SUPREME
COURT OF CANADA
Between:
Attorney General
of Canada
Appellant/Respondent
on cross-appeal
and
George
Hislop, Brent E. Daum, Albert McNutt, Eric Brogaard
and
Gail Meredith
Respondents/Appellants
on cross-appeal
‑ and ‑
Attorney
General of Ontario, Attorney General of Quebec,
Attorney
General of Alberta and Egale Canada Inc.
Interveners
Coram:
McLachlin C.J. and Bastarache, Binnie, LeBel, Deschamps, Abella and
Rothstein JJ.
Reasons for
Judgment:
(paras. 1 to 136)
Concurring
Reasons:
(paras. 137 to 165)
|
LeBel and Rothstein JJ. (McLachlin C.J. and Binnie,
Deschamps and Abella JJ. concurring)
Bastarache J.
|
______________________________
Canada (Attorney General) v. Hislop, [2007] 1 S.C.R. 429,
2007 SCC 10
Attorney General of Canada Appellant/Respondent
on cross‑appeal
v.
George Hislop, Brent E. Daum,
Albert McNutt, Eric Brogaard
and Gail Meredith Respondents/Appellants
on cross‑appeal
and
Attorney General of Ontario, Attorney General of Quebec,
Attorney General of Alberta and Egale Canada Inc. Interveners
Indexed as: Canada (Attorney General) v. Hislop
Neutral citation: 2007 SCC 10.
File No.: 30755.
2006: May 16; 2007: March 1.
Present: McLachlin C.J. and Bastarache, Binnie, LeBel,
Deschamps, Abella and Rothstein JJ.
on appeal from the court of appeal for ontario
Constitutional law — Charter of Rights — Equality
rights — Sexual orientation — Survivorship pensions — Whether provisions of
Canada Pension Plan limiting eligibility for survivor benefits to same‑sex
partners of deceased contributors who died on or after January 1, 1998 infringe
right to equality — If so, whether infringement justified — Canadian Charter of
Rights and Freedoms, ss. 1 , 15(1) — Canada Pension Plan, R.S.C. 1985,
c. C‑8, ss. 44(1.1) , 72 .
Constitutional law — Charter of Rights — Equality
rights — Sexual orientation — Survivorship pensions — Whether estate has
standing to advance equality rights claim on behalf of deceased survivor —
Canadian Charter of Rights and Freedoms, s. 15(1) — Canada Pension Plan,
R.S.C. 1985, c. C‑8, s. 60(2) .
Constitutional law — Remedy — Retroactive remedy —
Circumstances under which courts may limit retroactive remedy — Constitution
Act, 1982, s. 52 .
Following the 1999 decision of this Court in M.
v. H., the federal government amended the Canada Pension Plan (“CPP ”)
in 2000 to extend survivor benefits to same‑sex partners by changing the
definition of “spouse” to conform with the equality rights provisions of
s. 15(1) of the Canadian Charter of Rights and Freedoms and by
adding new provisions. However, under s. 44(1.1) of the CPP ,
eligibility was limited to same‑sex partners whose “spouse” died on or
after January 1, 1998. Benefits were not retroactive to
April 17, 1985, when s. 15(1) came into force, or the date of
death of the “spouse”, whichever occurred later. Section 72(2) of the CPP
precluded payments to same‑sex survivors for any month before
July 2000, the month s. 72(2) came into force. The effect of
s. 72(2) came to an end as of June 2001, because after that month,
same‑sex and opposite‑sex survivors benefited from application of
the general rule in s. 72(1) , which limits survivors’ arrears benefits to
not more than 12 months prior to the month in which the application is
received. Finally, s. 60(2) of the CPP , a general provision not
affected by the 2000 amendments, limits the right of estates of survivors
from obtaining benefits if the application for the benefits is not made within
12 months after the death of the survivor. A class action was commenced
by same‑sex survivors (“Hislop class”), challenging the constitutionality
of the four sections. The Ontario Superior Court of Justice held that
ss. 44(1.1) and 72(2) violated s. 15(1) and were not justified under
s. 1 of the Charter . Same‑sex survivors were granted a
constitutional exemption from ss. 60(2) and 72(1) . Survivors were
entitled to benefits retroactive to the later of the date of death of their
partner and April 17, 1985, with interest. The Court of Appeal
affirmed the finding that ss. 44(1.1) and 72(2) were unconstitutional, but
held that ss. 60(2) and 72(1) did not infringe s. 15(1) . Sections
44(1.1) and 72(2) were struck down, leaving a same‑sex surviving spouse
entitled to survivor benefits subject only to the 12‑month cap on arrears
and the limitation on estate claims (as were all benefits claimants).
Held: The
appeal and cross‑appeal should be dismissed.
Per McLachlin C.J. and
Binnie, LeBel, Deschamps, Abella and Rothstein JJ.: Essential
to the question of differential treatment in the s. 15(1) analysis is the
choice of comparator group. Under s. 44(1.1), the appropriate comparison
is between same‑sex and opposite‑sex survivors whose partners died
before January 1, 1998; under s. 72(2) , it is between same‑sex
and opposite‑sex survivors who applied for survivor’s pensions between
July 2000 and June 2001. To frame, as suggested by the government,
the comparator group in terms of the express distinction made in
ss. 44(1.1) and 72(2) between two groups of survivors of same‑sex
relationships, based on the date on which the relationship ended as a result of
the death of one of the partners, would be to miss the fundamental reason for
the 2000 CPP amendments — namely, to remedy unequal
treatment between same‑sex and opposite‑sex couples. Here, the
Court of Appeal was right to conclude that ss. 44(1.1) and 72(2) infringe
s. 15(1) of the Charter . Both provisions treat same‑sex
survivors differently from their comparator group; sexual orientation is
recognized as an analogous ground of discrimination under s. 15(1) , and
the differential treatment discriminated in a substantive sense. [24] [37‑38]
[60]
Section 44(1.1) is not saved under s. 1 of
the Charter . In an appropriate case, the matching of benefits conferred
under the CPP with obligations imposed on same‑sex partners under
other legislation may be a pressing and substantial objective of limits on the
extension of social benefit legislation as there are complex relationships
between income tax credits, pensions, supplements and other entitlements, often
based on conjugal relationships. It is not enough, however, to only make
general reference to these relationships. There must be an explanation
supported by relevant evidence as to what those relationships are, why they are
relevant and why they justify the limit on the Charter right that has
been found to be violated. In the absence of such evidence in this case,
matching cannot be considered to be a pressing and substantial objective of
s. 44(1.1) . The absence of evidence also precludes a finding of rational
connection between s. 44(1.1) and its objective and of minimal impairment of
a Charter right. [47] [52‑54]
Likewise, the government has failed to establish a
s. 1 justification for s. 72(2) . Where the issue is solely whether Charter
relief claimants should be entitled to the same retroactive benefits expressly
available to their comparator group, absent cost considerations, it is
difficult to see how denial of these benefits has a pressing and substantial
objective. Here, there is an absence of evidence of cost justifying the
provision. In the circumstances, there is no rational connection between
s. 72(2) and its objective, nor does s. 72(2) minimally impair the Charter
rights of the Hislop class. [65‑66]
With respect to s. 60(2) of the CPP , the
estates of those survivors who died more than 12 months before the coming
into force of the 2000 amendments do not have standing to claim a
s. 15(1) Charter right on behalf of the deceased survivor. The use
of the term “individual” in s. 15(1) was intentional and indicates that
s. 15(1) applies to natural persons only. In the context in which the
claim is made here, an estate is just a collection of assets and liabilities of
a person who has died. It is not an individual and it has no dignity that may
be infringed. Mr. Hislop’s situation, however, is different. Although he
died between the time his notice of appeal was filed in this Court and the
hearing of this appeal, he obtained judgment while he was still alive. When a
judgment is obtained, the cause of action upon which the judgment is based is merged
in the judgment. Therefore, where a party dies pending appeal, the appeal
survives even if the original cause of action would not. Although
s. 15(1) rights are personal, the constitutional issues raised here are
issues of public importance. Given the public interest in ensuring that
questions of law related to such rights be correctly decided, an appeal from a
judgment raising such issues must be allowed to survive the party’s death
pending the appeal. Moreover, it is a long‑standing principle of law
that a litigant should not be prejudiced by an act of the court. Accordingly,
the estate of any class member who was alive on the date that argument
concluded in the Superior Court of Justice and who otherwise met the
requirements under the CPP is entitled to the benefit of this judgment.
[71‑77]
Although the Hislop class frames its s. 72(1)
argument as an adverse effect discrimination claim, what it is seeking is a
retroactive Charter relief. Since such a relief is unavailable in this
case, it is not necessary to undertake a s. 15(1) analysis in respect of
s. 72(1) . Because courts are adjudicative bodies that, in the usual
course of things, are called upon to decide the legal consequences of past
happenings, they generally grant remedies that are retroactive to the extent
necessary to ensure that successful litigants will have the benefit of the
ruling. However, where the law changes through judicial intervention, it may
be appropriate for a court to limit the retroactive effect of its judgment.
While a substantial change in the law is necessary, it is not sufficient to
justify purely prospective remedies. Once the “substantial change” threshold
is met, factors such as good faith reliance by governments, fairness to the
litigants and the need to respect the constitutional role of legislatures must
be considered to determine whether it is appropriate to limit the retroactive
effect of the remedy. Here, the M. v. H. decision marked a departure
from pre‑existing jurisprudence on same‑sex equality rights, and
all the other relevant factors also weigh in favour of limiting retroactive
relief. First, given the state of the jurisprudence prior to M. v. H.,
the exclusion of same‑sex partners from the former CPP was based
on a reasonable understanding of the state of s. 15(1) jurisprudence as it
existed at the relevant time. Second, the government did not act in bad faith
in failing to extend survivors’ benefits to same‑sex couples prior to M.
v. H. Third, in seeking payment of arrears back as far as 1985, the
Hislop class effectively overlooks the evolution in the jurisprudence of same‑sex
equality rights. Its claim for a retroactive remedy is tantamount to a claim
for compensatory damages flowing from the underinclusiveness of the former CPP .
Imposing that sort of liability on the government, absent bad faith,
unreasonable reliance or conduct that is clearly wrong, would undermine the
important balance between the protection of constitutional rights and the need
for effective government that is struck by the general rule of qualified
immunity. [69] [86] [99‑100] [107‑117]
With respect to ss. 44(1.1) and 72(2) , the
appropriate remedy under s. 52 of the Constitution Act, 1982 is to
strike down both provisions. Here, the extent of the inconsistency with the
equality guarantee is co‑extensive with the scope of these two
provisions. As such, the inconsistency can be cured without distorting or
interfering with the rest of the legislative scheme. A declaration that
ss. 44(1.1) and 72(2) are of no force and effect is in keeping with the
scheme and obvious purpose of the 2000 amendments to extend the survivors’
benefit to same‑sex survivors. Finally, a temporary suspension of
the declaration of invalidity is not appropriate in the present case.
Suspensions should only be used where striking down the legislation without
enacting something in its place would pose a danger to the public, threaten the
rule of law or where it would result in the deprivation of benefits from
deserving persons without benefiting the rights claimant. None of these
factors are present here. As a result, a class member who was precluded by
s. 44(1.1) or s. 72(2) from receiving the survivor’s benefit and who
otherwise meets the eligibility requirements, will be entitled to payment of
that benefit. The relevant date for the purpose of that payment is the date on
which application was received or where no application was made because of the
unconstitutional provisions, the date on which the statement of claim was
filed. In no event are benefits payable in respect of a month prior to August
1999, which is the earliest month in respect of which a class member who
applied for the survivor’s benefit on the day the 2000 amendments came
into force could have been eligible. [121] [134]
Lastly, pre‑judgment interest is available in
the instant case. The CPP is silent on the issue of pre‑judgment
interest and cannot, as such, reasonably be interpreted as creating an
exception to the entitlement created by s. 31 of the Crown Liability
and Proceedings Act . [135]
Per Bastarache
J.: Retroactivity of a constitutional remedy granted under s. 52(1)
of the Constitution Act, 1982 is generally the norm in our
constitutional jurisprudence, and a decision to deny such
remedy — and the Hislop class’s challenge to s. 72(1) of the CPP
in this case — should be based on the balancing of interests that
must take place in any claim for relief from an unconstitutional law. Reliance
on the existence of a substantial change of law is not an appropriate
consideration in the context of Charter rights. Such an approach
implies that these rights depend on judicial recognition of a new or newly
recognized social environment for their genesis. By attaching importance
to changing social conditions, it makes Charter rights dependent on how
the majority of Canadians perceive the claimants’ rights. Society’s views of Charter
claimants — especially in the context of vulnerable minorities — should
not be a factor for determining whether a Charter right was part of the
Constitution in 1985, or whether it sprung into existence later and
thereby be a basis for denying retroactive relief. Our Constitution may change
to correspond to new realities, but that does not mean that every time a new
constitutional interpretation is adopted or a previous decision is overturned
that the fundamental rights and freedoms guaranteed in our Charter have
changed or that new ones have been created. There is a difference between
changes in constitutional interpretation and actual constitutional change.
Even if the existence of a substantial change in the law was an appropriate
threshold criterion, no such change actually occurred in this case. A review
of the case law after 1985 suggests that this Court did not upset
established law when it handed down its decision in M. v. H. [137] [140]
[143‑147] [157]
The normal retroactive effect of judgments, however,
may need to be tempered in certain circumstances in order to protect other
legitimate interests. Reasonable reliance, good faith, fairness to litigants
and Parliament’s role are important considerations to be taken into account in
deciding whether a retroactive constitutional remedy should be denied.
Different considerations apply when deciding whether a suspended declaration of
invalidity should be granted. A suspended declaration is a valid measure when
a declaration of invalidity would pose a danger to the public, threaten the
rule of law or deprive deserving persons of benefits without thereby benefiting
the individual whose rights had been violated. A suspended declaration of
invalidity is ultimately only a temporary limit on retroactivity; it does not
determine whether governments are entitled to deny retroactive relief. It will
not always be necessary to establish reasonable reliance in order to deny
retroactive relief. In this case, given the state of the law pre‑M.
v. H., it would be more accurate to emphasize the fact that the government
was acting in good faith in the face of jurisprudential uncertainty. Finally,
the nature of the constitutional violation is not relevant to deciding whether
to deny a retroactive remedy. Subject to these clarifications, there is general
agreement with how the majority has applied the other factors to this case.
[159‑164]
Cases Cited
By LeBel and Rothstein JJ.
Distinguished: Kingstreet
Investments Ltd. v. New Brunswick (Finance), [2007] 1 S.C.R. 3,
2007 SCC 1; considered: R. v. Big M Drug Mart
Ltd., [1985] 1 S.C.R. 295; Schachter v. Canada, [1992]
2 S.C.R. 679; Miron v. Trudel, [1995] 2 S.C.R. 418; Nova
Scotia (Workers’ Compensation Board) v. Martin, [2003]
2 S.C.R. 504, 2003 SCC 54; referred to: M. v. H.,
[1999] 2 S.C.R. 3; Law v. Canada (Minister of Employment and
Immigration), [1999] 1 S.C.R. 497; Egan v. Canada, [1995]
2 S.C.R. 513; Benner v. Canada (Secretary of State), [1997]
1 S.C.R. 358; R. v. Oakes, [1986] 1 S.C.R. 103; Stinson
Estate v. British Columbia (1999), 70 B.C.L.R. (3d) 233,
1999 BCCA 761; Edmonton Journal v. Alberta (Attorney General),
[1989] 2 S.C.R. 1326; Lew v. Lee, [1924] S.C.R. 612,
aff’d [1925] A.C. 819; Reid v. Batty, [1933] O.W.N. 496, aff’d
[1933] O.W.N. 817; Turner v. London and South‑Western Railway Co.
