Supreme Court Judgments

                                                 SUPREME COURT OF CANADA

Between:

David Dunsmuir

Appellant

v.

Her Majesty the Queen in Right of the Province of New Brunswick

as represented by Board of Management

Respondent

Coram: McLachlin C.J. and Bastarache, Binnie, LeBel, Deschamps, Fish, Abella, Charron and Rothstein JJ.

______________________________

Dunsmuir v. New Brunswick, [2008] 1 S.C.R. 190, 2008 SCC 9

David Dunsmuir                                                                                                                 Appellant

v.

Her Majesty the Queen in Right of the Province of

New Brunswick as represented by Board of Management                                          Respondent

Indexed as:  Dunsmuir v. New Brunswick

Neutral citation:  2008 SCC 9.

File No.:  31459.

2007:  May 15; 2008:  March 7.

Present:  McLachlin C.J. and Bastarache, Binnie, LeBel, Deschamps, Fish, Abella, Charron and Rothstein JJ.

on appeal from the court of appeal for new brunswick

Administrative law — Judicial review — Standard of review — Proper approach to judicial review of administrative decision makers — Whether judicial review should include only two standards:  correctness and reasonableness.

Administrative law — Judicial review — Standard of review — Employee holding office “at pleasure” in provincial civil service dismissed without alleged cause with four months’ pay in lieu of notice — Adjudicator interpreting enabling statute as conferring jurisdiction to determine whether discharge was in fact for cause — Adjudicator holding employer breached duty of procedural fairness and ordering reinstatement — Whether standard of reasonableness applicable to adjudicator’s decision on statutory interpretation issue — Public Service Labour Relations Act, R.S.N.B. 1973, c. P‑25, ss. 97(2.1), 100.1(5) — Civil Service Act, S.N.B. 1984, c. C‑5.1, s. 20.

Administrative law — Natural justice — Procedural fairness — Dismissal of public office holders — Employee holding office “at pleasure” in provincial civil service dismissed without alleged cause with four months’ pay in lieu of notice — Employee not informed of reasons for termination or provided with opportunity to respond — Whether employee entitled to procedural fairness — Proper approach to dismissal of public employees.

D was employed by the Department of Justice for the Province of New Brunswick.  He held a position under the Civil Service Act and was an office holder “at pleasure”.  His probationary period was extended twice and the employer reprimanded him on three separate occasions during the course of his employment.  On the third occasion, a formal letter of reprimand was sent to D warning him that his failure to improve his performance would result in further disciplinary action up to and including dismissal. While preparing for a meeting to discuss D’s performance review the employer concluded that D was not right for the job.  A formal letter of termination was delivered to D’s lawyer the next day.  Cause for the termination was explicitly not alleged and D was given four months’ pay in lieu of notice.

D commenced the grievance process under s. 100.1 of the Public Service Labour Relations Act (“PSLRA”), alleging that the reasons for the employer’s dissatisfaction were not made known, that he did not receive a reasonable opportunity to respond to the concerns, that the employer’s actions in terminating him were without notice, due process or procedural fairness, and that the length of the notice period was inadequate.  The grievance was denied and then referred to adjudication.  A preliminary issue of statutory interpretation arose as to whether, where dismissal was with notice or pay in lieu thereof, the adjudicator was authorized to determine the reasons underlying the province’s decision to terminate.  The adjudicator held that the referential incorporation of s. 97(2.1) of the PSLRA into s. 100.1(5) of that Act meant that he could determine whether D had been discharged or otherwise disciplined for cause.  Ultimately, the adjudicator made no finding as to whether the discharge was or was not for cause.  In his decision on the merits, he found that the termination letter effected termination with pay in lieu of notice and that the termination was not disciplinary.  As D’s employment was hybrid in character, the adjudicator held that D was entitled to and did not receive procedural fairness in the employer’s decision to terminate his employment.  He declared that the termination was void ab initio and ordered D reinstated as of the date of dismissal, adding that in the event that his reinstatement order was quashed on judicial review, he would find the appropriate notice period to be eight months.

On judicial review, the Court of Queen’s Bench applied the correctness standard and quashed the adjudicator’s preliminary decision, concluding that the adjudicator did not have jurisdiction to inquire into the reasons for the termination, and that his authority was limited to determining whether the notice period was reasonable.  On the merits, the court found that D had received procedural fairness by virtue of the grievance hearing before the adjudicator.  Concluding that the adjudicator’s decision did not stand up to review on a reasonableness simpliciter standard, the court quashed the reinstatement order but upheld the adjudicator’s provisional award of eight months’ notice.  The Court of Appeal held that the proper standard with respect to the interpretation of the adjudicator’s authority under the PSLRA was reasonableness simpliciter, not correctness, and that the adjudicator’s decision was unreasonable.  It found that where the employer elects to dismiss with notice or pay in lieu of notice, s. 97(2.1) of the PSLRA does not apply and the employee may only grieve the length of the notice period.  It agreed with the reviewing judge that D’s right to procedural fairness had not been breached.

Held:  The appeal should be dismissed.

Per McLachlin C.J. and Bastarache, LeBel, Fish and Abella JJ.:  Despite its clear, stable constitutional foundations, the system of judicial review in Canada has proven to be difficult to implement.  It is necessary to reconsider both the number and definitions of the various standards of review, and the analytical process employed to determine which standard applies in a given situation.  Notwithstanding the theoretical differences between the standards of patent unreasonableness and reasonableness simpliciter, any actual difference between them in terms of their operation appears to be illusory.  There ought to be only two standards of review: correctness and reasonableness.  [32] [34] [41]

When applying the correctness standard in respect of jurisdictional and some other questions of law, a reviewing court will not show deference to the decision maker’s reasoning process; it will rather undertake its own analysis of the question and decide whether it agrees with the determination of the decision maker; if not, the court will substitute its own view and provide the correct answer.  A court conducting a review for reasonableness inquires into the qualities that make a decision reasonable.  Reasonableness is concerned mostly with the existence of justification, transparency and intelligibility within the decision‑making process and with whether the decision falls within a range of possible, acceptable outcomes which are defensible in respect of the facts and the law.  It is a deferential standard which requires respect for the legislative choices to leave some matters in the hands of administrative decision makers, for the processes and determinations that draw on particular expertise and experiences, and for the different roles of the courts and administrative bodies within the Canadian constitutional system.  [47‑50]

An exhaustive analysis is not required in every case to determine the proper standard of review.  Courts must first ascertain whether the jurisprudence has already determined in a satisfactory manner the degree of deference to be accorded to a decision maker with regard to a particular category of question.  If the inquiry proves unfruitful, courts must analyze the factors making it possible to identify the proper standard of review.  The existence of a privative clause is a strong indication of review pursuant to the reasonableness standard, since it is evidence of Parliament or a legislature’s intent that an administrative decision maker be given greater deference and that interference by reviewing courts be minimized.  It is not, however, determinative.  Where the question is one of fact, discretion or policy, or where the legal issue is intertwined with and cannot be readily separated from the factual issue, deference will usually apply automatically.  Deference will usually result where a decision maker is interpreting its own statute or statutes closely connected to its function, with which it will have particular familiarity.  While deference may also be warranted where an administrative decision maker has developed particular expertise in the application of a general common law or civil law rule in relation to a specific statutory context, a question of law that is of central importance to the legal system as a whole and outside the specialized area of expertise of the administrative decision maker will always attract a correctness standard.  So will a true question of vires, a question regarding the jurisdictional lines between two or more competing specialized tribunals, and a constitutional question regarding the division of powers between Parliament and the provinces in the Constitution Act, 1867 .  [52‑62]

The standard of reasonableness applied on the issue of statutory interpretation.  While the question of whether the combined effect of ss. 97(2.1) and 100.1 of the PSLRA permits the adjudicator to inquire into the employer’s reason for dismissing an employee with notice or pay in lieu of notice is a question of law, it is not one that is of central importance to the legal system and outside the specialized expertise of the adjudicator, who was in fact interpreting his enabling statute.  Furthermore, s. 101(1) of the PSLRA includes a full privative clause, and the nature of the regime favours the standard of reasonableness.  Here, the adjudicator’s interpretation of the law was unreasonable and his decision does not fall within the range of acceptable outcomes that are defensible in respect of the facts and the law.  The employment relationship between the parties in this case was governed by private law.  The combined effect of ss. 97(2.1) and 100.1 of the PSLRA cannot, on any reasonable interpretation, remove the employer’s right, under the ordinary rules of contract, to discharge an employee with reasonable notice or pay in lieu thereof without asserting cause.  By giving the PSLRA an interpretation that allowed him to inquire into the reasons for discharge, the adjudicator adopted a reasoning process that was fundamentally inconsistent with the employment contract and, thus, fatally flawed.  [66‑75]

On the merits, D was not entitled to procedural fairness.  Where a public employee is employed under a contract of employment, regardless of his or her status as a public office holder, the applicable law governing his or her dismissal is the law of contract, not general principles arising out of public law.  Where a dismissal decision is properly within the public authority’s powers and is taken pursuant to a contract of employment, there is no compelling public law purpose for imposing a duty of fairness.  The principles expressed in Knight v. Indian Head School Division No. 19 in relation to the general duty of fairness owed by public authorities when making decisions that affect the rights, privileges or interests of individuals are valid and important.  However, to the extent that Knight ignored the important effect of a contract of employment, it should not be followed.  In the case at bar, D was a contractual employee in addition to being a public office holder.  Section 20 of the Civil Service Act provided that as a civil servant he could only be dismissed in accordance with the ordinary rules of contract.  To consider a public law duty of fairness issue where such a duty exists falls squarely within the adjudicator’s task to resolve a grievance.  Where, as here, the relationship is contractual, it was unnecessary to consider any public law duty of procedural fairness.  By imposing procedural fairness requirements on the respondent over and above its contractual obligations and ordering the full “reinstatement” of D, the adjudicator erred and his decision was therefore correctly struck down.  [76‑78] [81] [84] [106] [114] [117]

Per Binnie J.: The majority reasons for setting aside the adjudicator ruling were generally agreed with, however the call of the majority to re‑evaluate the pragmatic and functional test and to re‑assess “the structure and characteristics of the system of judicial review as a whole” and to develop a principled framework that is “more coherent and workable” invites a broader reappraisal.  Judicial review is an idea that has lately become unduly burdened with law office metaphysics.  Litigants find the court’s attention focussed not on their complaints, or the government’s response, but on lengthy and arcane discussions of something they are told is the pragmatic and functional test.  The Court should at least (i) establish some presumptive rules and (ii) get the parties away from arguing about the tests and back to arguing about the substantive merits of their case.  [119-122] [133] [145]

The distinction between “patent unreasonableness” and reasonableness simpliciter is now to be abandoned.  The repeated attempts to explain the difference between the two, was in hindsight, unproductive and distracting.  However, a broad reappraisal of the system of judicial review should explicitly address not only administrative tribunals but issues related to other types of administrative bodies and statutory decision makers including mid‑level bureaucrats and, for that matter, Ministers.  If logic and language cannot capture the distinction in one context, it must equally be deficient elsewhere in the field of judicial review.  [121‑123] [134-135] [140]

It should be presumed that the standard of review of an administrative outcome on grounds of substance is reasonableness.  In accordance with the ordinary rules of litigation, it should also be presumed that the decision under review is reasonable until the applicant shows otherwise.  An applicant urging the non‑deferential “correctness” standard should be required to demonstrate that the decision rests on an error in the determination of a legal issue not confided (or which constitutionally could not be confided) to the administrative decision maker to decide, whether in relation to jurisdiction or the general law.  The logic of the constitutional limitation is obvious.  If the limitation did not exist, the government could transfer the work of the courts to administrative bodies that are not independent of the executive and by statute immunize the decisions of these bodies from effective judicial review.  Questions of law outside the administrative decision maker’s home statute and closely related rules or statutes which require his or her expertise should also be reviewable on a “correctness” standard whether or not it meets the majority’s additional requirement that it be “of central importance to the legal system as a whole”.  The standard of correctness should also apply to the requirements of “procedural fairness”, which will vary with the type of decision maker and the type of decision under review.  Nobody should have his or her rights, interests or privileges adversely dealt with by an unjust process.  [127‑129] [146‑147]

On the other hand when the application for judicial review challenges the substantive outcome of an administrative action, the judge is invited to cross the line into second‑guessing matters that lie within the function of the administrator.  This is controversial because it is not immediately obvious why a judge’s view of the reasonableness of an administrative policy or the exercise of an administrative discretion should be preferred to that of the administrator to whom Parliament or a legislature has allocated the decision, unless there is a full statutory right of appeal to the courts, or it is otherwise indicated in the conferring legislation that a “correctness” standard is intended.  [130]

Abandonment of the distinction between reasonableness simpliciter and patent unreasonableness has important implications.  The two different standards addressed not merely “the magnitude or the immediacy of the defect” in the administrative decision but recognized that different administrative decisions command different degrees of deference, depending on who is deciding what.  [135]

“Contextualizing” a single standard of “reasonableness” review will shift the courtroom debate from choosing between two standards of reasonableness that each represented a different level of deference to a debate within a single standard of reasonableness to determine the appropriate level of deference.  [139]

Thus a single “reasonableness” standard will now necessarily incorporate both the degree of deference owed to the decision maker formerly reflected in the distinction between patent unreasonableness and reasonableness simpliciter, and an assessment of the range of options reasonably open to the decision maker in the circumstances.  The judge’s role is to identify the outer boundaries of reasonable outcomes within which the administrative decision maker is free to choose.  [141] [149]

A single “reasonableness” standard is a big tent that will have to accommodate a lot of variables that inform and limit a court’s review of the outcome of administrative decision making.  “Contextualizing” the reasonableness standard will require a reviewing court to consider the precise nature and function of the decision maker including its expertise, the terms and objectives of the governing statute (or common law) conferring the power of decision including the existence of a privative clause and the nature of the issue being decided.  Careful consideration of these matters will reveal the extent of the discretion conferred.  In some cases the court will have to recognize that the decision maker was required to strike a proper balance (or achieve proportionality) between the adverse impact of a decision on the rights and interests of the applicant or others directly affected weighed against the public purpose which is sought to be advanced.  In each case careful consideration will have to be given to the reasons given for the decision.  This list of “contextual” considerations is non‑exhaustive.  A reviewing court ought to recognize throughout the exercise that fundamentally the “reasonableness” of the administrative outcome is an issue given to another forum to decide.  [144] [151‑155]

Per Deschamps, Charron and Rothstein JJ.: Any review starts with the identification of the questions at issue as questions of law, questions of fact or questions of mixed fact and law.  In the adjudicative context, decisions on questions of fact, whether undergoing appellate review or administrative law review, always attract deference.  When there is a privative clause, deference is owed to the administrative body that interprets the legal rules it was created to interpret and apply.  If the body oversteps its delegated powers, if it is asked to interpret laws in respect of which it does not have expertise or if Parliament or a legislature has provided for a statutory right of review, deference is not owed to the decision maker.  Finally, when considering a question of mixed fact and law, a reviewing court should show an adjudicator the same deference as an appeal court would show a lower court.  [158‑164]

Here, the employer’s common law right to dismiss without cause was the starting point of the analysis.  Since the adjudicator does not have specific expertise in interpreting the common law, the reviewing court can proceed to its own interpretation of the applicable rules and determine whether the adjudicator could enquire into the cause of the dismissal.  The applicable standard of review is correctness.  The distinction between the common law rules of employment and the statutory rules applicable to a unionized employee is essential if s. 97(2.1) of the PSLRA is to be applied mutatis mutandis to the case of a non‑unionized employee as required by s. 100.1(5) of the PSLRA.  The adjudicator’s failure to inform himself of this crucial difference led him to look for a cause for the dismissal, which was not relevant.  Even if deference had been owed to the adjudicator, his interpretation could not have stood.  Employment security is so fundamental to an employment relationship that it could not have been granted by the legislature by providing only that the PSLRA was to apply mutatis mutandis to non‑unionized employees.  [168‑171]

Cases Cited

By Bastarache and LeBel JJ.

