, are understood to have been automatically included in the "sample" of 93. This belies any suggestion that the sample of 93 was randomly selected.

[438] The Commission's 1991 evaluations were not conducted within the confines of the Commission's recommended committee structure. Commission staff evaluated individually, and came together periodically to compare ratings.

[439] Mr. Paul Durber of the Commission, indicated in his evidence to the Tribunal in June 1993 that he thought the evaluation process followed by the Commission in 1991 fell between a minimum and an ideal process, where the ideal would be a joint employer/employee study.¹³⁴ While he was "satisfied with the quality and consistency of the product" of the evaluations, Mr. Durber indicated that the evaluating team was not "...an orthodox committee...". For one thing, its process "...was something of a departure ... [as] normally the Commission had followed - always as far as [he knew], for group cases - had followed a strict committee approach...". By `orthodox', Mr. Durber indicated that he meant a committee structure where all its members come together, with their individual ratings and, by a process of consensus building, jointly reach a mutually acceptable conclusion. This was missing in the Commission's committee where its members worked individually, coming together to exchange ratings.

(iv) Professional Team's 1993/1994 Job Evaluations

[440] The Alliance and the Commission have chosen to rely exclusively on the evaluations performed in 1993/1994 by the Professional Team in attempting to substantiate the Complaint before this Tribunal. A number of changes in evaluation scores made in June 2000 have revised the Team's original evaluation values.

[441] At least two members of the Professional Team were extremely comfortable with the Hay Evaluation Plan, and with using it to conduct the evaluations for this Tribunal hearing. Both Dr. Wolf and Dr. Ingster had spent considerable time in the Hay organization and had worked over many years with clients using the Hay Plan. Moreover, in the interest of thoroughness, the Team applied what it termed the factor comparison, or classic Hay Standard, which it considered to be a more diligent, time-consuming approach than the "fast track" Hay guide chart options.

[442] Dr. Wolf, spokesperson for the Professional Team, and accepted as an expert in Hay-based job evaluation, indicated that it was his opinion that the Hay Plan was generally regarded as being capable of measuring the relative value of male-predominant and female-predominant jobs. Additionally, the Hay Plan's factors measured work value based on a composite of skill, effort, responsibility and working conditions, as required by the Act.

[443] Questions continued to be raised, however, concerning the suitability of the Hay Plan for use with clerical and `blue-collar' jobs, given its current widely known use for evaluation of management-level work. This was certainly a concern of the Alliance, as well as of Canada Post. Dr. Wolf stated that the Hay Plan, in its earliest days, had been used in the evaluation of `blue collar' work but, in more recent times, client demand had been increasingly in the management area. While Dr. Wolf did admit that most of Hay's job evaluation is now with supervisory, management, and professional jobs, he and his two colleagues felt it was quite adaptable to both `blue collar' and clerical jobs.¹³⁵ This opinion was especially true, he indicated, when one was dealing with evaluators experienced in applying Hay. In addition, the expansion of the Hay working conditions factor, by the Team, was illustrative of its adaptability.

[444] While the Hay Plan may be considered an acceptable one for use in these evaluations - particularly in the hands of the well-qualified Professional Team - the sources of much of the data used raise many questions. The Job Fact Sheet, intended, when first designed, to serve as the primary source of up-to-date job information, proved to be entirely unacceptable as noted earlier. Indeed, so unreliable did the Professional Team consider the Job Fact Sheet that the Team largely discarded it and used the job descriptions as its primary document for its CR evaluations.

[445] Perhaps the most telling comment about the Job Fact Sheet was that of Dr. Wolf himself. He testified before the Tribunal in April 1995, as follows:

The job documents that were done by the Canadian Human Rights Commission required an awful lot of interpretation because the form they used was abominable. The guy who developed it probably should be taken out and shot.

It violated the basic rule of any job documentation process which is that you ask individuals to describe the jobs, not evaluate them. The way it was structured ... asking people to self-evaluate. So we disregarded that totally because people can't evaluate their own jobs because they don't understand the process.¹³⁶

[446] In addition to its `abominable' structure, the Job Fact Sheet was designed, as noted earlier, to relate to the System One plan. There would seem to be potential conflict in using job information, even in a less than primary role, provided by means of an instrument based on an evaluation plan different from the one being used for the evaluations. The Commission's model stresses that "... use of a single plan to evaluate all jobs is essential".¹³⁷

[447] Use of the job descriptions as the primary document for the CR evaluations posed its own problem. Dr. Wolf testified that the Team had accepted the job descriptions as they were, provided they were in correct Canada Post format, on the assumption that they had been the basis of Canada Post's classification of those positions.

[448] Unfortunately, there was not a consistently compatible quality set of job descriptions for the Phase 1 and Phase 2 CR evaluations, including no standard format of presentation. As already noted, the job descriptions were of varying ages. Some went back many years. Others were not considered by the Alliance to be "official" versions, even though they were more current in age. A number were missing. Dr. Wolf acknowledged in cross-examination that 14 were missing from Phase 1, and 11 from Phase 2 evaluations. Subsequently, in June 1997, one of the missing CR job descriptions for Phase 1 was found, and three were found for Phase 2. Additionally, 5 available job descriptions were considered to be inconsistent with the Job Fact Sheet, and, therefore, were not used. There were even examples of missing pages from some job descriptions. Some 50-job descriptions were not signed or dated. Finally, the job descriptions did not generally include information on working conditions. These deficiencies raise the issue of the consistency, completeness and accuracy of that aspect of the job data.

[449] These problems seriously damaged the credibility of many of the job descriptions, especially in their role for the Professional Team as the primary source of CR job information. Even the Commission's model recognized that job descriptions are not always an up-to-date and accurate reflection of work done and often replicate prevailing stereotypes in the workplace, and should not be used as the primary source of data.

[450] The matter of sample size of a total job population is a factor which generally requires professional expertise. In this case, a random sample of CR positions should be of sufficient size that it can, with some degree of confidence, be characterized as representative of all the work undertaken by the total CR population. Adding or deleting from the sample size may impact on the average value of work within both the sample and the total population. The same is true of a stratified random sample, an example of which is the Commission's breakdown of the total 194 CR positions into individual samples for each of the CR-2, 3, 4, and 5 levels.

[451] The development in 1986 of the original CR sample size and the subsequent "re-calculations" culminating in the final sample size of 194 was a non-professional process which was further discredited by the reduction to 93 in 1991. The Professional Team did restore that reduced sample almost to its original level by evaluating a further 97 CR positions. Expert witnesses for Canada Post and for the Alliance gave evidence concerning how representative these samples were of the total CR population.

[452] Canada Post called a witness, Dr. David Bellhouse, who first appeared before the Tribunal in January 1996.

[453] Dr. Bellhouse attended as Professor of Statistics at the University of Western Ontario. He was also Chair of the Department of Statistical and Actuarial Sciences, having held that office since July 1992. He obtained bachelors and masters degrees at the University of Manitoba, and earned his doctorate at the University of Waterloo in 1975. His PhD dissertation was entitled "Some results in sampling from a finite population under superpopulation models, 1975". Dr. Bellhouse began his academic career in 1974 as Assistant Professor - Department of Statistics, University of Manitoba. He later joined the University of Western Ontario and moved through the ranks of Assistant Professor, Associate Professor (with tenure) and full professorship. As Chair of his department, he serves as Director of its Statistical Laboratory which undertakes the design and execution of statistical research and surveys for others on a cost-recovery basis. He estimated his research-teaching-administrative workload as being 40-20-40. Dr. Bellhouse has had many papers published in refereed (peer reviewed) journals in his field of expertise. He has identified some 40 such papers involving subjects such as sampling techniques and statistical modelling and analysis. In 1985, Dr. Bellhouse received his University's Gold Medal for Excellence in Teaching. Grants-in-aid-of-Research have been awarded annually, since 1976-77, by the Natural Sciences and Engineering Research Council in support of Dr. Bellhouse's undertakings. He is a Fellow of the American Statistical Association, his Fellowship having been awarded on the basis of his research in survey sampling. He is also an elected member of the International Statistical Institute.

[454] Dr. Bellhouse was recognized by the Tribunal as an expert in statistics, with specialization in survey sampling [survey sampling comprises data collection and data analysis]. He noted that when one wishes to have individual information about all units of a population, one requires a census of all those units because "the technique of random sample selection alone ... cannot provide any information on units which are not in the sample".¹³⁸ Dr. Bellhouse indicated in his Report that, in line with Dr. Wolf's agreement that the purpose of job evaluation is "to try to compensate each position in accordance with its value"¹³⁹, one must use a complete census rather than a sample of any kind, as "job evaluations of any sampled positions do not provide any information on the other positions which have not been sampled".¹⁴⁰

[455] If, however, sampling is to be used, Dr. Bellhouse emphasized the scientific nature of this instrument, with the need for strict guidelines in order to avoid bias and to reach conclusions which would be useful for evaluation purposes. Amongst his conclusions concerning the sampling performed by the Commission, Dr. Bellhouse found that the initial sampling was an appropriate method to discover whether or not the current assignment of positions to the CR classification levels created for use by federal government departments, and then used by Canada Post, conformed to the non-overlapping intervals of points in the Hay Plan.

[456] He went on to indicate that when it became obvious that CR classification levels were comprised of overlapping intervals of Hay Plan points, a census of all positions was necessary so that each and every position (or job) would be given a value through the job evaluation process.

[457] Dr. Bellhouse concluded that, even if it were appropriate for use in evaluation, the Commission's original 1986 sampling design for the CR population was flawed in that it followed employees rather than positions. He considered that this choice of sampling employees rather than positions "led to biases in the survey in the sense that there are positions that you are not able to sample".¹⁴¹ He went on to conclude that these biases were compounded by a level of non-responses in the survey for which corrective follow-up action was not taken.

[458] Dr. Bellhouse's conclusion concerning the PO occupational group was that since there was no "probability sample" of this occupational group, there was no valid sampling estimate of the average job evaluation value for each PO level. Indeed, because the job value by levels was measured differently for the PO and the CR groups - the CR group having been evaluated using positions and the PO group, using job titles - these evaluations cannot be used for comparison of job values between the two groups.

[459] Dr. Bellhouse also concluded that there may be substantial selection bias in the PO sample because it was a selection of job titles rather than actual positions, and also because the job descriptions represented generic rather than actual job content. He cited the following example to illustrate his point:

there is the possibility of biases creeping into what is the actual value say for a letter carrier, because it is a generic job description and might not catch the variability in job value that is present in the general population.¹⁴²

[460] The Commission called a witness, Dr. John Kervin, who first appeared before the Tribunal in January 2002.

[461] Dr. Kervin attended in his capacity as Professor in the Department of Sociology and as a researcher at the Centre for Industrial Relations, both of the University of Toronto. He obtained his B.A. in Sociology from the University of British Columbia and earned his doctorate at John Hopkins University in Baltimore in 1972. His PhD thesis was entitled "An Information-Combining Model for the Formation of Performance Expectations in Small Groups". Dr. Kervin began his academic career in 1971 as Assistant Professor at the University of Toronto and became Associate Professor (tenured) in 1976. He was cross appointed to the Centre for Industrial Relations in 1977 where his research work is largely carried out. Dr. Kervin indicated that an important part of his research and teaching academic life has been with respect to methodology statistics and data analysis and the effect of gender on social interaction. He has undertaken research projects such as "Measuring Gender Bias in Wages" and has had a number of refereed articles published including one entitled "Where's the Bias?: Sources and Types of Gender Bias in Job Evaluation". Clients seeking Dr. Kervin's research services have included the Management Board of the Ontario government and Treasury Board Canada, some of which involved statistical data analysis in a "pay equity" context. An example of Dr. Kervin working jointly, with another party, in this case with Dr. Nan Weiner, is "Report on Possible Gender Bias in the Bank of Montreal's Hay-Points Compensation System - 1999". Dr. Kervin is a member of the Canadian Sociology and Anthropology Association and of the American Sociological Association. He is also a member of three industrial relations associations.

[462] The evidence of Dr. Kervin, presented as a witness for the Commission and qualified as an expert in data collection and data analysis (with `data analysis' including the use of statistics and statistical methodology), generally contradicts that of Dr. Bellhouse. Dr. Kervin's main criticism of Dr. Bellhouse's Report is that it is predicated upon an incorrect foundation. Whereas Dr. Bellhouse's Report stresses the analytical nature of statistical analysis and the need for scientific reliability, Dr. Kervin indicated that, when dealing with sociological phenomena, one must place the phenomena in the context of its social culture. In order to do this, the key component to data collection concerning the phenomena being studied is the initial formulation of the right question. He stated in his Report that Dr. Bellhouse never did ask the pertinent question, given the context of the Complaint.¹⁴³ Dr. Bellhouse's concern was that the data analysis, using the data collected, follow the scientific method. That the analysis employ procedures and empirical evidence and tests that are replicable appeared to be his main concern. He grounded this concern in a belief that "pay equity" stands for the philosophy that everyone should be paid according to what his or her job is worth.

[463] This is not, according to Dr. Kervin, correct. He felt that the real "pay equity" issue is "is there a gender-based wage gap, controlling for the value of the work?"¹⁴⁴ This, Dr. Kervin's Report indicates, is because the concept of "pay equity" assumes that there is a gender-based bias in wages. What one wishes to do when testing for that assumption is to identify the jobs, the gender composition of those jobs, the job values, and the wages.

[464] Dr. Kervin stated that he would want to look at the jobs both quantitatively as well as qualitatively, and with the judgement that comes from the "art" aspect of dealing with sociological questions. Dr. Bellhouse's Report did not take into consideration the "art" demands as well as the "science" demands of the Complaint. This difficulty, combined with his insistence that there be a correlation between the wages and classification levels, should negate Dr. Bellhouse's reported concerns about the Commission's sampling process, according to Dr. Kervin. The latter concern about the correlation between wages and classification levels indicates, according to Dr. Kervin, that Dr. Bellhouse does not understand the systemic foundations of the "pay equity" issue.

[465] Dr. Kervin stated in his Report that Dr. Bellhouse's conclusions address, principally, issues of sampling and job value measurement. Dr. Kervin generally disagrees with these conclusions.

[466] Specifically, Dr. Kervin was of the view that there is no need for a census of the total population concerned to meet the objective of "pay equity". A representative sample is more than adequate. He also argued that Dr. Bellhouse's rationale for use of a census - the overlap of Hay job points across CR classification levels - is invalid because the overlap is not due to sampling. It is due solely to the use of a different measure of the value of jobs. The overlap remains regardless of the type of sampling.

[467] Dr. Kervin found invalid Dr. Bellhouse's argument that the 1986 sample design was flawed because it followed employees rather than positions and thereby led to biases and higher rates of non-response. Dr. Kervin stated that he discovered no evidence to support this position.

[468] Dr. Kervin agreed with Dr. Bellhouse that the PO sample was not a probability or random sample, but rather a judgement sample. Unlike Dr. Bellhouse, however, Dr. Kervin argued that it was likely to be reasonably accurate. He felt that this was one of the examples where he and Dr. Bellhouse were pursuing different "research questions" - Dr. Bellhouse searching for statistical accuracy and significance, and Dr. Kervin addressing the needs of a "pay equity" situation.

[469] Dr. Kervin disagreed with Dr. Bellhouse that there may be substantial selection bias in the PO sample. He argued that Dr. Bellhouse devoted no discussion to the manner of selection or to the possibility of selection bias due to the use of job titles and generic descriptions.

[470] Finally, with respect to the measurement of job value at the level of job titles for the PO's and positions for the CR's, Dr. Kervin classified this as a difference in the unit of analysis and not as a difference in measurement. Dr. Kervin further stated that it is a situation that can be easily remedied.

[471] Clearly, there is an appreciable difference of opinion, if not a contradiction, between these two expert witnesses on the issues of sampling and job value measurement.

[472] The `job specifications' used for the 10 `generic' PO jobs were, in some ways, similar to the CR Job Fact Sheets. The information accumulated for them was provided by Canada Post, which also provided a number of PO job descriptions based on the job titles of the 10 "generics". Some of these job descriptions were "unofficial", not having had the endorsement of the relevant unions. While generally containing more up-to-date information than the CR Job Fact Sheets, the `job specifications' do not represent specific incumbent-held positions, but rather are an amalgam of functions for 10 commonly held job types. This results in a less than equal and compatible comparison between complainant and comparator jobs. Added to this is the fact that the gathering of the data was undertaken at different times - 1986 in the case of the CR's, and 1990-1991 in the case of the PO's. Such a time difference is usually considered unacceptable in a job evaluation exercise.

[473] Canada Post argued, additionally, that the 10 `generic' PO jobs had been undervalued because certain aspects of "sub-jobs", such as the rotational work of the PO-4's, had been excluded during the Professional Team's evaluation process.

[474] Although evaluation of the `generic' PO jobs was not dependent on the Job Fact Sheet, as the CR evaluations were originally meant to be, it relied heavily on the `job specifications' developed by the Commission. As already noted, these specifications were created, based on a variety of data acquired through a series of meetings with Canada Post management personnel. While appreciating why the Commission took this particular route, evidence before the Tribunal indicates that most job evaluation experts would not regard this information gathering methodology as a propitious approach. According to Dr. Pat Armstrong, an expert witness for the Commission, managers are generally too far removed from operational work to know it in the depth required for evaluation purposes. Contact with individual job incumbents is the favoured route. In fact, the job profiles made available by Canada Post for use by the Commission in developing its 10 PO `job specifications' posed their own difficulty because they were labelled "draft" and did not have union approval.

[475] Because of the unusual nature of the information gathering techniques used by the Commission, the Professional Team had an over-abundance of job information about the PO community. In addition to the `job specifications', certain job descriptions, job profiles and behavioural dimensions were available. They also had access to various Canada Post manuals, handbooks, and other materials. When the evaluation work was done by the Team in June 2000, it had access to the very considerable evidence and supporting material of the Canada Post witnesses who testified largely about the functions and activities of PO workers. Although the Team's examination of all this documentation at that time led to few changes in their original evaluations, it did serve to fortify the PO job data already in hand, even if only in the configuration of the 10 `generic' PO jobs.

[476] The 10 `generic' PO jobs do not represent any of the many jobs in the PO supervisor sub-group. Yet, the samples of CR incumbents do include some supervisors at the CR-5 level. This raises questions of consistency and completeness between the complainant and comparator groups.

[477] Items such as organization charts and the Interview Guide were regarded as secondary and tertiary sources of CR job data. Difficulties with these two instruments have already been identified - dated and missing versions of the charts, and the dissatisfaction of both Canada Post and the Alliance primarily about the content and proposed use of the Interview Guide. In fact, Dr. Wolf testified that the Interview Guide did not add anything in particular to the Team's understanding of the CR incumbents' job duties.

[478] Having access to the Commission's Rationale Statement brought criticism from Canada Post and its three key expert witnesses whose principal testimony is considered later in this section. Their concern related to the Professional Team's use of the Rationale Statement in preparing its primary listing of CR positions and PO jobs that it would evaluate. The Team's list was based on the Commission's total evaluation point scores in ascending order, which were identified in the Rationale Statement. The experts appearing before the Tribunal for Canada Post considered this to be unacceptable for an evaluation process, and felt that such use could have unfairly influenced eventual evaluation ratings.

[479] The telephone interviews with CR incumbents conducted in advance of the actual evaluations undertaken by the Professional Team in Phases 1 and 2 were a well-intentioned exercise designed to improve the Team's job knowledge. Dr. Wolf acknowledged that the interviews focused, primarily, on working conditions and, with one or two exceptions, did not add anything significant beyond that aspect. Also, completed interviews of 63% in Phase 1 and 57% in Phase 2 raise questions about whether the positions of the significant number of incumbents who did not participate, for whatever reasons, were disadvantaged in some way. Given the importance of consistency and completeness, and even fairness of treatment, there is, therefore, probably some limitation to the full benefit of this additional information.

[480] The process by which the Professional Team undertook the individual job evaluations was decidedly superior to that of the Commission in either its 1987 or 1991 evaluations. The Team operated as an entity, jointly reaching its decisions either unanimously or by consensus. This is the committee methodology recommended by evaluation experts, and follows the model presented in the Commission's booklet.

[481] The Professional Team's process did, however, have a number of weaknesses. For example, only one of the three members had ever been in a Canada Post facility or was familiar with postal operations. It did not, of course, have representation from the relevant organizational levels of the employer. Given the circumstances of this particular case, the Team was operating as an outside contracted body, without any contact with the employer and little or no contact with its employees. It had little background or direct knowledge of the nature, history and dynamic of the organization involved, except what the third member, having some Canada Post experience, could provide.

[482] The Professional Team's Audit Trail record led to difficulty in tracking precisely how Dr. Wolf and his colleagues had used the Hay Standard, and the rationale for their choices and evaluation ratings. In cross-examination, Dr. Wolf admitted that their Audit Trail did not reflect all their reasoning behind the ratings. He said it was "rudimentary" and "a rough outline as opposed to detailed". He went on to admit:

...if you are suggesting we did a sloppy job of putting down an audit trail, I will concede that.¹⁴⁵

(v) The Professional Team's Position re: its 1993/1994 Job Evaluations

[483] Dr. Wolf acknowledged that he and his two colleagues found many shortcomings in the available job information and data. Indeed, as already noted, he went so far as to label the Job Fact Sheet, originally intended to be the primary source instrument for obtaining up-to-date information on both CR and PO positions, as "abominable".

[484] With respect to his Team's job understanding based upon the materials before it, he testified as follows:

...I would have to say, with the exception of the four jobs which we passed (sic), that our understanding was adequate but not necessarily ideal...¹⁴⁶

[485] What is the meaning of "adequate"? The Oxford Concise Dictionary defines "adequate" as "sufficient, satisfactory; barely sufficient". Webster's Dictionary defines it as "enough for what is required; sufficient; suitable". In turn, "sufficient" is defined by Oxford as "sufficing, adequate, enough" and by Webster as "as much as is needed, enough, adequate".

[486] Based on these definitions, Dr. Wolf and his Team must have felt that the job information before them was enough for what they required to undertake the job evaluations. It was sufficient.

[487] With respect to the Hay Standard, and how the Professional Team employed it in their evaluations, Dr. Wolf testified, as follows:

Q. How does the standard here compare with commercial standards?

A. As I think I have indicated, we took a more rigorous approach or more exacting than we would normally be. So it certainly at least meets, and in my opinion probably exceeds, the typical commercial standard, if you will, what consultants from Hay or other consulting firms are doing for their clients.¹⁴⁷

(vi) Canada Post's Position re: the Commission's 1991 and the Professional Team's 1993/1994 Job Evaluations

[488] Canada Post's stance on the reliability of the Professional Team's work is clear. It submitted that the job information and data are not sufficiently reliable to substantiate the Complaint, and, specifically, cannot be relied upon to determine whether or not there is a gender-based wage gap within the meaning of section 11 of the Act. Canada Post also takes the position that the Hay System is not appropriate for "pay equity" evaluations involving `blue collar' and clerical workers, and that the process undertaken by the Professional Team was faulty.

[489] In support of its position, Canada Post called the following three expert witnesses, listed in order of their appearance before this Tribunal (the first two of whom provided testimony on both the Commission's 1991 evaluations and the Professional Team's 1993/1994 evaluations):

Ms. Nadine Winter first appeared before the Tribunal in April 1996 as President of N. Winter Consulting Inc., a firm specializing in job evaluation, pay research and compensation management. Prior to establishing her own company in 1989, Ms. Winter was with Hay Management Consultants Canada Ltd. from 1982 to 1988. In her role as Director of Equal Employment Programs, she advised Hay consultants and clients and was involved in the modification of the Hay System to comply with requirements for gender neutrality. She identified one of her accomplishments as the implementation of the Hay System, in a "pay equity" context, in the Government of Manitoba. She became a partner with Hay Canada in 1987, with the title Director of Employment Equity Practices. She was qualified by the Tribunal as an expert in job evaluation and compensation management, including consulting expertise in pay equity and equal pay for work of equal value.
Mr. Norman D. Willis first appeared before the Tribunal in May 1996. At that time, he had been retired for two years. He started his job evaluation career with Hay & Associates in 1968, in the United States. By 1971, he had formed his own company, specializing in management training and human resources studies. He developed his own job evaluation plan, which was conceptually similar to the Hay System, in 1974. That plan has evolved since then, incorporating changes to meet client requirements. His initial focus was clients in the Seattle area. His first Canadian job evaluation proposal began with a presentation on equal pay to the Government of the Yukon in 1985. Subsequently, he handled job evaluation studies in Prince Edward Island, the Northwest Territories, Alberta and Manitoba. Willis & Associates was hired by the Joint Union-Management Initiative Committee (JUMI), a combined Canadian government and public service unions "pay equity" study, to assist the Committee in its work. Eventually, the Committee decided to use the Willis Evaluation Plan provided it could be changed to meet the criteria of the enabling legislation, the Canadian Human Rights Act, section 11. Those changes were made. Later, Mr. Willis attended as an expert witness before the Treasury Board tribunal which heard the union complaint, brought under section 11 after the JUMI Study had broken down. He was qualified as an expert in pay equity and in job evaluation by this Tribunal.
Mr. P.G. Wallace first appeared before the Tribunal in June 2002. At that time, he was Senior Vice-President of Aon Consulting Inc., an organization offering consulting services on the management of compensation practices. He has had considerable experience in job evaluation and compensation design, having participated in the introduction of the Hay System at the Bank of Montreal in the 1970's. He also managed the Hay job evaluation process corporately for Shell Canada Ltd. and integrated it with Hay worldwide for the parent company, Royal Dutch Shell. In his current role, he consults with a wide range of companies in designing, implementing and administering various job evaluation programs. He was qualified by this Tribunal as an expert in job evaluation.

a) Ms. Winter's Testimony

[490] After examining the Commission's 1991 job evaluation work, Ms. Winter reached a number of conclusions, the principal ones being the following:

in adopting the Hay XYZ Plan, the Commission chose a method that fails to measure, accurately and completely, all aspects of the work found in clerical and blue collar positions (while also noting that neither the Alliance nor Canada Post had approved use of the Hay Plan);
the majority of CR positions were evaluated individually by the raters and not as a committee; in the case of the 10 `generic' PO jobs, all but one were evaluated by at least two raters, resulting in a different rating process between the CR's and the PO's;
the Commission's rater, who evaluated the largest number of CR positions, had no previous job evaluation experience;
the Commission's raters, as a whole, had inadequate knowledge of the CR and PO positions;
the data collection tools for the CR's and the PO's were incapable of collecting accurate, consistent and complete descriptions of the work concerned;
in particular, the Job Fact Sheet could not generate accurate, consistent, and complete position information; CR employees were asked to evaluate their own positions rather than provide factual position information; instructions and guidelines presented to the respondents were inadequate and confusing;
because the Job Fact Sheet was based on System One, information not relevant to that system was not collected; this meant that information in areas such as human relations skills and working conditions was lost;
the intended purpose of the Interview Guide was unclear and there was apparently no common set of guidelines and definitions available to assist interviewers; the same deficiencies found in the Job Fact Sheet were replicated in the Interview Guide; several changes were made in the Interview Guide's design after the interview process had begun; sometimes, there were conflicts or differences within position descriptions but these were not clarified by the interviews;
the PO position information collected was incomplete and did not reflect actual positions; the uniqueness and variations of individual positions was not recognized; job rotation requirements at the PO-4 level were not acknowledged; the `job specifications' for the PO group were a subjective compilation by one person, with no confirmation by employees or on-site observation;
there is no indication that the 1991 CR and PO Benchmarks were representative of the full range of positions involved; the quality of the Benchmark information was seriously deficient; how the Benchmarks were used in guiding the evaluation of other positions seems to be unclear, even, according to his evidence, to the head investigator;
the sample group of 93 CR positions and the 10 `generic' PO jobs do not make visible all the work of the total population of CR's and PO's;
the Commission may have biased the evaluation process by initially evaluating CR positions in the order in which they were filed, by CR classification level;
the Commission's Rationale Statements fail to provide adequate justification to explain and defend the ratings;
the Commission failed to check, formally and systematically, the consistency and correctness of the evaluations; many of the formal `sore thumb' reviews, that are integral to the Hay System, do not appear to have been applied.

[491] In summary, Ms. Winter concluded that "the process to collect position information was seriously flawed..." and that "the Commission did not apply a disciplined, rigorous or defensible process to determine the value of the CR positions and PO jobs which they rated".¹⁴⁸

[492] Ms. Winter's principal conclusions, arising from her examination of the job evaluation work of the Professional Team, included the following:

she expressed the same concern as she had expressed about the Commission's 1991 evaluations using the Hay Plan; she claimed that this plan fails to measure, accurately and completely, the work found in clerical and `blue collar' positions;
the Professional Team did not follow the standard application of the Hay Guide Chart-Profile Method; it relied on job-to-job comparisons rather than on job-to-Guide Chart definition comparisons;
by beginning to order positions according to their classification levels, and by the Commission's ratings, the Team may have biased the evaluation process altogether;
a representative and consistent set of benchmark evaluations was not established at the start of the evaluation process;
both Canadian and U.S. Hay job description standards for salary survey purposes were used in the evaluations, creating a consistency problem;
Phase 1 positions were not used as formal Benchmarks, nor were they selected as reference positions in any rigorous and systematic manner;
some positions, with few job characteristics in common with other positions or with deficient supporting documentation, were used as reference positions;
all three evaluators lacked experience in evaluating `blue collar' and clerical positions with the Hay Guide Chart-Profile method;
there were serious deficiencies with the position descriptions primarily in terms of age, accuracy and official status and the evaluators' knowledge of the positions was extremely limited;
the Team failed to administer a disciplined, formal `sore thumb' review of evaluation results;
several other issues were cited, including the time lapse between Phases 1 and 2 evaluations, the poor audit trail record and a flawed definition of `magnitude' in the Accountability factor.

