Canadian Union of Public Employees, Local 933 v. Cape Breton (Regional Municipality)

Canadian Union of Public Employees, Local 933 v. Cape Breton (Regional Municipality) Court Court of Appeal Date 2006-06-29 Citation 2006 NSCA 80 Docket CA 249471 Judge/Registrar/Adjudicator Bateman, Nancy J. (Honourable Justice) (CA); Cromwell, Thomas A. (Honourable Justice); MacDonald, J. Michael (Honourable Chief Justice) (CA) Document Type Decision Relations Library Sheet - Canadian Union of Public Employees, Local 933 v. Cape Breton (Regional Municipality) - 2006 NSCA 80 - 2006-06-29 - Library Sheet Decision Content NOVA SCOTIA COURT OF APPEAL Citation: Canadian Union of Public Employees, Local 933 v. Cape Breton (Regional Municipality), 2006 NSCA 80 Date: 20060629 Docket: CA 249471 Registry: Halifax Between: Canadian Union of Public Employees, Local 933 Appellant v. Cape Breton Regional Municipality Respondent Judges: MacDonald, C.J.N.S.; Bateman and Cromwell, JJ.A. Appeal Heard: March 24, 2006, in Halifax, Nova Scotia Held: Appeal allowed per reasons for judgment of Cromwell, J.A.; MacDonald, C.J.N.S. and Bateman, J.A. concurring. Counsel: Lionel G. Clarke, for the appellant Eric Durnford, Q.C. and Rebecca Pitts, for the respondent Reasons for judgment: I. INTRODUCTION: [1] A judge quashed a labour arbitrator’s award and the union appeals. The dispute between the parties concerns how to apply a re-opener clause in their collective agreement. [2] The re-opener clause permits an arbitrator to set new wage rates if there have been substantial increases in employees’ duties or responsibilities. A group of employees claimed that their duties and responsibilities had increased substantially. The employer did not agree and the dispute went to arbitration. The arbitrator found there had been substantial changes. He directed the parties to negotiate rates of pay and retained jurisdiction in case they could not agree. [3] The critical question before the arbitrator and on judicial review, was what changes in duties the arbitrator could consider. For the employer, each collective agreement is a fresh start. The parties had agreed to rates of pay in the collective agreement which reflected the duties at that time. It follows, said the employer, that only changes after that should count in applying the re-opener clause. The union took a different view, which might be called the “everything counts” approach. According to the union, the arbitrator could look at all changes, whenever they occurred because that is what the agreement says. [4] The arbitrator ended up taking a middle road between these two positions. He found that he was entitled to look at all of the changes over time, but could only grant relief if there were sufficient changes since the signing of the agreement to “crystalize” the substantial change. This might be called the “accretion” approach because it recognizes that there may have been a gradual accretion of duties over time. [5] The employer sought judicial review of the award, arguing that the arbitrator had made a reviewable error in adopting this “accretion” approach. The chambers judge agreed and quashed the award. He found that the appropriate standard of review was reasonableness and that the arbitrator had been unreasonable by failing to give effect to the intention of the parties. Hence, the union’s appeal. [6] There are two issues to be resolved. The first is whether the judge applied the correct standard of review. The second is whether the arbitrator made a reviewable error. [7] In my respectful view, the judge applied the wrong standard of review and erred in quashing the award. My conclusion is that the applicable standard of review is patent unreasonableness and that the arbitrator’s award should be upheld on review using that standard. I am also of the view that the judge incorrectly applied the reasonableness standard of review which he thought was applicable. He simply substituted his view for that of the arbitrator about how the collective agreement should be interpreted. He ought instead to have asked himself whether there was any line of reasoning in the award which provided tenable support for the arbitrator’s conclusion. I would therefore allow the appeal, set aside the order of the learned chambers judge and restore the arbitrator’s award. II. FACTS AND DECISIONS: [8] To put the case in context, I must briefly review the basic facts relating to the dispute, the relevant provision of the collective agreement, the arbitrator’s award and the judge’s decision to quash it. A. The Dispute and the Collective Agreement: [9] Under the collective agreement between the parties, they are to negotiate the new rates of pay when management substantially increases the duties or responsibilities in any classification. If they cannot agree, the issue may be submitted to arbitration. This is as provided for in Article 35: 35.01 When the duties or responsibilities in any classification are substantially increased by management, or where the union alleges that an employee is incorrectly classified, or when a position not covered in Appendix “A” is established during the term of this Agreement, the rate of pay shall be