Referred to:  Chalmers (Dr. Everett) Hospital v. Mills (1989), 102 N.B.R. (2d) 1; Knight v. Indian Head School Division No. 19, [1990] 1 S.C.R. 653; Alberta Union of Provincial Employees v. Lethbridge Community College, [2004] 1 S.C.R. 727, 2004 SCC 28; Crevier v. Attorney General of Quebec, [1981] 2 S.C.R. 220; Dr. Q v. College of Physicians and Surgeons of British Columbia, [2003] 1 S.C.R. 226, 2003 SCC 19; Executors of the Woodward Estate v. Minister of Finance, [1973] S.C.R. 120; U.E.S., Local 298 v. Bibeault, [1988] 2 S.C.R. 1048; Baker v. Canada (Minister of Citizenship and Immigration), [1999] 2 S.C.R. 817; Suresh v. Canada (Minister of Citizenship and Immigration), [2002] 1 S.C.R. 3, 2002 SCC 1; Mount Sinai Hospital Center v. Quebec (Minister of Health and Social Services), [2001] 2 S.C.R. 281, 2001 SCC 41; C.U.P.E. v. Ontario (Minister of Labour), [2003] 1 S.C.R. 539, 2003 SCC 29; Chamberlain v. Surrey School District No. 36, [2002] 4 S.C.R. 710, 2002 SCC 86; Canadian Union of Public Employees, Local 963 v. New Brunswick Liquor Corp., [1979] 2 S.C.R. 227; Canada (Director of Investigation and Research) v. Southam Inc., [1997] 1 S.C.R. 748; Pushpanathan v. Canada (Minister of Citizenship and Immigration), [1998] 1 S.C.R. 982; Toronto (City) v. C.U.P.E., Local 79, [2003] 3 S.C.R. 77, 2003 SCC 63; Law Society of New Brunswick v. Ryan, [2003] 1 S.C.R. 247, 2003 SCC 20; Council of Canadians with Disabilities v. Via Rail Canada Inc., [2007] 1 S.C.R. 650, 2007 SCC 15; Voice Construction Ltd. v. Construction & General Workers’ Union, Local 92, [2004] 1 S.C.R. 609, 2004 SCC 23; Canada (Attorney General) v. Mossop, [1993] 1 S.C.R. 554; Canadian Broadcasting Corp. v. Canada (Labour Relations Board), [1995] 1 S.C.R. 157; Toronto (City) Board of Education v. O.S.S.T.F., District 15, [1997] 1 S.C.R. 487; McLeod v. Egan, [1975] 1 S.C.R. 517; Cartaway Resources Corp. (Re), [2004] 1 S.C.R. 672, 2004 SCC 26; Westcoast Energy Inc. v. Canada (National Energy Board), [1998] 1 S.C.R. 322; Nova Scotia (Workers’ Compensation Board) v. Martin, [2003] 2 S.C.R. 504, 2003 SCC 54; United Taxi Drivers’ Fellowship of Southern Alberta v. Calgary (City), [2004] 1 S.C.R. 485, 2004 SCC 19; Regina Police Assn. Inc. v. Regina (City) Board of Police Commissioners, [2000] 1 S.C.R. 360, 2000 SCC 14; Quebec (Commission des droits de la personne et des droits de la jeunesse) v. Quebec (Attorney General), [2004] 2 S.C.R. 185, 2004 SCC 39; Canada Safeway Ltd. v. RWDSU, Local 454, [1998] 1 S.C.R. 1079; Moreau‑Bérubé v. New Brunswick (Judicial Council), [2002] 1 S.C.R. 249, 2002 SCC 11; Ridge v. Baldwin, [1963] 2 All E.R. 66; Nicholson v. Haldimand‑Norfolk Regional Board of Commissioners of Police, [1979] 1 S.C.R. 311; Martineau v. Matsqui Institution Disciplinary Board, [1980] 1 S.C.R. 602; Kane v. Board of Governors of the University of British Columbia, [1980] 1 S.C.R. 1105; Attorney General of Canada v. Inuit Tapirisat of Canada, [1980] 2 S.C.R. 735; Cardinal v. Director of Kent Institution, [1985] 2 S.C.R. 643; Reglin v. Creston (Town) (2004), 34 C.C.E.L. (3d) 123, 2004 BCSC 790; Gismondi v. Toronto (City) (2003), 64 O.R. (3d) 688; Seshia v. Health Sciences Centre (2001), 160 Man. R. (2d) 41, 2001 MBCA 151; Rosen v. Saskatoon District Health Board (2001), 202 D.L.R. (4th) 35, 2001 SKCA 83; Hanis v. Teevan (1998), 111 O.A.C. 91; Gerrard v. Sackville (Town) (1992), 124 N.B.R. (2d) 70; Malloch v. Aberdeen Corp., [1971] 2 All E.R. 1278; Hughes v. Moncton (City) (1990), 111 N.B.R. (2d) 184, aff’d (1991), 118 N.B.R. (2d) 306; Rosen v. Saskatoon District Health Board, [2000] 4 W.W.R. 606, 2000 SKQB 40; Wells v. Newfoundland, [1999] 3 S.C.R. 199; School District No. 5 (Southeast Kootenay) and B.C.T.F. (Yellowaga) (Re) (2000), 94 L.A.C. (4th) 56; Wallace v. United Grain Growers Ltd., [1997] 3 S.C.R. 701.

By Binnie J.

Referred to: Canadian Union of Public Employees, Local 963 v. New Brunswick Liquor Corp., [1979] 2 S.C.R. 227; Cooper v. Wandsworth Board of Works (1863), 14 C.B. (N.S.) 180, 143 E.R. 414; Ocean Port Hotel Ltd. v. British Columbia (General Manager, Liquor Control and Licensing Branch), [2001] 2 S.C.R. 781, 2001 SCC 52; U.E.S., Local 298 v. Bibeault, [1988] 2 S.C.R. 1048; Associated Provincial Picture Houses Ltd. v. Wednesbury Corp., [1947] 2 All E.R. 680; Baker v. Canada (Minister of Citizenship and Immigration), [1999] 2 S.C.R. 817; Mount Sinai Hospital Center v. Quebec (Minister of Health and Social Services), [2001] 2 S.C.R. 281, 2001 SCC 41; Idziak v. Canada (Minister of Justice), [1992] 3 S.C.R. 631; Suresh v. Canada (Minister of Citizenship and Immigration), [2002] 1 S.C.R. 3, 2002 SCC 1; Operation Dismantle Inc. v. The Queen, [1985] 1 S.C.R. 441; Attorney General of Canada v. Inuit Tapirisat of Canada, [1980] 2 S.C.R. 735; Westcoast Energy Inc. v. Canada (National Energy Board), [1998] 1 S.C.R. 322; Law Society of New Brunswick v. Ryan, [2003] 1 S.C.R. 247, 2003 SCC 20; Moreau‑Bérubé v. New Brunswick (Judicial Council), [2002] 1 S.C.R. 249, 2002 SCC 11; C.U.P.E. v. Ontario (Minister of Labour), [2003] 1 S.C.R. 539, 2003 SCC 29; Roncarelli v. Duplessis, [1959] S.C.R. 121.

By Deschamps J.

Referred to: H.L. v. Canada (Attorney General), [2005] 1 S.C.R. 401, 2005 SCC 25; Toronto (City) v. C.U.P.E., Local 79, [2003] 3 S.C.R. 77, 2003 SCC 63.

Statutes and Regulations Cited

Civil Service Act, S.N.B. 1984, c. C‑5.1, s. 20.

Constitution Act, 1867 , ss. 96  to 101 .

Employment Standards Act, S.N.B. 1982, c. E‑7.2.

Extradition Act, R.S.C. 1985, c. E‑23.

Human Rights Act, R.S.N.B. 1973, c. H‑11.

Interpretation Act , R.S.C. 1985, c. I‑21 , s. 23(1) .

Interpretation Act, R.S.N.B. 1973, c. I‑13, s. 20.

Public Service Labour Relations Act, R.S.N.B. 1973, c. P‑25, ss. 92(1), 97, 97(2.1) [ad. 1990, c. 30, s. 35], 100.1 [idem, s. 40], 101(1) [idem, s. 41], (2) [idem].

Authors Cited

Black’s Law Dictionary, 8th ed.  St. Paul, Minn.:  West, 2004, “pleasure appointment”.

Brown, Donald J. M., and John M. Evans.  Judicial Review of Administrative Action in Canada.  Toronto:  Canvasback, 1998 (loose‑leaf updated July 2007).

Cromwell, Thomas A.  “Appellate Review:  Policy and Pragmatism”.  In 2006 Pitblado Lectures, Appellate Courts:  Policy, Law and Practice.  Winnipeg:  Fort Garry, 2006, V‑1.

de Smith, Stanley A. Judicial Review of Administrative Action, 5th ed. By Lord Woolf and Jeffrey Jowell.  London:  Sweet & Maxwell, 1995.

Dyzenhaus, David.  “The Politics of Deference:  Judicial Review and Democracy”, in Michael Taggart, ed., The Province of Administrative Law.  Oxford:  Hart Publishing, 1997, 279.

England, Geoff.  Employment Law in Canada, 4th ed.  Markham, Ont.:  LexisNexis Butterworths, 2005 (loose‑leaf updated March 2007, release 10).

Hogg, Peter W., and Patrick J. Monahan.  Liability of the Crown, 3rd ed.  Scarborough, Ont.:  Carswell, 2000.

Mullan, David J.  Administrative Law.  Toronto:  Irwin Law, 2001.

Mullan, David J.  “Recent Developments in Standard of Review”, in Taking the Tribunal to Court:  A Practical Guide for Administrative Law Practitioners. Canadian Bar Association (Ontario), October 20, 2000.

Mullan, David J. “Establishing the Standard of Review:  The Struggle for Complexity?” (2004), 17 C.J.A.L.P. 59.

Sossin, Lorne, and Colleen M. Flood, “The Contextual Turn:  Iacobucci’s Legacy and the Standard of Review in Administrative Law” (2007), 57 U.T.L.J. 581.

Wade, Sir William.  Administrative Law, 8th ed. by Sir William Wade and Christopher Forsyth.  New York:  Oxford University Press, 2000.

APPEAL from a judgment of the New Brunswick Court of Appeal (Turnbull, Daigle and Robertson JJ.A.) (2006), 297 N.B.R. (2d) 151, 265 D.L.R. (4th) 609, 44 Admin. L.R. (4th) 92, 48 C.C.E.L. (3d) 196, 2006 CLLC ¶220‑030, [2006] N.B.J. No. 118 (QL), 2006 CarswellNB 155, 2006 NBCA 27, affirming a judgment of Rideout J. (2005), 293 N.B.R. (2d) 5, 43 C.C.E.L. (3d) 205, [2005] N.B.J. No. 327 (QL), 2005 CarswellNB 444, 2005 NBQB 270, quashing a preliminary ruling and quashing in part an award made by an adjudicator.  Appeal dismissed.

J. Gordon Petrie, Q.C., and Clarence L. Bennett, for the appellant.

C. Clyde Spinney, Q.C., and Keith P. Mullin, for the respondent.

The judgment of McLachlin C.J. and Bastarache, LeBel, Fish and Abella JJ. was delivered by

Bastarache and LeBel JJ. —

I.  Introduction

[1]                              This appeal calls on the Court to consider, once again, the troubling question of the approach to be taken in judicial review of decisions of administrative tribunals. The recent history of judicial review in Canada has been marked by ebbs and flows of deference, confounding tests and new words for old problems, but no solutions that provide real guidance for litigants, counsel, administrative decision makers or judicial review judges.  The time has arrived for a reassessment of the question.

A.  Facts

[2]                              The appellant, David Dunsmuir, was employed by the Department of Justice for the Province of New Brunswick.  His employment began on February 25, 2002, as a Legal Officer in the Fredericton Court Services Branch.  The appellant was placed on an initial six-month probationary term.  On March 14, 2002, by Order‑in‑Council, he was appointed to the offices of Clerk of the Court of Queen’s Bench, Trial Division, Administrator of the Court of Queen’s Bench, Family Division, and Clerk of the Probate Court of New Brunswick, all for the Judicial District of Fredericton.

[3]                              The employment relationship was not perfect.  The appellant’s probationary period was extended twice, to the maximum 12 months.  At the end of each probationary period, the appellant was given a performance review.  The first such review, which occurred in August 2002, identified four specific areas for improvement.  The second review, three months later, cited the same four areas for development, but noted improvements in two.  At the end of the third probationary period, the Regional Director of Court Services noted that the appellant had met all expectations and his employment was continued on a permanent basis.

[4]                              The employer reprimanded the appellant on three separate occasions during the course of his employment.  The first incident occurred in July 2002.  The appellant had sent an email to the Chief Justice of the Court of Queen’s Bench objecting to a request that had been made by the judge of the Fredericton Judicial District for the preparation of a practice directive.  The Regional Director issued a reprimand letter to the appellant, explaining that the means he had used to raise his concerns were inappropriate and exhibited serious error in judgment.  In the event that a similar concern arose in the future, he was directed to discuss the matter first with the Registrar or the Regional Director.  The letter warned that failure to comply would lead to additional disciplinary measures and, if necessary, to dismissal.

[5]                              A second disciplinary measure occurred when, in April 2004, it came to the attention of the Assistant Deputy Minister that the appellant was being advertised as a lecturer at legal seminars offered in the private sector.  The appellant had inquired previously into the possibility of doing legal work outside his employment.  In February 2004, the Assistant Deputy Minister had informed him that lawyers in the public service should not practise law in the private sector.  A month later, the appellant wrote a letter to the Law Society of New Brunswick stating that his participation as a non-remunerated lecturer had been vetted by his employer, who had voiced no objection.  On June 3, 2004, the Assistant Deputy Minister issued to the appellant written notice of a one-day suspension with pay regarding the incident.  The letter also referred to issues regarding the appellant’s work performance, including complaints from unnamed staff, lawyers and members of the public regarding his difficulties with timeliness and organization.  This second letter concluded with the statement that “[f]uture occurrences of this nature and failure to develop more efficient organized work habits will result in disciplinary action up to and including dismissal.”

[6]                              Third, on July 21, 2004, the Regional Director wrote a formal letter of reprimand to the appellant regarding three alleged incidents relating to his job performance.  This letter, too, concluded with a warning that the appellant’s failure to improve his organization and timeliness would result in further disciplinary action up to and including dismissal.  The appellant responded to the letter by informing the Regional Director that he would be seeking legal advice and, until that time, would not meet with her to discuss the matter further.

[7]                              A review of the appellant’s work performance had been due in April 2004 but did not take place. The appellant met with the Regional Director on a couple of occasions to discuss backlogs and organizational problems.  Complaints were relayed to her by staff but they were not documented and it is unknown how many complaints there had been.  The Regional Director notified the appellant on August 11, 2004, that his performance review was overdue and would occur by August 20.  A meeting had been arranged for August 19 between the appellant, the Regional Director, the Assistant Deputy Minister and counsel for the appellant and the employer.  While preparing for that meeting, the Regional Director and the Assistant Deputy Minister concluded that the appellant was not right for the job.  The scheduled meeting was cancelled and a termination notice was faxed to the appellant.  A formal letter of termination from the Deputy Minister was delivered to the appellant’s lawyer the next day.  The letter terminated the appellant’s employment with the Province of New Brunswick, effective December 31, 2004.  It read, in relevant part:

I regret to advise you that I have come to the conclusion that your particular skill set does not meet the needs of your employer in your current position, and that it is advisable to terminate your employment on reasonable notice, pursuant to section 20 of the Civil Service Act.  You are accordingly hereby advised that your employment with the Province of New Brunswick will terminate on December 31, 2004.  Cause for termination is not alleged.

To aid in your search for other employment, you are not required to report to work during the notice period and your salary will be continued until the date indicated or for such shorter period as you require either to find a job with equivalent remuneration, or you commence self‑employment.

. . .

In the circumstances, we would request that you avoid returning to the workplace until your departure has been announced to staff, and until you have returned your keys and government identification to your supervisor, Ms. Laundry as well as any other property of the employer still in your possession . . . .

[8]                              On February 3, 2005, the appellant was removed from his statutory offices by order of the Lieutenant-Governor in Council.

[9]                              The appellant commenced the grievance process under s. 100.1 of the Public Service Labour Relations Act, R.S.N.B. 1973, c. P-25 (“PSLRA”; see Appendix), by letter to the Deputy Minister on September 1, 2004. That provision grants non-unionized employees of the provincial public service the right to file a grievance with respect to a “discharge, suspension or a financial penalty” (s. 100.1(2)). The appellant asserted several grounds of complaint in his grievance letter, in particular, that the reasons for the employer’s dissatisfaction were not made known; that he did not receive a reasonable opportunity to respond to the employer’s concerns; that the employer’s actions in terminating him were without notice, due process or procedural fairness; and that the length of the notice period was inadequate.  The grievance was denied.  The appellant then gave notice that he would refer the grievance to adjudication under the PSLRA.  The adjudicator was selected by agreement of the parties and appointed by the Labour and Employment Board.

[10]                          The adjudication hearing was convened and counsel for the appellant produced as evidence a volume of 169 documents.  Counsel for the respondent objected to the inclusion of almost half of the documents.  The objection was made on the ground that the documents were irrelevant since the appellant’s dismissal was not disciplinary but rather was a termination on reasonable notice.  The preliminary issue therefore arose of whether, where dismissal was with notice or pay in lieu thereof, the adjudicator was authorized to assess the reasons underlying the province’s decision to terminate.  Following his preliminary ruling on that issue, the adjudicator heard and decided the merits of the grievance.

B.  Decisions of the Adjudicator

(1)  Preliminary Ruling (January 10, 2005)

[11]                          The adjudicator began his preliminary ruling by considering s. 97(2.1) of the PSLRA.  He reasoned that because the appellant was not included in a bargaining unit and there was no collective agreement or arbitral award, the section ought to be interpreted to mean that where an adjudicator determines that an employee has been discharged for cause, the adjudicator may substitute another penalty for the discharge as seems just and reasonable in the circumstances.  The adjudicator considered and relied on the decision of the New Brunswick Court of Appeal in Chalmers (Dr. Everett) Hospital v. Mills (1989), 102 N.B.R. (2d) 1.

[12]                          Turning to s. 100.1 of the PSLRA, he noted the referential incorporation of s. 97 in s. 100.1(5).  He stated that such incorporation “necessarily means that an adjudicator has jurisdiction to make the determination described in s. 97(2.1), i.e. that an employee has been discharged or otherwise disciplined for cause” (p. 5).  The adjudicator noted that an employee to whom s. 20 of the Civil Service Act, S.N.B. 1984, c. C-5.1 (see Appendix), applies may be discharged for cause, with reasonable notice or with pay in lieu of reasonable notice.  He concluded by holding that an employer cannot avoid an inquiry into its real reasons for dismissing an employee by stating that cause is not alleged.  Rather, a grieving employee is entitled to an adjudication as to whether a discharge purportedly with notice or pay in lieu thereof was in fact for cause.  He therefore held that he had jurisdiction to make such a determination.