[493] In summary, Ms. Winter concluded that the Professional Team diverged significantly from the standard application of the Hay Guide Chart-Profile method of job evaluation and accepted significant deficiencies in position content. She also concluded that "the quality of the position information for both the CR and PO positions was not adequate for evaluation purposes".¹⁴⁹

[494] Ms. Winter also undertook a comparison of the Commission's (1991) and the Professional Team's (1993) CR evaluation rating results. Both sets of raters evaluated the same 93 CR positions with almost the same data and with essentially the same Hay Plan. Under these circumstances, Ms. Winter indicated that respective ratings should be very similar. If they are not, the results cannot be accepted as reliable.

[495] Ms. Winter concluded that the hierarchies of value, reflected in the two sets of evaluation results "...are clearly inconsistent with each other. Given the inconsistency in results, the Tribunal cannot rely on either set of rating results". Apart from the different evaluation `discipline' adopted by the two groups of raters, she believed that there were three explanations for the number and size of the inconsistencies of the two results. First, the Hay method is not an appropriate tool to measure the value of CR and PO positions. Second, the application of the Hay method requires a consistent set of Benchmark evaluations to guide subsequent evaluations. No Benchmarks were used by the Professional Team and the ones used by the Commission were inadequate. Third, the lack of quality position information made it impossible to arrive at consistent and meaningful evaluation results.¹⁵⁰

b) Mr. Willis' Testimony

[496] Mr. Willis summarized the conclusion of his examination of the Commission's 1991 evaluations, as follows:

Compared to the disciplined approach required in the conduct of a sound Pay Equity study, the CHRC's effort was a poorly designed overall plan and a casually implemented process using data that could not form the basis of acceptable evaluations. It would not be possible to depend on the evaluations by this group for a viable Pay Equity result.¹⁵¹

[497] More specifically, Mr. Willis concluded the following:

the Commission allowed the Alliance to `cherry pick' the male comparator jobs by not including in its investigation any jobs not mentioned in the Complaint; the GS and GL&T group of employees was ignored;
the Hay Guide Chart-Profile system used should have been satisfactory as an evaluation instrument, provided the evaluators received adequate training; the Working Conditions factor which was apparently developed by Hay Canada provided for a wide range of points within each level or each sub-factor, thereby making it extremely difficult to evaluate consistently; there was evidence of mis-application of the Human Relations Skills sub-factor on a number of occasions; the evaluators discarded the Profile step (a means of checking the inter-factor relationships) in the evaluation process because the evaluators "were getting some rather strange profiles"¹⁵²; the wording in the Accountability, Magnitude, and Impact sub-factors was arbitrarily changed by the evaluators during the evaluations, which could have modified ratings and caused inconsistency;
Benchmark job evaluations are appropriate for this case but should be representative of the jobs within the total group being evaluated and commonly understood by the evaluators - this was not done by the Commission; the Benchmarks should have been evaluated by the full Committee and not by two evaluators, independently;
the process of selecting position/job samples fell short of meeting the stringent sampling requirement needed for a "pay equity" project; the sampled jobs should represent a full range of the depth and breadth of the organization; they did not in this case; additionally, the reduction of the 194 to 93 samples of CR's could not be considered sound or objective;
the quality of job information utilized was unacceptable and could not be expected to produce a fair, equitable evaluation result; "...the Job Fact Sheet was hopelessly inadequate for Pay Equity evaluation purposes"; the job descriptions "...were only of minimal usefulness ... in support of the Job Fact Sheet..."; "...a 14-page Interview Guide, incorporated into the information provided to evaluators was more of a liability than an asset..."; the job data were collected at two different times, 1986 for the CR's and 1990/91 for the PO's;¹⁵³
while the PO `job specifications' came closer to providing factual job information, they were based on management and not employee-supplied data, and were acquired by a totally different process from that used for the CR's;
the Commission's sore-thumbing step was not a proper one - it should be a tightly structured, group-led comparison of the evaluations concerned, sub-factor by sub-factor and factor point level by factor point level.

[498] Mr. Willis summarized the conclusions of his examination of the Professional Team's 1993/1994 evaluations as follows:

the Team "...misused the Hay evaluation plan by employing it as a factor comparison system";
the Team used essentially the same inadequate job content information as was used by the Commission evaluators;
the Team was "...ill-equipped to successfully complete their charge due to their lack of knowledge and backgrounds to undertake the assignment, and due to an approach to the evaluations that lacked the necessary discipline. The evaluation process utilized was unacceptable considering what is needed for a successful Pay Equity Study".¹⁵⁴

[499] Mr. Willis' final statement was "In my considered judgement, the efforts of the consultants retained by PSAC cannot be relied on for accurate Pay Equity evaluation results".¹⁵⁵

[500] With more specificity, Mr. Willis concluded the following:

he disputed the wisdom of using the factor comparison approach for the Hay Plan in a "pay equity" case and favoured the point-factor method of the Hay Guide Chart-Profile approach to job value measurement; this preference was based on his opinion that the point-factor method is more suitable for evaluating a wide variety of jobs in a "pay equity" context; further, he was of the view that the factor comparison approach is more acceptable in a traditional single occupational group of jobs;
the Professional Team had the same inadequate job information used by the Commission, including the Job Fact Sheet, Interview Guide, job descriptions, and `job specifications'; there were missing CR job descriptions and those of the PO's were "unofficial";
he agreed with Dr. Wolf's opinion that the telephone interviews conducted by the Professional Team, "...with one or two exceptions, outside of the working conditions arena ... did not add anything significant"¹⁵⁶;
knowledge of each job's classification level was available to the three Team members and this knowledge could have affected their perception of the jobs and, consequently, their evaluations;
use of the "black box analysis" approach to job evaluation could not be expected to provide a useful assessment of what was contained within the black box, given the poor quality of job information input;
it was a "highly questionable practice" to use the Hay U.S. job description standard, particularly for a "pay equity" case, as this standard was intended for survey comparisons and it cannot be assumed that U.S. and Canadian job titles, even if the same or similar, are the same in job content.

c) Mr. Wallace's Testimony

[501] Mr. Wallace indicated that his firm had been asked by Canada Post to review and comment on the process undertaken by the Professional Team in 2000 in addressing the possible impact on its 1993/1994 evaluations, of the considerable additional evidence that had arisen from a number of Canada Post witnesses. Mr. Wallace stated that he used what he termed "...standard criteria against which job evaluation exercises within companies are measured".¹⁵⁷ What he wanted to know was if the Team's job evaluation results were accurate, consistent, and credible.

[502] Mr. Wallace further indicated that "in conducting this review the goal is to determine whether or not the process and discipline employed by the Committee [Professional Team] was capable of yielding results meeting these criteria".¹⁵⁸

[503] The principal conclusions reached by Mr. Wallace included the following:

the processes followed and particular actions taken by the Professional Team fell short of many of what he considered to be industry standard practices for the evaluation of jobs, having the effect of significantly compromising the accuracy, consistency and credibility of the results;
"the lack of complete and consistent documentation on what is being measured by the job evaluation process ... and the failure of the Committee [Professional Team] to apply the safeguards of the Hay system, directly affects the consistency of the evaluation results"¹⁵⁹;
the lack of discipline and rigour in the Professional Team's process should call into question the credibility of the evaluation results.

[504] More particularly, Mr. Wallace concluded the following:

the Hay Guide Chart-Profile method, while an excellent job evaluation tool, was inappropriate for clerical and production/operations jobs;
the Hay factor comparison approach is dated and not in keeping with either Hay training materials or Mr. Wallace's 30 years of experience with the Hay method;
the factor comparison approach is best suited for jobs that are similar in nature; "it is difficult to impossible to create an accurate ranking of dissimilar jobs through total reliance upon the factor comparison methodology without linking the evaluations to the guide charts"¹⁶⁰;
the Professional Team, while including recognized experts in the use of Hay methodology, had limited experience in postal operations; the potential for bias was high, given the Team was mandated by two of the parties in the case;
while the Professional Team met for five days to deliberate over the impact of the additional evidence on its original evaluations, there was no apparent structure to the process followed; Dr. Wolf had no notes with him from his review of the material and no summary or analysis of the job data they had been considering;
the initial screening of the additional material by a single member of the Team, Dr. Wolf, raises questions, particularly given the absence of documentation of what items were considered irrelevant or inappropriate; nor is there any documentation of the selected additional testimony;
the job descriptions were incomplete and inconsistent and the additional information provided was not analyzed and documented appropriately; it is essential that an evaluating group be entirely satisfied with its job descriptions;
Mr. Wallace disagreed with the Professional Team's decision "...to evaluate PO jobs only if they determined there was a just noticeable difference between their recollection of their original job understanding and their new job understanding"¹⁶¹; the Team should have justified its "just noticeable difference" judgement by use of the Hay Guide Charts and re-evaluated each job incorporating all the additional data;
the weakness of the Audit Trail of the 1993/1994 evaluation process compromised the process of 2000 as there was no record of solid evaluations in the earlier period that could have served as reference positions for the 2000 evaluations, or of earlier controversial decisions that might have been altered by the additional material.

(vii) The Alliance's Position h3:

a) Ms. Winter's Testimony

[505] The Alliance's submissions focus on three features relating to Ms. Winter and her evidence, as follows:

her knowledge of Hay, and the Hay Job Evaluation Plan,
her credibility,
her Reports (Exhibits R-235, R-249, R-253, R-254 and R-278).

[506] It is the Alliance's submission that, as the cornerstone of Canada Post's attack on the Commission's investigation and evaluations, and on the Professional Team's evaluations, Ms. Winter's evidence fails to meet the requirements demanded of an expert witness and was so lacking in credibility that it should be given no weight.

[507] Ms. Winter joined the Hay organization in 1982 and began practical training as a job evaluation consultant in 1985, but did not regard herself as a full-fledged Hay evaluation consultant until early 1986. She was involved in the practice of job evaluation for some 2½ years while assuming other Hay responsibilities not directly related to job evaluation. Canada Post did not seek to qualify her as an expert in the Hay System of job evaluation.

[508] The Alliance cites Ms. Winter's involvement, while still with the Hay Canada Company, in a legal action which Hay had brought against Norman Willis for what she termed a "violation of intellectual property laws". There were a series of mis-statements and revisions made by Ms. Winter as she gave evidence on this issue. The Alliance argued that these resulted from her desire to avoid acknowledging the strong and obvious similarities between the Willis and the Hay job evaluation plans. Given that the Willis Plan had been used, successfully, in several "pay equity" applications, including clerical and `blue collar' work, the Alliance asserted that Ms. Winter was attempting to mislead the Tribunal regarding those similarities.

[509] The Alliance questioned Ms. Winter's credibility further, citing her failure to refer the Tribunal to the post JUMI tribunal decision (the Treasury Board case) which contradicted her opinion concerning the standard of reliability required for job information.¹⁶² She had called for a standard of correctness, as was demanded by her interpretation of the Ontario Haldimand-Norfolk decision.¹⁶³ The Treasury Board tribunal decision was based on a standard of reasonableness. She claimed, when being cross-examined on this point, that she did not know what "reasonableness" meant in the context of "pay equity" evaluations.

[510] Additional instances of Ms. Winter's lack of credibility were cited by the Alliance, as follows:

she omitted key material when giving her opinion concerning the quality of job information;
she had a tendency to highlight the negative, and to ignore the positive when giving her opinions, rather than answering questions directly;
her responses to questions posed were often inclined to be argumentative;
as her evidence continued for many days, her inclination to revise points made on the previous day, often at considerable length, became more and more evident.

[511] In summary, the Alliance argued that, when viewed as a whole, Ms. Winter's expert evidence was biased towards the Respondent's position, and fell far short of meeting the standards, including independence, required of a credible expert witness. Therefore, the Alliance argued, Ms. Winter's evidence should be given no weight.

[512] Most particularly, the Alliance's submissions concerning Ms. Winter's expert evidence focus on her opinion about the job data collection tools, the process and quality of the position information, and the Professional Team's job evaluation process. This opinion was presented in her viva voce evidence, and in Chapters 4 and 6 of her Reports (Exhibits R-235 and R-249).

[513] Ms. Winter's Report contends that the "quality of the position information for both the CR and the PO positions was not adequate for evaluation purposes".¹⁶⁴ She illustrated her contention by examining two CR positions which served as Benchmarks. After having reviewed all the pertinent documentation concerning these positions, she held that she had had difficulty understanding what the jobs were all about. One job, in particular, gave her great difficulty. In cross-examination of Ms. Winter, the Alliance demonstrated that the relevant supervisor for this job had identified another position that was essentially identical. Ms. Winter had used, as an example of a Benchmark which had such poor documentation that she could not understand what it was all about, the Benchmark position which referenced the nearly identical position. Ms. Winter, had, however, failed to refer to that nearly identical position. Had she done so, she would have enlightened her understanding of what she considered to be a difficult Benchmark position; the Professional Team had readily recognized the cross-over between the two positions.

[514] In the opinion of the Alliance, Ms. Winter's evidence, in cross-examination, concerning her position as co-chair of the Hay Job Evaluation Process in the Manitoba Pay Equity Study illustrated her ability to integrate information found in different incumbent responses in order to obtain acceptable job understanding. This ability stands in sharp contrast, argued the Alliance, to Ms. Winter's approach to job understanding in the Complaint before this Tribunal. She often offered her opinions in isolation, and made little or no effort to integrate information available to her from several arenas.

[515] Further examples of categorical or exaggerated opinions presented in her Reports, and in examination-in-chief, which were moderated in cross-examination, were cited by the Alliance.

[516] Notwithstanding modifications and explanations made during cross-examination, Ms. Winter stood by her reported conclusions. More particularly, she stated that the Professional Team's evaluation results "cannot be considered reliable, accurate or reflective of the work performed" because the job information used was significantly deficient and the Professional Team's evaluation process deviated significantly from the "standard" application of Hay as she understood it.¹⁶⁵

[517] This latter opinion was based on Ms. Winter's refusal to acknowledge that Hay could be applied, reliably, by means of a factor comparison approach, as was used by the Professional Team. The Alliance argued that this opinion should be rejected by the Tribunal based upon the opposite but more credible opinion of Dr. Martin Wolf, the only witness to be qualified as an expert in the use of the Hay Plan. The fact that Ms. Winter did not evaluate either the CR positions or the PO positions, using any job evaluation plan, underlined the theoretical nature of her opinion. Based, as it was, on theory which had been modified during her cross-examination, the Alliance submitted that Ms. Winter's opinion concerning the Professional Team's factor comparison approach to the Hay Evaluation Plan should be rejected.

[518] In the portion of her Report (R-254) concerning the wage adjustment methodologies which the Commission and the Alliance had used, Ms. Winter stressed her belief that gender identification, valuing of work and any wage adjustment should take place at the level of the job itself rather than at the broader grouping and level in which the jobs are found. She conceded, in cross-examination, that the Treasury Board tribunal decision adopted the identification of gender at the level of the group rather than in the individual job.

[519] Ms. Winter made reference to particular provisions of various provincial "pay equity" acts but made virtually no references to the Canadian Human Rights Act. She also failed to address sections 12 and 13 of the 1986 Equal Wage Guidelines, although she was criticizing the Commission and the Alliance for complying with these sections of the Guidelines.

[520] The Commission and the Alliance approach, calculating an average Hay score for each CR level and a wage gap based on that average value, was, in the opinion of Ms Winter, flawed. This approach, however, has been used elsewhere, including the Treasury Board tribunal decision.

[521] Ms. Winter also argued that each female-dominant CR position should be evaluated. She did not alter her opinion when, in cross-examination, it was pointed out to her that the Treasury Board tribunal decision had endorsed a sampling of the CR population.

[522] Ms. Winter was critical of the make-up of the male comparator group, comprising PO-INT and PO-EXT sub-groups. She made particular reference to Ontario's Pay Equity Act and its requirement to seek a comparator within the complainant group.¹⁶⁶ She also referred to other provincial jurisdictions that require the negotiation of comparator groups. In cross-examination, she admitted that these provincial provisions are not required under the federal legislation.

[523] To summarize, the Alliance submitted that the Tribunal ought to draw a negative inference from Ms. Winter's Reports based on her failure to address the relevance of federal tribunal decisions, the Act and its Guidelines, all of which do not support her version of "pay equity". Her opinions were based almost solely on portions of provincial legislation.¹⁶⁷

[524] Ms. Winter returned to present R-278, a Report which was to address the implications of newly-found CR job information. Her conclusion in that Report was a reiteration of her opinion that the job data used by all evaluators were inadequate. The Alliance submitted, in its argument, that Ms. Winter simply used this new Report as a pretext to revisit her earlier criticisms, while she continued to act as an advocate for the Respondent.

b) Mr. Willis' Testimony

[525] The Alliance's submissions focus on three features relating to Mr. Willis and his evidence, as follows:

his knowledge of Hay Job Evaluation,
his credibility,
his criticisms of the Professional Team's process, the job information that process used, and their approach to Hay, all as expressed in his Report (Exhibit R-455).

[526] While qualified as an expert in job evaluation and "pay equity" by this Tribunal, Mr. Willis was not qualified as an expert in the Hay method of job evaluation. He had only three years of direct experience in applying that method during his employment with the Hay organization from 1968 to 1971. He maintained links with a number of Hay installations after his departure from Hay.

[527] The Alliance questioned the credibility of Mr. Willis. Its argument found a basis in a statement made by the Chair (now resigned) of this Tribunal during an earlier appearance by Mr. Willis. At that time, the Chair expressed reservations about Mr. Willis' expert opinion, given that he had not examined the relevant evaluations and supporting material before giving his opinion.

[528] The Alliance submitted that Mr. Willis' Report, was prepared in a similar manner, without the proper foundation necessary to give an expert opinion. Mr. Willis admitted, when cross-examined about his Report that he had not read all the documentation sent to him by the Respondent. He had been asked, however, to prepare an expert's view on that documentation. Notwithstanding this admission, his Report, and his evidence concerning that Report, did not indicate that his opinions were based only on a partial review of the materials sent. Therefore, the Alliance submitted that his opinion should not be accepted as wholly credible.

[529] The Alliance also pointed to a number of occasions when Mr. Willis, in evidence-in- chief, "insisted vehemently that he believed statements made by Dr. Wolf, in sworn testimony, to be fallacious". Such statements of opinion, argued the Alliance, "raise significant concerns regarding Mr. Willis' credibility as an objective and professional expert witness".¹⁶⁸

[530] An inability to make even a slight concession to cross-examining counsel was also cited as an example of an expert witness who was intransigent and defensive of his position. In other words, Mr. Willis' expert opinions should be rejected as the opinions of someone who lacked the independent, professional approach, required of a credible expert witness.

[531] The Alliance also challenged Mr. Willis' opinions, based on his lack of experience with a "pay equity" process which was taking place in an atmosphere of litigation rather than in a cooperative union-management atmosphere. Mr. Willis' "pay equity" experience had always been, according to his evidence, with the latter type of work where he acted as a facilitator who had the full support of employer management in undertaking the job evaluation studies and the gathering of job information upon which those evaluations would be made. He was, according to the Alliance, either unable, or refused, to comprehend the nature and context of the Complaint before the Tribunal. Indeed, Mr. Willis made one concession during cross-examination. He indicated that the process which he advocated in his Report and during his evidence-in-chief, would, except for the choice of the evaluation plan, be impossible without the full co-operation of the employer.

[532] Based on that concession, the Alliance submitted that to accept Mr. Willis' Report and evidence would be to place the standard necessary for the successful presentation of a prima facie case by a complainant in any "pay equity" complaint under the Canadian Human Rights Act so high that a refusal by the employer to co-operate would always result in the failure of the complaint.

c) Mr. Wallace's Testimony

[533] The Alliance's submissions focus on three features relating to Mr. Wallace and his testimony, as follows:

his knowledge of Hay Job Evaluation,
his credibility,
his Report (Exhibit R-615).

[534] While qualified as an expert in job evaluation by this Tribunal, Mr. Wallace was not qualified as an expert in Hay job evaluation. He did, however, have considerable experience with the use of Hay as a client working under the direction of a Hay consultant. He had never worked for Hay in any capacity, nor had he ever been qualified as an expert in Hay before this Tribunal or any other tribunal or court.

[535] The Alliance was of the view that Mr. Wallace's credibility was put in doubt when he criticized the job understanding of Dr. Wolf and his colleagues, having read only a fraction of the material read, analyzed and evaluated by the Professional Team. Mr. Wallace admitted that he had not been provided with, and did not seek access to, certain original job documentation that had been available to the Professional Team. In fact, Mr. Wallace conceded, in cross-examination, that the Professional Team had a better understanding of the jobs concerned than he had.

[536] The Alliance also submitted that Mr. Wallace tended to highlight the negative and to ignore the positive in the Professional Team's Reports, and in Dr. Wolf's testimony.

[537] Mr. Wallace's Report must, according to the Alliance's submissions, be read in context. That context is that he did not have access to much of the material upon which the Professional Team had relied to make its evaluations.

[538] His Report is a critique of the Professional Team's methodology, process, and the job information used in its evaluations. It stresses that job evaluations, especially for "pay equity" purposes, must be done at a standard used in industrial practice. This is Mr. Wallace's usual methodology. This simply underlines his familiarity with employer-supported studies, and his lack of familiarity with a process taking place in a litigious arena.

[539] The Alliance submitted that "the most significant criticism that Mr. Wallace brought against the Hay Plan was found in his explanation of Appendix `A' to his Report, which shows a weak correlation between Hay points and money paid to CR's. Mr. Wallace admitted, in cross-examination, that he had not considered whether this wage gap for women performing work which appeared to have equal value with men could have been caused by a "pay equity" problem.¹⁶⁹

[540] According to the Alliance, Mr. Wallace's criticisms of the process followed by the Professional Team, compared with the one he proposed, reveal an essential difference in approach. The Professional Team began their 2000 process with the assumption that their earlier evaluations were `correct' and would only change if the additional and new material altered their previous understanding of each job. Mr. Wallace's proposed process was an internal, non-litigious, job evaluation appeal process. This is unlike the process necessitated by the facts of this case.

[541] In sum, the Alliance submitted the following, concerning Mr. Wallace's evidence:

Mr. Wallace's Report and testimony must be read and appreciated in light of the acknowledged fact that he did not have access to much of the material read, analyzed and evaluated by the Professional Team;
Mr. Wallace failed to appreciate the fundamental difference between the complaint before the Tribunal, a complaint which was litigious almost from the beginning, and the employer-managed "pay equity" job evaluation process with which he was familiar;
Mr. Wallace acknowledged that the Professional Team had a greater understanding of the jobs and positions being evaluated than he did at the time he constructed his Report, and gave his evidence.

(viii) The Commission's Position re: Canada Post's Expert Witnesses - Ms. Winter, Messrs. Willis and Wallace

[542] The Commission first noted that none of the three experts presented by Canada Post had actually worked with any of the job data which they found to be unacceptable. Secondly, their opinions were all based on incomplete information. Either the materials provided to them by Canada Post were incomplete, or the witness had not read all the material presented to him/her as the basis for the expert's report requested.

[543] Only the Professional Team had reviewed all of the job information, including that led by Canada Post in its defence. The Commission submitted that the two weaknesses noted are sufficient for this Tribunal to discount the evidence of all three experts presented by Canada Post to report on the evaluation process, the evaluation methodology chosen, and the job information used.

[544] The Commission re-iterated the submission that Ms. Winter's Report wrongly relied on a standard of `correctness' rather than the standard of reasonableness to criticize the collection of job data, and the job information used by the Professional Team during its evaluation process. In her Report, Ms. Winter had noted that the decision of the Ontario "pay equity" tribunal in a complaint involving a Haldimand-Norfolk hospital had concluded that the standard to be applied was that of correctness, and she indicated in her viva voce evidence the following:

Q. ...indicate that a standard of correctness was applied by the Ontario pay equity hearings tribunal in the Haldimand-Norfolk case. Now, ... I take it that you adopt that standard or that you have adopted that standard in your report?

A. Of correctness, yes.¹⁷⁰

[545] Indeed, Ms. Winter went on to indicate that she did not know what was meant by `reasonableness'.

[546] The Commission argued that this reliance on the standard of correctness as a foundation for the expert opinion presented in her Report and in her evidence before the Tribunal should be reason for the Tribunal to discount her evidence. This reliance on a standard of correctness is, according to the Commission, merely an example of the tendency of Ms. Winter to define "pay equity" principles rigidly. Her rigidity was noted, adversely, by the Ontario Court (General Division) in the Service Employees International Union case, where the Court preferred the expert evidence of Dr. Pat Armstrong compared to that of Ms. Winter.¹⁷¹

[547] The Commission submitted that rigid principles are inconsistent with a standard of reasonableness. Indeed, the Commission contends that Ms. Winter herself did not apply the exacting standard of correctness to job information available to the Manitoba Pay Equity Study, for which she was jointly responsible during her days with the Hay organization. In fact, she indicated in evidence that it was necessary, in that case, to "work with" the Manitoba job information. This attitude is analogous to the evidence of Dr. Wolf that the Professional Team had to "work with" the job data which was available to them.

[548] Ms. Winter's own admission of a less than rigid approach to job information during her work with the Manitoba Pay Equity Study is in sharp contrast to her unrelenting criticism of most, if not all, of the job documentation in this case, and her condemnation of the approach taken by the Professional Team to the job documentation. Accordingly, the Commission submitted that Ms. Winter lacks credibility as an expert giving an opinion concerning the job information upon which the Professional Team founded its evaluation process.

[549] The Commission submitted that Mr. Willis' expert opinion concerning the job information and the evaluation methodology and process should also be given less weight than that of the expert opinion of Dr. Wolf. It bases this submission upon its indication to the Tribunal that Mr. Willis was unclear, even ambiguous and evasive, about how long he had taken to review the job documentation materials presented to him by Canada Post to be a basis for his expert opinion.

[550] Additionally, Mr. Willis' experience in "pay equity" evaluation processes was largely based on joint studies in which job data was generated with the agreement of all parties involved. Although he was, as the facilitator in the study which used his `Willis Plan' as the basis for its process, qualified as an expert witness in the Treasury Board case, that complaint involved primarily the issue of the reliability of the methodology chosen to deal with the joint union-management "pay equity" study and evaluation process.

[551] According to the Commission, Mr. Wallace's critical opinion of the job information should be largely discounted because he did not, nor was he asked by Canada Post, to review most of the job documentation. His review was largely limited to an examination of the Professional Team's consideration, in 2000, of the additional evidence provided by Canada Post's many defence witnesses. Additionally, he was provided with portions of data contained in the transcripts of Dr. Wolf's cross-examination. Therefore, the Commission submits that Mr. Wallace's view of the job information is "only through the eyes of Canada Post" and his opinions are not independent of his client.

H. Reliability of Methodology Used by the Professional Team: - Tribunal's Analysis

(i) Introduction

[552] Human rights legislation demands constant attention to the purposive interpretation of the statute involved. In a recent decision of the Federal Court of Appeal concerning the interpretation of "establishment", Evans, J.A. stressed this need, as follows:

Any analysis of a statutory human rights issue must be undertaken with a view to the purposes of the legislative scheme and of the policy objectives of the particular provisions in dispute. A search for the meaning of human rights legislation, including subordinate legislation, must both start with, and be informed throughout by, its essential objective.¹⁷²

[553] As suggested in paragraph [412], the Tribunal accepts that the standard that ought to apply in "pay equity" complaints brought under section 11 of the Act, such as that before this Tribunal, is the standard of `reasonableness' in determining the reliability of the job evaluation system chosen, the process followed, and the job information used.

[554] The Tribunal rejects the submissions of Canada Post that a rigid standard of `correctness' is necessary for the purposes of a "pay equity" process. Rather, the Tribunal finds that the standard of `reasonableness' accepted by the tribunal in the Treasury Board case is more conducive to the interpretation of human rights legislation, and section 11 of the Act in particular. In the case before this Tribunal, the evidence of most experts, including those of Canada Post, presented the concept of job evaluation as "more an art than a science". Therefore, any standard which could not accommodate this concept should be rejected.

[555] Each of the elements necessary in testing reasonable reliability should be examined. In other words, the job evaluation system chosen should be reasonably reliable, the process and methodology used in evaluating the relevant jobs/positions should be reasonably reliable, and the job information and its sources should be reasonably reliable. The findings of the Tribunal should be based on the civil standard of a balance of probabilities.

[556] In this regard, the Tribunal has already noted the importance of examining the evidence provided by a number of expert witnesses in a systematic manner. The components of the systematic model employed by the Tribunal were identified in paragraph [419].

[557] The first several components of the model have already been addressed. The Tribunal has noted the party who called each expert witness, the expert's field of expertise, and each expert's mandate and conclusions reached. The expert witnesses have been identified as Dr. Wolf, Ms. Winter, Mr. Willis, Mr. Wallace, Dr. Bellhouse, and Dr. Kervin.

[558] While the first several components of the systematic model receive further attention in this analysis, they are examined, as appropriate, in the context of the remaining components of the model. In other words, the Tribunal will comment on the extent of each expert's knowledge, experience and standing in his/her field of expertise, and how each expert fulfilled his/her mandate and presented his/her conclusions to the Tribunal.

(ii) The Job Evaluation System Chosen

[559] The job evaluation system used by the Professional Team was the factor comparison approach to the Hay Plan. Canada Post, based upon the opinion of each of its three expert witnesses, submitted that the Professional Team's decision to employ this model of Hay was questionable. Indeed, Canada Post submitted that the use of the Hay Plan itself was not appropriate to a process which would evaluate, for "equal pay" purposes, diverse jobs in the clerical and operations spheres of the corporation.

[560] Although Mr. Willis, an expert witness for Canada Post, acknowledged that the Hay Plan, especially in its Guide Chart-Profile application, is a satisfactory job evaluation instrument provided the evaluators receive adequate training, his opinion was that the Professional Team "mis-used" the Hay Plan by engaging it in its factor comparison mode. Mr. Wallace, another expert witness for Canada Post, also contended that the Hay Guide Chart-Profile method was an acceptable job evaluation tool but regarded it as inappropriate for clerical and production/operations jobs. Mr. Wallace further felt the factor comparison method was long out-of-date. An additional Canada Post expert witness, Ms. Winter, condemned the Hay Plan, generally, for not being suitable to evaluate work of a clerical or "blue collar" nature.

