I.
INTRODUCTION
[1]
This is a judicial review of a decision of the
Canadian Human Rights Tribunal (CHRT or “the Tribunal”), 2011 CHRT 11. The
complainants initiated a complaint pursuant to the Canadian Human Rights Act,
RSC 1985, c H-6 [CHRA or the Act] against Air Canada and the Air Canada
Pilots Association [ACPA] (together “the respondents”), claiming an alleged
discriminatory practice relating to age with respect to the mandatory
retirement rule in their collective agreement. A judicial review was also
sought of a related further decision of the Tribunal, 2012 CHRT 9, concerning
an amended remedy pending an appeal to the Federal Court of Appeal; this has
been dismissed.
[2]
The complainants are a group of individual
applicants gathered into the “Fly Past 60 Coalition”. They are past members of
ACPA employed by Air Canada, an organization which employs over 2,800 pilots in
total. Other similar cases of retired pilots may be waiting to be heard after
this matter is disposed of.
[3]
The complainants allege that Air Canada and ACPA
contravened sections 7 and 10 of the CHRA by requiring them to retire as each
reached the age of 60 at various dates between 2005 and 2009, regardless of
merit or ability to continue flying, none of which is denied.
[4]
Air Canada pilots are the best paid in Canada, with generous benefits and excellent working conditions, particularly as pilots
gain seniority. The benefits include a lucrative defined benefit pension
awaiting them upon retirement, along with good job prospects to fly for other
airlines after retiring on account of their training and experience.
Accordingly, this case should be circumscribed to its facts of a mandatory
retirement provision [MRP] in an area of scarce good jobs with relatively
little financial hardship on retirement.
[5]
Since
1957, the Air Canada pension plan has stipulated that 60 is the compulsory age
of retirement for pilots. As of the early 1980's, provisions mandating
retirement at age 60 were included as part of the collective agreement in force
between Air Canada and its pilots' union. Since 1995, ACPA has been the union
representing Air Canada pilots. Under
the terms of the collective agreement and pension plan between Air Canada and ACPA, Air Canada pilots are required to retire on the first day of the month following
their 60th birthday.
[6]
A relevant constraint is
that Canada adheres to International Civil Aviation Organization (ICAO)
standards. Until March 2006, ICAO set a maximum age of 60 for a pilot in
command and recommended, but did not require, that a co-pilot on an
international flight not fly past his or her 60th birthday. In
March 2006, coming into force in November 2006, ICAO set 65 as the maximum age
for pilots and set as the standard that if one pilot was over 60, the other
must be under 60.
[7]
In 2011 CHRT 11, the Tribunal made an initial
finding of prima facie discrimination, which was never in dispute due to
legacy proceedings which will be described below, and the fact that the
respondents did not contest it.
[8]
The Tribunal also rejected the Bona Fide Occupational Requirement (BFOR)
defences of Air Canada and ACPA. Ultimately, however, the allegations of a
discriminatory practice were not upheld because the Tribunal concluded that 60
was the “normal age of retirement” in the Canadian passenger airline industry,
thereby denying liability pursuant to section 15(1)(c) of the Act.
[9]
All three parties filed judicial review
applications seeking to set aside that aspect of the Tribunal’s decision which
was not in their favour. The complainants challenged the finding on normal age
of retirement (Court file T-1428-11) while Air Canada and ACPA sought to set
aside the Tribunal’s decision rejecting their BFOR defences (T-1453-11 and
T-1463-11 respectively).
[10]
The Canadian Human Rights Commission
(Commission) also applied for judicial review (T-1456-11, now discontinued),
seeking a determination of the constitutionality of section 15(1)(c).
The Commission discontinued this application when the CHRT agreed to consider
the question.
[11]
On April 18, 2012, the Tribunal issued 2012 CHRT
9, finding that section 15(1)(c) violated the guarantee of equality in section 15(1)
of the Canadian Charter of Rights and Freedoms, Part I of the Constitution
Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11
[the Charter]. This reversed the previous outcome of the complaint, as
Air Canada and ACPA no longer had a valid defence to the prima facie
discrimination. Air Canada applied for judicial review of this new decision
(T-971-12), as did ACPA (T-979-12).
[12]
However, shortly afterwards, on July 17, 2012,
in Air Canada Pilots Association v Kelly, 2012 FCA 209 [“Kelly
FCA”], the Federal Court of Appeal, considering a previous series of Tribunal
and Federal Court decisions on the Air Canada mandatory retirement age of 60,
upheld the constitutionality of section 15(1)(c). Leave to appeal was
denied by the Supreme Court in [2013] SCCA No 395
(QL). The Tribunal’s decision in 2012 CHRT 9 is therefore no longer
valid, and Court files T-971-12 and T-979-12 are dismissed.
[13]
In the reasons that follow, I allow the
application of the complainants, setting aside the Tribunal’s decision on the
normal age of retirement of pilots at age 60. Conversely, I dismiss Air Canada’s application to set aside the Tribunal’s decision rejecting its BFOR defence. Most
significantly however, I allow ACPA’s application to set aside the Tribunal’s
decision dismissing its BFOR defence and I send the matter back for
redetermination by the same panel.
[14]
I say significantly because my decision is based
on an important distinction in the facts from those in the many similar cases
that have preceded this one. In the present case, ACPA based its hardship argument
on new evidence demonstrating an adverse differential financial impact
affecting its younger members in the event of the elimination of the age 60
rule in the collective agreement. The Tribunal nevertheless dismissed ACPA’s
application, finding that ACPA’s evidence on hardship presented a “close call”.
[15]
In this regard, I conclude that the Tribunal
failed to properly justify its decision in a transparent fashion. By that I
mean the Tribunal omitted to consider important aspects of the evidence that
demonstrated a significant adverse financial impact on younger pilots. Most
importantly, it mischaracterized the impact of extending the pilots’ careers as
a matter of delaying retirement, without considering ACPA’s submission that the
pilots would be working an additional three years to achieve the net total
revenues situation portrayed at age 63. In the case of the younger pilots, this
would mean working for highly reduced effective rates of pay or even for free
during those three years.
[16]
However, it was not a straightforward matter to
set aside the Tribunal’s decision concerning ACPA’s application and simply send
it back for reconsideration on the hardship issue. Before dismissing the
hardship claims, the Tribunal had already concluded that ACPA’s application would
fail, because as a union, it did not meet the requirements of steps one and two
of the test in British Columbia (Public Service Employee Relations Commission) v British
Columbia Government and Service Employees' Union (BCGSEU) (Meiorin Grievance), [1999] 3 S.C.R. 3, [1999] SCJ
No 46 (QL) [Meiorin] at para 54. The Tribunal’s
hardship analysis was carried out as a matter of “prudence”, most likely to
demonstrate to ACPA that its application was dismissed on substantive grounds.
[17]
Accordingly, setting aside the Tribunal’s ruling
on ACPA’s BFOR defence required surmounting a number of preliminary obstacles
and reconsidering previous jurisprudence touching on these issues. For this
purpose, I first concluded that in accordance with the Supreme Court decision
in Central
Okanagan School District v Renaud, [1992] 2 SCR 970, [1992] SCJ No
75 (QL) [Renaud], the Meiorin test could be modified to avoid
imposing absolute liability on ACPA. The Meiorin BFOR test was therefore
amended to reflect ACPA’s joint liability with Air Canada. I also added a
fourth step to the Meiorin test, as appeared to be the Supreme Court’s
direction in Renaud, requiring weighing the importance of preventing the
discriminatory practice in allowing a defence of hardship.
[18]
As an adjunct to the proscription in Renaud
against imposing absolute liability, and for other reasons, I also respectfully
disagreed with this Court’s previous decision in Vilven v Air Canada,
2009 FC 367, [Vilven] that held that the categories of hardship should
be confined to those expressly enumerated in section 15(2) of the CHRA, being
factors of safety, health and costs.
[19]
In addition, although this issue was not raised
by the respondents, I concluded that the new evidence on adverse differential
impact required a fresh consideration of whether the MRP was substantively
discriminatory. This consideration was premised on comments of the Supreme
Court in Law v Canada (Minister of Employment and
Immigration), [1999] 1 S.C.R. 497, [1999] SCJ No 12 (QL) [Law]
and Withler v
Canada (Attorney General), 2011 SCC 12, [2011] SCJ No 12 (QL)
[Withler] that together allow for a suggestion that the mandatory
retirement rule should be seen as serving an ameliorative purpose to provide
for the beneficial age-based equal distribution of benefits among ACPA’s
members, as opposed to serving to perpetuate stereotypes and prejudice.
[20]
Where the introduction of adverse differential
impact evidence appears capable of affecting conclusions on both discrimination
and hardship issues, it does not make sense in a redetermination to consider
only one of the issues, and not the other.
[21]
In a similar vein, I queried whether in the 21st
century it remains realistic to argue that there exist widespread attitudinal
stereotypes and prejudice that disadvantage older workers in the workplaces.
Accordingly, my direction to the Tribunal includes instructions permitting the
introduction of evidence with the view to reconsider past judicial notice
conclusions of the Supreme Court, principally from McKinney v University of
Guelph, [1990] 3 S.C.R. 229, [1990] SCJ No 122 (QL) [McKinney], relied
on in Vilven to support the Tribunal’s finding that the retirement rule
perpetuated stereotypes and prejudice against older workers.
[22]
This direction is based on the comments of LeBel
J. in Québec (AG) v A, 2013 SCC 5, [2013] SCJ No 5 (QL) [Québec v A]
at para 154 that “the court can take judicial notice of certain facts or
matters but must be careful not to use judicial notice to recognize social
phenomena that may not truly exist.”
[23]
My reasons in support of the foregoing rulings
and directions follow below.
II.
JUDICIAL HISTORY
[24]
In order
to provide context for these issues, it is necessary to understand the lengthy
procedures which have preceded the present review hearing. The numerous tribunal
and court rulings result from different issues being decided at different steps
in the judicial history, giving rise to further additional issues as decisions
of the Tribunal were set aside and new issues set out for consideration.
Unfortunately, this pattern of overturning the Tribunal’s decision and sending
the matter back for redetermination on different issues does not end with this
case.
[25]
Several
aspects of the case currently before me are legacy issues from a previous
series of challenges to Air Canada’s mandatory retirement provision, those
being the Vilven and Kelly cases described in the following
sections.
A. Vilven
v Air Canada, 2007 CHRT 36 [Vilven Tribunal #1]
[26]
Two issues were
decided by the Tribunal in this matter. First, it concluded that the normal age
of retirement of airline pilots in comparator airlines was age 60. This finding
resulted in a dismissal of the complaints on the basis that the MRP in the
collective agreement was saved by section 15(1)(c) of the CHRA, which provides
that it is not a discriminatory practice if termination results “because that
individual has reached the normal age of retirement for employees working in
positions similar to the position of that individual”. Second, the Tribunal determined that para 15(1)(c) did not
infringe section 15(1) of the Charter. Both decisions were overturned by
the Federal Court.
[27]
The basic facts were
similar to those in the present case. Two retired pilots, George Vilven and
Robert Neil Kelly, complained of age discrimination to the CHRC in 2007. Mr.
Vilven had been employed from May 26, 1986 until the day after he turned 60 on
August 30, 2003; Mr. Kelly had been employed from September 11, 1972 until the
day after he turned 60 on April 30, 2005.
[28]
Mr. Vilven had risen
to the position of First Officer on an A340 aircraft based in Vancouver, after
which he chose not to become a pilot in command but instead used his seniority
to remain on that aircraft type and in Vancouver near his family. Upon
retiring from Air Canada, he was entitled to a pension of $6,094.04 per month
until the age of 65 and $5,534.33 thereafter. He pursued his flying career
with a smaller airline. Mr. Kelly had risen to the position of pilot in
command on an A340. He was entitled to a pension of $10,233.96 per month until
the age of 65 and $9,477.56 thereafter. He too pursued his flying career with
smaller airlines after leaving Air Canada.
[29]
The complainants
established a prima facie case of discrimination against Air Canada
under sections 7 (refusing to continue to employ an individual on the basis of
age, a prohibited ground of discrimination) and 9 (depriving individuals of employment
opportunities on a prohibited ground) of the CHRA. A similar finding was made
against ACPA under section 10(b) (an employee organization entering into
an agreement that deprives individuals of employment opportunities on a
prohibited ground).
[30]
The Tribunal determined
that the proper comparator group by which to establish the normal age of
retirement in the airline industry was “pilots who fly with regularly scheduled
international flights with a major international airline.” The parties
produced a joint statement of facts listing 22 major international comparator
airlines of which only six were Canadian. For those major international
airlines for which complete data was available, 80% of pilot positions had
required mandatory retirement at age 60 or younger in 2003, and the Tribunal
concluded that this remained the case in 2005. Thus, age 60 was held to be the
mandatory age of retirement for the majority of positions similar to those of
the complainants.
[31]
The Tribunal further found that section 15(1)(c) of the
CHRA did not contravene section 15(1) of the Charter. In doing so, it
applied Law, in which the Supreme Court had stated that the overriding
concern was to protect and promote human dignity. It concluded that to
continue an arrangement which constituted prima facie age discrimination
based on a justification of section 15(1)(c) of the CHRA did not have a
negative impact on the complainants’ dignity, in the context of a system which
was designed to allocate/spread the responsibilities and benefits of being an
Air Canada pilot over different stages in pilots’ careers.
[32]
Because of its
finding on the section 15 Charter issue, the Tribunal did not have to
decide whether section 15(1)(c) of the CHRA could be justified under
section 1 of the Charter, nor whether the mandatory retirement policy
was a BFOR under sections 15(1)(a) and 15(2) of the CHRA.
B. Vilven
v Air Canada, 2009 FC 367 [Vilven]
[33]
Messrs. Vilven and
Kelly applied for judicial review of the Tribunal’s two decisions referred to
above. With respect to the normal age of retirement issue, Justice Mactavish
rejected the Tribunal’s test for determining the comparator airlines. She
concluded that it erred by focusing on the subjective perceptions of pilot
positions such as status or prestige, when the characteristics of comparator
airlines should have been based on the objective duties and functional
responsibilities of the position in question, that is “what pilots actually
do”.
[34]
Rather than setting
aside the decision, however, since the factual foundation was based upon the
Agreed Statement of Facts describing the major international airlines, she
concluded that the five Canadian airlines on that list should constitute the
comparator airlines for the purpose of determining normal age of retirement. On
the basis of those airlines, as well as Air Canada, the Court upheld the
Tribunal’s decision that 60 was the normal age of retirement for individuals
employed in positions similar to those occupied by Messrs. Vilven and Kelly
prior to their retirement.
[35]
A series of decisions
on the constitutionality of section 15(1)(c)
followed, but ultimately the provision was found not to infringe the Charter.
Accordingly, Justice Mactavish’s decision on the normal age of retirement
resulted in the dismissal of the Vilven and Kelly complaints. Its legacy in the
present matter relates to the interpretation of her test to determine
comparator airlines, which formed the basis of the Tribunal’s decision in this
matter in favour of Air Canada. Ultimately, I set aside the Tribunal decision
for not having properly applied Justice Mactavish’s test
[36]
Justice Mactavish thereafter reviewed the Tribunal’s
decision that section 15(1)(c) did not infringe the Charter. The Tribunal
had considered McKinney, Gosselin v Quebec (AG), 2002 SCC 84, [2002] 4 S.C.R. 429
[Gosselin] and subsequent jurisprudence. It
reached the conclusion that the loss of the opportunity to challenge the MRP
had not violated the dignity of the complainants nor failed to recognize them
as full and equal members of society.
[37]
Turning to the
constitutional question, Justice Mactavish asked whether section 15(1)(c)
of the CHRA violated section 15(1) of the Charter. She reviewed the
Supreme Court jurisprudence on mandatory retirement, including the Law case
and R v Kapp, 2008 SCC
41, [2008] 2 S.C.R. 483 [Kapp], as well as the Tribunal’s decision on the issue. She
commented that the focus of section 15(1) of the Charter was on
preventing governments from making distinctions based on enumerated or
analogous grounds which had the effect of perpetuating group disadvantages and
prejudice or which imposed disadvantages on the basis of stereotyping. For
there to be discrimination, there first had to be a distinction, and then that
distinction had to be shown to create a disadvantage. She took note of the
Supreme Court’s recognition that “human dignity,” as an abstract and subjective
notion, posed difficulties as a test, and that the perpetuation of disadvantage
or stereotyping was a preferable test.
[38]
She noted that the
objective of section 15(1)(c) of the CHRA had previously been described
as allowing for the continuation of a socially desirable employment regime,
which included pensions, job security, wages, and benefits. It was intended to
create an exception to the quasi-constitutional rights otherwise provided by
the Act.
[39]
Examining section
15(1)(c), she found that by exempting mandatory retirement from conduct
which would otherwise amount to prima facie age discrimination, it
created a distinction based on an enumerated ground. She noted that the
relevant comparison was between older workers having exceeded the normal age of
retirement for their positions and younger workers in similar positions who had
not yet reached that age. The effect of the provision was to deny the older
workers the equal protection and benefit of the CHRA.
[40]
She then considered
whether this distinction created a disadvantage by perpetuating prejudice or
stereotyping. Air Canada had argued, citing Gosselin, that age-based
distinctions are a common way of ordering our society and do not automatically
evoke a pre-existing disadvantage. Justice Mactavish commented that this was
based on comments made by the Supreme Court in a case involving a statutory
age-based distinction that had an adverse differential effect on younger
individuals, and that age-based section 15 claims were typically brought by
older people, who were presumed to lack abilities which they might in fact
possess.
[41]
The Tribunal found that
Messrs. Vilven and Kelly were members of a group identified as older workers, a
group the Supreme Court had repeatedly recognized as suffering from pre-existing
disadvantages and stereotyping. In addition to its comments in Gosselin
and Law, the Supreme Court referred in McKinney to “the
stereotype of older persons as unproductive, inefficient, and lacking in
competence.” By denying the benefit of, in that case, the Ontario provincial Human
Rights Code, RSO 1990, c H.19, to older workers, the effect was to
reinforce “the stereotype that older employees are no longer useful members of
the labour force and their services may therefore be freely and arbitrarily
dispensed with.”
[42]
As discussed below,
when raising the issue as to whether the differential adverse impact evidence
not before Justice Mactavish should give rise to an issue of whether the MRP is
discriminatory, I query the validity of a widespread negative stereotype
against older workers in our present society.
[43]
In the cases of
Messrs. Vilven and Kelly, the Tribunal accepted this general proposition, but
found that there was no indication that either complainant had personally
experienced such age-related disadvantages or stereotypes. Justice Mactavish
observed that first, to the extent that the analysis was of the group to which
the claimants belonged – older workers – it was clear that there was
pre-existing disadvantage, vulnerability, stereotyping, or prejudice. Second,
although there was no concern with the individual abilities of Messrs. Vilven
and Kelly, they were nonetheless disadvantaged by being forced to leave
positions that they clearly loved, merely because they had reached the age of
60. She was satisfied that this had the effect of perpetuating a group
disadvantage, suggesting that the MRP violated section 15(1) of the Charter.
[44]
She examined whether
the provision had an ameliorative purpose or effect which could save it. Air Canada had argued that it had the effect of freeing up positions for younger workers.
However, Justice Mactavish found that there had been no suggestion that younger
workers constituted a disadvantaged group which was being targeted by the CHRA.
[45]
As for whether the MRP
is discriminatory as opposed to the constitutionality of section 15(1)(c), to the extent that there is overlapping of relevant considerations, I
conclude that there is an important distinction between the factual foundation
in the matter that was before Justice Mactavish and the current matter. Based
on the evidentiary record which was placed before me, I find that the increase
in the retirement age to 63 would result in a significant adverse differential
impact on younger pilots.
[46]
In Vilven,
Justice Mactavish pointed out that the Supreme Court had stated in McKinney that
legislation that had as its objective the forcible retirement of older workers
in order to make way for younger workers would be in itself discriminatory,
since it would assume that the continued employment of some individuals was
less important and of less value to society, than the employment of other
individuals, based solely on age.
[47]
In the case of
Messrs. Vilven and Kelly, the interest at stake was the ability to continue to
work in the career of their choice. The importance of this could not be
overstated, commented Justice Mactavish. She concluded that section 15(1)(c)
of the Act violated section 15(1) of the Charter, by denying the
equal protection and equal benefit of the law to workers over the normal age of
retirement for similar positions.
[48]
Consequently, she
quashed the Tribunal's decision as it related to the Charter, and
remitted the matter to the Tribunal for determination of whether section 15(1)(c)
of the Act could be demonstrably justified as a reasonable limit in a
free and democratic society under section 1 of the Charter.
[49]
In the event that the
Tribunal determined that section 15(1)(c) of the CHRA was not saved
under section 1 of the Charter, she directed that it address the issue of
whether the age 60 retirement rule was a BFOR for Air Canada within the meaning
of section 15(1)(a) of the CHRA.
C. Vilven
v Air Canada, 2009 CHRT 24 [Vilven Tribunal #2]
[50]
As a result of the
Federal Court decision in Vilven, two further issues were remitted to the
Tribunal for determination: whether section 15(1)(c) of the CHRA could
be justified under section 1 of the Charter and whether Air Canada and
ACPA had established a BFOR for mandatory pilot retirement at age 60.
[51]
In assessing whether
section 15(1)(c) of the CHRA was saved under section 1 of the Charter,
the Tribunal applied the test articulated by the Supreme Court in R v Oakes,
[1986] 1 S.C.R. 103, [1986] SCJ No 7 (QL) [Oakes]. It concluded that
section 15(1)(c) of the CHRA could not be justified under any of the
elements of the test.
[52]
It was thus necessary
for the Tribunal to consider whether Air Canada and ACPA had demonstrated that
mandatory retirement at age 60 constituted a BFOR for Air Canada pilots.
[53]
In
answering this question, the Tribunal applied the Meiorin test
established by the Supreme Court.
[54]
According
to the Tribunal, neither Messrs. Vilven and Kelly nor the Commission disputed
that the first two components of the Meiorin test had been satisfied (that
the MRP be rationally connected to the performance of the job and good faith
adoption of the provision). This finding was later disputed before the Federal
Court. In any event, the Tribunal only considered what it saw as the "real
issue": whether Messrs. Vilven and Kelly could be accommodated without
causing undue hardship to Air Canada and/or ACPA.
[55]
In analyzing
undue hardship, the Supreme Court indicated in Meiorin that the factors
of health, safety, and costs listed at subsection 15(2) of the CHRA were not
entrenched unless expressly included or excluded by a statute being
considered. Further support for the non-exhaustive nature of this list could
be found in McGill University Health Centre v Syndicat des employés de
l’Hôpital général de Montréal, 2007 SCC 4, [2007] 1 S.C.R. 161, where the
Supreme Court emphasized that the factors which would support a finding of
undue hardship were to be applied with flexibility and common sense.
[56]
In the
case before the Tribunal, Air Canada argued that the ICAO standards caused
undue hardship – that it would have imposed major financial burdens and
disruption to operations to accommodate captains over the age of 60 under the
pre-November 2006 rules, which banned them from flying internationally. The
Tribunal considered the submissions as to the hardship resulting from the
difficulty of scheduling pilots over 60. After examining the evidence the
Tribunal found that Air Canada had not established that the retirement of Air Canada pilots at age 60 caused hardship and dismissed its BFOR defence.
[57]
ACPA, in a
different situation as a union and not an employer, argued that the hardship
from its point of view, as in Renaud, was whether there would be
prejudice to its members if the accommodation measures were adopted.
[58]
ACPA
submitted that removal of the MRP would limit the number of positions available
to pilots under 60, dilute their seniority, interfere with their ability to
plan for retirement in terms of timing and pension, and have a negative effect
on morale. It also claimed that the difference
in salary which would result from younger pilots being blocked from moving up
would be measured in the tens of thousands of dollars. It offered expert
opinion evidence to the effect that three to ten percent of pilots could be
expected to work longer if allowed to, and that assuming they worked for an
average of three extra years each, this would delay promotion for younger
pilots by one to four months.
[59]
The Tribunal
found that a delay in career progression and salary increases for younger
employees was not a substantial interference with their rights, and that it was
not more important to make way for younger workers than to continue the
employment of older workers. In this regard it concluded that “It is not as
ACPA stated, that the over 60 pilots would be taking money out of the younger
workers' pockets if the age 60 rule was removed.” Rather, the younger pilots
would take longer to achieve the salary increases that they desired. It also
found that “Offset against the delay in career progression would be the fact that
the younger pilots would have the freedom - when they reached age 60 - to work
as long as they needed or wished to work.”
[60]
The Tribunal
also rejected the proposition that younger pilots would not be able to enjoy
the benefits of the seniority system if the older pilots were not forced to
retire, as not having been established on the evidence and as being
inconsistent with the human rights principle stated in Renaud.
Insisting that the absolute preservation of a younger pilot’s seniority took
precedence over the continued employment of older colleagues was a purely
age-based and therefore arbitrary judgment about the relative worth to society
of the work performed by each age group, and the relative importance to the
individual of being employed. It concluded that ACPA had also not established
that the retirement of Air Canada pilots at age 60 constituted a BFOR.
D. Vilven v Air Canada, 2010 CHRT 27 [Vilven
Tribunal Damages]
[61]
In this
decision, the CHRT found that the appropriate way to remedy the discrimination
against Messrs. Vilven and Kelly was to order their reinstatement as pilots
with Air Canada. This involved “unwinding” the previous pension transactions;
the two successful complainants were to repay the pension payments they had
received and remit the pension contributions they would have made.
E. Air Canada Pilots Association v Kelly, 2011 FC 120
[Kelly]
[62]
Air
Canada and ACPA both applied for judicial review of Vilven
Tribunal #2, concluding that section 1 of the Charter
did not justify section 15(1)(c) of the CHRA. Justice Mactavish
dismissed their applications, which – until this was reversed by the Court of
Appeal - foreclosed any argument regarding the normal age of retirement issue.
Ultimately, however, by the Court of Appeal’s decision in Air Canada Pilots’ Association v Kelly, 2012
FCA 209, the
complainants’ case was dismissed on this issue based upon Justice Mactavish’s
conclusions in Vilven on normal age of retirement.