subject to negotiations between the employer and the union. If the parties are unable to agree on the reclassification and/or rate of pay of the job in question, such dispute shall be submitted to grievance. [10] Employees in two classifications, the communications operators and the communications operators supervisors, thought that their job duties or responsibilities had increased substantially. The employer did not agree and they filed a group grievance. B. The Arbitrator’s Award: [11] The grievance eventually proceeded to arbitration before Milton Veniot, Q.C. An issue arose as to the time period to be used for assessing whether there had been a substantial increase in duties and responsibilities. The employer, adopting a “fresh start” approach, contended that the rates of pay for all classifications were set out in the collective agreement and, therefore, the arbitrator could look only at changes since it had been signed on February 13, 2002. The union’s position – an “everything counts” approach – was that the arbitrator could consider changes in the positions over a longer period of time. [12] This dispute about the time frame confronted the arbitrator with a difficult dilemma. The collective agreement did not expressly set out the relevant time frame. Excluding evidence of all changes prior to the signing of the current agreement could perhaps be unjust in some circumstances. One could imagine relatively minor changes occurring over the course of a succession of agreements and that their cumulative effect might only become substantial during the course of the current agreement. Some might find it wrong to ignore that sort of “accretion” of duties over time. On the other hand, allowing the union to rely on all changes, whenever occurring, could undercut the finality of collective bargaining. What is the point of signing a contract, it could be asked, if the union, having not achieved what it wanted at the negotiating table, could try again at arbitration? [13] Faced with this dilemma, the arbitrator selected a sort of middle ground. His approach, which I have called the “accretion” approach, did not absolutely exclude evidence of changes to the jobs before the signing of the current agreement. However, it did require that there be changes to the job duties and responsibilities which crystalized the grievance during the current collective agreement. [14] This approach was founded on the award authored by arbitrator Hamilton in Selkirk and District General Hospital and Canadian Union of Public Employees (CUPE) Local 1601 (Re), [1993] M.G.A.D. No. 9. Under the Selkirk principles, earlier changes may be considered if, when coupled with the changes under the current agreement, they crystalize the grievance during the life of the current agreement. As arbitrator Hamilton put it: We concur with Mr. Myers that there can be an accretion of new duties over a (reasonable) period of time which can bring Article 29.01 into operation. Minor changes or revisions to a particular classification's duties may not, in both parties' view, be a "substantial change" but the addition of some other critical duties may lead to the conclusion that a substantial change has then taken place and this would justify addressing the issue of salary levels at that time. ... [15] While suggesting that the Selkirk principles might support a broader approach, Arbitrator Veniot nonetheless adopted what he called the “more modest test established by Selkirk.” He found that he could act under the re-opener clause in Article 35 of the agreement in two situations: first, if a “... substantial change has occurred entirely within the present agreement’s term ...” and second, where changes, not in themselves substantial, occurred in the term of the current agreement which, “ ... when added to those accreted under the previous agreement, “crystallize” within the term of the present agreement, the right to refer to the sum of the whole of the “new” and “accreted” changes as ‘substantial’.” [16] Applying that approach, he found that there had been a number of new duties and/or responsibilities which had been imposed since the signing of the current agreement. He set them out in detail. He then concluded that those changes, considered in light of previous changes, crystallized the grievance. He accordingly directed the parties to attempt to negotiate new rates of pay and, if they could not agree, retained jurisdiction to set the rates as contemplated by Article 35.01 [17] The arbitrator dealt at length with the employer’s two objections to this approach. [18] First, the employer submitted that an arbitrator receives his authority under the current agreement and has no authority to act under any other agreement. In response to this point, the arbitrator referred first to the text of the agreement. He noted that it did not expressly bar his consideration of evidence before the signing of the current agreement. The arbitrator found that to do so would not be acting outside the scope of the current agreement but giving effect to it. [19] Following the Selkirk award, he found that admitting evidence of changes which occurred before the current agreement had been