[561] Notwithstanding the opinions of both Mr. Willis and Mr. Wallace concerning the generally satisfactory nature of the Hay Plan as a job evaluation instrument, Canada Post's submissions faulted its use in evaluating clerical and `blue collar' work. In particular, counsel for the Corporation argued that Hay "under-weights" the working conditions factor, which, in turn, leads to under-valuing of elements important to clerical and `blue collar' types of positions.

[562] Mr. Willis' opinion was qualified by his observation that the factor-comparison approach to the Hay Plan might be acceptable for use when it is applied to a single occupational group of jobs. In a "pay equity" case, involving a wide variety of male-dominant and female-dominant jobs, however, he believed that the "point-factor" approach to the Hay Plan was more appropriate. Indeed, Dr. Wolf had agreed, in cross-examination, that, although the factor comparison approach to Hay can be, and is, used with dissimilar jobs, it is easier to apply, and perhaps works better, with jobs of a similar nature.¹⁷³

[563] Dr. Wolf indicated that it was the opinion of the Professional Team that the Hay Plan was quite able to accommodate clerical and `blue collar' jobs, particularly with the strengthened working conditions factor. This would be especially true when it was being applied by evaluators who had experience with the Hay methods. The Commission, in undertaking its 1991 evaluations, had also felt that the Hay Plan was capable of measuring `blue collar' and office-environment jobs.

[564] In this case, the Professional Team which evaluated the sample CR positions, and the `generic' PO jobs was composed of two members who were former Hay associates, with many, many years of experience in working with Hay and other clients. The Tribunal finds that Dr. Wolf was the only expert witness who was sufficiently qualified to assess the validity and reliability of the Hay Plan generally, and the factor comparison approach, in particular, vis-à-vis the requirements of this Complaint. Although he did not appear as an expert witness, Dr. Ingster, the first member of the Professional Team to be approached by the Alliance, confirmed, in letters presented as exhibits in this Complaint, the acceptability of the Hay Plan for use with the materials presented to the Team for evaluation.

[565] The Tribunal notes the Professional Team's statement, included in its Report, entered as PSAC-29, that its mandate was "that the Hay method was to be applied to the job content in accordance with the `best practices' of senior level Hay consultants considered to be expert in the use of the Guide Chart-Profile process". The request had been presented to Dr. Ingster when he was approached by the Alliance to participate in the evaluation process. The expert opinion, expressed by Dr. Wolf in his viva voce evidence concerning the Report of the Professional Team, reinforced the Team's conclusion that the mandate had been followed diligently.

[566] While all three of the expert witnesses called by Canada Post had some experience in working with the Hay Plan, the Tribunal finds that only Dr. Wolf demonstrated that he had the in-depth historical knowledge of how the Hay Plan's original design and use had evolved, in its various configurations, over its many years of application. He testified that he had spent thirty years in job evaluation including about twenty years dealing with the Hay methodology. As observed earlier in this Decision, he estimated that he had evaluated upwards of 10,000 jobs using the Hay process, including office clerical and payroll systems jobs, and in his early days, even `blue collar' jobs. Moreover, when employed with the Hay organization, he had served as a "correlator" or "keeper of the flame" in the role of maintaining the integrity of the Hay system.

[567] Additionally, the Team included Dr. Ingster, the person with whom the Alliance spoke first, and who had received the original Alliance mandate. Dr. Ingster spent an early part of his career with the Hay organization and later had an independent affiliation with Hay while working with a wide-range of clients.

[568] Accordingly, the Tribunal gives significant weight to the ability of Dr. Wolf and his colleague, Dr. Ingster, to choose, under the unusual circumstances of this case, the most suitable configuration of the Hay Plan for use in the evaluation of the jobs/positions involved.

[569] In contrast, the Tribunal finds that the expert opinion of Ms. Winter, in categorically dismissing the Hay Plan, was presented in a rigid and immoveable fashion, leaving the impression of being a witness who was espousing the position of her client rather than being an independent expert who was attempting to help the Tribunal understand difficult concepts outside its realm of expertise.

[570] The evidence of Mr. Willis, while not supportive of the factor-comparison approach, did not dismiss the Hay Plan as such for job evaluation in a "pay equity" context. Nor did the evidence of Mr. Wallace. He stated, on page 2 of his Report, that the Hay Guide Chart-Profile method was "...an excellent job evaluation tool...", but he considered it inappropriate for use with "...clerical and production/operations jobs".¹⁷⁴

[571] Therefore, the Tribunal finds that, on a balance of probabilities, the Hay Plan, whether using the factor comparison method or other approaches, is, in the hands of competent evaluators as were the members of the Professional Team, a suitable overall job evaluation scheme which will address the issues of this "pay equity" Complaint in a reasonably reliable manner.

(iii) The Process

[572] The Tribunal must answer the following question:

Is it more likely than not that the process followed by the Professional Team in undertaking its evaluations of the complainant and the comparator positions and jobs, in this case, was a reasonably reliable one?

[573] As noted earlier in this Decision, the Professional Team, having been engaged by the Alliance, undertook its evaluations as an entity, jointly reaching its decisions either unanimously or by consensus. This is, according to all of the expert evidence before the Tribunal, in line with what most experts in the field recommend. It is also in accord with the Commission's own model. The size of the Team was smaller and its make-up quite different from what one would normally expect. Usually an evaluation committee would be internal to the organization involved in the "pay equity" process. There might be external advisory or facilitation assistance. In this case, however, the situation which set the Professional Team in motion was far from normal.

[574] The existence of the Professional Team was somewhat unique in the field of "pay equity" job evaluation undertakings. The process by which the Professional Team evaluations were completed was equally unique, as there was no direct involvement from the employer or employees as part of the evaluation committee. Obviously, this was not a standard approach to "pay equity" job evaluation. Given the litigious setting, however, this was a process chosen to evaluate the complainant positions and the 10 `generic' PO jobs using a "rigorous application of the Hay Guide Chart-Profile Method of job evaluation...".¹⁷⁵

[575] Canada Post submitted that neither Dr. Wolf nor Dr. Ingster had experience with Canadian postal operations. Indeed, Dr. Wolf admitted in cross-examination that he had never been inside a Canadian post office facility. Although the third member of the Team, Judith Davidson-Palmer, had been an administrative management-level employee of Canada Post, she had little experience with the Hay Plan.

[576] In addition to Canada Post's concerns regarding the make-up of the Professional Team, it submitted that there was a dearth of written records of exactly how the Team had conducted its work. The "audit trail", usually a vital part of any evaluation process, was weak at best, leaving little documentation to support the Team's conclusions.

[577] Canada Post's three witnesses with experience in job evaluation processes denigrated the work of the Professional Team. Ms. Winter noted that the Team members lacked experience in dealing with clerical and `blue collar' work, and she was of the opinion that they ran an undisciplined `sore thumb' review of evaluation results. She faulted the Team for not establishing a representative and consistent set of Benchmark evaluations at the start of the evaluation process. Furthermore, Ms. Winter believed that the Team's initial ordering of positions by classification levels and by the Commission's evaluation ratings may have biased the overall process.

[578] Mr. Willis expressed his opinion that the Team lacked a certain discipline in its deliberations. For example, he felt that the "sore thumb" step should have been more tightly structured. Additionally, he was opposed to the `black box' techniques that Dr. Wolf had used in the evaluation process, techniques drawn from his background in engineering. Mr. Willis also criticized the Team's use of the U.S. Hay job description standard.

[579] Mr. Wallace noted that two members of the Team, while being recognized as experts, lacked postal operations experience. He also commented unfavourably on what he regarded as the Team's lack of discipline and rigour in the evaluation process, including the obvious weakness of the "audit trail".

[580] The Tribunal notes that the Professional Team had to tackle its assignment in a rather unusual manner. It had, over a relatively short period of time, to sift through literally thousands of pages of documentation and exhibits compiled by others. The evidence indicates that the Team members were not able to adhere to the exact discipline and rigour which would normally be expected in a joint union-management process, where the job evaluation committee would be composed of administrative staff members as well as other employees. The difficulty of evaluating in such a setting was largely off-set, however, by the particular competence of the three-member committee. Two of the members had many years of experience in evaluating jobs using the Hay Plan and other systems. The third member of the Professional Team had been a Canada Post employee, working in management and organization development. Working together, they believed they were able to complete the task assigned. Their Report and Dr. Wolf's viva voce evidence indicated that the Team considered that the process by which it conducted its evaluations, although unconventional, was done with reasonable diligence and discipline.

[581] The Tribunal accepts this opinion, and finds that the process, operating in the context of a unique litigious situation, and with an approach dependent upon the competence and experience of the small Professional Team, was reasonably reliable. Inevitably, it was not the superior process which might have resulted had the parties been working in the more usual, co-operative manner. It was, however, in the view of the Tribunal, a reasonably reliable process, given the circumstances under which it was accomplished.

[582] This acceptance of the opinion of the Professional Team, and that of Dr. Wolf, in particular, that their evaluation process was reasonably reliable is based upon the Tribunal's finding that the expert evidence of Dr. Wolf is more credible than that proffered by Canada Post's experts, Ms. Winter, Mr. Willis, and Mr. Wallace. Dr. Wolf was present during the process of job evaluation, and participated as a member of the Team. His evidence was factual as well as that of an expert.

[583] In giving more weight to Dr. Wolf's evidence, the Tribunal has already acknowledged that the process followed by the Professional Team was by no means of the highest level.

[584] The weight, if any, which the Tribunal has given to the expert evidence presented by Canada Post, however, has not been sufficient to overcome the opinion of Dr. Wolf that the Professional Team participated in a reasonably reliable evaluation process.

[585] Ms. Winter, a self-made businesswoman and President of her own consulting company, had spent several successful years as an employee of Hay Canada, attaining the rank of Sr. Vice-President and Partner. A sometimes argumentative witness, her opinions in both her viva voce evidence and in her Reports, displayed a tendency to rigidity and a requirement for absoluteness when measuring reliability.

[586] Mr. Willis began his job evaluation career with the Hay organization in the United States and formed his own consulting company in 1971. Having developed his own job evaluation plan in 1974, he had handled many job evaluation studies in both the U.S. and in Canada. He provided advice to the JUMI Study Committee that preceded the Treasury Board case. That Committee used the Willis Plan in its job evaluations.

[587] Mr. Willis indicated in evidence that he was accustomed to working in a non-litigious and co-operative setting. Although his Report did not so indicate, his viva voce evidence confirmed that he had not read all of the fairly detailed material sent to him by Canada Post. His responses given in cross-examination were somewhat evasive about the amount of time he had spent on the work he was requested to do by Canada Post. While obviously a successful, now-retired consultant, having run his own firm for some 25 years, his manner before the Tribunal was, at times, rather curt.

[588] Mr. Wallace began his considerable experience in job evaluation when he participated in the introduction of the Hay system at the Bank of Montreal in the 1970's. He went on to manage the Hay process for Shell Canada and integrated it with Hay world-wide for the parent company, Royal Dutch Shell. He had been Sr. Vice-President of a consulting company since late 1996.

[589] Mr. Wallace's mandate was limited to a review of the process undertaken by the Professional Team in 2000, in addressing the impact on its 1993/94 evaluations of the additional evidence that had subsequently arisen from a number of Canada Post witnesses.

[590] He presented himself before the Tribunal as someone who had learned, on the job, a great deal about job evaluation and compensation design. His opinion, critical of the Professional Team process, was predicated on industrial standards with which he was familiar and which were, in his opinion, rather strict standards. He gave no deference to the need to give a large and liberal interpretation to human rights legislation.

[591] The Tribunal, while giving no weight, for the reasons noted, to Ms. Winter's evidence concerning the evaluation process used by the Professional Team, does not totally dismiss the evidence presented by Messrs. Willis and Wallace. Rather, some of the criticisms expressed by them address features which, under normal circumstances, one would probably prefer not to be present in a job evaluation process. It is, therefore, a question of determining the degree of weight to be afforded.

[592] Accordingly, after considering the expert evidence, the Tribunal has given less weight to the evidence of Canada Post's witnesses than it has to Dr. Wolf and the Professional Team.

[593] Therefore, the Tribunal finds that it is more likely than not that the evaluation process which the Professional Team used in its work was reasonably reliable.

VI. RELIABILITY OF JOB INFORMATION SOURCES AND RESULTING JOB INFORMATION USED BY THE PROFESSIONAL TEAM

A. Background

[594] All three parties have agreed on the vital importance of using reliable information and data in job evaluation plans of whatever stripe. Canada Post's three expert job evaluation witnesses stressed the need to work with reliable job information and data, although Ms. Winter's opinion encouraged a standard higher than reasonable reliability as her exchange with counsel for the Alliance, in cross-examination, confirmed:

Q. So, are you suggesting then, or stating rather, that in order to accurately evaluate positions you need 100 percent job knowledge?

A. Yes.¹⁷⁶

[595] Both Dr. Wolf and Mr. Willis testified that, in their respective opinions, one's understanding of the jobs to be evaluated (including the quality of the job information) is of first importance. Of second importance is the process to be followed:

Dr. Wolf: Even to be a Hay consultant, the amount of time that you need with the job evaluation process is less important than your understanding of the jobs. The rule of job evaluation is like in computers: it is garbage in, garbage out. If you don't understand the job, you can't evaluate it properly, no matter how much you know about the technology of the process.¹⁷⁷

Mr. Willis: Next to the quality of the information, I'd say that the process is second in importance.¹⁷⁸

[596] As noted earlier in paragraph [413], the generally accepted standard of the job evaluation industry, of which all expert witnesses were aware, is to seek, to the extent possible, accuracy, consistency and completeness of job information being used for job evaluation purposes. Given the Tribunal's decision in this case to apply a reliability standard of `reasonableness' (paragraphs [412], [553]), this calls for reasonable accuracy, reasonable consistency and reasonable completeness.

[597] Accordingly, reasonably reliable job information and data is an essential ingredient of job evaluation as a concept, given its inherent dependence on subjective human judgement. Decisions of evaluators who are using reasonably accurate, consistent and complete job information should, understandably, and indeed, logically, produce more realistic and acceptable results than using job information that may be questionable or flawed.

[598] Although the Tribunal has already tracked the sources and the nature of the job information used in this case, in excruciating detail, it has decided to re-examine those sources

and job information in condensed form in two stages, which the Tribunal has labelled for convenience of reference, as follows:

FACTS I: These are the factual job information sources and the job information and data that resulted from those sources that existed prior to the date when the Professional Team began its work for the Alliance.

FACTS II: These are the additional relevant data and evidence to which the Professional Team had access once it began its work.

B. FACTS I

(i) Composition

[599] In summary form, FACTS I is comprised of the following:

the Job Fact Sheet/Questionnaire which was completed in 1986 by sampled employees; it was designed to relate to the not fully developed System One job evaluation plan, by a senior Commission officer, without professional assistance. Although it was intended to serve as the primary source of up-to-date information for both the CR and the PO positions, it was used by the Commission, in fact, only for the CR positions;
the Interview Guide was also designed by Commission staff, and based on System One. It was meant to be completed by Commission interviewers as a follow-up to the Job Fact Sheet/Questionnaire. It was used for the CR employees only; interviews were completed by December 1986;
job descriptions and organization charts which were expected to be attached to the Job Fact Sheet/Questionnaire by the incumbent employee completing that form. Most CR job descriptions were "unofficial", including several CR Benchmark position descriptions; many CR organization charts were out-of-date;
PO position samples were attempted with new 1989 lists of PO employees without success. In place of samples, 10 `generic' PO job specifications were drawn up in 1990/1991 by the Commission, with Canada Post management personnel supplying the foundation information and materials, including "unofficial" job descriptions, and job profiles which identified responsibilities of each job. These job specifications were developed in a very different manner from that used for the CR positions which were determined by means of what the Commission called a random sample of CR employees occupying actual positions;
the 10 `generic' jobs represented an amalgam of functions for 10 commonly held job types in the Postal Operations Group. They did not represent actual jobs or positions and did not have union approval. The PO supervisory sub-group, which constituted an element of the comparator group in the original Complaint, was not represented in the 10 `generic' PO job specifications, in contrast to the CR sample which included some supervisory representatives at the CR-5 level.

[600] These were the key sources and the nature of the job information that resulted from those sources during the pre-Professional Team period of 1986 to 1991.

[601] Related to these sources and the resulting job information, are four facts which have already been addressed elsewhere in this Decision, but are worthy of note in the context of FACTS I as they have potential impact on the nature of the job information arising from the sources.

[602] The first is the uncertainty that surrounds the various unprofessional calculations of the CR sampling size. The original CR sample was developed by a Commission officer in 1986. In 1987, the Commission did not act upon Statistics Canada's advice regarding the CR sample size. The CR sample was subsequently reduced in 1991 by Commission staff.

[603] The second is the fact that the Job Fact Sheet and the Interview Guide were both designed around the uncompleted System One evaluation system which had the endorsement of neither the Alliance nor Canada Post.

[604] The third is that the job data were gathered at different times. The data for the CR positions were gathered in 1986. The bulk of the data for the `generic' PO jobs was assembled in 1990/1991.

[605] The fourth is the apparent incompatibility between the job information collected for the CR incumbent-held positions, and the job "specifications" compiled by the Commission for the non-incumbent-held `generic' PO jobs.

[606] The Tribunal notes that it is undisputed by all parties that all of FACTS I occurred well before the Professional Team had been approached by the Alliance or had begun its work.

(ii) Submissions of the Parties and Expert Witnesses

[607] Focussing exclusively on the FACTS I elements, what were the principal arguments made by the parties and the expert witnesses in their respective submissions, about these elements?

[608] There was virtual unanimity between Dr. Wolf and two of Canada Post's expert witnesses with respect to the 1986 Job Fact Sheet/Questionnaire. It was Dr. Wolf who classified this document as `abominable', adding that "...the guy who developed it probably should be taken out and shot", (paragraph [445]). Both Mr. Willis and Ms. Winter dismissed it, Mr. Willis calling it "...hopelessly inadequate for Pay Equity evaluation purposes", (paragraph [497]). Canada Post, too, faulted it on the grounds it was self-evaluative, which was widely held, in job evaluation circles, to be unacceptable.

[609] Despite Dr. Wolf's condemnation of the design of the Job Fact Sheet, he testified that he and his two colleagues still made some use of it by disregarding the self-evaluation aspects of the responses and focussing on the "job description information" that could be found in the completed document. In response to a question from Alliance counsel, Dr. Wolf said:

So you had to work against the tide, if you will, with these documents unfortunately, but there was information there. You just had to be selective in using it to make sure you didn't pay any attention to the extraneous part.¹⁷⁹

[610] Both Mr. Willis and Ms. Winter also faulted the 1986 Interview Guide believing it replicated many of the deficiencies of the Job Fact Sheet while noting that it too, was based on the not-fully-developed System One evaluation plan. Dr. Wolf, himself, testified that the Interview Guide did not add anything of significance to the Professional Team's understanding of the CR incumbents' duties.

[611] Dr. Wolf acknowledged that some 50 of the CR job descriptions were unsigned and/or undated, and that others were often out-of-date or "unofficial," and even sometimes missing. Some did not include information on working conditions. Similarly, supporting organization charts were not always available or up-to-date.

[612] Two of Canada Post's expert witnesses identified serious deficiencies in the job descriptions primarily in terms of age, accuracy, and official status and all three witnesses stressed the importance of those undertaking evaluations being entirely satisfied with the job descriptions.

[613] Dr. Wolf admitted that because of the general unacceptability of the Job Fact Sheet which had been intended as the primary source document for job evaluation purposes, the Professional Team were compelled to rely on the available job descriptions as their primary document for the CR evaluations. He testified that he and his colleagues accepted the job descriptions as they were, provided they were in correct Canada Post format.

[614] Dr. Wolf clarified what was meant by treating the job descriptions as the primary document in comparison with the Job Fact Sheet and the Interview Guide, in the following cross-examination exchange with Canada Post counsel:

...when we say primary document, what we are saying is that when there is some question of the consistency of the documents, which one do we defer to, the answer is that we would defer to the Position Description as the official Canada Post document. So when we say it is the primary document, that is the one we would defer to.¹⁸⁰

[615] In those situations where there were no position descriptions - and this occurred in a fair number of instances (paragraph [448]) - Dr. Wolf agreed that such a deferment would, obviously, not be possible.

[616] Mr. Willis, while recognizing that the PO job specifications came closer to providing factual job information than the CR documentation, cautioned that the former were based on management-supplied, not the more appropriate employee-supplied, job data. Moreover, he pointed out that, contrary to accepted practice, the job data for each of the CR and PO employee groups were collected by totally different processes and at two different periods of time.

[617] Ms. Winter faulted the PO job specifications for not reflecting actual positions and being incomplete by not including the job rotational requirements of the PO-4 job level.

[618] In considering the first of the four related facts - the uncertainty about CR sampling mentioned in paragraph [602] - it is necessary to turn to the evidence of the two expert witnesses, Drs. Bellhouse and Kervin.

[619] There is an appreciable difference of opinion between these two expert witnesses. As noted earlier, Dr. Bellhouse argued that the Commission's original 1986 sampling design, upon which the selection of CR survey incumbents was based, was flawed. He also argued that the Commission's sampling of employees rather than positions led to biases in the sample which were compounded by the level of non-response in the survey for which corrective action was not taken.

[620] In particular, under the circumstances of this case, where Dr. Bellhouse understood that the CR classification levels were comprised of overlapping intervals of Hay Plan points, he considered that a full census of each CR position would be the required route. When questioned by Alliance counsel about whether overlapping Hay points and pay based on those points was a compensation rather than a statistical issue, Dr. Bellhouse answered that, given the existence of such an overlap, his recommendation would be a census.

[621] During the re-examination of Dr. Bellhouse by Canada Post counsel, Dr. Bellhouse agreed, however, that a full census could be avoided if one were seeking an average value per CR job title instead of per position. One would redefine the CR community by all of its job titles and, assuming a good deal of homogeneity, take an appropriate sample within each job title.¹⁸¹

[622] The Tribunal notes that in earlier evidence relating to the failure of the Commission to implement a Statistics Canada recommendation in 1987 to augment the CR sample, Statistics Canada was commenting on the design of an acceptable random sample rather than a census.

[623] With respect to the 10 PO `generic' jobs, Dr. Bellhouse did not regard them as a probability sample. He considered them, at best, to be a "judgement sample" with the possibility of substantial bias because it was a selection of particular job titles (paragraphs [454]-[459]).

[624] Dr. Kervin's opinion was that Dr. Bellhouse over-emphasized the analytical nature of statistical analysis and the need for scientific quantitative reliability. Dr. Kervin's point was that Dr. Bellhouse failed to recognize the sociological, qualitative, and systemic issues involved in a "pay equity" case. Dr. Kervin further indicated that the representative sample of CR employees which formed the basis for the collection of data for the Complainant group was more than adequate. He agreed with Dr. Bellhouse that the 10 PO `generic' jobs constituted a "judgement sample" (paragraphs [466]-[468]).

[625] The Tribunal is cognizant of the different backgrounds of these two experts. One gave his opinion evidence based upon his expertise as a professional statistician, the other based upon his expertise as a professional sociologist. Dr. Bellhouse, a Professor of Statistics, was qualified as an expert in statistics, with specialization in survey sampling. Dr. Kervin, a Professor of Sociology, was qualified as an expert in data collection and data analysis. As noted earlier, both have had considerable experience in working with paying clients, as well as students, in their respective fields of expertise. Judging from the evidence presented to qualify each as an expert, both are well regarded in their fields.

[626] At this juncture, the issue is the reasonable reliability of the sampling methodology and the sample size employed for the CR population. Much of the expert evidence about implementing a "pay equity" study in normal circumstances underlined the need to seek expert advice from professionals. In this instance, expertise in the design and implementation of statistical survey sampling technology was necessary, but not sought initially for the 1986 sample.

[627] Dr. Kervin was qualified as an expert in data collection and data analysis. He did not consider himself to be a professional statistician. In response to a question from Commission counsel, he said:

I am not a statistician. I don't generate new statistics.

I don't look at the properties of statistics. Instead, I use them...¹⁸²

[628] On the other hand, Dr. Bellhouse referred to himself as a "sampling statistician".¹⁸³

[629] Mr. Willis testified that the process of selecting the CR sample fell short of meeting the stringent sampling requirements of a "pay equity" case. Although he did not indicate that a census was a necessity, his opinion seemed to be closer to that of Dr. Bellhouse than that of Dr. Kervin.

[630] Accordingly, the Tribunal is faced with two conflicting opinions about the CR random sample. Ideally, one would want to re-examine the sampling methodology employed. But, the Tribunal has noted that the original sample of incumbents was a significant one, representing almost ten percent of the total CR population. The Professional Team deliberately chose to evaluate the positions of the full original sample and not limit itself to the Commission's subsequently reduced sample level. Furthermore, there was no solid factual evidence provided to demonstrate that the full sample was unrepresentative of the total CR community.

[631] The second related fact (paragraph [603]) concerns the reality that the Job Fact Sheet and the Interview Guide were both designed around the uncompleted System One job evaluation plan. The Professional Team employed the Hay factor comparison plan in conducting its job evaluations. Using one plan to design instrumentation - even if only partially used - and another to undertake job evaluations, is generally regarded, in the industry, as an unacceptable practice.

[632] Certainly, the Commission's booklet published in 1992 entitled "Implementing Pay Equity in the Federal Jurisdiction", makes it very clear that "...use of a single plan to evaluate all jobs is essential," (paragraph [358]), which would at least imply, if not explicitly state, that a single evaluation plan should govern all aspects of a particular job evaluation undertaking, including the gathering of job information documentation.

[633] The third fact (paragraph [604]) relates to the gathering of job data at different times. The data for the CR positions were collected in 1986, the data for the PO `generic' jobs primarily in 1990/1991. Mr. Willis commented on the desirability of collecting job information for all jobs being compared within a reasonable period of time of each other.

[634] The fourth fact (paragraph [605]) concerns the incongruity between the job information collected from incumbent CR employees and the job "specifications" compiled by the Commission for the non-incumbent `generic' PO jobs. Two of Canada Post's experts questioned such an approach.

(iii) Credibility of Evidence of Expert Witnesses

[635] What has been the position of each of the parties concerning the source materials that make up what the Tribunal has called FACTS I?

[636] Canada Post has said, in effect, that the information documented in FACTS I cannot be relied upon to determine reliable job evaluations.

[637] The Alliance has questioned the credibility of all three of Canada Post's expert witnesses - Ms. Winter largely on the basis of not meeting the standard of independence of an expert witness, and being too categorical or exaggerated in presenting her opinions; Mr. Willis for not having read all the documentation sent to him by Canada Post, and also on the grounds of either being unable, or refusing, to comprehend the litigious context of the Complaint; and Mr. Wallace for not having had access to much of the documentation involved in the case, and his tendency to highlight the negative while ignoring the positive.

[638] The Commission discounted the evidence of Ms. Winter essentially on the basis that she relied on the standard of correctness as the foundation for her expert opinions. Mr. Willis' credibility was questioned by the Commission on the basis of being unclear, even ambiguous and evasive about how long he had taken to review the job materials sent to him by Canada Post. The Commission submitted that Mr. Wallace's critical opinion of the job information should be largely discounted because he did not, nor was he asked, to review most of the job documentation.

[639] Insofar as the three Canada Post expert witnesses are concerned, the Tribunal concludes that the evidence of Messrs. Willis and Wallace should not be completely dismissed. Aspects of their evidence deserve some weight. As for Ms. Winter, in the Tribunal's view, her absolutist standard of correctness on virtually all fronts requiring a judgement about reliability, rendered her opinions beyond acceptance.

[640] Mr. Willis, however, was a witness with considerable years of experience in the job evaluation industry. Perhaps, he was somewhat evasive, and even acerbic on occasion, as when, for example, in response to a question from Canada Post's counsel, he said:

...the CHRC and PSAC's three consultants were faced with having to do work with inadequate data. I think it was so inadequate that neither one of them could - without additional input, without additional information, I don't see how either one of them could have done a satisfactory job.

In my overall analysis, I cannot back down one step: they are both junk...¹⁸⁴

[641] Mr. Willis' attitude should not completely overrule his expertise. His service to the Joint Union-Management Initiative, and subsequently as an expert witness on the Treasury Board case, are illustrative of the depth of his knowledge and experience, and his reputation in his field.

[642] The Tribunal appreciates the sheer volume of materials that existed in this case. There were over 400 transcripts and about a thousand supporting exhibits alone. Although not all of this documentation was sent to Mr. Willis, he did receive a significant amount of material to review. For an expert witness of the calibre and continental standing of Mr. Willis, some of the documentation in FACTS I, such as that concerning the CR sample methodology, the design of the Job Fact Sheet and the Interview Guide, would have pointed to deficiencies which were readily apparent without having read all the materials.

[643] Mr. Wallace also had considerable experience in the field of job evaluation. While his mandate, in this case, covered a more limited aspect and period of time, involving far less documentation than Mr. Willis, his knowledge and depth in applying the objectives and principles of job evaluation, particularly in the private sector, and in "pay equity" situations, were impressive.

C. FACTS II

(i) Composition and Impact

[644] FACTS II constitutes the additional relevant information, data and evidence, beyond FACTS I, to which the Professional Team had access in undertaking its CR and PO job evaluations in 1993/1994 (supplemented in 2000).

[645] One such additional item which the Alliance had provided to the Team was Hay documentation which the Commission had originally received from the Hay organization, including Guide Charts and a variety of samples of definitions of Hay evaluating factors. Dr. Ingster advised in correspondence with the Alliance that the Hay documentation had not been tailored for Commission use but was rather general Hay presentational material.

[646] The Professional Team also had access, in 1993, to the Commission's Rationale Statements which recorded its ratings, and reasons therefor, for its 1991 job evaluations of the reduced sample of 93 CR incumbents. The Commission's 1991 CR evaluations were based on the XYZ Hay Plan. The Team also drew on its own evaluations of the original 93 sample to provide Reference Positions for its second phase of 97 CR's which it undertook in 1994. Notes taken by Professional Team members during their telephone interviews with CR incumbents conducted in May 1993 and September 1994, were further CR position materials in the Team's possession.