[63]
With
respect to the BFOR issue, only Air Canada sought judicial review of the Tribunal’s
decision. It again argued the hardship imposed by the ICAO standards governing
international flights.
[64]
As for
the scope of the undue hardship factors with regard to accommodation, Justice
Mactavish rejected the Tribunal’s determination that it could look at factors
other than health, safety, and cost. Air Canada had argued that elimination of
the MRP would interfere with employee morale, and that this was an eligible
factor for consideration.
[65]
Justice
Mactavish analyzed section 15(2) in light of both the interpretive principle of
expressio unius est exclusio alterius (to express one thing is to
exclude another) and the principle that defences within human rights statutes
should be interpreted narrowly. She concluded that the express inclusion of
three specific factors was to be construed as limiting consideration to those
three factors.
[66]
She also
found that as human rights legislation is “the final refuge of the
disadvantaged and the disenfranchised” and “the last protection of the most
vulnerable members of society” (Zurich Insurance Co v Ontario (Human Rights Commission),
[1992] 2 S.C.R. 321, [1992] SCJ No 63 (QL) [Zurich] at para 18), only
matters of a sufficient gravity as to have a demonstrable impact on the
operations of an employer in a way that related to health, safety, or cost
should be taken into account as defences. In addition, she imputed knowledge to
Parliament of decisions by the Supreme Court which provided unlimited scope to
hardship factors as demonstrating the legislative intent to narrow these to the
factors expressly referred to in the provision. As I respectfully disagree with
this interpretation of section 15(2), these and other considerations will be
reviewed below.
[67]
Justice Mactavish set aside
the Tribunal’s decision dismissing Air Canada’s BFOR defence. She found that
it had failed to address significant evidence on hardship as it pertained to
the impact on scheduling of retaining pilots older than the ICAO standards
permitted on international flights which meant that this element of its
decision lacked the transparency and accountability required of a reasonable
decision.
[68]
She also
found that, contrary to the Tribunal’s view, no concession had been made on
steps one and two of the Meiorin test, and sent that issue back for
reconsideration.
F. Kelly
& Vilven v Air Canada & ACPA, 2011 CHRT 10 [Kelly Tribunal]
[69]
By the direction of
the Federal Court in Kelly, the Tribunal was required to consider first
whether Air Canada had met the first two steps of the Meiorin test and
second, whether it had demonstrated hardship for the period after 2006 due to
scheduling complications caused by the elimination of the MRP and costs
associated with those complications.
[70]
The Tribunal noted
that the first and second steps of the Meiorin test require an
assessment of the legitimacy of the standard's general purpose, and the
employer's intent in adopting it. This was to ensure that, when viewed both
objectively and subjectively, the standard does not have a discriminatory
foundation.
[71]
The CHRT found that
for decades Air Canada had engaged in a legitimate and meaningful bargaining
process with the pilots’ union that resulted in an enduring collective
agreement which enshrined seniority and provided for mandatory retirement at
age 60 with a generous pension. As a result, Air Canada had been able to
effectively manage the introduction of new pilots to replace a predictable
number of retiring pilots. Assessing this situation both subjectively and
objectively, the Tribunal concluded on a balance of probabilities that the MRP
did not have a discriminatory foundation. In this
matter, the complainants in the current proceedings argue that Air Canada did
not consider measures that could have been implemented to accommodate the
retiring pilots, allowing them to fly without causing undue hardship, an argument
I reject as not being feasible in the circumstances.
[72]
The Tribunal, in what I would describe as an
abridged analysis, also found that Air Canada had proved that it would suffer
undue hardship in accommodating the complainants due to the restrictions of the
ICAO over/under rule. It concluded that abolishing mandatory retirement would
have negative consequences for Air Canada by significantly increasing
operational costs and inefficiency in the scheduling of pilots and, to a lesser
extent, causing negative ramifications for the pilots’ pension plan and the
collective bargaining agreement, by affecting the rule of seniority. In this
regard, it would appear that the Tribunal ignored the direction in Kelly
limiting hardship to the factors in section 15(2).
[73]
In the matter before me, it should be noted that
the Tribunal came to a different conclusion, rejecting Air Canada’s BFOR defence. I dismiss Air Canada’s application seeking to set aside this later Tribunal
decision for the reasons described below.
G. Adamson
v Air Canada, 2011 CHRT 11 [“Adamson” or “this matter”]
[74]
This is the matter
with which I am seized. A detailed analysis follows below. There are two legacy
issues arising out of the previous decisions, but now grounded in new evidence:
from Vilven, the normal age of retirement defence raised by Air Canada; and from Kelly, the BFOR defences raised by Air Canada and ACPA.
H. Air Canada Pilots Association v Kelly, 2012 FCA 209
[Kelly FCA]
[75]
As already mentioned,
the Federal Court of Appeal overturned the decision in Kelly that
rejected Air Canada and ACPA’s argument that section 15(1)(c) of the
CHRA could be justified pursuant to section 1 of the Charter. The Court
found that this question had been decided by the Supreme Court in McKinney and that both the Tribunal and the Federal Court had erred in concluding
that they were not required to follow that precedent. It therefore returned
the matter to the Tribunal with the direction to dismiss the complaints.
[76]
Upon refusal of leave
to appeal, the Vilven proceeding was brought to end as ratified in Vilven
v Air Canada, 2013 FC 368, which dismissed the appeal from the Kelly Tribunal
decision (2011 CHRT 10), concluding that “the central legal question [was]
fully determined by the decision of the Court of Appeal”.
III.
ISSUES
[77]
The issues in this matter are:
a.
The appropriate standard of review;
b.
With respect to determining the normal age of
retirement pursuant to section 15(1)(c) of the Act, did the Tribunal
err in its interpretation of the test applied to determine the comparator
airlines which had employees working in positions similar to the positions of
Air Canada pilots?
c.
Did the Tribunal err in deciding that Air Canada
was not entitled to advance a BFOR defence under sections 15(1)(a) and
15(2) of the Act?
d.
Did the Tribunal err in deciding that ACPA was
not entitled to advance a BFOR defence under sections 15(1)(a) and 15(2)
of the Act, with specific reference to the following issues:
i.
Whether the defence under section 15(1)(a)
applies to ACPA as an employee organization?
ii.
Whether section 15(2) is limited to the hardship
factors of health, safety and cost?
iii.
Whether ACPA is barred by the CHRA from
justifying the rule of retirement at age 60 in the collective agreement by
demonstrating hardship in accommodating the change to the rule and if not,
whether the Tribunal should apply a modified application of the test in Meiorin
in relation to the principles in Renaud?
iv.
Whether the Tribunal erred in its finding that
no undue hardship to ACPA would arise from the elimination of compulsory
retirement at age 60?
e.
Did the Tribunal err in finding that the rule in
the collective agreement imposing retirement at age 60 was discriminatory?
IV.
ANALYSIS
A. Standard
of Review
[78]
As the Supreme Court noted at para 54 of Dunsmuir
v New Brunswick, 2008 SCC 9, [2008] 1 S.C.R. 190 [Dunsmuir],
the deferential standard of reasonableness will normally be called for where a Tribunal
is interpreting its own statute or statutes closely connected to its function.
However, there
are broader questions of law at play in this matter which are of "central
importance to the legal system ... and outside the ... specialized area of
expertise", as described in Dunsmuir at para 60.
[79]
The standard of review is
correctness for the legal questions of whether unions are entitled to advance a BFOR defence under section 15(1)(a) and
the scope of the hardship factors in section 15(2) of the CHRA. Because another judge of the Federal Court
has made rulings on aspects of the BFOR and hardship issues, these issues must
be reviewed on a correctness standard. The question of the
modification of the Meiorin test as applied to unions has not previously
been decided, nor has the requirement to establish substantive discrimination as
an element of the term “discriminatory practice” under the CHRA.
[80]
The standard of review for
the CHRT’s application of the direction from Vilven and its application
of the comparator group test is reasonableness. Reasonableness is also the
standard of review for the CHRT’s overall decision in light of the pension
evidence available to it. As the Supreme Court
explained in Dunsmuir at para 47, “reasonableness is concerned mostly
with the existence of justification, transparency and intelligibility within
the decision-making process. But it is also concerned with whether the decision
falls within a range of possible, acceptable outcomes which are defensible in
respect of the facts and law.”
[81]
Justification, transparency and intelligibility
all share the purpose of providing a reasonable explanation for the
basis of a decision, without which it lacks legitimacy.
[82]
Something is justified when the components of
the explanation line up logically to establish the conclusion. Something is
intelligible when the explanation is understandable, such that the
justification can be discerned from the reasons. Transparency is really a
subset of both justification and intelligibility. A failure of transparency is
most often a failure to consider properly a significant outlier issue or fact
that stands in the way of a logical and understandable explanation. The
substantial failure to meet any of the three requirements is normally fatal to
the decision.
[83]
Correctness applies where there is not a range
of acceptable answers, but only a single right one. The Supreme Court
explained in Dunsmuir at para 50 that in such cases a reviewing court “will not show deference to the decision maker's reasoning
process; it will rather undertake its own analysis of the question. The
analysis will bring the court to decide whether it agrees with the
determination of the decision maker; if not, the court will substitute its own
view and provide the correct answer. From the outset, the court must ask
whether the Tribunal's decision was correct.”
[84]
As well, for this
matter it is important to distinguish between reweighing the evidence and not
considering factors or conclusions that flow from undisputed facts based on the
evidence. Weighing evidence has to do with “believability”. It relates to the
credibility and reliability of the evidence in proving a probative relevant
conclusion of some kind. Weighing of evidence is for the ultimate determination
of the trier of fact, and not the trier of law.
[85]
However, a court
conducting judicial review may interfere if it concludes that a decision maker
has mischaracterized or failed to consider a significant factor or conclusion
that is logically or patently obvious from undisputed facts. Suresh v Canada (Minister of Citizenship and Immigration), 2002 SCC 1, [2002] 1 S.C.R. 3 supports
this proposition at para 37 as follows:
37 The passages in Baker
referring to the "weight" of particular factors (see paras. 68 and
73-75) must be read in this context. It is the Minister who was obliged to give
proper weight to the relevant factors and none other. Baker does not authorize
courts reviewing decisions on the discretionary end of the spectrum to engage
in a new weighing process, but draws on an established line of cases concerning
the failure of ministerial delegates to consider and weigh implied limitations
and/or patently relevant factors: see Anisminic Ltd. v. Foreign
Compensation Commission, [1969] 2 A.C. 147 (H.L.); Re Sheehan and Criminal
Injuries Compensation Board (1974), 52 D.L.R. (3d) 728 (Ont. C.A.); Maple Lodge
Farms Ltd. v. Government of Canada, [1982] 2 S.C.R. 2; Dagg, supra at paras.
111-12, per La Forest J. (dissenting on other grounds).
[Emphasis added]
B. Section 15(1)(c): Normal Age of Retirement
[86]
Section 15(1)(c) has now been repealed.
It read as follows when in force:
|
15. (1) It is not a
discriminatory practice if
[. . .]
(c) an individual’s employment is terminated because
that individual has reached the normal age of retirement for employees
working in positions similar to the position of that individual;
|
15. (1) Ne constituent pas
des actes discriminatoires:
[. . .]
c) le fait de mettre fin à
l’emploi d’une personne en appliquant la règle de l’âge de la retraite en
vigueur pour ce genre d’emploi;
|
[87]
As described above, the defence based on section
15(1)(c) of the CHRA was successfully relied upon in Vilven as a
defence to the complaints. Although section 15(1)(c) has now been
repealed, it continues to govern these proceedings as it was in force during
the life of the collective agreement and at the time the complaints were
initiated. The provision also survived a constitutional challenge when upheld
by the Federal Court of Appeal.
[88]
The proper characterization of the test
developed in Vilven (“the Vilven test”) to determine the
comparator pilots was the central issue before the Tribunal in this matter. I
am, of course, not bound to apply the Vilven test were I to conclude
that it would be inappropriate to do so. However, I agree that Vilven described
the appropriate test, if correctly characterized, which is what this issue
turns on. As I have indicated, I am in agreement with the complainants’
characterization of the test and in that guise adopt it as the appropriate test
to determine the comparator pilots in this matter.
(1) The Vilven
Test
[89]
The principal purpose of the Vilven test
was to determine which airlines (“the Comparator Airlines”) had positions
similar to those of the pilots in Air Canada. The methodology settled upon was
described as “statistical”, but in reality it was based on very simple
mathematics. The pilots of the Comparator Airlines are added to those of Air Canada, the total which serves as the denominator. From this total, the pilots of the
airlines with a mandatory retirement age of 60 are added up to constitute the numerator.
If the numerator is more than 50% of the denominator, then 60 is the normal age
of retirement and the defence under section 15(1)(c) applies.
[90]
Thus, the case turns on who the Comparator
Airlines are. If there are fewer of them, Air Canada wins by its dominance of
the industry. If there are many of them, the complainants would appear to
succeed because most other airlines use a retirement age of 65, and combined
they outweigh Air Canada and the few other airlines which have an age 60
retirement provision.
[91]
In her ruling, Justice Mactavish rejected the Tribunal’s
conclusion that international airlines could be used as the Comparator
Airlines. Instead, she substituted a test comprising Canadian airlines
possessing a list of attributes that she described in paras 111 and 112, that
are relied upon by the respondents, and then again in a slightly different
version in para 170. These two reiterations of the list can be interpreted
differently. The respondents rely upon the first version, at paras 111-112,
while the complainants rely upon the second version, at para 170, and they also
interpret them differently. The two versions are as follows:
In the first
description;
[111] The essence of what Air Canada pilots do is to fly aircraft of
varying sizes and types, transporting passengers to both domestic and
international destinations, through Canadian and foreign airspace.
[112]… In light of the essential features of Messrs. Vilven and
Kelly’s positions, the appropriate comparator group should have been pilots
working for Canadian airlines who fly aircraft of varying sizes and types,
transporting passengers to both domestic and international destinations,
through Canadian and foreign airspace.
[Emphasis
added]
Versus
the second description;
[170] However, as was explained earlier, I am of the view that the
tribunal erred in its identification of the "positions similar" to
those occupied by Messrs. Vilven and Kelly. It is pilots working for Canadian
airlines flying aircraft of various sizes and types to domestic and
international destinations through Canadian and foreign airspace that form the
proper comparator group.
[92]
All of the above formulations of the Vilven test
contain the same four determinative attributes of what Canadian airline pilots
do, i.e. flying aircraft of (1) various sizes and (2) various types to (3)
domestic destinations and (4) international destinations. The difference in the
positions of the parties and the outcome in determining the comparator pilots
is whether these factors should be applied conjunctively, based in large part
on the inclusion of the word “both” in paragraphs 111 and 112 above, or disjunctively
as argued by the complainants.
[93]
The distinction is significant inasmuch as the
conjunctive formulation will exclude airlines which do not exhibit all
of the factors in the Court's test. Conversely, treating the factors
disjunctively includes the pilots of any airline exhibiting any single
characteristic from the list of attributes described by Justice Mactavish.
[94]
In applying the test in Vilven the Court
was limited to evidence on comparator airlines contained in an Agreed Statement
of Facts. Because the parties appeared to have agreed that the Comparator
Airlines comprised major airlines that flew internationally, the evidence
comprised 22 airlines only six of which were Canadian which flew to
international destinations. The Canadian airlines included the four major
competitors at that time, WestJet, Transat Air, Skyservice, and CanJet, which
along with Air Canada and its former subsidiary Jazz Air constituted the
totality of Canadian airlines found by Justice Mactavish to have similar pilot
positions. On the basis of this group of airlines dominated by Air Canada and Jazz, the majority normal age of retirement of pilots in Canadian airlines was
age 60.
[95]
There is no Agreed Statement of Facts before the
Tribunal in the present matter. Instead, the parties introduced evidence on 38
airlines, the pilots of which it could be argued did what Air Canada pilots do
based on the attributes described in the Vilven test.
[96]
In determining the Comparator Airlines, the Tribunal
adopted Air Canada’s conjunctive interpretation of the Vilven test as
implicitly described in paragraph 55 of its decision and demonstrated in the
application of the test to eliminate airlines from the list of those exhibiting
one or more of the attributes described by Justice Mactavish.
55 Last
but certainly not least, Captain Prentice has constructed his own formula as
to what criteria should be used to determine the comparator group. He
includes only two of the five criteria enunciated by the Court. I have
decided that all five should be used to determine the comparator.
[Emphasis added]
[97]
Of the total number of airlines on which
evidence was introduced before the Tribunal, only eleven smaller airlines, in
addition to Jazz Air, met the conjunctive Vilven test. As mentioned, due
to the numerical dominance of Air Canada and its former subsidiary Jazz Air,
pilots with a retirement policy of age 60 were found to make up the majority of
pilots, i.e. 56% of all pilots of Comparator Airlines.
[98]
I find it significant that the application of the
conjunctive test had the effect of eliminating Air Canada's 10 closest major
competitors. Of the comparator airlines considered in Vilven, Transat
Air, Skyservice and CanJet were struck from the list because they do not fly
domestically, while WestJet was not included because it only flew one type of
jet, although in a variety of sizes.
[99]
An outcome that eliminates Air Canada’s major competitors on a test meant to compare airlines based on the similarity of
their pilots’ functions and duties is patently unreasonable and results from the
application of wrong principles. I will now turn to an explanation of the Tribunal’s
failure to apply proper principles, which led to the unacceptable outcome which
I determine must be set aside despite the deference owed the decision-maker.
(2) The Tribunal’s Reasoning
[100]
The Tribunal sought to apply what it considered
to be the essence of the directions of the Federal Court decision in Vilven
#1. For purposes of completeness and as an aid to my analysis, I set out
the full Tribunal reasoning in paragraphs 6 to 12 and 20 to 25 of its decision,
marked with my emphasis at significant passages:
6 The
Court stated that s. 15(1)(c) requires two questions to be answered. First,
"what is the proper comparator group to identify the positions that are similar to that occupied
by Air Canada and secondly, what is the normal age of retirement".
7 And
when assessing whether a position is "similar" to that occupied by
the complainants, the focus should be on the objective duties and functional
responsibilities of the position in question. That is, what do Air Canada pilots actually do?
8 For
the Federal Court, "the essence of what Air Canada pilots do is to
fly aircraft of varying sizes and types, transporting passengers to both
domestic and international destinations, through Canadian and international
airspace", (para. 111).
9 Thus
the appropriate comparator group for the complainants should be "pilots working for
Canadian airlines who fly aircraft of varying sizes and types to both
domestic and international destinations, through Canadian and international
airspace", (para.112) ("Test").
10 The
Court reiterated the Test later in paragraph 125 of its decision where it said,
"To
summarize my findings to this point: the essence of what Air Canada pilots do
can be described as flying aircraft of varying sizes and types, transporting
passengers to both domestic and international destinations, through
Canadian and foreign airspace. There are many Canadian pilots in similar
positions, including those working for other Canadian airlines. These pilots
form the comparator group for the purposes of paragraph 15(1)(c) of the
CHRA."
11 The
Court observed that, as of the date of the Agreed Statement of Facts that the
parties submitted to the Vilven
tribunal, there were five airlines in Canada, apart from Air Canada, that transported passengers to domestic and international destinations. They were
Jazz, Air Transat, CanJet, Skyservice and WestJet. However, there is nothing
in this observation that suggests that the Court accepted that these five
airlines satisfied all of the comparator criteria set out in paragraphs 112 and
125.
12 Interestingly,
the Court went on to say in paragraph 170 of its decision that, "as explained earlier,
the Tribunal erred in its identification of 'positions similar'". It is
"pilots working for Canadian airlines flying aircraft of various sizes to
domestic and international destinations, through Canadian foreign airspace that
form the proper comparator group". The words,
"transporting passengers" and "both" in reference to
destinations and "types" in reference to aircraft are absent in this
formulation of the test for the comparator group.
[. . .]
B. What Should be the Test
for the Comparator Group?
20 For
the comparator group, the Respondents accept and rely on the Court's formula in
paragraphs 112 and 125. The Complainants have a different opinion. Both the
Complainants and the Canadian Human Rights Commission ("CHRC") assert
that the formula prescribed by the Court in these two paragraphs of its
decision should not be literally applied in determining the appropriate
comparators. Rather, the test that the Court set out in Vilven was dictated by the facts in that case
and should only be considered as guidelines to assist the Tribunal in defining
the comparator group.
21 Both
the Complainants and the CHRC point to paragraph 170 of the Vilven decision position
as support and propose that the Test should be "pilots working for
Canadian airlines flying aircraft to either domestic or international
destinations through Canadian or foreign airspace".
They say that the Court inserted "both" in relation to domestic and
international destinations in the Test to emphasize that the Tribunal erred in
limiting the comparator group to those airlines that fly only to international
destinations. The Court did not intend that the definition of the comparator
group to be more restrictive than that of the Tribunal. It should be read
disjunctively to include an airline whether it operates only domestically or
only internationally.
22 Further,
both the Complainants and the CHRC assert that the absence of "varying
types" in paragraph 170 makes sense. Otherwise, two of Air Canada's major competitors, who fly only one type of aircraft would be excluded from the
comparator group.
23 This
is so even though their pilots do essentially do what Air Canada pilots do, fly
passengers to domestic and international destinations. The CHRC would also drop
Varying Sizes from the Test arguing that size does not matter. Whether an
aircraft is small, medium or large, the essence of what a pilot does is the
same.
24 It
is unfortunate that the Court went on to state the comparator test for yet a
third time and somewhat differently as it did in paragraph 170 of its decision.
There is no explanation from the Court, at this late stage of its decision, as
to why the comparator test should be modified. In my view, it should be regarded
more as a matter of inadvertence rather than a restatement of the comparator
group test.
25 What
the Tribunal must do in this case is what the Court did in Vilven,
which is to ask and answer the question, what is the essence of what Air Canada
pilots do? The evidence in this case demonstrates that what Air Canada
pilots do is as described by the Court in Vilven
in paragraphs 112 and 125. Thus, the criteria to be applied in this case will
be the same as the criteria applied in Vilven to
determine the appropriate group.
[Emphasis added]
(3) Errors in the Tribunal’s Decision
(a) The Unreasonableness of the Elimination of Air Canada’s Competitors
[101]
While deference is owed to the Tribunal, it is
unreasonable to eliminate Air Canada’s main competitors from a list of airlines
in Canada whose pilots actually do the same thing as Air Canada pilots. This illogical result is compounded by a number of errors of principle in
the Tribunal’s analysis, which are described under the various headings that
follow.
(b) The Failure to Conduct a Functional
Analysis of the Positions
[102]
In Vilven the Court concluded that the
functions and duties of Air Canada pilots, what they actually do, should be the
overriding consideration in determining appropriate comparators.
[103]
As the complainants argued, given that Vilven
was based on an Agreed Statement of Facts, Justice Mactavish was not
required, nor able, to apply her test based upon the functions and duties of
Air Canada to determine whether there were similar positions in other airlines.
Vilven's only purpose, therefore, was to serve as a guide on how to
proceed.
[104]
In contradistinction to Vilven, in the
present matter the parties introduced concrete evidence on the characteristics
of a large variety of airlines. The Tribunal was therefore required to perform
the functional analysis described in Vilven to determine whether the
pilots of those airlines actually did the same thing as Air Canada pilots. Unless it first analyzed the functions of Air Canada pilots and determined
that every position possessed every one of the five criteria, it could not
eliminate pilot positions with other airlines on the basis that the
organization did not present one of the factors, such as flying only one type
of airplane or flying only domestically, and that this rendered the positions
dissimilar.
[105]
Take as an example the characteristics of
different sizes and types of airplanes. The employer had to discharge its onus
of demonstrating on the basis of the duties and functions of the Air Canada
pilots that the function of operating more than one size and more than one type
was sufficiently distinct from the function exercised by pilots in other
airlines operating only one type or only one size of airplane so as to exclude
the latter from being Comparator Airlines. Only then could the Tribunal decide
that a certain type or a certain size of airplane should be a limiting factor.
The same methodological requirement applied for airlines flying only
domestically or only internationally. This was not done.
[106]
I am fairly certain, however, that it would be
difficult to establish that the competitor airlines should be excluded on the
basis of a disjunctive interpretation, as there was no suggestion made by the
parties that the functions and duties of the pilots of, for instance, Air
Transat differed from those of Air Canada when flying to foreign destinations
such that any basis existed to eliminate Air Transat as a Comparator Airline.
[107]
Likewise, there does not appear to have been any
basis to have eliminated WestJet as a Comparator Airline on the functionally
irrelevant consideration that its pilots flew only one type of aircraft. Its
pilots obviously “actually do” what Air Canada pilots “actually do”. As its
principal competitor vying for the same passenger base, WestJet pilots “actually
do” what Air Canada pilots “actually do”: flying the same flights, with similar
types of airplanes.
[108]
The evidence showed that Air Canada pilots are
only permitted to fly one type of airplane at a time after extensive training
on that specific airplane, until they change to another plane and are retrained
on it. The actual duties and functions of a pilot at any given time relate to
only one type of aircraft.
[109]
The obviousness of the fact that WestJet and
Transat should qualify as Comparator Airlines demonstrates that the Tribunal
did not properly follow the directions in Vilven, which required the
comparison of the functions and duties of pilots and did not allow for the
elimination of an airline due to a criterion that had no impact on what the
pilots actually do.
(c) The
Failure to Conduct Contextual Analysis of the Vilven Reasoning
[110]
The interpretation of any term or definition
requires that meaning be found in the context of all the surrounding elements
that are said to be connected to the proposed term or definition. With respect
to the Tribunal’s interpretation of the concept of comparator airlines, my
first conclusion is that its reasoning is circular and tautological.
[111]
In its analysis, the Tribunal noted the Court's
different statements of the Comparator Airline test as regards the use of the
term "both" in relation to domestic and international destinations.
By concluding that the transportation of passengers by an airline had to be
both domestic and international, it set the tone for the treatment of the other
factors in Justice Mactavish’s list as also being conjunctive.
[112]
No logical explanation was provided by the Tribunal
for the inconsistent inclusion of the word “both”. Relying on the enunciation
of the test twice with the word “both,” and only once without, without a
contextual analysis of the Court’s intention in Vilven, was a
superficial and unreasonable explanation for why the restatement not including
“both” should be “regarded more as a matter of inadvertence rather than a
restatement of the comparator group test”.