(2)  Ruling on the Merits (February 16, 2005)

[13]                          In his decision on the merits, released shortly thereafter, the adjudicator found that the termination letter of August 19 effected termination with pay in lieu of notice.  The employer did not allege cause.  Inquiring into the reasons for dismissal the adjudicator was satisfied that, on his view of the evidence, the termination was not disciplinary.  Rather, the decision to terminate was based on the employer’s concerns about the appellant’s work performance and his suitability for the positions he held.

[14]                          The adjudicator then considered the appellant’s claim that he was dismissed without procedural fairness in that the employer did not inform him of the reasons for its dissatisfaction and did not give him an opportunity to respond.  The adjudicator placed some responsibility on the employer for cancelling the performance review scheduled for August 19.  He also opined that the employer was not so much dissatisfied with the appellant’s quality of work as with his lack of organization.

[15]                          The adjudicator’s decision relied on Knight v. Indian Head School Division No. 19, [1990] 1 S.C.R. 653, for the relevant legal principles regarding the right of “at pleasure” office holders to procedural fairness.  As the appellant’s employment was “hybrid in character” (para. 53) _ he was both a Legal Officer under the Civil Service Act and, as Clerk, an office holder “at pleasure” _ the adjudicator held that the appellant was entitled to procedural fairness in the employer’s decision to terminate his employment.  He declared that the termination was void ab initio and ordered the appellant reinstated as of August 19, 2004, the date of dismissal.

[16]                          The adjudicator added that in the event that his reinstatement order was quashed on judicial review, he would find the appropriate notice period to be eight months.

C.  Judicial History

(1)  Court of Queen’s Bench of New Brunswick (2005), 293 N.B.R. (2d) 5, 2005 NBQB 270

[17]                          The Province of New Brunswick applied for judicial review of the adjudicator’s decision on numerous grounds.  In particular, it argued that the adjudicator had exceeded his jurisdiction in his preliminary ruling by holding that he was authorized to determine whether the termination was in fact for cause. The Province further argued that the adjudicator had acted incorrectly or unreasonably in deciding the procedural fairness issue.  The application was heard by Rideout J.

[18]                          The reviewing judge applied a pragmatic and functional analysis, considering the presence of a full privative clause in the PSLRA, the relative expertise of adjudicators appointed under the PSLRA, the purposes of ss. 97(2.1) and 100.1 of the PSLRA as well as s. 20 of the Civil Service Act, and the nature of the question as one of statutory interpretation.  He concluded that the correctness standard of review applied and that the court need not show curial deference to the decision of an adjudicator regarding the interpretation of those statutory provisions.

[19]                          Regarding the preliminary ruling, the reviewing judge noted that the appellant was employed “at pleasure” and fell under s. 20 of the Civil Service Act.  In his view, the adjudicator had overlooked the effects of s. 20 and had mistakenly given ss. 97(2.1) and 100.1 of the PSLRA a substantive, rather than procedural, interpretation.  Those sections are procedural in nature.  They provide an employee with a right to grieve his or her dismissal and set out the steps that must be followed to pursue a grievance.  The adjudicator is bound to apply the contractual provisions as they exist and has no authority to change those provisions.  Thus, in cases in which s. 20 of the Civil Service Act applies, the adjudicator must apply the ordinary rules of contract.  The reviewing judge held that the adjudicator had erred in removing the words “and the collective agreement or arbitral award does not contain a specific penalty for the infraction that resulted in the employee being discharged or otherwise disciplined” from s. 97(2.1).  Those words limit s. 97(2.1) to employees who are not employed “at pleasure”.  In the view of the reviewing judge, the adjudicator did not have jurisdiction to inquire into the reasons for the termination.  His authority was limited to determining whether the notice period was reasonable.  Having found that the adjudicator had exceeded his jurisdiction, the reviewing judge quashed his preliminary ruling.

[20]                          With respect to the adjudicator’s award on the merits, the reviewing judge commented that some aspects of the decision are factual in nature and should be reviewed on a patent unreasonableness standard, while other aspects involve questions of mixed fact and law which are subject to a reasonableness simpliciter standard.  The reviewing judge agreed with the Province that the adjudicator’s reasons do not stand up to a “somewhat probing examination” (para. 76).  The reviewing judge held that the adjudicator’s award of reinstatement could not stand as he was not empowered by the PSLRA to make Lieutenant-Governor in Council appointments.  In addition, by concluding that the decision was void ab initio owing to a lack of procedural fairness, the adjudicator failed to consider the doctrine of adequate alternative remedy.  The appellant received procedural fairness by virtue of the grievance hearing before the adjudicator.  The adjudicator had provisionally increased the notice period to eight months _ that provided an adequate alternative remedy.  Concluding that the adjudicator’s decision did not stand up to review on a reasonableness simpliciter standard, the reviewing judge quashed the reinstatement order but upheld the adjudicator’s provisional award of eight months’ notice.

(2)  Court of Appeal of New Brunswick (2006), 297 N.B.R. (2d) 151, 2006 NBCA 27

[21]                          The appellant appealed the decision of the reviewing judge.  The Court of Appeal, Robertson J.A. writing, held that the proper standard with respect to the interpretation of the adjudicator’s authority under the PSLRA was reasonableness simpliciter and that the reviewing judge had erred in adopting the correctness standard.  The court reached that conclusion by proceeding through a pragmatic and functional analysis, placing particular emphasis on the presence of a full privative clause in the PSLRA and the relative expertise of an adjudicator in the labour relations and employment context. The court also relied on the decision of this Court in Alberta Union of Provincial Employees v. Lethbridge Community College, [2004] 1 S.C.R. 727, 2004 SCC 28.  However, the court noted that the adjudicator’s interpretation of the Mills decision warranted no deference and that “correctness is the proper review standard when it comes to the interpretation and application of caselaw” (para. 17).

[22]                          Applying the reasonableness simpliciter standard, the court held that the adjudicator’s decision was unreasonable.  Robertson J.A. began by considering s. 20 of the Civil Service Act and noted that under the ordinary rules of contract, an employer holds the right to dismiss an employee with cause or with reasonable notice or with pay in lieu of notice.  Section 20 of the Civil Service Act limits the Crown’s common law right to dismiss its employees without cause or notice.  Robertson J.A. reasoned that s. 97(2.1) of the PSLRA applies in principle to non-unionized employees, but that it is only where an employee has been discharged or disciplined for cause that an adjudicator may substitute such other penalty as seems just and reasonable in the circumstances.  Where the employer elects to dismiss with notice or pay in lieu of notice, however, s. 97(2.1) does not apply.  In such circumstances, the employee may only grieve the length of the notice period.  The only exception is where the employee alleges that the decision to terminate was based on a prohibited ground of discrimination.

[23]                          On the issue of procedural fairness, the court found that the appellant exercised his right to grieve, and thus a finding that the duty of fairness had been breached was without legal foundation.  The court dismissed the appeal.

II.  Issues

[24]                          At issue, firstly is the approach to be taken in the judicial review of a decision of a particular adjudicative tribunal which was seized of a grievance filed by the appellant after his employment was terminated.  This appeal gives us the opportunity to re-examine the foundations of judicial review and the standards of review applicable in various situations.

[25]                          The second issue involves examining whether the appellant who held an office “at pleasure” in the civil service of New Brunswick, had the right to procedural fairness in the employer’s decision to terminate him.  On this occasion, we will reassess the rule that has found formal expression in Knight.

[26]                          The two types of judicial review, on the merits and on the process, are therefore engaged in this case.  Our review of the system will therefore be comprehensive, which is preferable since a holistic approach is needed when considering fundamental principles.

III.  Issue 1: Review of the Adjudicator’s Statutory Interpretation Determination

A.  Judicial Review

[27]                          As a matter of constitutional law, judicial review is intimately connected with the preservation of the rule of law.  It is essentially that constitutional foundation which explains the purpose of judicial review and guides its function and operation.  Judicial review seeks to address an underlying tension between the rule of law and the foundational democratic principle, which finds an expression in the initiatives of Parliament and legislatures to create various administrative bodies and endow them with broad powers. Courts, while exercising their constitutional functions of judicial review, must be sensitive not only to the need to uphold the rule of law, but also to the necessity of avoiding undue interference with the discharge of administrative functions in respect of the matters delegated to administrative bodies by Parliament and legislatures.

[28]                          By virtue of the rule of law principle, all exercises of public authority must find their source in law.  All decision-making powers have legal limits, derived from the enabling statute itself, the common or civil law or the Constitution.   Judicial review is the means by which the courts supervise those who exercise statutory powers, to ensure that they do not overstep their legal authority.  The function of judicial review is therefore to ensure the legality, the reasonableness and the fairness of the administrative process and its outcomes.

[29]                          Administrative powers are exercised by decision makers according to statutory regimes that are themselves confined.  A decision maker may not exercise authority not specifically assigned to him or her.  By acting in the absence of legal authority, the decision maker transgresses the principle of the rule of law.  Thus, when a reviewing court considers the scope of a decision-making power or the jurisdiction conferred by a statute, the standard of review analysis strives to determine what authority was intended to be given to the body in relation to the subject matter.  This is done within the context of the courts’ constitutional duty to ensure that public authorities do not overreach their lawful powers: Crevier v. Attorney General of Quebec, [1981] 2 S.C.R. 220, at p. 234; also Dr. Q v. College of Physicians and Surgeons of British Columbia, [2003] 1 S.C.R. 226, 2003 SCC 19, at para. 21.

[30]                          In addition to the role judicial review plays in upholding the rule of law, it also performs an important constitutional function in maintaining legislative supremacy. As noted by Justice Thomas Cromwell, “the rule of law is affirmed by assuring that the courts have the final say on the jurisdictional limits of a tribunal’s authority; second, legislative supremacy is affirmed by adopting the principle that the concept of jurisdiction should be narrowly circumscribed and defined according to the intent of the legislature in a contextual and purposeful way; third, legislative supremacy is affirmed and the court-centric conception of the rule of law is reined in by acknowledging that the courts do not have a monopoly on deciding all questions of law” (“Appellate Review: Policy and Pragmatism”, in 2006 Isaac Pitblado Lectures, Appellate Courts: Policy, Law and Practice, V-1, at p. V-12). In essence, the rule of law is maintained because the courts have the last word on jurisdiction, and legislative supremacy is assured because determining the applicable standard of review is accomplished by establishing legislative intent.

[31]                          The legislative branch of government cannot remove the judiciary’s power to review actions and decisions of administrative bodies for compliance with the constitutional capacities of the government.  Even a privative clause, which provides a strong indication of legislative intent, cannot be determinative in this respect (Executors of the Woodward Estate v. Minister of Finance, [1973] S.C.R. 120, at p. 127).  The inherent power of superior courts to review administrative action and ensure that it does not exceed its jurisdiction stems from the judicature provisions in ss. 96  to 101  of the Constitution Act, 1867 : Crevier.  As noted by Beetz J. in U.E.S., Local 298 v. Bibeault, [1988] 2 S.C.R. 1048, at p. 1090, “[t]he role of the superior courts in maintaining the rule of law is so important that it is given constitutional protection”.  In short, judicial review is constitutionally guaranteed in Canada, particularly with regard to the definition and enforcement of jurisdictional limits.  As Laskin C.J. explained in Crevier:

Where . . . questions of law have been specifically covered in a privative enactment, this Court, as in Farrah, has not hesitated to recognize this limitation on judicial review as serving the interests of an express legislative policy to protect decisions of adjudicative agencies from external correction.  Thus, it has, in my opinion, balanced the competing interests of a provincial Legislature in its enactment of substantively valid legislation and of the courts as ultimate interpreters of the British North America Act and s. 96 thereof.  The same considerations do not, however, apply to issues of jurisdiction which are not far removed from issues of constitutionality.  It cannot be left to a provincial statutory tribunal, in the face of s. 96, to determine the limits of its own jurisdiction without appeal or review. [pp. 237-38]

See also D. J. Mullan, Administrative Law (2001), at p. 50.

[32]                          Despite the clear, stable constitutional foundations of the system of judicial review, the operation of judicial review in Canada has been in a constant state of evolution over the years, as courts have attempted to devise approaches to judicial review that are both theoretically sound and effective in practice.  Despite efforts to refine and clarify it, the present system has proven to be difficult to implement.  The time has arrived to re-examine the Canadian approach to judicial review of administrative decisions and develop a principled framework that is more coherent and workable. 

[33]                          Although the instant appeal deals with the particular problem of judicial review of the decisions of an adjudicative tribunal, these reasons will address first and foremost the structure and characteristics of the system of judicial review as a whole. In the wake of Baker v. Canada (Minister of Citizenship and Immigration), [1999] 2 S.C.R. 817, Suresh v. Canada (Minister of Citizenship and Immigration), [2002] 1 S.C.R. 3, 2002 SCC 1, Mount Sinai Hospital Center v. Quebec (Minister of Health and Social Services), [2001] 2 S.C.R. 281, 2001 SCC 41, and C.U.P.E. v. Ontario (Minister of Labour), [2003] 1 S.C.R. 539, 2003 SCC 29, it has become apparent that the present system must be simplified.  The comments of LeBel J. in Chamberlain v. Surrey School District No. 36, [2002] 4 S.C.R. 710, 2002 SCC 86, at paras. 190 and 195, questioning the applicability of the “pragmatic and functional approach” to the decisions and actions of all kinds of administrative actors, illustrated the need for change.

B.  Reconsidering the Standards of Judicial Review

[34]                          The current approach to judicial review involves three standards of review, which range from correctness, where no deference is shown, to patent unreasonableness, which is most deferential to the decision maker, the standard of reasonableness simpliciter lying, theoretically, in the middle.  In our view, it is necessary to reconsider both the number and definitions of the various standards of review, and the analytical process employed to determine which standard applies in a given situation.  We conclude that there ought to be two standards of review _ correctness and reasonableness.

[35]                          The existing system of judicial review has its roots in several landmark decisions beginning in the late 1970s in which this Court developed the theory of substantive review to be applied to determinations of law, and determinations of fact and of mixed law and fact made by administrative tribunals.  In Canadian Union of Public Employees, Local 963 v. New Brunswick Liquor Corp., [1979] 2 S.C.R. 227 (“CUPE”), Dickson J. introduced the idea that, depending on the legal and administrative contexts, a specialized administrative tribunal with particular expertise, which has been given the protection of a privative clause, if acting within its jurisdiction, could provide an interpretation of its enabling legislation that would be allowed to stand unless “so patently unreasonable that its construction cannot be rationally supported by the relevant legislation and demands intervention by the court upon review” (p. 237).  Prior to CUPE, judicial review followed the “preliminary question doctrine”, which inquired into whether a tribunal had erred in determining the scope of its jurisdiction.  By simply branding an issue as “jurisdictional”, courts could replace a decision of the tribunal with one they preferred, often at the expense of a legislative intention that the matter lie in the hands of the administrative tribunal.  CUPE marked a significant turning point in the approach of courts to judicial review, most notably in Dickson J.’s warning that courts “should not be alert to brand as jurisdictional, and therefore subject to broader curial  review, that which may be doubtfully so” (p. 233).  Dickson J.’s policy of judicial respect for administrative decision making marked the beginning of the modern era of Canadian administrative law.

[36]                          CUPE did not do away with correctness review altogether and in Bibeault, the Court affirmed that there are still questions on which a tribunal must be correct.  As Beetz J. explained, “the jurisdiction conferred on administrative tribunals and other bodies created by statute is limited, and . . . such a tribunal cannot by a misinterpretation of an enactment assume a power not given to it by the legislator” (p. 1086).  Bibeault introduced the concept of a “pragmatic and functional analysis” to determine the jurisdiction of a tribunal, abandoning the “preliminary question” theory.  In arriving at the appropriate standard of review, courts were to consider a number of factors including the wording of the provision conferring jurisdiction on the tribunal, the purpose of the enabling statute, the reason for the existence of the tribunal, the expertise of its members, and the nature of the problem (p. 1088).  The new approach would put “renewed emphasis on the superintending and reforming function of the superior courts” (p. 1090).  The “pragmatic and functional analysis”, as it came to be known, was later expanded to determine the appropriate degree of deference in respect of various forms of administrative decision making.

[37]                          In Canada (Director of Investigation and Research) v. Southam Inc., [1997] 1 S.C.R. 748, a third standard of review was introduced into Canadian administrative law.  The legislative context of that case, which provided a statutory right of appeal from the decision of a specialized tribunal, suggested that none of the existing standards was entirely satisfactory. As a result, the reasonableness simpliciter standard was introduced.  It asks whether the tribunal’s decision was reasonable.  If so, the decision should stand; if not, it must fall.  In Southam, Iacobucci J. described an unreasonable decision as one that “is not supported by any reasons that can stand up to a somewhat probing examination” (para. 56) and explained that the difference between patent unreasonableness and reasonableness simpliciter is the “immediacy” or “obviousness” of the defect in the tribunal’s decision (para. 57).  The defect will appear on the face of a patently unreasonable decision, but where the decision is merely unreasonable, it will take a searching review to find the defect. 

[38]                          The three standards of review have since remained in Canadian administrative law, the approach to determining the appropriate standard of review having been refined in Pushpanathan v. Canada (Minister of Citizenship and Immigration), [1998] 1 S.C.R. 982.