[647] For the PO community, the Professional Team had access to behavioural dimensions for each job which the Commission had obtained from Canada Post as well as the Commission's Rationale Statements indicating its 1991 job evaluation ratings, and reasons therefor, of the 10 `generic' jobs based on the XYZ Hay Plan. Dr. Ingster had advised, however, in his letter of July 21, 1993 to the Alliance, that the behavioural dimensions and job profiles had not been provided for four of the 10 PO `generic' jobs.

[648] A Commission-prepared document was also furnished which included descriptions of the knowledge and skill, problem solving, responsibility and working conditions characteristics of the 10 PO `generic' jobs. Finally, the Team also referred to a variety of Canada Post operator handbooks, postal guides and related materials.

[649] Subsequent to the Professional Team's 1993/1994 CR and PO job evaluations, there was, in 1997, newly-found CR documentation which included several, but not all, of the previously missing job descriptions. However, Dr. Wolf, on behalf of the Team, concluded that this additional material was not significant enough to re-evaluate the Team's earlier evaluations.

[650] In the year 2000, the Professional Team undertook a review of the possible impact on its 1993/1994 job evaluations of a considerable amount of evidence which had been submitted to the Tribunal during the period of 1995 to 2000, by a number of Canada Post witnesses.

[651] This voluminous new evidence consisted of approximately 4,000 pages of written material including transcripts of many days of testimony-in-chief and of cross-examination concerning job content, primarily for the 10 `generic' PO jobs. A considerable number of exhibits were also involved such as Canada Post manuals, handbooks and training materials. Several other exhibits were provided by the Alliance and the Commission.

[652] Dr. Wolf screened this extensive documentation and extracted material that he considered was not relevant to job evaluation. The balance was then referred to the full Professional Team. Canada Post's counsel questioned whether this was an acceptable practice to discard material before it had been seen by his two colleagues.

[653] Dr. Wolf indicated in his evidence that he and his two colleagues had concluded, based on this new material, that each of the 10 `generic' jobs probably described few, if any, of the many incumbents of those jobs. Few, if any, would be performing all of the duties described. In response, and in fairness to all incumbents, Dr. Wolf and his two colleagues chose to re-evaluate the 10 `generic' jobs based on the assumption that all incumbents were performing all of the respective duties concerned.

[654] The Professional Team concluded that the new evidence had no impact on their CR evaluations but had some impact on their PO `generic' job evaluations. For example, Dr. Wolf testified that:

...the range of content within any one of the PO jobs was much greater than we had originally realized. The 10 jobs really represent many more than that.¹⁸⁵

[655] Dr. Wolf reported that several changes in evaluation point values resulted from the re-evaluation of the 10 `generic' jobs. Five of the PO `generic' jobs had no changes in their evaluation point values, and three jobs had minimal changes of three points or less. Two `generic' jobs had significant changes in point value. The Counter Clerk `generic' went up in value while the Relief Mail Services Courier `generic' went down in value.

[656] Consequently, FACTS II provided the Professional Team with a fair amount of additional job information, data and background material to that provided by FACTS I. The question, therefore, arises: how useful did this FACTS II additional job information and background material prove to be?

[657] Undoubtedly, it added to the Professional Team's overall perspective of the nature and work of the employee groups involved in the Complaint. While a good portion of the new material related solely to the Postal Operations Group (PO) of employees, there was also additional material covering the CR community.

[658] For example, the Professional Team's CR employee telephone interview notes served as new material which was helpful in providing a focus on the working conditions factor of the CR positions involved.

[659] As already noted, Dr. Wolf confirmed that the Professional Team's review of the newly found CR documentation in 1997 was not sufficiently significant to re-evaluate the Team's earlier evaluations. On the other hand, the Team's review of the new evidence arising from the Canada Post witnesses over the period 1995 to 2000 did result in major changes in evaluated point values for two of the 10 PO generic jobs.

[660] All told, the Tribunal finds that the evidence supports the view that while much of the new job information and background materials that made up FACTS II did not add a substantive new dimension to the core job information base of FACTS I, it did augment and fortify the Professional Team's understanding of the jobs and positions to be evaluated.

D. FACTS I and II Compared to Reliability Standard of the Job Evaluation Industry

[661] How did the job information and data of FACTS I and II measure-up to the generally accepted standard of the job evaluation industry, as described in paragraph [596]? In other words, how reasonably accurate, how reasonably consistent and how reasonably complete were the job information/data used by the Professional Team (FACTS I and II) in undertaking its job evaluations in 1993/1994 (supplemented in 2000)?

[662] The deficiencies already well documented above in the job descriptions which the Professional Team came to regard as their primary source documents for the CR positions are, perhaps, one of the best illustrations of a general lack of accuracy, consistency and completeness. Dr. Wolf, himself, acknowledged the many deficiencies including out-of-date, incomplete, unofficial and even missing CR job descriptions.

[663] Canada Post was supported by Mr. Willis and Ms. Winter in commenting on the lack of accuracy, consistency and completeness in many of the CR job descriptions, noting that generally they did not include information on working conditions. Mr. Wallace also stated in his report that there was a "...lack of complete and consistent documentation".¹⁸⁶

[664] Even the Commission cautioned about the use of job descriptions in its booklet on implementing "pay equity", as follows:

...job descriptions should not be used on their own or treated as the primary source of data, since they often replicate prevailing stereotypes and are not always an up-to-date, accurate reflection of work done, (paragraph [358]).

[665] An inconsistency also occurred in the use of the Interview Guide with CR incumbents. Certain changes in its original design, proposed by a representative of the Alliance, were accepted by the Commission after interviews had already begun, resulting in two versions of the Interview Guide having been in the system.

[666] Questions of inconsistency and incompleteness also arose in evidence about the CR sample which included supervisors at the CR-5 level, while the PO supervisor's sub-group had been dropped by the Commission from the PO `generic' jobs. Similarly, lack of consistency was expressed over the appreciable difference in the dates of information collection - 1986 for the CR's and 1990/1991 for the PO `generic' jobs. Mr. Willis, for example, indicated that all data involved in job evaluation should, ideally, be collected during the same time period and as near as practicable, to the date of performance of the job evaluations. He considered this to be important because of the tendency of jobs to change over time.

[667] The Alliance and the Commission did not really directly address the industry standard of attaining reasonable accuracy, consistency and completeness of job information used in job evaluation undertakings. Both parties tended to discredit the evidence, in this regard, of Messrs. Willis and Wallace on the basis of not having read all the relevant documentation and not having performed CR and PO job evaluations themselves. They contended, therefore, that the opinions of Messrs. Willis and Wallace were based on incomplete information, and should be rejected. As only Dr. Wolf and his two colleagues had worked through all the job documentation and actively performed job evaluations, the Commission and the Alliance urged that Dr. Wolf's opinions be accepted.

[668] Interestingly, Dr. Wolf openly acknowledged particular inaccuracies, inconsistencies and incompleteness in job information, which were referred to above.

E. Tribunal's Analysis

(i) A Daunting Task

[669] In undertaking this final analysis of the job information used in this case, the Tribunal is reminded of the following two factors which were observed earlier in this Decision.

[670] Both Dr. Wolf and Mr. Willis confirmed that in conducting job evaluations, the quality of the job information and one's understanding of the jobs are paramount, out-matching either the evaluation plan or the process involved.

[671] The Alliance and the Commission chose to rely exclusively on the Professional Team's job evaluations performed in 1993/1994, (supplemented in 2000), to substantiate the Complaint. In effect, the Alliance and the Commission have asked that the Commission's job evaluations performed in each of 1987 and 1991 be ignored in favour of those conducted by the Professional Team. What cannot be ignored, however, is the fact that a portion of the source materials used by the Team to conduct its evaluations was, essentially, the information that the Commission had employed in its own earlier evaluations, that is FACTS I.

[672] The Tribunal's assessment and weighing of the evidence submitted by each of the parties and expert witnesses concerning this issue of the reliability of the job information has been a daunting task.

[673] There is little doubt that the job information (FACTS I and II) employed by the Professional Team in conducting its job evaluations did not meet the standard that one would normally expect from a joint employer-employee "pay equity" study. But, given the somewhat painful and prolonged circumstances of the case before this Tribunal, was the job information "good enough", on a balance of probabilities, to generate reasonably reliable job/position values that, in turn, could be used to demonstrate whether or not there was a wage gap?

[674] Dr. Wolf, as spokesman for the Professional Team, acknowledged that he and his two colleagues found many shortcomings in the available job data. But he also said that their job understanding was "...adequate but not necessarily ideal..." (paragraph [484]).

[675] By "adequate", the Tribunal suggested in paragraph [486], one might consider "sufficient" as an acceptable synonym.

[676] The Tribunal sees little value in attempting to attribute blame for the state of the job information but notes that the Alliance counsel in his final oral argument stated the following:

If we have less than an ideal view of the PO work, I suggest to you that, in large measure, that has been caused by decisions made by Canada Post.¹⁸⁷

[677] At the same time, the Tribunal notes that section 43 of the Act furnishes the Commission with certain powers to obtain relevant documentation from a respondent while conducting its investigation of a complaint. Hence, the Tribunal accepts that the Commission, and perhaps even the Complainant, could also be held partially accountable for the job information available in this case.

[678] Another aspect which the Tribunal believes deserves mention and over which the Professional Team had no direct input relates to the CR sample. While the conflicting positions of expert witnesses Drs. Bellhouse and Kervin have been documented above, it is helpful to note the following final oral argument made by the Alliance's counsel which adds yet another dimension to that issue:

I will point out one factor that you might consider, however, and that is that when you are looking at the data, one of the things that stuck in my mind, the representivity of the data, you go back to the root cause of the alleged problems with the CR sample, and the problem comes from the fact that Canada Post gave the Commission an out-of-date employee list. That is the origin, that is the genesis of the need to go and get other employees.

I am not going to tell you that the Commission was perfect in what they did, but nor was Canada Post. It certainly would be inequitable for Canada Post to now come to the Tribunal and say that you can't rely on the data when they were responsible for providing the information to the Commission.¹⁸⁸

[679] To answer the question of whether or not the job information was reasonably reliable, the Tribunal found the following excerpt helpful:

In Anglo-Canadian law ... the courts have consistently held that if the plaintiff establishes that a loss has probably been suffered, the difficulty of determining the amount of it can never excuse the wrongdoer from paying damages. If the amount is difficult to estimate, the tribunal must simply do its best on the material available, though of course if the plaintiff has not adduced evidence that might have been expected to be adduced if the claim were sound, the omission will tell against the plaintiff. In Ratcliffe v. Evans, Bower L.J. said:

As much certainty and particularity must be insisted on, both in pleading and proof of damage, as is reasonable, having regard to the circumstances and to the nature of the acts themselves by which the damage is done. To insist upon less would be to relax old and intelligible principles. To insist upon more would be the vainest pedantry.¹⁸⁹

[680] While the aforementioned excerpt relates to the law of damages, the Tribunal finds that it addresses an approach that may be analogous to what the Tribunal considers to be the spectrum of reasonable reliability. The Tribunal has already accepted (paragraph [596]) that a standard of `reasonableness' should apply in determining the reliability of the job evaluation system chosen, the process followed and the job information used. An absolutist standard of `correctness' has been dismissed as has the evidence of Ms. Winter on these same grounds.

[681] The publication of Professor Waddams, cited above, to which, interestingly, both the Commission and the Alliance made reference in the context of their submissions on Remedy, refers to the need for a tribunal, in determining damages, to "do its best on the material available," because the difficulty of determining the amount of damages "can never excuse the wrongdoer from paying damages." The citation goes on to state from the Ratcliffe v. Evans case that to insist upon more than reasonable certainty in pleading and proof of damage "would be the vainest pedantry," and to insist upon less "would be to relax old and intelligible principles".¹⁹⁰

[682] In other words, absolutism should probably be avoided at both ends of the spectrum. A standard of 100% correctness is unacceptable at the top end of the spectrum as is a standard at the lower end which simply dismisses everything as being entirely worthless. This conception of a spectrum is certainly relevant to the Tribunal's decision concerning the reasonable reliability of the documentation used to conduct the evaluations in this Complaint.

[683] In view of the circumstances of this particular case and the remedial nature of human rights legislation calling for a purposive, broad and liberal interpretation, the Tribunal finds that a similarly broad and liberal approach, using the analogy of the spectrum, is appropriate to a decision concerning the reasonable reliability of the job information. While the job information may not meet the degree of reliability that should normally be sought for a "pay equity" situation, is it "adequate", as Dr. Wolf indicated it was, for this specific situation? Alternatively, should the job information used by the Professional Team, with its various deficiencies, be dismissed as being entirely worthless, and as absolutely without merit, along the lines of Mr. Willis' opinion?

[684] The Tribunal believes these questions are best answered in the context of the total job evaluation undertaking - that is the job evaluation plan selected, and above all, the evaluators involved.

[685] Given the very considerable job evaluation experience of Drs. Wolf and Ingster of the Professional Team, including their application, over many years, of the Hay system to a wide-range of jobs involving a variety of job information, the Tribunal considers that their opinion concerning the reliability of the available job information was particularly compelling.

[686] Moreover, Dr. Wolf was not hesitant to identify deficiencies in instruments such as the Job Fact Sheet and certain job descriptions. He also demonstrated an ability to adapt to the situation before him as illustrated in his remarks about being "selective" in using data included in the Job Fact Sheet (paragraph [609]). He obviously knew how to avoid the most offensive aspects of that document. He and his two colleagues were therefore very aware of the imperfections, including certain inconsistencies and even incompleteness, in the job information, but still concluded that the material was "adequate" for the work being performed by the Professional Team.

[687] The Professional Team also benefited from the augmentation of the FACTS I base data by means of the FACTS II material. Examples include the new working conditions information that arose from the telephone interviews with the CR sample members contacted. Another, was the re-evaluation of the 10 PO `generic' jobs in the year 2000 based on considerable new evidence from a number of Canada Post witnesses, resulting in two significant job value revisions.

[688] In responding to a question from Alliance counsel about his "comfort level" with the overall evaluations performed by the Professional Team, Dr. Wolf replied as follows, implying that the job information was at least adequate:

I feel that these evaluations are valid representations of the particular jobs at hand. I wouldn't have evaluated the job if I didn't feel we could evaluate it in a meaningful and appropriate way. That's why we punted on the four we punted on, because we felt we just couldn't accurately evaluate those jobs.¹⁹¹

[689] The Tribunal must confess that navigating the job information through the straits of "reasonable reliability" has not been a relaxing passage. Yet, balancing the evidence presented by all parties and expert witnesses, and under the unique circumstances of this case in the realm of proscribed discrimination human rights legislation, the Tribunal finds that the job information, in the hands of the Professional Team, was more likely than not, "reasonably reliable", or "adequate" as that Team described it, despite certain imperfections.

(ii) Definitions and Sub-bands of Acceptability

[690] Recognizing the significance of this conclusion, the Tribunal wishes to record how it decided to handle what it considers to be one of the most challenging aspects of this case. To assist itself in analysing the many facets of the available job information and testing for the "reasonable reliability" of that information, the Tribunal found it necessary to re-examine the definitions with which it was working.

[691] Firstly, the expression used by Dr. Wolf in describing the Professional Team's job understanding was " ... adequate but not necessarily ideal...". The Tribunal drew on Webster's and Oxford's definitions of the word "adequate" in paragraph [485]. Both dictionaries included "sufficient" as one of their definitions of "adequate". In turn, both dictionaries have defined "sufficient" as being "adequate" or "enough". The word "ideal" is defined by Webster as "a standard of perfection or excellence", and by Oxford as "conceived as perfect in its kind".

[692] How do these definitions compare to the meaning of "reasonably reliable"? Webster defines "reasonable" as "not exceeding the limit prescribed by reason, not excessive, moderate". Oxford defines "reasonable" as "not going beyond the limit assigned by reason, not extravagant or excessive, moderate". Webster defines "reliable" as "consistently dependable in character, judgements, performances or result". Oxford defines "reliable" as "in which confidence may be put, trustworthy, safe, sure".

[693] "Reasonably reliable" job information can therefore, be interpreted as being job information that is consistently, moderately dependable or in which moderate confidence can be put. The words "adequate" and "sufficient" are interchangeable. While some might consider "consistently, moderately dependable" or "moderate confidence" to be more demanding than "adequate" or "sufficient", in terms of the level of quality, the Tribunal concluded that they are generally equivalent for the situation at hand.

[694] Accordingly, the Tribunal regarded the term "reasonably reliable" and the words "adequate" and "sufficient" as being interchangeable for the purpose of determining the state of reliability of the job information available in this case. Obviously, the word "ideal" sets a standard of correctness well beyond the standard of "reasonable reliability".

[695] Having clarified the terminology governing the standard of reliability, the Tribunal concluded that it is more likely than not that there is no one exact point that represents "reasonable" reliability, or "adequate" reliability. Rather, it is more likely than not that a range or band of acceptability represents what is meant by "reasonable" or "adequate" reliability. Some candidate items may fit more comfortably than others within that band, but all that pass the test of entry are considered to be reasonably or adequately or sufficiently reliable.

[696] While the Tribunal concluded that it is difficult, and probably unwise, to attempt to be quantitatively precise about the width of the range or band of acceptability, it found that it was administratively useful to think in terms of three possible sub-bands. The first sub-band represents the upper percentiles of the band, the second sub-band represents the mid-percentiles, and the third sub-band the lower percentiles. The Tribunal called these respectively, "upper reasonable reliability", "mid reasonable reliability" and "lower reasonable reliability".

[697] In undertaking its study of the massive sea of material and testimony of this multi-year case, the Tribunal evolved to the view that ultimate fairness to all parties in a "pay equity" case would probably be achieved when the quality of the job information concerned fell comfortably into the "upper reasonable reliability" sub-band. The higher the sub-band level within the band of "reasonable reliability", the higher the quality and the more accurate the eventual values attributed to the jobs and positions involved - at least in theory.

[698] Thus, while all three sub-bands meet the test of "reasonable reliability", the upper sub-band meets the test more abundantly and should, in the Tribunal's view, be the preferred choice for a "pay equity" situation.

[699] At this point, the Tribunal asked itself, into which sub-band would it place the job information which the Professional Team employed in this case? Given the number of reservations and imperfections in the available job information already identified in the foregoing review, the Tribunal concluded that that information could not be comfortably classified as "upper reasonably reliable". Nor was the Tribunal content to accept the job information as fitting, however tightly, into the "mid reasonably reliable" sub-band. The Tribunal did, however, agree that the most suitable home for the job information was the "lower reasonably reliable" sub-band.

[700] Hence, the Tribunal found, as stated in paragraph [689], that it was more likely than not that the job information utilized by the Professional Team in conducting its job evaluations of the CR and PO positions/jobs pertinent to this case, was reasonably reliable, albeit at the "lower reasonably reliable" sub-band level.

VII. WAGE GAP AND WAGE ADJUSTMENT METHODOLOGY

A. Introduction

[701] Having found that it is more likely than not, that the "off-the-shelf" Hay Plan being used in the traditional factor comparison methodology, the process followed and the job information utilized by the Professional Team in conducting its CR and PO positions/jobs evaluations were reasonably reliable, the next questions to be addressed are:

How reliable were the resulting job evaluation values attributed by the Professional Team to the CR positions and PO jobs concerned?

Was a "wage gap" demonstrated between the female and male predominant groups performing work of equal value?

[702] The Tribunal has already established the credibility of the three members of the Professional Team, having noted their qualifications in paragraph [382]. More particularly, Dr. Wolf was qualified as an expert in Hay-based job evaluation and Hay-based compensation.

[703] Accordingly, the Tribunal concludes that it is more likely than not that the aforementioned reasonably reliable Hay Plan, process and job information, in the hands of competent evaluators, as were the Professional Team, would result in reasonably reliable job evaluation values being attributed to the work performed by CR and PO employees.

[704] The Professional Team in its Final Report on the Hay Method Evaluation concluded:

Having found that a substantial portion of the CR jobs are of a value equal to, or greater than, that of the PO jobs, the logical next step was to identify the nature of the wage gap, if any, between the male-dominated PO jobs and the female-dominated CR jobs.¹⁹²

[705] In comparing its CR and PO job evaluation values with CR and PO hourly compensation rates, the Professional Team stated in its Report (Exhibit PSAC-30), that it did so for each of three years: 1983, representing the year the Complaint was filed; 1989, the year the Commission used for its wage analysis, and 1995, the year of the Professional Team's Report. The hourly wage rates were supplied by the Alliance and were assumed to be correct. The top rate was used in all cases.

[706] As its wage adjustment methodology, the Professional Team employed a level-to-line method in which a male wage line was developed against which female pay rates were compared. The Team used this level-to-line approach because it is an indirect comparison frequently used in "pay equity" situations involving large organizations as distinct from a direct job-to-job approach. The male wage line represents a "pattern" of male jobs plotted on a graph, using total job evaluation point values as one axis of the graph and wages as the other. Once this is done, the intersection of point values and wages for the female jobs, plotted similarly using the same graph, is compared to the male line at particular intersection positions.

[707] The Professional Team developed eight different male wage or pay lines which it designated Approaches A to H inclusive, including one for each of PO-INT jobs (Approach A), PO-EXT jobs (Approach B), and an average of PO-INT and EXT jobs (Approach E), for each of the three chosen years. One job-to-job comparison was also undertaken, or "position matching" as the Team called it (Approach H).

[708] Analyses by the Professional Team of the pay relationships between the male-dominated PO jobs and the female-dominated CR positions revealed a wage gap no matter which level-to-line method was used. A similar result occurred with the "position matching" approach. These findings were equally true for each of the years 1983, 1989, and 1995.

[709] The Professional Team concluded that "all of these approaches show a significant gap between the wages paid to CR's and to PO's performing work of equal value".¹⁹³ Although there were differences in the size of the wage gap across the various approaches, the Professional Team characterized these differences as "relatively minor."

[710] From a compensation perspective, the Professional Team's expert opinion was that the best measure of the pay discrepancy between the PO and CR jobs/positions was based on the PO-INT male wage line (Approach A) or the "position matching" option (Approach H). The Team felt this was so because the INT jobs seemed to be a closer match to the CR positions in terms of job content.

B. Submissions of the Parties

(i) The Alliance

[711] The Alliance submitted that once a wage gap has been established, it is necessary to decide upon the most appropriate method of closing it between the CR's and the PO's. It also submitted that closing the gap is best done by means of employing a "wage adjustment methodology".

[712] Of the eight male pay lines drawn up by the Professional Team, the Alliance's preferred approach was an adjustment to the average of the INT and EXT pay line - Approach E - because the Alliance felt it best represented the pay practices of the male comparator group. As already noted, the Professional Team's preference was the INT pay line (Approach A) though it accepted that Approach E was also workable.

[713] The Alliance's witness, Mr. Terrence Ranger, whose evidence was heard in November 1995, supported the Alliance's position. The Tribunal was not asked to qualify Mr. Ranger as an expert witness.

[714] Mr. Ranger stated that he was, at the time he gave his evidence, employed by the Alliance as Section Head of Research in the Collective Bargaining Branch. This branch included six research officers who served on negotiating teams and provided background support for bargaining demands. The officers also undertook compensation analysis in the areas of direct wages and benefits such as pension, dental and health care plans. He was first engaged by the Alliance in 1976 as a senior research officer. He confirmed that he had been called as a witness before the tribunal which handled the Treasury Board case.¹⁹⁴ There, he had been asked to calculate the monies required to close a wage gap that had been determined by one of the statistical consultants involved in that case.

[715] Mr. Ranger testified that he had reviewed Dr. Lee's Report¹⁹⁵ and the Professional Team's Reports¹⁹⁶, and had, as his objective, to cost for each year of the Complaint, the payments that would be necessary to close the wage gaps that had been determined by the Professional Team in its eight different approaches. Since the Team's calculations had been limited to the years 1983, 1989, and 1995, Mr. Ranger indicated that he produced pay lines based on the Team's methodology for each of the years from 1981 to 1995. The starting year of 1981 reflected the year that Canada Post became a Crown Corporation. He used what he called "annual equivalent" direct wage rates for each 12-month period and excluded "indirect wages" from his calculations in line with Dr. Lee's conclusions.

[716] In effect, Mr. Ranger testified, he had replicated what the Professional Team had done, using the same methodology, with the addition of employee population levels. These, he understood were Canada Post numbers which had been obtained either by the Alliance or directly from the Agreed Statement of Facts. For the years 1981 to 1985, he assumed the population figures would be the same as for 1986. Mr. Ranger then multiplied the wage gap identified by the Professional Team for each of the eight approaches, by the number of working hours per year and by the number of employees concerned.

[717] Mr. Ranger's conclusion was that the Professional Team's Approach E - the average of INT and EXT male pay lines - offered the best representation of the comparator population and therefore, the most appropriate remedy for this case. He recognized that the payout would be less under Approach E than under the Professional Team's preferred Approach A.

[718] Mr. Ranger cautioned that the employee population data he used should be regarded as "estimates" as the data were taken at particular points in time and were assumed to be full-time employees for the entire 12-month period. In reality, some full-time employees may not have been employed in a given position for the full 12 months of a year, and some are believed to have been part-time employees. Consequently, this would, Mr. Ranger indicated, impact on his costing.

[719] Mr. Ranger went on to testify that the total possible wage gap payout to CR employees that he had calculated as being just over $124 million (excluding "indirect wages"), should be regarded as "an estimate", and maybe even "...a ballpark estimate that's on the high side".

[720] He also testified as follows:

I am not suggesting that the $124 million be the settlement cost; it is some indication of the global costing of this complaint, but I don't think that anyone here could come up with that amount. I think that what should be identified are the hourly amounts that are required and then each individual employee will receive the amount that is due to them based on those hourly amounts and the amount of time they worked during the period. The final cost in fact won't be known to us, to anybody, until all of this is implemented.¹⁹⁷

[721] When he was cross-examined by Canada Post's counsel, Mr. Ranger agreed that any equal pay adjustments should be "pay for all purposes". He explained that this meant the adjustment would also include any statutory remittances that Canada Post might have to make arising from the basic adjustment. He cited examples such as the employer's obligations involved in items like superannuation, employment insurance and health tax. Although these would add to Canada Post's costs, they had not been quantified by Mr. Ranger.

[722] The Alliance was strongly supportive of the "pay for all purposes" concept and submitted that monies payable to CR's to close the wage gap must be so classified. Appropriate adjustments would have to be made to reflect the cost of such statutory remittances.

[723] Under further questioning by Canada Post's counsel, Mr. Ranger confirmed that for purposes of his wage gap calculations he used the maximum wage rate allotted in the range of wage rates for a given position or job. Although positions or jobs with particular evaluation points will fall into specific wage bands and incumbents may be earning different rates within a given band depending on length of time in that position or job, for Mr. Ranger's quantification of the cost of closing the wage gap, he assumed each job or position was paid at the highest level allowed.

[724] The Alliance also submitted that the "fold-in" principle should be implemented at the time of the wage adjustment to close the wage gap. The submission was that, as of the date of the final Tribunal Decision, an adjustment would be made to the base wage rates in the CR collective agreement such that wages for CR's would be the same as wages for PO's for work of equal value. The adjustment would be folded into the base CR wage rate.

[725] As an alternative to Mr. Ranger's endorsement of the Professional Team's wage adjustment methodology, the Alliance submitted that the job-based methodology of Dr. Kervin would be acceptable for this case. The Alliance considered that Dr. Kervin's approach produced very similar wage gaps to those found by the Professional Team and Mr. Ranger.

[726] In conclusion, the Alliance submitted that the final actual costing of the wage gaps will require additional work. For example, an examination of individual employee records held by Canada Post will be necessary. Additionally, the Professional Team's changes in certain PO job evaluations, arising from its review, in June 2000, of the evidence of Canada Post witnesses, will impact on wage gap calculations. These changes were identified in the Professional Team's Report (Exhibit PSAC-180).

(ii) The Commission

[727] The Commission submitted that an award of lost wages is warranted to address the wage gap that has been demonstrated to exist between the CR and the PO employee groups in this case. It was also submitted by the Commission that the most appropriate wage adjustment methodology to use in calculating this particular award is a level-to-line approach using a combined male line such as the Professional Team's Approach E.

[728] It was noted by the Commission that most "pay equity" cases which are systemic in nature involve occupational groups which do not always lend themselves to direct comparisons of the value of the work performed. In the instant case, since the jobs of the PO-INT and PO-EXT sub-groups do not provide direct comparators for the CR positions, the Commission submitted that, as permitted by section 15 of the 1986 Guidelines, an indirect comparison between female and male work must be used. This, argued the Commission, involves the construction of a male pay line using regression analysis.

[729] The Commission also argued that such an indirect comparison, by means of the level-to-line wage adjustment methodology, would be compatible with what Drs. Wolf and Kervin and Mr. Ranger had recommended. It would also be consistent with the approach taken in the Treasury Board case.¹⁹⁸

[730] The male pay line, according to the Commission's submissions, should combine all relevant male data available. This would best be accomplished using Approach E from the Professional Team's options. It would also accord with the choices of Dr. Kervin and Mr. Ranger, and is consistent with section 14 of the 1986 Guidelines. That section notes that when a comparison is made between the occupational group that filed a complaint alleging a difference in wages, and other occupational groups, those other groups are deemed to be one group.

[731] The Commission submitted that the most appropriate wage adjustment approach for the female job values was one based upon a level as opposed to a female wage line. Hence, the level-to-line designation. Dr. Wolf and Mr. Ranger had based their female level on the average of the evaluation scores of the positions within each CR level.

[732] The Commission submitted that, while neither it nor the Alliance had the ability to determine which positions comprised the jobs for the CR sample, Canada Post in May 1999 provided, through its witness Mr. Brian Wilson, a Report on rolling-up the evaluated CR positions into jobs.¹⁹⁹ Mr. Wilson, a Canada Post retiree since late 1995, had spent over 30 years with the company, having started as a letter carrier and moved progressively through the labour relations and human resources supervisory and management ranks.