[113] Instead, the Tribunal illogically relied upon its own evidence to
interpret the meaning of the direction from Vilven: “The
evidence in this case demonstrates that what Air Canada pilots do is as
described by the Court in Vilven in paragraphs 112 and 125”. With
respect, the Tribunal could not rely on the facts before it to make an ex
post facto
determination of the Vilven Court’s intention in prescribing the test to apply to such
facts. Furthermore, the evidence could also be said to show that pilots do
what was described in Vilven if the Court’s direction was interpreted
disjunctively, particularly as no pilot at any time is accomplishing functions
which simultaneously have all the attributes in the Vilven test.
[114]
In this regard, the Tribunal failed to respond
to the submissions of the complainants, beyond merely referring to them, which
demonstrates a lack of transparency. By not describing why it rejected the
principled submissions of the complainants supporting a disjunctive application
of the test in Vilven,
the Tribunal’s decision lacked transparency and justification. This further
vitiates the unreasonable outcome of eliminating Air Canada’s closest
competitors as Comparator Airlines. I find the complainants’ submissions
compelling and largely adopt them in my reasoning that follows.
(d) An Overlooked Reference to Explain the
Meaning of “Both”
[115]
As part of a contextual analysis of the
reasoning in Vilven, I have already referred to paragraph 170 of Justice
Mactavish’s decision where she identified similar positions in reference to
domestic and international destinations without using the term “both”. The fact
that she does not include the word “both” in that paragraph does not seem to be
inadvertent in light of her further explanation at paragraph 173. The latter
refers to Air Canada pilots and those of the Comparator Airlines
exhibiting the same characteristics required to meet her test:
[173]
The statistical information before the tribunal with respect to airline pilots
working for both Air Canada and other Canadian airlines flying aircraft of
various sizes to domestic and international destinations, through Canadian
and foreign airspace, reveals that at the time that Messrs. Vilven and Kelly
were forced to leave their positions at Air Canada, several Canadian airlines
allowed their pilots to fly until they were 65, and one had no mandatory
retirement policy whatsoever. Nevertheless, 56.13% of Canadian airline pilots
retired by the time they reached the age of 60.
[Emphasis
added]
(e) An
Implication that Vilven was Based on a Misapprehension of Evidence
[116]
Similarly, at paragraph 113 of the reasons, the
Court referred to the evidence from the agreed statement of facts relating to
the “five principal airlines in Canada (apart from Air Canada) that were engaged in transporting passengers to domestic and international
destinations. These were Jazz, Air Transat, CanJet, Skyservice and WestJet.”
[117]
Thus, while Air Transat, Skyservice and CanJet
were included in the list before Justice Mactavish and were used as comparators
to count employees in similar working positions to those of the Air Canada pilots
in Vilven, they were rejected as comparator airlines by the Tribunal in
this matter. The Tribunal at paragraph 174 of its decision concluded that they
did not meet the definition of "domestic", because as charter
airlines they flew only to international destinations.
[118]
From Justice Mactavish’s reasons the Tribunal
was aware, or at least should have been, that she had concluded that other
Canadian airlines were flying aircraft of various sizes to domestic and
international destinations, which could only be possible if the factors in her
test were treated disjunctively.
[119]
If it did not accept this explanation, the Tribunal
would have had to conclude that Justice Mactavish misapprehended the evidence
and made an erroneous finding that Air Transat, Skyservice or CanJet flew
domestically, as well as internationally.
[120]
I do not find that a reasonable conclusion given
the more logical explanation that the factors in the test were intended to be
treated disjunctively. It is also only in this manner that one avoids an
obviously unreasonable outcome that excludes Air Canada’s main competitors,
even though expectations would be that their pilots would occupy positions
having functions and duties the most similar to those of Air Canada.
(f) Too Limited a Comparator Group
[121]
Another contextual aspect of Justice Mactavish’s
reasoning was her discomfiture with limiting the comparator airlines to those
operating in Canada, due to Air Canada’s dominance thereby establishing the
norm in a statistical defence to a discriminatory provision. This was a
concern of the Court, as indicated in paragraph 171 of the reasons:
[171] I
also agree with the tribunal’s observation that there are problems associated
with using Canadian data for comparison purposes. Citing the tribunal decision
in Campbell, the tribunal noted that because of Air Canada’s dominant position within the Canadian airline industry, a comparison of pilot positions
within Canada would result in Air Canada setting the industry norm. This
would allow Air Canada to effectively determine the “normal age of retirement”
for the purposes of section 15(1)(c) of the Act.
[Emphasis
added]
[122]
Given the concern about Air Canada’s dominance
of the industry skewing the results of any survey of the normal age of
retirement of Canadian pilots, it is unlikely that the Court intended to
propose a test which would greatly narrow the comparator airlines, as opposed
to one which would tend to be more inclusive. In my view, this was a contextual
factor from the reasons in Vilven that the Tribunal should have taken
into consideration in applying Justice Mactavish’s test.
(g) A Contextual Interpretation
of “Both”
[123]
“Both” is an inherently ambiguous word because
it may be employed in different circumstances for different purposes. For
example, it may be used to emphasize an inclusive response to a question on
choice. In answer to a question from a host as to which type of vegetable the
guest would like with the rest of her serving, the answer could be “both”,
meaning to request that “both” types of food be placed on her plate. However,
in answer to a question from the same host as to whether the guest generally
likes to drink tea or coffee, the answer could also be “both”, this time
meaning that “either” would be welcome but not that she was expecting both
simultaneously.
[124]
The Court found itself effectively being asked a
similar question by the Tribunal’s definition of the attributes of a comparator
airline. It was confronted by the Vilven Tribunal #1 decision that
raised the question as to the essence of the positions occupied by Air Canada pilots. The Tribunal suggested that the essential attribute "was that they
flew on regularly scheduled international flights [or] on wide-bodied
aircraft to many international destinations with a major international
airline".
[125]
Contextually, the Tribunal was in effect asking
the Court whether it should use only international flights as the
salient comparator characteristic, to which the Court answered “No, use both
domestic and international flights.” This answer applied with similar force to
wide-bodied airplanes and smaller airplanes.
[126]
In the circumstances, it is understandable that
Justice Mactavish would have wanted to emphasize that the comparator pilots of
airlines should not be limited to those operating one type of airplane
(wide-bodied airplanes) or to airlines flying only to international
destinations.
[127]
I am satisfied that in the context of its
reasons the Court used the term “both” to ensure that it was understood that
flying only to international destinations was insufficient as a limiting
comparator factor, just as the test was not to be restricted to airlines flying
one size or type of aircraft.
(4) Conclusion
on Normal Age of Retirement
[128]
Recognizing the deference owed the Tribunal, I
nevertheless find that the Tribunal erred in principle in its interpretation of
the direction of the Court in Vilven as imposing a rule consisting of a
series of factors to be considered conjunctively, when the decision interpreted
in its context clearly directed the Tribunal to apply those factors
disjunctively.
[129]
On the basis of the foregoing, I adopt the
reasons of Justice Mactavish in Vilven as properly determining the
attributes of Comparator Airlines in so far as the enumerated factors are to be
applied disjunctively. Otherwise, I would respectfully disagree with her
decision on the basis of my reasons described above, which in my view require
the enumerated factors identified in her decision to be applied disjunctively
in order to avoid the unreasonable outcome of Air Canada’s major competitors
being eliminated as Comparator Airlines.
[130]
The Tribunal’s error in applying a restricted
exclusionary test based on a conjunctive application of the factors resulted in
an unreasonable outcome that eliminated suitable comparators in the form of Air
Canada’s competitors whose pilots obviously performed duties and functions
similar to those of Air Canada pilots.
[131]
The appropriate test required the Tribunal to
consider the functions and duties of Air Canada pilots as demonstrated by the
employer and then to consider the functions and duties of pilots at other
airlines based on the agreed statement of facts. Only then could it conclude
that, based on the position characteristics evaluated in accordance with Vilven,
the functions and duties of pilots in other airlines were sufficiently similar
or different so as to require their inclusion or elimination from consideration
as comparators. It should be remembered that Air Canada bore the onus
throughout on this issue.
[132]
Accordingly, the application of the complainants
is allowed and the decision of the Tribunal in 2011 CHRT 11 in respect of
normal age of retirement is set aside and remitted for reconsideration. Upon
reconsideration, the Tribunal is directed to apply the factors of the test in Vilven
disjunctively. It is also to determine attributes of similarity based on what
pilots actually do, i.e. whether the attributes of positions for pilots flying
large and small planes are similar, and so on.
C. AIR CANADA’S DEFENCE OF A BONA FIDE OCCUPATIONAL
REQUIREMENT
[133]
As described above, Air Canada was successful in Kelly in setting aside the CHRT’s decision rejecting its
BFOR defence. The redetermination ordered by the Federal Court never occurred
because the Federal Court of Appeal brought that litigation to an end when it
concluded that the defence of normal age of retirement found to apply in Vilven
was constitutional.
[134]
Air Canada is now attempting to achieve the same
result as in Kelly. It seeks to set aside the CHRT’s 2011 decision
concluding that the mandatory retirement rule did not constitute a BFOR for its
pilots as envisaged in section 15(1) of the CHRA. Air Canada submits that the Tribunal’s evaluation of uncontested key evidence failed to meet
the tests of intelligibility, transparency and justification such that its
decision must be overturned as unreasonable.
[135]
I reject these submissions. I find that the Tribunal
properly considered the evidence and that it was justified in rejecting Air Canada’s BFOR defence.
[136]
In considering the
three-step BFOR analysis of the Supreme Court’s decision in Meiorin, the Tribunal, for the second time, as it had done in Vilven Tribunal
#2, avoided ruling on whether Air Canada met the first two steps
regarding the legitimacy of the standard’s purpose and the good faith of the
employer in adopting the standard. It noted only that “this is not so much in
dispute between the parties as is their disagreement on step three, the
accommodation obligation.”
[137]
When later dealing with ACPA’s BFOR defence, the
Tribunal appeared to make reference back to Air Canada’s situation, by
commenting that it was difficult to see how the age of the pilot bore any
relationship to the performance of the job when Transport Canada did not impose any maximum age restriction.
[138]
It is not clear whether this statement was
intended to suggest that Air Canada did not meet the first factor of the Meiorin
test because age was not related to a standard of pilot performance. I cannot
leave any doubt on this issue because, as shall become apparent below, Air Canada’s fulfilment of the first step of the Meiorin test is relevant to my
conclusions on ACPA’s entitlement to plead a BFOR defence.
[139]
I conclude that being older than 60 was indeed
an appropriate BFOR consideration in relation to the ability of pilots to
perform their jobs in the years in question. The evidence established that
mandatory rules imposed on Canadian airlines pursuant to ICAO treaties would
prevent Air Canada from operating approximately 90 percent of its flights when
the Captain is over 60 years old, unless he or she was accompanied by a First
Officer under the age of 60.
[140]
Accordingly, depending upon the ages of the
available Captains and First Officers, there is potential gridlock that will
prevent pilots from performing their duties and will require costly remedial
measures. If these potential gridlocks were to materialize they would likely
constitute a legitimate BFOR defence.
[141]
Justice Mactavish in Kelly also found
that the age of pilots was a consideration for Air Canada’s BFOR defence based
on the scheduling problems which pilots over age 60 could cause due to the
application of the said ICAO rules. The failure of the Tribunal to properly
consider the issue resulted in her setting aside its decision and remitting it
for reconsideration. The Tribunal in the present matter did take the age factor
into consideration.
[142]
The complainants limited their submissions on
the first two steps of Meiorin before this Court to issues concerning
the employer’s failure to investigate alternative approaches to perform the job
that did not have a discriminatory effect. As I am in agreement with the Tribunal
that no undue hardship has been established by Air Canada, I do not need to
consider the issue at this time with respect to Court file T-1453-11.
[143]
However, with respect to Court file T-1463-11,
since I conclude that ACPA would suffer undue hardship due to the infringement
of its members’ employment rights, I will state my view that procedural
accommodation does not arise for the union to consider. Its members being in a
form of “chow line”, no alternatives for partial accommodation exist because
any attempted accommodation to favour one member of the union unavoidably
adversely impacts on another member.
[144]
Not having taken a clear position on the first
two steps of the Meiorin test, the Tribunal nevertheless proceeded to
consider the accommodation issue, and found no undue hardship caused to Air Canada by the abolishment of the MRP. Air Canada argued that the application of the ICAO
rules would cause a scheduling gridlock requiring it to hire and train new
staff and that this would represent undue hardship in the form of the extra
costs incurred in alleviating the problem.
[145]
The Tribunal accepted the premise of Air Canada’s argument but found that the contention was speculative because in practice the age
profile of Captains and First Officers was such that gridlock was unlikely to
occur.
[146]
The evidence before the Tribunal was different
from that in Kelly when that decision considered the BFOR issue. The
conclusions reported by the Court in that case depicted different scenarios for
gridlock occurring at lower percentages (see paragraphs 446 to 451 of the
decision). In addition, the issue of predicting future Captain-to-First Officer
ratios does not appear to have figured in the Tribunal’s decision.
[147]
For the present judicial review, Air Canada cited three areas in the Tribunal’s reasons where it contended that the CHRT had
either misapprehended the evidence or misdirected itself. The most significant
of these was the allegation that the Tribunal failed to understand the nature
of the evidence of Mr. Tarapasky, the Manager of Crew
Scheduling in charge of automation and process for flight operations,
which evidence Air Canada described in its Memorandum as follows:
Any
possible future changes to the demographics of Air Canada pilots, were the
mandatory retirement policy to be abolished, will necessarily be speculative:
the mandatory retirement policy eliminates the possibility of now knowing with
any certainty the actual demographics of its pilots if mandatory retirement
were to be eliminated.
[148]
In support of this submission, Air Canada points
out that in Kelly, the Court set aside the Tribunal’s decision on BFOR
because it had failed to understand that the “experiment did not require
consideration of the actual number of over-60 Captains and First Officers in
Vancouver at the time. The purpose of the experiment was to determine whether a
flight schedule could be produced if 10 percent of each group was potentially
restricted.”
[149]
I disagree with Air Canada’s submission and
moreover conclude that Kelly has no bearing in this case given the
significantly different nature of the evidence before the Tribunal in Kelly
and in this matter. The evidence in this matter does not demonstrate difficulty
in assigning flight schedules unless a high proportion of First Officers are
over 60 years of age.
[150]
In this matter, the Tribunal specifically
referred to the evidence of Mr. Tarapasky and in my view properly captured Air
Canada’s concern that application of the ICAO standard, by changing the age of
retirement to 65 could result in potential pairings of Captains and First
Officers over age 60 that would require costly remedial measures to avoid. I
cite for example paragraph 270 of the Tribunal’s decision:
270 Mr.
Tarapasky used the same methodology for the rest of the experiments. For June
2009 Vancouver B777 FOs [First Officers], the actual number of pilots was 46
CAs and 73 FOs. In this experiment, a No solution was only reached at the level
of 50 percent potentially restricted CAs and 30 percent potentially restricted
FOs.
[151]
In paragraphs 271 to 274 of its decision, the Tribunal
described similar instances from Mr. Tarapasky’s evidence, noting similar
combinations of Captains and First Officers where gridlock would result for
other types of airplanes at Vancouver, Toronto and Montréal.
[152]
In the conclusory portion of its decision, the Tribunal
pointed out the challenges to the assumptions underlying Mr. Tarapasky’s
conclusions, which were described as requiring that First Officers remain first
officers and that a high proportion of First Officers be over the age of 60.
[153]
I quote paragraphs 418 to 421 from the decision
as follows:
[418]
The evidence is that certain combinations of potentially restricted
captains and first officers reach a threshold whereby PBS cannot produce a
block solution. This is with respect to Air Canada’s wide-bodied aircraft. Mr.
Tarapasky’s assumption was that the pilots on these flights would be the most
senior and the oldest of the Air Canada pilot groups.
[419]
Professor Kesselman challenged the validity of this assumption on the basis
that in the absence of mandatory retirement, first officers would not
remain such and would progress to captain status. Further, in his opinion,
Mr. Tarapasky’s assumptions in his model that first officers would remain first
officers and a high proportion of first officers would be over the age of 60
are not necessarily valid, given the variant ages at hiring and
other factors affecting promotions.
[420]
Professor Kesselman said that the pilots at the top of the seniority in each
classification are not necessarily the ones that will be over 60. Seniority is
by years of service at Air Canada; pilots may be hired at different ages and
therefore there could be some individuals with higher seniority but lesser age
and vice versa. Rather, said Professor Kesselman, more first officers will tend
to be younger and over time without mandatory retirement, it would be easier to
satisfy the over/under rule.
[421]
Mr. Tarapasky's model requires much higher percentages of FOs being over age
60, in most cases 30%, 40% or even 50% before it runs into this
gridlock, where it can't do the matching, can't satisfy the over/under
rule.
[154]
In addition, the evidence of Professor
Kesselman, who was qualified as an expert in labour economics, played an
important role in the Tribunal’s decision. It is referred to at paragraphs 288
and 299 of the reasons, where Professor Kesselman is quoted as noting the lack
of empirical studies, such that one could not even “ballpark” the number of
pilots choosing to work beyond age 60, describing the estimates as
“guesstimates” at best, based upon indirect evidence.
[155]
Evidence from the transcript of the hearing
includes the Tribunal Chairperson questioning Mr. Tarapasky on this issue. He agrees
that the model percentages of Captains and First Officers may never be achieved
(Volume VII of Air Canada Application Record (T-1453-11), page 3428 [page 1900-1
of transcript excerpt]).
[156]
Similarly, Captain Duke, the General Manager of Crew Resources at Air Canada stated under cross-examination that “I
do not know what the actual impact [of removing mandatory retirement at age 60]
will be, because I do not know how many are actually going to stay.”
(Transcript Volume 10, p. 2337).
[157]
The Tribunal also reviewed the evidence of
witnesses Captain Duke and Mr. Rikk Salamat, an expert in collective agreement
analysis, on age distribution of First Officers, pointing out that very few
First Officers were at the highest seniority level, a factor I rely on in respect
of ACPA’s appeal.
[158]
Accordingly, I reject Air Canada’s submission that the Tribunal misconstrued Mr. Tarapasky’s experiments by assuming
“that pilots on wide-bodied aircraft flights would be the most senior and
oldest of the Air Canada pilot group”. The Tribunal’s conclusions pertained
only to the age of First Officers, and were made in reliance upon Professor
Kesselman’s evidence, which described and challenged the underlying assumption
of Mr. Tarapasky’s experiments.
[159]
Air Canada also criticized the Tribunal for its characterization
of Mr. Tarapasky’s conclusions as “a snapshot only for one month”. The Tribunal
specifically relied upon the evidence of Captain Duke, who acknowledged the
weakness of Mr. Tarapasky’s model based on one month’s data.
[160]
In conclusion, the Tribunal’s decision is
reasonable in its justification of the outcome and intelligible in its
explanation in concluding that Air Canada had not met the burden of proving
that it would suffer undue hardship due to the elimination of the age 60 retirement
rule and therefore could not rely upon the BFOR defence under section 15(1)(a)
of the CHRA.
[161]
Accordingly, Air Canada’s application is
dismissed.
D. ACPA’S
DEFENCE OF A BONA FIDE OCCUPATIONAL REQUIREMENT
(1) Introduction
[162]
The Tribunal also rejected ACPA’s BFOR defence,
concluding that it failed to meet the requirements of all three steps of the Meiorin
test. The Tribunal opined that there was no connection between age, performance
of the job as an airline pilot and membership in a union, such that ACPA had
failed to satisfy the first and second steps of the Meiorin test.
[163]
It also appears that had hardship been
established, the Tribunal would have rejected the BFOR defence on the basis
that hardship involving representation of union members is excluded by section
15(2) because it is not one of the three categories that Parliament enumerated in the provision, i.e. health, safety
and cost. The sanctity of the collective agreement would certainly not fall
within any of these categories. Even were this not the case, because of the
limited scope of section 15(2) delineated in the Kelly decision,
the only hardship evidence presented by ACPA was financial in nature.
[164]
Despite these hurdles to the union succeeding in
its application, the Tribunal stated that “it is nonetheless prudent” (para
347) to consider whether the union members suffered undue hardship as a result
of eliminating the MRP. While the basis for such prudence was not stated, it
was likely due to the comments of the Supreme Court in Renaud, recognizing that a union may be relieved
of a duty to accommodate on account of a discriminatory practice if the
accommodation would cause undue hardship to other union members.
[165]
For whatever reason subsumed by “prudence”, the Tribunal
analyzed the relative economic hardship of the financial consequences to Air Canada pilots caused by the elimination of the MRP. It based the analysis on a theoretical
extension of the average age of retirement to 63 based on the American
experience of extending the age of retirement to age 65. It concluded that
while “The choice is difficult” (para 401), on balance the adverse differential
impact on the pilots did not reach the level of undue hardship.
[166]
Had the Tribunal concluded that undue hardship
occurred, the union's BFOR defence would still have been rejected for its
failure to meet steps 1 and 2 of the Meiorin test. To that extent the Tribunal
was not placed in the unsatisfactory situation of finding against ACPA in the
face of a conclusion that eliminating the MRP would cause hardship to its
members. Indeed, handing down an award against the respondent despite it
establishing undue hardship would undermine the purposes of the CHRA by
imposing absolute liability when its actions were justified.
[167]
Unfortunately, I am not spared this dilemma
inasmuch as I conclude that removing the MRP would cause undue financial
hardship to union members. As discussed below, I find that the Tribunal failed
to consider obvious factors and conclusions arising from Mr. Salamat’s evidence
and applied wrong principles. This resulted in an unreasonable conclusion that
the airline’s younger pilots would not sustain undue hardship caused by the
adverse differential impact that would result from the elimination of the MRP
in the collective agreement.
[168]
Because the Supreme Court’s decision in Renaud
declared that liability should not be imposed on a union without allowing it to
justify its actions, I feel obliged to reconsider the legal obstacles which
would prevent ACPA from fully defending its actions.
[169]
This entails a reconsideration of sections
15(1)(a) and 15(2) of the CHRA and the applicability of Meiorin
in the case of a union. As a result of this analysis, I proceed to apply a
modified four-step Meiorin test to consider ACPA’s BFOR defence. The
modified test includes an extra step - referred to in Renaud - that
requires the Court to consider whether the importance of preventing
discrimination based on age prevents adoption of a lower standard despite the
undue hardship caused.
(2) CHRA Provisions Regarding BFOR
[170]
The complainants’ arguments based on statutory
provisions were twofold. Firstly, they contended that ACPA was not entitled to
advance a BFOR defence because unions were not included in section 15(1)(a).
Secondly, they argued that the nature of the union’s hardship, if established,
was excluded by section 15(2), which limited this defence to issues of health, safety
and costs. The relevant statutory provisions from the CHRA with my emphasis
added are reproduced as follows:
|
2. 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 or in respect of which a record suspension
has been ordered.
|
2. La présente loi a pour objet de compléter la législation
canadienne en donnant effet, dans le champ de compétence du Parlement du
Canada, au principe suivant : le droit de tous les individus, dans la
mesure compatible avec leurs devoirs et obligations au sein de la société, à
l’égalité des chances d’épanouissement et à la prise de mesures visant à la
satisfaction de leurs besoins, indépendamment des considérations fondées
sur la race, l’origine nationale ou ethnique, la couleur, la religion, l’âge,
le sexe, l’orientation sexuelle, l’état matrimonial, la situation de famille,
la déficience ou l’état de personne graciée.
|
|
7. It is a discriminatory
practice, directly or indirectly,
(a) to
refuse to employ or continue to employ any individual, or
(b) in
the course of employment, to differentiate adversely in relation to an
employee,
on a
prohibited ground of discrimination.
|
7. Constitue un acte
discriminatoire, s’il est fondé sur un motif de distinction illicite, le
fait, par des moyens directs ou indirects :
a) de refuser d’employer ou de
continuer d’employer un individu;
b) de le défavoriser en cours
d’emploi.
|
|
10. It is a discriminatory
practice for an employer, employee organization or employer
organization
(a) to
establish or pursue a policy or practice, or
(b) to
enter into an agreement affecting recruitment, referral, hiring, promotion,
training, apprenticeship, transfer or any other matter relating to employment
or prospective employment,
that deprives
or tends to deprive an individual or class of individuals of any employment
opportunities on a prohibited ground of discrimination.
|
10. Constitue un acte
discriminatoire, s’il est fondé sur un motif de distinction illicite et s’il
est susceptible d’annihiler les chances d’emploi ou d’avancement d’un
individu ou d’une catégorie d’individus, le fait, pour l’employeur,
l’association patronale ou l’organisation syndicale :
a) de fixer ou d’appliquer des
lignes de conduite;
b) de
conclure des ententes touchant le recrutement, les mises en rapport,
l’engagement, les promotions, la formation, l’apprentissage, les mutations ou
tout autre aspect d’un emploi présent ou éventuel.
|
|
15. (1) It is not a
discriminatory practice if
(a) any
refusal, exclusion, expulsion, suspension, limitation, specification or
preference in relation to any employment is established by an employer
to be based on a bona fide occupational
requirement;
[.
. .]
(d) the terms and conditions of any
pension fund or plan established by an employer, employee organization or
employer organization provide for the compulsory vesting or locking-in of
pension contributions at a fixed or determinable age in accordance with
sections 17 and 18 of the Pension Benefits Standards Act, 1985;
[.
. .]
(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
[.