[39]                          The operation of three standards of review has not been without practical and theoretical difficulties, neither has it been free of criticism.  One major problem lies in distinguishing between the patent unreasonableness standard and the reasonableness simpliciter standard.  The difficulty in distinguishing between those standards contributes to the problem of choosing the right standard of review.  An even greater problem lies in the application of the patent unreasonableness standard, which at times seems to require parties to accept an unreasonable decision.

[40]                          The definitions of the patent unreasonableness standard that arise from the case law tend to focus on the magnitude of the defect and on the immediacy of the defect (see Toronto (City) v. C.U.P.E., Local 79, [2003] 3 S.C.R. 77, 2003 SCC 63, at para. 78, per LeBel J.).  Those two hallmarks of review under the patent unreasonableness standard have been used consistently in the jurisprudence to distinguish it from review under the standard of reasonableness simpliciter.  As it had become clear that, after Southam, lower courts were struggling with the conceptual distinction between patent unreasonableness and reasonableness simpliciter, Iacobucci J., writing for the Court in Law Society of New Brunswick v. Ryan, [2003] 1 S.C.R. 247, 2003 SCC 20, attempted to bring some clarity to the issue.  He explained the different operations of the two deferential standards as follows, at paras. 52-53:

[A] patently unreasonable defect, once identified, can be explained simply and easily, leaving no real possibility of doubting that the decision is defective.  A patently unreasonable decision has been described as “clearly irrational” or “evidently not in accordance with reason” . . . . A decision that is patently unreasonable is so flawed that no amount of curial deference can justify letting it stand.

A decision may be unreasonable without being patently unreasonable when the defect in the decision is less obvious and might only be discovered after “significant searching or testing” (Southam, supra, at para. 57).  Explaining the defect may require a detailed exposition to show that there are no lines of reasoning supporting the decision which could reasonably lead that tribunal to reach the decision it did.

[41]                          As discussed by LeBel J. at length in Toronto (City) v. C.U.P.E., notwithstanding the increased clarity that Ryan brought to the issue and the theoretical differences between the standards of patent unreasonableness and reasonableness simpliciter, a review of the cases reveals that any actual difference between them in terms of their operation appears to be illusory (see also the comments of Abella J. in Council of Canadians with Disabilities v. Via Rail Canada Inc., [2007] 1 S.C.R. 650, 2007 SCC 15, at paras. 101‑3).  Indeed, even this Court divided when attempting to determine whether a particular decision was “patently unreasonable”, although this should have been self-evident under the existing test (see C.U.P.E. v. Ontario (Minister of Labour)).  This result is explained by the fact that both standards are based on the idea that there might be multiple valid interpretations of a statutory provision or answers to a legal dispute and that courts ought not to interfere where the tribunal’s decision is rationally supported.  Looking to either the magnitude or the immediacy of the defect in the tribunal’s decision provides no meaningful way in practice of distinguishing between a patently unreasonable and an unreasonable decision.  As Mullan has explained:

[T]o maintain a position that it is only the “clearly irrational” that will cross the threshold of patent unreasonableness while irrationality simpliciter will not is to make a nonsense of the law.  Attaching the adjective “clearly” to irrational is surely a tautology.  Like “uniqueness”, irrationality either exists or it does not.  There cannot be shades of irrationality.

See D. J. Mullan, “Recent Developments in Standard of Review”, in Canadian Bar Association (Ontario), Taking the Tribunal to Court: A Practical Guide for Administrative Law Practitioners (2000), at p. 25.

[42]                          Moreover, even if one could conceive of a situation in which a clearly or highly irrational decision were distinguishable from a merely irrational decision, it would be unpalatable to require parties to accept an irrational decision simply because, on a deferential standard, the irrationality of the decision is not clear enough.  It is also inconsistent with the rule of law to retain an irrational decision.  As LeBel J. explained in his concurring reasons in Toronto (City) v. C.U.P.E., at para. 108:

In the end, the essential question remains the same under both standards: was the decision of the adjudicator taken in accordance with reason?  Where the answer is no, for instance because the legislation in question cannot rationally support the adjudicator’s interpretation, the error will invalidate the decision, regardless of whether the standard applied is reasonableness simpliciter or patent unreasonableness . . . .

See also Voice Construction Ltd. v. Construction & General Workers’ Union, Local 92, [2004] 1 S.C.R. 609, 2004 SCC 23, at paras. 40-41, per LeBel J.

C.  Two Standards of Review

[43]                          The Court has moved from a highly formalistic, artificial “jurisdiction” test that could easily be manipulated, to a highly contextual “functional” test that provides great flexibility but little real on-the-ground guidance, and offers too many standards of review. What is needed is a test that offers guidance, is not formalistic or artificial, and permits review where justice requires it, but not otherwise. A simpler test is needed.

(1)  Defining the Concepts of Reasonabless and Correctness

[44]                          As explained above, the patent unreasonableness standard was developed many years prior to the introduction of the reasonableness simpliciter standard in Southam.  The intermediate standard was developed to respond to what the Court viewed as problems in the operation of judicial review in Canada, particularly the perceived all-or-nothing approach to deference, and in order to create a more finely calibrated system of judicial review (see also L. Sossin and C. M. Flood, “The Contextual Turn: Iacobucci’s Legacy and the Standard of Review in Administrative Law” (2007), 57 U.T.L.J. 581).  However, the analytical problems that arise in trying to apply the different standards undercut any conceptual usefulness created by the inherently greater flexibility of having multiple standards of review.  Though we are of the view that the three-standard model is too difficult to apply to justify its retention, now, several years after Southam, we believe that it would be a step backwards to simply remove the reasonableness simpliciter standard and revert to pre-Southam law.  As we see it, the problems that Southam attempted to remedy with the introduction of the intermediate standard are best addressed not by three standards of review, but by two standards, defined appropriately.

[45]                          We therefore conclude that the two variants of reasonableness review should be collapsed into a single form of “reasonableness” review.  The result is a system of judicial review comprising two standards _ correctness and reasonableness.  But the revised system cannot be expected to be simpler and more workable unless the concepts it employs are clearly defined.

[46]                          What does this revised reasonableness standard mean?  Reasonableness is one of the most widely used and yet most complex legal concepts.  In any area of the law we turn our attention to, we find ourselves dealing with the reasonable, reasonableness or rationality.  But what is a reasonable decision?  How are reviewing courts to identify an unreasonable decision in the context of administrative law and, especially, of judicial review?

[47]                          Reasonableness is a deferential standard animated by the principle that underlies the development of the two previous standards of reasonableness: certain questions that come before administrative tribunals do not lend themselves to one specific, particular result.  Instead, they may give rise to a number of possible, reasonable conclusions.  Tribunals have a margin of appreciation within the range of acceptable and rational solutions.  A court conducting a review for reasonableness inquires into the qualities that make a decision reasonable, referring both to the process of articulating the reasons and to outcomes.  In judicial review, reasonableness is concerned mostly with the existence of justification, transparency and intelligibility within the decision-making process.  But it is also concerned with whether the decision falls within a range of possible, acceptable outcomes which are defensible in respect of the facts and law.

[48]                          The move towards a single reasonableness standard does not pave the way for a more intrusive review by courts and does not represent a return to pre-Southam formalism.  In this respect, the concept of deference, so central to judicial review in administrative law, has perhaps been insufficiently explored in the case law.  What does deference mean in this context?  Deference is both an attitude of the court and a requirement of the law of judicial review.  It does not mean that courts are subservient to the determinations of decision makers, or that courts must show blind reverence to their interpretations, or that they may be content to pay lip service to the concept of reasonableness review while in fact imposing their own view.  Rather, deference imports respect for the decision-making process of adjudicative bodies with regard to both the facts and the law.  The notion of deference “is rooted in part in a respect for governmental decisions to create administrative bodies with delegated powers” (Canada (Attorney General) v. Mossop, [1993] 1 S.C.R. 554, at p. 596, per L’Heureux-Dubé J., dissenting).  We agree with David Dyzenhaus where he states that the concept of “deference as respect” requires of the courts “not submission but a respectful attention to the reasons offered or which could be offered in support of a decision”: “The Politics of Deference: Judicial Review and Democracy”, in M. Taggart, ed., The Province of Administrative Law (1997), 279, at p. 286 (quoted with approval in Baker, at para. 65, per L’Heureux-Dubé J.; Ryan, at para. 49).

[49]                          Deference in the context of the reasonableness standard therefore implies that courts will give due consideration to the determinations of decision makers.  As Mullan explains, a policy of deference “recognizes the reality that, in many instances, those working day to day in the implementation of frequently complex administrative schemes have or will develop a considerable degree of expertise or field sensitivity to the imperatives and nuances of the legislative regime”: D. J. Mullan, “Establishing the Standard of Review: The Struggle for Complexity?” (2004), 17 C.J.A.L.P. 59, at p. 93.  In short, deference requires respect for the legislative choices to leave some matters in the hands of administrative decision makers, for the processes and determinations that draw on particular expertise and experiences, and for the different roles of the courts and administrative bodies within the Canadian constitutional system.

[50]                          As important as it is that courts have a proper understanding of reasonableness review as a deferential standard, it is also without question that the standard of correctness must be maintained in respect of jurisdictional and some other questions of law.  This promotes just decisions and avoids inconsistent and unauthorized application of law.  When applying the correctness standard, a reviewing court will not show deference to the decision maker’s reasoning process; it will rather undertake its own analysis of the question.  The analysis will bring the court to decide whether it agrees with the determination of the decision maker; if not, the court will substitute its own view and provide the correct answer. From the outset, the court must ask whether the tribunal’s decision was correct.

(2)  Determining the Appropriate Standard of Review

[51]                          Having dealt with the nature of the standards of review, we now turn our attention to the method for selecting the appropriate standard in individual cases.  As we will now demonstrate, questions of fact, discretion and policy as well as questions where the legal issues cannot be easily separated from the factual issues generally attract a standard of reasonableness while many legal issues attract a standard of correctness.  Some legal issues, however, attract the more deferential standard of reasonableness.

[52]                          The existence of a privative or preclusive clause gives rise to a strong indication of review pursuant to the reasonableness standard.  This conclusion is appropriate because a privative clause is evidence of Parliament or a legislature’s intent that an administrative decision maker be given greater deference and that interference by reviewing courts be minimized.  This does not mean, however, that the presence of a privative clause is determinative.  The rule of law requires that the constitutional role of superior courts be preserved and, as indicated above, neither Parliament nor any legislature can completely remove the courts’ power to review the actions and decisions of administrative bodies.  This power is constitutionally protected.  Judicial review is necessary to ensure that the privative clause is read in its appropriate statutory context and that administrative bodies do not exceed their jurisdiction.

[53]                          Where the question is one of fact, discretion or policy, deference will usually apply automatically (Mossop, at pp. 599-600; Dr. Q, at para. 29; Suresh, at paras. 29-30).  We believe that the same standard must apply to the review of questions where the legal and factual issues are  intertwined with and cannot be readily separated.

[54]                          Guidance with regard to the questions that will be reviewed on a reasonableness standard can be found in the existing case law.  Deference will usually result where a tribunal is interpreting its own statute or statutes closely connected to its function, with which it will have particular familiarity: Canadian Broadcasting Corp. v. Canada (Labour Relations Board), [1995] 1 S.C.R. 157, at para. 48; Toronto (City) Board of Education v. O.S.S.T.F., District 15, [1997] 1 S.C.R. 487, at para. 39.  Deference may also be warranted where an administrative tribunal has developed particular expertise in the application of a general common law or civil law rule in relation to a specific statutory context: Toronto (City) v. C.U.P.E., at para. 72.  Adjudication in labour law remains a good example of the relevance of this approach.  The case law has moved away considerably from the strict position evidenced in McLeod v. Egan, [1975] 1 S.C.R. 517, where it was held that an administrative decision maker will always risk having its interpretation of an external statute set aside upon judicial review.

[55]                          A consideration of the following factors will lead to the conclusion that the decision maker should be given deference and a reasonableness test applied:

_     A privative clause: this is a statutory direction from Parliament or a legislature indicating the need for deference.

_     A discrete and special administrative regime in which the decision maker has special expertise (labour relations for instance).

_     The nature of the question of law.  A question of law that is of “central importance to the legal system . . . and outside the . . . specialized area of expertise” of the administrative decision maker will always attract a correctness standard (Toronto (City) v. C.U.P.E., at para. 62).  On the other hand, a question of law that does not rise to this level may be compatible with a reasonableness standard where the two above factors so indicate.

[56]                           If these factors, considered together, point to a standard of reasonableness, the decision maker’s decision must be approached with deference in the sense of respect discussed earlier in these reasons.  There is nothing unprincipled in the fact that some questions of law will be decided on the basis of reasonableness.  It simply means giving the adjudicator’s decision appropriate deference in deciding whether a decision should be upheld, bearing in mind the factors indicated.

[57]                           An exhaustive review is not required in every case to determine the proper standard of review.  Here again, existing jurisprudence may be helpful in identifying some of the questions that generally fall to be determined according to the correctness standard (Cartaway Resources Corp. (Re), [2004] 1 S.C.R. 672, 2004 SCC 26). This simply means that the analysis required is already deemed to have been performed and need not be repeated.

[58]                           For example, correctness review has been found to apply to constitutional questions regarding the division of powers between Parliament and the provinces in the Constitution Act, 1867 : Westcoast Energy Inc. v. Canada (National Energy Board), [1998] 1 S.C.R. 322.  Such questions, as well as other constitutional issues, are necessarily subject to correctness review because of the unique role of s. 96 courts as interpreters of the Constitution: Nova Scotia (Workers’ Compensation Board) v. Martin, [2003] 2 S.C.R. 504, 2003 SCC 54; Mullan, Administrative Law, at p. 60.

[59]                           Administrative bodies must also be correct in their determinations of true questions of jurisdiction or vires. We mention true questions of vires to distance ourselves from the extended definitions adopted before CUPE. It is important here to take a robust view of jurisdiction.  We neither wish nor intend to return to the jurisdiction/preliminary question doctrine that plagued the jurisprudence in this area for many years.  “Jurisdiction” is intended in the narrow sense of whether or not the tribunal had the authority to make the inquiry.  In other words, true jurisdiction questions arise where the tribunal must explicitly determine whether its statutory grant of power gives it the authority to decide a particular matter. The tribunal must interpret the grant of authority correctly or its action will be found to be ultra vires or to constitute a wrongful decline of jurisdiction: D. J. M. Brown and J. M. Evans, Judicial Review of Administrative Action in Canada (loose‑leaf), at pp. 14-3 to 14-6.  An example may be found in United Taxi Drivers’ Fellowship of Southern Alberta v. Calgary (City), [2004] 1 S.C.R. 485, 2004 SCC 19.  In that case, the issue was whether the City of Calgary was authorized under the relevant municipal acts to enact bylaws limiting the number of taxi plate licences (para. 5, per Bastarache J.).  That case involved the decision-making powers of a municipality and exemplifies a true question of jurisdiction or vires.  These questions will be narrow.  We reiterate the caution of Dickson J. in CUPE that reviewing judges must not brand as jurisdictional issues that are doubtfully so.

[60]                           As mentioned earlier, courts must also continue to substitute their own view of the correct answer where the question at issue is one of general law “that is both of central importance to the legal system as a whole and outside the adjudicator’s specialized area of expertise” (Toronto (City) v. C.U.P.E., at para. 62, per LeBel J.).  Because of their impact on the administration of justice as a whole, such questions require uniform and consistent answers.  Such was the case in Toronto (City) v. C.U.P.E., which dealt with complex common law rules and conflicting jurisprudence on the doctrines of res judicata and abuse of process _ issues that are at the heart of the administration of justice (see para. 15, per Arbour J.).

[61]                           Questions regarding the jurisdictional lines between two or more competing specialized tribunals have also been subject to review on a correctness basis: Regina Police Assn. Inc. v. Regina (City) Board of Police Commissioners, [2000] 1 S.C.R. 360, 2000 SCC 14; Quebec (Commission des droits de la personne et des droits de la jeunesse) v. Quebec (Attorney General), [2004] 2 S.C.R. 185, 2004 SCC 39.

[62]                           In summary, the process of judicial review involves two steps.  First, courts ascertain whether the jurisprudence has already determined in a satisfactory manner the degree of deference to be accorded with regard to a particular category of question.  Second, where the first inquiry proves unfruitful, courts must proceed to an analysis of the factors making it possible to identify the proper standard of review.

[63]                           The existing approach to determining the appropriate standard of review has commonly been referred to as “pragmatic and functional”.  That name is unimportant. Reviewing courts must not get fixated on the label at the expense of a proper understanding of what the inquiry actually entails.  Because the phrase “pragmatic and functional approach” may have misguided courts in the past, we prefer to refer simply to the “standard of review analysis” in the future.

[64]                           The analysis must be contextual.  As mentioned above, it is dependent on the application of a number of relevant factors, including: (1) the presence or absence of a privative clause; (2) the purpose of the tribunal as determined by interpretation of enabling legislation; (3) the nature of the question at issue, and; (4) the expertise of the tribunal.  In many cases, it will not be necessary to consider all of the factors, as some of them may be determinative in the application of the reasonableness standard in a specific case.

D.  Application

[65]                           Returning to the instant appeal and bearing in mind the foregoing discussion, we must determine the standard of review applicable to the adjudicator’s interpretation of the PSLRA, in particular ss. 97(2.1) and 100.1, and s. 20 of the Civil Service Act.  That standard of review must then be applied to the adjudicator’s decision.  In order to determine the applicable standard, we will now examine the factors relevant to the standard of review analysis.