[733] It was the Commission's submission that Dr. Mark Killingsworth, another of Canada Post's witnesses, then determined the average values of these jobs, using the evaluated CR positions and their gender predominance from Mr. Wilson's material. The Commission argued that it was from Dr. Killingsworth's determinations that Dr. Kervin, the Alliance's witness, was able to provide his expert advice on how to calculate the level-to-line "pay equity" adjustments.

[734] Dr. Killingsworth was qualified by the Tribunal as an expert witness in labour economics including econometrics. He obtained his Bachelor of Arts degree "with high distinction" in Economics from the University of Michigan in 1967. He then attended Oxford University on a Rhodes Scholarship, earning a B. Phil in Economics in 1969 and his D. Phil in Economics in 1977. At the time of his first appearance before the Tribunal in May 1999, he was Professor of Economics at Rutgers University in New Jersey, a position he had held since 1988. He was also serving as Research Economist for the National Bureau of Economic Research, having been in that post since 1984. He is the author of a number of publications on Comparable Worth (the U.S. term for "pay equity") and has undertaken considerable research in that and related fields.

[735] Because Dr. Kervin's analysis involved the use of CR jobs rather than CR positions, this is the Commission's favoured approach. The Commission argued that many experts agree that the preferred unit of analysis for equal pay studies is "jobs". A "job" is a collection of duties usually performed by several or many individual employees occupying "positions".

[736] Whether the "jobs" or the "positions" approach was used to calculate the female level, the Commission submitted that the level-to-line methodology was preferable in this case. It does, however, assume that each level is based on a representative sample of predominantly female jobs. The Commission reminded the Tribunal that Dr. Kervin's expert opinion was that the CR sample was adequately representative.

[737] The Commission's submissions underlined the use of the Hay Plan, whereby the random-sampled CR positions were evaluated in accordance with the four factors of skill, effort, responsibility required in the performance of the work, and the conditions under which the work was performed, as called for in subsection 11(2) of the Act. This resulted in the range of job values for each of the CR levels.

[738] The Commission concluded that Dr. Kervin's and Mr. Ranger's methodologies, both involving a composite male pay line, were the most appropriate. Accordingly, the Commission submitted two sets of calculations of adjustments required for each CR level to achieve equal pay for work of equal value for the years from 1981 to 2002.

[739] The first calculation was based on Dr. Kervin's methodology using the Professional Team's evaluation job values, under which he had determined a wage gap for the year 1995. From this base year, the Commission extrapolated wage gaps for each of the other years. The second calculation was that of Mr. Ranger's whose wage gap determination by CR level by year, was based upon the methodology employed by the Professional Team.

[740] The Commission indicated that the wage rates used in its wage adjustment calculations were based on those recorded in the relevant collective agreements. Specifically, the wage rates for the years 1981 to 1994 were those stipulated in Mr. Ranger's Report. Those for 1995 were drawn from the Professional Team's Reports. Those for the years 1996 to 2002 were taken directly from the collective agreements by the Commission, since neither the Professional Team nor Dr. Kervin provided calculations beyond 1995.

[741] The Commission indicated that while both Dr. Kervin's and Mr. Ranger's calculations provided satisfactory estimates of the wage gap for any given year, it preferred Dr. Kervin's wage adjustment estimates. Reasons for this preference included Dr. Kervin's use of an average job value in each CR level as opposed to the average position value of Dr. Wolf's team. Also the Commission considered Dr. Kervin's calculations to be more accurate and up-to-date than those of Mr. Ranger.

[742] The Commission acknowledged that both sets of calculations represented wage gap estimates only and the determination of actual payouts for individual CR employees must, of necessity, be subject to the examination of employee records with appropriate employer input.

[743] While the Commission argued that tribunals do not determine total individual payouts, the Tribunal, should it find the determination of evaluated job values to be reasonably reliable, must decide which wage adjustment methodology is appropriate, given the circumstances of the case.

[744] The Commission indicated its agreement with the Alliance's "fold-in" principle that, in making any back-pay adjustment, it would be folded into the CR base wage rates in its collective agreement²⁰⁰

[745] The Commission was also supportive of the "pay for all purposes" concept and submitted that:

...it is essential that the pay equity adjustments include not only adjustments to base salary, but also for all purposes, i.e. pensions, overtime, sick leave, acting pay, and long term disability payments.²⁰¹

[746] The Commission did, however, accept the expert evidence of Dr. Lee in concluding that there was no material difference between the non-wage compensation of CR's and PO's, based on current contracts for full-time employees. Dr. Lee had reviewed the "historical differences" for the previous 12 years and concluded that they could not be calculated reliably without a complete file of employee experience for each benefit provision. Where differences did exist, Dr. Lee considered them to be minor and, in most cases, without a wage equivalent value of significance for "pay equity" purposes. Therefore, Dr. Lee had presented his opinion that there was no overall difference in non-wage compensation that should be incorporated into the calculation of wage adjustments that would favour either the CR's or the PO's.

[747] It was also the Commission's submission that the exclusion of the PO-SUP subgroup of jobs from the PO occupational group during both the Investigation Stage and the subsequent "pay equity" process, had no bearing on the reasonable reliability of the wage adjustment methodology.

[748] The Commission urged the Tribunal to retain jurisdiction after submitting its Decision, to assist the parties as may be appropriate, in the event difficulties are encountered at a later date in determining the specifics of individual payouts.

(iii) Canada Post

[749] Canada Post submitted that the question of Remedy, including the wage adjustment methodology need not arise if the Tribunal were to decide in Canada Post's favour and dismiss the Complaint for one or more of the reasons it had already argued. However, Canada Post did choose to respond to the submissions of the Alliance and the Commission on remedial issues.

[750] Canada Post argued that the importance of a remedial award and its impact on all parties demands a high degree of confidence in the methodology used to determine any wage adjustments. In this regard, Canada Post considered the Alliance's wage adjustment calculations to be "exaggerated in every respect". Canada Post estimated that based on Mr. Ranger's methodology, the total award would be approximately $2.4 billion, and approximately $443 million based on Dr. Kervin's methodology.

[751] It was noted by Canada Post that the Alliance's submission and Mr. Ranger's calculations were based on the Professional Team's wage adjustment analysis which included development of a regression line for the male jobs by fitting a line through the PO data points manually by eye, using a ruler. The pay line was then extended beyond the range of the PO data by extrapolation which Canada Post stated was not considered an acceptable technique by the Equal Pay Division of Labour Canada.

[752] It was Canada Post's submission that there were four significant flaws in the Professional Team's wage adjustment analysis that rendered it inadequate as a basis for calculating appropriate wage adjustments in this case.

[753] The first flaw, in Canada Post's opinion, was the manual setting of the male pay lines. Dr. Wolf had testified that while there is a difference of opinion among statisticians as to the minimum number of observations required for the proper use of regression analysis, most regard less than 25 to 30 to be questionable. Dr. Wolf and his two colleagues had five PO-INT observations and five PO-EXT observations, derived from "generic" job titles. Yet, the Professional Team actually drew regression lines by hand based on this limited number of observations.

[754] The second flaw was the exclusion by the Professional Team of one of the five observations in drawing its PO-EXT regression line which it considered to be anomalous because the job concerned was paid significantly more that the other jobs. Canada Post submitted that there may have been a more plausible reason for its exclusion. If it had been included, it might, in Canada Post's view, have demonstrated that there was no relationship between wages and job values.

[755] The third flaw was the Professional Team's extrapolation of the PO-INT and PO-EXT pay lines which, Canada Post argued, demonstrated that the majority of the Hay point scores for the CR 2, 3, 4, and 5's fell outside the range of PO point scores. Only through extrapolation was it possible to make this comparison since there were no PO jobs that were equal in value to many of the lower value CR jobs. Accordingly, Canada Post was of the opinion that on the basis of the data collected by the Commission, it was simply not possible to make a wage adjustment for those lower value CR jobs.

[756] The fourth flaw related to the Professional Team's selection of wage adjustment models which Canada Post considered to be unjustified. Having developed seven level-to-line pay lines and one position-matched option for only three of the years involved, Canada Post argued that with little explanation, Dr. Wolf and his two colleagues recommended the most expensive option for each of the three years. Both the Alliance and the Commission preferred the lower priced average composite PO-INT and PO-EXT option.

[757] With respect to Mr. Ranger's analysis, Canada Post submitted that since it was based upon the Professional Team's work, the defects in the latter's reasoning and analysis were continued in Mr. Ranger's work. Further, Canada Post questioned Mr. Ranger's method of developing PO job pay lines for the years not addressed by the Professional Team, particularly where he based his pay lines on just two observation points.

[758] Canada Post submitted that as a result of the fundamental flaws in the analysis of the Professional Team, compounded by the questionable approach of Mr. Ranger, Mr. Ranger's analysis and calculations must be rejected out of hand.

[759] Canada Post argued that Dr. Kervin's wage adjustment methodology was also seriously defective and identified its four basic steps as follows:

STEP

A male pay line was determined using regression analysis of the male-dominated PO-INT and PO-EXT jobs.
An average evaluation points score value was determined for the jobs in each of the four female-dominated CR levels.
Steps 1 and 2 were used to obtain a "predicted male wage" for each CR level.
The difference between the actual hourly wage for each CR level and its "predicted male wage" for that CR level was determined which represented the amount of wage adjustment to be made for that CR level.

[760] Canada Post maintained that there were six flaws in Step 1, ranging from the fact that the male pay line included one of ten PO jobs that was not male-dominated but was between 33% and 53% female (the PO-4), to the fact that Dr. Kervin constructed his pay lines and computed pay adjustments for all years in 1995 dollars only.

[761] In Step 2, Canada Post demonstrated that, by assuming that all jobs in the same CR level 2, 3, 4, or 5, had the same "job value" (the average job value within that level), some difficult and even absurd anomalies resulted because of the substantial overlap of job values between CR levels. Canada Post argued that a census rather than a sample might, indeed, be necessary at the adjustments stage to ensure a fair and equitable wage adjustment.

[762] Canada Post faulted Step 3 of Dr. Kervin's methodology for treating all the employees in PO jobs as though they were men, and all PO jobs as though they were male-dominated, even though in Canada Post's submission, neither of these assumptions were true.

[763] Finally, Canada Post argued that Step 4 would result in all jobs in a given CR level receiving the same wage adjustment regardless of their actual job value or their actual female percentage. For example, if one particular job at the CR-2 level was not female dominated, that job would obtain the same wage adjustment as a CR-2 job that was female dominated. Canada Post alleged that only 69.1% of the CR-2 incumbents were female, and were therefore, only "weakly" and not "predominantly", female (effective year not specified). Yet, Dr. Kervin's approach would provide CR-2 jobs with the largest wage adjustment of any CR level.

[764] Canada Post also claimed that three of the four CR levels would, under Dr. Kervin's model, receive a wage adjustment substantially in excess of what Dr. Killingsworth derived in his "Model 1" as the maximum difference in pay that can be attributed to gender. This was unusual, argued Canada Post, as Dr. Kervin had, when presenting his evidence, fully endorsed and accepted Dr. Killingsworth's "Model 1".

[765] In conclusion, Canada Post submitted that a wage gap award based on Dr. Kervin's methodology would be seriously flawed and should be rejected by the Tribunal.

[766] Canada Post proposed an alternative wage adjustment methodology that would, in its opinion, avoid the flaws contained in the proposals of Mr. Ranger and Dr. Kervin while furthering the objective of section 11 of the Act. The basic principles of Canada Post's wage adjustment proposal were expressed along the following lines:

PRINCIPLE

Since the purpose of section 11 is to eliminate sex-based wage gaps, wages would only be increased for those jobs in a CR level that are strongly female.
Since section 11 requires wage adjustments only to the extent that there is unequal pay for work of equal value, adjustments should be made at the level of the job, where job value is reasonably precise.
Wage adjustments should correct only for wage differences attributable to the gender make-up of the job and should only close the gap between predominantly female jobs and jobs performed by men.
Wage adjustments should be made only as "back-pay" to compensate for past discrimination. The Alliance and Canada Post have agreed in their current collective agreement that pay rates are compliant with section 11 of the Act.

[767] To accomplish these principles, Canada Post outlined a detailed procedure which will not be described here. Suffice it to say that the procedure included use of Dr. Killingsworth's Models 1 and 7 for each year from 1981 to 2001, and of actual PO and CR wage rates including benefits and the 6.7% "paid lunch" allowance for the years in which it was in effect. It also provided for determining the percentage female in each predominantly female job.²⁰²

[768] Canada Post submitted that in the event the Tribunal ordered a wage adjustment in this case, it should direct the parties to use this approach in calculating the relevant amounts, while recognizing that the agreement of all parties would be required as how best to implement that approach.

[769] Two other issues were addressed by Canada Post. The first related to what has been termed "paid lunch", which Canada Post argued was included or assumed in the calculations of Drs. Wolf and Kervin and Mr. Ranger. This gross-up of the PO wage rates by 6.7% should, in Canada Post's view, be removed. Canada Post argued that the evidence of Messrs Edward Fournier and Harold Dunstan, two of its witnesses, demonstrated that this benefit was achieved through collective bargaining and awarded on the basis of enhancing productivity. Known as "pay for performance", Canada Post submitted, it was clearly related to productivity and not to gender. Therefore, Canada Post argued that, because subsection 16(a) of the 1986 Guidelines indicates that a difference in wages between male and female employees performing work of equal value is justified by different performance ratings, this particular difference cannot be part of a wage gap.

[770] The second issue was the impact on Canada Post of the collective bargaining and labour relations situation that prevailed for several years following the proclamation of the Canada Post Corporation Act in October 1981. Upon becoming a Crown Corporation, the then existing bargaining units and agents of the former Post Office Department remained in place until 1985 to assist in providing transitional stability. This, and the introduction of the federal "6 and 5" cost control legislation in 1982, pre-empted an early start by Canada Post on the expected reform of the collective bargaining process.

[771] Consequently, Canada Post has now argued, in this case, that it "cannot be fairly said to have either established or maintained wage differences before at least 1985."²⁰³

[772] Canada Post took this argument one step further by referring to the fact that the Canada Labour Relations Board did not release its decision on the appropriate bargaining units structure for Canada Post until 1988 and its effect was not felt until the next round of bargaining in 1989-1992. Hence, Canada Post's following statement:

Again, Canada Post cannot reasonably be said to have infringed section 11 because it had no real power to participate in establishing or maintaining wages until after that time.²⁰⁴

(iv) Reply Submissions of the Alliance

[773] In addressing Canada Post's criticisms of the manual fitting of the male pay line, the Alliance countered that while "the results may be somewhat less accurate" than regression analysis, the technique used by the Professional Team had been successfully used by Mr. Willis in his Washington State Study. The Alliance also understood the Professional Team's results were extremely close to those of Dr. Kervin.

[774] The Alliance argued that Canada Post's questioning of Mr. Ranger's work in generating PO pay lines for each year of the Complaint was "entirely without merit." Mr. Ranger's calculations were derived from basic mathematical principles and precisely replicated the Professional Team's methodology.

[775] The Alliance reiterated that when the analyses of the Professional Team and Mr. Ranger were conducted, it was only possible to compare CR positions to PO jobs. The work of both Mr. Wilson and Dr. Killingsworth made it possible for Dr. Kervin to gross-up the CR positions into CR jobs enabling him to undertake his analysis on this basis. Despite this difference, the Alliance claimed that Dr. Kervin's end results were "strikingly similar" to those of the Professional Team and Mr. Ranger, and in the alternative, the Alliance adopted Dr. Kervin's methodology.

[776] The Alliance made reference to the fact that a wage increase of nearly 20% had been granted by Canada Post to CR employees in 1995/1996 which would reduce the wage gap for the years subsequent to 1995.

[777] With respect to Canada Post's proposed alternative wage adjustment methodology, the Alliance noted that this approach had never been advanced explicitly during the hearing and that it remained uncosted. The Alliance indicated that it adopted the submissions of the Commission regarding the inherent problems with such an approach.

[778] The Alliance questioned Canada Post's use of subsection 16(a) of the 1986 Guidelines to justify excluding the value of the "paid lunch" in the calculation of PO wages on the basis of section 17 of the 1986 Guidelines. Section 17 requires that the reasonable factor invoked must be equally available to all male and female employees concerned. Canada Post failed, alleged the Alliance, to establish that performance pay was equally open to both the male comparator and the female complainant groups because the evidence before the Tribunal would indicate that CR performance pay does not exist.

[779] The response of the Alliance to Canada Post's defences that it should not be responsible for wage discrimination under section 11 of the Act for a number of years during the 1980's was that "[t]hese defences are unmeritorious". While Canada Post has always maintained that there was no wage discrimination, the Alliance argued that Canada Post initially maintained and later independently established discriminatory wages between male and female employees performing work of equal value, an offence under section 11. The Alliance also argued that Canada Post could have rectified the situation by making voluntary equal pay adjustments, outside of collective agreements and without affecting base wage rates, as did the Government of Canada in the JUMI undertaking. Canada Post chose not to do so.

[780] It was, therefore, the Alliance's submission that the specified time period during which Canada Post argued it lacked the power to establish or maintain wage differences should not be excluded from the ambit of this Complaint.

(v) Reply Submissions of the Commission

[781] The Commission acknowledged that it articulated its position on wage adjustment methodology after much of the evidence had been led. It was only after hearing the evidence of Mr. Wilson regarding the roll-up of CR positions to jobs, and having heard the expert evidence of Dr. Kervin, that the Commission felt comfortable taking a firm position on the choice of level-to-line methodology.

[782] The Commission submitted that Canada Post had misinterpreted the wage adjustment methodology used by itself and the Alliance by claiming that both are treating the average of each CR level as a "job". The Commission responded by stating that its wage adjustment methodology does not treat each level as a job but averages the job values by CR level because all CR's in a given level are treated the same for pay purposes.

[783] Canada Post's calculations of a possible total award based on Mr. Ranger's approach of approximately $2.4 billion, and of approximately $443 million based on Dr. Kervin's model, were questioned by the Commission. Neither supporting evidence indicating how the figures had been determined, nor evidence to enable one to replicate the calculations, were furnished, according to the Commission.

[784] In response to what the Commission regarded as grossly inflated calculations by Canada Post, the Commission submitted that it had undertaken its own calculations based on Dr. Kervin's methodology. The Commission stressed that its calculations were but estimates, as the exact amounts would be dependent upon a number of factors about which information is not known at this time. Examples are: the exact number of employees at Canada Post over each of the many years of this case, their length of service, and the individual entitlements associated with "pay for all purposes".

[785] Based on Mr. Wilson's reported CR population numbers and assuming a 1956.6 hour year, the Commission estimated the possible cost of correcting the CR wage gaps to be of the following order:

(1) Using the Courts of Justice Act interest rate:

compounded semi-annually approximately $527.5 million

simple annually approximately $357.4 million

(2) Using the Canada Savings Bonds interest rate:

compounded semi-annually approximately $375.2 million

simple annually approximately $301.1 million

[786] The Commission cautioned that it had made what it considered to be certain reasonable assumptions and choices in arriving at these estimates, for the purpose of providing the Tribunal with a more realistic indication of potential cost implications than what had been provided to date.

[787] The Commission submitted that Canada Post's attacks on Dr. Kervin's wage adjustment methodology can best be dismissed by addressing several key issues, as follows.

[788] The first issue was Canada Post's flawed interpretation of section 11 of the Act and a misunderstanding or refusal to accept section 13 of the 1986 Guidelines. It would seem, argued the Commission, that Canada Post was simply opposed to the manner in which section 13 of the 1986 Guidelines determines gender predominance and preferred some other method or the use of an alternative percentage cut-off point. The Commission believed that the jurisprudence supports its approach in applying section 13 of the 1986 Guidelines.

[789] The second issue was the use of the linear regression line. The Commission believed that Canada Post had repeatedly misunderstood the purpose of regression analysis in wage adjustment approaches. It is not, as Canada Post had argued, to "explain" pay, but rather to summarize the wages for the male and female groups, so that comparisons can be made "on average" in relation to a new measure of job value derived by means of a gender neutral job evaluation plan.

[790] The third issue was the use of extrapolation. Contrary to Canada Post's assertion, the Commission submitted that Dr. Kervin did not question the use of extrapolation. Nor did he state that there were too few jobs to calculate a regression line for the male PO jobs. What Dr. Kervin did say was that it would be more difficult to extrapolate if the line were not linear. In this case, however, the PO pay line was linear.

[791] The fourth issue was Canada Post's criticism of the level-to-line wage adjustment methodology which, in the Commission's opinion, provided "on-average fairness" within the existing CR classification structure. The Commission emphasized that one of the purposes of the level-to-line approach, as with any regression line model, is to allow wage adjustments where direct job-to-job comparisons are not feasible.

[792] The fifth issue was a misinterpretation of the significance of Dr. Killingsworth's Model 1 as it related to the calculation of "pay equity" adjustments. In asserting that the level-to-line approach can produce "absurd" results when compared with Dr. Killingsworth's Model 1, the Commission argued that Canada Post had failed to realize that Model 1 was fundamentally different from a level-to-line model and measured a different variable. It was the Commission's position that, while Dr. Kervin acknowledged that Model 1 could show the existence of a "pay equity" problem, he did not accept it as suitable for determining exact wage gap adjustments.

[793] With respect to Canada Post's own proposed alternative wage adjustment methodology, the Commission called it a "novel approach" but argued that, in formulating it, Canada Post had ignored or misinterpreted section 11 of the Act, the 1986 Guidelines and the case law. It appeared to the Commission to be an attempt to avoid the application of the Act and would lead to a result which would fail to advance the purpose of section 11 because it would not address systemic gender-based wage discrimination stemming from occupational segregation and the undervaluing of women's work.

[794] The Commission questioned Canada Post's statement that its proposed methodology had been used in the past, and particularly that it had been "widely supported," and argued that the source cited by Canada Post failed to substantiate this proposition. In the Commission's view, Canada Post's proposal had been neither used nor supported under any federal equal pay for work of equal value policy, nor in any Canadian jurisdiction.

[795] The Commission argued further that since Canada Post's proposal hinges upon the use of "percent female" instead of the concept of "gender predominance," it runs counter to what expert witnesses Drs. Armstrong and Kervin had advocated - addressing occupational segregation and the undervaluing of women's work as intended by section 11 of the Act.

[796] Finally, the Commission remarked that it had had no opportunity to attempt to replicate Canada Post's proposed adjustment methodology, and no evidence, expert or otherwise, had been called to elaborate upon it.

[797] In sum, the Commission submitted that the Tribunal should take an extremely cautious approach to Canada Post's wage adjustment methodological proposal and should reject it.

C. Tribunal's Analysis

(i) Preliminary

[798] The Tribunal has already concluded that it is more likely than not that the reasonably reliable Hay Plan, process and job information, in the hands of the competent Professional Team, would result in reasonably reliable job evaluation values being attributed to the work performed by CR and PO employees (paragraph [703]). In determining the value of the work performed by those employees, the Professional Team applied the composite of the skill, effort and responsibility required in the performance of the work, and the conditions under which the work was performed, all in line with the requirements of subsection 11(2) of the Act.

[799] Comparing the resulting CR and PO job evaluation values, the Professional Team found that a significant portion of the CR positions were of a value equal to or greater than that of the PO jobs. The next step was to identify whether or not there was a wage gap between the male-dominated PO jobs and the female-dominated CR positions after comparing the evaluation values and CR and PO hourly wage rates. The Professional Team concluded that there was a wage gap between CR's and PO's performing work of equal value.

[800] Establishing a wage gap in this context is a most crucial step since it is the wage gap that evidences the discriminatory practice prohibited by section 11 of the Act. It has been said several times in this Decision that the essential purpose of section 11 is to eliminate systemic discrimination - to achieve "pay equity" between male and female employees employed in the same establishment who are performing work of equal value. It is therefore, by demonstrating a difference in wages between such male and female employees that systemic discrimination is proven under section 11, on a balance of probabilities, provided the employer has not demonstrated that the difference is attributed to one of the reasonable factors prescribed in section 16 of the 1986 Guidelines, and also provided that the methodology used in determining the wage gap meets the requirements of the Act.

[801] The Tribunal accepts that the evidence of the Professional Team, both through the viva voce evidence of Dr. Wolf and also through the presentation of the Team's Reports to the Tribunal, is sufficient, on a balance of probabilities, to demonstrate a wage gap when the work of the predominantly female CR's was compared with the work of equal value being performed by the predominantly male PO's at Canada Post. As Mme Justice L'Heureux-Dube indicated, when she wrote of the difficulty of comparing work of equal value in the SEPQA decision:

One element of difficulty is the concept of equality. The prohibition against wage discrimination is part of a broader legislative scheme designed to eradicate all discriminatory practices and to promote equality in employment. In this larger context section 11 addresses the problem of the undervaluing of work performed by women. As this objective transcends the obvious prohibition against paying lower wages for strictly identical work, the notion of equality in section 11 should not receive a technical or restrictive interpretation.²⁰⁵

[802] Further, the Tribunal accepts that Canada Post has not demonstrated that there was, in its personnel or wage policies, a reasonable factor prescribed in section 16 of the 1986 Guidelines which could explain this wage gap as caused by other than systemic sex discrimination.

[803] Having accepted that there is a wage gap, and, consequently, there is proof, on a balance of probabilities, that there has been systemic discrimination in this "pay equity" complaint, the next step is to select the most appropriate wage adjustment methodology to use to calculate an award of lost wages and to eliminate the gap. Given the many circumstances of this case, how appropriate are the wage adjustment proposals presented herein, and is there one or more that the Tribunal finds acceptable?

[804] To select the most appropriate wage adjustment methodology, it is helpful to be reminded of what Mr. Justice Evans said in the Treasury Board decision about the role of the Commission and the tribunal in such a choice:

Section 11 provides only a broad legal framework within which problems of wage discrimination between men and women are to be tackled in light of the facts of the particular employment situation, the evidence of expert witnesses, and the underlying purposes of the statute. In my view it would be inconsistent with both the underlying purpose of section 11, and the legislative record, to interpret the section as impliedly prescribing with the particularity suggested by counsel for the Attorney General the characteristics of the permitted comparative methodologies. Much must inevitably be left to be decided by the Commission and the Tribunal case by case, with the assistance of experts. (emphasis added)²⁰⁶

[805] The Tribunal has concluded from the evidence before it, and influenced by Mr. Justice Evans' Treasury Board decision, that testing for the appropriateness of the proposed wage adjustment methodologies is best handled in the following manner:

Are the methodologies compatible with the purpose of the Act, its Guidelines and its remedial provisions?
What expert evidence was heard and what supportive case law was identified with respect to each wage adjustment methodology?
What experience has the Commission had with such or similar methodologies?
Are the methodologies sufficiently compatible with how work and wages are structured and organized by the employer, so that should a Remedy be recommended, it could be implemented without undue difficulty?

[806] The Tribunal has identified three wage adjustment methodologies that it considers deserve assessment, as follows:

The Professional Team/Ranger Proposal in the PO INT - PO EXT composite version, labelled Approach E: This version was based on the Professional Team's methodology subsequently replicated by Mr. Ranger who also undertook calculations on 12 month periods; supported by the Alliance, accepted by the Commission, and found to be unacceptable by Canada Post.
The Kervin/Commission Proposal in a PO composite version: This version used the Professional Team's base material and Dr. Kervin's methodology in which he rolled up CR positions into CR jobs, courtesy of Mr. Wilson's and Dr. Killingsworth's workings; calculated by Dr. Kervin for 1995, all other years extrapolated by the Commission. Preferred option of the Commission, acceptable to the Alliance, unacceptable to Canada Post.
The Canada Post Alternative Proposal. Proposed by Canada Post to avoid what it considered to be flaws in the other two proposals; presented in Canada Post's final submissions. Acceptable to Canada Post, unacceptable to the Alliance and the Commission.

(ii) Review of the Wage Adjustment Methodology Proposals

[807] Canada Post submitted that its Proposal would further the objective of section 11 of the Act by not awarding compensation where there was no demonstrated wage gap or where a wage gap was not based on sex. Wages would only be increased for those jobs in a CR level that were strongly female. Adjustments would have to be made at the level of the job where job value is reasonably precise.

[808] However, the Commission argued that Canada Post, in developing its Proposal, had ignored or misinterpreted section 11 of the Act and the 1986 Guidelines, as well as case law, by failing to address systemic gender-based wage discrimination stemming from occupational segregation in the undervaluing of women's work.

[809] Unfortunately, there was no expert or other evidence led by Canada Post in defence of its position. There was, however, evidence from the Alliance's expert witness, Dr. Kervin, who argued that Dr. Killingsworth's Model 1, which constituted an element of Canada Post's Proposal, was not suitable for determining exact wage gap adjustments. Indeed, because Dr. Killingsworth's methodology ignored the gender predominance requirements in the Act and the Guidelines, his Models would not, in Dr. Kervin's opinion, be appropriate for the "pay equity" Complaint before the Tribunal.

[810] The Commission disputed Canada Post's statement that its proposed methodology had been used in the past and had been widely supported. It had, to the Commission's knowledge, never been used nor supported under the federal equal pay for work of equal value policy, nor in any Canadian jurisdiction.

[811] The Tribunal notes that Canada Post's Proposal lacked any estimated costing of its possible impact as a wage adjustment approach. It was also noted that the Commission had not had an opportunity to replicate the methodology of this alternative option.

[812] As a result, the Tribunal found it very difficult to assess this methodology vis-à-vis the other two Proposals, quite apart from attempting to put in context the many criticisms submitted by the Commission and endorsed by the Alliance. While one or two aspects of Canada Post's Proposal sparked the investigative interest of the Tribunal, it has concluded that it would be grossly unfair, not only to Canada Post, the initiator, but also to the other two parties, to attempt to make a decision about the appropriateness of a fairly complex matter, based on the very slim amount of factual evidence available and the complete absence of expert evidential input.

[813] Under these circumstances, the Tribunal is compelled to excise the Canada Post Proposal from further consideration.