. .]
|
15. (1) Ne constituent pas
des actes discriminatoires :
a) les refus, exclusions,
expulsions, suspensions, restrictions, conditions ou préférences de
l’employeur qui démontre qu’ils découlent d’exigences professionnelles
justifiées;
[. . .]
d) le fait que les conditions et
modalités d’une caisse ou d’un régime de retraite constitués par l’employeur,
l’organisation patronale ou l’organisation syndicale prévoient la
dévolution ou le blocage obligatoires des cotisations à des âges déterminés
ou déterminables conformément aux articles 17 et 18 de la Loi de 1985
sur les normes de prestation de pension;
[. . .]
f) le fait pour un employeur,
une organisation patronale ou une organisation syndicale d’accorder à
une employée un congé ou des avantages spéciaux liés à sa grossesse ou à son
accouchement, ou d’accorder à ses employés un congé ou des avantages spéciaux
leur permettant de prendre soin de leurs enfants;
[. . .]
|
|
(2) For
any practice mentioned in section (1)(a) to be considered to be
based on a bona fide occupational requirement
and for any practice mentioned in section (1)(g) to be considered to
have a bona fide justification, it must be
established that accommodation of the needs of an individual or a class of
individuals affected would impose undue hardship on the person who
would have to accommodate those needs, considering health, safety and cost.
|
(2) Les
faits prévus à l’alinéa (1)a) sont des exigences professionnelles
justifiées ou un motif justifiable, au sens de l’alinéa (1)g), s’il
est démontré que les mesures destinées à répondre aux besoins d’une personne
ou d’une catégorie de personnes visées constituent, pour la personne qui doit
les prendre, une contrainte excessive en matière de coûts, de santé et de
sécurité.
|
(3) Section 15(1)(a): Are Unions Entitled
to Advance a BFOR Defence?
[171]
Before the Tribunal, the complainant pilots
argued that a BFOR defence is not available to a union under section 15(1)(a)
because this provision makes no reference to an employee organization. They sought to apply the interpretive principle of expression unius est exclusio alterius, pointing out that sections 15(1)(d) and 15(1)(f) of the CHRA specifically
made the BFOR defence available to an employee organization as well as to an
employer. Having made specific reference to its inclusion in other provisions
in the same section, they argued, Parliament must have intended it to be absent
from section 15(1)(a). To soften the blow somewhat, the
complainants submitted that a union could provide
evidence to support an employer's BFOR defence, but not benefit from the provision
itself.
[172]
The Tribunal’s response to the complainants’
argument was threefold. Firstly, in support of ACPA, the Tribunal pointed out
that the complainants’ discrimination claim was based on section 10 of the CHRA.
It concluded that it made no sense either in terms of policy or logic that a
discriminatory practice could be committed under section 10 by an employer, an
employee organization or an employer organization and yet of those three, only
an employer could raise a BFOR defence. I am in agreement with this view.
[173]
Secondly, the Tribunal made reference to Renaud
for the proposition that the union must have the same right as an employer to
justify discrimination. Again, I agree with this premise to permit a union to raise
a BFOR defence, which I will discuss in more detail in respect of section 15
(2) below.
[174]
Thirdly, and this time in support of the
complainant’s submissions, the Tribunal concluded that
the reference in Meiorin only to the “employer” gave
weight to the argument excluding unions under section 15(1)(a).
If I agreed with the Tribunal that Meiorin could stand for the
proposition that the Supreme Court intended to deny the existence of a BFOR
defence for unions, I would conclude that such a direction would outweigh the
previous two considerations in favour of a liberal interpretation of section
15(1)(a).
[175]
However, I was not presented with any Supreme
Court jurisprudence considering whether and how a union might raise a section
15(1)(a) BFOR defence when its members suffer from undue hardship.
Indeed, a union was the applicant in Meiorin, so it is difficult to
comprehend how that case could stand for any limitation on a union’s hardship
claim.
[176]
Moreover, as I shall describe below, on the
basis of statements in Renaud that unions should not be denied the right
to justify their conduct by demonstrating undue hardship, I conclude that the
appropriate approach to the application of Meiorin to novel facts
involving a union requires a modification of its test to meet the circumstances
of unions facing hardship in accommodating discriminatory rules. Accordingly, I
reject the concept that Meiorin could somehow support a restrictive
interpretation of section 15(1)(a).
(4) Is Section 15(2) Limited to Its Enumerated
Hardship Factors?
[177]
In Kelly, Justice Mactavish
rejected the Tribunal’s interpretation that section 15(2) was not bound by the
three enumerated heads of hardship. She conceded that other factors such as the
impact on employee morale may be taken into consideration if of sufficient
gravity to have a demonstrable impact on the operations of an employer in a way
that relates to the three enumerated heads in section 15(2) of “health, safety
and cost”.
[178]
In the present matter, although not citing the Kelly
decision, the Tribunal confined section 15(2) to its three enumerated factors,
applying the same reasoning relating to the
interpretive doctrine of unius est exclusio alterius
and the requirement to narrowly construe exceptions limiting the scope of human
rights.
[179]
ACPA did not challenge the Tribunal’s BFOR in Kelly.
Nevertheless, Air Canada raised hardship issues in its application with respect
to the impact which eliminating the MRP would have on the morale of pilots and
their seniority rights. Accordingly, the court in Kelly was required to
consider the issue of relying on hardship factors not falling within health,
safety or costs, the three expressly contained in section 15(2) for the
employer.
[180]
The situation faced in this matter is novel. It
involves direct hardship on unions by undermining the central principles by
which they operate, in addition to an indirect form of surrogacy hardship
suffered by the union’s members on a real and personal basis. It raises a whole
host of considerations not of central importance to employers, but only to
unions, which may not have been contemplated by the drafters of the CHRA. Yet
they are nevertheless very relevant to discrimination, in particular inasmuch
as they raise concerns over the imposition of unjust results if the statute is
applied in a literal fashion. It is these concerns that largely inform my
interpretations of these provisions.
[181]
I find my situation different from those who
have ruled on this matter ahead of me. Previously, discussion of limiting the
scope of hardship factors was “theoretical” in the sense that no serious
hardship was advanced on hard facts. In this matter, there was detailed
evidence of an adverse differential impact on the younger pilots of Air Canada which would result from accommodating the complainants; an impact which I am
satisfied constitutes undue hardship when properly considered. This was not the
situation before the courts and tribunals that have previously been called upon
to interpret section 15(2).
[182]
For example, the Tribunal’s findings on adverse
differential impact were set out at paragraphs 139 and 140 of Vilven
Tribunal #2 as follows:
[139]
A delay in career progression would also mean a delay in salary increases. It
is not as ACPA stated, that the over 60 pilots would be taking money out of the
younger workers' pockets if the age 60 rule was removed. Rather, the
younger pilots would take longer to achieve the salary increases that they
desire.
[140]
There was no evidence that a delay in the career progression and salary
increases of younger pilots would cause a substantial interference with the
rights of these employees. [. . .]
[Emphasis added]
[183]
Relying on this evidence, the Court stated at
paragraphs 372 and 373 of its reasons in Kelly:
[372]
Insofar ACPA was concerned, the Tribunal examined the issue of hardship to the
union in light of the principles articulated by the Supreme Court in Central
Okanagan School District v. Renaud, 1992 CanLII 81 (SCC), [1992] 2 S.C.R.
970, [1992] S.C.J. No. 75. The Tribunal had particular regard for the
effect that accommodative measures would have had on other ACPA members.
[373]
The Tribunal found that there was no evidence to show that a delay in the
career progression and salary increases of younger pilots would cause
substantial interference with the rights of these employees: Tribunal
decision #2 at para.140. [. . .]
[Emphasis
added]
[184]
While there was no finding of an adverse
differential impact before in Kelly, its decision regarding section
15(2) cannot be distinguished on this basis. Its analysis was in relation to a
factor not relating to safety, health or costs, but regarding the impact on
employee morale by the elimination of the MRP.
[185]
Confronted with this Court’s recent legal
interpretation of section 15(2), I considered attempting to characterize all
aspects of my decision on the undue hardship suffered by the affected pilots in
ACPA as a matter of cost, pertaining substantially to reductions in the
affected union members’ salaries that would result from the elimination of the MRP.
However, I conclude that the issue is much too entwined with matters of
intangible benefits such as those attached to hours and routes, life-style
factors of access to financial returns to raise families, the unequal impact on
seniority rights in violation of the sanctity of the collective agreement, and
the infringement of employment rights and benefits. As a result, I have no
alternative but to confront the correctness of the Court’s interpretation of
section 15(2) in Kelly.
[186]
By the doctrine of comity that binds judges, I
am required either to apply the legal interpretation of section 15(2) in Kelly
or provide cogent reasons in support of my dissent. The doctrine was recently
summarized by Noël J.A. of the Federal Court of Appeal in Apotex Inc v
Allergan Inc, 2012 FCA 308, at paras 46-48. Justice Noël noted that:
[47]
In the Federal Court, the above passage has been referred to as authority for
the proposition that while the decisions rendered by colleagues are persuasive
and should be given considerable weight, a departure is authorized where a
judge is convinced that the prior decision is wrong and can advance cogent
reasons in support of this view (Dela Fuente v. Canada (Minister of
Citizenship and Immigration), 2005 FC 992 (CanLII), 2005 FC 992, para. 29; Stone
v. Canada (Attorney General), 2012 FC 81 (CanLII), 2012 FC 81, para. 12).
[187]
After carefully considering the issue and with
all due respect, I conclude that the hardship factors encompassed by section
15(2) are not limited to those specifically enumerated therein. My reasons to
support an expansive interpretation of the provision follow below.
(a)
Lack of Policy Rationale for the Imposition of Absolute Liability
[188]
As with the application of section 15(1)(a)
to unions, I can think of no logic or policy reason that could be advanced to
deny them the right to justify their actions on the basis of accommodation
causing undue hardship. Nor have I seen any policy grounds or logical
explanation why imposition of liability should occur without the right to
defend oneself.
[189]
By that I mean that I cannot conceive why a
court or legislature would place strict limitations on any evidence that might
be introduced to prove hardship. Discrimination can arise from unlimited
factual circumstances. Moreover, if discrimination concerns such varied
intangibles as dignity, self-esteem, vulnerability and attitudes of prejudice
and stereotypes, why limit the consideration of disadvantage to three fixed
categories of hardship?
[190]
The requirement to categorize evidence under one
of the three hardship factors adds another layer of complexity to an already
complex subject. For example, in this matter does “cost” include a delay in
receiving salary or benefits to members? Indeed, costs are something that
concern employers. For workers, salaries are the equivalent of the employers’
revenues (the loss of which would also not be a “cost” to an employer, unless
prepared to abandon a literal interpretation); costs for workers are what they
spend their salaries on. Are the life-style factors that often tend to be more
important than salaries to workers costs? Does hardship include the relative
needs of the pilots at different ages due to their family and other situations?
If yes, why not the similar impact on worker morale or the dissension in union
ranks caused by removing the mandatory age retirement rule? They are all
indirect factors relating to monetary issues.
[191]
In order to prove something in the context of
legal proceedings, all of the relevant and admissible evidence on the issue (in
this case the issue of the disadvantage caused to one party by accommodating
another) must be presented and considered. The Court’s ultimate task is to
assess the entirety of the evidence in accordance with understandable and
practical precepts and decide whether the person, or in this case the
collective of persons, would suffer undue hardship. Hardship, like fairness, is
an indefinable term that is a so multifaceted that it must be left to judges to
determine based on all the relevant evidence. That is because hardship is not
definable in terms of circumstances, but rather depends on circumstances for
its determination.
[192]
Moreover, in terms of purposes of the CHRA,
section 2 provides that successful complainants should “have their needs
accommodated, consistent with their duties and obligations as members of
society.” Reference to the duties and obligations of members of society cannot
just cut in one direction, i.e, in respect of the claimants only. Accordingly,
I would submit that this language contradicts any suggestion that the purpose
of the legislation is to impose accommodation requirements on persons that are
inconsistent with their duties and obligations as members of society, such as
would result from arbitrary limits on the scope of hardship factors.
(b) Avoidance of Absolute Liability
[193]
Despite the reference to Renaud, it does
not appear that the court in Kelly had been directed that the consequence
of denying a right to argue hardship would equate to the imposition of absolute
liability on a union. Similarly, the Tribunal in the present matter, while
referring to Renaud, did not appear to consider the explanation for
the Supreme Court’s conclusion that unions must be able to justify their
position because to do otherwise would impose absolute liability on them. Its
only reference to Renaud was to note that a “union must have the same
right as an employer to justify the discrimination” (para 340 of Adamson).
[194]
The rationale for asserting the union’s right to
justify the discriminatory standard is found at paragraph 32 of Justice La Forest’s reasons in Renaud:
….
Moreover, any person who discriminates is subject to the sanctions which the
Act provides. By definition (s.1) a union is a person. Accordingly,
a union which causes or contributes to the discriminatory effect incurs
liability. In order to avoid imposing absolute liability, a union
must have the same right as an employer to justify the
discrimination. In order to do so it must discharge its duty to
accommodate.
[Emphasis added]
[195]
I base my conclusion on the requirement for an
inclusive interpretation of section 15(2) to avoid imposing absolute liability.
This springs not simply from the sense of injustice that arises in imposing
liability when evidence would establish that undue hardship would result from
accommodation of the complainant. There is also a rule of statutory
construction that statutes ought not to be interpreted so as to impose absolute
liability unless there are express words to that effect.
[196]
This doctrine was stated by the House of Lords
in London Guarantee & Accident Co, Ltd v Northwestern Utilities, Ltd,
[1936] AC 108, [1935] JCJ No 2 (QL) at para 18 by the Master of
the Rolls, Lord Wright as follows:
18 In
Hammond v. St. Pancras Vestry (1874), L.R. 9 C.P.
316, where the Act imposed on the Vestry the duty of properly cleansing their
sewers, it was held that as these words were susceptible of meaning either that
an absolute duty was imposed or that the duty was only to exercise due and
reasonable care, the latter meaning was to be preferred, since the absolute
duty could not be held to be imposed save by clear words. That case was
followed in Stretton's Derby Brewery Co. v. Derby Corp.,
[1894] 1 Ch. 431.
[Emphasis added]
[197]
The Supreme Court has applied this doctrine to
regulatory penalties. See Lévis (City) v Tétreault; Lévis (City) v 2629-4470 Québec
inc, 2006 SCC 12, [2006] 1 S.C.R. 420 paras 13-19; R v Sault
Ste Marie (City), [1978] 2 S.C.R. 1299, [1978] SCJ No 59 (QL). In R v Desousa, [1992] 2
SCR 944, [1992] SCJ No 77 (QL) at para 21, Sopinka J. explained the rule as
follows:
It
is axiomatic that in criminal law there should be no responsibility without
personal fault. A fault requirement was asserted to be fundamental aspect of
our common law by this Court in R. v. City of Sault Ste. Marie, [1978] 2 S.C.R.
1299, and as a matter of constitutional law under s. 7 of the Charter in Re
B.C. Motor Vehicle Act, [1985] 2 S.C.R. 486. As a matter of statutory
interpretation, a provision should not be interpreted to lack any element of
personal fault unless the statutory language mandates such an interpretation in
clear and unambiguous terms.
[Emphasis
added]
[198]
If the rule of not imposing absolute liability
without unambiguous words to that effect has not yet been recognized as a maxim
of interpretation in civil matters, I would nonetheless think that only logical
and cogent policy grounds, such as may be found in workplace safety or products
liability law, could support the imposition of absolute liability on a party.
[199]
Consequently, on the one hand there appears to
be no policy ground advanced supporting the truncation of hardship factors; while
on the other, there is an opposing policy consideration that prevents the
imposition of absolute liability by a doctrine of statutory interpretation that
requires a clear statement of intent for that purpose.
(c) Expressio Unius Est Exclusio Alterius
[200]
The Court in Kelly relied upon the
interpretive doctrine of expressio unius est exclusio alterius to
support its conclusion that Meiorin dictates that hardship factors are
not entrenched, unless they are expressly included or excluded by statute. In
explaining her reliance upon this maxim, Justice Mactavish stated at paras
393-6 of her reasons in Kelly:
393 As
the Tribunal itself noted, the Supreme Court stated in Meiorin that the factors to be considered in
determining whether accommodation imposes undue hardship "are not
entrenched, unless they are
expressly included or excluded by statute": at para. 63,
emphasis added. In this case, Parliament has chosen to specifically identify
the matters that may be taken into account by the Tribunal in an accommodation
analysis: see Russel Zinn, The
Law of Human Rights in Canada: Practice and Procedure,
loose-leaf, (Aurora: Canada Law Book, 1996) at s. 14:60:2.
394 Moreover,
there are two different interpretative principles that were not addressed by
the Tribunal, both of which suggest that the factors identified in section
15(2) of the Canadian Human
Rights Act should be read as an exhaustive list. These are the
principle of expressio unius est
exclusio alterius; and the approach that is to be taken in
interpreting human rights statutes.
395 The
"expressio unius est
exclusio alterius" maxim refers to a general principle of
statutory interpretation which suggests that to express one thing is to exclude
another: see Ruth Sullivan, Sullivan
on the Construction of Statutes, 5th ed. (Markham: LexisNexis,
2008) at p. 244.
396 That
is, the failure of Parliament to mention a thing in a list will give rise to
the inference that it was deliberately excluded. As Professor Sullivan says,
"The force of the implication depends on the strength and
legitimacy of the expectation of express reference. The better the
reason for anticipating express reference to a thing, the more telling the
silence of the legislature": at p. 244.
[Emphasis
added]
[201]
I have two comments with respect to this
reasoning. Firstly, if the Meiorin decision requires a factor to be
“expressly excluded”, it cannot be excluded by implication. Excluding
something expressly can only be achieved by clear unambiguous words that
require no interpretive maxim to determine their meaning. Interpretive
doctrines are only needed where the words are ambiguous. This is acknowledged
in Sullivan’s text, because if recourse is required to a maxim of statutory
interpretation where “The force of the implication depends on…” the
meaning of the provision, the point cannot be considered to be expressly
stated. [Emphasis added]
[202]
Secondly, again with reference to the Sullivan
text whereby “the strength and legitimacy of the implication depends upon the expectation
of express references”, [Emphasis added] I would think that the expectation of naming
all of the hardship factors in a statutory provision should be very low.
Determining hardship is entirely circumstantial, depending upon the nature and
significance of the impugned standard and “the duties and obligations as
members of society” to accommodate complainants. Being entirely circumstantial,
the categories of hardship are never closed, as indicated by the Supreme Court
in Hydro-Québec v. Syndicat des employé-e-s de techniques professionnelles
et de bureau d'Hydro-Québec, section locale, 2000 (SCFP-FTQ), 2008 SCC 43,
[2008] 2 S.C.R. 561 at paragraph 12 of its decision:
[12]
… What is really required is not proof that it is impossible to integrate an
employee who does not meet a standard, but proof of undue hardship, which
can take as many forms as there are circumstances. This is clear from the
additional comments on undue hardship in Meiorin (at para. 63): […]
[Emphasis
added]
[203]
If the categories of hardship are never closed,
there should be no expectation that hardship factors be expressly listed
in a statute, and therefore the “expressio” rule should not apply.
(d) Exceptions to
Human Rights Legislation should be Narrowly
Construed
[204] The second
interpretive principle relied upon in Kelly was that exceptions and
defences to human rights legislation should be narrowly construed; ergo,
when only three hardship factors are referred to in legislation, the provision
should be narrowly construed and limited to those factors alone. The Court’s
reasoning is set out at paragraphs 399 to 401 as follows:
399
My conclusion that section 15(2) of the Canadian Human Rights Act should
be interpreted as limiting the factors to be taken into account in an
accommodation analysis to health, safety and cost is reinforced when the issue
is examined in light of the principles to be applied when interpreting human
rights legislation.
400
That is, while the quasi-constitutional rights conferred by human rights
legislation are to be broadly interpreted, this is not so with respect to the
defences provided in the human rights statute in question. Defences to the
exercise of those rights are to be interpreted narrowly: see Brossard
(Town) v. Québec (Commission des droits de la personne), [1988] 2 S.C.R.
279, [1988] S.C.J. No. 79 (QL) at para. 56, and Dickson at para. 17.
401
As Justice Sopinka observed in Zurich Insurance Co. v. Ontario (Human Rights
Commission) [1992] 2 S.C.R. 321, [1992] S.C.J. No. 63, human rights
legislation is often "...the final refuge of the disadvantaged and the
disenfranchised". He went on to observe that "As the last protection
of the most vulnerable members of society, exceptions to such legislation
should be narrowly construed ...": at para. 18.
[205]
I think there is a distinction to be made in
strictly interpreting a hardship factor and strictly interpreting hardship to
eliminate obvious relevant factors that cause hardship. I also would suggest
caution in reliance upon dicta from Supreme Court cases such as Zurich
Insurance to support an argument that hardship categories should be strictly
constrained. The opposite point was made in McGill University Health Centre
(Montreal General Hospital) v Syndicat des employés de l'Hôpital général de
Montréal, 2007 SCC 4, [2007] 1 S.C.R. 161. It made specific reference at para
15 to hardship factors that would apply to unions and their members, such as
the disruption of collective agreements and infringing their employment rights,
stating as follows:
15
The factors that will support a finding of undue hardship are not
entrenched and must be applied with common sense and flexibility (Meiorin
at para. 63; Commission scolaire régionale de Chambly v. Bergevin, 1994
CanLII 102 (SCC), [1994] 2 S.C.R. 525, at p. 546; and Central Alberta Dairy
Pool v. Alberta (Human Rights Commission), 1990 CanLII 76 (SCC), [1990] 2
S.C.R. 489, at pp. 520‑21). For example, the cost of the possible accommodation method, employee
morale and mobility, the interchangeability of facilities, and the prospect
of interference with other employees’ rights or of disruption of the
collective agreement may be taken into consideration. Since the right to
accommodation is not absolute, consideration of all relevant factors can lead
to the conclusion that the impact of the application of a prejudicial standard
is legitimate.
[Emphasis added]
[206]
Secondly, the circumstances of the
complainant, i.e. being “disadvantaged” and “disenfranchised”
(Zurich, at para 18) should not exclude consideration of relevant
hardship factors of the respondent. This does not mean however, that the
seriousness of the discrimination may not outweigh a hardship factor in
considering whether it is “undue”.
[207]
Moreover, not all
complainants are disadvantaged and disenfranchised, by which I understand to
mean that they are vulnerable or needy. If that was the criterion, this case
would not have reached first base. There is no sense of the complainants being
“disadvantaged” and “disenfranchised” when a case deals
with pilots earning over $200,000 a year with benefits including an
advantageous life-style founded on a compressed 8-day work month and the
ability to retire with generous pension benefits and ready opportunities for
work with other airlines, as demonstrated by Messrs. Vilven and Kelly.
[208]
On the other hand judicial notice can probably
be taken of the notorious fact that the younger generation tends to
disenfranchise itself by its failure to participate in the political process;
while in terms of financial position and general living standards, the
complainant pilots collectively are surely better off in comparison to those
who are subsidizing the continuation of their advantageous working conditions
and high economic returns.
(e) Parliamentary Intention
[209]
In Kelly, the Court took note at paragraph
397 of its reasons that “Parliament would thus have been well aware that
factors such as impact on employee morale and interference with the rights of
other employees had been identified as relevant considerations in an
accommodation analysis”, “well before the addition of subsection 15(2) to the CHRA
in 1998”. This was said to support a strong inference that “Parliament intended
the list set out in subsection 15(2) of the CHRA to be an exhaustive
one” (Kelly, para 398).
[210]
It is difficult to disagree with that position,
except to conclude that it is doubtful whether either the drafters or
Parliament contemplated hardship factors pertaining to unions or their members.
[211]
Nevertheless, it is arguable that the Supreme
Court in Meiorin appears to have gone out of its way to maintain its
position that recourse should be had to all relevant hardship factors despite
section 15(2) of the CHRA. Although the term “hardship” was not included in the
British Columbia statute under consideration, the Court mentioned the need
for a unified approach to justification by hardship, including making specific
reference to the amendment to section 15(2) of the CHRA, but without
mentioning the limitations on the scope of hardship factors:
52
Furthermore, some provinces have revised their human rights statutes so that
courts are now required to adopt a unified approach: see s. 24(2) of the
Ontario Human Rights Code, R.S.O. 1990, c. H.19; s. 12 of the Manitoba Human
Rights Code, S.M. 1987-88, c. 45, and, in a more limited sense, s. 7 of the
Yukon Human Rights Act, S.Y. 1987, c. 3. Most recently, the Canadian Human
Rights Act, R.S.C., 1985, c. H-6, was amended (S.C. 1998, c. 9, s. 10) so that
s. 15(2) of the Act now expressly provides that an otherwise
discriminatory practice will only constitute a BFOR if the employer establishes
that the needs of the individual or class of individuals cannot be accommodated
without imposing undue hardship.
53
Finally, judges of this Court have not infrequently written of the need to
adopt a simpler, more common-sense approach to determining when an
employer may be justified in applying a standard with discriminatory effects.
See Bhinder, supra, at pp. 567-68, per Dickson C.J. (dissenting); Central Alberta Dairy Pool, supra, at pp. 528-29, per Sopinka J.; Large, supra, at para. 56, per
L’Heureux-Dubé J. It is noteworthy that even Wilson J., writing for the
majority of this Court in Central Alberta Dairy Pool, supra, arguably
recognized that a form of accommodation -- the search for proportionate,
reasonable alternatives to a general rule -- had a certain place within the
BFOR analysis, then applicable only to cases of direct discrimination. See in
particular her references, at pp. 518-19, to Brossard, supra, and Saskatoon, supra.
[Emphasis
added]
[212]
While it may be supposition on my part, I think the
Court, out of concern that the hardship factors were expressed too narrowly in
the CHRA, may have directly addressed the proper approach to the
inclusion or exclusion of the categories of hardship, including with reference
to Renaud, when it stated at para 63 as follows:
.
. . Among the relevant factors are the financial cost of the possible method of
accommodation, the relative interchangeability of the workforce and facilities,
and the prospect of substantial interference with the rights of other
employees. See also Renaud, supra, at p. 984, per Sopinka J. The various
factors are not entrenched, except to the extent that they are expressly
included or excluded by statute. In all cases, as Cory J. noted in Chambly, supra, at p. 546, such considerations “should be applied with common sense and
flexibility in the context of the factual situation presented in each case”
[Emphasis added]
[213]
Although irrelevant to the British Columbia statute
that did not mention hardship, but where specific mention was made to section
15(2) of the CHRA without reference to its three limiting factors, the Supreme
Court proscribed the entrenchment (obviously referring to legislation)
of factors, unless expressly excluded by statute. In other words, the zeitgeist
of the Supreme Court’s thinking on this issue appears to be that all factors
not expressly excluded in section 15(2) should be included.
[214]
In addition to the grounds above, I refer to the
reasoning at paragraphs 376-380 which describe how the justification of a
claim, including a claim of fundamental rights, would be proscribed by a
literal interpretation of section 15(2).