(1874), L.R. 17 Eq. 561; Gunn v. Harper (1902),
3 O.L.R. 693; Hubert v. DeCamillis (1963), 41 D.L.R.
(2d) 495; Monahan v. Nelson (2000), 186 D.L.R. (4th) 193,
2000 BCCA 297; In re Spectrum Plus Ltd. (in liquidation),
[2005] 2 A.C. 680, [2005] UKHL 41; Reference re Remuneration
of Judges of the Provincial Court of Prince Edward Island, [1998]
1 S.C.R. 3; R. v. Demers, [2004] 2 S.C.R. 489,
2004 SCC 46; Reference re Manitoba Language Rights, [1985]
1 S.C.R. 721; Edwards v. Attorney‑General for Canada,
[1930] A.C. 124; Reference re Same‑Sex Marriage, [2004]
3 S.C.R. 698, 2004 SCC 79; Attorney General of Quebec v.
Blaikie, [1979] 2 S.C.R. 1016; Re Residential Tenancies Act,
1979, [1981] 1 S.C.R. 714; Law Society of Upper Canada v.
Skapinker, [1984] 1 S.C.R. 357; Hunter v. Southam Inc.,
[1984] 2 S.C.R. 145; United States v. Johnson,
457 U.S. 537 (1982); Chevron Oil Co. v. Huson,
404 U.S. 97 (1971); Mackin v. New Brunswick (Minister of Finance),
[2002] 1 S.C.R. 405, 2002 SCC 13; Guimond v. Quebec
(Attorney General), [1996] 3 S.C.R. 347; Tétreault‑Gadoury
v. Canada (Employment and Immigration Commission), [1991]
2 S.C.R. 22.
By Bastarache J.
Referred to: Nova
Scotia (Workers’ Compensation Board) v. Martin, [2003]
2 S.C.R. 504, 2003 SCC 54; Jim Pattison Industries Ltd.
v. The Queen, [1984] 2 F.C. 954; Davidson v. Davidson (1986),
33 D.L.R. (4th) 161, leave to appeal refused, [1987] 1 S.C.R. vii; Rath
v. Kemp (1996), 46 Alta. L.R. (3d) 1; Reference re Manitoba
Language Rights, [1985] 1 S.C.R. 721; Miron v. Trudel,
[1995] 2 S.C.R. 418; R. v. Hess, [1990]
2 S.C.R. 906; Reference re Same‑Sex Marriage, [2004]
3 S.C.R. 698, 2004 SCC 79; M. v. H., [1999]
2 S.C.R. 3, aff’g (1996), 96 O.A.C. 173; Andrews v.
Ontario (Minister of Health) (1988), 64 O.R. (2d) 258; Vogel
v. Manitoba (1992), 90 D.L.R. (4th) 84, rev’d (1995), 126 D.L.R.
(4th) 72; Veysey v. Canada (Commissioner of the Correctional Service),
[1990] 1 F.C. 321; Knodel v. British Columbia (Medical Services
Commission) (1991), 58 B.C.L.R. (2d) 356; Canada (Attorney
General) v. Mossop, [1993] 1 S.C.R. 554; Egan v. Canada,
[1995] 2 S.C.R. 513; McKinney v. University of Guelph, [1990]
3 S.C.R. 229; Kane v. Ontario (Attorney General) (1997),
152 D.L.R. (4th) 738; Vriend v. Alberta, [1998]
1 S.C.R. 493; Rosenberg v. Canada (Attorney General) (1998),
38 O.R. (3d) 577; Schachter v. Canada, [1992]
2 S.C.R. 679; Kingstreet Investments Ltd. v. New Brunswick
(Finance), [2007] 1 S.C.R. 3, 2007 SCC 1.
Statutes and Regulations Cited
Amendments
Because of the Supreme Court of Canada Decision in M. v. H. Act, 1999, S.O. 1999, c. 6, s. 68(2).
Canada Pension Plan, R.S.C. 1985, c. C‑8, ss. 2(1) “common-law
partner” [ad. 2000, c. 12, s. 42(2)], “spouse” [ad. c. 30 (2nd Supp.), s. 1(3) ;
rep. 2000, c. 12, s. 42(1)], 42(1) “survivor” [ad. 2000, c. 12, s. 44(3)],
44(1)(d) [repl. idem, s. 45(1) ], 44(1.1) [ad. idem, s.
45(2) ], 60(2), 72 [am. idem, s. 54 ].
Canadian Charter of Rights and
Freedoms, ss. 1 , 15(1) , 24(1) .
Constitution Act, 1982, s. 52(1) .
Crown Liability and Proceedings
Act, R.S.C. 1985, c. C‑50,
s. 31 .
Income Tax Act, R.S.C. 1985, c. 1 (5th Supp .).
Modernization of Benefits and
Obligations Act, S.C. 2000, c. 12,
ss. 42 , 45(2) .
Authors Cited
Blackstone, William. Commentaries
on the Laws of England, vol. 1. Oxford: Clarendon Press, 1765.
Canada. Parliament. Special
Joint Committee on the Constitution of Canada. Minutes of Proceedings and
Evidence of the Special Joint Committee of the Senate and of the House of
Commons on the Constitution of Canada (1980-81), Issue No. 43,
January 22, 1981, pp. 43:39‑43:44; Issue No. 44,
January 23, 1981, pp. 44:6‑44:10; Issue No. 47,
January 28, 1981, p. 47:88; Issue No. 48, January 29, 1981,
pp. 48:4‑48:49.
Choudhry, Sujit, and
Kent Roach. “Putting the Past Behind Us? Prospective Judicial and
Legislative Constitutional Remedies” (2003), 21 S.C.L.R.
(2d) 205.
Driedger, Elmer A. Construction
of Statutes, 2nd ed. Toronto: Butterworths, 1983.
Driedger, Elmer A.
“Statutes: Retroactive Retrospective Reflections” (1978), 56 Can. Bar
Rev. 264.
Fisch, Jill E.
“Retroactivity and Legal Change: An Equilibrium Approach” (1997),
110 Harv. L. Rev. 1055.
Hogg, Peter W. Constitutional
Law of Canada, vol. 2, loose‑leaf ed. Scarborough,
Ont.: Carswell, 1992 (updated 2005, release 1).
Lord Reid. “The Judge as Law Maker” (1972-1973), 12 J.S.P.T.L. 22.
Roach, Kent. Constitutional
Remedies in Canada. Aurora, Ont.: Canada Law Book, 1994
(loose-leaf updated 2006, release 13).
Sampford, Charles.
Retrospectivity and the Rule of Law. Oxford: Oxford University Press,
2006.
APPEAL and CROSS‑APPEAL from a judgment of the
Ontario Court of Appeal (Charron, Feldman and Lang JJ.A.) (2004),
73 O.R. (3d) 641, 246 D.L.R. (4th) 644,
192 O.A.C. 331, 124 C.R.R. (2d) 1, 12 R.F.L.
(6th) 71, [2004] O.J. No. 4815 (QL), setting aside in part a decision
of Macdonald J. (2003), 234 D.L.R. (4th) 465, 114 C.R.R.
(2d) 303, 50 R.F.L. (5th) 26, [2003] O.J. No. 5212 (QL). Appeal
and cross‑appeal dismissed.
Roslyn J. Levine,
Q.C., and Paul Vickery, for the appellant/respondent on cross‑appeal.
J. J. Camp, Q.C.,
R. Douglas Elliott, Sharon D. Matthews,
Patricia A. LeFebour, R. Trent Morris and Sean M. Grayson,
for the respondents/appellants on cross‑appeal.
Daniel Guttman and
Janet E. Minor, for the intervener the Attorney
General of Ontario.
Hugo Jean, for the
intervener the Attorney General of Quebec.
Nick Parker, for
the intervener the Attorney General of Alberta.
Written submissions only by Cynthia Petersen
for the intervener Egale Canada Inc.
The judgment of McLachlin C.J. and Binnie, LeBel,
Deschamps, Abella and Rothstein JJ. was delivered by
LeBel and Rothstein
JJ. —
I. Introduction
1
This is an appeal by the Attorney General of Canada (“government”) from
a decision of the Ontario Court of Appeal upholding in part the decision of the
Ontario Superior Court of Justice and a cross-appeal by the respondents
(“Hislop class”). The government appeals the finding of unconstitutionality of
ss. 44(1.1) and 72(2) of the Canada Pension Plan, R.S.C. 1985, c. C-8 (“CPP ”),
as enacted by the Modernization of Benefits and Obligations Act, S.C.
2000, c. 12 (“MBOA ”). The Hislop class cross-appeals the finding of
constitutionality of ss. 72(1) and 60(2) of the CPP and the denial of a
remedy with respect thereto. We propose to dismiss both the appeal and the
cross-appeal.
II. Overview
2
Under the CPP , the spouse of a contributor was entitled to apply
for a survivor’s pension after the death of the contributor. If the survivor’s
pension was approved, it would be payable for each month following the death of
the contributor.
3
However, if the application of the survivor was not received by the
government within 12 months of the death of the contributor, the arrears that
could be claimed by the survivor were limited to a 12-month period preceding
the receipt of the application. For example, if the contributor died in
January 1995, and the survivor’s application was not received by the government
until March 1998, the maximum monthly pension arrears that could be claimed
would be for 12 months preceding March 1998.
4
Until July 2000, for purposes of entitlement to a survivor’s pension
under the CPP , the survivor had to have been married to the contributor
or had to be a person of the opposite sex who was cohabiting with the
contributor in a conjugal relationship at the time of the contributor’s death.
Same-sex conjugal relationships were not recognized and the survivor of a
same-sex conjugal relationship was not eligible to receive a survivorship
pension under the CPP .
5
In M. v. H., [1999] 2 S.C.R. 3, this Court struck down the
opposite-sex definition of spouse in the Family Law Act, R.S.O. 1990, c.
F.3, as contrary to s. 15(1) of the Canadian Charter of Rights and Freedoms .
The declaration of invalidity was suspended for six months to give the Ontario
government time to review the legislation. As a consequence of M. v. H.,
the federal government and the governments of other provinces also amended a
number of their statutes to address the constitutional deficiency. The federal
government enacted the MBOA , which amended 68 pieces of legislation.
This appeal involves a constitutional challenge to remedial amendments made to
the CPP by the MBOA to recognize same-sex conjugal relationships
for the purpose of entitlement to survivor’s pensions under the CPP .
6
The Hislop class commenced a class action claiming that the MBOA ’s
remedial amendments to the CPP were contrary to s. 15(1) of the Charter
as being underinclusive. There are four issues to consider. First, the
amendments restricted eligibility for survivor benefits under the CPP
to survivors of same-sex conjugal relationships whose partners died on or after
January 1, 1998. The Hislop class argued that survivors of contributing
partners of same-sex conjugal relationships who died anytime after
s. 15(1) of the Charter became effective (April 17, 1985) should be
eligible to make an application for CPP survivorship benefits. The
Ontario Court of Appeal agreed with the Hislop class and found that the eligibility
restriction violated s. 15(1) of the Charter and that the restriction
could not be justified under s. 1 . In other words, survivors of same-sex
conjugal relationships in which the contributors died any time after April 17,
1985 should be eligible to receive survivor’s pensions. We agree with that
conclusion.
7
Second, for eligible same-sex survivors, the MBOA remedial
amendments granted monthly pension payments commencing in July 2000,
when the amendments came into force. As such, the amendments did not grant
benefits retroactively. Irrespective of when the same-sex survivor became
eligible, the monthly pension payments would not commence before July 2000.
8
The Hislop class argued that survivors of same-sex conjugal
relationships should be entitled to retroactive benefits from the time of death
of the same-sex conjugal contributing partner. The Ontario Court of Appeal
dismissed the Hislop class’s claim for retroactive benefits.
9
The third and fourth issues are of less pervasive significance. In
respect of the third issue, as explained earlier, general provisions of the CPP
unaffected by MBOA amendments allowed for monthly pension arrears for up
to 12 months preceding the receipt by the government of an application for
benefits. The Ontario Court of Appeal found that the MBOA amendments
that granted benefits commencing in July 2000, without entitling a same-sex
survivor the opportunity of up to 12 months of arrears as was available to
opposite-sex survivors, were contrary to s. 15 of the Charter and could
not be justified under s. 1 .
10
As to the fourth issue, the Ontario Court of Appeal held that estates of
same-sex survivors have no standing to claim a remedy for breach of s. 15(1) of
the Charter .
III. Statutory Provisions at Issue
11
Prior to the MBOA , “spouse” was defined in the CPP as:
Canada
Pension Plan, R.S.C. 1985, c. C-8
2. (1) . . .
“spouse”, in relation to a contributor, means,
(a) except in or in relation to section 55,
(i) if there is no person described in subparagraph (ii), a person
who is married to the contributor at the relevant time, or
(ii) a person of the opposite sex who is cohabiting with the
contributor in a conjugal relationship at the relevant time, having so
cohabited with the contributor for a continuous period of at least one year,
and
(b) in or in relation to section 55, a person who is married
to the contributor at the relevant time,
and, in the case of a contributor’s death, the “relevant time”, for
greater certainty, means the time of the contributor’s death;
12
The MBOA came into force on July 31, 2000. Section 42(1) of the MBOA
repealed the definition of spouse in the CPP . Section 42(2) of the MBOA
substituted a definition of “common-law partner”. In that definition there is
no express or implied reference to the gender of the partners. The definition
of “common-law partner” reads:
2. (1) In this Act,
.
. .
“common-law partner”, in relation to a
contributor, means a person who is cohabiting with the contributor in a
conjugal relationship at the relevant time, having so cohabited with the
contributor for a continuous period of at least one year. For greater
certainty, in the case of a contributor’s death, the “relevant time” means the
time of the contributor’s death.
13
By a series of other MBOA amendments to the CPP , survivors
of same-sex conjugal relationships became entitled to receive benefits under
the CPP . However, by s. 45(2) of the MBOA , s. 44(1.1) of the CPP
was added whereby no survivorship pension would be payable to survivors of
same-sex conjugal relationships unless they became survivors on or after
January 1, 1998. In other words, if the partner of the same-sex survivor died
before January 1, 1998, the survivor would not be eligible for a survivorship
pension under the CPP . Section 44(1.1) provides:
44. . . .
(1.1) In the case of a common-law partner who was
not, immediately before the coming into force of this subsection, a person
described in subparagraph (a)(ii) of the definition “spouse” in
subsection 2(1) as that definition read at that time, no survivor’s pension
shall be paid under paragraph (1)(d) unless the common-law partner
became a survivor on or after January 1, 1998.
14
A second provision added by the MBOA was s. 72(2) of the CPP .
Section 72(2) operated to preclude payments to same-sex survivors for any
month before July 2000, the month s. 72(2) came into force. Section 72(2)
provides:
72. . . .