signed did not offend the general principle of arbitration law that a board appointed under a collective agreement only has authority to address breaches alleged to have occurred under that agreement. As the arbitrator put it, “... it does no damage to that principle if events which occurred prior to the time of the new agreement are used to establish a breach which is alleged to have occurred in the term of the new agreement.” He found that to adopt the opposite rule, would confuse the alleged breach with the evidence by which it is to be proved. [20] He found, applying this principle, that the “... union easily meets the test of having changes in the duties and responsibilities of the classifications occur during the term of the agreement, as there have been a number of new duties and/or responsibilities which have been imposed since February 13, 2002.” [21] The employer’s second objection was that the union was estopped from or had waived the right to assert changes predating the signing of the current collective agreement. The argument, in a nutshell, was that by agreeing to the wage rates set out for these classifications in the collective agreement, the union could not now allege that those same rates had been not proper at the time they were agreed upon. [22] The arbitrator rejected this contention for two reasons. [23] First, he found that as a result of the inclusion of Article 35, the text of the collective agreement expressly provides the union with the right to grieve where the duties or responsibilities in any classification are substantially increased and the parties cannot agree on new rates of pay. He noted that Article 35 does not specifically limit these substantial increases to those occurring during the term of the agreement. The arbitrator then considered whether such a limit should be implied. He found that such an implication would be contrary to the intention of the parties. He noted, referring to the “third scenario” in Article 35, that where the parties wanted to limit the operation of the article to events which occurred entirely within the agreement term, they did so expressly. [24] For ease of reference, I will reproduce Article 35.01 again, but in a way that shows more clearly the three “scenarios” to which the arbitrator refers: 1. when the duties or responsibilities in any classification are substantially increased by management, or 2. where the union alleges that an employee is incorrectly classified, or 3. when a position not covered in Appendix “A” [that is, the list of rates of pay for each classification] is established during the term of this Agreement, the rate of pay shall be subject to negotiations between the employer and the union. If the parties are unable to agree on the reclassification and/or rate of pay of the job in question, such dispute shall be submitted to grievance. [25] As the arbitrator noted, this article specifically sets out the relevant time frame with respect to the “third scenario” which it addresses, that is, management establishing a new classification. The time is restricted to “during the term of this Agreement”. However, in the other two “scenarios” addressed in the article, no time period is specified. Thus, the arbitrator concluded that “... it was a part of the bargain signed on February 13, 2002 that the union could, after negotiating a wage, attempt to increase it after the agreement was signed by using Article 35.” [26] The arbitrator had a second reason for rejecting the employer’s position. He found that the evidence did not support any estoppel of, or waiver by, the union in relation to grievances under Article 35.01. C. Judicial Review: [27] The employer sought judicial review and LeBlanc, J. in Supreme Court chambers quashed the award. After a thorough review of the intricate case law on standard of review, he held that reasonableness was the applicable standard. Applying it, the judge found the award to be unreasonable. In doing so, the judge substituted his view of what the parties intended for the view of the arbitrator. The judge expressed his critical conclusion as follows: [72] I conclude the parties must have contemplated that the date of the analysis of whether there had been an increase in duties and responsibilities for this classification unit must have been from the beginning of the current collective agreement. I believe that it was unreasonable to go back in history to the beginning of the relationship between the parties to determine if there had been a significant increase in duties and responsibilities. ... [78] I find that the arbitrator acted unreasonably in considering increases in duties and responsibilities that occurred prior to February 13, 2002. Article 35.01 must be read with reference to the object and purpose of the collective agreement. The Arbitrator failed to consider the importance of Article 1.02. He thereby failed to consider the intentions of the parties in concluding the collective agreement, leading to a result that cannot reasonably be supported by the reasons given. (Emphasis added) III. ANALYSIS: A. Standard of Review: [28] The appellant submits that the judge applied the wrong standard of judicial review