[215]
For the reasons outlined above, I respectfully
disagree with the decision in Kelly and conclude that the categories of
hardship under section 15(2) are not limited to health, safety and costs.
(5) Modification of the Meiorin Test to Apply to
Unions
[216]
The Supreme Court in Renaud stressed that
a union which is subject to the same liability as the employer “must have the
same right as an employer to justify the discrimination. In order to do so it
must discharge its duty to accommodate” (para 32).
[217]
I consider these observations to be strong dicta
in support of a conclusion that ACPA must be afforded the same right as Air Canada to justify the discrimination in order to avoid being subjected to absolute
liability. I also take cognizance of ACPA’s submission that its joint liability
with Air Canada for the consequences of the impugned MRP should provide it with
an avenue to justify its conduct.
[218]
As already noted, Meiorin did not involve
a union as a respondent; the union in that case was the claimant advocating on
behalf of a member. I was not presented with any case that had considered Meiorin
in light of the principles in Renaud, i.e. how to provide a union with
an opportunity to justify its conduct in a BFOR case.
[219]
Recognizing that I am in somewhat uncharted
waters, I would think that a reasonable approach to reconciling the principles
of Meiorin and Renaud in a novel situation involving a union is
to suggest a test that would reconcile the purposes of the two decisions
to the extent possible. In my view, this would entail maintaining the Meiorin
test, but adapting each step of the test to accommodate the different
circumstances of a union as a function of representing its members.
[220]
On this basis, a hybrid BFOR test
incorporating the requirements of Meiorin and Renaud could be
advanced for the purpose of determining whether ACPA’s joint participation in
the discriminatory practice with the employer is justified. It would impose the
following requirements:
a.
The employer adopted the standard for a
purpose rationally connected to the performance of the job;
b.
the union adopted the particular standard
in an honest and good faith belief that it was in the collective best interests
of its membership;
c.
the standard is reasonably necessary to the
accomplishment of the legitimate work-related purposes of the union.
For a union to show that the standard is reasonably necessary, it must be
demonstrated that it cannot accommodate the individual members of the union
sharing the characteristics of the claimant without imposing undue hardship
on other members of the union; and
d.
the degree of hardship must be weighed against
the nature of the discrimination to ensure that the importance of promoting
freedom from the discriminatory conduct, in this case freedom from age
discrimination, can admit a lower standard.
(a) First Step – Piggyback on the
Employer
[221]
I am in agreement with the Tribunal’s view that
imposing on the union the first step of the Meiorin test - demonstrating
that it adopted the standard for a purpose rationally connected to the
performance of the job - is a non-starter. If not modified in some manner,
this stands in the way of any union’s ability to justify its participation in
an employer’s discriminatory practice.
[222]
It is not a union’s responsibility to propose
standards or qualifications for the employees’ performance. This falls within
the employer’s exclusive right to manage its affairs. Therefore, the only
manner by which a union can be provided with a right to justify its practice is
to piggyback on the employer’s responsibility to define the standards of
performance. In this regard, the Court adopts ACPA’s argument that the Meiorin
test should be modified to reflect that the workplace standards were agreed to
by the employer and the trade union jointly, making them jointly liable.
[223]
On this basis the union shares liability with
the employer. If no rational connection of the standard with the performance of
the job can be demonstrated, the union cannot be in a better position than the
employer. For similar reasons, it would be hard to imagine how elimination of
the standard would infringe the rights of other employees if it bore no
relationship to performance of their duties.
(b) Second Step – Good Faith of the Union
[224]
The second step in the hybrid test simply
reflects that the union must be acting in good faith in agreeing to adopt the
standard in its collective agreement. This step would normally be “a given”
considering the democratic nature of unions. All that would be required was to
demonstrate that the standard was adopted in the best collective interests of
its members.
(c) Third Step – Hardship
[225]
With respect to the third step of demonstrating
undue hardship, the point to recall is that section 15(2) is not limited to
factors of health, safety, and cost. All forms of disadvantage visited on the
comparator pilots and the union should be considered.
(d) Fourth Step - Weighing
Hardship and the Nature of the Discrimination
[226]
The principles described in Renaud
suggest a two-step process, as described at para 38 of the Supreme Court’s
reasons:
38
…. I agree with the submissions of the respondent union and C.L.C. that
the focus of the duty differs from that of the employer in that the
representative nature of a union must be considered. [1] The
primary concern with respect to the impact of accommodating measures is
not, as in the case of the employer, the expense to or disruption of the
business of the union but rather the effect on other employees. The
duty to accommodate should not substitute discrimination against other
employees for the discrimination suffered by the complainant. Any
significant interference with the rights of others will ordinarily justify the
union in refusing to consent to a measure which would have this effect.
Although the test of undue hardship applies to a union, it will often be met by
a showing of prejudice to other employees if proposed accommodating measures
are adopted. As I stated previously, [2] this test is grounded
on the reasonableness of the measures to remove discrimination which are taken
or proposed. Given the importance of promoting religious freedom in the
workplace, a lower standard cannot be defended.
[Emphasis
added, with added numbers in square brackets]
[227]
Any significant interference with the rights of
other members caused by an accommodation will ordinarily justify a union’s refusal
to consent to such a measure. However, the test remains the reasonableness of
the measures to remove discrimination which are taken or proposed. The
importance of preventing the form of discrimination involved may nevertheless
override the hardship caused to the other members of the union. It is a matter
of weighing the two considerations.
(6) Applying the Modified Meiorin Test to ACPA
(a) Step One – Rational Connection
and Attempts to Accommodate
[228]
In this case, the first step of the Meiorin
test remains unchanged from that applied to the employer: the employer must
have adopted the standard for a purpose rationally connected to the performance
of the job.
[229]
ACPA argued that the Tribunal found that Air
Canada had satisfied the first (and second) Meiorin step. I disagree. At
paragraph 402 of the Tribunal’s decision in Adamson, in reference to the
first two steps, it stated only that “This is not so much in dispute between
the parties as is their disagreement on step 3, the accommodation obligation.”
Thereafter, the Tribunal dealt at length with the hardship issue and rejected
Air Canada’s arguments, without determining whether the first two steps had
been satisfied.
[230]
In reviewing the submissions of the complainants
before the Tribunal and in the present proceedings, I find that they made no
distinction between steps 1 and 2 of the Meiorin test, inasmuch as their
principal argument was that procedurally Air Canada made no attempt to
accommodate the pilots, simply applying its blanket MRP without exception. In
other words, the issue raised by the complainants was not so much whether there
was a rational connection of the standard to the performance of the job as
whether its effects could be mitigated by some lesser form of accommodation
than eliminating the MRP from the collective agreement.
[231]
As I indicated during the hearing, I do not consider
the situation to be one where it was possible to accommodate the complainant pilots
without causing the airline and union members undue hardship.
[232]
In the course of the hearing, Air Canada presented evidence demonstrating that gridlock in scheduling would occur with
certain combinations of Captains and First Officers if the MRP were eliminated,
resulting in a costly requirement to hire new pilots while keeping the older
pilots on staff without work to do.
[233]
The evidence placed before the Tribunal demonstrated
that scheduling more than 2,800 pilots, with the challenges of seniority and
the vast number of flight options to cope with, is already a horrifically
complicated procedure. Redefining the inputs to satisfy the 70 complainants and
other pilots as they reach 60 years of age while still operating under the old
system is not a requirement that I would impose on Air Canada pending the outcome of the Tribunal’s decision.
[234]
Moreover, on each occasion that one allows a
pilot to work over the age of 60 it imposes hardship on the younger pilots,
with no means to recover their lost income and future lost working advantages.
The complainants, on the other hand, may recover damages for lost income
similar to that awarded in the Tribunal Vilven Damages decision.
[235]
This is clearly what I would describe as an
all-or-nothing accommodation situation. It presents no scope to allow
individual exceptions to the mandatory age of retirement requirement.
Accordingly, I am satisfied that Air Canada met the first step of the Meiorin
test, including any requirement to demonstrate that partial accommodation was
not possible.
(b) Step Two – Good Faith of ACPA
[236]
Despite accepting that the union acted honestly
and in good faith, the Tribunal found that it failed the second step of the Meiorin
test because the mandatory retirement requirement was not for the fulfillment
of a work-related purpose. I respectfully disagree. In my view, the gravamen
of the second step relates to the mental element or motive in adopting the
standard. Therefore, ACPA did not fail the second step of the Meiorin test
as it was clearly acting in good faith in adopting the MRP. The issue was
simply no longer relevant given the Tribunal’s conclusion on the first step.
[237]
With respect to the proposed modified second
step of the union Meiorin test, the evidence proves that the union
overwhelmingly adopted the standard in an honest and good faith belief that it
was in the collective best interests of its members. There is no dispute that
the system of seniority was fundamental to the larger array of benefits
provided to ACPA members by the collective agreement. As shall be seen, I
conclude that it is an ameliorative rule. Moreover, when adopted, the mandatory
retirement requirement was part of the international standards directing flight
operations.
(c) Step Three – Hardship
(i) The Evidence on Undue
Hardship to the Comparator Pilots
[238]
Mr. Salamat provided evidence upon which the
issue of ACPA’s hardship was determined. The basic methodology employed focused
on the salary losses the comparator pilots would incur over their careers due
to their advancement being delayed by the complainants continuing to occupy
positions, versus the total salary gains they would make after turning 60 by
working to the fictive retirement age of 63.
[239]
Because the advancement of Air Canada pilots is entirely determined by seniority, a system described by Mr. Salamat as
being similar to a “chow line”, everything about a pilot’s advancement and
therefore his or her income relative to other pilots can be determined based on
the hiring date. The only exception occurs when pilots leave before age 60,
which does not affect the comparability of the financial outcomes. The fact
that the age and seniority of Air Canada’s pilots was known provided the basis
upon which Mr. Salamat could accurately plot the relative earnings situation of
every pilot employed at the time the complainants retired until 35 years in the
future, based on the career of the youngest pilot at the date of the
application.
[240]
Mr. Salamat used Air Canada’s pilot seniority
list as it existed in 2009. He extended it into the future, removing pilots as
they turned 60 and then backfilling those positions in the order of seniority
of the remaining pilots. This entire process was repeated until all the
employees on the current list were retired.
[241]
He then applied the relevant pay scales, also
determined by the collective agreement, to the pilots in their varying
positions throughout their career. Using a discount rate of 3 percent on future
earnings, he calculated a lump sum aggregate total which was described as the net
present value of the potential career earnings (“the NPV”) of the comparator
pilots at age 60.
[242]
Mr. Salamat conducted the same analysis for
retirement years of 61 through 65. Ultimately, he settled on age 63 as the
average age of retirement based on pilot retirement data from the United States
after the age of retirement there was extended to 65 in 2007.
[243]
Using the average retirement age of 63, he
recalculated the NPV of the comparator pilots at age 63. The results of his
analysis were initially displayed in chart form in Exhibit 1 of his report. I
have set it out below for the purpose of better comprehending his evidence.
[244]
The numbers in parentheses at the bottom of the
chart indicate the average loss of NPV in total earnings for each of the seven
age groups of pilots up to age of 60. As the reader moves to the right along
the year axis, the average retirement date for each of the age groups is
encountered. At that point the chart provides an aggregate average loss of
salary for the age group caused by deferral of promotions to age 60. As salary
is earned after age 60, the plot line climbs above the zero NPV axis, ending at
the 63 year mark for the last member of the age group. At the top of each age
group the NPV of total earnings at age 63 is displayed.
[245]
Mr. Salamat thereafter re-grouped the pilots
into four categories, intended to reflect the manner in which the potential
benefits and risks were distributed among each pilot group after the
elimination of the mandatory retirement age. This means that he calculated the
maximum potential benefit and the maximum potential damage for each pilot and
based on the two extremes, rated the risk as nil, moderate, significant or no
benefit from the elimination of the MRP.
[246]
These results were charted in Exhibit 6 to his
report. It plots the loss of NPV of each pilot prior to turning 60 and the
increase of NPV gained by working to age 63, along a seniority axis. It is not
in a form that can be reproduced, as it relies on colours to distinguish the
four risk groupings of pilots, while the tiny individual points representing
the 2,957 pilots’ NPVs are interspersed throughout the chart.
[247]
His conclusions from the re-grouping of the
pilots are as follows: for Group 1, representing 14% of the pilots (generally
those with the highest seniority), there is no risk whatsoever; for Group 2,
representing 57% of the pilots, there is a moderate risk; for Group 3,
representing 28% of the pilots, there is a significant risk; and for Group 4,
representing only one percent of the pilots, they obtain no benefit, i.e. only
losses, from the elimination of the MRP.
[248]
The general conclusions of Mr. Salamat’s report
are as follows:
For
only 14% of the active Air Canada pilots does the elimination of mandatory
retirement present a clear financial advantage. Because the decrease in the
number of pilots retiring will translate into a delay in advancement
opportunities, the other 86% of pilots will have to work beyond 60 in order to
maintain the same potential earnings they would have if the mandatory
retirement age was left in place. Within that group, there are some for whom
the decrease in career opportunities is relatively small compared to the value
of being able to work additional years. However, approximately 29 percent of
pilots will have to work additional years for relatively little additional
career earnings, as the near-term impacts they face are relatively large.
(ii) The Tribunal’s Decision
on Hardship
[249]
To determine the relative hardship for the
pilots, the Tribunal applied what I would describe as a salary gains and losses
“balancing” methodology used by Mr. Salamat, by which he attempted to weigh the
hardship in terms of the potential loss of NPV suffered by complainants who could
not work beyond age 60 in comparison with the potential loss to a comparator
group of pilots if the MRP was eliminated. The comparator group comprised the
pilots under the age of 60 who had the option of retiring at age 60 or working
to age 63. The Tribunal’s conclusions were set out at paragraphs 398 to 401 of
its reasons in Adamson as follows:
398 The
question is what is the extent of the hardship that is acceptable in this case.
Of the total number of active pilots employed by Air Canada in January 2009, 99%
stand to benefit with positive earnings (if) they worked to retirement age 63.
There is no doubt that the more senior pilots would benefit to a greater
extent, the value depending on the relative seniority levels. The downside
for the less senior pilots is that their ability to choose to retire at age 60
would be negatively impacted in terms of their earnings.
399 The most significant impact would be on the
most junior pilots. They would suffer a significant loss of earnings at
retirement age 60 and at best would be in a no benefit position even if they
work to retirement age 63.
If
some degree of hardship is acceptable should it be the pilots whose ability to
continue working at Air Canada was cut off at age 60 no reason other than that
they attained age of 60? Or should it be the majority of pilots for whom
their ability to choose to retire at age 60 would now be more constrained but
who would still be in a positive earnings position if they worked to age 63?
This is not to underestimate the more serious impact on the younger pilots with
the least seniority, but this group of pilots constitutes only 1% of the total
Air Canada pilot cohort.
The choice is difficult. But in my
opinion, the impact of eliminating the age 60 retirement rule does not reach
the threshold of “undue” hardship. I have concluded therefore that ACPA has not
satisfied the third step of the Meiorin test.
[Emphasis
added]
[250]
I conclude that it is necessary to send this
matter back for a redetermination inasmuch as I have found that the novel
circumstances presented should give rise to modifications of some of the legal
principles on BFOR in order to avoid unjust results. I have also argued for a
broad interpretation of the hardship factors under section 15 (2). These changes
in legal principles affect how this case would have been prepared and argued. For
example, there are significant qualitative lifestyle issues, issues relating to
relative need of the opposing groups in the union, and issues about the impact
on the collective agreement and worker morale at stake, all of which are
germane to determining relative hardship. I also conclude that the data on
adverse differential income breathes a new life into the significance of the
sanctity of the collective agreements that augment and distribute benefits
equally to all members.
[251]
As well, I have incorporated the principles
expressed in Renaud into the Meiorin test. This recognizes the
unusual nature of a union’s hardship in terms of the effect of accommodation on
the rights of other members of the union, a factor not just limited to
economic considerations. There are significant employment rights of the
comparator pilots not considered by the Tribunal, which ought to have been
considered.
[252]
However, even aside from the changes to the Meiorin
test and the scope of hardship factors, I find that there are deficiencies in
the Tribunal’s decision from its failure to consider significant factors
arising from Mr. Salamat’s evidence that lead me to conclude that the decision does
not fall within the range of possible reasonable outcomes based on the law and
facts. I discuss these below along with some comments on employment rights
affected by the elimination of the age 60 retirement rule.
(iii) Errors in the Tribunal’s Decision
1. Reduced Salary During the Make-up Period
[253]
I conclude that the Tribunal mischaracterized
the comparator pilots’ loss, significantly understating the economic
disadvantage that would befall the “majority of pilots for whom their ability
to choose to retire at age 60 would now be more constrained but
who would still be in a positive earnings position if they worked to age
63.” [Emphasis added]
[254]
The error of this statement is that the Tribunal
ignores ACPA’s most important point: that the comparator pilots have to work at
effectively no salary, or for a reduced salary, for three years to “make up”
for the loss of income accrued to age 60. Mr. Salamat’s report points this out:
“The youngest pilots, however, will suffer significant losses should they
retire at 60 and enjoy only negligible benefits for three additional years
of work.”
[255]
The younger pilots would not accept that they
were in a “positive” aggregate earnings position when their NPV position at age
63 after three years of additional work places them little ahead of what they
would have earned at age 60 with an MRP in place, meaning that they worked
those three additional years at a highly reduced effective salary rate.
[256]
As an example, considering the median 40-44 age
group from Exhibit 1 of Mr. Salamat’s report, the cumulative NPV after working
three years to age 63 is $157,858 above what their NPV would have been at age
60, which is a negative NPV of $153,536 (all expressed
in discounted dollars). In effect, these pilots would earn approximately
$52,500 annually, rather than their $200,000 plus salary normally earned
at that age, to obtain their increased cumulative NPV. This is because they would
have to work for free (i.e. not financial benefit to themselves) until achieving
a positive NPV amount.
[257]
Obviously the returns for working the three
extra years fall even more precipitously for the age 25-29 group, who would be
earning an additional $1,087 in NPV at age 63 for their three years of extra
work, or $362 per year on the basis of an annual salary, to eliminate their
negative NPV at age 60 of $246,874. There is nothing “positive” about working
without pay for three additional years to end up in the same financial
situation that they would have been in if retiring at age 60 under the
mandatory retirement regime.
[258]
This does not even account for the fact that if
retiring at age 60, this age group would be entitled to approximately 2/3 of their
salary earnings in the form of pension benefits for those three years, which I
discuss below.
[259]
The final point, which is another aspect of the
“unfairness” of the deferred make-up deal, is that because all of the age
groups (except the age 55-59 group) start from negative NPV cumulative earnings
at age 60, if for some reason those pilots do not work the extra three years to
age 63, there will be a further shortfall in their career cumulative earnings.
2. Permanence of Lost Wages due to Deferral of
Salary until Age 60
[260]
The Tribunal failed to comprehend that working
for free or at some reduced salary to make up for the reduced MVP at age 60 is
an indication that the loss is permanent and cannot be made up. It is an
illusion to state that the pilots will be in a positive earnings position
relative to retiring at age 60.
[261]
As is plainly evident from Mr. Salamat’s Exhibit
1, all age groups sustain an annual salary reduction in the neighbourhood of
$7000 to $8000 annually due to the elimination of the MRP. The aggregate loss
of this income at age 60 for pilots under age 55 ranges from $70,000 to
$247,000, the latter being the loss for the youngest group of pilots retiring
in 31 to 35 year from 2010.
[262]
The Tribunal mischaracterized Mr. Salamat’s
conclusions on working an extra three years as a form of recouping losses such
that they were in a “no benefit” or a “positive earnings position if they
worked to age 63”. This fails to comprehend that loss of salary to age 60 is
permanent and cannot be made up in any fashion by working longer. The only way in
which this loss could be eliminated would be if they were paid at a higher rate
than under the collective agreement during the make-up period, which would then
constitute hardship to the employer.
[263]
Working longer to compensate for a debt on which
one cannot collect is not repayment. It is merely working longer to make up for
what somebody else has taken away. The debt is never repaid. Each pilot’s loss
of income to age 60 has to be understood therefore as an unrecoverable loss.
For that reason the complainants should understand that their gain would be at
the expense of the pilots that follow them.
[264]
Neither is there traction in the complainants’ argument
that the situation is no different for any worker who instead of retiring early
at age 55, decides to work to 60. This scenario gives rise to no loss of
income by other workers in the same workplace. As long as there is a benchmark retirement
age, no negative consequences ensue until the retirement age is exceeded. The
maximum NPV of a pilot at hire was calculated going forward on the basis of
working to age 60. It is only when that age is exceeded by others that the
losses will occur relative to what the pilot was entitled to earn when first
hired by Air Canada, as then agreed by the employer, the employees, and the
union.
[265]
The Tribunal therefore mischaracterized the
comparator pilots’ loss by concluding that “their ability to choose to retire
at age 60 would now be more constrained but [they] would still be in a positive
earnings position if they worked to age 63”. At age 60, the youngest pilots are
short approximately $250,000 which has been permanently taken from them as a
result of the elimination of the MRP. That is exactly what is shown from Mr.
Salamat’s Exhibit 1.
[266]
Moreover, reference to constraining retirement
simply throws anyone analyzing the problem off the trail by transforming a
financial loss issue into one about delayed life-style on retirement. This also
ignores the hardship impact from a lack of financial resources on the younger pilots
to meet pressing needs such as for housing or raising families. In addition,
this mischaracterization of the pilots’ loss turns it into a non-financial hardship
factor falling outside the narrow interpretation that the Tribunal was applying
to section 15(2).
3. The
Windfall Earnings of the Complainants
[267]
Another factor that the Tribunal failed to
consider in its adverse differential impact comparison was the windfall nature
of the complainants’ extra earnings from working past age 60. Instead it
focussed on the hardship of “the pilots whose ability to continue working at
Air Canada was cut off at age 60 for no reason other than that they attained
age 60.”
[268]
The Tribunal’s sympathetic characterization of
the complainants’ financial loss ignores the evidence that the complainants
would be obtaining windfall earnings and benefits at the other employees’
expense. No other employees, before or after them, except for Captains
also near retirement when the age 60 rule is eliminated, would receive a
similar prolongation of their salary and benefits at the level reached during
the most lucrative period of their piloting career.
[269]
Mr. Salamat described the windfall effects
provided to the complainants in the following terms:
Again
for the junior pilots the cost of retiring at 60 will be $222,000 and for the
most senior it will be nothing. So they, you know, it’s the same pattern that
we saw with the other ones. One of the, I mean one of the more interesting
things about this view of looking at the impact is that pretty much all pilots
start off down at the bottom here, as the first officers in the little plane,
and as people retire they move up and it’s really just incredible luck, you
know should the age change happen, and you happen to be at the top of the list,
because you would have had a career-long benefit of pilots having retired at 60
and moving up, and then at the very moment when you’re supposed to retire, you
get to stay. So you know really it’s quite a windfall if you happen to be
there.
(Tribunal
transcript, pages 2575-2576)
[270]
The purpose of the collective agreement was to
prevent such unequal distribution of the fruits of working for Air Canada amongst members of the union. Windfall gains, the undermining of the purpose of the
collective agreement, and impact on worker morale and harmony within the
workforce are not cost factors or otherwise falling within a narrow
interpretation of hardship factors under section 15(2). This may explain why
the respondents and the Tribunal failed to consider what I would describe as intangible
or equitable hardship issues arising from the elimination of the MRP.
4. The Infringement of the Comparator Pilots’
Employment Rights
[271]
Because the hardship factors were limited to
purely financial matters, the parties could not present arguments relating to
the infringement of basic employment rights, which also meant that the Tribunal
did not consider the impact of eliminating the age 60 rule from that
perspective. I believe that this omission also resulted from the Tribunal’s failure
to grasp that the comparator pilots would suffer an unrecoverable loss of
income by the elimination of the retirement rule. In any event, the Tribunal
did not consider whether, by normal employment law standards, sustaining a loss
of income during the pilots’ careers to age 60 would constitute an infringement
of their rights, even if the loss could be made up after age 60.
[272]
An employer who unilaterally reduces an
employee’s salary without reasonable cause will face a claim for constructive
dismissal. It would be possible in some work situations to insert employees at
higher paying positions to the same effect, but that would not be permitted
under Air Canada’s collective agreement.
[273]
In any event, that is not what is happening
here. The scenario described is that of the comparator pilots’ income being
reduced; in effect transferring their loss of income to the complainants
in the form of windfall gains. Employment law makes no exception for
transferring income from one group of employees to another.
[274]
It is not as though the comparator pilots have
contributed to this situation. Most of them (75%) voted, when the issue first
arose, to adhere to the terms of the collective agreement. It is clear why. As
shall be discussed when considering below whether the MRP is discriminatory,
the complainants at age 60 would have already benefited by their predecessor
pilots having adhered to the collective agreement and thereby allowing them to
enjoy the higher salary levels in the later period of their careers intended by
the collective agreement. They now propose to extend these late career high
salary benefits without having “paid into” the agreement, thereby imposing the additional
costs of these benefits on the pilots that follow them.
[275]
Also, as shall be seen below, the purpose of the
collective agreement is to distribute benefits equally, including on a
generational basis. By eliminating the MRP in the agreement this fundamental
foundation of the collective agreement has been overturned. This seriously
impacts the “sanctity of the collective agreement” issue which I discuss in the
context of whether prima facie discrimination is established. I find
that this issue fits more readily into the union’s argument that the MRP is not
discriminatory, as opposed to the union’s hardship. Nevertheless, as it is not
clear that my conclusion that the MRP is not discriminatory will be upheld, the
negative impact on ACPA’s collective agreements may also be considered a
hardship factor.
[276]
Nor is there any principle in employment law
that allows for a salary reduction based on a deferral of income that can be
made up by the employee later by working longer and earning the same or more
total income, even were this the case, which it is not.
[277]
This scenario is also unfair in that the younger
pilots are required to assume all the risk of a contingent “bargain” that the Tribunal
sees as amounting to the pilots arriving back at a positive earnings position
20 to 35 years in the future.
[278]
Common sense dictates that no pilot under 40
would contemplate accepting such an offer if given the choice. It is
unreasonable and perverse to suggest that it is fair for the younger pilots to
defer present income based on an opportunity that it may be recovered in the
far future.