(2) In the case of a survivor who was the
contributor’s common-law partner and was not, immediately before the coming
into force of this subsection, a person described in subparagraph (a)(ii)
of the definition “spouse” in subsection 2(1) as that definition read at that
time, no survivor’s pension may be paid for any month before the month in which
this subsection comes into force.
15
The effect of s. 72(2) was to entirely preclude retroactive pension
benefits to same-sex survivors. The Hislop class’s position is that same-sex
survivors should be entitled to retroactive benefits to the month following the
death of their same-sex conjugal partner.
16
The incidental effect of s. 72(2) was to preclude for same-sex survivors
the opportunity to seek up to 12 months of pension arrears prior to July 2000,
an opportunity that was available to opposite-sex survivors under s. 72(1).
Section 72(1) provides in relevant part:
72. (1) Subject to subsection (2) and
section 62, where payment of a survivor’s pension is approved, the pension is
payable for each month commencing with the month following
.
. .
but in no case earlier than the twelfth month preceding the month
following the month in which the application was received.
The effect of
s. 72(2) came to an end as of June 2001, because after that month, same-sex and
opposite-sex survivors benefited from the application of the general rule in s.
72(1).
17
As indicated, s. 72(1) limits survivors’ arrears benefits to not more
than 12 months prior to the month in which the application is received.
The Hislop class says that s. 72(1) imposes adverse effect discrimination on
survivors of same-sex conjugal relationships. While s. 72(1) is facially
neutral, access to survivorship pensions for same-sex survivors was not
recognized until the enactment of the MBOA . Thus, even if s. 72(2)
was struck down, the effect of s. 72(1) is to limit same-sex survivors from
claiming benefits for more than 12 months prior to when their application was
made. The Hislop class argues that s. 72(1) should be suspended to enable
same-sex survivors to obtain payment of pension benefits back to when their
same-sex partner died.
18
Finally, s. 60(2) of the CPP , a general provision not affected by
the MBOA , limited the right of estates of survivors from obtaining
benefits if the application for the benefits is not made within 12 months after
the death of the survivor. Section 60(2) provides:
60. . . .
(2) Notwithstanding anything in this Act, but
subject to subsections (2.1) and (2.2), an application for a benefit, other
than a death benefit, that would have been payable in respect of a month to a
deceased person who, prior to the person’s death, would have been entitled on
approval of an application to payment of that benefit under this Act may be
approved in respect of that month only if it is made within 12 months after the
death of that person by the estate, the representative or heir of that person or
by any person that may be prescribed by regulation.
19
The Hislop class says that s. 60(2) should be suspended to enable
estates of survivors of same-sex conjugal relationships to claim retroactive
benefits for the period during which the survivor would have been entitled to
benefits had same-sex survivors been recognized by the CPP as of the
date of the death of the contributor. The suspension of s. 60(2) would only
avail to the benefit of estates if ss. 44(1.1) and 72(2) are found to be
unconstitutional.
IV. Judicial History
A. Ontario Superior Court of Justice (2003),
234 D.L.R. (4th) 465
20
Macdonald J. found that all the respondents in this case were in
conjugal relationships with their same-sex partners for a number of years. She
considered whether ss. 44(1.1) and 72(2) of the CPP breached s. 15(1) of
the Charter , using the three‑part analysis set out in Law v.
Canada (Minister of Employment and Immigration), [1999] 1 S.C.R. 497. The
comparator group chosen by the Hislop class was married heterosexual couples.
The trial judge agreed and found that the legislation treats same-sex survivors
differently than survivors of married or opposite-sex common-law
relationships. She observed that in Egan v. Canada, [1995] 2
S.C.R. 513, at para. 13, it was found that “‘sexual orientation’ is an
analogous ground under s. 15(1) ”. The judge also found that ss. 44(1.1) and
72(2) offended Hislop’s human dignity. She concluded that ss. 44(1.1) and
72(2) infringed s. 15(1) of the Charter .
21
Macdonald J. considered the government’s argument that the Hislop
class’s claim amounted to an impermissible claim for retroactive benefits for a
breach of the Charter . She held that this Court’s judgment in Benner
v. Canada (Secretary of State), [1997] 1 S.C.R. 358, provided “a complete
answer to the arguments raised by the Crown” (para. 99). In her view, “. . .
discrimination was experienced after the coming into force of s. 15 of the Charter .
The plaintiffs have actually experienced discrimination since before the
passage of the Charter , but they seek to apply the Charter
prospectively, to discrimination that took place after April 17, 1985” (ibid.).
She held that the government had failed to demonstrate that the exclusion of
same‑sex survivors’ pensions was a reasonable limit on the Hislop class’s
Charter rights under s. 1 of the Charter .
22
Macdonald J. held that ss. 44(1.1) and 72(2) violated s. 15(1) of the Charter
and were not justified under s. 1 . Under s. 52(1) of the Constitution Act,
1982 , she granted same‑sex survivors a constitutional exemption from
the two general sections (ss. 60(2) and 72(1) ) of the CPP . She
also awarded interest under s. 31 of the Crown Liability and Proceedings Act,
R.S.C. 1985, c. C-50 , on the unpaid survivors’ pensions from February 1, 1992
or one month after the death of the contributor, whichever is later. She also
awarded costs at approximately $1.3 million. She dismissed the Hislop class’s
claims for symbolic damages of $20,000 for each class member under s. 24(1) of
the Charter , and their claims for damages for breach of fiduciary duty
and unjust enrichment.
B. Ontario Court of Appeal (Charron,
Feldman and Lang JJ.A.) (2004), 73 O.R. (3d) 641
23
The Court of Appeal affirmed Macdonald J.’s decision with regard to the
unconstitutionality of ss. 44(1.1) and 72(2) . However, the court queried why
the trial judge considered ss. 60(2) and 72(1) in her analysis of remedy but
did not review them in her s. 15(1) or s. 1 analysis. The court was informed
that the trial judge’s remedy analysis followed the argument as presented at
trial by the Hislop class’s counsel. The court invited written submissions as
to the constitutional validity of ss. 60(2) and 72(1). The court then found
that it had the necessary factual foundation and jurisdiction to consider the
constitutional validity of these sections and that they must be subjected to a
s. 15 and a s. 1 Charter analysis.
24
The court concluded that ss. 44(1.1) and 72(2) offended s. 15(1) of the Charter .
The court found that the appropriate comparator group is opposite‑sex
surviving partners and that, although the impugned sections contain temporal
distinctions as the government argued, those distinctions apply only to same‑sex
surviving partners and not to opposite‑sex survivors. Both ss. 44(1.1)
and 72(2) treat same‑sex survivors differently from their comparator
group. The court said that sexual orientation is now recognized as an
analogous ground under s. 15(1) of the Charter . The differential
treatment discriminated in a substantive sense, and the discrimination could
not be justified under s. 1 .
25
In dealing with s. 60(2), the Court of Appeal held that estates could
not make a s. 15(1) Charter application because s. 15(1) only applies to
“[e]very individual”. Survivors’ estates enjoy no s. 15(1) rights and there
was no basis to assess whether such rights were breached by s. 60(2) .
26
With respect to s. 72(1), the court held that it did not limit same-sex
survivors’ rights. The court observed, at para. 106, that “[i]t is only once
s. 44(1.1) and s. 72(2) are declared unconstitutional that s. 72(1) may have an
adverse effect” on the Hislop class, by limiting pension entitlement to 12
months of arrears, no matter when the partner died. Most same-sex survivors
did not apply for the survivor’s pension when their partners died, because they
were not eligible to apply under the CPP . The court held that s. 72(1)
did not violate s. 15(1) of the Charter because, “in the context of the MBOA ,
it had no adverse effect on the claimants, and at this stage, it is not clear
that it will have an adverse effect on the class” (para. 111). Therefore,
there was no need to consider s. 1 .
27
The Court of Appeal upheld the declaration of invalidity in respect of
ss. 44(1.1) and 72(2). The court held that the two general sections, ss.
60(2) and 72(1) of the CPP , do not breach s. 15(1) of the Charter .
Consequently, there was no remedy to be granted in respect of those sections.
The court set aside the constitutional exemption which was ordered by the trial
judge with respect to ss. 60(2) and 72(1) .
28
The Court of Appeal held that the trial judge properly exercised her
discretion on pre‑judgment interest. She was entitled to award pre‑judgment
interest and there is no basis to interfere with that disposition. Costs were
awarded on a partial indemnity basis.
C. Appeal and Cross-Appeal to This Court
29
The government now appeals the declaration of invalidity of ss. 44(1.1)
and 72(2) and the awarding of pre-judgment interest on pension arrears. The
Hislop class cross-appeals the finding of the Court of Appeal which set aside
the constitutional exemptions ordered by the trial judge with respect to ss.
60(2) and 72(1) .
V. The Issues
30
On October 6, 2005, the Chief Justice stated the following
constitutional questions:
1. Does s. 44(1.1) of the Canada Pension
Plan, R.S.C. 1985, c. C‑8 , infringe s. 15(1) of the Canadian
Charter of Rights and Freedoms ?
2. If so, is the infringement a reasonable
limit prescribed by law as can be demonstrably justified in a free and
democratic society under s. 1 of the Canadian Charter of
Rights and Freedoms ?
3. Does s. 72(2) of the Canada Pension Plan,
R.S.C. 1985, c. C‑8 , infringe s. 15(1) of the Canadian Charter
of Rights and Freedoms ?
4. If so, is the infringement a reasonable
limit prescribed by law as can be demonstrably justified in a free and
democratic society under s. 1 of the Canadian Charter of
Rights and Freedoms ?
5. Does s. 60(2) of the Canada Pension Plan,
R.S.C. 1985, c. C‑8 , infringe s. 15(1) of the Canadian Charter
of Rights and Freedoms ?
6. If so, is the infringement a reasonable
limit prescribed by law as can be demonstrably justified in a free and
democratic society under s. 1 of the Canadian Charter of Rights and Freedoms ?
7. Does s. 72(1) of the Canada Pension Plan,
R.S.C. 1985, c. C‑8 , infringe s. 15(1) of the Canadian Charter
of Rights and Freedoms ?
8. If so, is the infringement a reasonable
limit prescribed by law as can be demonstrably justified in a free and
democratic society under s. 1 of the Canadian Charter of
Rights and Freedoms ?
31
By way of cross-appeal, the Hislop class submitted three questions:
(1) Can estates advance s. 15(1)
rights?
(2) Is there an entitlement to
interest on survivor’s pension arrears?
(3) Did the Court of Appeal err in
setting aside the constitutional exemption in respect of ss. 60(2) and 72(1)?
32
In essence, the issues are:
(1) Should survivors of same-sex
conjugal relationships in which the contributing partner died prior to January
1, 1998 be eligible to receive CPP survivorship benefits?
(2) Should survivors of same-sex
conjugal relationships in which the contributing partner died at any time after
April 17, 1985 be entitled to retroactive CPP benefits from the month
following the death of the contributing partner?
(3) Should estates of deceased
survivors be entitled to claim Charter relief which the survivors would
have been entitled to claim?
VI. Analysis
33
The MBOA was enacted in response to this Court’s decision in M.
v. H. The Act seeks to eliminate from federal legislation distinctions
based on sexual orientation. In the context of survivorship benefits under the
CPP , the MBOA amendments entitle survivors of conjugal
relationships, whether they be of the same or opposite sex, access to the same
benefit regime. In other words, the gender of the parties to a conjugal
relationship is now irrelevant in the survivorship provisions of the CPP .
Even though the MBOA is remedial legislation, the complaint is that the
relief granted by the MBOA does not go far enough, in that it fails to
confer eligibility on survivors whose same-sex partner died prior to January 1,
1998 and fails to grant retroactive relief.
A. Section 44(1.1)
34
It will be convenient to first deal with the eligibility restriction.
The MBOA recognized that the former CPP regime violated s. 15(1)
of the Charter because it failed to recognize survivors of same-sex
conjugal relationships. Section 44(1.1) continues to fail to recognize
survivors of same-sex conjugal relationships whose contributing partner died
before January 1, 1998.
35
Section 15(1) of the Charter provides:
Equality
Rights
15. (1) Every individual is equal
before and under the law and has the right to the equal protection and equal
benefit of the law without discrimination and, in particular, without
discrimination based on race, national or ethnic origin, colour, religion, sex,
age or mental or physical disability.
36
The test applicable to s. 15(1) of the Charter was articulated by
this Court in Law. In short, there must be (1) differential treatment
on the basis of a personal characteristic, (2) that is an enumerated or
analogous ground, (3) which is discriminatory in purpose or effect.
(1) Differential Treatment
37
As the Court of Appeal observed, essential to the question of
differential treatment is the choice of comparator group. Throughout this
litigation, the government has argued that s. 44(1.1) draws a temporal
distinction only. The government’s position is that the provisions of the MBOA
do not differentiate between same-sex couples and opposite-sex couples, but
rather, between two groups of survivors of same-sex relationships, based on the
date their relationships ended as a result of one partner’s death. It cannot,
therefore, violate s. 15(1) because a temporal basis for a distinction is not
an enumerated or recognized analogous ground of discrimination. In our
opinion, the courts below were correct in rejecting this argument.
38
To frame the comparator group in terms of the express distinction made
in s. 44(1.1) between survivors whose partners died before January 1, 1998
and those whose partners died on or after that date would be to miss the
fundamental reason for the enactment of the MBOA . In M. v. H.,
this Court held that the distinction in the spousal support regime between
same-sex and opposite-sex couples was unconstitutional and that it could not be
saved under s. 1 . The MBOA was expressly intended to extend equal
treatment to same-sex partners in a wide range of statutes. It is the purpose
of the MBOA itself that determines the appropriate comparator group.
What must be compared is the subset of same-sex survivors that remains excluded
from the CPP survivor’s benefits, i.e., those whose partners died before
January 1, 1998, and similarly situated opposite-sex survivors. The
appropriate comparator group in respect of the s. 44(1.1) analysis is survivors
of opposite-sex conjugal relationships whose partners died before January 1, 1998.
39
If the government was correct, remedial legislation intended to address
the constitutional infirmity of existing legislation, but which limited
eligibility for relief on a temporal basis, could never be the subject of a
successful s. 15(1) Charter challenge. That is because a temporal basis
of distinction is not one based upon grounds enumerated in s. 15(1) or grounds
analogous thereto. When the government enacts remedial legislation, that
legislation may still violate s. 15(1) requirements. The fact that it is
remedial legislation does not immunize it from Charter review.
40
In this Court, the government’s primary challenge to the s. 15(1)
decision of the Court of Appeal was based on the choice of comparator group.
We find the Court of Appeal did not err in the selection of comparator group.
41
The government cursorily argued that the Charter should be
interpreted in a manner recognizing the evolution of societal opinions and the
incremental recognition of analogous grounds. The government says the MBOA
amendments restricting eligibility to survivors whose partners died on or after
January 1, 1998 are consistent with that evolution. With respect, we do not
see how the evolution of societal opinions and the incremental recognition of
analogous grounds bear on whether s. 44(1.1) continues the discrimination which
the MBOA amendments are intended to remedy. The question is not whether
there was recognition of same-sex discrimination prior to 1998. Rather, the
question is whether the prior discrimination which was recognized by the MBOA
is or is not now being remedied. If survivors of same-sex conjugal
relationships whose partners died before 1998 continue to be ineligible for CPP
survivorship pensions, the prior discrimination that has been recognized by
Parliament in enacting the MBOA continues for such survivors. For these
reasons, we do not accept the government’s evolution argument as a valid
response to the s. 15(1) claim in respect of s. 44(1.1) .