[814] Both the Alliance and the Commission submitted that the two other Proposals, despite Canada Post's criticisms of them, were consistent with the purpose of the Act which, in line with Mr. Justice Laforest's 1987 decision in the Robichaud case, is remedial.²⁰⁷

[815] The Tribunal also recognizes the importance of addressing systemic remedies when one is dealing with systemic discrimination. Remedial measures should remedy the past, present and future effects of such discrimination. As Mr. Justice Dickson, then Chief Justice of the Supreme Court of Canada, pronounced in 1987 "statutes are deemed to be remedial and are thus to be given such fair, large and liberal interpretation as will best ensure that their objects are attained ... [and] the purpose of the [Canadian Human Rights] Act is ... to prevent discrimination".²⁰⁸ After a lengthy discussion of systemic discrimination and the methods necessary to combat it, the Chief Justice concluded that "it is essential to look to the past patterns of discrimination and to destroy those patterns in order to prevent the same type of discrimination in the future".²⁰⁹

[816] In 1996, Mr. Justice Hugessen noted that there was a "presumption that systemic discrimination will have produced the same effects in the past as it does in the present, although that presumption clearly becomes weaker the further it is extended into the past".²¹⁰

[817] The Tribunal accepts the Commission's argument that while wage adjustment methodology can be a fairly technical matter, its real purpose is to operationalize equality under section 11 of the Act and it should, therefore, not receive a restrictive, overly technical interpretation. Complementing the goal of equality is the discretionary non-prescriptive remedial authority of the Tribunal under subsection 53(2) of the Act. The scope of the Tribunal's remedial jurisdiction is, therefore, not only discretionary but also broadly based in all of the circumstances concerned.

[818] In this case, the Tribunal accepts the evidence of Mr. Durber, an early Commission expert witness, who dealt with the concept of "on-average fairness" as a determinative factor when dealing with the concept of "pay equity". He made the pertinent point that one is not dealing with one-to-one, job-to-job comparisons in seeking wage equality when large organizations and large numbers of employees are involved, but rather with employee group comparisons that provide "on-average fairness".

[819] To put Mr. Durber's comments about "on-average fairness" in context, it is helpful to refer to his testimony in May 1993. During his direct evidence, while he was explaining why it was not feasible to do a job-to-job comparison, Mr. Durber pointed out that "averaging means that there are jobs whose values are higher than that mean and jobs whose values are lower".²¹¹

[820] Additionally, Mr. Durber concluded that there were reasons why individual job-to-job comparisons were neither practical nor even desirable, in this particular case:

Fairness on average will end up with a system that is fair on average, that when one is designing a remedy, one probably cannot afford to design a remedy for all individuals because you are dealing with a systemic issue in which you are trying to see if whole structures ought to be adjusted on some reasonable basis. (emphasis added)²¹²

[821] Indeed, the Commission's booklet "Implementing Pay Equity in the Federal Jurisdiction" referred to in earlier sections of this Decision (Exhibit HR-1), identified the "wage line approach" which presents job values and wages for employees in male-dominated jobs on an average wage line. The Tribunal accepts that the level-to-line approach is appropriate for a "pay equity" case such as this Complaint, and notes that both Proposals under review supported the composite PO level-to-line male wage line approach.

[822] The Tribunal is cognizant that the Treasury Board case, in which the Commission was actively engaged, involved considerable evidence from expert witnesses called to give opinions concerning the selection of an acceptable wage adjustment methodology.²¹³ The tribunal in that case favoured a level-to-segment methodology after hearing from at least four witnesses with significant expertise in the subject matter.

[823] The Tribunal accepts that the two Proposals recognize that the wage adjustment device should be crafted in such a way that it can be implemented at the level in the organization at which the wage inequality manifests itself. In the current case, it is a question of adjusting the pay rates for the different CR classification levels because the goal of section 11 is to remedy discrimination within the existing job classification system of Canada Post. This will enable male predominant and female predominant employees, performing work of equal value in the same establishment, to receive equal pay.

[824] Both Proposals have relied, understandably, on certain core evaluation material generated by the Professional Team, the job information for which was categorized by the Tribunal as being of "lower reasonable reliability". Additionally, there has continued to be a question of how representative the CR sample was, in fact, given the different opinions of Drs. Kervin and Bellhouse. Dr. Kervin considered the sample to be sufficiently representative to provide adequate representation at each CR level for use in his methodology.

[825] With regard to the "paid lunch" issue, the evidence before the Tribunal was that the PO wage rates used in the calculations of the Professional Team, and of Dr. Kervin and Mr. Ranger, reflected the 6.7% additional value.²¹⁴ A similar addition to the CR wage rates was not applied, as "paid lunch" is not applicable to the CR employees. Furthermore, the Tribunal finds that no evidence was presented to support Canada Post's argument that the value of "paid lunch" should be excluded under subsection 16(a) of the 1986 Guidelines as a "pay for performance" provision, which would require equal applicability to both the PO and CR employees under section 17 of the 1986 Guidelines.

[826] There does, however, appear to be some disparity between what Canada Post called for in its own Alternative Proposal and what it submitted about the Professional Team/Ranger and Kervin/Commission Proposals, insofar as the "paid lunch" issue is concerned. The evidence indicated that a "paid lunch" allowance was included for PO employees for the years it was in effect in Canada Post's Proposal, but was questioned by Canada Post when included in the other two Proposals.²¹⁵

[827] Mr. Ranger and Dr. Kervin, and the Commission, all cautioned that their respective calculations have resulted in estimates only. Additional work will be required to achieve final costing of individual adjustments per CR employee after having access to employee records in consultation with Canada Post.

[828] Mr. Ranger made particular reference to his best estimates of employee population data which were taken at one point in time and were assumed to be all full-time employees. He also indicated he had used maximum wage rates throughout his calculations.

(iii) Sum-up

[829] Determining the most appropriate wage adjustment methodology upon which to apply the principle of equal pay for work of equal value enacted by Parliament in section 11 of the Act and sections 13 and 14 of the 1986 Guidelines, has not been an easy task. In the foregoing analysis, the Tribunal has examined the available methodology options presented by the Parties by means of the four eligibility criteria identified in paragraph [805] - that is, consistency with the purpose and remedial provisions of the Act and with its Guidelines; reference to expert opinions and case law; reference to Commission experience; and compatibility with the employer's work and wage structures.

[830] The Tribunal has identified three alternative wage adjustment methodology options that it has accepted for consideration. It has examined all three against the four eligibility criteria.

[831] Given the complexity of the subject matter, the Tribunal concluded that the assistance of experts familiar with "pay equity" and wage adjustment methodology was of particular importance in its consideration of each option, not only in terms of helping the Tribunal to understand the nature of each wage adjustment methodology, but also in terms of contextualizing each methodology while interpreting the purpose and principles of the Act and its Guidelines.

[832] Despite the Tribunal's interest in studying, further, Canada Post's Alternative Proposal, it was found to be based on very little factual evidence and the complete absence of expert evidential input. Under these circumstances, the Tribunal concluded that it was not a viable option for this case.

[833] The Tribunal found that each of the two remaining options - the Professional Team/Ranger Proposal and the Kervin/Commission Proposal meet, on a balance of probabilities, the four eligibility criteria and can, therefore, be considered appropriate wage adjustment methodologies for this case.

[834] The Tribunal, however, does not accept as conclusive the monetary values provided by the parties and witnesses for each of the two Proposals. The Commission and the Alliance, as well as the expert and other witnesses canvassed, have cautioned that their respective calculations are only estimates requiring additional work to achieve final costing of individual CR employee adjustments.

[835] Access to individual employee records in consultation with Canada Post will be necessary. Additionally, a number of variables may require detailed review. These may include the actual employee populations and their full time or part time status, the various wage rates used and their sources, and the individual employee entitlements for "pay for all purposes".

[836] Since the two Proposals meet the four eligibility criteria, and ignoring the cost implications of each, either of the two could be considered as appropriate wage adjustment methodologies for this case, subject to the additional work mentioned in the preceding paragraph.

[837] Which of the two is preferable? The Commission urged that the Tribunal accept the Kervin/Commission Proposal as "jobs" were used as the basis for its conclusions. More up-to-date information was also included in that Proposal.

[838] Dr. Kervin commented extensively on the necessity, in "pay equity" cases, that the notion of systemic discrimination be a foundation for decisions made. He indicated that an allegation of systemic discrimination, in the "pay equity" context, demanded a close scrutiny of "jobs" in the organization involved.

[839] This commentary reflects very much that of Dr. Pat Armstrong, who had introduced the Tribunal to the concept of "pay equity". She, too, had underlined the necessity to concentrate on "jobs" rather than on individuals, to see the concept of "pay equity" as a natural evolution of the history of business philosophy.

[840] As Dr. Kervin observed:

Pay equity is about systemic discrimination. It is about discrimination decisions that are not made on a one-to-one basis, but rather are inherent as part of the view of how an organization approaches jobs and their compensation. That is part of the reason it moves beyond equal pay for the same work, which would deal with non-systemic discrimination; in other words, discrimination targeted to some specific individual: "I don't like your mouth, so I am going to make sure you don't get paid as much".

In this case, we are talking about decisions that are made about jobs and the content of jobs and the important thing - and one of the reasons I believe the focus is on jobs rather than positions is that you want those decisions to be about work content.

If pay equity deals with decisions about work content and the value of that work content to organizations, then you can see that the decision takes place when you think about the job.

If there are no jobs, then it will take place at the level of the position. But there one has to be careful to make sure that the characteristics of the incumbent - height, eye colour, race, whether or not he or she is bald, those kinds of things - don't enter into the decision-making. Those are discriminatory behaviours of a different sort.

With pay equity, it's jobs of equal work. The job itself in terms of how much it's worth to the organization ought to be paid fairly based on that worth to the organization.²¹⁶

[841] Therefore, Dr. Kervin took from both Mr. Wilson's and Dr. Killingsworth's work their aggregation of the CR positions to create "jobs" with the characteristics of those positions rated by the Professional Team. Their work converted the 194 CR positions into a number of jobs by combining the information on all the positions into "some jobs". Dr. Kervin described this work as "aggregat[ing] the ratings of the positions into ratings of jobs" (Exhibit HR-93A).

[842] Using these jobs, according to Dr. Kervin, one can create a comparison of the PO jobs and the CR jobs, using a level-to-line technique, and from that comparison, determine how to close the wage gap.

[843] The male wage line is created, according to Dr. Kervin, by observing the wage data and the Hay points for the male-dominated PO jobs. Then, the "mean" or "average" value for each CR level can be calculated by using the information for the CR "jobs". The question then becomes, "what would the wage at the CR level be if the jobs in that category were paid according to the male wage line?" This new "female" wage is the wage necessary to close the gap.

[844] This emphasis on "jobs" is important to the concept of "pay equity". Based upon this ability to deal positively with the concept of closing the wage gap, while using information about the respective "jobs" involved, the Tribunal prefers the Kervin/Commission Proposal.

[845] The exclusion of statutory payments ("pay for all purposes") from the calculations in both Proposals is the subject of the next Section of this Decision entitled Non-Wage Forms of Compensation.

VIII. Non-Wage Forms of Compensation

A. Background

[846] The provisions of the Act pertinent to this Section of the Decision are found in subsections 11(1) and 11(7) of the Act, which have already been identified in the Decision but are restated, below, for ease of reference:

Equal wages

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. (emphasis added)

Definition of "wages"

11(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 of 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. (emphasis added)

[847] For the sake of clarity, the Tribunal notes that two terms have been used, somewhat interchangeably, in the evidence and the testimony to address the subject at hand. One is the term non-wage forms of compensation, the title of this Section of the Decision. The other is pay for all purposes.

[848] Non-wage forms of compensation, sometimes also called in the evidence "non-wage benefits", "indirect compensation", "indirect wages", "indirect remuneration", or even "non-cash wages" or "non-cash remuneration", is the term that has arisen most frequently to include all forms of compensation itemized in subsection 11(7) of the Act, other than base wage or salary.

[849] Pay for all purposes seems to have been used primarily in the context of Remedy. For example, in being questioned by Canada Post counsel whether or not Canada Post would have to make statutory remittances on any wage adjustments from a wage gap on items such as employment insurance, health tax, or pensions, Mr. Ranger responded, as follows:

A. It is certainly my view that the equal pay adjustments on the new rates are pay for all purposes.

Q. So that would include, in your view, making statutory remittances to any pertinent statute?

A. Yes.²¹⁷

[850] Another example is the Commission's use of the term on page 187 (Chapter 13) of its Submissions, as follows:

Thus, a pay for all purposes remedy will account for the reality that any benefit which is linked to the base salary will be likewise affected by any wage rate adjustments.

[851] In effect, the Tribunal is examining the question of "indirect compensation" at two levels. The first level is the impact of the non-base salary elements specified in subsection 11(7) on the definition of "wages". The second level is the inclusion in any wage gap adjustments, not only adjustments to the base salary or direct wage, but also to the non-salary elements, as may be appropriate.

[852] Presumably, this becomes an issue only if the non-salary elements have not been costed, and therefore, not included in the employer's definition of "wages". If they have been costed and included in "wages", their value will be reflected in any resulting wage adjustments. The employer will, understandably, have to deduct from an individual employee's gross wage adjustment such items as relevant income tax and incremental pension plan contributions. The employer will have, however, an additional liability in terms of making remittances against any wage adjustments that may be required under statutory-based non-salary obligations which are jointly supported by employee and employer, such as pension plan contributions.

[853] It is important to note that in determining that there was a `wage gap' as described in the preceding Section of this Decision, the Professional Team's analysis relied on the base "direct wages" of the CR positions and PO jobs which did not include the non-wage benefits. The Commission, in undertaking its job evaluations during the Investigation Stage of the Complaint, had not specifically costed the non-wage benefits.

[854] All parties did acknowledge, however, that non-wage compensation, within the meaning of subsection 11(7) of the Act, must be considered in an equal pay complaint. Indeed, the Alliance engaged, in 1995, an expert resource in the person of Dr. Don Lee, to address the costing of these components for the 12-year period of 1983 to 1995. His Report is the subject of review later in this Section of the Decision.

B. Submissions of the Parties

(i) The Commission's Position

[855] The Commission's Final Investigation Report, dated January 24, 1992, included the following sentence in paragraph 58:

There appears on balance to be a wage gap in non-cash remuneration insofar as the available evidence indicates.

[856] Mr. Paul Durber, giving his evidence-in-chief in June 1993, was asked by Commission counsel to elaborate on that statement in the Final Investigation Report. The following comments were included in his response:

We will recall from some of the earlier evidence that section 11, of course, requires that we look not only at salaries but all forms of remuneration. That is, what is referred to here as non-cash wages, indirect remuneration.²¹⁸

[857] When asked by Commission counsel what effect such non-cash remuneration would have on the alleged wage gap, Mr. Durber responded as follows:

It would appear on balance to enlarge the gap somewhat. It is difficult to say by how much. We know that non-cash remuneration accounts for the smaller portion of total remuneration - that is, perhaps 30 per cent - and we know that much of that 30 per cent in turn is driven by direct wages...

(...)

So we don't have all the evidence. We know this investigation is incomplete on that score. We have not been able to get the information and we draw a very general conclusion that there may be an additional wage gap. We don't think that indirect or non-cash remuneration will reduce the gap...²¹⁹

[858] In its final written submission, the only reference the Commission appears to have made to non-wage compensation is an indirect one, in the context of wage adjustment methodology, when it stated as follows:

...it is essential that the pay equity adjustments include not only adjustments to base salary, but also for all purposes, i.e. pensions, overtime, sick leave, acting pay, and long term disability payments.

(...)

Pay for all purposes will ensure that all necessary adjustments are made in respect of all pay-related benefits and premiums. As a result, entitled employees should be compensated in respect of any monetary benefit which has a nexus to the base wage rate.²²⁰

(ii) The Alliance's Position

[859] The Alliance called Dr. Don Lee as a witness before the Tribunal in October 1995. He was qualified by the Tribunal as an expert in contract analysis and non-wage compensation valuation.

[860] Dr. Lee obtained his Bachelor of Arts degree in 1968 from the University of Waterloo and his Master of Arts in political science from Queen's University. He later studied under a Canada Council Fellowship at the London School of Economics from which he was awarded a Ph.D. in political studies in 1980. He began his career as an Actuarial Assistant with a competitor of the Hay organization, handling the calculations associated with pension plan valuations, work which was compatible with his earlier mathematical studies at the undergraduate level. Subsequently, he served as Assistant Director of Research and Legislation with the Canadian Labour Congress and was engaged in the development of pension plan benefits policy and the provision of technical advice to unions in bargaining with respect to pension plans and other benefits such as retirement health insurance coverage. In the period of 1977-1978, he was contracted by the Ontario Federation of Labour to coordinate its participation in the Ontario Royal Commission on Pensions. Dr. Lee has been operating as an independent consultant since 1979. His firm, Union Pension Services Ltd., focuses on two principal aspects of retirement pensions and related benefits: firstly, providing technical support to unions at the bargaining table and secondly, offering education and training to union members. Dr. Lee's clients have included unions in many prominent Canadian industries and several Ontario public sector unions.

[861] Dr. Lee stated that he had been contracted by the Alliance in June 1995 to compare the non-wage forms of compensation of the CR group and the PO EXT and PO INT sub-groups, for the lifetime of the Complaint from August 1983 to the summer of 1995. He indicated that he had examined a variety of federal government benefit plans and certain consultants' reports on non-wage compensation. He also reviewed 14 collective agreements covering this period of time for the CR and PO employees concerned. Dr. Lee's Report has been designated as Exhibit PSAC-55.

[862] Dr. Lee reported that he had been instructed by the Alliance to exclude from his study those benefits provided for in paragraph 15(1)(f) of the Act. These are benefits which are provided but that provision is not deemed to be a discriminatory practice. Paragraph 15(1)(f) reads, as follows:

15 (1) It is not a discriminatory practice if....

(f) an employer, employee organization or employer organization grants a female employee special leave or benefits in connection with pregnancy or child-birth or grants employees special leave or benefits to assist them in the care of their children.

[863] In comparing the individual non-wage compensation provisions of the CR and PO groups, Dr. Lee determined where differences existed between the two groups and then calculated the value of those differences in wage-equivalent terms. To maintain a consistent framework, he classified some 51 non-wage provisions into 9 general categories of compensation.

[864] Dr. Lee stated that, in undertaking his study for the Alliance, he accepted the principle (which he understood the Commission also endorsed) that a determination of equality of non-wage compensation should be simple and workable for employers and comprehensible for employees. Therefore, recognizing the difficulty of being precise about valuating certain non-wage compensatory components, he considered differences representing less than 1/10^th of 1 % of wages not to be significant. Individual differences of this order were, therefore, not reflected in Dr. Lee's comparison calculations.

[865] Dr. Lee indicated that his detailed valuation analysis was confined to the non-wage compensation provisions included in the then-current collective bargaining agreements. He was, therefore, dealing with those agreements in effect in the summer of 1995, which concerned the full-time CR and PO employees.

[866] Dr. Lee confirmed, when giving his evidence, that his Report did not include, in his list of non-wage compensation items, the provision of uniforms and protective clothing for the PO employees which clearly favoured that group. He also did not include a provision for job security/technological change.

[867] Dr. Lee did, however, undertake a general review of what he called "historical differences" arising from collective agreements that were in effect from 1983 to 1994. His general review revealed that many of the "historical differences" were minor. Many more were temporary and were subsequently eliminated through the normal process of collective bargaining. He decided, therefore, that it was not feasible to calculate wage-equivalent values for such differences without a complete file of employee experience with each of the compensating provisions over the twelve-year period.

[868] The last paragraph of page 21 of Dr. Lee's Report reads as follows:

And finally, wherever any judgment enters directly into a calculation, I have attempted to overstate the value of the differences favouring the CR group and to understate the value of differences favouring the PO group.

[869] Dr. Lee illustrated the meaning of this point by citing an example where he was in doubt whether or not to classify a particular CR non-wage benefit provision as being equivalent to or better than the companion PO provision. Where he had to make a judgement call, he sought to classify the current CR provision as being better for the complainant group than the current PO provision was for the comparator group.

[870] Should his judgement call be questioned, he said he would be able to claim that he had erred on the side of underestimating the value of the specific benefit that favoured the CR group. He went on to say that "this is a sort of general principle of actuarial practice: when in doubt, make an assumption which tends to favour the opposite of your conclusion".²²¹

[871] Dr. Lee concluded from his study that, based on the then-current 1995 collective bargaining agreements in effect with full-time CR and PO employees concerned, there were essentially no differences in non-wage compensation between the two groups of employees.

[872] Specifically, he indicated as follows:

The essential point, I think, is that the overwhelming body of non-wage compensation provisions are currently the same or equivalent, and the extent to which they are the same or equivalent represents something like 34 per cent of wages.²²²

[873] He further concluded that there were no reasonable grounds for calculating differences in non-wage compensation for the two employee groups for the period 1983 to 1994 since any differences that were identified were relatively minor and often temporary. Moreover, a reliable calculation of the value of differences would require intensive study of employee experience for each of the compensatory provisions, over the twelve-year period.

[874] Accordingly, Dr. Lee concluded that there were no differences in non-wage compensation between full-time CR and PO employees that should be considered in determining if there were any "differences in wages" under subsection 11(1) of the Act.

(iii) Canada Post's Position

[875] Canada Post called Mr. Robert Bass as a witness in April 2000, and he was qualified by the Tribunal as an expert in costing compensation.

[876] Mr. Bass obtained his extended honours Bachelor of Science degree in mathematics and computers in 1974 from the University of Waterloo, under its co-op alternating classroom/workplace program. He began his career with the Toronto Board of Education where his primary responsibility was to support the Board's teachers' collective bargaining team, particularly with respect to complex costing issues. He became Director - Research and Information for the Ontario Hospital Association in 1977 where he developed a fully functioning research department whose role was to provide research and data support to the bargaining teams of member hospitals. Mr. Bass developed a computer-based total compensation-costing model for the Association, tracking wages and other compensatory clauses in hospital collective agreements. In the early 1980's, Mr. Bass set up his own consulting firm, known more currently as Bass Associates Ltd., providing a full range of labour relations support to management clients in the public sectors in Ontario, Alberta and B.C. Clients have included a broad range of providers in fields of service such as education, health, policing, and retirement homes. Mr. Bass specializes in providing employers with the costing and database analysis that is particularly crucial to management in the collective bargaining process. Since the passing of Ontario's Pay Equity Act in 1987, Mr. Bass and his associates have become increasingly engaged in developing with clients, gender neutral pay equity plans, often province-wide, requiring the involvement of both employers and unions. Inevitably, such plans demand an accurate assessment of the total value of compensation packages (wages and non-wage benefits) for comparison purposes in the context of the collective bargaining process.

[877] Mr. Bass indicated in his evidence that it is not unusual for "costing" to be a major issue in addressing wage and non-wage benefits in collective bargaining negotiations. He testified that it is important for both the employer and the union/employees concerned to know the cost of particular demands that arise in negotiations. It is acutely important for the employer to have this information, as it is the employer who must ensure that the business has the ability to carry all costs negotiated.

[878] When one is dealing with collective bargaining situations, according to Mr. Bass, there is often a requirement that the parties to negotiations develop a total compensation-costing model for both wages and non-wage benefits. This model must address the cost of existing wage and non-wage benefits for a base year, and then cost precisely the improved or diminished benefits for the years being negotiated.

[879] This requirement for precision, according to Mr. Bass's evidence, may entail the use of data retrieval techniques, identification of assumptions, and computer simulation modelling to gather and analyse pertinent employee usage and other information for each benefit. Of critical importance in such valuations is the modelling of the rate of expected change and the cost of such change to the employer.

[880] Mr. Bass confirmed that his mandate from Canada Post was to review and critique Dr. Lee's Report and "...to look at its methods, assumptions and methodology and give my comments".²²³ Mr. Bass's Report has been identified as Exhibit R-547.

[881] Mr. Bass's critique of Dr. Lee's Report faulted it on several grounds.

[882] First, Mr. Bass indicated that Dr. Lee made a methodological error by basing his analysis on the then-current 1995 non-wage compensation provisions. Mr. Bass stated that he would have used 1983 as his base year because he would have been looking for the differences in non-wage benefits from the time the Complaint was filed.

[883] Second, Mr. Bass disagreed with Dr. Lee's decision to dismiss individual differences in non-wage benefits of less than 1/10^th of 1% on the grounds that the sum of a large number of small numbers can equal a large number. Mr. Bass would have included such differences in his analysis.

[884] Third, Mr. Bass was not in agreement with Dr. Lee's exclusion of job security from his analysis and referred to Dr. Lee's response to the question of job security as a benefit which he "...[had] not been able to attach any wage equivalent value to...".²²⁴ Mr. Bass said that, in his experience, job security was one of the principal issues in collective bargaining. He added that it had such import in collective bargaining that it often became the basis for trade-offs in the negotiating process. In this case, Mr. Bass indicated that the generous job security provisions in Canada Post's union agreements meant that it was important to cost job security. He believed that it could be costed and identified several basic steps for doing so.

[885] Fourth, Mr. Bass challenged the exclusion of non-wage benefits arising from paragraph 15(1)(f) of the Act. Although Mr. Bass acknowledged that Dr. Lee was following the direction given to him by the Alliance when he did not address these non-wage benefits, he indicated that the items excluded should have been costed. He based this opinion, again, on the fact that these items are often major issues in collective bargaining and, as such, should be deserving of valuation. This opinion was refined somewhat when Mr. Bass indicated that he would, at least, cost those benefits which go beyond minimum employment standards. He indicated that such benefits as paternity leave and leave for family responsibilities might be examples of non-wage benefits which are not sufficiently widespread to be considered to fall within the realm of minimum employment standards.

[886] Mr. Bass concluded that each of the four faults described above could, individually, distort the valuation of the non-wage forms of compensation applicable to the employee groups involved in this case. He considered all four faults to be "fatal flaws" in Dr. Lee's analysis, thereby rendering the results of his work as unreliable.

(iv) Tribunal's Analysis

[887] There is no dispute among the parties about the intent of subsection 11(7) of the Act. It defines what is meant by "wages", and includes therein those forms of non-wage compensation specifically identified in addition to the remuneration paid for work performed by individual employees. All parties also recognize that paragraph 11(7)(e) calls for "any other advantage received directly or indirectly from the individual's employer" to be included in the definition of "wages".

[888] As well, there is probably no dispute among the parties about applying a "pay for all purposes" Remedy, should one be called for. While Canada Post does not appear to have used the expression "pay for all purposes" in its submissions, it was, as noted earlier, used by the Alliance's witness, Mr. Ranger, in response to a question from Canada Post counsel.

[889] The Federal Court of Appeal, in its decision of March 18, 2004 in the `Airlines Case', reinforced what it called this "very broadly" defined term of "wages".²²⁵

[890] The Commission's investigation reached the very general conclusion that there was likely an addition to the wage gap between the CR complainant and the PO comparator groups when one took into consideration the non-wage benefits, or indirect remuneration. Although it offered no direct evidence of such, it did not think that the non-wage remuneration would reduce the wage gap in direct remuneration between the comparator and the complainant groups.

[891] The Alliance's position was that there was no difference in value in non-wage compensation between the CR and the PO employee groups. This position was underlined by the Report of Dr. Lee which was based upon the collective bargaining agreements in effect in 1995 with full-time CR and PO employees. A review of the "historical differences" between 1983 and 1994 led to Dr. Lee's conclusion that there were relatively minor and often temporary differences in value of non-wage compensation between the two groups for that period.

[892] Canada Post presented its position that Dr. Lee's Report was flawed and, therefore unreliable through the evidence of its witness, Mr. Bass. His evidence was a critique of the Lee Report, and offered no determination of valuations or comparisons of values of the non-wage compensation components of the employee groups. Rather, he indicated to the Tribunal that the flaws identified could have led to a deviation or a distortion in the results of Dr. Lee's work.

[893] As always, the Tribunal must consider reliability in the context of the circumstances involved in the Complaint and will use the standard of reasonableness, based on the civil standard of the balance of probabilities.

[894] As already noted, having been accepted as an expert in contract analysis and non-wage compensation valuation, Dr. Lee's mandate was to determine and compare the value of the non-wage elements of compensation for the CR and PO employee groups.

[895] Dr. Lee noted that, from the inception of the Complaint to 1995, when he delivered his Report, there had been 13 collective agreements involving Canada Post and the Alliance and the unions representing the PO's. The PO group was first represented by LCUC which later merged with CUPW. Four agreements of the 13 involved Canada Post and LCUC.

[896] Dr. Lee's main area of scrutiny was the current (1995) collective bargaining agreement for each of the complainant and the comparator groups. Although he was aware of the prior agreements, information was not available for certain non-wage benefits for some years, and sometimes it was simply not existent at all. For those benefits he could compare in prior years, he noted that the differences were minor over the years, and often were temporary.

[897] For the current (1995) period, Dr. Lee examined 51 provisions of the collective agreements involving the CR's and the PO's. He excluded eight provisions entirely and two partially pursuant to paragraph 15(1)(f) of the Act, as requested by the Alliance. Of the 41 remaining provisions, he classified 24 as giving the same or an equivalent non-wage benefit to each group. An additional eight provisions had had differences in past agreements but were currently the same or equivalent.

[898] Eight provisions favoured one or other of the CR's or PO's. Six favoured the PO group and two, the CR group. Dr. Lee had been instructed by the Alliance that, when in doubt, he should choose the option that would reduce any wage gap. He had tried to do this, as he noted, (paragraphs [868]-[870]) using his principle of actuarial practice "to favour the opposite of your conclusion".

[899] Based upon his comparison of the collective bargaining agreements and his calculations of the values of the individual benefits, Dr. Lee concluded that there was no substantial difference between non-wage compensation for the two groups with the exception of the uniforms and protective clothing allowance for the PO group, which he estimated "may amount to as much as 2.08% of wages".

[900] As noted earlier, Canada Post's expert in the costing of compensation, Mr. Bass, was mandated to critique the work of Dr. Lee. He was not asked to determine and compare the value of the non-wage compensation components of both employee groups.