[279]
In this regard, I find that the Tribunal also
erred in ignoring the cogent evidence, admittedly from previous related cases,
of the referendum on the removal of the MRP from the collective agreement. As
previously mentioned, removal was overwhelmingly rejected by 75% of the members.
[280]
The pilots’ decisive vote is highly probative
evidence attesting to the negative consequences in terms of the hardship that
the pilots consider will be imposed on them by eliminating mandatory
retirement. No one knows the impact of a change in the collective agreement better
than those directly affected by it.
[281]
Nor can this vote be considered the “tyranny of
the majority,” as was suggested by the Commission. The issue before the Tribunal
was one of comparative economic hardship. The pilots voted in a common sense
fashion to reject a proposal that would significantly change the status quo to
their detriment, impairing their employment rights to the benefit of the
complainants and older pilots who voted for the change because they would be
better off.
[282]
Accordingly, taken without colour of right, the
elimination of the MRP in the collective agreement would infringe the
comparator pilots’ employment rights as described in the Renaud
decision.
[283]
It also would undermine the founding principles
of the collective agreement, which should be understood as an ameliorative
scheme for ensuring equalization in the distribution of the fruits of labour
among the members of the union.
5. The Impact of Pensions on
Working to Age 63
[284]
The impact of pensions in reducing the net value
of earnings of pilots at age 63 from those recorded on Mr. Salamat’s NPV
calculations was raised by the Commission’s counsel during the hearing. The Tribunal
peremptorily rejected pensions as a factor affecting Mr. Salamat’s data on
Exhibit 1. It nevertheless went on to consider Mr. Salamat’s pension evidence
concerning the increase in the pilots’ average income during their best five
years – the baseline for calculating the increase in the value of their
pensions as a further advantage of working to age 63. This was misleading and
inaccurate due to its incompleteness.
[285]
I am at a loss as to why ACPA would lead
evidence supporting the complainants’ position which was inaccurate and
misleading, yet fail to introduce easily calculable evidence showing that the
financial benefits to the pilots at age 63 in Mr. Salamat’s Exhibit 1 were
significantly overstated because they were not calculated as the net of the
pension income that would have been earned during the additional three years
along with the cost of pension contributions not required to be paid during the
same period.
[286]
Not having entered into evidence the impact of
pension income not taken on the NPV projections at age 63, ACPA obviously did
not argue the issue. However, it was raised by the Commission during the
hearing along with an argument for its consideration. It was peremptorily
rejected by the Tribunal.
[287]
I conclude that the Tribunal made obvious errors
regarding pensions in its reasons: on the one hand considering misleading and
incomplete evidence favouring the complainants; and on the other rejecting any
evidence demonstrating that the NPV numbers were significantly overstated at
age 63.
- The Increase in the Value of Pensions at Age 63
[288]
In its decision, the Tribunal described Mr.
Salamat’s evidence on pensions at paragraphs 374 to 376 as follows:
374 Mr.
Salamat next produced a Table showing the impact on the average income for the
last five years of employment by retirement at age 60. He assumed that the
Pension was based on the average of the last five years of earnings. (Air Canada’s Pension is based on the best 60 consecutive months).
375 Mr.
Salamat explained that, whereas the previous scenarios illustrated what happens
during the employment phase, this Table is intended to show the impact on
Pensionable earnings if pilots choose to retire at age 60 when the retirement age
rises to age 61-65.
376 What
the Table ultimately concludes is that the average impact on a pilot's earnings
for Pension purposes is $3,762 or 3% less if they retire at age 60 when the
average retirement rises to age 61. If the average retirement age goes to 63
and the average pilot retires at 60 when the average age rises to 63, the
negative impact will be $13,900, or eight percent less for Pension purposes.
And so on for the years, 62, 64 and 65.
[Emphasis added]
[289]
The $13,900 [rounded to $14,000] “negative impact”
for “pension purposes”, is expressed as a disadvantage to the comparator
pilots if they continue to retire at age 60 instead of working to age 63. This
is factually inaccurate because evidence of a $14,000 increase in the best
five-year average income by working to age 63 has no relevance unless used as a
factor in the valuation of the pension which in turn is used to increase
the total advantage from working to age 63.
[290]
However, introducing evidence increasing the
valuation at age 63 would be even more misleading. Any gain in the value of
pensions by working three more years would normally be more than offset
by the decrease in economic position if one deducted pension benefits available
and contributions not made from age 60 to 63.
[291]
Therefore, if the Tribunal wished to refer to
evidence on the average increase to the pilots’ best five-year salary, it
should only have been for the purpose of questioning Mr. Salamat as to why he
was introducing this evidence without ACPA having used it to calculate the increase
in the pension value at age 63 in comparison with age 60, and more importantly,
why no deduction was being made for pension benefits not drawn on to age 63.
This assumes, that workers in the federal public service understand that the
value of their pension in terms of benefits received is based on the two factors
of their best five year income average and the number of years worked, which together
determine the percentage of that average salary that they will receive based on
a 2 percent rule for every year worked.
[292]
Pension administrators or actuaries could
produce the increase in the value of the Air Canada pension earned by working
to age 63 as they do all the time, for example in the context of family law.
They would use the $14,000 increase in the five-year average while also
utilizing the 6% increase in annual salary gained by working the three
additional years. The calculations would have to be generalized somewhat and
grouped into age cohorts, but Mr. Salamat’s evidence was already generalized in
this fashion.
[293]
The increase in the value of the pension would
have been considerably greater than the $14,000 increase in the best five-year
average income. However, the additional value of the pensions at age 63 would
be considerably less than the pension income available but not taken during the
three extra years worked plus pension contributions not paid over the same
period.
- The NPV at Age 63 Net of Pension Benefits
[294]
The Chairperson was aware from his decision in Vilven
Tribunal Damages, 2010 CHRT 27, that the true calculation of the
differential in wages when working an additional three years from age 60 to age
63 would have to be net of the pension amounts that the pilots would have been
paid during that same period.
[295]
In Vilven Tribunal Damages, he calculated
the lost salary incurred for Messrs. Vilven and Kelly from their retirement at
age 60 to their date of reinstatement as being net of pension benefits received
during the same period. His order at para. 174(7) stated as follows:
174(7)
The compensation for lost wages shall be net of the amounts of the Pension paid
to the complainants from September 1, 2009 to the date of their reinstatement.
[296]
There was a similar deduction for pension
contributions that the complainants would have had to pay had they remained on
staff after age 60.
[297]
Commission counsel posed a number of questions
to Mr. Salamat about his failure to deduct pension benefits for the three
additional years worked to age 63. There was also an exchange with the Tribunal
over whether it was prepared to consider the issue. Counsel’s purpose was to
demonstrate that the NPV numbers at age 63 of Mr. Salamat’s report were
inflated because they did not deduct pension amounts available but not taken by
pilots working beyond age 60. The transcript of these exchanges is as follows:
T-1453-11 application record of the Applicant Air Canada, Vol VII of
IX: transcript, page 2614, line 16 to page 2615, line 8:
Mr.
Poulin: Versus – but you did not take into account the amounts of money that
he’d be making you know, if he stayed out, the pension he would receive. And
thus that there would be a much smaller difference.
Mr.
Salamat: Yeah. No, the pension impacts on an individual are beyond
the scope of this analysis.
Mr.
Poulin: But the position of individuals is very important, as you showed in
your display when you pointed to that lone little red dot at the top left.
Mr.
Salamat: Mm’hmm.
Mr.
Poulin: The position of any individual at any point can have quite an impact
on any potential income or potential revenue or potential losses.
Mr.
Salamat: That’s correct.
[Emphasis
added]
T-1453-11 application record of the Applicant Air Canada, Vol VII of
IX: transcript, page 2616, line 24 to page 2619, line 15:
Mr.
Poulin: Yes, so that’s the – what I’m basically saying to you and to Mr.
Salamat, and I think he agrees with me, is that number in reality is quite
smaller since those pilots would have received a pension. There’s quite a
difference.
Mr.
Salamat: I – no, I couldn’t actually specifically agree to that. I question
that it would be quite smaller. I have no doubt that there is a pension
impact, but like I say, pensions is not my area of expertise. You know if
you work past the age when you’re supposed to retire, that means one thing. But
if you can continue to contribute that means another thing. And so, you know
this is not my area of expertise, and so this is why we didn’t really approach
it in the analysis.
Mr.
Poulin: Can we, can we agree that – sorry, I interrupted you. I have a bad
tendency. Can we agree then just to say that there is an actual impact
between the, what you’ve calculated as a potential benefit and what
would happen in reality?
Mr.
Salamat: Because of, because of the pension?
Mr.
Poulin: Because of a number of uncertain factors including pension – how much
money one would receive in pension, depending on the size of this pension –
Mr.
Salamat: Well yeah. I mean the two things that jump to mind are pensions and
taxes, right? So those two things alone will mean that $600,000 is almost
guaranteed to be not the amount the pilot would realize.
Mr.
Poulin: Okay.
Mr.
Salamat: So, but then again I hope I didn’t imply that this was taking taxes or
pensions into account.
Mr.
Poulin: Okay.
The
Chairperson: I think, Mr. Poulin, I think what you’re getting at is if you just
look at this model and it says if you work three years longer, if you’re in the
age group say – if you look in Exhibit 1 – 50 to 59 – your NPV as of this year
(inaudible) $470,965. And I take it that you’re saying is well I may not earn
that amount of money, because I’m going to look at how much I’m going to earn
if I were to retire, and I want to look at what my pension is going to be when
I retire. And so I may choose not to stay for three more years, because it’s
not worth it for me, because I’m not going to make $479,069, because I may make
that or I may make a pension that’s so close or relatively close that it’s not
worth it for me to work three more years, after you take away taxes and all
these other considerations. Is that your point that you’re trying to make?
Mr.
Poulin: Yeah, in the end that’s the point I’m basically saying, is you don’t
know. The problem is with Mr. Salamat’s model is you don’t know, and there’s
no way of knowing. What is the potential issues and potential benefit is so
– it’s so –
The
Chairperson: Well you can argue that. What I understand is Exhibit 1
then – if you’re in that age group and you work till age 63 instead of leaving at
60, given (inaudible) that income level, that’s your potential benefit.
Mr.
Poulin: True.
The
Chairperson: That amount. Period.
[Emphasis
added]
[298]
In his evidence Mr. Salamat agreed that
deducting the available pension benefits of the pilots that could have been
earned to age 63, if they instead retired at age 60, would have “quite an
impact on any potential income or potential revenue or potential losses” on his
charts. He later resiled somewhat from that position when he would not
“actually specifically agree” that in reality the amounts paid to pilots
receiving a pension would be quite a bit smaller. He had placed himself in a
difficult position in light of his earlier admission of failure to consider an
item that would have “quite an impact on any potential income or potential
revenue or potential losses”, which undermined his opinions. Ultimately he fell
back to a safer position that pensions were not his area of expertise, which he
stated was the reason why he did not consider pensions in his analysis.
[299]
Of more significance is the Chairperson’s
intervention on the point that demonstrates that he misapprehended Commission
counsel’s point, which was that the potential benefits working to age 63
described in Mr. Salamat’s report could not be determined without including the
impact of pensions. Instead, the Tribunal characterized counsel’s explanation
as whether it would be “worth it for me [pilots] to work three more years, after
you take away taxes and all these other considerations”. It is unreasonable
to rationalize a decision not to consider the impact of pension income earned
by reference to taxes imposed by the state. Pension revenue is earned by
pilots as part of their compensation for working. It is also paid to pilots as
a revenue stream similar to the salary earned. By omitting its consideration,
the Board fails to properly consider the compensation in earnings that the
pilots would have at age 63 by retiring or not retiring, which was the purpose
of Mr. Salamat’s evidence. If, pension income is not relevant why did Mr.
Salamat lead evidence on the best five years of pension amounts And why did the
Board refer to these results in its decision?
[300]
After not comprehending the point being made by
Commission counsel that the problem with Mr. Salamat’s model was that you
needed pension information to determine the benefits or costs of retiring at
age 60 versus at age 63, the Chairperson dismissed any suggestion that the
total earnings at age 63 should be netted out after deductions for pension benefits.
He stated in reply to counsel, “Well you can argue that”. Thereafter, he rendered
his decision on the point by referring to Mr. Salamat’s Exhibit 1 and stating
“that’s your potential benefit”, without any consideration of the impact of
pensions on Mr. Salamat’s calculations of NPV of total earnings at age 63.
[301]
The significant impact of pension benefits, in
particular for younger pilots, may be understood by considering the situation
of the youngest age group of pilots aged 25 to 29, as depicted in the evidence
before the Tribunal. While the same effect should apply to most pilots, I
illustrate this with the youngest age group because the correlation between
seniority and age is nearly perfect. By that I mean that the 25-29 age
group cannot contain members whose age and seniority do not correlate. For
example, the 30-34 age group may contain both pilots who joined Air Canada when
25 to 29 years old and have accumulated up to a decade of seniority as well as
pilots who have just joined aged 30 to 34 years old and have accumulated little
or no seniority.
[302]
Assuming the 27-year-old pilot represents the
average age of the 25-29 group, his or her pension at age 60 will be 2/3, or
66%, of the best consecutive 60 months [60 less 27 = 33 years times 2 equals
66% of income]. With the highest seniority possible amongst all the pilots at the
age of retirement, the pilot will have attained the rank of Captain with the
highest pay [by the terms of the collective agreement this must be the group of
most senior pilots on retirement] earning $200,000-plus assuming today’s salary
for these purposes. Accordingly, applying a deduction for pension benefits, the
total of three years’ earnings of $600,000 would be reduced by 66% or $400,000,
in addition to three years of unpaid pension contributions - approximately
$25,000, based on the pension contributions in the Vilven Tribunal Damages
decision).
[303]
Anyone who has a pension would know that no
increase in the value of the pilot’s pension based on a 6% increase in the
average income of $14,000 is going to approach anywhere close to the $425,000
deduction that should be applied to the NPV of these pilots at age 63,
particularly as the pilots’ pension is not even indexed.
[304]
Even pegging the increase in the value of the
pension at an exaggerated amount, the NPV of total earnings on Mr. Salamat’s
chart will be significantly overstated. This was the point that Mr. Salamat
originally acknowledged and the point which Commission counsel was trying to
make, that the true NPV of total earnings was greatly overstated because it did
not net out the pension income not drawn on by pilots who continued to work.
[305]
This means that besides working for free for
three years to achieve the same NPV at age 63 as at age 60, the youngest pilots
will still be in a significant deficit position at age 63, representing three
years of untaken pension, as opposed to breaking even, as shown in Mr.
Salamat’s numbers.
[306]
In my view, the Tribunal erred in refusing to
consider the evidence from Mr. Salamat’s questioning and Commission counsel’s
submissions demonstrating that Mr. Salamat’s NPV numbers were too high and
could not be relied upon for comparative purposes of working to age 63 versus
retiring at age 60 without pension information.
- Conclusion on Pension Evidence
[307]
In regard to pensions, the Tribunal’s errors are
twofold. Firstly, it should not have given any credence to the allegedly
positive increase of $14,000 in the 60-month average income used as the pension
baseline, as presented in Mr. Salamat’s evidence. This evidence was used to
weaken the comparator pilots’ position even though it was incomplete and served
no probative value unless integrated into the valuation of the pension at age
63, in which case a greater deduction would also have been required for the
pension benefits not taken from age 60 to age 63.
[308]
Secondly, the Tribunal misapprehended the
evidence and the purpose of the questioning of Mr. Salamat by Commission
counsel, which demonstrated that the true NPV of the comparator pilots at age
63 would be incomplete and misleading in demonstrating the economic advantage
of working to age 63 without considering the net amount after deducting for
untaken pension income.
[309]
That being said, however, I acknowledge that the
principal problem on this issue is that ACPA only advanced evidence not
helpful to its case, without providing the information necessary for the Tribunal
to properly consider pension factors in rendering its decision.
[310]
Nevertheless, the Commission’s questions were
sufficient to draw the Tribunal’s attention to the significant understatement
of the comparator pilots’ true economic situation at age 63, such that it was
compelled to react to the evidence in some form or fashion. Because it failed
to recognize the significant misstatement of the NPV calculations at age 63
without the inclusion of pension factors, it foreclosed its opportunity to
consider what procedure would best meet its mandate in the circumstances.
[311]
Two factors weigh heavily in the Court’s
decision in analyzing what the Tribunal should have done had it not
peremptorily dismissed the Commission’s evidence and submissions regarding
pension income. Firstly, the CHRT is an administrative Tribunal dealing with
issues of social equity involving the rights of citizens to be free of
discriminatory disadvantages that are unfair or objectionable, the
accommodation of which does not create undue hardship.
[312]
In circumstances where this Tribunal should have
concern about the insufficiency of evidence that would be determinative of, or
significantly affect the outcome, I conclude that it has a duty to intervene in
the fact determination process to encourage the parties to bring forward all of
the evidence necessary to allow it to properly decide the matter. In that respect,
I consider its role to be different from that of courts, which to a greater
extent must leave the parties to develop their evidence.
[313]
Secondly, given the intervention of Commission
counsel, which should have drawn the attention of the Tribunal to the
inadequacies of ACPA’s evidence for want of consideration of the impact of
pensions, the Tribunal erred in peremptorily stating that it was relying on Mr.
Salamat’s evidence on NPV values at age 63 to determine that the younger pilots
were in a positive economic position if working to age 63. Had the position of
the Commission been understood, any serious reflection on the issue would have
led the Tribunal to conclude that the evidence on NPV values at age 63 was
compromised for the purpose of demonstrating the relative economic positions of
age groups working to 63.
[314]
In my view, the appropriate action for the Tribunal
would have been to ask the parties whether they wished to introduce evidence on
the effects of pension income and contributions in order to provide the basis
for an accurate comparative analysis of the effect of eliminating the MRP and
working to age 63.
[315]
I say parties, also including the complainants.
Commission counsel may not have realized it, but his questions could only
undermine the reliability of Mr. Salamat’s NPV evidence at age 63. There is no
similar difficulty with pensions affecting Mr. Salamat’s evidence on the loss
of accumulated earnings up to age 60. His evidence accurately reflects
salary loss caused by the delay in pilot advancement dictated by the effects of
the collective agreement.
[316]
Accordingly, if the NPV data at age 63 is
compromised, ACPA has nevertheless made its case demonstrating the significant
loss to pilots at age 60 with the elimination of the MRP. The evidentiary
burden would then fall upon the complainants, who would be left without the
means to attempt to demonstrate that it could be recovered by the pilots
working longer.
(iv) Hardship Factors Other Than Costs
[317]
In light of previous interpretations of the Meiorin
decision which did not contemplate unions advancing hardship submissions as an
aspect of a BFOR defence in addition to the previously limited scope of
hardship factors under section 15(2) of the CHRA, it is understandable that the
non-economic factors that represent a disadvantage to younger pilots by the
elimination of the MRP were not raised.
[318]
As I am setting aside the decision and sending
it back to the Tribunal for a fresh determination based on the union-Meiorin
test described above, the parties may wish to lead evidence on the non-economic
factors such as the relative needs of the complainants and the comparator
pilots. This evidence appears to be relevant and significant when comparing the
hardship of the claimants and the comparator pilots.
[319]
ACPA may also lead evidence of the effects of
sustaining an annual salary loss in the order of $8,000 over 20 to 35 years as
a result of the elimination of the MRP in terms of hardship to the younger
pilots. The evidence on the value of the pensions of the retiring pilots is
also a factor in terms of the relative net worth of the groups of pilots.
[320]
The parties may also wish to lead evidence
concerning the types of expenditures which the different groups of pilots face
on average at their position on the career timelines, i.e. for younger pilots
raising families, educating their children, purchasing homes, other capital
expenditures and other family burdens of the “sandwich generation” compared
with the expenditures of the older pilots.
[321]
Reference has been made in earlier cases to the
issue of non-financial considerations such as those referenced at paragraph 220
of Vilven (2009 FC 367):
[220] That
said, the tribunal also accepted that when Messrs. Vilven and Kelly reached age
60 and had to retire from Air Canada, each experienced a blow to his
self-esteem. Both complainants had testified that they missed the prestige and
exciting work they had as Air Canada pilots. Mr. Kelly had also testified to
missing the friendships that he had formed at Air Canada.
[322]
The parties can lead evidence on other
non-financial considerations that result or not from the elimination of the MRP.
This would include the effects on employee morale and similar disruptive
factors to the union and employer. I would think that dissension in the ranks
of members and employees would be a significant hardship factor for all
concerned if it can be demonstrated by probative evidence.
[323]
In addition to the personal hardship suffered by
pilots, ACPA may discuss the impact of eliminating the MRP on the collective
agreement, the undermining of a fair and equitable distribution income and
benefits system and the infringement of standard employment rights on the
premise that one can reduce salary and somehow make up for it many years in the
future.
[324]
None of the foregoing is intended to limit the
initiatives of the parties to lead other non-financial evidence relating to the
hardship of the pilots or the union proper.
(d) Fourth Step - Whether a Lower Standard May be Defended
[325]
In recasting the hardship test for unions by
incorporating the principles of the Renaud decision, an issue arises
that was not considered by the Tribunal. The Supreme Court concluded that
despite a finding of hardship to other members of a union, a decision-maker
must nevertheless determine whether the importance of promoting the right that
was the subject of the discrimination was such that it prevented a lower
standard from being defended. Justice La Forest’s comments on this point at
paragraph 38 of Renaud are as follows:
As
I stated previously, this test [of undue hardship] is grounded on the
reasonableness of the measures to remove discrimination which are taken or
proposed. Given the importance of promoting religious freedom in the
workplace, a lower standard cannot be defended.
[326]
As this is a new issue, I will attempt to
provide some guidelines to the Tribunal for its consideration.
(i) A Nuanced Approach to Age Discrimination
[327]
The importance of promoting freedom from age
discrimination was an issue implicitly considered by the Supreme Court in McKinney . In upholding the constitutionality of the MRP in that case, the Court
pointed out the need for a nuanced and balanced approach towards age
discrimination in comparison with more emotionally charged forms of
discrimination. It explained the difference at paragraph 88 of McKinney:
.
. . Racial and religious discrimination and the like are generally based on
feelings of hostility or intolerance. On the other hand, as Professor Ely
has observed, "the facts that all of us once were young, and most expect
one day to be fairly old, should neutralize whatever suspicion we might
otherwise entertain respecting the multitude of laws . . . that comparatively
advantage those between, say, 21 and 65 vis-à-vis those who are younger or
older", Democracy and Distrust (1980), at p. 160. The truth
is that, while we must guard against laws having an unnecessary deleterious
impact on the aged based on inaccurate assumptions about the effects of age on
ability, there are often solid grounds for importing benefits on one age group
over another in the development of broad social schemes and in allocating
benefits. The careful manner in which the General Assembly Resolution on
the rights of the aged is framed is worth noting. Its recommendation
discouraging discriminatory practices in employment based exclusively on age is
prefaced by the words that this be done "wherever and whenever the overall
situation allows".
[Supreme
Court’s emphasis]
[328]
Justice Mactavish in Kelly, after a
careful review of McKinney and developments since its release, concluded
that many of the factors that informed its reasoning had less relevance in the
intervening years, thereby undermining its precedential value. Her decision was
reversed by the Federal Court of Appeal, which concluded that McKinney
continued to bind the lower courts. As a result, McKinney can be understood
as confirmation that the importance of protection against age discrimination
would not prevent a defence of hardship justifying a lower standard.
(ii) Adverse Differential Impact
[329]
The Supreme Court in Gosselin pointed out
at paragraphs 31 and 32 that age-based distinctions are a common way of
ordering our society and do not automatically evoke pre-existing disadvantage
suggesting discrimination and marginalization in the way that other enumerated
or analogous grounds might.
[330]
Justice Mactavish in Vilven, in the
finding that the MRP was discriminatory under section 15(1) of the Charter,
distinguished these conclusions in Gosselin by noting that they were
made in the context of an adverse differential effect in relation to younger
individuals. By this distinction, she implied that an adverse differential
impact would be a factor in concluding a rule was discriminatory. This would
support a conclusion that hardship factors may justify a lower standard of
promoting freedom from age discrimination where it relates to an adverse
differential impact suffered by the respondent.
[331]
I therefore, do not see any suggestion in the
foregoing passages from the McKinney, Gosselin or Vilven
decisions that the importance of promoting age discrimination would prevent a
lower standard in being justified by hardship.
(e) Conclusion and Directions on ACPA’s BFOR Defence
[332]
In conclusion, although entirely understandable,
I find that ACPA’s BFOR defence was premised on wrong principles such that no
proper hearing was held on the issue. First, the Tribunal applied a BFOR test
for employers, when the test should have been one that would permit ACPA to
advance defences justifying the mandatory retirement rule. Second, the case was
conducted on the basis that the hardship factors had to be limited to safety,
health and costs, thereby limiting hardship considerations to the financial
impact of extending the age of retirement of pilots. As a result of these two
errors in principle, the matter must be redetermined using an appropriate BFOR
test at a hearing at which the parties may lead evidence and advance arguments
on all factors of hardship not considered in these proceedings.
[333]
I deal below under the heading of discrimination
with what I describe as the notion of upholding the sanctity of the union’s
collective agreement rights based on the principle of equal treatment of
members. Infringement of the union’s fundamental principles, and thereby its
rights, could be considered a hardship factor under the broadened scope of section
15(2), if it turns out that it is rejected as a rationale to disprove substantive
discrimination occurred. Accordingly, undermining the basic collective
agreement rights of ACPA should also be considered a hardship factor, as was
indicated in Meiorin.
[334]
Based on all of the above, I conclude that the
decision on hardship is not justified or transparent and does not fall within
the range of possible reasonable outcomes. Accordingly, ACPA’s application is
granted and the decision of the Tribunal rejecting ACPA’s BFOR defence is set
aside. In returning the matter before the same panel, I provide the following
directions:
a.
The Tribunal is to apply the four-step hybrid Meiorin
test as described at para 220 above.
b.
The Tribunal shall not limit the evidence on
hardship to health, safety or costs, but shall consider any evidence of
disadvantage to the comparator pilots and the union caused by the elimination
of the MRP.
c.