42
This disposes of the government’s s. 15(1) arguments. The government
did not argue any other error in the Charter analysis of the Court of
Appeal. We agree with that analysis and need not duplicate it here.
(2) Section 1
43
Section 1 of the Charter provides:
Guarantee
of Rights and Freedoms
1. The Canadian Charter of Rights and
Freedoms guarantees the rights and freedoms set out in it subject only to
such reasonable limits prescribed by law as can be demonstrably justified in a
free and democratic society.
Under s. 1,
the government has the burden to demonstrate that a discriminatory provision is
a reasonable limit on a s. 15(1) Charter right. If it meets this
burden, the law will be saved as being a demonstrably justified reasonable
limit on that right.
44
The framework for a s. 1 analysis is the well-known Oakes test
(see R. v. Oakes, [1986] 1 S.C.R. 103, at pp. 138-40). The Oakes
test may be formulated as two main tests with subtests under the second branch,
but it may be easier to think of it in terms of four independent tests. If the
legislation fails under any one test, it cannot be justified. The four tests
ask the following questions:
(1) Is the objective of the
legislation pressing and substantial?
(2) Is there a rational connection
between the government’s legislation and its objective?
(3) Does the government’s legislation
minimally impair the Charter right or freedom at stake?
(4) Is the deleterious effect of the Charter
breach outweighed by the salutary effect of the legislation?
45
As the MBOA was the legislative response to this Court’s decision
in M. v. H., the Act obviously has a pressing and substantial purpose.
However, in M. v. H., the Court also held that “when dealing with
underinclusive legislation it is important also to consider the impugned
omission when construing the objective” (para. 100).
46
It is therefore necessary to consider whether the objective of s.
44(1.1) is pressing and substantial. The effect of s. 44(1.1) is to limit
eligibility for CPP survivorship pensions in the case of survivors of
same-sex conjugal relationships. No survivorship pension is payable where the
partner died prior to January 1, 1998.
47
The government’s pressing and substantial objective argument in this
Court is different than in the Ontario courts. In the Ontario courts, the
government’s pressing and substantial argument was based on the evolutionary
recognition of Charter rights and that, prior to M. v. H., there
would have been no rights to same-sex partners. In this Court, the government
argues that the pressing and substantial objective of s. 44(1.1) is the
matching of benefits conferred under the CPP with obligations imposed on
same-sex partners under other legislation.
48
In the government’s factum and in its oral argument there was no
reference to evidence before the trial judge supporting the matching of
benefits and obligations argument. The reasons of the trial judge and the
Court of Appeal make no reference to the matching argument. It appears that
this argument was raised for the first time in this Court.
49
We acknowledge that there may be cases in which the pressing and
substantial objective of legislation and the impugned provisions are obvious
and may be deduced from the legislation itself. However, in the majority of
cases, in order to satisfy the pressing and substantial objective test, the
government must adduce some evidence to support its argument.
50
In the case of s. 44(1.1), the issue is one of eligibility only. It is
not a provision that purports to make benefits retroactive or that would confer
a lump-sum windfall on a survivor. Therefore, it is not obvious why the
matching argument supports the limitation imposed by s. 44(1.1).
51
The government seems to relate the January 1, 1998 eligibility date in
the CPP to provisions of the Income Tax Act, R.S.C. 1985, c. 1
(5th Supp .), allowing for a three-year limitation period for the government to
re-assess or the individual to refile individual tax returns under that Act.
Again, it is not clear why re-assessment or refiling under the Income Tax
Act would have any relation to eligibility under the CPP . The
eligibility date of January 1, 1998 in the CPP has no bearing on income
of years prior to 2000 which is the first year in which same-sex survivors
would actually receive CPP benefits. Therefore the matching argument
relying on limitation periods in the Income Tax Act is perplexing.
52
If there was something to the matching argument in respect of
eligibility, one would think that s. 44(1.1) would not have made the
eligibility date January 1, 1998, but rather would have related it to the date
the legislation came into force, July 2000. The fact that the government saw
no difficulty in granting eligibility as of January 1, 1998 suggests that
matching cannot be considered to be a pressing and substantial objective of s.
44(1.1).
53
We do not say that in an appropriate case the matching of benefits and
obligations may not be a pressing and substantial objective of limits on the
extension of social benefit legislation. We also acknowledge the government’s
argument that there are complex relationships between income tax credits,
pensions, supplements and other entitlements, often based on conjugal
relationships. But it is not enough to make general reference to these
relationships. There must be an explanation supported by relevant evidence as
to what those relationships are, why they are relevant and why they justify the
limit on the Charter right that has been found to be violated.
54
The absence of evidence justifying the matching argument also precludes
a finding of rational connection between s. 44(1.1) and its objective. Without
evidence, it is impossible to understand why s. 44(1.1) is rationally connected
to the matching of benefits and obligations. For the same reason there is no
evidence as to why s. 44(1.1) is minimally impairing of the s. 15 right to
which same-sex survivors have been found to be entitled.
55
More than a vague argument raised for the first time in this Court by
the government will be necessary if a s. 1 argument is to succeed. It cannot
succeed in this case. Because the government has failed to establish the
pressing and substantial objective of the provision, its rational connection to
its objective and that it minimally impairs the Hislop class’s Charter
rights, its argument fails in respect of s. 44(1.1) of the CPP .
B. Section 72(2)
56
Section 72(2) became effective in July 2000. The earliest month for
which survivorship benefits may be payable to a survivor of a same-sex
relationship is therefore July 2000.
57
The Hislop class says that s. 72(2) is to be contrasted with s. 72(1)
which is the provision which applies generally to all survivors eligible for CPP
survivorship pensions. Section 72(1) provides that, in the general case,
survivorship pension arrears are payable for a period of up to 12 months
preceding the month following the month in which the application for the
pension was received.
58
After June 2001, survivors of same-sex and opposite-sex relationships
are equally entitled to up to 12 months of arrears benefits under s. 72(1).
However, s. 72(2) provides that, in the transitional period from July 2000 to
June 2001, same-sex survivors could not access arrears prior to July 2000. By
reason of s. 72(2), same-sex and opposite-sex survivors are treated differently
in that transitional period.
59
The government makes the same “comparator group” argument in respect of
s. 72(2) as it did in respect of s. 44(1.1) . It says the MBOA
amendments did not differentiate between same-sex and opposite-sex couples but
rather between two groups of survivors of same-sex relationships based on the
date on which the relationship ended as a result of the death of one of the
partners.
60
The government is correct that s. 72(2) creates a temporal distinction.
But that does not make its choice of comparator group correct. The appropriate
comparison is between same-sex and opposite-sex survivors who applied for the
survivor’s pensions between July 2000 and June 2001. As with s. 44(1.1) , the
government’s challenge to the s. 15(1) decision of the Court of Appeal is based
solely on the choice of comparator group and the evolutionary recognition of
analogous grounds. For the reasons we have expressed in our analysis of the s.
15(1) arguments in respect of s. 44(1.1) , we do not find merit in the
government’s submissions in respect of s. 72(2).
61
With respect to s. 1, the government seems to make two arguments in
support of s. 72(2). One is that the general principle is that legislation
operates from the time of enactment forward and that this approach is
consistent with prior amendments to the CPP . The other is that the
focus was to fix the liability of the CPP as of July 2000.
62
We agree that remedial legislation generally operates prospectively.
However, the only effect of striking down s. 72(2) is to render s. 72(1)
applicable to survivors of same-sex conjugal relationships during the July 2000
to June 2001 period. For those same-sex survivors who applied for survivors’
pensions during that time, they, like opposite-sex survivors, could be entitled
to up to 12 months of arrears depending upon when the contributor died.
63
The effect of s. 72(1) applying to same-sex survivors during the period
from July 2000 to June 2001 means that survivors who applied during that period
could be entitled to arrears, at most, back to August 1999. While the striking
down of s. 72(2) will grant access of up to 12 months of pension arrears, this
is a result of retroactive benefits which Parliament itself granted by enacting
s. 72(1).
64
We recognize that costs may be a factor in a s. 1 analysis. The
government says that July 2000 was selected in order to fix the liability of
the CPP as of that date. However, as of July 2000, the CPP had
to anticipate up to 12 months of arrears liability arising from opposite-sex
relationships. The government did not refer to evidence to suggest that cost
was the reason to deny same-sex survivors up to 12 months of benefit
arrears during the transitional period.
65
We are not persuaded that, in the circumstances of this case, s. 72(2)
is justified under s. 1 of the Charter . Access of up to 12 months of
arrears payments prior to July 2000 is provided by existing legislation, namely
s. 72(1) . Parliament itself has provided for limited retroactive benefits.
Where the issue is solely whether Charter relief claimants should be
entitled to the same retroactive benefits expressly available to their
comparator group, absent cost considerations, it is difficult to see how denial
of these benefits has a pressing and substantial objective. Here, there is an
absence of evidence of cost justifying the provision. In the circumstances, we
cannot find there is a rational connection between s. 72(2) and its objective,
or indeed, that s. 72(2) minimally impairs the Charter rights of the
Hislop class.
66
The government has failed to establish a s. 1 justification for s.
72(2).
C. Section 72(1)
67
Because we have found that s. 72(2) violates s. 15(1) and is not saved
by s. 1 of the Charter , same-sex survivors will be entitled to up
to 12 months of pension arrears for the period from August 1999 to July 2000,
pursuant to s. 72(1) .
68
However, the Hislop class takes issue with s. 72(1) itself. The Hislop
class argues that although s. 72(1) is facially neutral, its 12-month
limitation on pension arrears has an adverse effect on same-sex survivors. It
says that same-sex survivors were unable, prior to July 2000, to make a claim
for survivorship benefits. Based upon s. 44(1.1) and s. 72(2) being struck
down, it is argued that same-sex survivors should be entitled to claim
retroactive benefits from the time they became survivors after April 17, 1985
when s. 15(1) of the Charter came into force. For example, if an
individual became a survivor of a same-sex relationship in 1995, the survivor
should be entitled to a lump sum of benefits for the period from 1995 to the
time the application was received. The Hislop class argues that the operation
of s. 72(1) should be suspended for same-sex survivors to enable them to obtain
those retroactive benefits.
69
Although the Hislop class frames the s. 72(1) argument as an adverse
effect discrimination claim, the issue which the argument raises is, in fact,
one of remedy. What the Hislop class is seeking is retroactive Charter relief.
Their request for a constitutional exemption from the limitation on arrears in
s. 72(1) is, in effect, a request for a remedy in respect of their exclusion
from the survivors’ benefits by the pre-MBOA CPP between 1985 and
2000. As will be explained hereafter, this Court has been explicit in
restricting entitlement to retroactive Charter relief of this nature.
Because the remedy sought by the Hislop class is unavailable in any event, it
is not necessary to undertake a s. 15(1) analysis in respect of s. 72(1) .
D. Section 60(2)
70
Where a survivor entitled to a survivor’s CPP pension dies
without making application for that pension, the survivor’s estate may apply
and obtain the pension benefit to which the survivor would have been entitled,
provided the estate makes application within 12 months after the death of the
survivor. The Hislop class submits that, because some same-sex survivors had
been deceased for over 12 months when the MBOA amendments to the CPP came
into effect, their estates should be able to apply for the benefits to which
the survivors would have been entitled and that the 12-month limitation in s.
60(2) should be suspended so as not to bar such estate claims.
71
The threshold issue is whether the estates of those survivors who died
more than 12 months before the coming into force of the MBOA amendments
to the CPP may have standing to claim a s. 15(1) Charter right on
behalf of the deceased survivor. Only if they have such standing may the Court
even entertain an argument that s. 60(2) should not apply to such estates. The
Hislop class relies on R. v. Big M Drug Mart Ltd., [1985] 1 S.C.R. 295,
at pp. 322-23. Big M Drug Mart dealt with s. 2 of the Charter
which uses the term “[e]veryone”. The term used in s. 15(1) is more precise
and narrower, as it allows rights to “[e]very individual”.
72
The government submits, on the basis of the British Columbia Court of
Appeal judgment in Stinson Estate v. British Columbia (1999), 70
B.C.L.R. (3d) 233, 1999 BCCA 761, that s. 15(1) rights cannot be enforced by an
estate because those rights are personal and terminate with the death of the
affected individual. The government also submits that estates are not
individuals but artificial entities incapable of having their human dignity
infringed. In addition, the government relies on the Special Joint Committee
on the Constitution (see Minutes of Proceedings and Evidence of the Special
Joint Committee of the Senate and of the House of Commons on the Constitution
of Canada (1980-81), Issue No. 43, January 22, 1981, at
pp. 43:39-43:44; see also Issue No. 44, January 23, 1981, at pp.
44:6-44:10; Issue No. 47, January 28, 1981, at p. 47:88; and Issue No. 48,
January 29, 1981, at pp. 48:4-48:49), which substituted the words “every
individual” for “everyone” in s. 15(1) in response to the Minister of Justice’s
desire “to make it clear that this right would apply to natural persons only”
(p. 43:41). The government further argues that this Court has held that s.
15(1) rights could not be claimed by other entities such as corporations (see Edmonton
Journal v. Alberta (Attorney General), [1989] 2 S.C.R. 1326, at p. 1382, per
La Forest J.).
73
In our opinion, the government’s submissions have merit. In the context
in which the claim is made here, an estate is just a collection of assets and
liabilities of a person who has died. It is not an individual and it has no
dignity that may be infringed. The use of the term “individual” in s. 15(1)
was intentional. For these reasons, we conclude that estates do not have
standing to commence s. 15(1) Charter claims. In this sense, it may be
said that s. 15 rights die with the individual.
74
Mr. Hislop’s individual situation, however, is different. Although he
died between the time his notice of appeal was filed in this Court and the
hearing of this appeal, he obtained judgment while he was still alive.
75
When a judgment is obtained, the cause of action upon which the judgment
is based is merged in the judgment: Lew v. Lee, [1924] S.C.R. 612,
aff’d on this point [1925] A.C. 819 (P.C.); Reid v. Batty, [1933] O.W.N.
496 (H.C.J.), aff’d [1933] O.W.N. 817 (C.A.). In Lew, Anglin C.J.
explained that, because of the doctrine of merger, the issue in an appeal is
not the original cause of action but rather the legality and validity of the
judgment. As such, where a party dies pending appeal, the appeal survives even
if the original cause of action would not.
76
It should be noted that Anglin C.J. relied in part on a provision in the
British Columbia Supreme Court Rules (currently r. 15(2)), which
provided that whether a cause of action survives or not, the death of either
party between verdict or finding of the issues of fact and judgment will not
give rise to abatement and that judgment may be entered notwithstanding death.
He reasoned that, a fortiori, the right to enforce a judgment or defend
it on appeal must also survive. In our view, his analysis is applicable in the
instant case irrespective of any legislative provision. Although s. 15(1)
rights are personal, the constitutional issues raised here are issues of public
importance. Given the public interest in ensuring that questions of law
related to such rights be correctly decided, an appeal from a judgment raising
such issues must be allowed to survive the party’s death pending the appeal.
77
Although the preceding comments are sufficient to dispose of the issue
in relation to Mr. Hislop himself, because this is a class action, it is
appropriate to clarify with more precision the time at which s. 15(1) rights
crystallize. Merger, as we have explained, occurs when judgment is entered.