[901] Mr. Bass faulted Dr. Lee's Report on several grounds and concluded that each fault identified could have changed Dr. Lee's conclusion by distorting the valuation of the non-wage forms of compensation.

[902] Mr. Bass indicated that the choice of year is critical when doing a comparative analysis of non-wage compensation, as the year chosen will be the basis for future collective bargaining and eventual agreement. In this case, he noted that the year the complaint was brought would have been a better choice. Mr. Bass would have liked to have been able to see clearly the changes in cost of the non-wage compensation as the years unfolded from 1983 to 1995, and, thereby, make note of what Dr. Lee called minor or transitory differences in order to come to a definitive conclusion.

[903] Mr. Bass did admit that he understood that information for certain non-wage benefits was either not readily available or not available at all for all the years concerned. In other cases, an automated capacity was not available, necessitating a manual and more expensive means of data retrieval. He stated that, while his methodology depended upon obtaining accurate data, it was not uncommon to encounter situations where employers felt they could not provide suitable data. In such circumstances, he believed that one had to probe deeper, and perhaps deal with employer representatives who are close to the working operation. In terms of voluminous manual records, Mr. Bass stated that he might be prepared to work with a reasonable sample size, provided he could be assured that he would get representative data.

[904] The Tribunal finds that Mr. Bass's first criticism of Dr. Lee's work relating to his selection of the base year of 1995 rather than 1983 has merit as a theoretical statement of the most suitable year to begin the analysis of non-wage compensation differences between groups. In the circumstances, however, this cannot be given such weight as to overcome the work on the available 1995 data. There was a lack of certain information. Other information was not readily available. Dr. Lee, a witness being called by the Alliance to give evidence at the Tribunal hearing, could not simply ask a representative of the Respondent, Canada Post, to supply information to him. Even if he had been able to do so, subsequent evidence demonstrated that some of that information was not available.

[905] Mr. Bass's second criticism related to Dr. Lee's decision to dismiss differences in value of non-wage compensation benefits of less than 1/10^th of 1%. Although the Tribunal understands Mr. Bass's point about the possible total cost of a series of individual benefits with less than 1/10^th of 1% difference, he gave no concrete evidence to illustrate what impact this might have on the definition of wages.

[906] Therefore, the Tribunal does not give significant weight to this second alleged fault in Dr. Lee's Report.

[907] Dr. Lee and Mr. Bass disagreed about the ability to evaluate a benefit such as "job security", Mr. Bass's third area of concern. While Dr. Lee identified job security as an employee-benefitting category of non-wage compensation, he concluded that there were not reasonable means of attaching a wage-equivalent value to such a benefit, as it depends on future usage which cannot be reliably predicted.

[908] Mr. Bass disagreed. He felt that the costing of job security was particularly important in industries where technological change is prevalent, such as in postal operations. He did, however, note that to cost such a benefit would require access to appropriate employee data such as the number of employees actually or potentially at risk of being declared surplus, the time elements involved, and other related factors.

[909] While both expert witnesses acknowledged that job security is a non-wage benefit, it is evident, under the circumstances of this case, that Dr. Lee did not have access to the necessary information to undertake a costing, however approximate, of its value. Nor was the timing of his study propitious in the midst of the current hearing.

[910] Given the nebulous nature of costing job security and the credibility of both witnesses in their respective fields of expertise, the Tribunal accepts that job security is probably one of those non-wage benefits that is likely to be of equivalent value to virtually all of the CR and PO employees. Furthermore, the evidence suggests that this could be particularly so in an industry where technological change is prevalent, such as a modern postal collection, processing and delivery organization.

[911] Finally, Mr. Bass indicated that a thorough comparison of value of non-wage compensation for the CR's and the PO's would necessitate the valuation of the various provisions excluded by Dr. Lee. They concern primarily leave, with or without pay, for maternity, paternity and adoption, as well as leave for family responsibilities or for parental needs.

[912] These provisions arise under paragraph 15(1)(f) of the Act and represent benefits which are deemed by the Act not to be discriminatory. For this reason, Dr. Lee testified that he had been instructed by the Alliance to exclude them from his analysis.

[913] Mr. Bass stated that he would have costed the paragraph 15(1)(f) benefits - or at least, those that were above the minimum employment standards. He said that it was not usually regarded as an achievement in collective bargaining circles to negotiate only up to employment standards. It would, however, be an achievement to negotiate and succeed with benefits that go beyond those standards and, therefore, vital to know the costed value of the benefits.

[914] In the circumstances of Dr. Lee's Report, Mr. Bass admitted that such a costed valuation would be impossible without the availability of the usage data for each of the benefits which he did not have. Therefore, Dr. Lee's acceptance of his instruction to exclude the provisions which fell under paragraph 15(1)(f) of the Act was understandable given the nature of the Complaint, the clear wording of the paragraph, and the unavailability of the necessary data.

[915] Similarly, without an ability to make a costed valuation, Mr. Bass's comment that a majority of the paragraph 15(1)(f) provisions appeared to favour the CR group was sheer supposition.²²⁶

[916] Nor was Dr. Lee above making a supposition concerning provisions which he was unable to value or which he believed were part of Canada Post's administrative practice. He commented, as follows, on benefits such as Career Development Leave with Pay, and Examination Leave with Pay, both of which were available to CR employees for many years prior to 1995 when the PO group negotiated their availability for its group:

...I find it difficult to imagine that Canada Post would not have allowed POs time off for important examinations up until their most recent contract.

I suspect that in the case of some of these differences, that there have been administrative management practices ... which have been in place, perhaps, for years but were never written into the contract...²²⁷

[917] Neither expert's supposition had a basis in the evidence heard by the Tribunal.

[918] The Report of Dr. Lee, however, does present the picture of the value of the non-wage compensation available to the complainant and comparator groups as being, more likely than not, of equivalent value and tied, in a negotiated pattern, to the value of the wages paid to the two groups.

[919] The critique of Mr. Bass has not persuaded the Tribunal that Dr. Lee's Report should be rejected. Indeed, it has underlined the need for availability of data and materials in order to do more precise work. In the circumstances of this Complaint, the ability to be more precise was substantially reduced because of a lack of suitable data necessary to that precision. This lack of precision is a long way, however, from stating that the evidence presented by Dr. Lee's Report cannot be accepted as proving, on a balance of probabilities, that it is more likely than not that the non-wage compensation of both the complainant and the comparator groups was generally equivalent.

[920] There remains one other issue to consider which relates to Canada Post's statement, on page 264 of Chapter 11 of its written submissions, viz:

Any wage-gap analysis under section 11 is incomplete and inaccurate without an analysis of all forms of non-wage compensation. A full analysis is not only necessary; it is a required element of Section 11, and therefore a required element of a prima facie case.

[921] The resulting question that the Tribunal must address is: Has a reasonably reliable analysis been undertaken by the Complainant with respect to the non-wage components of compensation to meet the requirements of section 11 of the Act?

[922] First, the Tribunal accepts, from the evidence, that the manner in which the examination of the non-wage components was handled in this case was far from what one would expect from a joint employer-employee "pay equity" study. It was even removed from what one might consider as normal, accepted practice. Neither the Commission nor the Alliance dealt with the matter during the Investigation Stage. It was then left to the Alliance, during the hearing process of the Complaint, to engage Dr. Lee, an expert in contract analysis and non-wage compensation valuation to undertake a detailed study. This involved, as already noted, the examination of the many collective agreements and benefit plans that existed during the life of the Complaint. These agreements and plans involved Canada Post and the two unions representing the CR complainant group and the PO comparator group.

[923] Dr. Lee's finding, as expressed in his Report PSAC-55, challenged methodologically by Mr. Bass, was that current contract provisions for most forms of non-wage compensation were either "precisely the same or generally equivalent". Where differences did exist, Dr. Lee found them to be minor and, in most cases, did "not have a wage equivalent value which would be considered significant for the purposes of pay equity". He went on to conclude that "there is no difference in non-wage compensation which should be incorporated into the calculation of adjustments" which may arise for the CR complainant group.

[924] Does Dr. Lee's work, despite Mr. Bass's criticisms which were reviewed earlier in this Section of the Decision, constitute a reasonably reliable response to the need to consider the non-wage elements under section 11 of the Act?

[925] During the Investigation Stage of the Complaint, no costing of non-wage benefits was done by the Complainant or by the Commission despite the fact that subsection 11(7) of the Act calls for an identification of these elements as part of the definition of "wages".

[926] Dr. Lee was engaged by the Complainant well into the hearing, and, as a result of the lateness of his employment, he had no ability to verify data pertaining to the benefits or to cost them. He did the best he could given the situation he faced.

[927] Given these circumstances, the Tribunal drew on the spectrum approach used earlier to deal with the reasonable reliability of the job information used by the Professional Team in its job evaluations (paragraph [696]). The Tribunal concluded that it could not categorize Dr. Lee's Report and its results as being "upper reasonably reliable" or "mid-reasonably reliable". The Tribunal finds, however, that his methodology and its results were "lower reasonably reliable".

[928] The fact that the analysis was accomplished at an awkward time, by one expert witness, and demonstrated that the wage equivalent values for the non-wage compensation components for the complainant CR's and the comparator PO's essentially balanced each other out, does not, in the Tribunal's view, mean that subsection 11(7) of the Act was not respected. Nor does it mean that the wage gap analysis was incomplete or necessarily inaccurate.

[929] The Tribunal, therefore, finds that the Alliance, through the evidence of Dr. Lee, has met the requirements of the Act in considering the non-wage compensation elements of subsection 11(7) of the Act as part of the definition of "wages".

[930] The Tribunal also finds that the Alliance has fulfilled this requirement of a prima facie case under section 11 of the Act by proving, based on Dr. Lee's evidence and, on a balance of probabilities, that there were essentially no differences in non-wage compensation between the subject CR and PO employees that should be considered in determining if there were any "differences in wages" under subsection 11(1).

IX. Remedy

A. Background

[931] The provisions of the Act dealing with the Tribunal's jurisdiction to award remedies are as follows:

Complaint Substantiated

53(2) If at the conclusion of the inquiry the member or panel finds that the complaint is substantiated, the member or panel may, subject to section 54, make an order against the person found to be engaging or to have engaged in the discriminatory practice and include in the order any of the following terms that the member or panel considers appropriate:

(...)

(c) that the person compensate the victim for any or all of the wages that the victim was deprived of and for any expenses incurred by the victim as a result of the discriminatory practice;

(d) that the person compensate the victim for any or all additional costs of obtaining alternative goods, services, facilities or accommodation and for any expenses incurred by the victim as a result of the discriminatory practice; and

(e) that the person compensate the victim, by an amount not exceeding twenty thousand dollars, for any pain and suffering that the victim experienced as a result of the discriminatory practice.

Special Compensation

53(3) In addition to any order under subsection (2), the member or panel may order the person to pay such compensation not exceeding twenty thousand dollars to the victim as the member or panel may determine if the member or panel finds that the person is engaging or has engaged in the discriminatory practice wilfully or recklessly.

Interest

53(4) Subject to the rules made under section 48.9, an order to pay compensation under this section may include an award of interest at a rate and for a period that the member or panel considers appropriate.

[932] Clearly, the Tribunal has been bestowed broad remedial powers under section 53 of the Act to remedy the effects of discrimination when a human rights complaint has been substantiated under the Act.

[933] As the Commission pointed out in its submissions, these remedial powers should ensure that the victims of discriminatory treatment or practice are "made whole." In addressing the assessment of the damages recoverable by a victim in either tort or human rights law, the

Commission referred to the following statement made by Mr. Justice Marceau in the Morgan case:

In both fields, the goal is exactly the same: make the victim whole for the damage caused by the act [sic] source of liability. Any other goal would simply lead to an unjust enrichment and a parallel unjust impoverishment.²²⁸

[934] One can, perhaps, best interpret "making the victim whole" as meaning restoring the victim to the position or status he or she would have been in had the substantiated discrimination not occurred.

[935] It is also pertinent to note, as did the Commission in its submissions, Mr. Justice Hugessen's examination of paragraph 53(2)(c) of the Act in which he and his colleagues were considering an equal pay for work of equal value case:

As I read this provision, it is a simple and straightforward authority to order the payment to a victim of lost wages resulting from a discriminatory practice. Such an order will always be backward looking and will result from the answer to the question 'what wages was this victim deprived of as a result of the discriminatory practice?' Nothing in the text justifies the view that such an award should be 'minimally afforded' or that its starting point should be restricted 'to the moment the complaint was filed'. A complaint of discrimination necessarily relates to practices which have predated the complaint itself; one can hardly complain of discrimination which has not yet occurred. Of course, the discrimination may be continuing so that the Tribunal will also order remedies for the future, but that fact should not blind us to the obvious need to remedy what has taken place in the past.²²⁹

[936] In the same aforementioned DND case, Mr. Justice Hugessen made the following statement, supported by the very extract identified earlier in paragraph [679] from S.M. Waddam's publication:

In my view, it is well settled law that once it is known that a plaintiff has suffered damage, a court cannot refuse to make an award simply because the proof of the precise amount thereof is difficult or impossible. The judge must do the best he can with what he has.²³⁰

[937] There is another important contextual factor to consider which relates to the possible impact on an award for damages of any uncertainty about the nature, extent and value of the losses involved. This matter was addressed by Mr. Justice Marceau as follows:

It seems to me that the proof of the existence of a real loss and its connection with the discriminatory act should not be confused with that of its extent. To establish that real damage was actually suffered creating a right to compensation, it was not required to prove that, without the discriminatory practice, the position would certainly have been obtained. Indeed, to establish actual damage, one does not require a probability. In my view, a mere possibility, provided it was a serious one, is sufficient to prove its reality. But, to establish the extent of that damage and evaluate the monetary compensation to which it could give rise, I do not see how it would be possible to simply disregard evidence that the job could have been denied in any event. The presence of such uncertainty would prevent an assessment of the damages to the same amount as if no such uncertainty existed. The amount would have had to be reduced to the extent of such uncertainty. (emphasis added)²³¹

[938] Two more recent cases have come to the Tribunal's attention which illustrate the principle enunciated by Mr. Justice Marceau. One is Chopra v. Department of National Health and Welfare in which the complainant sought compensation for wage loss arising from his failure to attain acting status and to qualify as an eligible candidate in a competition for a more senior position in his department.²³² The complainant was awarded damages by the tribunal which were very significantly reduced on the grounds of the relatively high level of uncertainty of his being successful in the final competition. This decision is currently the subject of a judicial review application before the Federal Court.

[939] The second case is Singh v. Statistics Canada [1998] in which the tribunal found that the complainant had been discriminated against because of his age when his name was not added to an eligibility list for particular positions at Statistics Canada.²³³ Damages were, however, based on providing him with a position and allowances two classification rungs below the one sought by the complainant and the CHRC, on the grounds that "it is by no means certain that Mr. Singh's progress would have followed" the path he claimed.²³⁴ The decision was upheld by the Federal Court.²³⁵

[940] While the presence of uncertainty in determining the extent of damages should not, indeed must not, inhibit the Tribunal from awarding damages, that uncertainty can, nevertheless, result in a reduction, under some circumstances very appreciable, in the assessed value of the damages.

[941] Given the classification, by the Tribunal, of the job information used in evaluating the CR positions and PO jobs, as "lower reasonably reliable," (paragraph [699]) the Tribunal finds there is present a significant degree of uncertainty. This uncertainty arises from the lowest rating on the "band of acceptance" which pre-empts an assessment of the wage loss damages to the amount that could be expected had the job information been rated at the "upper reasonably reliable" level - the most desirable level for a "pay equity" case.

[942] A similar further element of uncertainty arises from the classification, by the Tribunal, of the non-wage forms of compensation as also being "lower reasonably reliable" (paragraph [927]).

[943] Taking into account these elements of uncertainty which affect the very crucial aspect of determining the extent of the wage gap, it is, in the Tribunal's view, more likely than not that if the job information and the non-wage benefits had been "upper reasonably reliable," the resulting wage gap would have more accurately reflected reality. In other words, the greater the reliability of the job information and the non-wage benefits, the greater the accuracy of the wage gap determination. This determination is seminal to the extent of the award of damages.

[944] Recognizing these elements of uncertainty in the state of the job information and non-wage benefits documentation, the Tribunal finds that it cannot accept the full extent of the wage gap as claimed by the Alliance and endorsed by the Commission.

B. Remedial Components

(i) Award of Lost Wages

[945] On balance, the Tribunal favours the level-to-line Kervin/Commission methodology in determining the extent of the wage gap to be closed (paragraph [844]). Both Dr. Kervin and the Commission stated however, that their wage gap calculations were only estimates. Final costing of individual adjustments for each CR employee will require review of employee records in concert with Canada Post.

[946] As noted above, the Tribunal finds that the job information and the non-wage compensation issues have created uncertainty in the determination of the wage gap. That uncertainty, in turn, calls for a discounting of the award of lost wages.

[947] Unlike other sections of the Act, section 11 does not present a clear distinction between proof of liability and proof of damages. Nor does it present a clear methodology to measure damages arising after a finding of discrimination. It is, therefore, necessary for the Tribunal to address the distinction between proof of liability and of damages, as well as the means of measuring the possible damages.

[948] Following the spectrum analysis already completed for the two elements of uncertainty, the Tribunal concludes that a wage gap determination based upon "upper reasonable reliability" evidence should, logically, give rise to a 100% award of lost wages, a determination based upon "mid reasonable reliability" to a 75% award, and a determination based upon "lower reasonable reliability" to an award of 50% or less.

[949] Accordingly, the Tribunal concludes that the finally determined award of lost wages for each eligible CR employee, by whatever methodology, should be discounted by 50% in line with the lower reasonable reliability status of the relevant job information and non-wage forms of compensation.

(ii) Back-Pay - The Compensation Period

[950] The Commission has called for the awarding of lost wages back to October 16, 1981, the date of incorporation of Canada Post as a Crown Corporation. The Commission cited Mr. Justice Hugessen's decision in the DND case which recognized the ability of a tribunal to deviate from the Commission's practice of one year prior to the filing of a complaint, as follows:

In ordinary circumstances, the present limit set by the Commission's practice of one year prior to the filing of the complaint seems to me to strike a reasonable balance between the competing interests involved. Like any limitation period, it is, of course somewhat arbitrary and I would temper such arbitrariness by holding that it could be varied by a tribunal if the facts in any particular case indicated that a longer or shorter period was warranted.²³⁶

[951] The Alliance also cited this case and submitted that the effective date for the calculations of lost wages should be October 16, 1981. Anything less would be, in the Alliance's view, unfair to the victims of systemic discrimination.

[952] Canada Post submitted that the Alliance had not succeeded in establishing a prima facie case that any wage gap for which Canada Post could be liable under section 11 existed prior to the filing of the Complaint. Canada Post also pointed out that the tribunal in the Treasury Board (Phase II) case²³⁷, where the respondent admitted liability under section 11, did not award back-pay for any period prior to the filing of the complaint. The wage adjustment was ordered to run from the starting date of the JUMI Study - some 3½ months after the complaint had been filed. It was Canada Post's position that the facts in this current case support no retroactive payments.

[953] Mr. Justice Hugessen in DND indicated that the reach of paragraph 53(2)(c) of the Act "will always be backward looking" in ordering the payment of lost wages resulting from a discriminatory practice.²³⁸

[954] The Complaint was filed on August 24, 1983 but during 1984 and 1985, it was not actively investigated by the Commission. The Alliance and Canada Post were pre-occupied during this period with other matters, particularly the development of the proposed System One job evaluation plan. The Commission reactivated its investigation in October 1985.

[955] The Tribunal considers that given the systemic nature of the discrimination under section 11, there should be recognition of some period of retroactivity. At the same time, the Federal Court of Appeal in Morgan affirmed that in creating a period of compensation, common sense should apply and some limits need to be placed upon liability.

[956] Given that there was no evidence presented to underline an argument that this Complaint should be treated differently, the Tribunal concludes that adherence to the Commission's frequent practice of limiting that period to one year prior to the filing of the Complaint would be a reasonable balance under all the circumstances. Therefore, the Tribunal finds that August 24, 1982 is the appropriate date to begin the compensation period.

[957] In terms of the period of time the backpay should cover, it is noted that Alliance counsel advised the Tribunal, in June 2003, that the Alliance and Canada Post had entered into a Letter of Understanding, as of June 6, 2002, under their then current Collective Agreement, providing for the introduction of a new Job Evaluation Plan. The new Plan superceded the original Treasury Board classification standards and created six administrative levels A1 to A6, into which former CR's were to be reclassified, as appropriate. The implementation date for the new Plan was set for August 3, 2002, with the new wage rates taking retroactive effect as of June 3, 2002.

[958] Alliance counsel stated that the resulting range of wage rates for the CR's arising from the new Plan was, in general, quite a bit higher than the wage rates for PO INT's and PO EXT's under the CUPW agreement expiring in January 2003. What does this all mean?

[959] What it meant to Alliance counsel, in terms of its impact on this Complaint, was expressed as follows:

"...it marks the outside boundary of this pay equity complaint".²³⁹

"...in all likelihood, this does represent the end of the complaint, the outer limit, the outer parameter of the complaint".²⁴⁰

[960] The position of Alliance counsel is reinforced by the fact that the Letter of Understanding of June 6, 2002, was a direct result of a Memorandum of Understanding (Appendix `D') of the then current Canada Post / Alliance Collective Agreement, expiring October 31, 2004. The Memorandum of Understanding recorded the concurrence of Canada Post and the Alliance (and Union of Postal Communications Employees) that the proposed new Job Evaluation Plan would "be free of gender bias and shall meet the requirements of section 11 of the Canadian Human Rights Act".

[961] Accordingly, the Tribunal concludes that the back-pay period will extend from August 24, 1982 to June 2, 2002, after which there should be no wage gap between the complainant CR's and the comparator PO's. Furthermore, the Tribunal concludes that with the new wage rates for the former CR's taking effect from June 3, 2002, there will be no need for a "fold-in" into the base wage rates per level at that date, since the successor administrative A1 to A6 levels, as of that same date, were compatible with section 11.

(iii) Interest

[962] Subsection 53(4) of the Act provides for the inclusion of an award of interest at a rate and for a period that the Tribunal considers appropriate, in any order to pay compensation under section 53. Such an order is subject to the rules made pursuant to subsection 48.9(2) of the Act which reads as follows:

Tribunal Rules of Procedure

48.9(2) The Chairperson may make rules of procedure governing the practice and procedure before the Tribunal, including, but not limited to, rules governing...

(i) awards of interest.

[963] The Commission indicated that in the past there was debate as to whether the Canadian Human Rights Tribunal had the authority to make awards of interest as there was no provision in the Act providing for such awards. The 1998 amendments to the Act added such a provision in subsections 48.9 and 53(4).

[964] All three parties referred to Rule 9(12) of the Canadian Human Rights Tribunal Interim Rules of Procedure (dated January 8, 2000) which reads as follows:

Awards of Interest

9(12) Unless the Panel orders otherwise, any award of interest under s. 53(4) of the Canadian Human Rights Act shall

a) be simple interest calculated on a yearly basis at the Canada Savings Bond rate; and

b) begin accruing from the date on which the discriminatory practice occurred.

[965] The Commission submitted that interest consists of two elements, namely, compensation for the loss of use of money, and compensation for the decline of its value. It also noted that Rule 9(12) is an interim and not a final Rule of Procedure and permits the Tribunal to award interest otherwise than as delineated in the Rule.

[966] The circumstances of this case with affected employees waiting over 20 years for a wage adjustment was, in the Commission's view, justification for a more generous award than Rule 9(12) would provide. The Commission called for an award of compound interest at the Courts of Justice Act rate, thereby providing fuller compensation.²⁴¹

[967] The Alliance's submission was that the Tribunal should exercise its discretion to award interest at a higher rate and also favoured compound interest, calculated semi-annually, at the rate established by the Courts of Justice Act.

[968] Both the Alliance and the Commission considered that their respective positions on the matter of compound interest were supported by Waddams, supra, which concluded that compound interest, as a principle, could be warranted under certain conditions.

[969] Canada Post's position was that it would be inappropriate to award compound interest without demonstrating, as indicated in Morgan, a special need or circumstance to compensate for the actual loss sustained. In the opinion of Canada Post, neither the Commission nor the Alliance had presented any evidence to support an exceptional compound interest award or a rate of interest higher than the Canada Savings Bond rate.

[970] Despite the protracted period of the current case and the resulting delays in reaching a conclusion for the complainant employees, the Tribunal finds that no special needs or circumstances were demonstrated in the evidence before it which would justify an award of compound interest.

[971] The Tribunal therefore, concludes that simple interest should be payable on the finally determined award of lost wages, which should in turn be discounted by 50%. The simple interest should be at the Canada Savings Bond rate, beginning from the commencement date of the retroactive period, August 24, 1982. Calculations should be annual, using the Canada Savings Bond rate in effect on September 1^st of each year concerned.

(iv) Post- Judgement Interest

[972] With the likelihood that there will be delays between the date of the Tribunal's final decision and the date of payment of awards to entitled employees, the Commission and the Alliance submitted that post-judgement interest should be paid in accordance with the provisions of the Courts of Justice Act.

[973] The Tribunal agrees, subject to the proviso that any such post-judgement interest shall be calculated from the date of its Order, on the finally determined award of lost wages discounted by 50%.

(v) Special Compensation

[974] The Alliance submitted that the evidence supported a finding by the Tribunal that Canada Post wilfully engaged in a discriminatory practice and that special compensation should be ordered under subsection 53(3) of the Act.

[975] The Alliance argued that, through no fault of their own, the affected employees have been waiting some 20 years for their Complaint to be resolved. The Alliance also argued that its earlier efforts to negotiate a solution at the bargaining table were rebuffed by the employer.

[976] It was, as well, the Alliance's view that Canada Post had consciously avoided satisfying itself as to whether the work of the female-dominated CR group was of equal value to that of the male-dominated PO group. Instead, Canada Post had chosen solely to advance criticisms and arguments of a technical and legal nature designed to avoid its obligations under section 11 of the Act.

[977] A further argument presented by the Alliance was the profound impact the Respondent's violation of section 11 has had on the mostly low wage earners of the CR group. Being deprived of the income comprising the wage gap can reasonably be seen, in particular cases, to have led to "a host of lost opportunities" which should be reflected in the Tribunal's remedial order.

[978] Finally, the Alliance submitted that the disappointment and frustration associated with the length of time it has taken to obtain redress should be compensated in the form of damages for hurt feelings.

[979] While recognizing that no amount of monetary compensation can properly eradicate the impact of so many years of underpayment, the Alliance submitted that employees in the CR group should each receive compensation pursuant to subsection 53(3) of the Act, to be determined at the Tribunal's discretion. The Alliance proposed that the award be prorated based on the number of full years worked by each eligible employee.

[980] The Alliance further submitted that simple interest calculated pursuant to the Ontario Courts of Justice Act should be paid on the special compensation awards.

[981] The Commission noted that pursuant to paragraph 53(2)(e) and subsection 53(3), the Tribunal is empowered to award compensation for pain and suffering and special compensation. The Commission indicated however, that in the circumstances of this case, it had taken no position on the Alliance's claim.

[982] Canada Post submitted that it was not appropriate to award special compensation in this case and argued that the Alliance did not direct any evidence to the Tribunal of the nature required to support such a request.

[983] Canada Post also submitted that the law respecting special damages is clear and was set out well by the tribunal in the Treasury Board (Phase II) case, as follows:

We are of the view that an entitlement under s. 53(3)(b) of the Act requires an evidentiary basis outlining the effects of the discriminatory practice on the individuals concerned. An award for hurt feelings is personal and is usually awarded in the context of direct discrimination. During the course of a hearing a tribunal will assess entitlement after hearing from individuals about the effects of the discrimination upon him or her. (see R. v. Cranston (1997), T.D. 1/97 (C.H.R.T.)). In this manner the Tribunal is able to observe the complainant's demeanour while testifying and come to some conclusion whether, in the circumstances, an award for hurt feelings is called for. In our view the impact of delays giving rise to disappointments, frustrations, maybe even sadness or anger, although legitimate reactions, do not measure up, in our opinion, to the degree and extent of hurt feelings and loss of self-respect that s. 53(3)(b) is directed towards remedying.

The discriminatory practice in this case has its genesis in societal attitudes and history, shared by both males and females. Attitudes about female work are undergoing change with increased awareness, education and legislation. The problem here is systemic and it has occurred in the Employer's pay system. To grant the Commission's and the Alliance's request would amount to an award for hurt feelings, en masse, which is not, in our view, what is contemplated by s. 53(3)(b) of the Act.

We do not doubt some Complainants have experienced a sense of loss, which in some cases may be felt more strongly by some than others. We also appreciate the impracticality of individuals in this case testifying before the Tribunal as to the effects of the discriminatory practice upon them. However, these factors cannot compel us to make an award, en masse, under s. 53(3)(b).²⁴²

[984] The Tribunal notes that, in the aforementioned excerpt from the Treasury Board (Phase II) case, the tribunal was addressing paragraph 53(3)(b) from the pre-1998 amended version of the Act which read as follows:

53(3) In addition to any order that the Tribunal may make pursuant to subsection (2), if the Tribunal finds that

(a) a person is engaging or has engaged in a discriminatory practice wilfully or recklessly, or

(b) the victim of the discriminatory practice has suffered in respect of feelings or self-respect as a result of the practice,

the Tribunal may order the person to pay such compensation to the victim, not exceeding five thousand dollars, as the Tribunal may determine.

[985] The tribunal in the Treasury Board case was, therefore, dealing with the Commission's argument that, as a result of the discriminatory practice, the victims suffered from hurt feelings and a loss of self-respect (former paragraph 53(3)(b)). That tribunal was not addressing in this context whether a person had been engaged in a discriminatory practice wilfully or recklessly (former paragraph 53(3)(a)).

[986] On the other hand, in the current case, the Alliance has submitted that Canada Post wilfully engaged in a discriminatory practice under subsection 53(3) of the current Act. It has made no direct mention of "pain and suffering" under paragraph 53(2)(e) of the current Act.