In determining whether financial hardship is
occasioned to the comparator pilots, the Tribunal will give due consideration
to the areas of concern described above.
d.
The Tribunal is to provide the parties with an
opportunity to present evidence on the net economic effect resulting
from extending the age of retirement from age 60 to 63, which may include evidence
on the impact of pensions, but not taxes.
E.
Is the
Mandatory Retirement Provision Discriminatory?
(1)
Introduction
[335]
My initial analysis of this case was from the
perspective of hardship. Hardship was the focus of the previous related
decisions concerning mandatory retirement. In addition, Air Canada and ACPA limited their defences to proving that there was hardship or that the
normal age of retirement in the Canadian airline industry was 60. Renaud
also considered the disadvantages to union members caused by accommodation and
their rights under the rubric of hardship. It made no reference to separate and
distinct hardship for the union caused by the undermining of its collective
agreement.
[336]
Despite this background and the fact that this
matter is being remitted for a redetermination on hardship, I nevertheless have
concluded that my direction should include an option that ACPA (and Air Canada) may argue that the MRP in the collective agreement did not cause substantive
discrimination against the complainants.
[337]
Somewhat like the Tribunal chairperson herein
who thought it prudent to consider the hardship of the union, I think it
prudent for a number of reasons to direct the Tribunal to consider the
complaint in a two-stage process: firstly from the perspective of discrimination,
and thereafter hardship.
[338]
Firstly, I find that the evidence demonstrating
an adverse differential impact on the comparator pilots is applied for
different purposes to the union and the member pilots. For the union, its
primary concern is defending what I have called the sanctity of its collective
agreement rights. ACPA originally argued that the retirement rule was not
discriminatory as it provided for a “life-cycle” egalitarian treatment of its
members. Seen in this light, the mandatory retirement rule appears to be a
meritorious ameliorative rule intended to distribute the benefits of the
workplace, using age as a basis, in a fair and equal fashion.
[339]
The purpose of the rule is to ensure equal
favourable treatment of all union members over time. I conclude that this
purpose was not completely apparent prior to Mr. Salamat’s evidence that
demonstrated that by passing the burden of their gains to the pilots left
behind, the complainants would introduce unequal treatment of all employees.
ACPA’s argument, therefore, is at the level of the rule; that there is no
discrimination by the retirement rule because it serves a meritorious or
ameliorative purpose, which laws are supposed to promote, not prevent.
[340]
ACPA’s members, on the other hand, personally suffer
from the consequences of the adverse differential treatment resulting from
elimination of the MRP in their lowered incomes and delayed careers, plus other
non-financial disadvantages. They argue that not accommodating the
complainants’ rights is justified because to do so would cause them personal undue
financial and other personal hardship which varies with their personal
circumstances. They could also claim that their union rights have been
breached, but I do not see this as occurring at the level of the individual
member. I think that this issue arises at the collective level and falls under the
union’s responsibility as an extension of its mandate to negotiate and enforce the
mandatory retirement rule.
[341]
Because the Court in Renaud saw the
union’s hardship in terms of the disadvantage to its members, it categorized adverse
impact under hardship. I have already raised the point that unions could also
suffer hardship in the infringement of their basic collective rights, but I
think this is truly a matter related to discrimination, as perhaps is the
question of members’ rights under the collective agreement.
[342]
I see the factor of union rights benefiting
members by their equal treatment as an issue arising under the rubric of
discrimination, rather than justifying the discrimination by demonstrating
hardship and a failure to accommodate. The issue of a beneficial rule would
seem to arise when determining whether prima facie discrimination has
been proven, because it relates to the substantive nature of the deviation
in treatment of the complainants as meritorious or not. One should not have to
justify meritorious deviations. Only non-meritorious deviations in treatment
of complainants require justification by respondents relating to the burden of
accommodating the discrimination by the impact it has on respondents.
[343]
Because of my concern that the ameliorative
effect of a rule may not fit into the hardship tests, I consider it necessary
to direct the Tribunal to consider whether the retirement rule is ameliorative
in the nature of its distinction and whether a case of prima facie
discrimination exists.
[344]
Even if I am wrong in concluding that the issue
of the rule being meritorious can only be considered at the prima facie
determination stage, I would nevertheless direct the Tribunal to consider the
issue. That is because to find a solution to what appears to be an unjust
result, I have already travelled far into uncharted waters on the principles of
hardship for a trial level court, which usually navigates by the rules of other
courts. This includes modifying the Supreme Court’s Meiorin test so as
to allow unions to justify their actions and to avoid the application of
absolute liability. I have also upheld the Tribunal’s interpretation of section
15(1) and taken a broader interpretation of the scope of hardship factors under
section 15(2) despite the language of both sections and also contrary to what was
previously thought to apply for section 15(2) by a highly respected judge of
this Court. If I am proven incorrect in these conclusions, turning to the
definition of a discriminatory practice under the CHRA appears the only
alternative to avoid what appears to be an injustice to the younger pilots if
the MRP is struck down.
[345]
In addition, even though this issue was not
raised by the parties, I believe that I am correct in concluding that it is an
overriding error in principle to decide a matter on an incorrect
characterization of a fundamental issue before the Tribunal, regardless of
whether the issue was argued or not. Similarly, because there was no probative
evidence of a differential impact from the elimination of a mandatory
retirement rule in any of the previous CHRT and Federal Court decisions treating
this subject, I do not feel constrained by their reasoning or decisions.
[346]
Finally, I would not be so far from shore on any
of these tacks if I did not believe that the new evidence on adverse
differential impact is significant new evidence affecting the field of
mandatory retirement for two reasons. Firstly, I strongly suspect that
differential impacts will occur in all work places where good jobs are scarce
and there is a correlation in advancement and compensation with jobs opening
up. In other words, the negative effect of extending careers does not require
a perfect “chow line” workplace structure, such as is the case for Air Canada pilots. The uniqueness of their situation merely allows for a precise calculation of
the effects of extending careers. This brought to light an unknown negative
consequence of eliminating MRPs, affecting not just entry-level workers, but likely
all employees in an organization, which should have general application beyond
a workplace driven entirely by seniority.
[347]
Secondly, the adverse differential impact
evidence should call into question some of the conclusions from previous
jurisprudence on MRPs that were rendered without the benefit of the complete
evidence on adverse differential impact. Once it is understood that the issue
in mandatory retirement rules is not just making way for younger unemployed
workers, as was argued in McKinney, but also avoiding the imposition of
unfair differential treatment on all workers on the job who follow those who
would obtain windfall benefits by extending their careers, the past
jurisprudence needs to be reconsidered because its pronouncements were based on
incomplete facts.
[348]
Furthermore, these conclusions are significant
and may extend into other areas which are presently the subject of controversy
in our society. For example, I would think that this evidence and the Courts’
conclusions on MRPs are relevant to the debate of extending retirement ages to
cover pension shortfalls. The focus to date has largely been on concerns over
having sufficient funding to pay the pensions of the mass of Baby Boomers who
are in the process of retiring. The evidence in this case suggests that
consideration should be given to the impact of these initiatives on persons
already in the workforce, as well as those seeking to gain entry. By that I
mean that Mr. Salamat’s evidence suggests that all workers, and particularly
the younger generation, may be forced to subsidize the older generation by
extending the careers of older workers to pay their pensions.
[349]
As discussed below, I find that the McKinney
decision of the Supreme Court supports these initiatives to fund pensions by
describing any MRP as per se discriminatory and therefore in principle
not worthy of maintaining. This case provides the opportunity to revisit the characterization
of a mandatory retirement rule as discriminatory. Without such a reconsideration,
the significant momentum to extend workers’ careers in areas of scarce good
jobs, already generated by past court decisions, will continue unabated. In
light of Mr. Salamat’s evidence, I believe that it is arguable that eliminating
retirement rules in the area of scarce good jobs may cause greater hardship to younger
workers than gains made by extending older workers’ careers judged by equitable
measures of vulnerability and need.
[350]
By all present accounts the younger generation, in
the area of scarce good jobs at least, are facing more difficult economic
challenges than the older generation at the age of retirement. Moreover, the financial
insecurity of members of the younger generation may be having a profound impact
on the fundamental institutions of our society for example by delaying marriages
and the raising of families. I realize that these observations are perhaps too
controversial to be taken as proper judicial notice. But this is a realm that has
relied heavily in the past on judicial notice, some of which I find highly
controversial and indeed inapplicable to Canadian society in the 21st
century. One of the possible benefits of this decision may be to obtain some
hard evidence on generational inequalities in the Tribunal’s redetermination.
[351]
Accordingly, in addition to my directions, I
propose to consider some of the issues that I see arising for the Tribunal in
considering whether the MRP in the collective agreement could properly be
described as a discriminatory practice under the CHRA.
(2) Scope of “Employment Opportunities” in Section 10 of the CHRA
[352]
One proposition that I considered but rejected
was that the wording of section 10 of the CHRA may be interpreted to conclude
that an MRP in a collective agreement causes no deprivation of an “employment
opportunity”.
[353]
In considering this issue, I reject the
application of section 7 to establish prima facie discrimination. The
reference is only to “employees” which I take to mean that its application is
only to employers. To assist in analysis of this issue, section 10 of the Act
is repeated below:
|
10. It is a
discriminatory practice for an employer, employee organization or
employer organization
(a) to establish or pursue a policy or practice, or
(b) to enter into an agreement affecting recruitment,
referral, hiring, promotion, training, apprenticeship, transfer or any other
matter relating to employment or prospective employment,
that deprives
or tends to deprive an individual or class of individuals from any employment
opportunities on a prohibited ground of discrimination.
[Emphasis
added]
|
10. Constitue un acte discriminatoire, s’il est fondé sur
un motif de distinction illicite et s’il est susceptible d’annihiler les
chances d’emploi ou d’avancement d’un individu ou d’une catégorie
d’individus, le fait, pour l’employeur, l’association patronale ou
l’organisation syndicale :
a) de fixer ou d’appliquer des lignes de conduite;
b) de conclure des ententes touchant le recrutement, les
mises en rapport, l’engagement, les promotions, la formation,
l’apprentissage, les mutations ou tout autre aspect d’un emploi présent ou
éventuel.
[Je souligne]
|
[354]
It is arguable on the language of the provision that
the entering into of the agreement by ACPA and Air Canada was not for the
purpose of depriving union members of employment opportunities. Rather, it was intended
to entitle the union members, over the course of their careers, to enjoy generous
employment opportunities, in particular including significant income to pilots
in the latter years of their careers, along with the ability to retire with
excellent pensions and other benefits.
[355]
In considering the interpretation of the
employment opportunities resulting from the collective agreement, it seems
reasonable to submit that the Court should not adopt a timeframe perspective at
the end of the pilots’ careers. This results in the characterization of the
situation as one of deprivation going forward. A more realistic timeframe of
the pilots’ employment opportunities should be from the perspective when they
joined Air Canada. At that time employment opportunities were viewed as an entitlement
over a career with a definite and beneficial endpoint.
[356]
The difficulty with this line of reasoning is
that it depends upon establishing that the collective agreement including the
MRP is an ameliorative arrangement. The meaning of employment opportunities
includes employment. Therefore, a deprivation of these opportunities occurs at
age 60 on account of an age rule in the collective agreement. I do not think
the term “opportunity” can exclude from its scope a situation of the termination
of work because of an age rule.
[357]
Therefore, the success of this argument would
depend entirely on the characterization of the collective agreement as an
ameliorative set of rules intended to distribute benefits equally among workers
over a time period with a definite endpoint. It is only on this basis that the
concept of deprivation can be challenged and turned into an entitlement, and
with that the Court’s perspective in time changed from the end to the beginning
of the agreement.
[358]
If an ameliorative arrangement, then the explanation
is that there is no substantive discrimination. Basically by that
argument, a rule is not discriminatory if it is a meritorious deviation in
treatment of the complainants; in this case to treat all pilots the same and in
a mutually beneficial manner, including their retirement at a pre-determined
age. If that is the legal result and the basis for changing the time
perspective to see the “employment opportunity” as an entitlement as opposed to
a deprivation, I think it is better to call a spade a spade. By this I mean it
is preferable to explain the underlying generally applicable rule why it is not
discriminatory based on legal principles, rather than say on the words of the
statute that it is not a deprivation of an employment opportunity based on the
specific facts.
(3) Formalistic versus Substantive Prima Facie
Discrimination
[359]
The foregoing analysis goes to the heart of the issue
of discrimination: whether its meaning may be interpreted on a substantive
basis as opposed to a formalistic approach, which appears to have been the methodology
adopted by the Tribunal. Its brief reasons on discrimination are set out at
paragraphs 2 and 3 of its decision as follows:
[2]
To succeed in their complaints, the Complainants must establish a prima
facie case of discrimination and once having done so, the onus shifts to
the Respondents to establish a defence on a balance of probabilities.
[3]
Under the terms of the collective agreement and pension plan between Air Canada and ACPA, Air Canada pilots are required to retire on the first day of the month following
their 60th birthday. Amended Schedule A, Complainants Employment History
(January 4, 2010) provided by Air Canada, lists each complainant’s name; date
of birth; date of 60th birthday; and date of retirement. This shows that all of
the complainants were retired on the prescribed date. Their employment was
terminated solely because of their age. This is not disputed by the
Respondents. Accordingly, the Complainants have established a prima facie
case of discrimination.
[360]
The Tribunal’s decision adheres to the standard
practice of requiring an initial determination of prima facie
discrimination, after which arguments of justification may be considered. The
methodology employed is formalistic in defining discrimination. By that I mean
that it rests on the notion that any departure from identical treatment of
individuals on the basis of an enumerated or analogous ground violates equality
and therefore demands justification.
[361]
The Supreme Court has repeatedly rejected a
formalistic approach to defining discrimination for the purposes of applying
section 15(1) of the Charter. Instead numerous cases such as Law,
Kapp, and Withler have emphasized that the differential
treatment must be shown to discriminate in a substantive sense to be found
discriminatory for the purposes of section 15(1).
[362]
I use as an example the Supreme Court’s most
recent restatement of the section 15(1) Charter test in Quebec v A,
cited above, requiring substantive discrimination, made by Lebel J at paragraph
154. I choose to cite this passage because it also raises caveats about the
overuse of judicial notice, which is another consideration I discuss below.
154.
To resolve the third issue and thus determine whether the differential
treatment discriminates in a substantive sense and brings the purpose of s.
15 (1) into play, the court must undertake a full contextual inquiry
concerning the circumstances of the claimant’s claim. This inquiry must be
undertaken from the point of view of a reasonable person in circumstances
similar to those of the claimant who takes the relevant context into account.
Whereas the claimant must prove on a balance of probabilities that the impugned
provision discriminates in a substantive sense, the court can take judicial
notice of certain facts or matters but must be careful not to use judicial
notice to recognize social phenomena that may not truly exist.
[Emphasis
added]
[363]
There is some debate as to the extent to which a
formal Charter discrimination analysis should apply in the human rights
context. In this regard, it is necessary to consider the recent Supreme Court
decision of Moore v British Columbia (Education), 2012 SCC 61, [2012] 3
SCR 360 [Moore]. This was an appeal from the decision of the British
Columbia Court of Appeal (2010 BCCA 478), wherein a debate had ensued as to the
application of substantive discrimination Charter principles to
establish prima facie discrimination under section 8 of British
Columbia’s Human Rights Code, RSBC 1996, c 210.
[364]
The issue in Moore was circumscribing the
universe of people potentially entitled to equal treatment in relation to the
subject matter of the claim (special education) and to decide whether reference
should be made to other children with special education needs or to the general
education population as a matter of achieving substantive equality.
[365]
Rowles JA, whose dissent was upheld by the
Supreme Court, described the role of a Charter analysis in determining
discrimination under a human rights code as follows at paragraphs 111 and 112
of her reasons:
VI. Comparator
Analysis
[111]
A considerable amount of argument on this appeal concerns whether a formal
comparator group analysis is required, given that this claim is made under the
Code (as opposed to the Charter), and that it is an “accommodation” claim (as
opposed to a claim for identical treatment). In my opinion, nothing much turns
on this question.
[112]
The detailed comparator group analysis is a product of s. 15 Charter
jurisprudence. It began in Andrews, where McIntyre J. stated that equality “is
a comparative concept” (at 164), was first formalized by Law, and then
subsequently developed by cases such as Granovsky v. Canada (Minister of
Employment and Immigration), 2000 SCC 28 (CanLII), 2000 SCC 28, [2000] 1 S.C.R.
703 [Granovsky], and Hodge v. Canada (Minister of Human Resources Development),
2004 SCC 65 (CanLII), 2004 SCC 65, [2004] 3 S.C.R. 357 [Hodge]. Traditionally
then, a formal comparator group analysis was not done for statutory human
rights claims, and explicit mention of comparator groups remains conspicuously
absent. However, in Gibbs, which predated Law, Sopinka J. for the majority
engaged in a form of comparator group analysis in respect of discrimination
under Saskatchewan’s Human Rights Code, S.S. 1979, c. S-24.1. As well, Huddart
J.A. in the Teachers case held that “[r]easoning by analogy from the analysis
developed to consider alleged breaches of s. 15 of the Charter, implicit in the
establishment of prima facie discrimination [under the Code] are considerations
of the appropriate comparator ...” (para. 17).
[Emphasis
added]
[366]
At paragraph 89 of her reasons, Rowles JA noted
that the characterization of the service that best defined the comparator group
was public education generally because this “best accords with the purposes of
the Code, substantive equality and equality jurisprudence generally”.
[367]
Rowles JA’s decision on the appropriate
comparator group was unanimously upheld, with Abella J writing for the Supreme
Court. However, no comment on the issue of the application of substantive
discrimination accompanied Justice Abella’s reasons, which she summarized at
paragraph 30 of the Supreme Court’s decision as follows:
[30]
To define ‘special education’ as the service at issue also risks descending
into the kind of “separate but equal” approach which was majestically discarded
in Brown v. Board of Education of Topeka, 347 U.S. 483 (1954).
Comparing Jeffrey only with other special needs students would mean that the
District could cut all special needs programs and yet be immune from a claim of
discrimination. It is not a question of who else is or is not experiencing
similar barriers. This formalism was one of the potential dangers of
comparator groups identified in Withler v. Canada (Attorney General), 2011
SCC 12 (CanLII), [2011] 1 S.C.R. 396.
[368]
I do not interpret the Supreme Court’s decision
as rejecting the application of Charter principles or the requirement that
substantive equality constitute a component of prima facie
discrimination where appropriate; as stated by Rowles JA, applying the general
education population as the comparator best accorded with “substantive
equality”.
[369]
But that conclusion does not provide much help.
The more difficult question is whether substantive equality Charter
principles should be used to determine whether a prima facie situation
of discrimination arises for a rule said to serve an ameliorative purpose for
the equal distribution of benefits among workers. To properly answer this
question, I think, depends upon providing an explanation why it would be
inappropriate not to resort to Charter principles of substantive
discrimination to determine the content of prima facie discrimination in
the specific context in this matter.
[370]
In searching for an answer to support the
argument that prima facie discrimination under the CHRA should include a
substantive element, I turn to an article by Professor Donna Greschner, “Does
Law Advance the Cause of Equality?” (2001) 27 Queen’s LJ 299. This article was
cited with approval by the Supreme Court in Kapp to support its
reasoning to vary the focus of a discrimination analysis towards that of the
perpetuation of stereotypes and prejudice. I find the article useful in that
the author provides an explanation as to why, and therefore when, substantive
as opposed to formal equality should be required in a section 15(1) Charter
analysis:
4
In Law and other recent cases, the Court has stated that the goal of section 15
is to promote substantive equality. [. . .]
5
To begin, what does substantive equality mean? The term is used in contrast,
usually positive contrast, to formal equality. Hence, at a minimum, substantive
equality must mean something different, and better, than formal equality. Thus,
what does formal equality mean?
[…]
16
However, the two-step approach is formalistic. It rests on the notion that any
departure from identical treatment of people, on the basis of an enumerated or
analogous ground, violates equality and therefore demands justification. The
two-step approach contains only one basic question, which typically has a
fairly easy answer: does the impugned law create a distinction (or fail to draw
a distinction) on the basis of an enumerated or analogous ground? In almost
every case, protestations about contextualism notwithstanding, one can answer
that question by merely examining the words of the statute. [...]
[…]
18
In my view, labelling every distinction on enumerated or analogous grounds as a
violation of equality (the first step of the two-step approach) is inconsistent
with substantive equality itself. […]
19
The formalist approach to section 15 also cheapens rights talk. It
labels any deviation from identical treatment, even for the most meritorious
reasons, as an infringement upon constitutional rights. […]
[Emphasis
added]
[371]
Professor Greschner provides a double-barrelled
explanation for why a substantive discrimination analysis should be applied.
Firstly, it is illogical that a meritorious deviation be considered prima
facie discrimination. As already pointed out, laws are intended to prevent
injurious conduct. They should have no application to meritorious behaviour
resulting from an ameliorative rule. An ameliorative retirement rule would fit
the description of a meritorious deviation.
[372]
Secondly, discrimination is a highly opprobrious,
and therefore effective, label used to discourage a type of behaviour rejected
by our society. It should not be cheapened by overuse where common sense
suggests no discrimination has occurred. Like crying wolf too often, the term
will lose its punch and legitimacy as an effective deterrent to discriminatory
conduct.
[373]
An example of cheapening rights talk I think is
occurring in this case by telling the younger pilots that the older pilots
suffered “discrimination” by the retirement rule. One scoffs at the suggestion
that a rule which was intended to ensure that all pilots would be treated
equally, and which when eliminated results in windfall benefits to the complainants
at the expense of all of their colleagues, was based on a prima facie
stereotype of prejudice against older pilots. Equally it is hard to accept
that the rule perpetuated a disadvantage to well-off senior pilots of the Baby Boomer
generation because they were victims from among a disenfranchised and destitute
group that suffers age discrimination in the 21st century.
[374]
If on the one hand, an explanation exists why
substantive discrimination factors should be considered to find a prima
facie discriminatory practice under the CHRA, conversely I can think of no
policy ground or explanation why the courts should not resort to a substantive
definition of discrimination so as to avoid mislabelling a genuinely
meritorious deviation as discriminatory.
[375]
Moreover, logically, I cannot see how section
15(1)(c) of the CHRA can be labelled discriminatory for Charter
purposes based on permitting an impugned offending rule (a retirement
provision), while the retirement rule is also found to be discriminatory under
the Act, unless the same principles apply for both determinations. I assume
that discrimination for Charter purposes does not exist “in the air”,
i.e. where no discriminatory conduct can be demonstrated on the facts of the
case upon which a pronouncement on Charter discrimination is made. This
logically means that the retirement provision must be found to be
discriminatory under the Act and under the Charter on the same
principles. I believe that this was another point made in Meiorin which
was in part rationalized upon a principle that there should not be too extensive
a dissonance between human rights analysis and Charter analysis. See Meiorin
at paras 47 to 49.
[376]
There is a third reason that an ameliorative
rule should not be branded as prima facie discriminatory. It does not
occur in this situation, but arises in more challenging circumstances involving
incompatible fundamental rights. As was pointed out in McKinney, because rights
involving age in the workplace tend to be somewhat nuanced, they do not fall
into the category of those requiring more stringent protection. Other rights
relating for example, to race, religion or gender where historical patterns
existed of discriminatory stereotyping and prejudice require more vigilance
against intrusion by justification.
[377]
When incompatibility arises where these rights
clash, such as occurs for example when a gender based rule is in conflict with
a religious right, I think a strong case can be made that such issues should
not be determined on the principles of hardship. Rather the claimant should
demonstrate that he or she was the subject of substantive discrimination before
any issue of justification by hardship arises.
[378]
This conclusion is based on two premises.
Firstly, formalistic discrimination does not establish that discrimination has
in fact occurred. That is because substantive discrimination involves harmful
rules of distinction, which determination is not necessarily the formalistic
test. A human rights procedure that avoids consideration of the substantive
discrimination requirement and moves directly into the issue of justification
by hardship, eliminates or severely diminishes the opportunity to argue that
the rule is not discriminatory. This would be all the more so if hardship
factors were limited to safety, health and costs as per section 15(2) of the
CHRA. If the literal interpretation of this provision is sustained, arguments
based on upholding the values of competing rights such as those involving
gender for example, would be excluded from consideration.
[379]
Secondly, shifting the focus from the issue of
discrimination, to hardship usually entails matters of accommodation. This
issue places an ameliorative rule at a disadvantage by the generous hardship
accommodation principles that are designed to seek a middle ground of
compromise, which normally is to be encouraged. However, a pattern of
accommodation over a period of time, I think, will lead to the undermining of
the fundamental right by the process of compromise. None should have been made
in the first place, where no substantive discrimination occurred.
[380]
In a democratic society, some values must take
precedent over others. It serves no purpose therefore to avoid the hard choices
when they clash if the results are pernicious to the principles upon which our
society is founded. Avoidance of these decisions cannot occur where the prima facie
discrimination is required to be substantive.
[381]
With this background, I do not understand that
discrimination or a “discriminatory practice” as defined in the CHRA is in fact
discriminatory if the test applied is merely formalistic and does not amount to
substantive discrimination, thereby allowing for the application of appropriate
Charter principles for the purpose of the Court’s analysis.
[382]
On that basis, it would be an error to interpret
the CHRA as allowing a claimant to establish a discriminatory practice
involving a mandatory retirement rule where on the legal onus the
claimant has not established substantive discrimination.
[383]
That is not to say that the Tribunal was
incorrect in concluding that the complainants had made out a case of prima
facie discrimination, and that the onus shifted to Air Canada and ACPA to demonstrate that their conduct was not discriminatory. Establishing a
first stage formalistic case of discrimination shifts the evidentiary onus to
the respondents who, unless they establish by evidence the ameliorative effect
of the rule, are left with the prima facie finding of discrimination.
That is what happened here when no challenge was made to the finding of prima
facie discrimination.
[384]
My point however, is that it would appear that
the respondents could have met that evidentiary onus by proving, on a “balance
of probabilities” onus in their favour, that the complainants were not
the subject of substantive discrimination in being required to retire at the
age of 60. Only after that issue was determined against them would they have
had the further onus of establishing justification using hardship principles,
on which issue the legal onus would remain with them throughout.