Nevertheless, it is a long-standing principle of law that a litigant should not
be prejudiced by an act of the court (actus curiae neminem gravabit): Turner
v. London and South-Western Railway Co. (1874), L.R. 17 Eq. 561. Based on
this principle, in cases where a plaintiff has died after the conclusion of
argument but before judgment was entered, courts have entered judgment nunc
pro tunc as of the date that argument concluded: see Gunn v. Harper (1902),
3 O.L.R. 693 (C.A.); Hubert v. DeCamillis (1963), 41 D.L.R. (2d) 495
(B.C.S.C.); Monahan v. Nelson (2000), 186 D.L.R. (4th) 193, 2000 BCCA
297. We affirm the correctness of this approach and conclude that the estate
of any class member who was alive on the date that argument concluded in the
Ontario Superior Court, and who otherwise met the requirements under the CPP ,
is entitled to the benefit of this judgment.
E. Remedies
78
We now turn to the question of remedy. In challenging s. 72(1) of the CPP ,
the appellants seek a fully retroactive remedy. They argue that a declaration
of invalidity under s. 52 of the Constitution Act, 1982 necessarily
operates back to the coming into force of s. 15 of the Charter . In
order to determine the validity of this claim, we must first consider the
nature of constitutional remedies and the circumstances under which courts may
limit a retroactive remedy and craft a prospective remedy. In the present
appeal, we conclude that a prospective remedy would be appropriate. A
retroactive remedy would be unwarranted in respect of s. 72(1) CPP . We
will also determine the appropriate remedies in respect of s. 44(1.1) and s.
72(2) CPP .
79
In substance, the position of the appellants is predicated on the
traditional — often called Blackstonian — view that judges never make law, but
merely discover it. In this perspective, the courts are said to apply the law
as it really was or has been rediscovered. As a consequence of the declaration
of nullity, the appellants claim that they are entitled to the full benefits of
the law, in conformity with an understanding of the Constitution, which is
deemed to have never changed.
80
The supremacy clause, now enshrined at s. 52 , is silent about the
remedies which may flow from a declaration of nullity. Does it mean that such
a declaration is always both prospective and retroactive? This does not appear
to have been the position of our Court throughout the incremental development
of the law of constitutional remedies after the adoption of the Charter . A
body of jurisprudence now accepts the legitimacy of limiting the retroactive
effect of a declaration of nullity and of fashioning prospective remedies in
appropriate circumstances.
(1) Retroactive and Prospective Remedies
Under the Charter
81
The Constitution empowers courts to issue constitutional remedies with both
retroactive and prospective effects: see, e.g., Schachter v. Canada,
[1992] 2 S.C.R. 679, at p. 719. Section 24(1) of the Charter enables
individuals who have had their Charter rights violated to seek redress
for those past wrongs and “obtain such remedy as the court considers
appropriate and just”. Section 24(1) may also, in some situations, enable the
claimant to recover damages, which are necessarily retroactive: Schachter,
at pp. 725-26.
82
Section 52(1) instructs courts to declare unconstitutional legislation
of no force or effect. When a court issues a declaration of invalidity, it
declares that, henceforth, the unconstitutional law cannot be enforced. The
nullification of a law is thus prospective. However, s. 52(1) may also operate
retroactively so far as the parties are concerned, reaching into the past to
annul the effects of the unconstitutional law: see, e.g., Miron v. Trudel,
[1995] 2 S.C.R. 418.
83
This Court has applied in many cases the “declaratory approach” to
constitutional remedies, which implies that s. 52(1) remedies are often given
retroactive effect. See, for example, Nova Scotia (Workers’ Compensation
Board) v. Martin, [2003] 2 S.C.R. 504, 2003 SCC 54, at para. 28,
Gonthier J. On this view, s. 52(1) remedies are deemed to be fully retroactive
because the legislature never had the authority to enact an unconstitutional
law. In the words of Professor Hogg, a declaration of constitutional invalidity
“involves the nullification of the law from the outset” (P. W.
Hogg, Constitutional Law of Canada (loose-leaf ed.), vol. 2, at p. 55-2
(emphasis added)). If the law was invalid from the outset, then any government
action taken pursuant to that law is also invalid, and consequently, those
affected by it have a right to redress which reaches back into the past.
84
As mentioned above, the declaratory approach is derived from
Blackstone’s famous aphorism that judges do not create law but merely discover
it: W. Blackstone, Commentaries on the Laws of England (1765), vol. 1,
at pp. 69-70. It reflects a traditional and widespread understanding of the
role of the judiciary in a democratic state governed by strong principles of
separation of powers between courts, legislatures and executives. In this
perspective, courts grant retroactive relief applying existing law or
rediscovered rules which are deemed to have always existed. On the other hand,
legislators fashion new laws for the future.
85
Blackstone’s declaratory approach has not remained unchallenged in modern
law. Commentators and courts have pointed out that judges fulfill a legitimate
law-making function. Judges do not merely declare law; they also make law.
These critics argue that Blackstone’s view is a fiction as judges make law,
especially in the common law world. See, e.g., Lord Reid, “The Judge as Law
Maker” (1972-1973), 12 J.S.P.T.L. 22. They say such a fiction should
not be turned into an ironclad principle.
86
However, this acknowledgement does not require abandoning Blackstone’s
declaratory approach altogether. The critique of the Blackstonian approach
applies only to situations in which judges are fashioning new legal rules or
principles and not when they are applying the existing law. In instances where
courts apply pre-existing legal doctrine to a new set of facts, Blackstone’s
declaratory approach remains appropriate and remedies are necessarily
retroactive. Because courts are adjudicative bodies that, in the usual course
of things, are called upon to decide the legal consequences of past happenings,
they generally grant remedies that are retroactive to the extent necessary to
ensure that successful litigants will have the benefit of the ruling: see S.
Choudhry and K. Roach, “Putting the Past Behind Us? Prospective Judicial and
Legislative Constitutional Remedies” (2003), 21 S.C.L.R. (2d) 205, at
pp. 211 and 218. There is, however, an important difference between saying
that judicial decisions are generally retroactive and that they are necessarily
retroactive. When the law changes through judicial intervention, courts
operate outside of the Blackstonian paradigm. In those situations, it may be
appropriate for the court to issue a prospective rather than a retroactive
remedy. The question then becomes what kind of change and which conditions
will justify the crafting of judicial prospective remedies.
87
The House of Lords recently adopted this view in the course of its
discussion of prospective overruling: In re Spectrum Plus Ltd. (in
liquidation), [2005] 2 A.C. 680, [2005] UHKL 41. The words of Lord
Nicholls at para. 34 are particularly apt:
[Blackstone’s declaratory] theory is still valid when applied to cases
where a previous decision is overruled as wrong when given. Most overruling
occurs on this basis. These cases are to be contrasted with [those] where the
later decision represents a response to changes in social conditions and
expectations. Then, on any view, the declaratory approach is inapt. In this
context [this] approach has long been discarded. It is at odds with reality.
88
Although Lord Nicholls’ statement arose in the common law context, this
Court has implicitly adopted a similar line of reasoning in constitutional
law. Despite this Court’s endorsement of the Blackstonian declaratory
approach, in its development of the law of constitutional remedies, it has
frequently seen fit to temper the retroactive effect of s. 52(1) remedies and
adopt a position similar to that of Lord Nicholls. On occasion, this Court has
expressly stated that the s. 52(1) remedy would “apply prospectively” only: Reference
re Remuneration of Judges of the Provincial Court of Prince Edward Island,
[1998] 1 S.C.R. 3, at para. 18. In that reference, the Court held that the
provinces’ remuneration scheme for provincial judges violated the right to a
trial before an independent and impartial tribunal. In fashioning its remedy,
the Court side-stepped Blackstonian doctrine and provided for a “transition
period of one year before th[e] requirement [took] effect” (para. 18).
89
The use of the “transition period” after a finding of
unconstitutionality is inconsistent with the declaratory approach. It keeps in
force by judicial decision a law that was deemed never to have been of force or
effect. Thus, the fact that this Court has occasionally prescribed transition
periods, intended to give a s. 52(1) remedy prospective effect only, suggests
that this Court is not wedded to the declaratory approach in all Charter
cases.
90
In another type of situation, which arises more frequently, the Court
has held that providing immediate and retroactive judicial remedies may be
“inappropriate” when “doing so would create a lacuna in the regime before
Parliament would have a chance to act”: R. v. Demers, [2004] 2 S.C.R.
489, 2004 SCC 46, at para. 57. In such cases, the Court has temporarily
suspended the declaration of invalidity of the unconstitutional legislation to
avoid creating a “legal vacuum” or “legal chaos” before Parliament or the
legislature has the opportunity to enact something in place of the
unconstitutional legislation: Reference re Manitoba Language Rights,
[1985] 1 S.C.R. 721 (“Manitoba Language Rights Reference”), at p. 747; Schachter.
In Schachter, this Court held that the suspended declaration of
invalidity was appropriate when giving immediate retroactive effect to the
Court’s declaration of invalidity would (a) “pose a danger to the public”; (b)
“threaten the rule of law”; or (c) “result in the deprivation of benefits from
deserving persons”, such as when the legislation was “deemed unconstitutional
because of underinclusiveness rather than overbreadth”: Schachter, at p.
719.
91
Like transition periods and other purely prospective remedies, the
suspended declaration of invalidity is not fully consistent with the
declaratory approach. By suspending the declaration of invalidity, the Court
allows the constitutional infirmity to continue temporarily so that the
legislature can fix the problem. In other words, the Court extends the life of
a law which, on the Blackstonian view, never existed.
92
Although if the legislature fails to comply with the Court’s order
within the period of suspension, the Court’s declaration would apply
retroactively, the purpose of a suspended declaration of invalidity can be to
facilitate the legislature’s function in crafting a prospective remedy. The
temporal delay in striking down the law also has the effect of extending the
life of an unconstitutional law. In such cases, to allow the claimants to
recover concurrent retroactive relief would be at cross-purposes with the
Court’s decision to grant a suspended declaration of invalidity: Schachter,
at p. 720.
93
The determination of whether to limit the retroactive effect of a s.
52(1) remedy and grant a purely prospective remedy will be largely determined
by whether the Court is operating inside or outside the Blackstonian paradigm.
When the Court is declaring the law as it has existed, then the Blackstonian
approach is appropriate and retroactive relief should be granted. On the other
hand, when a court is developing new law within the broad confines of the
Constitution, it may be appropriate to limit the retroactive effect of its
judgment.
94
The approach which our Court has adopted in respect of the crafting of
constitutional remedies also flows from its understanding of the process of
constitutional interpretation, which the “living tree” metaphor neatly
describes. From the time Lord Sankey L.C. used these words to characterize the
nature of the Canadian Constitution, courts have relied on this expression to
emphasize the ability of the Constitution to develop with our country (Edwards
v. Attorney-General for Canada, [1930] A.C. 124 (P.C.), at p. 136).
This Court has often stated that the Canadian Constitution should not be viewed
as a static document but as an instrument capable of adapting with the times by
way of a process of evolutionary interpretation, within the natural limits of
the text, which “accommodates and addresses the realities of modern life”:
Reference re Same-Sex Marriage, [2004] 3 S.C.R. 698, 2004 SCC 79, at
para. 22; see also Attorney General of Quebec v. Blaikie, [1979] 2
S.C.R. 1016, at p. 1029; Re Residential Tenancies Act, 1979, [1981] 1
S.C.R. 714, at p. 723; Law Society of Upper Canada v. Skapinker, [1984]
1 S.C.R. 357, at p. 365; Hunter v. Southam Inc., [1984] 2 S.C.R.
145, at p. 155.
95
It is true that the “living tree” doctrine is not wedded to a particular
model of the judicial function. At times, its application may reflect the fact
that, in a case, the Court is merely declaring the law of the country as it has
stood and that a retroactive remedy is then generally appropriate. In other
circumstances, its use recognizes that the law has changed, that the change
must be acknowledged and that, from a given point in time, the new law or the
new understanding of some legal principle will prevail.
96
The question is no longer the legitimacy of prospective remedies, but
rather when, why and how judges may rule prospectively or restrict the
retroactive effect of their decisions in constitutional matters. The key
question becomes the nature and effect of the legal change at issue in order to
determine whether a prospective remedy is appropriate. The legitimacy of its
use turns on the answer to this question.
97
There must be change, but what kind of change will be enough? Given the
often incremental nature of changes in judge-made law in a common law system,
the question is bound to raise difficulties. Various formulas have been
suggested. For example, at a time when it appeared to be moving towards a
broad recognition of prospective overruling, the U.S. Supreme Court relied on a
notion of a “clear break with the past”. The limitation on proactive remedies
could be considered after litigants had established such a change (see United
States v. Johnson, 457 U.S. 537 (1982), at p. 549).
98
On other occasions, the same court crafted different formulas. In an
earlier case, in 1971, it relied on a test which tended to look at both the
nature of the change and of some of its consequences (Chevron Oil Co. v.
Huson, 404 U.S. 97 (1971), at pp. 106‑7). It identified three relevant
considerations:
First, the decision to be applied nonretroactively must establish a new
principle of law, either by overruling clear past precedent on which litigants
may have relied . . . or by deciding an issue of first impression whose resolution
was not clearly foreshadowed. . . . Second, it has been stressed that “we must
. . . weigh the merits and demerits in each case by looking to the
prior history of the rule in question, its purpose and effect, and whether
retrospective operation will further or retard its operation.” . . . Finally,
we have weighed the inequity imposed by retroactive application, for “[w]here a
decision of this Court could produce substantial inequitable results if applied
retroactively, there is ample basis in our cases for avoiding the ‘injustice or
hardship’ by a holding of nonretroactivity.”
See also: J.
E. Fisch, “Retroactivity and Legal Change: An Equilibrium Approach” (1997), 110
Harv. L. Rev. 1055, at pp. 1060-63; C. Sampford, Retrospectivity and
the Rule of Law (2006), at pp. 211-12.
99
Change in the law occurs in many ways. “Clear break with the past”
catches some of its diversity. It can be best identified with those situations
where, in Canadian law, the Supreme Court departs from its own jurisprudence by
expressly overruling or implicitly repudiating a prior decision. Such clear
situations would justify recourse to prospective remedies in a proper context.
But other forms of substantial change may be as relevant, especially in
constitutional adjudication, where courts must give content to broad, but
previously undefined, rights, principles or norms. The definition of a yet
undetermined standard or the recognition that a situation is now covered by a
constitutional guarantee also often expresses a substantial change in the law.
The right may have been there, but it finds an expression in a new or newly
recognized technological or social environment. Such a legal response to these
developments properly grounds the use of prospective remedies, when the
appropriate circumstances are met. A substantial change in the law is
necessary, not sufficient, to justify purely prospective remedies. Hence, we
must now turn to what else must be considered once legal change has been
established.
100
Although the list of such factors should not be considered as closed,
some of them appear more clearly compelling. They may include reasonable or in
good faith reliance by governments (Miron, at para. 173; Mackin v.
New Brunswick (Minister of Finance), [2002] 1 S.C.R. 405, 2002 SCC 13, at
para. 78), or the fairness of the limitation of the retroactivity of the remedy
to the litigants. Courts ought also consider whether a retroactive remedy
would unduly interfere with the constitutional role of legislatures and democratic
governments in the allocation of public resources (Benner, at
para. 103; Schachter, at p. 710).