and that the appropriate standard is patent unreasonableness. It is submitted that the judge erred in his interpretation of the privative clauses that apply to this arbitrator and misapplied the Supreme Court’s decision in Voice Construction Ltd. v. Construction & General Workers’ Union , Local 92, [2004] 1 S.C.R. 609; S.C.J. No. 2 (Q.L.). [29] The respondent supports the judge’s conclusion concerning standard of review and submits he did not err in his assessment of the four contextual factors in this case or in his application of the Voice Construction decision. [30] In my respectful view, the judge erred in finding that reasonableness, rather than patent unreasonableness, is the applicable standard of review. Specifically, he erred in two respects. First, he did not give sufficient weight to the privative protection which the arbitrator has under the applicable legislation and the terms of the agreement. Second, he misinterpreted one aspect of the judgment in Voice Construction. I will summarize my reasons by examining each of the four contextual factors. But first it will be helpful to say something about the Supreme Court’s decision in Voice Construction. [31] At least since the 1998 decision in Pushpanathan v. Canada (Minister of Citizenship and Immigration), [1998] 1 S.C.R. 982, the Supreme Court of Canada has “... unequivocally accepted the primacy of the pragmatic and functional approach to determining the standard of judicial review of administrative decisions”: Dr. Q v. College of Physicians and Surgeons of British Columbia, [2003] 1 S.C.R. 226. This means that while “... the wisdom of past administrative law jurisprudence need not be wholly discarded ...”, the applicable standard of review must be determined by applying the pragmatic and functional analysis, not by slotting the case in to traditional pigeon-holes. [32] That said, the law has recognized for many years – both before and after the adoption of the pragmatic and functional approach to determining the standard of review – that the courts should afford great deference to labour arbitrators interpreting collective agreements: see Volvo Canada Ltd. v. International Union, United Automobile, Aerospace & Agricultural Implement Workers of America (UAW), Local 720, [1980] 1 S.C.R. 178, at p. 214; Alberta Union of Provincial Employees, Branch 63 v. Board of Governors of Olds College, [1982] 1 S.C.R. 923, at p. 935; United Brotherhood of Carpenters and Joiners of America, Local 579 v. Bradco Construction Ltd., [1993] 2 S.C.R. 316, at pp. 337-39; Canada Safeway Ltd. v. RWDSU, Local 454, [1998] 1 S.C.R. 1079, at paras. 58-60. [33] This highly deferential approach does not rest primarily on a minute parsing of the words used to give arbitrators privative protection. Rather, it rests on policy considerations relating to the centrality to the whole scheme of collective bargaining of grievance arbitration by expert arbitrators: United Brotherhood v. Bradco, supra at pp. 334 - 339. The policy basis of this deference was stressed in Dayco (Canada) Ltd. v. CAW-Canada, [1993] 2 S.C.R. 230, a case like Bradco in which the pragmatic and functional approach was applied. LaForest, J. at 250 - 251 of Dayco said this: ... This Court has stated in previous cases that courts should, as a matter of policy, defer to the expertise of the arbitrator in questions relating to the interpretation of collective agreements; ... the arbitrator is acting within his or her "home territory", and any judicial review of that interpretation must only be to a standard of patent unreasonableness. (Emphasis added) [34] While Voice Construction reminds us of the primacy of the pragmatic and functional approach, it does not question, let alone erase, the importance of judicial deference to labour arbitrators. Voice does not say that labour arbitrators interpreting the provisions of a collective agreement will always, or even generally, be reviewed for reasonableness: Nova Scotia Government and General Employees Union v. Capital District Health Authority, [2006] N.S.J. No. 153 (Q.L.)(C.A.). 1. Privative clause: [35] The first of the contextual factors is the presence and wording of any privative clause or statutory mechanism for review. In this case, the relevant statutory provisions, as well as the collective agreement, provide that the arbitrator’s award shall be “final and binding”. [36] The appellant and the respondent disagree about the extent of the preclusive effect of these “final and binding” clauses and about the judge’s assessment of it. The appellant says that the words “final and binding” are the equivalent of final and conclusive and should, therefore, be taken to have full privative effect. The respondent, on the other hand, submits that the final and binding language does not warrant a high level of deference. In my view, the judge underestimated the preclusive effect of the ‘final and binding’ provisions in this case. [37] The effect of a particular ‘final and binding’ clause depends on the context. Such clauses have, at times, been found to have no preclusive effect while, at other times, the same words have been taken as indicating full preclusive effect: see Donald Brown