(1)  Proper Standard of Review on the Statutory Interpretation Issue

[66]                           The specific question on this front is whether the combined effect of s. 97(2.1) and s. 100.1 of the PSLRA permits the adjudicator to inquire into the employer’s reason for dismissing an employee with notice or pay in lieu of notice.  This is a question of law.  The question to be answered is therefore whether in light of the privative clause, the regime under which the adjudicator acted, and the nature of the question of law involved, a standard of correctness should apply.

[67]                           The adjudicator was appointed and empowered under the PSLRA; s. 101(1) of that statute contains a full privative clause, stating in no uncertain terms that “every order, award, direction, decision, declaration or ruling of . . . an adjudicator is final and shall not be questioned or reviewed in any court”.  Section 101(2) adds that “[n]o order shall be made or process entered, and no proceedings shall be taken in any court, whether by way of injunction, judicial review, or otherwise, to question, review, prohibit or restrain . . . an adjudicator in any of its or his proceedings.”  The inclusion of a full privative clause in the PSLRA gives rise to a strong indication that the reasonableness standard of review will apply.

[68]                           The nature of the regime also favours the standard of reasonableness.  This Court has often recognized the relative expertise of labour arbitrators in the interpretation of collective agreements, and counselled that the review of their decisions should be approached with deference: CUPE, at pp. 235-36; Canada Safeway Ltd. v. RWDSU, Local 454, [1998] 1 S.C.R. 1079, at para. 58; Voice Construction, at para. 22.  The adjudicator in this case was, in fact, interpreting his enabling statute.  Although the adjudicator was appointed on an ad hoc basis, he was selected by the mutual agreement of the parties and, at an institutional level, adjudicators acting under the PSLRA can be presumed to hold relative expertise in the interpretation of the legislation that gives them their mandate, as well as related legislation that they might often encounter in the course of their functions.  See Alberta Union of Provincial Employees v. Lethbridge Community College.  This factor also suggests a reasonableness standard of review.

[69]                           The legislative purpose confirms this view of the regime.  The PSLRA establishes a time- and cost-effective method of resolving employment disputes.  It provides an alternative to judicial determination.  Section 100.1 of the PSLRA defines the adjudicator’s powers in deciding a dispute, but it also provides remedial protection for employees who are not unionized.  The remedial nature of s. 100.1 and its provision for timely and binding settlements of disputes also imply that a reasonableness review is appropriate.

[70]                           Finally, the nature of the legal question at issue is not one that is of central importance to the legal system and outside the specialized expertise of the adjudicator. This also suggests that the standard of reasonableness should apply. 

[71]                           Considering the privative clause, the nature of the regime, and the nature of the question of law here at issue, we conclude that the appropriate standard is reasonableness.  We must now apply that standard to the issue considered by the adjudicator in his preliminary ruling.

(2)  Was the Adjudicator’s Interpretation Unreasonable?

[72]                           While we are required to give deference to the determination of the adjudicator, considering the decision in the preliminary ruling as a whole, we are unable to accept that it reaches the standard of reasonableness.  The reasoning process of the adjudicator was deeply flawed.  It relied on and led to a construction of the statute that fell outside the range of admissible statutory interpretations.

[73]                           The adjudicator considered the New Brunswick Court of Appeal decision in Chalmers (Dr. Everett) Hospital v. Mills as well as amendments made to the PSLRA in 1990 (S.N.B. 1990, c. 30).  Under the former version of the Act, an employee could grieve “with respect to . . . disciplinary action resulting in discharge, suspension or a financial penalty” (s. 92(1)).  The amended legislation grants the right to grieve “with respect to discharge, suspension or a financial penalty” (PSLRA, s. 100.1(2)).  The adjudicator reasoned that the referential incorporation of s. 97(2.1) in s. 100.1(5) “necessarily means that an adjudicator has jurisdiction to make the determination described in subsection 97(2.1), i.e. that an employee has been discharged or otherwise disciplined for cause” (p. 5).  He further stated that an employer “cannot avoid an inquiry into its real reasons for a discharge, or exclude resort to subsection 97(2.1), by simply stating that cause is not alleged” (ibid. (emphasis added)).  The adjudicator concluded that he could determine whether a discharge purportedly with notice or pay in lieu of notice was in reality for cause.

[74]                           The interpretation of the law is always contextual.  The law does not operate in a vacuum.  The adjudicator was required to take into account the legal context in which he was to apply the law.  The employment relationship between the parties in this case was governed by private law. The contractual terms of employment could not reasonably be ignored.  That is made clear by s. 20 of the Civil Service Act.  Under the ordinary rules of contract, the employer is entitled to discharge an employee for cause, with notice or with pay in lieu of notice.  Where the employer chooses to exercise its right to discharge with reasonable notice or pay in lieu thereof, the employer is not required to assert cause for discharge.  The grievance process cannot have the effect of changing the terms of the contract of employment.  The respondent chose to exercise its right to terminate without alleging cause in this case.  By giving the PSLRA an interpretation that allowed him to inquire into the reasons for discharge where the employer had the right not to provide _ or even have _ such reasons, the adjudicator adopted a reasoning process that was fundamentally inconsistent with the employment contract and, thus, fatally flawed.  For this reason, the decision does not fall within the range of acceptable outcomes that are defensible in respect of the facts and the law.

[75]                           The decision of the adjudicator treated the appellant, a non-unionized employee, as a unionized employee.  His interpretation of the PSLRA, which permits an adjudicator to inquire into the reasons for discharge where notice is given and, under s. 97(2.1), substitute a penalty that he or she determines just and reasonable in the circumstances, creates a requirement that the employer show cause before dismissal. There can be no justification for this; no reasonable interpretation can lead to that result.  Section 100.1(5) incorporates s. 97(2.1) by reference into the determination of grievances brought by non-unionized employees.  The employees subject to the PSLRA are usually unionized and the terms of their employment are determined by collective agreement; s. 97(2.1) explicitly refers to the collective agreement context.  Section 100.1(5) referentially incorporates s. 97(2.1) mutatis mutandis into the non-collective agreement context so that non-unionized employees who are discharged for cause and without notice have the right to grieve the discharge and have the adjudicator substitute another penalty as seems just and reasonable in the circumstances.  Therefore, the combined effect of s. 97(2.1) and s. 100.1 cannot, on any reasonable interpretation, remove the employer’s right under contract law to discharge an employee with reasonable notice or pay in lieu of notice.

[76]                           The interpretation of the adjudicator was simply unreasonable in the context of the legislative wording and the larger labour context in which it is embedded. It must be set aside. Nevertheless, it must be acknowledged that his interpretation of the PSLRA was ultimately inconsequential to the overall determination of the grievance, since the adjudicator made no finding as to whether the discharge was or was not, in fact, for cause.  The decision on the merits, which resulted in an order that the appellant be reinstated, instead turned on the adjudicator’s decision on a separate issue _ whether the appellant was entitled to and, if so, received procedural fairness with regard to the employer’s decision to terminate his employment.  This issue is discrete and isolated from the statutory interpretation issue, and it raises very different considerations.

IV.  Issue 2: Review of the Adjudicator’s Procedural Fairness Determination

[77]                           Procedural fairness has many faces.  It is at issue where an administrative body may have prescribed rules of procedure that have been breached.  It is also concerned with general principles involving the right to answer and defence where one’s rights are affected.  In this case, the appellant raised in his grievance letter that the reasons for the employer’s dissatisfaction were not specified and that he did not have a reasonable opportunity to respond to the employer’s concerns.  There was, in his view, lack of due process and a breach of procedural fairness.

[78]                           The procedural fairness issue was dealt with only briefly by the Court of Appeal. Robertson J.A. mentioned at the end of his reasons that a duty of fairness did not arise in this case since the appellant had been terminated with notice and had exercised his right to grieve. Before this Court, however, the appellant argued that he was entitled to procedural fairness as a result of this Court’s jurisprudence. Although ultimately we do not agree with the appellant, his contention raises important issues that need to be examined more fully.

A.  Duty of Fairness

[79]                           Procedural fairness is a cornerstone of modern Canadian administrative law. Public decision makers are required to act fairly in coming to decisions that affect the rights, privileges or interests of an individual. Thus stated the principle is easy to grasp. It is not, however, always easy to apply. As has been noted many times, “the concept of procedural fairness is eminently variable and its content is to be decided in the specific context of each case” (Knight, at p. 682; Baker, at para. 21; Moreau‑Bérubé v. New Brunswick (Judicial Council), [2002] 1 S.C.R. 249, 2002 SCC 11, at paras. 74‑75).

[80]                           This case raises the issue of the extent to which a duty of fairness applies to the dismissal of a public employee pursuant to a contract of employment. The grievance adjudicator concluded that the appellant had been denied procedural fairness because he had not been granted a hearing by the employer before being dismissed with four months’ pay in lieu of notice. This conclusion was said to flow from this Court’s decision in Knight, where it was held that the holder of an office “at pleasure” was entitled to be given the  reasons for his or her dismissal and an opportunity to be heard before being dismissed (p. 683).

[81]                           We are of the view that the principles established in Knight relating to the applicability of a duty of fairness in the context of public employment merit reconsideration. While the majority opinion in Knight properly recognized the important place of a general duty of fairness in administrative law, in our opinion, it incorrectly analyzed the effects of a contract of employment on such a duty. The majority in Knight proceeded on the premise that a duty of fairness based on public law applied unless expressly excluded by the employment contract or the statute (p. 681), without consideration of the terms of the contract with regard to fairness issues. It also upheld the distinction between office holders and contractual employees for procedural fairness purposes (pp. 670-76).  In our view, what matters is the nature of the employment relationship between the public employee and the public employer.  Where a public employee is employed under a contract of employment, regardless of his or her status as a public office holder, the applicable law governing his or her dismissal is the law of contract, not general principles arising out of public law.  What Knight truly stands for is the principle that there is always a recourse available where the employee is an office holder and the applicable law leaves him or her without any protection whatsoever when dismissed.

[82]                           This conclusion does not detract from the general duty of fairness owed by administrative decision makers. Rather it acknowledges that in the specific context of dismissal from public employment, disputes should be viewed through the lens of contract law rather than public law.

[83]                           In order to understand why a reconsideration of Knight is warranted, it is necessary to review the development of the duty of fairness in Canadian administrative law. As we shall see, its development in the public employment context was intimately related to the distinction between public office holders and contractual employees, a distinction which, in our view, has become increasingly difficult to maintain both in principle and in practice.

(1)  The Preliminary Issue of Jurisdiction

[84]                           Before dealing with the scope of the duty of fairness in this case, a word should be said about the respondent’s preliminary objection to the jurisdiction of the adjudicator under the PSLRA to consider procedural fairness. The respondent argues that allowing adjudicators to consider procedural fairness risks granting them the inherent powers of a court. We disagree. We can see nothing problematic with a grievance adjudicator considering a public law duty of fairness issue where such a duty exists.  It falls squarely within the adjudicator’s task to resolve a grievance.  However, as will be explained below, the proper approach is to first identify the nature of the employment relationship and the applicable law. Where, as here, the relationship is contractual, a public law duty of fairness is not engaged and therefore should play no role in resolving the grievance.

(2)  The Development of the Duty of Fairness in Canadian Public Law

[85]                           In Canada, the modern concept of procedural fairness in administrative law was inspired by the House of Lords’ landmark decision in Ridge v. Baldwin, [1963] 2 All E.R. 66, a case which involved the summary dismissal of the chief constable of Brighton. The House of Lords declared the chief constable’s dismissal a nullity on the grounds that the administrative body which had dismissed him had failed to provide the reasons for his dismissal or to accord him an opportunity to be heard in violation of the rules of natural justice. Central to the reasoning in the case was Lord Reid’s distinction between (i) master‑servant relationships (i.e. contractual employment), (ii) offices held “at pleasure”, and (iii) offices where there must be cause for dismissal, which included the chief constable’s position.  According to Lord Reid, only the last category of persons was entitled to procedural fairness in relation to their dismissal since both contractual employees and office holders employed “at pleasure” could be dismissed without reason (p. 72).  As the authors Wade and Forsyth note that, after a period of retreat from imposing procedural fairness requirements on administrative decision makers, Ridge v. Baldwin “marked an important change of judicial policy, indicating that natural justice was restored to favour and would be applied on a wide basis” (W. Wade and C. Forsyth, Administrative Law (8th ed. 2000), at p. 438).  

[86]                           The principles established by Ridge v. Baldwin were followed by this Court in Nicholson v. Haldimand‑Norfolk Regional Board of Commissioners of Police, [1979] 1 S.C.R. 311. Nicholson, like its U.K. predecessor, marked the return to a less rigid approach to natural justice in Canada (see Brown and Evans, at pp. 7‑5 to 7‑9). Nicholson concerned the summary dismissal of a probationary police officer by a regional board of police commissioners. Laskin C.J., for the majority, at p. 328, declared the dismissal void on the ground that the officer fell into Lord Reid’s third category and was therefore entitled to the same procedural protections as in Ridge v. Baldwin.

[87]                           Although Ridge v. Baldwin and Nicholson were concerned with procedural fairness in the context of the dismissal of public office holders, the concept of fairness  was quickly extended to other types of administrative decisions (see e.g. Martineau v. Matsqui Institution Disciplinary Board, [1980] 1 S.C.R. 602; Kane v. Board of Governors of the University of British Columbia, [1980] 1 S.C.R. 1105; Attorney General of Canada v. Inuit Tapirisat of Canada, [1980] 2 S.C.R. 735). In Cardinal v. Director of Kent Institution, [1985] 2 S.C.R. 643, Le Dain J. stated that the duty of fairness was a general principle of law applicable to all public authorities:

This Court has affirmed that there is, as a general common law principle, a duty of procedural fairness lying on every public authority making an administrative decision which is not of a legislative nature and which affects the rights, privileges or interests of an individual . . . . [p. 653]

(See also Baker, at para. 20.)

[88]                           In Knight, the Court relied on the statement of Le Dain J. in Cardinal v. Director of Kent Institution that the existence of a general duty to act fairly will depend on “(i) the nature of the decision to be made by the administrative body; (ii) the relationship existing between that body and the individual; and (iii) the effect of that decision on the individual’s rights” (Knight, at p. 669).

[89]                           The dispute in Knight centred on whether a board of education had failed to accord procedural fairness when it dismissed a director of education with three months’ notice pursuant to his contract of employment. The main issue was whether the director’s employment relationship with the school board was one that attracted a public law duty of fairness. L’Heureux‑Dubé J., for the majority, held that it did attract such a duty on the ground that the director’s position had a “strong ‘statutory flavour’” and could thus be qualified as a public office (p. 672). In doing so, she specifically recognized that, contrary to Lord Reid’s holding in Ridge v. Baldwin, holders of an office “at pleasure”, were also entitled to procedural fairness before being dismissed (pp. 673‑74). The fact that the director’s written contract of employment specifically provided that he could be dismissed with three months’ notice was held not to be enough to displace a public law duty to act fairly (p. 681).

[90]                           From these foundational cases, procedural fairness has grown to become a central principle of Canadian administrative law. Its overarching purpose is not difficult to discern: administrative decision makers, in the exercise of public powers, should act fairly in coming to decisions that affect the interests of individuals. In other words, “[t]he observance of fair procedures is central to the notion of the ‘just’ exercise of power” (Brown and Evans, at p. 7‑3). What is less clear, however, is whether this purpose is served by imposing public law procedural fairness requirements on public bodies in the exercise of their contractual rights as employers.

(3)  Procedural Fairness in the Public Employment Context

[91]                           Ridge v. Baldwin and Nicholson established that a public employee’s right to procedural fairness depended on his or her status as an office holder. While Knight extended a duty of fairness to office holders during pleasure, it nevertheless upheld the distinction between office holders and contractual employees as an important criterion in establishing whether a duty of fairness was owed. Courts have continued to rely on this distinction, either extending or denying procedural protections depending on the characterization of the public employee’s legal status as an office holder or contractual employee (see e.g. Reglin v. Creston (Town) (2004), 34 C.C.E.L. (3d) 123, 2004 BCSC 790; Gismondi v. Toronto (City) (2003), 64 O.R. (3d) 688 (C.A.); Seshia v. Health Sciences Centre (2001), 160 Man. R. (2d) 41, 2001 MBCA 151; Rosen v. Saskatoon District Health Board (2001), 202 D.L.R. (4th) 35, 2001 SKCA 83; Hanis v. Teevan (1998), 111 O.A.C. 91; Gerrard v. Sackville (Town) (1992), 124 N.B.R. (2d) 70 (C.A.)).

[92]                           In practice, a clear distinction between office holders and contractual employees has been difficult to maintain:

Although the law makes such a sharp distinction between office and service in theory, in practice it may be difficult to tell which is which. For tax purposes “office” has long been defined as a “subsisting, permanent substantive position which has an existence independent of the person who fills it”, but for the purposes of natural justice the test may not be the same. Nor need an office necessarily be statutory, although nearly all public offices of importance in administrative law are statutory. A statutory public authority may have many employees who are in law merely its servants, and others of higher grades who are office-holders.

(Wade and Forsyth, at pp. 532-33)

[93]                           Lord Wilberforce noted that attempting to separate office holders from contractual employees

involves the risk of a compartmental approach which, although convenient as a solvent, may lead to narrower distinctions than are appropriate to the broader issues of administrative law. A comparative list of situations in which persons have been held entitled or not entitled to a hearing, or to observation of rules of natural justice, according to the master and servant test, looks illogical and even bizarre.