101
A careful consideration of reliance interests is critical to this
analytical process. Although legal mechanisms, such as the de facto doctrine,
res judicata or the law of limitations, may mitigate the
consequences of declaratory rulings in certain circumstances, they do not
address every situation. Fully retroactive remedies might prove highly
disruptive in respect of government action, which, on the basis of settled or
broadly held views of the law as it stood, framed budgets or attempted to
design social programs. Persons and public authorities could then become
liable under a new legal norm. Neither governments nor citizens could be reasonably
assured of the legal consequences of their actions at the time they are taken.
102
The strict declaratory approach also hardly appears reconcilable with
the well-established doctrine of qualified immunity in respect of the adoption
of unconstitutional statutes which our Court applied, for example, in cases
such as Mackin and Guimond v. Quebec (Attorney General), [1996] 3
S.C.R. 347. Where legislation is found to be invalid as a result of a judicial
shift in the law, it will not generally be appropriate to impose liability on
the government. As Gonthier J. wrote in Mackin, it is a general
rule of public law that “absent conduct that is clearly wrong, in bad faith or
an abuse of power, the courts will not award damages for the harm suffered as a
result of the mere enactment or application of a law that is subsequently
declared to be unconstitutional” (para. 78). The rationale for this qualified
immunity, which applies equally to actions for damages based on the general law
of civil liability and to claims for damages under s. 24(1) of the Charter ,
was aptly expressed by Gonthier J.:
Thus, the government and its representatives are required to exercise
their powers in good faith and to respect the “established and indisputable”
laws that define the constitutional rights of individuals. However, if they
act in good faith and without abusing their power under prevailing law and only
subsequently are their acts found to be unconstitutional, they will not be
liable. Otherwise, the effectiveness and efficiency of government action would
be excessively constrained. Laws must be given their full force and effect as
long as they are not declared invalid. Thus it is only in the event of conduct
that is clearly wrong, in bad faith or an abuse of power that damages may be
awarded. [para. 79]
The same
principles will apply in respect of claims for retroactive benefits under s. 15
of the Charter . Whether framed as a remedy under s. 52 or s. 24(1) , it
may be tantamount to a claim for compensatory damages flowing from the
underinclusiveness of the legislation.
103
People generally conduct their affairs based on their understanding of
what the law requires. Governments in this country are no different. Every
law they pass or administrative action they take must be performed with an eye
to what the Constitution requires. Just as ignorance of the law is no excuse
for an individual who breaks the law, ignorance of the Constitution is no
excuse for governments. But where a judicial ruling changes the existing law
or creates new law, it may, under certain conditions, be inappropriate to hold
the government retroactively liable. An approach to constitutional
interpretation that makes it possible to identify, in appropriate cases, a
point in time when the law changed, makes it easier to ensure that persons and
legislatures who relied on the former legal rule while it prevailed will be
protected. In this way, a balance is struck between the legitimate reliance
interests of actors who make decisions based on a reasonable assessment of the
state of the law at the relevant time on one hand and the need to allow
constitutional jurisprudence to evolve over time on the other.
104
Having regard to the above-mentioned criteria, it is possible to
distinguish this case from some cases where fully retroactive remedies were
granted. Miron provides an example of when it would not be
appropriate for courts to limit the retroactive effect of a s. 52(1) remedy.
In Miron, the appellant was injured while a passenger in a vehicle
driven by an uninsured driver. He made a claim for accident benefits against
the insurance policy of his unmarried partner but his claim was denied on the
basis that the policy covered only legally married spouses. Writing for the
majority, McLachlin J. (as she then was) held that the distinction based on
marital status was discriminatory under s. 15(1) of the Charter . She
concluded that retroactive reading-up of the legislation was an appropriate
remedy, which entitled the appellant to the retroactive benefit of his
partner’s insurance policy.
105
In Miron, it would not have been appropriate for the Court
to limit the retroactive effect of the remedy and grant a purely prospective
remedy. First, the government did not meet the threshold factor of showing a
substantial change from the existing law. As early as 1980, the Ontario
Legislature was able to agree on a formula to extend death benefits to certain
unmarried persons. And in 1981, in the context of the Ontario Human Rights
Code, the Legislature agreed on a definition of “spouse” as the person to
whom a person of the opposite sex is married or with whom the person is living
in a conjugal relationship outside marriage. In other words, Ontario’s vehicle
insurance legislation was out of step with the evolving understanding of
“spouse” as it existed in other Ontario statutes. Therefore, the Court’s
holding in Miron — that the vehicle insurance legislation’s definition
of “spouse” violated s. 15 — was not a substantial change from the existing
law. To the contrary, it reflected an understanding of s. 15 of the Charter
that was already understood by the Ontario Legislature in the context of
the Ontario Human Rights Code and other provincial legislation. Because
the finding of a s. 15 infringement in Miron did not represent a
substantial change in the law, it would have been inappropriate to limit the
retroactive effect of its decision.
106
However, even if the government had succeeded in meeting the substantial
change requirement, other factors militated against limiting the retroactive
effect of the remedy. In reaching her conclusion, McLachlin J. drew support
from three observations. First, she observed that the legislature had, since
the accident occurred, amended the applicable legislation to include unmarried
partners, thus allaying any concerns about interfering unduly with legislative
objectives. The amended legislation provided “the best possible evidence of
what the Legislature would have done had it been forced to face the problem the
appellants raise[d]” (para. 180). Second, considerations of fairness to the
successful litigant also weighed in favour of retroactivity, as providing a
retroactive remedy in this case was the only means of “cur[ing] an injustice
which might otherwise go unremedied”: ibid. Third, McLachlin J.
noted that the distinction based on marital status was unreasonable, even at
the time the impugned legislation was enacted (para. 173). Because
the Legislature ought to have known that the vehicle insurance legislation was
out of step with a modern understanding of “spouse”, it could not reasonably
exclude common law spouses from insurance coverage.
107
It should be noted that, in Miron, all of the factors discussed
above — good faith reliance by governments, fairness to the litigants and the
need to respect the constitutional role of legislatures — favoured a
retroactive remedy. In a number of cases, however, these factors may pull in
different directions, with some factors favouring a retroactive remedy and
others favouring a purely prospective remedy. In such cases, once the
“substantial change” threshold criterion is met, it may be appropriate to limit
the retroactive effect of the remedy based on a balancing of these other
factors. This balance must be struck on a case-by-case basis.
108
A second situation that must be distinguished was considered by this
Court in its recent judgment in Kingstreet Investments Ltd. v. New Brunswick
(Finance), [2007] 1 S.C.R. 3, 2007 SCC 1, wherein it held that taxes
collected pursuant to an ultra vires regulation are recoverable by the
taxpayer. The difference between the result in Kingstreet and the type
of situation in the present case may be understood in terms of a basic
distinction between cases involving moneys collected by the government and
benefits cases. Where the government has collected taxes in violation of the
Constitution, there can be only one possible remedy: restitution to the
taxpayer. In contrast, where a scheme for benefits falls foul of the s. 15
guarantee of equal benefit under the law, we normally do not know what the
legislature would have done had it known that its benefits scheme failed to
comply with the Charter . In benefits cases, a range of options is open
to government. The excluded group could simply be included in the existing
benefit scheme as was the result in Tétreault-Gadoury v. Canada (Employment
and Immigration Commission), [1991] 2 S.C.R. 22. It could also be
included in a modified benefit scheme, adopted by legislative amendments, as
occurred in Schachter. Also, in Schachter, the Court alluded to
the possibility of an elimination of the benefit (p. 702). In our political
system, choosing between those options remains the domain of governments. This
principle points towards limiting the retroactive effect of remedies in s. 15 benefits
cases in which the other above-mentioned criteria are met.
(2) The Appropriate Remedy in This Case
(a) Limits on the Retroactive Effect of the
Remedy in the Context of This Case
109
Same-sex equality jurisprudence since 1985 is illustrative of the sort
of legal shift that gives rise to new law and justifies consideration of
prospective remedies. The factors mentioned above also weigh in favour of
limiting the retroactive effect of the remedy in the context of this case.
(i) The Substantial Change in the Law
110
This Court’s decision in M. v. H. marked a departure from
pre-existing jurisprudence on same-sex equality rights. In 1995, a majority of
this Court upheld the exclusion of same-sex partners from old age security
legislation in Egan, with four judges finding no s. 15(1) violation, and
one judge concluding that the scheme was contrary to s. 15(1) but that it could
be justified under s. 1 . Four years later in M. v. H., eight members of
this Court held that the exclusion of same-sex partners from the spousal
support provisions under the Family Law Act was contrary to s. 15(1) and
could not be saved under s. 1 . M. v. H. thus marks a clear shift in the
jurisprudence of the Court, where it moved away from the plurality’s holding in
Egan and came to a new understanding of the scope of equality rights.
111
Bastarache J. disagrees with our conclusion on the nature of the change
brought about by M. v. H. He cites lower court decisions rendered prior
to Egan and before M. v. H. to show that the law on same-sex
equality rights remained unsettled until M. v. H. However, in our
system, the Supreme Court has the final word on the interpretation of the
Constitution: Manitoba Language Rights Reference, at
p. 745. A majority of the Court in Egan rejected the appellants’
claim for equal benefits under the law. It was a fact that this Court held in Egan
that the Constitution did not require equal benefits for same-sex couples. This
fact changed only after M. v. H. when this Court held that it was
unconstitutional to exclude same-sex couples from the definition of spouse in
the Family Law Act. The threshold requirement for limiting the
retroactive effect of the remedy has been satisfied. The Court must now
consider other relevant factors. In this case, reliance interests, fairness
concerns, the government’s good faith, and the need to respect Parliament’s
legislative role all weigh in favour of limiting retroactive relief.
(ii) Reasonable Reliance
112
Given the state of the jurisprudence prior to M. v. H., the
exclusion of same-sex partners from the former CPP was based on a
reasonable understanding of the state of s. 15(1) jurisprudence as it existed
after Egan and before M. v. H. Admittedly, the Court in Egan was
divided over whether to extend old age security benefits to same-sex couples,
with four judges dissenting. After M. v. H., it became apparent that
the Egan dissent had prevailed. However, the benefit of hindsight does
not undermine the government’s reasonableness in relying on Egan.
113
In holding that the government reasonably relied on the pre-M. v. H.
jurisprudence, we do not seek to justify the slowness of legislatures and
courts alike in recognizing Charter rights. Rather, we acknowledge the
fact that although the Constitution embodies the supreme law and the enduring
values of this country, it is up to the courts to interpret and apply those
provisions. In Manitoba Language Rights Reference, this Court
held:
The Constitution of a country is a statement of the
will of the people to be governed in accordance with certain principles held as
fundamental and certain prescriptions restrictive of the powers of the
legislature and government. It is, as s. 52 of the Constitution Act, 1982 declares,
the “supreme law” of the nation, unalterable by the normal legislative process,
and unsuffering of laws inconsistent with it. The duty of the judiciary is to
interpret and apply the laws of Canada and each of the provinces, and it is
thus our duty to ensure that the constitutional law prevails. [p. 745]
114
The text of the Constitution establishes the broad confines of the
supreme law, but it is up to the courts to interpret and apply the Constitution
in any given context. The inviolability of the Constitution ensures that our
nation’s most cherished values are preserved, while the role of the courts in
applying the Constitution ensures that the law is sufficiently flexible to
change over time to reflect advances in human understanding. But it also
means that the Constitution, at any snapshot in time, is only as robust as the
court interpreting it. If the judiciary errs or is slow to recognize that
previous interpretations of the Constitution no longer correspond to social
realities, it must change the law. However, in breaking with the past, the
Court does not create an automatic right to redress for the Court’s prior
ruling. Where the government’s reliance on the unconstitutional law was
reasonable because it was relying on this Court’s jurisprudence, it will be
less likely that a right to retroactive relief will flow from a subsequent
declaration of invalidity of the unconstitutional law.
(iii) Good Faith
115
Our comments above indicate that the government did not act in bad faith
in failing to extend survivors’ benefits to same-sex couples prior to M. v.
H. It is significant that the survivors’ benefit scheme under the former CPP
was never struck down by a court of competent jurisdiction. Rather,
recognizing the likely implications of this Court’s ruling in M. v. H.
for that scheme, Parliament endeavoured to pre-emptively correct the
constitutional deficiencies therein by enacting remedial legislation. Because
the government acted in good faith by attempting pre-emptively to correct a
constitutional infirmity soon after it was discovered, it would be
inappropriate to reach back further in time and impose a retroactive remedy.
(iv) Fairness to Litigants
116
In seeking payment of arrears back as far as 1985, the Hislop class
effectively asks this Court to overlook the evolution in the jurisprudence of
same-sex equality rights that has taken place and to declare that the
understanding to which we have come over that period of time was in fact the
law in 1985. This position cannot be sustained. Although M. v. H. declares
what the Constitution requires, it does not give rise to an automatic right to
every government benefit that might have been paid out had the Court always
interpreted the Constitution in accordance with its present-day understanding
of it. M. v. H. was not a case like Miron where limiting the retroactive
effect of the s. 52(1) remedy would have granted the “successful” claimant a
hollow victory. In contrast, a purely prospective remedy in M. v. H.
was not meaningless. M. v. H. resulted in wide-scale amendments to
federal and provincial legislation across the country to extend government
benefits to same-sex couples. Equally important, M. v. H. helped usher
in a new era of understanding of the equal human dignity of same-sex couples.
One could not say that M. v. H. granted those litigants only a Pyrrhic
victory.
(v) Respecting Parliament’s Role
117
Achieving an appropriate balance between fairness to individual
litigants and respecting the legislative role of Parliament may mean that Charter
remedies will be directed more toward government action in the future and less
toward the correction of past wrongs. In the present case, the Hislop class’s
claim for a retroactive remedy is tantamount to a claim for compensatory
damages flowing from the underinclusiveness of the former CPP . Imposing
that sort of liability on the government, absent bad faith, unreasonable
reliance or conduct that is clearly wrong, would undermine the important
balance between the protection of constitutional rights and the need for
effective government that is struck by the general rule of qualified immunity.
A retroactive remedy in the instant case would encroach unduly on the
inherently legislative domain of the distribution of government resources and
of policy making in respect of this process.
(vi) Conclusion
118
For the foregoing reasons, the retroactive relief sought by the Hislop
class is unavailable under the law applicable to constitutional remedies. It
is not therefore necessary to carry out a s. 15(1) analysis in respect of s.
72(1) .
(b) Remedies Arising From the Specific
Provisions: Sections 44(1.1) and 72(2)
119
We turn now to a consideration of the appropriate remedy for the
specific constitutional violations that have been identified in this case.
120
Writing for a majority of the Court in Schachter, Lamer C.J.
explained that three questions must be answered when s. 52 of the Constitution
Act, 1982 is engaged: (1) what is the extent of the inconsistency between
the impugned provision and the Charter ; (2) can that inconsistency be
dealt with alone, by way of severance or reading in, or is it too inextricably
linked to other parts of the legislation; and (3) should a declaration of
invalidity be temporarily suspended?