(4) Mandatory Retirement as an Essential Component of a Larger
Ameliorative Benefits Scheme
(a) Differing
Factors to Determine Discrimination
[385]
The Supreme Court has pointed out that where a
law has an ameliorative effect, affecting various groups, the impact on others
must be considered in determining whether the distinction in treatment
perpetuates a prejudice or stereotype.
[386]
In Withler, the Supreme Court
noted the need to consider different factors depending upon the circumstances
of the alleged discrimination. These included considerations such as whether
the impugned law was part of a larger benefits scheme, the ameliorative effect
of the law on others and the multiplicity of interests it attempted to balance,
which the Court indicated should colour the discrimination analysis. The point
is first made at paragraph 38 of the Court’s reasons and amplified upon at
paragraph 67 as follows:
[38]
Without attempting to limit the factors that may be useful in assessing a
claim of discrimination, it can be said that where the discriminatory effect is
said to be the perpetuation of disadvantage or prejudice, evidence that goes to
establishing a claimant’s historical position of disadvantage or to
demonstrating existing prejudice against the claimant group, as well as the
nature of the interest that is affected, will be considered. Where the claim is
that a law is based on stereotyped views of the claimant group, the issue will
be whether there is correspondence with the claimants’ actual characteristics
or circumstances. Where the impugned law is part of a larger benefits
scheme, as it is here, the ameliorative effect of the law on others and the
multiplicity of interests it attempts to balance will also colour the
discrimination analysis.
[…]
[67]
In cases involving a pension benefits program such as this case, the
contextual inquiry at the second step of the s. 15(1) analysis will typically
focus on the purpose of the provision that is alleged to discriminate, viewed
in the broader context of the scheme as a whole. Whom did the legislature
intend to benefit and why? In determining whether the distinction
perpetuates prejudice or stereotypes a particular group, the court will take
into account the fact that such programs are designed to benefit a number of
different groups and necessarily draw lines on factors like age. It will ask
whether the lines drawn are generally appropriate, having regard to the
circumstances of the persons impacted and the objects of the scheme.
Perfect correspondence between a benefit program and the actual needs and
circumstances of the claimant group is not required. Allocation of resources
and particular policy goals that the legislature may be seeking to achieve may
also be considered.
[Emphasis
added]
(b) The Ameliorative Purpose of the Mandatory Retirement
Provision in the Collective Agreement
[387]
In the Vilven litigation, the Tribunal in
its first decision (2007 CHRT 36, Vilven Tribunal #1) discussed the
nature and purpose of mandatory retirement policies in Air Canada’s collective
agreements, in various paragraphs commencing at 99 and ending at 110. Those of
relevance are as follows:
[99]
Mandatory retirement policies are usually in place where the employees have
considerable bargaining power, most commonly through trade union
representation. Indeed the overwhelming majority of mandatory retirement
policies are found in unionized workplaces. Labour economists Jonathan
Kesselman and Lorne Carmichael, testifying on behalf of the Commission and Air
Canada respectively, agreed that jobs in unionized workplaces are considered
to be the "good jobs", that is, jobs that pay well, have a high
degree of security, operate with a strong seniority system and have good
pension plans.
[100]
In the present case, ACPA and Air Canada agreed to retirement at age 60 in
exchange for the rich compensation package, including a pension plan that
put Air Canada pilots in an elite group of pensioners. Mr. Harlan Clarke,
Manager Labour Relations Flight Operations at Air Canada, identified an
important characteristic of a mandatory retirement policy, namely, that
employees, including Air Canada pilots, are not faced with the indignity of
retiring because they have been found to be incapable of performing the
requirements of their position or because of failing health. Rather, retirement
at age 60 for pilots is the fully understood and anticipated conclusion of a
prestigious and financially rewarding career.
[101]
The complainants testified that they were fully aware, when they began their
employment with Air Canada, that they would be required to retire at age 60.
They testified that becoming a pilot with Air Canada was every pilot's goal;
the pay was excellent, the work was interesting and there was significant
prestige associated with the position. However, they also knew that this would
not last indefinitely and that all pilots at Air Canada were required to retire
at age 60.
[…]
[106]
According to Professor Carmichael, the complainants, throughout their careers
at Air Canada, reaped the benefit of the mandatory retirement rule that
their union had negotiated on their behalf. As a result of the departure of
60-year old pilots from Air Canada, the complainants were able to progress
through their careers at a more rapid pace.
[107]
In addition, the pilots' status, income, the base from which they flew, the
choice of schedules and the pension plan benefits they received, among other
things, were negotiated on the basis of the mandatory retirement provision.
Having reaped the benefit of mandatory retirement, it should not be perceived
as unfair to require the complainants to ultimately bear the burden of
that policy.
[108]
The complainants may be unhappy about ending their rewarding careers as pilots
with Air Canada. But that situation cannot be viewed in isolation. It must be
seen in the context of a system that was designed to assign the
responsibilities and benefits of being an Air Canada pilot over different
stages in the pilots' careers. All pilots in Air Canada understand that they
will share these benefits and burdens equally at the appropriate stages in
their careers.
[109]
The denial of the right to challenge the final stage of that system -
retirement at age 60 - as a result of s. 15(1)(c) does not communicate the
message that the complainants are not valued as members of society, nor does it
necessarily marginalize them. It simply reflects the view that it is not unfair
to require the complainants to assume their final responsibility as Air Canada pilots. This message cannot reasonably be viewed as an affront to their dignity.
[110]
For these reasons, the Tribunal has concluded that the complainants' right to
equality under s. 15 of the Charter has not been violated by s. 15(1)(c)
of the CHRA. […]
[Emphasis
added]
[388]
The Tribunal’s findings, which were not
challenged, clearly point to the collective benefits achieved by ACPA’s members
by the terms of the collective agreement. These included generous pension and
other benefits on retirement.
[389]
Thus, considering the factors described above in
Withler, it is apparent that the Air Canada MRP is part of a larger
benefit scheme designed to meet the best collective interests of all of the
union’s members. The collective agreement and the included pension scheme can
therefore be seen to have an ameliorative effect on the members and to attempt
to serve and balance a multiplicity of interests, including the varying needs
of the younger and older workers.
[390]
It will accordingly be necessary to draw lines
on factors like age. The question for the Court is whether the lines drawn are
generally appropriate, having regard to the circumstances of the persons
affected and the objects of the scheme.
[391]
It is implicit that one of the purposes of the
pension scheme was to allow pilots to progress through their careers more
quickly. This facilitated the rapid accrual of salary and benefits while
allowing for an early retirement so as to avoid having to work longer to
achieve the same financial results, as demonstrated by Mr. Salamat’s data.
[392]
Accordingly, by extending the age of retirement,
the very purpose of the collective agreement would appear to be thwarted. For
instance, the younger pilots would have to work for three additional years,
effectively at no salary, to maintain the same total earnings as the mandatory
retirement provision would have leveraged for them at age 60.
[393]
Seen in this context, I believe that a strong
argument can be made that the distinction said to act against the complainants’
interest should be seen not as perpetuating a stereotype or prejudice. Quite
the opposite, it would seem that the purpose of the age 60 mandatory retirement
provision was to benefit the complainants by enabling them to retire early
having enjoyed generous working conditions, salary and benefits, with those
salary and benefits continuing in the form of pension benefits after
retirement.
[394]
Moreover, it was certainly not the intention of
the union to make the working conditions so exciting, interesting, or fun that
the pilots would regret the higher salary and benefits obtained over their
career, if this is the basis for their loss of esteem as was the evidence in Vilven
Tribunal #1.
(c) The Vilven Tribunal’s Conclusion of “No Unfairness”
[395]
Even though no adverse differential impact had
been demonstrated, the Tribunal in Vilven Tribunal #1 concluded that the
outcome of the complainants being required to adhere to the MRP provision would
“not be unfair”. This appeared to refer to the fact that the complainants had
taken the benefit of their careers advancing more quickly because other pilots
before them had retired at age 60 in accordance with the collective agreement.
However the Tribunal stopped there and made no mention of the windfall benefits
which would flow to the complainants by being the first not to adhere to the
collective agreement, as this point only came out through Mr. Salamat’s
evidence.
[396]
As seen above, the Tribunal used the term
“unfair” to describe the result of a party taking the benefit of other persons’
adhesion to an agreement, and when the party’s term under the agreement was
completed, changing or ignoring the terms to suit its purposes. Under normal
circumstances our sense of fairness does not appreciate agreeing to something
to gain benefits and then scrapping the agreement once the benefits are
obtained so as to satisfy our self-interest. It is the sense of not accepting
the sanctity of the agreement by which the complainants benefited that
led the Tribunal to conclude in Vilven Tribunal #1 that sticking to the
agreement was not unfair, in addition to not offending the complainants’
dignity since termination was not unexpected.
[397]
However, because there was no evidence of an
adverse impact on the comparator pilots in Vilven Tribunal #1, the
mandatory retirement provision was said to perpetuate a disadvantage that
constituted a discriminatory practice. The Court in Vilven thereby
concluded that it was the agreement that was unfair in its
treatment of older workers, not the complainants’ challenge to an agreement
that they had finished benefiting by.
[398]
One aspect that does not appear to have been
referred to in this discussion is that the collective agreement
was set up specifically to favour older workers under a rigid seniority
regime. As I understand the scheme, this was done by providing older pilots
with highly generous economic benefits and better working conditions as they
approached the end of their careers. This was made possible by the pilots
receiving considerably less in the way of returns, while requiring more effort
to achieve them, when younger. In other words, the generous benefits and
working conditions made available to older pilots depended on the effort and
hardship put in as younger workers.
[399]
Any claim therefore, for additional benefits by
staying on after the agreed-upon retirement date would not have been “earned”.
It would be similar to making contributions X to an RRSP to produce Y results
over 35 years. But when the time comes to withdraw funds after 35 years, the
pilot asks for Y, plus an additional amount on top of that.
[400]
Nor is it arguable that an exception should be
made to the age of retirement for those pilots who entered Air Canada late and therefore may have greater needs because their pension income will not be
equal to those other pilots whose careers are longer. They argue that they have
not been employed long enough to have the same generous benefits as other
pilots. They have no desire to take out more than they earned; just let them
work longer to generate more pension income and be treated like other pilots
who have the advantage of a longer career.
[401]
However, the consequences are the same whatever
the reason to extend someone’s age of retirement: those that follow must work
longer and subsidize the extended careers of those who work past 60. Newcomers
cannot expect the rule of equal treatment to be modified to suit their
circumstances when the rule was known and agreed to upon joining Air Canada and its violation has an adverse differential impact on other younger employees who
jointed before they did.
[402]
In a sense, therefore, it is difficult to
assimilate how a regime already favouring older workers would be described as
discriminatory because at age 60 it cut off the continuation of that generous
situation. Termination at that age was intrinsic to the functioning of the
collective agreement which based payment out at older ages on effort put in at
younger ages.
(d) Fairness Reconsidered Based on the Adverse Differential
Impact
[403]
In any event, Mr. Salamat’s evidence brings
unfairness back into the picture, but framed differently. The reference is with
respect to unfairness in outcome to the other pilots as opposed to being
“not unfair” to the complainants. It is now apparent that the advantage gained
by the complainants is not effect-neutral or without negative consequences on
others. The one-off gains by these complainants from working longer would be
subsidized by their fellow pilots, with the youngest pilots suffering the
greatest economic and other disadvantages.
[404]
It is not just that the MRP in the collective
agreement is part of a comprehensive benefits scheme intended to provide
equality of treatment of all members. It also cannot be altered in mid-route
except by agreement of all members. Any change to the agreement introduces unfair
inequality of treatment, by not only allowing older workers to gain more
than others, but also to have others pay for those gains.
[405]
The results will unfairly disadvantage the
present and future members of the union by requiring them to pay for the
claimants’ advantages through delays in entry and in progression of their
careers until retirement, with the youngest and the future members of the union
paying the highest price.
(5) Perpetuation of Stereotypes, Prejudice or Disadvantage
Against Older Workers
(a) Discriminatory Stereotypes of Older Workers
[406]
A further consideration in respect of the issue
of mandatory retirement provisions is to seriously question whether widespread
age discrimination exists in Canada, such that it is appropriate to give the
older worker the benefit of any disadvantage based on age.
[407]
At paragraph 272 of her reasons in Vilven,
Justice Mactavish concluded that it was clear from statements in Supreme Court
cases that older workers, as a group, suffer from a pre-existing disadvantage, vulnerability,
stereotyping or prejudice. I agree based on the jurisprudence.
[408]
In particular at paragraph 271 of her reasons,
she had referred to statements from previous cases to the effect that
discriminatory stereotyping existed against older people “who are presumed to
lack abilities that they may in fact possess” (Gosselin at para 32), or
are “unproductive, inefficient, and lacking in competence” and “no longer
useful members of the labour force”, such that “their services may therefore be
freely and arbitrarily dispensed with” (McKinney at para 347).
[409]
While the Federal Court has already experienced
a reversal in challenging the conclusions of McKinney in the above
cases, I nevertheless would respectfully suggest that conclusions on such
stereotypes of older persons which are based largely on judicial notice should
be revisited. They are the types of conclusions which would be controversial in
today’s society and evidence should be available to properly support any
conclusion on this subject.
[410]
In my view, the daily experience of members of
the older generation (Baby Boomers) would suggest that whatever the stereotypes
of the past, there has been a wholesale positive attitudinal change by the
older worker and by the rest of society which would strongly contradict any
suggestion that members of this generation can any longer be labelled as
victims of age discrimination in the workplace.
[411]
The older Canadian generation enjoys unheard of
advantages in comparison to the generations that preceded it. These advantages
are expressed in terms of the Baby Boomers’ wealth, their role in
decision-making, their health and physical condition, and their expanded level
of activity in society. Most of all, the older generation exhibits confidence,
self-reliance and self-importance which are the attributes of being the most
populous and self-indulgent generation ever, one which has seen a lifetime of
positive economic and social growth beyond anything experienced by past
generations.
[412]
Indeed, the new-found vitality, activism and
longevity of the older generation probably represent its best argument
for extending its work career. But that is entirely a different foundation than
describing the most prosperous and powerful generation in the history of
humankind as victims of negative stereotyping or prejudice, or viewed as
unproductive, incompetent or vulnerable workers.
[413]
Perhaps more to the point, I repeat that there
is no evidence that the older pilots at Air Canada suffer from any of the
negative attitudinal stereotyping or prejudice of the past. Their entire
situation is governed by the collective agreement and has nothing to do with
attitudes that diminish their abilities or competence to fly airplanes.
[414]
On this point, the CHRT concluded in Vilven
Tribunal #1 (2007 CHRT 36) that there was no indication that the
complainants had experienced age-related disadvantages or negative
stereotyping, which finding of fact was not challenged. Its reasons on this
point are set out below:
Does the distinction created by s. 15(1)(c) contribute to or
reinforce stereotyping or pre-existing disadvantage experienced by the
complainants?
[92]
One of the most compelling factor favouring a conclusion that differential
treatment imposed by legislation is discriminatory is pre-existing disadvantage
or vulnerability to stereotyping (Law at para. 63). While it is clear that
airline pilots, as pilots, do not constitute a group which suffers from
negative stereotyping or pre-existing disadvantage, the more appropriate focus
of the analysis here is whether the complainants, as members of the group of
older workers whose employment has been forcibly terminated, are subject to
pre-existing disadvantage or negative stereotyping.
[93]
The disadvantages suffered by older workers have been noted in the case law.
For example, in McKinney, La Forest J. stated that barring specific skills, it
is generally known that persons over 45 have more difficulty finding work than
others. They do not have the flexibility of the young, a disadvantage often
accentuated by the fact that the latter are frequently more recently trained in
the more modern skills (at p. 299). Moreover, while social security and private
pension schemes may afford some financial redress, many older people have need
of additional income, a situation that is becoming apparent as people live
longer (at p. 300).
[94]
In her dissenting reasons in McKinney, Wilson J. noted that there is a
stereotype that older people are unproductive, inefficient, and lacking in
competence. (at p. 413)
[95]
There was no indication that the complainants experienced these age-related
disadvantages or negative stereotyping. On the contrary, the evidence was
that as senior pilots, the complainants were fully up-to-date in the latest
technology and skills required to fly some of the most sophisticated aircraft
in a major international airline.
[96]
Very soon after their retirement from Air Canada, both were able to get work as
pilots with other airlines that did not have mandatory retirement policies. Mr.
Kelly testified that when he was returning his Security Pass to Air Canada following his last flight, he ran into a former colleague who offered him employment
with Skyservice Airlines. He readily accepted the offer.
[97]
The acceptance of employment with Skyservice meant that Mr. Kelly was able to
supplement the $124,000 income that he was receiving from his Air Canada
pension with what he earned as a pilot with Skyservice. At $72,000 per annum,
Mr. Vilven's retirement income was less at retirement than Mr. Kelly's because
Mr. Vilven started work at Air Canada later. However, Mr. Vilven was able to
supplement his pension income with the earnings he received working as a pilot
with Flair Airlines.
[Emphasis
added]
(b) The Impact of Retirement on
Pilots
[415]
I also have difficulty accepting the severity of
the impact of retirement on the older pilots in terms of maintaining a sense of
dignity and self-worth.
[416]
Retirement is something every worker will face.
Accordingly, the adverse experience of its impact cannot be avoided, but at
best postponed. The complainant pilots never had any expectation of working
beyond age 60; this was understood on the day they joined Air Canada. Nor could they have regrets about not achieving their goals, with the seniority
system having moved them up into the upper ranks of the organization. They are
generally in good economic circumstances with a wide range of options open to
them to overcome some of the challenges of a change of circumstances. Most of
all, if they want to continue to fly commercial airplanes, the evidence
indicates that they have no trouble finding work.
[417]
This is apparent from the testimony about loss
of self-esteem caused by retirement from good jobs of the complainants in Vilven.
The best that they could offer was that they missed the exciting work, prestige
and friendships formed as an Air Canada pilot. These disadvantages mostly
reflect the excellent working conditions of the positions that they held in Air
Canada. This would include for example, being able to earn the maximum salary
of well over $200,000 per annum which is available to pilots at the end of
their careers while operating on a compressed 8-day work schedule, all thanks
to the collective agreement.
(c) McKinney’s Rejection of Making Way for the Younger Worker
[418]
In McKinney, the Court recognized that
improving the prospects for younger members of the labour force was one of the
predominant purposes of mandatory retirement provisions in legislation. It also
recognized that generally older workers were in a better position to protect
themselves from the vicissitudes of unemployment than younger persons. The
Court nevertheless accorded no weight to the legislature’s views or to the
plight of the younger worker. It concluded that “the evidence on this is
conjectural”. The factual foundation is different in the present matter in that
Mr. Salamat’s evidence strongly suggests the financial prejudice, not only to
the younger generation, but to the younger members of the workforce by the fact
of extending the careers of the older pilots.
[419]
The Supreme Court placed the protection against
forced retirement of older workers on a higher plane by its outright
rejection of the concept of making way for the younger worker. The essence of
the Court’s conclusions is stated at paragraph 97 of McKinney as
follows:
97
As for the objective of reducing youth unemployment, it seems to me that such
an objective should not be accorded much weight. If the values and principles
essential to a free and democratic society include, according to Oakes,
"respect for the inherent dignity of the human person" and
"commitment to social justice and equality", then the objective of
forcibly retiring older workers in order to make way for younger workers is in
itself discriminatory since it assumes that the continued employment of
some individuals is less important to those individuals, and of less value
to society at large, than is the employment of other individuals, solely on the
basis of age.
[Emphasis
added]
[420]
In my view, this passage with its forceful
language spelled the death knell for the survival of any legislative mandatory
retirement provision in Canada. It is also cited in cases where the issue of
the rights of the older versus the younger worker are at play, such as in these
cases. I would, with respect, offer two comments on the Court’s reasoning.
[421]
First, the Court appears to be declaring that
the concept of forcibly retiring older workers in order to make way for younger
workers is per se, formalistically, discriminatory. This cannot be a
complete statement of the reasoning because the Court has declared on numerous
occasions that a contextual analysis that comprises all aspects of the problem
is required.
[422]
Secondly,
I am not aware of anyone suggesting that one generation’s right to work should
be viewed as less important or less valued to society. Rather the Court’s
reference to “commitment to social justice and equality” hits the fundamental
issue squarely on the head.
[423]
The essence of the problem in terms of social
justice and equity is that good jobs are a scarce resource. It also seems
apparent from the increasing concern about youth unemployment that good jobs are
now more scarce a resource than they were in the past. Moreover, the
ineluctable march of globalization in combination with the exponential growth
of technological innovation suggests that the trend towards the scarcity of
good jobs is only going to get worse.
[424]
As with any scarce resource, the real issue is
how best to distribute it; in this case between generations. This brings into
play the other reality that in many cases good jobs for the younger worker and
his or her economic advancement will depend on openings being created by
persons retiring from the workforce. But equally important, it appears that the
extension of careers also negatively affects those already employed, having the
greatest impact on the younger worker in the area of good scarce jobs.
[425]
In this matter the specific scarce jobs to be
distributed are senior positions at Air Canada. These had been made available
in accordance with a scheme based upon the delay of promotions with major
benefits at the end of careers, along with the sanctity of the collective
agreement that promoted equality of treatment of its members. Tangentially
however, the scarcity of good jobs extends to pilots wishing to gain entry to
Air Canada, as well as those looking for their first job as a commercial
airline pilot which could eventually open up when an Air Canada pilot completely
retires from commercial flying. This is more likely to happen upon their leaving
Air Canada.
[426]
In the end it is all about fixing on a fair age
of retirement in a world of scarce jobs and economic disadvantage due to
delayed careers in the context of a collective agreement that seeks to treat
all employees equally based on inputs and outputs over the life-cycle of the
agreement.
[427]
Mr. Salamat has demonstrated that altering the
line in favour of the older pilots without any other change to the seniority,
pay, and benefits system is done at a cost to the younger pilot and contravenes
the equity of treatment principle enshrined in the collective agreement. The
opposite economic effect is obviously achieved by forcing pilots to retire
early, say before age 65. But this has to be seen in the context of the
collective agreement and all the circumstances touching on need, vulnerability,
expectations and options that reflect the reality of the pilots’ positions.
[428]
Taking into consideration all circumstances
relevant to the interpretation of a statute promoting fair and equitable
treatment of persons, I do not see this as a case of a prima facie
discriminatory rule that must be justified on the basis of hardship. I would
think that a better conclusion would be to interpret the CHRA such that no
discrimination arises by maintaining a collective agreement provision of
retirement for Air Canada pilots at age 60 negotiated with the employer for an
ameliorative purpose intended to ensure an agreed upon scheme to achieve
benefits for all pilots at Air Canada and to distribute the fruits of the
agreement fairly and equally amongst the union’s members without exception.
(6) Conclusions on Revisiting Whether the Age 60 Rule of
Retirement is Discriminatory
[429]
I have concluded that Mr. Salamat’s evidence
demonstrating an adverse differential effect by the elimination of the MRP in
the collective agreement requires me to supplement my direction to the Tribunal
to permit ACPA and Air Canada to argue that the age 60 retirement rule is not
discriminatory.
[430]
On this basis it should be understood that the
evidence on the adverse differential impact is relevant to both issues of
discrimination and hardship. Nevertheless, the analysis of each issue, albeit
using similar evidence, must be conducted separately, commencing with the
discrimination analysis. Moreover, the hardship analysis would include a
consideration, if after determining that undue hardship would result from the
elimination of the MRP, whether the characteristics of the age discrimination
meant that the importance of promoting freedom from it could not justify a
lower standard. This entails a consideration of much of the discrimination
analysis described above.
[431]
The Tribunal may have regard to my foregoing
remarks on the subject, but must understand that its decision must be based
upon the evidence adduced before it. In that regard, the Tribunal exercises a
full uninhibited discretion to decide the case in accordance with the evidence
before it, particularly given that the Meiorin test has not previously
been applied to a union’s representation rights as in this case.
V.
Decisions
on the Applications of the Complainants, Air Canada and ACPA
[432]
First, the application of the complainants is
allowed and the decision of the Tribunal in respect of normal age of retirement
is set aside and remitted to the same panel for reconsideration. Upon
reconsideration, the Tribunal is directed to apply the factors of the test in Vilven
disjunctively as described above. It is also directed to determine
attributes of similarity of pilots of comparator airlines and those of Air
Canada based on what pilots actually do, e.g. are the attributes of positions
similar for pilots flying large and small planes in terms of the level of
skill, knowledge and responsibilities each requires?
[433]
Secondly, I dismiss Air Canada’s application seeking to set aside the Tribunal’s decision that it had not established that
the mandatory retirement provision in the collective agreement is a BFOR within
sections 15(1)(a) and 15(2) of the CHRA.
[434]
Thirdly, the Tribunal’s decision that ACPA had
not established that the mandatory retirement provision in the collective
agreement is a BFOR under sections 15(1)(a) and 15(2) of the Act is set
aside and returned to the same panel with the following directions:
a.
ACPA and Air Canada may lead evidence and argue
that the age 60 retirement rule in the collective agreement is not
discriminatory.
b.
Section 15(1)(a) of the CHRA regarding a
BFOR defence applies to employee organizations.
c.
The hardship factors in section 15(2) of the
CHRA are not limited to safety, health and costs.
d.
The Tribunal is to apply the four-step hybrid Meiorin
test as described at para 220 above.
e.
In determining whether hardship is occasioned to
the comparator pilots by the elimination of the mandatory retirement provision
in the collective agreement, the Tribunal will give due consideration to areas
of concern of the Court described above, including permitting the introduction
of admissible evidence on the effect of pensions in the determination of any
adverse differential effect caused by the elimination of the mandatory
retirement rule from the collective agreement.
f.
If undue hardship is established to the
comparator pilots, the Tribunal shall not dismiss the complaint against ACPA
unless satisfied that the importance of upholding age discrimination in all the
circumstances is such that it cannot justify a lesser standard.
g.
As ACPA and Air Canada are jointly liable for
having adopted the age 60 retirement provision, a dismissal of the complaint
against ACPA results in the dismissal of the complaint against Air Canada.
h.
Applications T-971-12 and T-979-12 are dismissed
without costs.
VI.
Costs
[435]
No costs are ordered in any of the applications.
Email this Content