(Malloch v. Aberdeen Corp., [1971] 2 All E.R. 1278 (H.L.), at p. 1294)

[94]                           There is no reason to think that the distinction has been easier to apply in Canada. In Knight, as has been noted, the majority judgment relied on whether the public employee’s position had a “strong ‘statutory flavour’” (p. 672), but as Brown and Evans observe, “there is no simple test for determining whether there is a sufficiently strong ‘statutory flavour’ to a job for it to be classified as an ‘office’” (p. 7-19). This has led to uncertainty as to whether procedural fairness attaches to particular positions.  For instance, there are conflicting decisions on whether the position of a “middle manager” in a municipality is sufficiently important to attract a duty of fairness (compare Gismondi, at para. 53, and Hughes v. Moncton (City) (1990), 111 N.B.R. (2d) 184 (Q.B.), aff’d (1991), 118 N.B.R. (2d) 306 (C.A.)). Similarly, physicians working in the public health system may or may not be entitled to a duty of fairness (compare Seshia and Rosen v. Saskatoon District Health Board, [2000] 4 W.W.R. 606, 2000 SKQB 40).

[95]                           Further complicating the distinction is the fact that public employment is for the most part now viewed as a regular contractual employment relationship. The traditional position at common law was that public servants were literally “servants of the Crown” and could therefore be dismissed at will. However, it is now recognized that most public employees are employed on a contractual basis: Wells v. Newfoundland, [1999] 3 S.C.R. 199.

[96]                           Wells concerned the dismissal without compensation of a public office holder whose position had been abolished by statute. The Court held that, while Wells’ position was created by statute, his employment relationship with the Crown was contractual and therefore he was entitled to be compensated for breach of contract according to ordinary private law principles. Indeed, Wells recognized that most civil servants and public officers are employed under contracts of employment, either as members of unions bound by collective agreements or as non‑unionized employees under individual contracts of employment (paras. 20‑21 and 29‑32). Only certain officers, like ministers of the Crown and “others who fulfill constitutionally defined state roles”, do not have a contractual relationship with the Crown, since the terms of their positions cannot be modified by agreement (Wells, at paras. 29‑32).

[97]                           The effect of Wells, as Professors Hogg and Monahan note, is that

[t]he government’s common law relationship with its employees will now be governed, for the most part, by the general law of contract, in the same way as private employment relationships. This does not mean that governments cannot provide for a right to terminate employment contracts at pleasure. However, if the government wishes to have such a right, it must either contract for it or make provision (expressly or by necessary implication) by way of statute.

(P. W. Hogg and P. J. Monahan, Liability of the Crown (3rd ed. 2000), at p. 240)

The important point for our purposes is that Wells confirmed that most public office holders have a contractual employment relationship. Of course, office holders’ positions will also often be governed by statute and regulations, but the essence of the employment relationship is still contractual. In this context, attempting to make a clear distinction between office holders and contractual employees for the purposes of procedural fairness becomes even more difficult.

[98]                           If the distinction has become difficult to maintain in practice, it is also increasingly hard to justify in principle. There would appear to be three main reasons for distinguishing between office holders and contractual employees and for extending procedural fairness protections only to the former, all of which, in our view, are problematic.

[99]                           First, historically, offices were viewed as a form of property, and thus could be recovered by the office holder who was removed contrary to the principles of natural justice. Employees who were dismissed in breach of their contract, however, could only sue for damages, since specific performance is not generally available for contracts for personal service (Wade and Forsyth, at pp. 531‑32). This conception of public office has long since faded from our law: public offices are no longer treated as a form of private property. 

[100]                       A second and more persuasive reason for the distinction is that dismissal from public office involves the exercise of delegated statutory power and should therefore be subject to public law controls like any other administrative decision (Knight, at p. 675; Malloch, at p. 1293, per Lord Wilberforce). In contrast, the dismissal of a contractual employee only implicates a public authority’s private law rights as an employer.

[101]                       A third reason is that, unlike contractual employees, office holders did not typically benefit from contractual rights protecting them from summary discharge. This was true of the public office holders in Ridge v. Baldwin and Nicholson. Indeed, in both cases the statutory language purported to authorize dismissal without notice. The holders of an office “at pleasure” were in an even more tenuous position since by definition they could be dismissed without notice and without reason (Nicholson, at p. 323; Black’s Law Dictionary (8th ed. 2004), at p. 1192 “pleasure appointment”). Because of this relative insecurity it was seen to be desirable to impose minimal procedural requirements in order to ensure that office holders were not deprived of their positions arbitrarily (Nicholson, at pp. 322-23; Knight, at pp. 674-75; Wade and Forsyth, at pp. 536-37).

[102]                       In our view, the existence of a contract of employment, not the public employee’s status as an office holder, is the crucial consideration. Where a public office holder is employed under a contract of employment the justifications for imposing a public law duty of fairness with respect to his or her dismissal lose much of their force. 

[103]                       Where the employment relationship is contractual, it becomes difficult to see how a public employer is acting any differently in dismissing a public office holder and a contractual employee. In both cases, it would seem that the public employer is merely exercising its private law rights as an employer.  For instance, in Knight, the director’s position was terminated by a resolution passed by the board of education pursuant to statute, but it was done in accordance with the contract of employment, which provided for dismissal on three months’ notice. Similarly, the appellant in this case was dismissed pursuant to s. 20 of the New Brunswick Civil Service Act, but that section provides that the ordinary rules of contract govern dismissal. He could therefore only be dismissed for just cause or on reasonable notice, and any failure to do so would give rise to a right to damages. In seeking to end the employment relationship with four months’ pay in lieu of notice, the respondent was acting no differently than any other employer at common law. In Wells, Major J. noted that public employment had all of the features of a contractual relationship:

A common‑sense view of what it means to work for the government suggests that these relationships have all the hallmarks of contract. There are negotiations leading to agreement and employment.  This gives rise to enforceable obligations on both sides. The Crown is acting much as an ordinary citizen would, engaging in mutually beneficial commercial relations with individual and corporate actors.  Although the Crown may have statutory guidelines, the result is still a contract of employment. [Emphasis added; para. 22.]

If the Crown is acting as any other private actor would in hiring its employees, then it follows that the dismissal of its employees should be viewed in the same way.

[104]                       Furthermore, while public law is rightly concerned with preventing the arbitrary exercise of delegated powers, the good faith exercise of the contractual rights of an employer, such as the right to end the employment relationship on reasonable notice, cannot be qualified as arbitrary. Where the terms of the employment contract were explicitly agreed to, it will be assumed that procedural fairness was dealt with by the parties (see, for example, in the context of collective agreements: School District No. 5 (Southeast Kootenay) and B.C.T.F. (Yellowaga) (Re) (2000), 94 L.A.C. (4th) 56). If, however, the contract of employment is silent, the fundamental terms will be supplied by the common law or the civil law, in which case dismissal may only be for just cause or on reasonable notice.

[105]                       In the context of this appeal, it must be emphasized that dismissal with reasonable notice is not unfair per se. An employer’s right to terminate the employment relationship with due notice is simply the counterpart to the employee’s right to quit with due notice (G. England, Employment Law in Canada (4th ed. (loose‑leaf)), at para. 13.3). It is a well-established principle of the common law that, unless otherwise provided, both parties to an employment contract may end the relationship without alleging cause so long as they provide adequate notice. An employer’s right to terminate on reasonable notice must be exercised within the framework of an employer’s general obligations of good faith and fair dealing: Wallace v. United Grain Growers Ltd., [1997] 3 S.C.R. 701, at para. 95.  But the good faith exercise of a common law contractual right to dismiss with notice does not give rise to concerns about the illegitimate exercise of public power. Moreover, as will be discussed below, where public employers do act in bad faith or engage in unfair dealing, the private law provides a more appropriate form of relief and there is no reason that they should be treated differently than private sector employers who engage in similar conduct.

[106]                       Of course, a public authority must abide by any statutory restrictions on the exercise of its discretion as an employer, regardless of the terms of an employment contract, and failure to do so may give rise to a public law remedy. A public authority cannot contract out of its statutory duties. But where a dismissal decision is properly within the public authority’s powers and is taken pursuant to a contract of employment, there is no compelling public law purpose for imposing a duty of fairness.

[107]                       Nor is the protection of office holders a justification for imposing a duty of fairness when the employee is protected from wrongful dismissal by contract. The appellant’s situation provides a good illustration of why this is so. As an office holder, the appellant was employed “at pleasure”, and could therefore be terminated without notice or reason (Interpretation Act, R.S.N.B. 1973, c. I‑13, s. 20). However, he was also a civil servant and, pursuant to s. 20 of the Civil Service Act, his dismissal was governed by the ordinary rules of contract. If his employer had dismissed him without notice and without cause he would have been entitled to claim damages for breach of contract. Even if he was dismissed with notice, it was open to him to challenge the length of notice or amount of pay in lieu of notice given. On the facts, the respondent gave the appellant four months’ worth of pay in lieu of notice, which he was successful in having increased to eight months before the grievance adjudicator.

[108]                       It is true that the remedy of reinstatement is not available for breach of contract at common law. In this regard, it might be argued that contractual remedies, on their own, offer insufficient protection to office holders (see de Smith, Woolf & Jowell: Judicial Review of Administrative Action (5th ed. 1995), at p. 187). However, it must be kept in mind that breach of a public law duty of fairness also does not lead to full reinstatement. The effect of a breach of procedural fairness is to render the dismissal decision void ab initio (Ridge v. Baldwin, at p. 81). Accordingly, the employment is deemed to have never ceased and the office holder is entitled to unpaid wages and benefits from the date of the dismissal to the date of judgment (see England, at para. 17.224). However, an employer is free to follow the correct procedure and dismiss the office holder again. A breach of the duty of fairness simply requires that the dismissal decision be retaken. It therefore is incorrect to equate it to reinstatement (see Malloch, at p. 1284).

[109]                       In addition, a public law remedy can lead to unfairness. The amount of unpaid wages and benefits an office holder is entitled to will be a function of the length of time the judicial process has taken to wend its way to a final resolution rather than criteria related to the employee’s situation. Furthermore, in principle, there is no duty to mitigate since unpaid wages are not technically damages. As a result, an employee may recoup much more than he or she actually lost (see England, at para. 17.224).

[110]                       In contrast, the private law offers a more principled and fair remedy. The length of notice or amount of pay in lieu of notice an employee is entitled to depends on a number of factors including length of service, age, experience and the availability of alternative employment (see Wallace, at paras. 81 ff.). The notice period may be increased if it is established that the employer acted in bad faith or engaged in unfair dealing when acting to dismiss the employee (Wallace, at para. 95). These considerations aim at ensuring that dismissed employees are afforded some measure of protection while looking for new employment.

[111]                       It is important to note as well that the appellant, as a public employee employed under a contract of employment, also had access to all of the same statutory and common law protections that surround private sector employment. He was protected from dismissal on the basis of a prohibited ground of discrimination under the Human Rights Act, R.S.N.B. 1973, c. H‑11.  His employer was bound to respect the norms laid down by the Employment Standards Act, S.N.B. 1982, c. E‑7.2. As has already been mentioned, if his dismissal had been in bad faith or he had been subject to unfair dealing, it would have been open to him to argue for an extension of the notice period pursuant to the principles laid down in Wallace. In short, the appellant was not without legal protections or remedies in the face of his dismissal.

(4)  The Proper Approach to the Dismissal of Public Employees

[112]                       In our view, the distinction between office holder and contractual employee for the purposes of a public law duty of fairness is problematic and should be done away with. The distinction is difficult to apply in practice and does not correspond with the justifications for imposing public law procedural fairness requirements. What is important in assessing the actions of a public employer in relation to its employees is the nature of the employment relationship. Where the relationship is contractual, it should be viewed as any other private law employment relationship  regardless of an employee’s status as an office holder.

[113]                       The starting point, therefore, in any analysis, should be to determine the nature of the employment relationship with the public authority. Following Wells, it is assumed that most public employment relationships are contractual. Where this is the case, disputes relating to dismissal should be resolved according to the express or implied terms of the contract of employment and any applicable statutes and regulations, without regard for whether the employee is an office holder. A public authority which dismisses an employee pursuant to a contract of employment should not be subject to any additional public law duty of fairness. Where the dismissal results in a breach of contract, the public employee will have access to ordinary contractual remedies.

[114]                       The principles expressed in Knight in relation to the general duty of fairness owed by public authorities when making decisions that affect the rights, privileges or interests of individuals are valid and important. However, to the extent that the majority decision in Knight ignored the important effect of a contract of employment, it should not be followed. Where a public employee is protected from wrongful dismissal by contract, his or her remedy should be in private law, not in public law.

[115]                       The dismissal of a public employee should therefore generally be viewed as a typical employment law dispute. However, there may be occasions where a public law duty of fairness will still apply. We can envision two such situations at present. The first occurs where a public employee is not, in fact, protected by a contract of employment. This will be the case with judges, ministers of the Crown and others who “fulfill constitutionally defined state roles” (Wells, at para. 31). It may also be that the terms of appointment of some public office holders expressly provide for summary dismissal or, at the very least, are silent on the matter, in which case the office holders may be deemed to hold office “at pleasure” (see e.g. New Brunswick  Interpretation Act , s. 20 ; Interpretation Act , R.S.C. 1985, c. I‑21, s. 23(1) ). Because an employee in this situation is truly subject to the will of the Crown, procedural fairness is required to ensure that public power is not exercised capriciously.

[116]                       A second situation occurs when a duty of fairness flows by necessary implication from a statutory power governing the employment relationship. In Malloch, the applicable statute provided that dismissal of a teacher could only take place if the teacher was given three weeks’ notice of the motion to dismiss. The House of Lords found that this necessarily implied a right for the teacher to make representations at the meeting where the dismissal motion was being considered. Otherwise, there would have been little reason for Parliament to have provided for the notice procedure in the first place (p. 1282). Whether and what type of procedural requirements result from a particular statutory power will of course depend on the specific wording at issue and will vary with the context (Knight, at p. 682).

B.  Conclusion

[117]                       In this case, the appellant was a contractual employee of the respondent in addition to being a public office holder. Section 20 of the Civil Service Act provided that, as a civil servant, he could only be dismissed in accordance with the ordinary rules of contract. In these circumstances it was unnecessary to consider any public law duty of procedural fairness. The respondent was fully within its rights to dismiss the appellant with pay in lieu of notice without affording him a hearing. The respondent dismissed the appellant with four months’ pay in lieu of notice. The appellant was successful in increasing this amount to eight months. The appellant was protected by contract and was able to obtain contractual remedies in relation to his dismissal. By imposing procedural fairness requirements on the respondent over and above its contractual obligations and ordering the full “reinstatement” of the appellant, the adjudicator erred in his application of the duty of fairness and his decision was therefore correctly struck down by the Court of Queen’s Bench.

V.  Disposition

[118]                       We would dismiss the appeal.  There will be no order for costs in this Court as the respondent is not requesting them.

The following are the reasons delivered by

[119]                       Binnie J. — I agree with my colleagues that the appellant’s former employment relationship with the respondent is governed by contract.  The respondent chose to exercise its right to terminate the employment without alleging cause.  The adjudicator  adopted an unreasonable interpretation of s. 20 of the Civil Service Act, S.N.B. 1984, c. C-5.1, and of ss. 97(2.1) and 100.1 of the Public Service Labour Relations Act, R.S.N.B. 1973, c. P-25.  The appellant was a non-unionized employee whose job was terminated in accordance with contract law.  Public law principles of procedural fairness were not applicable in the circumstances.  These conclusions are enough to dispose of the appeal.

[120]                       However, my colleagues Bastarache and LeBel JJ. are embarked on a more ambitious mission, stating that:

Although the instant appeal deals with the particular problem of judicial review of the decisions of an adjudicative tribunal, these reasons will address first and foremost the structure and characteristics of the system of judicial review as a whole.

                                                                      . . .

. . . The time has arrived to re-examine the Canadian approach to judicial review of administrative decisions and develop a principled framework that is more coherent and workable. [Emphasis added; paras. 33 and 32.]

[121]                       The need for such a re-examination is widely recognized, but in the end my colleagues’ reasons for judgment do not deal with the “system as a whole”.  They focus on administrative tribunals.  In that context, they reduce the applicable standards of review from three to two (“correctness” and “reasonableness”), but retain the pragmatic and functional analysis, although now it is to be called the “standard of review analysis”  (para. 63).  A broader reappraisal is called for.  Changing the name of the old pragmatic and functional test represents a limited advance, but as the poet says:

What’s in a name?  that which we call a rose

By any other name would smell as sweet;

(Romeo and Juliet, Act II, Scene ii)

[122]                       I am emboldened by my colleagues’ insistence that “a holistic approach is needed when considering fundamental principles” (para. 26) to express the following views.  Judicial review is an idea that has lately become unduly burdened with law office metaphysics.  We are concerned with substance not nomenclature.  The words themselves are unobjectionable.  The dreaded reference to “functional” can simply be taken to mean that generally speaking courts have the last word on what they consider the correct decision on legal matters (because deciding legal issues is their “function”), while administrators should generally have the last word within their function, which is to decide administrative matters.  The word “pragmatic” not only signals a distaste for formalism but recognizes that a conceptually tidy division of functions has to be tempered by practical considerations: for example, a labour board is better placed than the courts to interpret the intricacies of provisions in a labour statute governing replacement of union workers; see e.g. Canadian Union of Public Employees, Local 963 v. New Brunswick Liquor Corp., [1979] 2 S.C.R. 227.