121
In the present case, ss. 44(1.1) and 72(2) , although found within
remedial legislation, restrict the availability of that legislation to a
marginalized group. The extent of the inconsistency with the equality
guarantee is co-extensive with the scope of these two provisions. As such, the
inconsistency can be cured without distorting or interfering with the rest of
the legislative scheme. A declaration that ss. 44(1.1) and 72(2) are of no
force and effect is in keeping with the scheme and obvious purpose of the MBOA
to extend the survivors’ benefit to same-sex survivors. Finally, a temporary
suspension of the declaration of invalidity is not appropriate in the present
case. As Lamer C.J. noted in Schachter, at p. 716, such suspensions are
“serious matter[s] from the point of view of the enforcement of the Charter ”
because they allow an unconstitutional state of affairs to persist.
Suspensions should only be used where striking down the legislation without
enacting something in its place would pose a danger to the public, threaten the
rule of law or where it would result in the deprivation of benefits from
deserving persons without benefiting the rights claimant (p. 719). None of
these factors are present in the case at bar.
122
Throughout these proceedings, the Crown has taken the position that the
specific provisions are constitutionally unassailable because they merely
ensure that the remedial provisions in the MBOA apply prospectively
(i.e., not retroactively) from the time of the enactment. Further, the Crown
has relied on the argument that same-sex equality rights, as they are
understood today, are not what they were in 1985. In effect, it is the Crown’s
position that striking down the specific provisions would be tantamount to
applying today’s legal understanding of equality rights to past situations.
123
As we have explained above, the specific provisions violate s. 15(1) and
cannot be justified under s. 1 . The constitutional analysis related to these
provisions does not depend on any particular conception of the equality rights
of same-sex survivors prior to the enactment of the MBOA . The specific
provisions, as applied to same-sex survivors today, are discriminatory.
124
Any concerns about retroactivity in respect of s. 44(1.1) are misplaced
because they mistakenly construe the survivor’s benefit as a payment in respect
of a past event, namely the death of the survivor’s spouse or common law
partner, rather than the ongoing status of being a survivor. In our view, the
principles articulated by this Court in Benner are a complete answer to
the Crown’s s. 44(1.1) argument.
125
Benner involved a challenge to provisions of the Canadian
Citizenship Act that afforded different treatment to applicants for
Canadian citizenship, where those applicants were born before February 14,
1977, based on whether they were born of Canadian mothers or Canadian fathers.
Children of Canadian fathers were automatically entitled to Canadian
citizenship upon registration of their birth, while children of Canadian
mothers were required to apply for citizenship, swear an oath and pass a
security check. Benner challenged the legislative distinction on s. 15(1)
grounds, while the Crown resisted the claim in part on the basis that Benner’s claim
involved an impermissible retroactive application of the Charter to
events that had taken place before the Charter came into force. Writing
for a unanimous court, Iacobucci J. rejected the Crown’s argument and, in so
doing, provided a useful clarification of the concepts of retroactivity,
retrospectivity and prospectivity.
126
It should be noted that Benner engaged questions related to the
retroactive application of the Charter , while the instant case raises
questions related to the retroactive application of ordinary legislation and to
the availability of retroactive remedies for unconstitutional legislation.
Although retroactivity raises different concerns in the two contexts, the basic
conceptual distinctions between retroactivity, retrospectivity and
prospectivity remain. As such, Iacobucci J.’s elucidation of these concepts is
directly relevant to the present case.
127
First, Iacobucci J. addressed the difference between retroactivity and
retrospectivity, citing, at para. 39, E. A. Driedger’s explanation from
“Statutes: Retroactive Retrospective Reflections” (1978), 56 Can. Bar Rev. 264,
at pp. 268‑69:
A retroactive statute is one that operates as of a
time prior to its enactment. A retrospective statute is one that operates for
the future only. It is prospective, but it imposes new results in respect of a
past event. A retroactive statute operates backwards. A retrospective
statute operates forwards, but it looks backwards in that it attaches
new consequences for the future to an event that took place before the
statute was enacted. A retroactive statute changes the law from what it was; a
retrospective statute changes the law from what it otherwise would be with
respect to a prior event. [Emphasis in original.]
Next,
Iacobucci J. turned to the difference between retrospectivity and
prospectivity, again citing Professor Driedger. He reproduced the following
passage from Driedger’s Construction of Statutes (2nd ed. 1983), at p.
192, at para. 42 of his reasons:
These past facts may describe a status or
characteristic, or they may describe an event. It is submitted that where
the fact‑situation is a status or characteristic (the being something),
the enactment is not given retrospective effect when it is applied to persons
or things that acquired that status or characteristic before the enactment, if
they have it when the enactment comes into force; but where the fact‑situation
is an event (the happening of or the becoming something), then the enactment
would be given retrospective effect if it is applied so as to attach a new
duty, penalty or disability to an event that took place before the enactment.
[Emphasis added.]
It is true, as
Iacobucci J. observed, that the distinction between situations involving
discrete events and those involving ongoing conditions is not always clear.
For example, while the death of a spouse or common law partner is an event, the
fact of “being a survivor” is an ongoing condition. The challenge lies in
determining whether, in all the circumstances, a particular legislative scheme
relates primarily to the past event or the current condition resulting from it
(para. 46).
128
On the facts before the Court in Benner, Iacobucci J. concluded
that the appellant’s claim did not involve a retrospective application of the Charter
because it was a claim flowing from Benner’s ongoing status as the child
born outside Canada of a Canadian mother before the specified date. The date
on which Benner first acquired that status was, in Iacobucci J.’s view,
immaterial.
129
The same reasoning is controlling in the present case. As the Court of
Appeal found, the purpose of the CPP is “to provide a secure
government-controlled pension for retired persons in their senior years, and
thereafter to provide for their surviving spouses and partners” (para. 55).
The basis for the survivor’s benefit is the continuing status of being a
survivor. As Iacobucci J. observed in Benner, at para. 56, “the
important point is not the moment at which the individual acquires the status
in question, it is the moment at which that status is held against him or
disentitles him to a benefit”. There is, as such, no merit to the government’s
argument that striking down s. 44(1.1) would be tantamount to requiring
retroactive remedies for Charter violations. The remedy for the class
members in relation to s. 44(1.1) is prospective in that it entails granting
equal benefit of the law, prospectively, to people who are survivors of same‑sex
partners.
130
The government raises the additional argument that the Court should not
strike down s. 44(1.1) because doing so would have the unintended effect of
excluding same‑sex survivors from eligibility for the survivor’s
benefit. The Crown’s argument is predicated on the view that the survivor’s
benefit is paid in respect of a past event such that any entitlement to it
crystallizes at the time that the survivor’s partner dies and that, but for s.
44(1.1) , no benefit would be payable to any survivor whose same‑sex
common‑law partner died before July 2000. As we have stated, this is an
untenable view of what the survivor’s benefit is and how entitlement to it
works. A close look at the July 2000 amendments to the CPP reveals
that s. 44(1.1) is in fact a limiting provision, such that simply striking it
down leaves a coherent provision that provides for equal treatment, in terms of
eligibility, of opposite‑sex survivors and same‑sex survivors.
131
As part of the 2000 amendments, the opposite-sex definition of “spouse”
in s. 2(1) of the CPP was repealed. In its place, a definition for
“survivor” that includes both opposite-sex and same-sex spouses was added to
Part II, “Pensions and Supplementary Benefits”, in s. 42(1) . Finally,
s. 44(1)(d), which sets out the basic entitlement to the survivor’s
benefit, was amended by replacing reference to the statutory definition of
“spouse” with reference to the new statutory definition of “survivor”. The
amended provision provides that “subject to subsection (1.1), a survivor’s
pension shall be paid to the survivor of a deceased contributor . . .”.
132
Once eligibility is understood in terms of the current status of being a
survivor, it becomes clear that the effect of the amendments listed above is to
extend entitlement to the survivor’s benefit under s. 44(1)(d) to same‑sex
survivors. The function of s. 44(1.1) is purely restrictive: it limits that
entitlement based on the date on which the applicant became a survivor. As
such, striking down s. 44(1.1) has no unintended or undesirable effect.
133
The analysis in relation to s. 72(2) is somewhat different. Prior to
the enactment of the MBOA , it is true that same-sex survivors had no
right to the survivor’s benefit. Striking down s. 72(2) admittedly alters
the legal consequences of having been a survivor (a past situation) in the 12
months preceding July 2000, when the MBOA came into force. To the
extent that this involves a retroactive change in the law, it flows necessarily
from the fact that Parliament has included in the CPP a right to up to
12 months of arrears. It is clearly open to Parliament to legislate
retroactively, which it has done in s. 72(1) . The Charter simply
requires that same-sex survivors receive equal treatment as their opposite-sex
counterparts. To the extent that s. 72(2) withholds from same-sex survivors a
right to arrears to which similarly situated opposite-sex survivors are
entitled, it is of no force and effect.
134
In conclusion, class members who were precluded by s. 44(1.1) or s.
72(2) from receiving the survivor’s benefit, and who otherwise meet the
eligibility requirements, will be entitled to payment of that benefit. In the
circumstances, the relevant date for the purpose of that payment is the date on
which application was received or where no application was made because of the
unconstitutional provisions, the date on which the statement of claim was
filed. In no event are benefits payable in respect of a month prior to August
1999, which is the earliest month in respect of which a class member who
applied for the survivor’s benefit on the day the MBOA came into force
could have been eligible.
VII. Interest
135
For the reasons given by the Court of Appeal, we reject the government’s
contention that pre-judgment interest is not available in the instant case. In
short, s. 31 of the Crown Liability and Proceedings Act provides that,
subject to an express provision in another enactment, successful litigants are
entitled to pre-judgment interest from the Crown in the same way that they
would be so entitled as against any other litigant. The CPP is silent
on the issue of pre-judgment interest and cannot, as such, reasonably be
interpreted as creating an exception to the entitlement created by s. 31
of the Crown Liability and Proceedings Act .
VIII. Disposition
136
The appeal and cross-appeal are dismissed without costs. The
constitutional questions are answered as follows:
1. Does s. 44(1.1) of the Canada Pension
Plan, R.S.C. 1985, c. C‑8 , infringe s. 15(1) of the Canadian Charter
of Rights and Freedoms ?
Yes.
2. If so, is the infringement a reasonable
limit prescribed by law as can be demonstrably justified in a free and
democratic society under s. 1 of the Canadian Charter of
Rights and Freedoms ?
No.
3. Does s. 72(2) of the Canada Pension Plan,
R.S.C. 1985, c. C‑8 , infringe s. 15(1) of the Canadian Charter of
Rights and Freedoms ?
The following are the reasons delivered by
137
Bastarache J. — I have
read the joint reasons of my colleagues Justices LeBel and Rothstein, and,
while I agree with their disposition of this appeal, I cannot fully agree with
their approach to the question of retroactive constitutional remedies. In my
view, their reliance on the existence of a substantial change of law is an
inappropriate consideration in the context of rights guaranteed by the Canadian
Charter of Rights and Freedoms and is, in any event, inapplicable to this
appeal. The decision to deny retroactive relief, and the appellants’ challenge
to s. 72(1) of the Canadian Pension Plan, R.S.C. 1985, c. C-8 (“CPP ”),
specifically, should be based purely on the balancing of interests that must
take place in any claim for relief from an unconstitutional law.
The
Retroactivity of Constitutional Remedies
138
It is important to be clear about the nature and reasons for retroactive
constitutional remedies. The general and well-established rule is that remedies
for constitutional violations apply retroactively (S. Choudhry and K. Roach,
“Putting the Past Behind Us? Prospective Judicial and Legislative
Constitutional Remedies” (2003), 21 S.C.L.R. (2d) 205, at p. 211). The
basis for general retroactivity is not Blackstone’s declaratory theory, but the
Constitution itself. Section 52(1) of the Constitution Act, 1982 establishes
that any law which is inconsistent with the Constitution of Canada “is, to the
extent of the inconsistency, of no force or effect”. The Constitution exists
independently of judicial decisions and, as such, any law which is inconsistent
with it is invalidated from the moment the law came into effect. Gonthier J.
explained this principle clearly in Nova Scotia (Workers’ Compensation
Board) v. Martin, [2003] 2 S.C.R. 504, 2003 SCC 54, at para. 28:
The invalidity of a legislative provision inconsistent with the Charter
does not arise from the fact of its being declared unconstitutional by a court,
but from the operation of s. 52(1) . Thus, in principle, such a provision
is invalid from the moment it is enacted, and a judicial declaration to this
effect is but one remedy amongst others to protect those whom it adversely
affects. In that sense, by virtue of s. 52(1) , the question of constitutional
validity inheres in every legislative enactment. [Emphasis added.]
(See also P.
W. Hogg, Constitutional Law of Canada (loose-leaf ed.), vol. 2, at p.
55-2; K. Roach, Constitutional Remedies in Canada (loose-leaf), at para.
14.920.)
139
This rule applies equally in the context of the Charter . Laws
which violate Charter guarantees are invalid from the date that the
provisions of the Charter came into force: Jim Pattison Industries
Ltd. v. The Queen, [1984] 2 F.C. 954 (T.D.), at p. 957; Davidson v.
Davidson (1986), 33 D.L.R. (4th) 161 (B.C.C.A.), at p. 170, leave to appeal
refused, [1987] 1 S.C.R. vii; Rath v. Kemp (1996), 46 Alta. L.R. (3d) 1
(C.A.), at para. 25. In the case of a violation of s. 15 , this would mean that,
in theory, a law which violates the Charter would be invalid as of
April 17, 1985 or the date of its adoption, if later.
140
The general norm of retroactivity has been reaffirmed many times by this
Court. In Reference re Manitoba Language Rights, [1985] 1 S.C.R. 721,
the finding that the Province of Manitoba has been constitutionally required
since 1890 to enact its laws both in English and in French applied not only for
the future, but also retroactively to all of its laws that had been enacted in
English only since 1890. The declaration of invalidity was suspended in order
to preserve the rule of law, but there was no doubt that a finding of
unconstitutionality applied retroactively. In Miron v. Trudel, [1995] 2
S.C.R. 418, a violation of s. 15(1) of the Charter was cured by reading
the excluded group into the legislation, a remedy that applied retroactively.
In R. v. Hess, [1990] 2 S.C.R. 906, the Court severed an offending
portion of the Criminal Code, R.S.C. 1985, c. C-46 , and ordered a new
trial on the basis of the provision as amended even though, strictly speaking,
it was not the law of the land when the alleged crime had been committed. These
cases are only examples, but they confirm that retroactivity of a constitutional
remedy granted under s. 52(1) is the norm in our constitutional jurisprudence,
not the exception.
141
With respect, my colleagues ignore the fact that Blackstone’s comments
on the nature of judicial law making were made in the context of the common law
and find only limited application in the constitutional context. As it pertains
to the common law, Blackstone’s declaratory theory is necessarily a fiction
because there is no independent source for common law rules. The common law is
by definition judge-made. It does not exist in some jurisprudential ether for
judges to discover. Therefore when judges overturn precedents or establish a
new common law rule, they are undoubtedly making new law.
142
The same cannot be said for judicial decisions which interpret and apply
the Constitution. The Constitution exists independently of judicial decisions.
Judges do not “make” the Constitution every time they interpret its provisions.
Interpretations of what the Constitution requires may change, but the
underlying rights and freedoms endure. Charter rights are not created
every time a court expressly overrules or implicitly repudiates a prior
decision or gives “content to broad, but previously undefined, rights,
principles or norms” (LeBel and Rothstein JJ., at para. 99). The rights and
freedoms in the Charter were guaranteed to all Canadians from the moment
the Charter came into force.