[987] Canada Post's use of the Treasury Board (Phase II) case decision to support its dismissal of the Alliance's claim was therefore, in the Tribunal's view, somewhat misdirected as the Treasury Board decision focused squarely on the suffering of victims in respect of "feelings or self-respect" and not on a person engaged in a discriminatory practice "wilfully or recklessly".

[988] In fairness to Canada Post, however, the Alliance does appear to have considered aspects of both paragraph 53(2)(e) and subsection 53(3), although it limited its specific written request to subsection 53(3).²⁴³ For example, while alleging wilful engagement of Canada Post in a discriminatory practice, it also submitted that its victims experienced considerable disappointment and frustration and may have suffered in respect of "hurt feelings", which seems to come close to the pain and suffering element of paragraph 53(2)(e).

[989] Given the fact that the Alliance has limited its submission to the provisions of subsection 53(3), the Tribunal has focused on, but has not confined its decision to, that remedial dimension of section 53 of the Act. While the Alliance has argued that Canada Post contributed to the prolonged nature of the Complaint, and protracted the process through its technical and legal criticisms and arguments, the Tribunal does not find that sufficient detailed evidence was furnished to lead the Tribunal to conclude that Canada Post had been engaged in a discriminatory practice wilfully or recklessly.

[990] The subject discriminatory practice is, after all, systemic discrimination which, as a concept, has most often been found to be unintentional. As well, the Act is a statute that seeks remedial corrective action rather than one that seeks to cast blame and punishment.

[991] Based on the evidence presented, the Tribunal does not find that Canada Post has, more likely than not, been engaged in practicing systemic discrimination, wilfully or recklessly. Nor does the Tribunal find that sufficient evidence was provided to document the extent to which victims of the systemic discrimination, either individually, or corporately, may have experienced pain and suffering, difficult as this may be to demonstrate when dealing with a large body of employees.

[992] Accordingly, the Tribunal finds that no award is justified under subsection 53(3) or paragraph 53(2)(e) of the Act.

(vi) Legal Costs

[993] The legal costs, including fees and disbursements associated with the adjudication of the Complaint involving over 400 days of hearings, "are enormous," argued the Alliance. To ensure full compensation for the victims, the Alliance felt it essential that the legal costs be taken into account. This has, in the Alliance's submission, occurred in the past where complainant counsel have contributed an important dimension to the presentation of the complainant's case, as the Alliance has done in this case.

[994] The Alliance submitted that it is settled law that paragraph 53(2)(c) of the Act provides the Tribunal with the authority to award legal costs as well as compensation for other expenses incurred by the victim as a result of the discriminatory practice. Alliance counsel referred to a decision of the Canadian Human Rights Tribunal in Grover v. Canada (National Research Council), [1992] C.H.R.D. No. 12 (QL). In that case, the tribunal ordered the respondent to pay the complainant's legal costs pursuant to the assessment of the costs under the Federal Court scale.

[995] The Alliance also referred to a decision of the Federal Court in Canada (Attorney General) v. Thwaites, [1994] F.C.J. No. 364 (T.D.) and to a tribunal decision in Nkwazi v. Canada (Correctional Service), [2001] C.H.R.D. No. 43 (QL). Both decisions were in the Alliance's submission, supportive of the inclusion of reasonable legal costs in the compensation award to a successful complainant.

[996] Therefore, the Alliance has called for an order of the Tribunal to award its legal costs. It proposes that such costs be calculated on a substantial indemnity basis in accordance with the Tariff prescribed under the Ontario Courts of Justice Act and the Rules of Civil Procedure.

[997] The Commission submitted that while there has, in the past, been some inconsistency in the awarding of legal costs, recent jurisprudence supports the authority of tribunals to do so. In particular, the Commission cited Nkwazi and Premakumar v. Air Canada, drawing on the following excerpt from the latter tribunal's decision:

I am of the view that the remedial objects of the Canadian Human Rights Act are best attained by ensuring that successful complainants are able to recover their reasonable legal expenses associated with the prosecution of human rights complaints.²⁴⁴

[998] Given the 20-year life of the Complaint and the "immense" legal fees and expenses related to the complex litigation involving numerous expert and lay witnesses, the Commission supported a full award of legal costs in favour of the Complainant group. The Commission acknowledged the "very active role" played by the Alliance in the adjudication, "including bearing considerable costs for the gender neutral job evaluation process engaged in by Dr. Wolf and his colleagues".²⁴⁵

[999] The Commission concluded that these factors militate in favour of an award of full legal costs to the Alliance.

[1000] Canada Post submitted that the current case was not an appropriate one for an award of legal costs. While the proceedings have been protracted and complex, Canada Post cited the tribunal's decision in Treasury Board (Phase II) case, where an award of legal costs was deemed to be inappropriate.

[1001] Ambiguously, Canada Post in its concluding submission, requested the following ruling by the Tribunal:

Based upon a broad and liberal interpretation of section 53(1) of the Act, that Canada Post Corporation be reimbursed for its reasonable legal costs in this proceeding as against the CHRC and/or the PSAC.²⁴⁶

[1002] The Tribunal observes that this has been not only a very protracted and complex case but also a rather tortuous one. Allegations have been made about Canada Post's possible role in this tortuosity, but it can also be alleged that the Commission was not entirely responsibility-free -- that it, too, may have contributed to that tortuosity, by the way it managed the Investigation Stage of the Complaint. It can also be alleged that even the Alliance made its own contribution to that tortuosity by not ensuring, during the formative stage of the Complaint, that the non-wage elements of compensation under subsection 11(7) of the Act were included in the wage calculations.

[1003] However, as indicated earlier in this Decision, laying blame is not an objective of the Act, nor is it a course which the Tribunal has pursued. Having moved into the adjudicative tribunal arena, each party opted to act within its respective rights and decided and honed its strategies with respect to the Complaint. Each party engaged its own legal counsel and directed them accordingly.

[1004] Given the unintentional nature of the alleged systemic discrimination in this case, and after carefully considering all the arguments and evidence made available, the Tribunal finds that each party should assume responsibility for its own legal costs, including related disbursements.

(vii) Retention of Jurisdiction

[1005] Both the Alliance and the Commission have requested that the Tribunal retain jurisdiction to deal with issues that they expect may arise in the implementation of the Tribunal's decision, on an "as-needed" basis.

[1006] In the interest of assisting all parties, as may be appropriate, the Tribunal agrees with this request.

X. JOINT UNION-EMPLOYER LIABILITY FOR WAGE DISCRIMINATION

A. Canada Post's Submission

[1007] Canada Post submitted that, were the Tribunal to find that the Complaint has been substantiated, the Alliance, as the union representing the Complainant group, and Canada Post, as the employer, should be held jointly liable for the discriminatory practice.

[1008] The basis for this submission by Canada Post was their argument that as a principle, a union and an employer share liability for any clauses, including those that are discriminatory, which are negotiated in a collective agreement.

[1009] In support of this position, Canada Post relied on the decisions in Central Okanagan School District No. 23 v. Renaud, [1992] 2 S.C.R. 790, and Canada Safeway Ltd. v. Saskatchewan (Human Rights Commission), [1997] S.J. No. 502 (Sask. C.A.).

B. The Alliance's Position

[1010] The Alliance submitted that the wording of subsection 11(1) of the Act clearly precludes the imposition of liability on anyone other than an employer.

[1011] Further, the Alliance submitted that the decision in Bell Canada v. Communications, Energy and Paperworkers Union of Canada is binding authority concerning the interpretation of subsection 11(1).²⁴⁷

[1012] The Alliance urged the Tribunal to find that the cases cited by Canada Post could be distinguished.

C. The Commission's Position

[1013] The Commission's submissions underlined those of the Alliance.

D. Tribunal's Analysis

[1014] The decision of the Federal Court of Appeal in Bell Canada, cited by the Alliance as authority for their submissions concerning the interpretation of subsection 11(1) of the Act, contains strong statements to the effect that section 11 of the Act makes the employer alone liable for differences in wages with respect to work of equal value. For example, at paragraph 56, the Court noted as follows:

For reasons of its own Parliament has chosen, in section 11, to make the employer alone liable for differences in wages with respect to work of equal value. It would fly in the face of the clear wording of the Act and the obvious intent of Parliament to find the unions equally liable either implicitly under section 11 or indirectly through sections such as section 10 for having participated in the establishment of different wages with respect to work of equal value. It may at first blush appear to be self serving and unethical for a union to use the mechanism of a complaint under section 11 to force for all practical purposes the revision of a collective agreement it has freshly negotiated, but absent bad faith -- the Motions Judge did not make a specific finding of bad faith in the instant case... -- it is not legally wrong. The Court applies the Act as it is, not as it might have been.²⁴⁸

[1015] This decision was rendered by one of the Tribunal's supervisory courts. The decision included the above-mentioned discussion of the very legislative provision that is in issue in the present Complaint. Is the Tribunal bound, therefore, to follow this decision?

[1016] The Court prefaced its remarks in that same Bell Canada case with the stipulation that it was not providing a definitive interpretation of section 11 of the Act, as follows:

The Motions Judge erred in totally ignoring sections 43, 44, and 49 of the Act and in his premise that '[w]hat is principally at issue in this case is the correct interpretation of s. 11' (Paragraph (8) of his reasons [at page 85]) That was simply not the issue at this stage. The decision attacked is the decision to request the appointment of a Human Rights Tribunal. It will be the duty of the Tribunal to determine whether the complaints are well founded or not and the Tribunal will in no way be bound by the interpretation given to section 11 by the investigator and presumably adopted by the Commission. Those who expected this Court to resolve issues with respect to the interpretation and application of section 11 without the benefit of the decision of a tribunal on this issue in the instant case will be disappointed; whatever was said by the Motions Judge should be considered as obiter and I make no observations upon any of it.²⁴⁹

[1017] As can be seen from the comments of the Federal Court of Appeal, the decision in Bell Canada, upon which the Alliance and the Commission relied in their submissions was fundamentally concerned with the legality of the Commission's decision, under section 49 of the Act, to request the appointment of a Human Rights Tribunal.

[1018] Accordingly, this Tribunal has concluded that the question of joint union-employer liability under section 11 of the Act remains an open question for its decision.

[1019] Subsection 11(1) of the Act clearly indicates on whose shoulders liability must rest. It states as follows:

It is a discriminatory practice for an employer to establish or maintain differences in wages... (emphasis added)

[1020] This wording can be contrasted to that of other provisions of the Act, such as section 10 which explicitly addresses employee organizations as well as employers, and section 7 which contains no qualifying language.

[1021] Based on a clear and straightforward reading of subsection 11(1), the argument that a union may incur liability under this section must be rejected. Canada Post's reference to the Renaud and Safeway cases has not been helpful in this instance, as they both dealt with legislative provisions that addressed union liability, and are distinguishable on that basis. In this case, there is no such inclusion. In fact, the section of the Act is very clear in its notation that it is a discriminatory practice for an employer to establish or maintain differences in wages. There is no mention of other organizations, nor is there a lack of clarity in the wording.

[1022] Therefore, the Tribunal cannot accept the submission of Canada Post on this issue of "joint liability".

XI. ORDERS

[1023] Based on all of its foregoing findings and conclusions, including a breach of section 11 of the Act, the Tribunal Orders that:

The Respondent shall pay to each of its eligible Clerical and Regulatory employees an award for lost wages by closing the wage gap between employees of the Complainant and Comparator groups represented in this Complaint.
The wage gap between the Complainant group and the Comparator group shall be determined and calculated by a level-to-line technique, preferably following the Kervin/Commission Wage Adjustment Model.
The Respondent shall provide access to the individual employee records, as required, to enable final wage gap calculations to be determined.
The finally determined award of lost wages ("pay for all purposes") for each eligible CR employee, by whatever methodology, shall be discounted by 50%.
The back-pay compensation period shall extend from August 24, 1982 to June 2, 2002.
Simple interest shall be calculated annually on the amount of the 50% discounted award of lost wages, and paid to each eligible CR employee for each year, or fraction thereof, of the back-pay compensation period.
The simple interest shall be determined using the Canada Savings Bond rate in effect on September 1^st of each year concerned.
Between the date of this Decision and the date of the ultimate payment of the 50% back-pay award of lost wages, post-judgement simple interest shall be paid to each eligible CR employee at the applicable post-judgement rate prescribed by the Courts of Justice Act of Ontario or comparable provincial legislation.
The Respondent shall be responsible for making remittances, as necessary, that may arise as a result of any of these Orders, with respect to statutory-based non-wage forms of compensation.
The Complainant's claim for special compensation pursuant to paragraph 53(2)(e) or subsection 53(3) of the Act is hereby dismissed.
The claims for legal costs are hereby dismissed.
The Respondent's submission that both it and the Alliance should be jointly liable for any substantiated wage discrimination is hereby dismissed.
The Tribunal shall retain jurisdiction to deal with issues that may arise in the implementation of its Decision, on an "as needed" basis.

Signed by

Elizabeth Leighton

Signed by

Gerald T. Rayner

OTTAWA, Ontario

October 7, 2005

¹R.S.C. 1985, c. H-6.

²R.S.C. 1985, c. P-35.

³R.S.C. 1985, c. C-10.

⁴Ibid. at s. 5(2).

⁵R.S.C. 1985, c. L-2.

⁶G.A. Res. 217A(III), UN Doc. A/810 (1948) at 71.

⁷Convention concerning Equal Remuneration for Men and Women Workers for Work of Equal Value, 29 June 1951, I.L.O. C100.

⁸19 December 1966, 993 U.N.T.S. 3 (entered into force 3 January 1976, accession by Canada 19 May 1976).

⁹19 December 1966, 999 U.N.T.S. 171 (entered into force 23 March 1976, accession by Canada 19 May 1976).

¹⁰Ibid.

¹¹Supra note 8.

¹²Canada, Report of the Royal Commission on the Status of Women (Ottawa: Royal Commission on the Status of Women, 1970).

¹³18 December 1979, 1249 U.N.T.S. 13 (entered into force 3 September 1981, ratification by Canada 10 December 1981).

¹⁴at 918.

¹⁵Canada (P.G.) v. Mossop, [1993] 1 S.C.R. 554 at 612.

¹⁶at 156.

¹⁷2nd ed. (Toronto: Butterworths, 1983).

¹⁸3^rd ed. (Toronto: Butterworths, 1994).

¹⁹Supra note 17 at 87.

²⁰Supra note 18 at 131.

²¹Ibid. at 288.

²²Rizzo & Rizzo Shoes Ltd. (Re), [1998] 1 S.C.R. 27 at para. 22.

²³"More than 1,000 new Calls to the Bar" Ontario Lawyers Gazette (Fall/Winter 2002).

²⁴Parliament, Standing Committee on Justice and Legal Affairs, Minutes of Proceedings and Evidence, Issue No. 11 (17 May 1977).

²⁵Supra note 12.

²⁶ R.S.C. 1985, c. H-6, s. 27(2).

²⁷S.O.R./1986-1082.

²⁸Canada (Attorney General) v. Public Service Alliance of Canada, [1999] F.C.J. No. 1531 at para. 152 (F.C.T.D.).

²⁹Ontario (Human Rights Commission) v. Simpson-Sears Ltd., [1985] 2 S.C.R. 536 at 558.

³⁰Bell Canada v. Canadian Telephone Employees Association, [1998] F.C.J. No. 313 (F.C.T.D.).

³¹Canadian Telephone Employees Association v. Bell Canada (4 June 1997), Interim Ruling, T454/0991 at 19 (C.H.R.T.).

³²Supra note 30 at para. 154.

³³Public Service Alliance of Canada v. Canada Post Corporation (21 October 1998), Interim Ruling, T299/1392 at 24 (C.H.R.T.).

³⁴Ibid. at 23.

³⁵An Act to amend the Canada Evidence Act, Criminal Code and Canadian Human Rights Act, S.C. 1998, c. 9.

³⁶Canadian Telephone Employees Association v. Bell Canada (26 April 1999), Interim Ruling, T503/2098 (C.H.R.T.).

³⁷Bell Canada v. Canada (Human Rights Commission), [2000] F.C.J. No. 1747 (F.C.T.D.).

³⁸Bell Canada v. Canada (Human Rights Commission), [2001] F.C.J. No. 776 (F.C.A.).

³⁹Bell Canada v. Canadian Telephone Employees Association, [2003] 1 S.C.R. 884.

⁴⁰Ibid. at para. 47.

⁴¹Ibid. at para. 50.

⁴²Supra note 39.

⁴³Ibid. at para. 47

⁴⁴Ibid.

⁴⁵Equal Wages Guidelines, S.I./78-155 (1978), as am. by S.I./82-2 (1982).

⁴⁶Canada Post Submissions, Transcript, Vol. 409 at 45968.

⁴⁷Ibid. at 45981.

⁴⁸Ibid. at 45981-45983.

⁴⁹[1977] 1 S.C.R. 271.

⁵⁰Commission Reply Submissions at 16.

⁵¹Ibid. at 17.

⁵²Ibid. at 18.

⁵³[2001] A.J. No. 1535 (Alta. Q.B.), as cited in the Commission Reply Submissions at 26.

⁵⁴Ibid.

⁵⁵Supra note 7; Supra note 12.

⁵⁶Supra note 49; Supra note 18.

⁵⁷Supra note 18.

⁵⁸Ibid. at 514-515.

⁵⁹Ibid. at 517.

⁶⁰Ibid.

⁶¹Commission Submissions at para. 15-16; Transcript, Vol. 415 at 46841.

⁶²Supra note 39 at para. 47.

⁶³Supra note 53 at para. 21.

⁶⁴Ibid. at para. 22.

⁶⁵Supra note 18 at 537.

⁶⁶Friends of the Oldman River Society v. Canada (Minister of Transport), [1992] 1 S.C.R. 3.

⁶⁷Canada Post Submissions, Transcript, Vol. 409 at 46026-27.

⁶⁸Transcript, Vol. 414 at 46730-31.

⁶⁹Supra note 17 at 87; See also supra note 18 at 131 and 288.

⁷⁰Canada Post Submissions, Transcript, Vol. 408 at 45844.

⁷¹Canada Post Submissions, Transcript, Vol. 408 at 45873.

⁷²Supra note 28.

⁷³Public Service Alliance of Canada v. Canada (Department of National Defence), [1996] F.C.J. No. 842 (F.C.A).

⁷⁴Supra note 17 at 87.

⁷⁵Supra note 18 at 131.

⁷⁶R.S.C. 1985, c. I-21, s. 12.

⁷⁷Supra note 73 at para. 2.

⁷⁸[1987] 1 S.C.R. 1114.

⁷⁹Ibid. at paras. 34 and 40.

⁸⁰[1991] C.H.R.D. No. 4 (C.H.R.T) (QL).

⁸¹Supra note 28 at para. 141.

⁸²Ibid. at para. 150.

⁸³Ibid. at para. 151.

⁸⁴Transcript, Vol. 396 at 44388.

⁸⁵ Supra note 28 at para. 152.

⁸⁶Statement of Agreed Facts, Appendices J & L.

⁸⁷Ibid.

⁸⁸Statement of Agreed Facts, Appendix L; Transcript, Vol. 315 at 36800-3680l.

⁸⁹Exhibit HR-1, Tab 3 at 24.

⁹⁰Ibid. at 11.

⁹¹Transcript, Vol. 20 at 2626.

⁹²Exhibit HR-2, Tab 2.

⁹³Transcript, Vol. 20 at 2754.

⁹⁴Transcript, Vol. 38 at 5288.

⁹⁵Transcript, Vol. 57 at 7674-7675.

⁹⁶Transcript, Vol. 3 at 327-328.

⁹⁷Ibid. at 331.

⁹⁸Ibid. at 333.

⁹⁹Canadian Union of Public Employees (Airline Division) v. Canadian Airlines International Ltd., [1998] C.H.R.D. No. 8 (C.H.R.T.) (QL).

¹⁰⁰Canada (Canadian Human Rights Commission) v. Canadian Airlines International Ltd., [2001] F.C.J. No. 1258.

¹⁰¹Canada (Canadian Human Rights Commission) v. Canadian Airlines International Ltd., [2004] F.C.J. No. 483.

¹⁰²Ibid. at para. 20.

¹⁰³Ibid. at para. 52.

¹⁰⁴Ibid. at para. 32.

¹⁰⁵Ibid. at para. 92.

¹⁰⁶Ibid. at para. 25.

¹⁰⁷Ibid. at para. 49.

¹⁰⁸Ibid. at para. 46.

¹⁰⁹Transcript, Vol. 49 at 6652.

¹¹⁰Transcript, Vol. 213 at 27026.

¹¹¹Transcript, Vol. 214 at 27076.

¹¹²Transcript, Vol. 214 at 27113.

¹¹³Transcript, Vol. 216 at 27236.

¹¹⁴Transcript, Vol. 280 at 33124.

¹¹⁵Transcript, Vol. 295 at 34933.

¹¹⁶Transcript, Vol. 298 at 35204.

¹¹⁷Transcript, Vol. 296 at 35062.

¹¹⁸Transcript, Vol. 299 at 35253.

¹¹⁹Transcript, Vol. 318 at 37037.

¹²⁰Exhibit HR-1, Tab 22.

¹²¹Exhibit PSAC-29 at 1.

¹²²Exhibit PSAC-29, Appendix A.

¹²³Ibid.

¹²⁴Exhibit PSAC-180, Findings & Conclusions.

¹²⁵Transcript, Vol. 368 at 41399; Vol. 369 at 41430.

¹²⁶Public Service Alliance of Canada v. Canada (Treasury Board), [1996] C.H.R.D. No. 2 at para. 187 (C.H.R.T.) (QL).

¹²⁷Supra note 28 at para. 79.

¹²⁸Supra note 73 at para. 33.

¹²⁹Trojan Technologies, Inc. v. Suntec Environmental Inc., [2004] F.C.J. No. 636.

¹³⁰Merck & Co. v. Apotex Inc., [2004] F.C.J. No. 684 (T.D.).

¹³¹Exhibit HR-1, Tab 22.

¹³²Exhibit HR-31, Tab 6 at 201.

¹³³Transcript, Vol. 126 at 17241.

¹³⁴Transcript, Vol. 35 at 4803.

¹³⁵Transcript, Vol. 372 at 41789.

¹³⁶Transcript, Vol. 127 at 17350.

¹³⁷Exhibit HR-1, Tab 22.

¹³⁸Exhibit R-225 at 3.

¹³⁹Transcript, Vol. 136 at 18643.

¹⁴⁰Supra note 138.

¹⁴¹Transcript, Vol. 172 at 22797.

¹⁴²Ibid. at 22803.

¹⁴³Exhibit HR-93 A.

¹⁴⁴Ibid.

¹⁴⁵Transcript, Vol. 144 at 19837.

¹⁴⁶Transcript, Vol. 127 at 17345.

¹⁴⁷Ibid. at 17374.

¹⁴⁸ Exhibit R-235 at 66; Exhibit R-249 at 18.

¹⁴⁹Exhibit R-235 at 77; Exhibit R-249 at 32.

¹⁵⁰Exhibit R-249 at 80.

¹⁵¹ Exhibit R-455 at 22

¹⁵² Transcript, Vol. 105 at 14524.

¹⁵³Supra note 151 at 12-14 and 16.

¹⁵⁴Ibid. at 33.

¹⁵⁵Ibid.

¹⁵⁶Transcript, Vol. 130 at 17711.

¹⁵⁷Exhibit R-615 at 1

¹⁵⁸Ibid.

¹⁵⁹Ibid. at 3.

¹⁶⁰ Ibid. at 23.

¹⁶¹Ibid. at 25.

¹⁶²Supra note 126.

¹⁶³Ontario Nurses' Association v. Regional Municipality of Haldimand-Norfold (1991), 2 P.E.R. 10S (Pay Equity Hearings Tribunal).

¹⁶⁴Exhibit R-235 at 77.

¹⁶⁵Exhibit R-249 at chapter 6.

¹⁶⁶R.S.O. 1990, c. P-7.

¹⁶⁷Ibid.

¹⁶⁸Alliance Submissions at 113.

¹⁶⁹Ibid. at 167.

¹⁷⁰Transcript, Vol. 210 at 26679.

¹⁷¹Service Employees International Union, Local 204 v. Ontario (Attorney General), [1997] O.J. No. 3563.

¹⁷²Canada (Human Rights Commission) v. Canadian Airlines International Ltd., [2004] F.C.J. No. 483 at para. 51.

¹⁷³Transcript, Vol. 136 at 18700.

¹⁷⁴Exhibit R-615.

¹⁷⁵Exhibit PSAC-29.

¹⁷⁶Transcript, Vol. 194 at 25029.

¹⁷⁷Transcript, Vol. 125 at 17031.

¹⁷⁸Transcript, Vol. 163 at 21992.

¹⁷⁹Transcript, Vol. 127 at 17350-17351.

¹⁸⁰Transcript, Vol. 134 at 18298.

¹⁸¹Transcript, Vol. 174 at 22908.

¹⁸²Transcript, Vol. 376 at 42167.

¹⁸³Transcript, Vol. 173 at 22830.

¹⁸⁴ Transcript, Vol. 314 at 36739.

¹⁸⁵Transcript, Vol. 345 at 39204.

¹⁸⁶Exhibit R-615.

¹⁸⁷Transcript, Vol. 404 at 45223.

¹⁸⁸Ibid. at 45335.

¹⁸⁹S.M. Waddams, The Law of Damages, looseleaf (Toronto: Canada Law Book Inc., 2004) at 13-1 and 13-2.

¹⁹⁰Ibid.

¹⁹¹Transcript, Vol. 127 at 17373.

¹⁹²Exhibit PSAC-29.

¹⁹³Exhibit PSAC-30.

¹⁹⁴Supra note 126.

¹⁹⁵Exhibit PSAC-55.

¹⁹⁶Exhibits PSAC-29 and PSAC-30.

¹⁹⁷Transcript, Vol. 151 at 20794-20795.

¹⁹⁸Supra note 126.

¹⁹⁹Exhibits R-433 and R-436.

²⁰⁰See paragraph [724] above.

²⁰¹Commission Submissions, Chap. 13 at para. 532.

²⁰²Canada Post Submissions, Chap. 14 at 395.

²⁰³Ibid. at para. 89.

²⁰⁴ Ibid. at para. 95.

²⁰⁵Syndicat des employés de production du Québec et de l'Acadie v. Canada (Human Rights Commission), [1989] 2 S.C.R. 879 at para. 87.

²⁰⁶Supra note 28 at para. 116.

²⁰⁷Robichaud v. Canada (Treasury Board), [1987] 2 SCR 84.

²⁰⁸Supra note 78 at paras. 24 and 25.

²⁰⁹Ibid. at para. 44.

²¹⁰Supra note 73 at para. 49.

²¹¹Transcript, Vol. 26 at 3564.

²¹²Ibid. at 3565.

²¹³Supra note 126.

²¹⁴ See paragraphs [769] and [778] above.

²¹⁵ See paragraphs [767] and [769] above.

²¹⁶Transcript, Vol. 382 at 42856-57.

²¹⁷Transcript, Vol. 151 at 20835.

²¹⁸Transcript, Vol. 37 at 5120.

²¹⁹Ibid. at 5121-5122.

²²⁰Commission Submissions at 187-188.

²²¹Transcript, Vol. 148 at 20440.

²²²Transcript, Vol. 150 at 20600-20601.

²²³Transcript, Vol. 337 at 38705.

²²⁴Transcript, Vol. 148 at 20433.

²²⁵ Supra note 101 at para. 87.

²²⁶Transcript, Vol. 337 at 38727.

²²⁷Transcript, Vol. 148 at 20435.

²²⁸Canada (Attorney General) v. Morgan, [1991] F.C.J. No. 1105 at para. 19 (F.C.A.).

²²⁹Supra note 73 at para. 20.

²³⁰Ibid. at para. 44.

²³¹Supra note 228 at para. 15.

²³²[2004] C.H.R.D. No. 16 (C.H.R.T.) (QL).

²³³[1998] C.H.R.D. No. 7 (C.H.R.T.) (QL).

²³⁴Ibid. at para. 286.

²³⁵[2000] F.C.J. No. 417.

²³⁶Supra note 73 at para. 49.

²³⁷Public Service Alliance of Canada v. Canada (Treasury Board), [1998] C.H.R.D. No. 6 (C.H.R.T.) (QL).

²³⁸Supra note 73 at para. 20.

²³⁹Transcript, Vol. 414 at 46825.

²⁴⁰Ibid. at 46829.

²⁴¹R.S.O. 1990, c. C-43.

²⁴²Supra note 237 at paras. 496-498.

²⁴³Alliance Submissions, Chap. 17 at 551.

²⁴⁴[2002] C.H.R.D. No. 17 at para. 11 (C.H.R.T.) (QL).

²⁴⁵Commission Submissions, Chap. 13 at 570.

²⁴⁶Canada Post Submissions, Chap. 15 at 427.

²⁴⁷[1998] F.C.J. No. 1609 (F.C.A.).

²⁴⁸Ibid.

²⁴⁹Ibid. at para. 37.

PARTIES OF RECORD

TRIBUNAL FILE:

T299/1392

STYLE OF CAUSE:

Public Service Alliance of Canada

v. Canada Post Coroporation

DATE AND PLACE OF HEARING:

September 21, 1992 to August 27, 2003

(415 days)

Ottawa, Ontario

DECISION OF THE TRIBUNAL DATED:

October 7, 2005

APPEARANCES:

James Cameron

for the Complainant

Peter Engelmann

Fiona Campbell

Fiona Keith

Rosemary Morgan (1993-1994)

for the Canadian Human Rights Commission

Russell G. Juriansz, (1992-March 1998)

Roy L. Heenan, (1998-2005)

E. Joy Noonan, (1998-2005)

Anne Irwin, (1993-2003)

Guy Dufort, (1998-2003)

Rob Grant, (2002-2003)

Zygmunt Machelak, (1992) and (1998)

Neelam Jolly, (1997-1998)

Stephen Bird, (1998)

Nitya Iyer, (2003)

for the Respondent

You are being directed to the most recent version of the statute which may not be the version considered at the time of the judgment.

Do not display this message anymore