[123]                       Parliament or a provincial legislature is often well advised to allocate an administrative decision to someone other than a judge.  The judge is on the outside of the administration looking in. The legislators are entitled to put their trust in the viewpoint of the designated decision maker (particularly as to what constitutes a reasonable outcome), not only in the case of the administrative tribunals of principal concern to my colleagues but (taking a “holistic approach”) also in the case of a minister, a board, a public servant, a commission, an elected council or other administrative bodies and statutory decision makers.  In the absence of a full statutory right of appeal, the court ought generally to respect the exercise of the administrative discretion, particularly in the face of a privative clause.

[124]                       On the other hand, a court is right to insist that its view of the correct opinion (i.e. the “correctness” standard of review) is accepted on questions concerning the Constitution, the common law, and the interpretation of a statute other than the administrator’s enabling statute (the “home statute”) or a rule or statute closely connected with it; see generally D. J. M. Brown and J. M. Evans, Judicial Review of Administrative Action in Canada (loose-leaf), at para. 14:2210.

[125]                       Thus the law (or, more grandly, the “rule of law”) sets the boundaries of potential administrative action.  It is sometimes said by judges that an administrator acting within his or her discretion “has the right to be wrong”.  This reflects an unduly court-centred view of the universe.  A disagreement between the court and an administrator does not necessarily mean that the administrator is wrong.

A.  Limits on the Allocation of Decision Making

[126]                       It should not be difficult in the course of judicial review to identify legal questions requiring disposition by a judge.  There are three basic legal limits on the allocation of administrative discretion.

[127]                       Firstly, the Constitution restricts the legislator’s ability to allocate issues to administrative bodies which s. 96  of the Constitution Act, 1867  has allocated to the courts.  The logic of the constitutional limitation is obvious.  If the limitation did not exist, the government could transfer the work of the courts to administrative bodies that are not independent of the executive and by statute immunize the decisions of these bodies from effective judicial review.  The country would still possess an independent judiciary, but the courts would not be available to citizens whose rights or interests are trapped in the administration.

[128]                       Secondly, administrative action must be founded on statutory or prerogative (i.e. common law) powers .  This too is a simple idea.  No one can exercise a power they do not possess.  Whether or not the power (or jurisdiction) exists is a question of law for the courts to determine, just as it is for the courts (not the administrators) to have the final word on questions of general law that may be relevant to the resolution of an administrative issue.  The instances where this Court has deferred to an administrator’s conclusion of law outside his or her home statute, or a statute “intimately” connected thereto, are exceptional.  We should say so.  Instead, my colleagues say the court’s view of the law will prevail

where the question at issue is one of general law “that is both of central importance to the legal system as a whole and outside the adjudicator’s specialized area of expertise”. [para. 60]

It is, with respect, a distraction to unleash a debate in the reviewing judge’s courtroom about whether or not a particular question of law is “of central importance to the legal system as a whole”.  It should be sufficient to frame a rule exempting from the correctness standard the provisions of the home statute and closely related statutes which require the expertise of the administrative decision maker (as in the labour board example).  Apart from that exception, we should prefer clarity to needless complexity  and hold that the last word on questions of general law should be left to judges.

[129]                       Thirdly, a fair procedure is said to be the handmaiden of justice.  Accordingly, procedural limits are placed on administrative bodies by statute and the common law.  These include the requirements of “procedural fairness”, which will vary with the type of decision maker and the type of decision under review.  On such matters, as well, the courts have the final say.  The need for such procedural safeguards is obvious.  Nobody should have his or her rights, interests or privileges adversely dealt with by an unjust process.  Nor is such an unjust intent to be attributed easily to legislators.  Hansard is full of expressions of concern by Ministers and Members of Parliament regarding the fairness of proposed legislative provisions.  There is a dated hauteur about judicial pronouncements such as that the “justice of the common law will supply the omission of the legislature” (Cooper v. Wandsworth Board of Works (1863), 14 C.B. (N.S.) 180, 143 E.R. 414 (C.P.), at p. 420).  Generally speaking, legislators and judges in this country are working with a common set of basic legal and constitutional values.  They share a belief in the rule of law.  Constitutional considerations aside, however, statutory protections can nevertheless be repealed and common law protections can be modified by statute, as was demonstrated in Ocean Port Hotel Ltd. v. British Columbia (General Manager, Liquor Control and Licensing Branch), [2001] 2 S.C.R. 781, 2001 SCC 52.

B.  Reasonableness of Outcome

[130]                       At this point, judicial review shifts gears.  When the applicant for judicial review challenges the substantive outcome of an administrative action, the judge is invited to cross the line into second-guessing matters that lie within the function of the administrator.  This is controversial because it is not immediately obvious why a judge’s view of the reasonableness of an administrative policy or the exercise of an administrative discretion should be preferred to that of the administrator to whom Parliament or a  legislature has allocated the decision, unless there is a full statutory right of appeal to the courts, or it is otherwise indicated in the conferring legislation that a “correctness” standard is intended.

[131]                       In U.E.S., Local 298 v. Bibeault, [1988] 2 S.C.R. 1048, Beetz J. adopted the view that “[t]o a large extent judicial review of administrative action is a specialized branch of statutory interpretation” (p. 1087 (emphasis deleted)).  Judicial intervention in administrative decisions on grounds of substance (in the absence of a constitutional challenge) has been based on presumed legislative intent in a line of cases from Associated Provincial Picture Houses Ltd. v. Wednesbury Corp., [1947] 2 All E.R. 680 (C.A.) (“you may have something so absurd that no sensible person could ever dream that it lay within the powers of the authority” (p. 683)) to Canadian Union of Public Employees, Local 963 v. New Brunswick Liquor Corp. (“was the Board’s interpretation so patently unreasonable that its construction cannot be rationally supported by the relevant legislation . . . ?” (p. 237)).  More recent examples are Baker v. Canada (Minister of Citizenship and Immigration), [1999] 2 S.C.R. 817 (para. 53), and Mount Sinai Hospital Center v. Quebec (Minister of Health and Social Services), [2001] 2 S.C.R. 281, 2001 SCC 41 (paras. 60-61).  Judicial review proceeds on the justified presumption that legislators do not intend results that depart from reasonable standards.

C.  The Need to Reappraise the Approach to Judicial Review

[132]                       The present difficulty, it seems, does not lie in the component parts of judicial review, most of which are well entrenched in decades of case law, but in the current methodology for putting those component parts into action.  There is afoot in the legal profession a desire for clearer guidance than is provided by lists of principles, factors and spectrums.  It must be recognized, of course, that complexity is inherent in all legal principles that must address the vast range of administrative decision making. The objection is that our present “pragmatic and functional” approach is more complicated than is required by the subject matter.

[133]                       People who feel victimized or unjustly dealt with by the apparatus of government, and who have no recourse to an administrative appeal, should have access to an independent judge through a procedure that is quick and relatively inexpensive.  Like much litigation these days, however, judicial review is burdened with undue cost and delay.  Litigants understandably hesitate to go to court to seek redress for a perceived administrative injustice if their lawyers cannot predict with confidence even what standard of review will be applied.  The disposition of the case may well turn on the choice of standard of review.  If litigants do take the plunge, they may find the court’s attention focussed not on their complaints, or the government’s response, but on lengthy and arcane discussions of something they are told is the pragmatic and functional test.  Every hour of a lawyer’s preparation and court time devoted to unproductive “lawyer’s talk” poses a significant cost to the applicant.  If the challenge is unsuccessful, the unhappy applicant may also face a substantial bill of costs from the successful government agency.  A victory before the reviewing court may be overturned on appeal because the wrong “standard of review” was selected.  A small business denied a licence or a professional person who wants to challenge disciplinary action should be able to seek judicial review without betting the store or the house on the outcome.  Thus, in my view, the law of judicial review should be pruned of some of its unduly subtle, unproductive, or esoteric features.

D.  Standards of Review

[134]                       My colleagues conclude that three standards of review should be reduced to two standards of review.  I agree that this simplification will avoid some of the arcane debates about the point at which “unreasonableness” becomes “patent unreasonableness”.  However, in my view the repercussions of their position go well beyond administrative tribunals.  My colleagues conclude, and I agree:

Looking to either the magnitude or the immediacy of the defect in the tribunal’s decision provides no meaningful way in practice of distinguishing between a patently unreasonable and an unreasonable decision. [para. 41]

More broadly, they declare that “the analytical problems that arise in trying to apply the different standards undercut any conceptual usefulness created by the inherently greater flexibility of having multiple standards of review” (para. 44), and “any actual difference between them in terms of their operation appears to be illusory” (para. 41).  A test which is incoherent when applied to administrative tribunals does not gain in coherence or logic when applied to other administrative decision makers such as mid-level bureaucrats or, for that matter, Ministers.  If logic and language cannot capture the distinction in one context, it must equally be deficient elsewhere in the field of judicial review.  I therefore proceed on the basis that the distinction between “patent unreasonableness” and “reasonableness simpliciter” has been declared by the Court to be abandoned.  I propose at this point to examine what I see as some of the implications of this abandonment.

E.  Degrees of Deference

[135]                       The distinction between reasonableness simpliciter and patent unreasonableness was not directed merely to “the magnitude or the immediacy of the defect” in the administrative decision (para. 41).  The distinction also recognized that different administrative decisions command different degrees of deference, depending on who is deciding what.

[136]                       A minister making decisions under the Extradition Act, R.S.C. 1985, c. E-23, to surrender a fugitive, for example, is said to be “at the extreme legislative end of the continuum of administrative decision-making” (Idziak v. Canada (Minister of Justice), [1992] 3 S.C.R. 631, at p. 659).  On the other hand, a ministerial delegate making a deportation decision according to ministerial guidelines was accorded considerably less deference in Baker (where the “reasonableness simpliciter” standard was applied).  The difference does not lie only in the judge’s view of the perceived immediacy of the defect in the administrative decision.  In Suresh v. Canada (Minister of Citizenship and Immigration), [2002] 1 S.C.R. 3, 2002 SCC 1, a unanimous Court adopted the caution in the context of counter-terrorism measures that “[i]f the people are to accept the consequences of such decisions, they must be made by persons whom the people have elected and whom they can remove” (para. 33).  Administrative decision makers  generally command respect more for their expertise than for their prominence in the administrative food chain.  Far more numerous are the lesser officials who reside in the bowels and recesses of government departments adjudicating pension benefits or the granting or withholding of licences, or municipal boards poring over budgets or allocating costs of local improvements.  Then there are the Cabinet and Ministers of the Crown who make broad decisions of public policy such as testing cruise missiles, Operation Dismantle Inc. v. The Queen, [1985] 1 S.C.R. 441, or policy decisions arising out of decisions of major administrative tribunals, as in Attorney General of Canada v. Inuit Tapirisat of Canada, [1980] 2 S.C.R. 735, at p. 753, where the Court said:  “The very nature of the body must be taken into account in assessing the technique of review  which has been adopted by the Governor in Council.”

[137]                       Of course, the degree of deference also depends on the nature and content of the question.  An adjudicative tribunal called on to approve pipelines based on “public convenience and necessity” (Westcoast Energy Inc. v. Canada (National Energy Board), [1998] 1 S.C.R. 322) or simply to take a decision in the “public interest” is necessarily accorded more room to manoeuvre than is a professional body, given the task of determining an appropriate sanction for a member’s misconduct (Law Society of New Brunswick v. Ryan, [2003] 1 S.C.R. 247, 2003 SCC 20). 

[138]                       In our recent jurisprudence, the “nature of the question” before the decision maker has been considered as one of a number of elements to be considered in choosing amongst the various standards of review.  At this point, however, I believe it plays a more important role in terms of substantive review.  It helps to define the range of reasonable outcomes within which the administrator is authorized to choose.

[139]                       The judicial sensitivity to different levels of respect (or deference) required in different situations is quite legitimate.  “Contextualizing” a single standard of review will shift the debate (slightly) from choosing between two standards of reasonableness that each represent a different level of deference to a debate within a single standard of reasonableness to determine the appropriate level of deference.  In practice, the result of today’s decision may be like the bold innovations of a traffic engineer that in the end do no more than shift rush hour congestion from one road intersection to another without any overall saving to motorists in time or expense.

[140]                       That said, I agree that the repeated attempts to define and explain the difference between reasonableness simpliciter and “patent” unreasonableness can be seen with the benefit of hindsight to be unproductive and distracting.  Nevertheless, the underlying issue of degrees of deference (which the two standards were designed to address) remains.

[141]                       Historically, our law recognized “patent” unreasonableness before it recognized what became known as reasonableness simpliciter.  The adjective “patent” initially underscored the level of respect that was due to the designated decision maker, and signalled the narrow authority of the courts to interfere with a particular administrative outcome on substantive grounds.  The reasonableness simpliciter standard was added at a later date to recognize a reduced level of deference.  Reducing three standards of review to two standards of review does not alter the reality that at the high end “patent” unreasonableness (in the sense of manifestly indefensible) was not a bad description of the hurdle an applicant had to get over to have an administrative decision quashed on a ground of substance.  The danger of labelling the most “deferential” standard as “reasonableness” is that it may be taken (wrongly) as an invitation to reviewing judges not simply to identify the usual issues, such as whether irrelevant matters were taken into consideration, or relevant matters were not taken into consideration, but to reweigh the input that resulted in the administrator’s decision as if it were the judge’s view of “reasonableness” that counts.  At this point, the judge’s role is to identify the outer boundaries of reasonable outcomes within which the administrative decision maker is free to choose.

F.  Multiple Aspects of Administrative Decisions

[142]                       Mention should be made of a further feature that also reflects the complexity  of the subject matter of judicial review.  An applicant may advance several grounds for quashing an administrative decision.  He or she may contend that the decision maker has misinterpreted the general law.  He or she may argue, in the alternative, that even if the decision maker got the general law straight (an issue on which the court’s view of what is correct will prevail), the decision maker did not properly apply it to the facts (an issue on which the decision maker is entitled to deference).  In a challenge under the Canadian Charter of Rights and Freedoms  to a surrender for extradition, for example, the minister will have to comply with the Court’s view of Charter  principles (the “correctness” standard), but if he or she correctly appreciates the applicable law, the court will properly recognize a wide discretion in the application of those principles to the particular facts.  The same approach is taken to less exalted decision makers (Moreau-Bérubé v. New Brunswick (Judicial Council), [2002] 1 S.C.R. 249, 2002 SCC 11).  In the jargon of the judicial review bar, this is known as “segmentation”.

G.  The Existence of a Privative Clause

[143]                       The existence of a privative clause is currently subsumed within the “pragmatic and functional” test as one factor amongst others to be considered in determining the appropriate standard of review, where it supports the choice of the patent unreasonableness standard.  A single standard of “reasonableness” cannot mean that the degree of deference is unaffected by the existence of a suitably worded privative clause.  It is certainly a relevant contextual circumstance that helps to calibrate the intrusiveness of a court’s review.  It signals the level of respect that must be shown.  Chief Justice Laskin during argument once memorably condemned the quashing of a labour board decision protected by a strong privative clause, by saying “what’s wrong with these people [the judges], can’t they read?”  A system of judicial review based on the rule of law ought not to treat a privative clause as conclusive, but it is more than just another “factor” in the hopper of pragmatism and functionality.  Its existence should presumptively foreclose judicial review on the basis of outcome on substantive grounds unless the applicant can show that the clause, properly interpreted, permits it or there is some legal reason why it cannot be given effect.

H.  A Broader Reappraisal

[144]                       “Reasonableness” is a big tent that will have to accommodate a lot of variables that inform and limit a court’s review of the outcome of administrative decision making.

[145]                       The theory of our recent case law has been that once the appropriate standard of review is selected, it is a fairly straightforward matter to apply it.  In practice, the criteria for selection among “reasonableness” standards of review proved to be undefinable and their application unpredictable.  The present incarnation of the “standard of review” analysis requires a threshold debate about the four factors (non-exhaustive) which critics say too often leads to unnecessary delay, uncertainty and costs as arguments rage before the court about balancing expertise against the “real” nature of the question before the administrator, or whether the existence of a privative clause trumps the larger statutory purpose, and so on.  And this is all mere preparation for the argument about the actual substance of the case.  While a measure of uncertainty is inherent in the subject matter and unavoidable in litigation (otherwise there wouldn’t be any), we should at least (i) establish some presumptive rules and (ii) get the parties away from arguing about the tests and back to arguing about the substantive merits of their case.

[146]                       The going-in presumption should be that the standard of review of any administrative outcome on grounds of substance is not correctness but reasonableness (“contextually” applied).  The fact that the legislature designated someone other than the court as the decision maker calls for deference to (or judicial respect for) the outcome, absent a broad statutory right of appeal.  Administrative decisions generally call for the exercise of discretion.  Everybody recognizes in such cases that there is no single “correct” outcome.  It should also be presumed, in accordance with the ordinary rules of litigation, that the decision under review is reasonable until the applicant shows otherwise.

[147]                       An applicant urging the non-deferential “correctness” standard should be required to demonstrate that the decision under review rests on an error in the determination of a legal issue not confided (or which constitutionally could not be confided) to the administrative decision maker to decide, whether in relation to jurisdiction or the general law.  Labour arbitrators, as in this case, command deference on legal matters within their enabling statute or on legal matters intimately connected thereto.