143
By justifying the denial of retroactive relief in part on the existence
of a “substantial change in the law”, my colleagues give Charter rights
an uncertain status. I cannot accept an approach that, for remedial purposes,
implies that Charter rights can be here one day and gone the next or,
conversely, that they depend on judicial recognition of “a new or newly recognized
technological or social environment” (para. 99) for their genesis. Such
reasoning represents a watering down of the promise made to all Canadians when
the Charter was enacted. By attaching importance to changing social
conditions, it makes Charter rights dependent on how the majority of
Canadians perceive the claimants’ rights. With respect, I cannot see why
society’s views of Charter claimants — especially in the context of
vulnerable minorities — should be a factor for determining whether a Charter
right was part of the Constitution in 1985, or whether it sprung into
existence later and thereby be a basis for denying retroactive relief.
144
I should note that I am not advocating for a view of the Constitution
that says that it is frozen in time. The “living tree” metaphor aptly describes
how through time our Constitution may change to correspond to new realities.
But that does not mean that every time a new constitutional interpretation is
adopted or a previous decision is overturned that the fundamental rights and
freedoms guaranteed in our Charter have changed or that new ones have
been created. There is a difference between changes in constitutional
interpretation and actual constitutional change. Furthermore, the “living tree”
doctrine is a doctrine of “progressive interpretation” (Reference re
Same‑Sex Marriage, [2004] 3 S.C.R. 698, 2004 SCC 79, at paras. 22-23
(emphasis added)), necessary to ensure that our Constitution does not become
rigid and unresponsive to Canadian society. It should not be used as a
justification for denying relief to a particular group of Charter
claimants.
145
For these reasons, I cannot accept my colleagues’ critique of the
declaratory approach as a basis for denying retroactive relief. As will be
explained below, there are important reasons for denying retroactive relief in
certain circumstances. I am largely in agreement with my colleagues on what
they are. But I cannot agree that they have anything to do with the success or
failure of the “Blackstonian paradigm” in the context of constitutional law.
146
The dangers of my colleagues’ approach are adequately evidenced when
applied to the claimants in this appeal. The starting point of their analysis
is that there was a substantial change in the law between 1985, when s. 15(1)
of the Charter came into force, and this Court’s decision in M. v. H.,
[1999] 2 S.C.R. 3. The implication is that the right of same-sex spouses
not to be excluded from survivor benefits did not form part of the Constitution
until 1999. To put it bluntly, s. 15(1) of the Charter did not
extend to same-sex couples until this Court said it did. I note that my
colleagues are not simply saying that this Court’s interpretation of the
Constitution had changed between 1985 and 1999. If that were the case, it would
be sufficient to base their denial of retroactive relief solely on the good
faith reliance of the government. Instead, by relying on a critique of the
declaratory theory and the “living tree” doctrine, my colleagues assert, in
essence, that the Constitution actually changed between 1985 and 1999 and that
the claimants, unlike other Canadians, were not entitled to its protection in
1985. Such an approach runs counter to the spirit of the Charter and
should not be countenanced.
147
Furthermore, it is not at all clear that a substantial change in the law
actually occurred sufficient to trigger a departure from the norm of
retroactivity under my colleagues’ threshold approach. A review of the case law
before and after Egan suggests that it was not the “final word” on s.
15(1) and discrimination of same-sex couples.
148
Courts did not begin pronouncing on the extent of protection that s. 15
afforded to same-sex couples until the late 1980s, and then with conflicting
results. In Andrews v. Ontario (Minister of Health) (1988), 64 O.R. (2d)
258, the Ontario High Court of Justice held that it was not discriminatory to
deny provincial health insurance benefits to same-sex dependants based on the
definition of “spouse” in the applicable legislation. In Vogel v. Manitoba (1992),
90 D.L.R. (4th) 84 (Man. Q.B.), exclusion of a same-sex partner from an
employer’s spousal benefit plan was held not to be discriminatory.
149
In other cases, the opposite result was achieved. In Veysey v. Canada
(Commissioner of the Correctional Service), [1990] 1 F.C. 321, the Federal
Court—Trial Division found that it was an infringement of s. 15(1) to deny
visitation rights to the same-sex partner of an inmate and that the
infringement could not be saved under s. 1 . In Knodel v. British Colombia
(Medical Services Commission) (1991), 58 B.C.L.R. (2d) 356 (S.C.), in
direct opposition to Andrews, regulations excluding same-sex dependants
from claiming provincial health insurance benefits were found to be an
unjustified infringement of s. 15(1) and a declaration that same-sex couples be
read into the definition of “spouse” was made.
150
This Court first came close to the issue in Canada (Attorney General)
v. Mossop, [1993] 1 S.C.R. 554, which involved a provision of a collective
agreement that restricted bereavement leave to members of an employee’s
“immediate family”. The complainant had sought to have the day he took off work
to attend the funeral of his same-sex partner’s father count as bereavement
leave. The majority (in a 4-3 decision) decided the issue on the narrow ground
that the Canadian Human Rights Act, R.S.C. 1985, c. H-6 , did not include
sexual orientation as a ground of discrimination. Lamer C.J. noted, however,
that the case could have been argued on the basis that the Charter
required sexual orientation to be read into the Canadian Human Rights Act .
Because the Charter argument had not been made, he felt the Court had to
apply the Act as it stood. He made it clear that he would have perhaps decided
the case differently if sexual orientation had been a prohibited ground of
discrimination (p. 582).
151
The next time this Court dealt with the issue was in Egan v. Canada,
[1995] 2 S.C.R. 513. In a narrow 5-4 decision, this Court held that the
definition of “spouse” under the Old Age Security Act, R.S.C. 1985, c.
O-9 , for the purposes of spousal allowances under the Act, was an infringement
of s. 15(1) , but was saved under s. 1 . Sopinka J. cast the deciding vote.
Central to his reasoning was that governments should be accorded some leeway in
extending social benefits. In this regard, he agreed with La Forest J.’s
suggestion in McKinney v. University of Guelph, [1990] 3 S.C.R. 229,
that governments should be allowed time to take incremental measures to deal
with discrimination (para. 105).
152
It was soon clear that Egan was not the final word on the matter.
The Manitoba Court of Appeal rendered its judgment in Vogel v. Manitoba
(1995), 126 D.L.R. (4th) 72, a few months after Egan. It reversed the
trial judge’s decision and held that it was discriminatory under provincial
human rights legislation to exclude same-sex partners from employee benefit
plans on the basis that same-sex partners were not included in the term
“spouse”. Helper J.A. and Scott C.J.M. relied on Egan to find that the
eligibility criteria of the scheme were discriminatory. A little more than a
year later, the Ontario Court of Appeal released its decision in M. v. H.
(1996), 96 O.A.C. 173. Charron J.A. (as she then was) found that the definition
of “spouse” contained in the Family Law Act, R.S.O. 1990, c. F.3, and
which excluded same-sex couples, was a violation of s. 15(1) and could not be
saved under s. 1 . Although she took notice of this Court’s decision in Egan,
she was of the view that it was not determinative since the objectives of the
legislation in each case were different (para. 84). The Court of Appeal’s
decision in M. v. H. was relied upon in Kane v. Ontario (Attorney
General) (1997), 152 D.L.R. (4th) 738 (Ont. Ct. (Gen. Div.)), to hold that
the exclusion of same-sex partners from the definition of “spouse” in Ontario’s
Insurance Act, R.S.O. 1990, c. I.8, was also unconstitutional.
153
In Vriend v. Alberta, [1998] 1 S.C.R. 493, this Court
rejected the incrementalist approach that was relied upon in part by Sopinka J.
in Egan to find that the constitutional violation in that case was
saved under s. 1 .
154
Relying on its own jurisprudence and this Court’s decision in Vriend,
the Ontario Court of Appeal held in Rosenberg v. Canada (Attorney General)
(1998), 38 O.R. (3d) 577, that a provision of the Income Tax Act, R.S.C.
1985, c. 1 (5th Supp .), that limited the definition of “spouse” to opposite sex
couples was unconstitutional.
155
Finally, in M. v. H., this Court upheld the Ontario Court of
Appeal’s decision that the exclusion of same-sex couples from the definition of
spouse in the Family Law Act was unconstitutional. Iacobucci J.
distinguished this Court’s holding in Egan on the same grounds as the
Court of Appeal had, namely that the legislative objectives at issue were
entirely different (para. 75). It was in response to this decision that the
federal government decided to enact the amendments to federal legislation,
including the CPP , contained in the Modernization of Benefits and
Obligations Act, S.C. 2000, c. 12 .
156
As can be seen, the majority of cases both before and after Egan indicated
that differential treatment of same-sex couples was discriminatory and could
not be justified under s. 1 . It is true that this Court has the final word on
the interpretation of the Constitution. However, I find it difficult to believe
that Egan was the final word — even for a time — on the application of
s. 15(1) to same-sex couples. Egan was decided on the basis of s. 1
arguments that were tailored to the specific legislative objectives at issue in
that case. After Egan, it was far from clear whether its reasoning would
apply outside the realm of social assistance schemes such as Old Age Security.
Indeed, a number of decisions, including this Court’s decision in M. v. H.,
distinguished Egan on this basis. It did not stand for the application
of s. 15(1) to all instances of legislative exclusions of same-sex couples.
Given the CPP ’s status as an insurance-based scheme, where contributions
are paid by employers and employees, it is difficult to see how Egan definitely
settled the issue with regard to the exclusion of same-sex couples from
survivor’s benefits under the CPP .
157
Furthermore, given the contradictory decisions both before and after Egan,
the closeness of the decision in that case, and the difficult nature of the
issues at stake, it is difficult to see Egan as definitively
establishing what the Constitution required. The reality is that it was for a
time unclear exactly how s. 15(1) would apply to same-sex couples. The judicial
process can be slow. It took time for this Court and others to articulate the
correct constitutional principles to be applied to legislative exclusions of
same-sex couples. That does not mean that this Court was upsetting established
law when it handed down its decision in M. v. H. In short, even if the
existence of a substantial change in the law was an appropriate threshold
criterion, no such change actually occurred in this case. We must therefore
look to other reasons for denying retroactive relief.
The Correct
Approach to Retroactive Remedies in This Case
158
It is well established in Canadian constitutional law that courts will
seek to balance competing interests when devising remedies for constitutional
violations (Roach, at paras. 3.680 to 3.780). The starting point is
that past constitutional violations should be corrected. This recognizes the
grave nature of constitutional infringements:
Constitutional remedies are matters of considerable
importance. Constitutional law expresses the most fundamental restraints and
obligations of governments. Violations of constitutional rights are a serious
matter both for those who suffer from the violation and the public in general.
(Roach, at para. 1.10)
Retroactive
constitutional remedies ensure that the claimants who have brought the
successful action, as well as similarly situated claimants, can benefit from
the judgment (Choudhry and Roach, at p. 210).
159
However, the normal retroactive effect of judgments may need to be
tempered in certain circumstances in order to protect other legitimate
interests (see Choudhry and Roach, at pp. 209-11). The use of transition
periods and suspended declarations of invalidity are accepted ways of
temporarily limiting the retroactive effect of constitutional remedies in order
to prevent legal vacuums and introduce new procedural requirements. They are
not evidence of the Court operating outside of the Blackstonian paradigm, but
rather a recognition that in certain circumstances other legitimate interests
may require retroactivity to be limited.
160
The question is what legitimate interests should be taken into account
in deciding to deny a retroactive constitutional remedy. For the most part, I
agree that reasonable reliance, good faith, fairness to litigants and
Parliament’s role are important considerations. I would simply like to add what
I see as a few points of clarification.
161
First, the question of whether to deny a retroactive remedy is different
from deciding whether to grant a suspended declaration of invalidity; the same
considerations do not apply to both. The latter will be a valid measure when a
declaration of invalidity would pose a danger to the public, threaten the rule
of law or deprive deserving persons of benefits without thereby benefiting the
individual whose rights had been violated (Schachter v. Canada,
[1992] 2 S.C.R. 679, at p. 719). A suspended declaration of invalidity is
ultimately only a temporary limit on retroactivity; it does not determine
whether governments are entitled to deny retroactive relief to the claimants in
acting to cure the constitutional defect. Reasonable reliance, good faith,
fairness to litigants and Parliament’s role will be appropriate to consider
when courts are confronted with the question of what remedy the claimant is
entitled to. This will occur, for example, when deciding whether a remedy of
“reading in” should apply retroactively or whether a legislative provision
enacted in response to a declaration of invalidity which limits retroactive
relief should be read down or severed.
162
Second, to my mind, establishing reasonable reliance will not always be
necessary in order to deny retroactive relief. In this case, even on my
colleagues’ analysis, the government’s reliance on Egan could not be a
justification for denying relief before Egan was decided. Indeed,
although it was not entirely clear, most of the case law before Egan suggested
that governments could not justifiably exclude same-sex couples. Given the
state of the law pre-M. v. H., it would be more accurate to emphasize
the fact that the government was acting in good faith in the face of
jurisprudential uncertainty. Although not determinative, this should be taken
into account in deciding whether to depart from the norm of retroactive
relief.
163
Finally, my colleagues seem to suggest in their discussion of this
Court’s recent decision of Kingstreet Investments Ltd. v. New Brunswick
(Finance), [2007] 1 S.C.R. 3, 2007 SCC 1, that the nature of the
constitutional violation is relevant to deciding whether to deny a retroactive
remedy. I do not agree. A government has no more right to discriminate in the
provision of benefits than it does to collect unconstitutionally levied taxes.
In Kingstreet, there were no legitimate concerns with applying the
general rule of retroactivity. The legislature retained the ability to enact
remedial legislation to cure any adverse effects (para. 25), and a purely
prospective remedy would have left the claimant empty-handed.
164
These clarifications made, I am in general agreement with how my
colleagues have applied the other factors to this appeal. Particularly
relevant, it seems to me, is the fact that the Modernization of Benefits and
Obligations Act was enacted in response to this Court’s decision in M.
v. H. In that case, a suspension of the declaration of invalidity was
ordered so as to allow the Ontario government flexibility to cure the
constitutional defect. That flexibility implicitly included the ability to
limit the retroactive effect of any remedial legislation. Indeed, this is what
the Ontario legislature chose to do. The remedial legislation was made
prospective from November 20, 1999 (Amendments Because of the Supreme Court
of Canada Decision in M. v. H. Act, 1999, S.O. 1999, c. 6, s.
68(2)). Similar flexibility should be accorded to the Canadian government in
this case. The legislative branch is better able to deal with distributional
concerns than are courts, and its choices should be respected so long as they
fall within the limits of the Constitution.
Conclusion
165
For these reasons, I agree that the appeal and cross-appeal should be
dismissed without costs.
Appeal and cross‑appeal dismissed.
Solicitor for the appellant/respondent on cross‑appeal: Attorney
General of Canada, Ottawa.
Solicitors for the respondents/appellants on cross‑appeal: Roy
Elliott Kim O’Connor, Toronto.
Solicitor for the intervener the Attorney General of
Ontario: Attorney General of Ontario, Toronto.
Solicitor for the intervener the Attorney General of
Quebec: Department of Justice, Sainte‑Foy.
Solicitor for the intervener the Attorney General of
Alberta: Alberta Justice, Edmonton.
Solicitors for the intervener Egale Canada Inc.: Sack
Goldblatt Mitchell, Toronto.