[148]                       When, then, should a decision be deemed “unreasonable”?  My colleagues suggest a test of irrationality (para. 46), but the editors of de Smith point out that “many decisions which fall foul of [unreasonableness] have been coldly rational” (de Smith, Woolf & Jowell: Judicial Review of Administrative Action (5th ed. 1995), at para. 13-003).  A decision meeting this description by this Court is C.U.P.E. v. Ontario (Minister of Labour), [2003] 1 S.C.R. 539, 2003 SCC 29, where the Minister’s appointment of retired judges with little experience in labour matters to chair “interest” arbitrations (as opposed to “grievance” arbitrations) between hospitals and hospital workers was “coldly rational” in terms of the Minister’s own agenda, but was held by a majority of this Court to be patently unreasonable in terms of the history, object and purpose of the authorizing legislation.  He had not used the appointment power for the purposes for which the legislature had conferred it.

[149]                       Reasonableness rather than rationality has been the traditional standard and, properly interpreted, it works.  That said, a single “reasonableness” standard will now necessarily incorporate both the degree of deference formerly reflected in the distinction between patent unreasonableness and reasonableness simpliciter, and an assessment of the range of options reasonably open to the decision maker in the circumstances, in light of the reasons given for the decision.  Any reappraisal of our approach to judicial review should, I think, explicitly recognize these different dimensions to the “reasonableness” standard.

I.  Judging “Reasonableness”

[150]                       I agree with my colleagues that “reasonableness” depends on the context.  It must be calibrated to fit the circumstances.  A driving speed that is “reasonable” when motoring along a four-lane interprovincial highway is not “reasonable” when driving along an inner city street.  The standard (“reasonableness”) stays the same, but the reasonableness assessment will vary with the relevant circumstances. 

[151]                       This, of course, is the nub of the difficulty.  My colleagues write:

In judicial review, reasonableness is concerned mostly with the existence of justification, transparency and intelligibility within the decision-making process.  But it is also concerned with whether the decision falls within a range of possible, acceptable outcomes which are defensible in respect of the facts and law. [para. 47]

I agree with this summary but what is required, with respect, is a more easily applied framework into which the judicial review court and litigants can plug in the relevant context.  No one doubts that in order to overturn an administrative outcome on grounds of substance (i.e. leaving aside errors of fairness or law which lie within the supervising “function” of the courts), the reviewing court must be satisfied that the outcome was outside the scope of reasonable responses open to the decision maker under its grant of authority, usually a statute.  “[T]here is always a perspective”, observed Rand J., “within which a statute is intended [by the legislature] to operate”: Roncarelli v. Duplessis, [1959] S.C.R. 121, at p. 140.   How is that “perspective” to be ascertained?  The reviewing judge will obviously want to consider the precise nature and function of the decision maker including its expertise, the terms and objectives of the governing statute (or common law) conferring the power of decision, including the existence of a privative clause and the nature of the issue being decided.  Careful consideration of these matters will reveal the extent of the discretion conferred, for example, the extent to which the decision formulates or implements broad public policy.  In such cases, the range of permissible considerations will obviously be much broader than where the decision to be made is more narrowly circumscribed, e.g., whether a particular claimant is entitled to a disability benefit under governmental social programs.  In some cases, the court will have to recognize that the decision maker was required to strike a proper balance (or achieve proportionality) between the adverse impact of a decision on the rights and interests of the applicant or others directly affected weighed against the public purpose which is sought to be advanced.  In each case, careful consideration will have to be given to the reasons given for the decision.  To this list, of course, may be added as many “contextual” considerations as the court considers relevant and material.

[152]                       Some of these indicia were included from the outset in the pragmatic and functional test itself (see Bibeault, at p. 1088).  The problem, however, is that under Bibeault, and the cases that followed it, these indicia were used to choose among the different standards of review, which were themselves considered more or less fixed.  In Law Society of New Brunswick v. Ryan, for example, the Court rejected the argument that “it is sometimes appropriate to apply the reasonableness standard more deferentially and sometimes less deferentially depending on the circumstances” (para. 43).  It seems to me that collapsing everything beyond “correctness” into a single “reasonableness” standard will require a reviewing court to do exactly that.

[153]                       The Court’s adoption in this case of a single “reasonableness” standard that covers both the degree of deference assessment and the reviewing court’s evaluation, in light of the appropriate degree of deference, of whether the decision falls within a range of reasonable administrative choices will require a reviewing court to juggle a number of variables that are necessarily to be considered together.  Asking courts to have regard to more than one variable is not asking too much, in my opinion.  In other disciplines, data are routinely plotted simultaneously along both an X axis and a Y axis, without traumatizing the participants.

[154]                       It is not as though we lack guidance in the decided cases.  Much has been written by various courts about deference and reasonableness in the particular contexts of different administrative situations.  Leaving aside the “pragmatic and functional” test, we have ample precedents to show when it is (or is not) appropriate for a court to intervene in the outcome of an administrative decision.  The problem is that courts have lately felt obliged to devote too much time to multi-part threshold tests instead of focussing on the who, what, why and wherefor of the litigant’s complaint on its merits.

[155]                       That having been said, a reviewing court ought to recognize throughout the exercise that fundamentally the “reasonableness” of the outcome is an issue given to others to decide.  The exercise of discretion is an important part of administrative decision making.  Adoption of a single “reasonableness” standard should not be seen by potential litigants as a lowering of the bar to judicial intervention.

J.  Application to This Case

[156]                       Labour arbitrators often have to juggle different statutory provisions in disposing of a grievance.  The courts have generally attached great importance to their expertise in keeping labour peace.  In this case, the adjudicator was dealing with his “home statute” plus other statutes intimately linked to public sector relations in New Brunswick.  He was working on his “home turf”, and the legislature has made clear in the privative clause that it intended the adjudicator to determine the outcome of the appellant’s grievance.  In this field, quick and cheap justice (capped by finality) advances the achievement of the legislative scheme.  Recourse to judicial review is discouraged.  I would therefore apply a reasonableness standard to the adjudicator’s interpretation of his “home turf” statutory framework.

[157]                       Once under the flag of reasonableness, however, the salient question before the adjudicator in this case was essentially legal in nature, as reflected in the reasons he gave for his decision.  He was not called on to implement public policy; nor was there a lot of discretion in dealing with a non-unionized employee.  The basic facts were not in dispute.  He was disposing of a lis which he believed to be governed by the legislation.  He was right to be conscious of the impact of his decision on the appellant, but he stretched the law too far in coming to his rescue.  I therefore join with my colleagues in dismissing the appeal.

The reasons of Deschamps, Charron and Rothstein JJ. were delivered by

[158]                       Deschamps J. — The law of judicial review of administrative action not only requires repairs, it needs to be cleared of superfluous discussions and processes. This area of the law can be simplified by examining the substance of the work courts are called upon to do when reviewing any case, whether it be in the context of administrative or of appellate review. Any review starts with the identification of the questions at issue as questions of law, questions of fact or questions of mixed fact and law.  Very little else needs to be done in order to determine whether deference needs to be shown to an administrative body.

[159]                       By virtue of the Constitution, superior courts are the only courts that possess inherent jurisdiction. They are responsible both for applying the laws enacted by Parliament and the legislatures and for insuring that statutory bodies respect their legal boundaries. Parliament and the legislatures cannot totally exclude judicial oversight without overstepping the division between legislative or executive powers and judicial powers. Superior courts are, in the end, the protectors of the integrity of the rule of law and the justice system. Judicial review of administrative action is rooted in these fundamental principles and its boundaries are largely informed by the roles of the respective branches of government.

[160]                       The judicial review of administrative action has, over the past 20 years, been viewed as involving a preliminary analysis of whether deference is owed to an administrative body based on four factors: (1) the nature of the question, (2) the presence or absence of a privative clause, (3) the expertise of the administrative decision maker and (4) the object of the statute. The process of answering this preliminary question has become more complex than the determination of the substantive questions the court is called upon to resolve.  In my view, the analysis can be made plainer if the focus is placed on the issues the parties need to have adjudicated rather than on the nature of the judicial review process itself. By focusing first on “the nature of the question”, to use what has become familiar parlance, it will become apparent that all four factors need not be considered in every case and that the judicial review of administrative action is often not distinguishable from the appellate review of court decisions.

[161]                       Questions before the courts have consistently been identified as either questions of fact, questions of law or questions of mixed fact and law. Whether undergoing appellate review or administrative law review, decisions on questions of fact always attract deference. The use of different terminology — “palpable and overriding error” versus “unreasonable decision” — does not change the substance of the review. Indeed, in the context of appellate review of court decisions, this Court has recognized that these expressions as well as others all encapsulate the same principle of deference with respect to a trial judge’s findings of fact: H.L. v. Canada (Attorney General), [2005] 1 S.C.R. 401, 2005 SCC 25, at paras. 55-56.  Therefore, when the issue is limited to questions of fact, there is no need to enquire into any other factor in order to determine that deference is owed to an administrative decision maker.

[162]                       Questions of law, by contrast, require more thorough scrutiny when deference is evaluated, and the particular context of administrative decision making can make judicial review different than appellate review. Although superior courts have a core expertise to interpret questions of law, Parliament or a legislature may have provided that the decision of an administrative body is protected from judicial review by a privative clause. When an administrative body is created to interpret and apply certain legal rules, it develops specific expertise in exercising its jurisdiction and has a more comprehensive view of those rules. Where there is a privative clause, Parliament or a legislature’s intent to leave the final decision to that body cannot be doubted and deference is usually owed to the body.

[163]                       However, privative clauses cannot totally shield an administrative body from review. Parliament, or a legislature, cannot have intended that the body would be protected were it to overstep its delegated powers. Moreover, if such a body is asked to interpret laws in respect of which it does not have expertise, the constitutional responsibility of the superior courts as guardians of the rule of law compels them to insure that laws falling outside an administrative body’s core expertise are interpreted correctly. This reduced deference insures that laws of general application, such as the Constitution, the common law and the Civil Code, are interpreted correctly and consistently. Consistency of the law is of prime societal importance.  Finally, deference is not owed on questions of law where Parliament or a legislature has provided for a statutory right of review on such questions.

[164]                       The category of questions of mixed fact and law should be limited to cases in which the determination of a legal issue is inextricably intertwined with the determination of facts. Often, an administrative body will first identify the rule and then apply it. Identifying the contours and the content of a legal rule are questions of law.  Applying the rule, however, is a question of mixed fact and law. When considering a question of mixed fact and law, a reviewing court should show an adjudicator the same deference as an appeal court would show a lower court.

[165]                       In addition, Parliament or a legislature may confer a discretionary power on an administrative body.  Since the case at bar does not concern a discretionary power, it will suffice for the purposes of these reasons to note that, in any analysis, deference is owed to an exercise of discretion unless the body has exceeded its mandate.

[166]                       In summary, in the adjudicative context, the same deference is owed in respect of questions of fact and questions of mixed fact and law on administrative review as on an appeal from a court decision.  A decision on a question of law will also attract deference, provided it concerns the interpretation of the enabling statute and provided there is no right of review.

[167]                       I would be remiss were I to disregard the difficulty inherent in any exercise of deference. In Toronto (City) v. C.U.P.E., Local 79, [2003] 3 S.C.R. 77, 2003 SCC 63, LeBel J. explained why a distinction between the standards of patent unreasonableness and unreasonableness simpliciter is untenable.  I agree. The problem with the definitions resides in attempts by the courts to enclose the concept of reasonableness in a formula fitting all cases. No matter how this Court defines this concept, any context considered by a reviewing court will, more often than not, look more like a rainbow than a black and white situation. One cannot change this reality. I use the word “deference” to define the contours of reasonableness because it describes the attitude adopted towards the decision maker. The word “reasonableness” concerns the decision. However, neither the concept of reasonableness nor that of deference is particular to the field of administrative law. These concepts are also found in the context of criminal and civil appellate review of court decisions. Yet, the exercise of the judicial supervisory role in those fields has not given rise to the complexities encountered in administrative law.  The process of stepping back and taking an ex post facto look at the decision to determine whether there is an error justifying intervention should not be more complex in the administrative law context than in the criminal and civil law contexts.

[168]                       In the case at bar, the adjudicator was asked to adjudicate the grievance of a non-unionized employee. This meant that he had to identify the rules governing the contract. Identifying those rules is a question of law. Section 20 of the Civil Service Act, S.N.B. 1984, c. C‑5.1, incorporates the rules of the common law, which accordingly become the starting point of the analysis. The adjudicator had to decide whether those rules had been ousted by the Public Service Labour Relations Act, R.S.N.B. 1973, c. P‑25 (“PSLRA”), as applied, mutatis mutandis, to the case of a non-unionized employee (ss. 97(2.1), 100.1(2) and 100.1(5)). The common law rules relating to the dismissal of an employee differ completely from the ones provided for in the PSLRA that the adjudicator is regularly required to interpret. Since the common law, not the adjudicator’s enabling statute, is the starting point of the analysis, and since the adjudicator does not have specific expertise in interpreting the common law, the reviewing court does not have to defer to his decision on the basis of expertise. This leads me to conclude that the reviewing court can proceed to its own interpretation of the rules applicable to the non-unionized employee’s contract of employment and determine whether the adjudicator could enquire into the cause of the dismissal. The applicable standard of review is correctness.

[169]                       It is clear from the adjudicator’s reasoning that he did not even consider the common law rules. He said:

An employee to whom section 20 of the Civil Service Act and section 100.1 of the PSLR Act apply may be discharged for cause, with reasonable notice or with severance pay in lieu of reasonable notice. A discharge for cause may be for disciplinary or non-disciplinary reasons. [p. 5]

[170]                       The employer’s common law right to dismiss without cause is not alluded to in this key passage of the decision. Unlike a unionized employee, a non-unionized employee does not have employment security. His or her employment may be terminated without cause.  The corollary of the employer’s right to dismiss without cause is the employee’s right to reasonable notice or to compensation in lieu of notice. The distinction between the common law rules of employment and the statutory rules applicable to a unionized employee is therefore essential if s. 97(2.1) is to be applied mutatis mutandis to the case of a non-unionized employee as required by s. 100.1(5). The adjudicator’s failure to inform himself of this crucial difference led him to look for a cause, which was not relevant in the context of a dismissal without cause. In a case involving dismissal without cause, only the amount of the compensation or the length of the notice is relevant. In a case involving dismissal for cause, the employer takes the position that no compensation or notice is owed to the employee.  This was not such a case. In the case at bar, the adjudicator’s role was limited to evaluating the length of the notice. He erred in interpreting s. 97(2.1) in a vacuum. He overlooked the common law rules, misinterpreted s. 100.1(5) and applied s. 97(2.1) literally to the case of a non-unionized employee.

[171]                       This case is one where, even if deference had been owed to the adjudicator, his interpretation could not have stood. The legislature could not have intended to grant employment security to non‑unionized employees while providing only that the PSLRA was to apply mutatis mutandis.  This right is so fundamental to an employment relationship that it could not have been granted in so indirect and obscure a manner.

[172]                       In this case, the Court has been given both an opportunity and the responsibility to simplify and clarify the law of judicial review of administrative action. The judicial review of administrative action need not be a complex area of law in itself. Every day, reviewing courts decide cases raising multiple questions, some of fact, some of mixed fact and law and some purely of law; in various contexts, the first two of these types of questions tend to require deference, while the third often does not. Reviewing courts are already amply equipped to resolve such questions and do not need a specialized analytical toolbox in order to review administrative decisions. 

[173]                       On the issue of natural justice, I agree with my colleagues. On the result, I agree that the appeal should be dismissed.

APPENDIX

Relevant Statutory Provisions

Civil Service Act, S.N.B. 1984, c. C-5.1

20  Subject to the provisions of this Act or any other Act, termination of the employment of a deputy head or an employee shall be governed by the ordinary rules of contract.

Public Service Labour Relations Act, R.S.N.B. 1973, c. P-25

92(1)  Where an employee has presented a grievance up to and including the final level in the grievance process with respect to

(a)  the interpretation or application in respect of him of a provision of a collective agreement or an arbitral award, or

(b)  disciplinary action resulting in discharge, suspension or a financial penalty,

and his grievance has not been dealt with to his satisfaction, he may, subject to subsection (2), refer the grievance to adjudication.

Public Service Labour Relations Act, R.S.N.B. 1973, c. P-25, as amended

97(2.1)  Where an adjudicator determines that an employee has been discharged or otherwise disciplined by the employer for cause and the collective agreement or arbitral award does not contain a specific penalty for the infraction that resulted in the employee being discharged or otherwise disciplined, the adjudicator may substitute such other penalty for the discharge or discipline as to the adjudicator seems just and reasonable in all the circumstances.

. . . 

100.1(2)  An employee who is not included in a bargaining unit may, in the manner, form and within such time as may be prescribed, present to the employer a grievance with respect to discharge, suspension or a financial penalty.

100.1(3)  Where an employee has presented a grievance in accordance with subsection (2) and the grievance has not been dealt with to the employee’s satisfaction, the employee may refer the grievance to the Board who shall, in the manner and within such time as may be prescribed, refer the grievance to an adjudicator appointed by the Board.

. . .

100.1(5)  Sections 19, 97, 98.1, 101, 108 and 111 apply mutatis mutandis to an adjudicator to whom a grievance has been referred in accordance with subsection (3) and in relation to any decision rendered by such adjudicator.

. . .

101(1)  Except as provided in this Act, every order, award, direction, decision, declaration or ruling of the Board, an arbitration tribunal or an adjudicator is final and shall not be questioned or reviewed in any court.

101(2)  No order shall be made or process entered, and no proceedings shall be taken in any court, whether by way of injunction, judicial review, or otherwise, to question, review, prohibit or restrain the Board, an arbitration tribunal or an adjudicator in any of its or his proceedings.

Appeal dismissed.

Solicitors for the appellant:  Stewart McKelvey, Fredericton.

Solicitor for the respondent:  Attorney General of New Brunswick, Fredericton.