and John Evans, Judicial Review of Administrative Action in Canada, looseleaf updated to July 2005 (Toronto: Canvasback, 1998) at para. 13.5330. In my view, the provisions in issue in this case, read in their full context, signal that significant deference should be accorded to the arbitrator’s interpretation of the collective agreement. [38] In the present case, the arbitrator’s decision receives privative protection from three sources: the collective agreement itself, the Trade Union Act, R.S.N.S. 1989, c. 475 (“TUA”) and the Arbitration Act, R.S.N.S. 1989, c. 19. The collective agreement in Article 11.11 provides that the decision of the arbitration board is “... final and binding and enforceable on all parties.” The TUA requires a collective agreement to contain a provision for final settlement of all differences between the parties concerning its meaning or violation. It also provides the text of a provision that is deemed to be part of a collective agreement that does not have such a provision. That deemed inclusion provides that the decision of the arbitrator is final and binding upon the parties: TUA s.42(1) and (2). Similarly, the Arbitration Act deems every submission to arbitration, in the absence of a contrary intention expressed in the submission, to contain a provision that the arbitrator’s award is final and binding on the parties. [39] The relevant statutory provisions must, like all statutes, be interpreted in accordance with the modern principle of statutory interpretation: the words must be read in their entire context, in their grammatical and ordinary sense harmoniously with the scheme of the Act, the object of the Act and the intention of Parliament: see, e.g. Rizzo and Rizzo Shoes (Re), [1998] 1 S.C.R. 27 and Bell ExpressVu Ltd. Partnership v. Rex, [2002] 2 S.C.R. 559. [40] Important context for the interpretation of these provisions is the purpose and scheme of collective bargaining statutes such as the TUA. Estey, J. summarized the broad context of modern collective bargaining law in St. Anne Nackawic Pulp and Paper Co Ltd v. Canadian Paper Workers Union Local 219, [1986] 1 S.C.R. 704 at 717-19: ... The legislature created the status of the parties [to a collective agreement] in a process founded upon a solution to labour relations in a wholly new and statutory framework at the centre of which stands a new forum, the contract arbitration tribunal. Furthermore, the structure embodies a new form of triangular contract with but two signatories, a statutory solution to the disability of the common law in the field of third party rights. These are but some of the components in the all-embracing legislative program for the establishment and furtherance of labour relations in the interest of the community at large as well as in the interests of the parties to those labour relations. . . . The collective agreement establishes the broad parameters of the relationship between the employer and his employees. This relationship is properly regulated through arbitration ... The more modern approach is to consider that labour relations legislation provides a code governing all aspects of labour relations, and that it would offend the legislative scheme to permit the parties to a collective agreement, or the employees on whose behalf it was negotiated, to have recourse to the ordinary courts which are in the circumstances a duplicative forum to which the legislature has not assigned these tasks. [41] It is within the context of this broad legislative purpose, which includes by statutory design a significant limitation of the role of the courts in collective bargaining law, that the final and binding provision in this case must be understood. [42] It is also important to remember that the nature of the privative protection afforded by a “final and binding” clause cannot be considered in the abstract, divorced from the other contextual factors. The nature of the issue raised before the arbitrator, in particular, goes hand in glove with the wording of the privative clause. [43] Dayco is a good example. In that case, as noted earlier, LaForest, J. affirmed in unequivocal language the deferential posture of the courts in relation to an arbitrator’s interpretation of a collective agreement. However, he also found, in the circumstances of that case, that the “final and binding” language of s. 44 of the Labour Relations Act, R.S.O.1980, c. 228 (which is very similar to s. 42 of the NS TUA) was “... more a limited shield against judicial review...”: p. 264. He was careful to point out that this determination was made in the context of a question which he characterized as “jurisdictional”. As he said at p. 268, “I cannot accept that courts should mechanically defer to a tribunal simply because of the presence of a “final and binding” or “final and conclusive” clause. These finality clauses can clearly signal deference, but they should also be considered in the context of the type of question and the nature and expertise of the tribunal.”(emphasis added) [44] The Dayco approach was referred to with approval by the Court in Alberta Union of Provincial Courts v. Lethbridge Community College, [2004] 1 S.C.R. 727 at para. 16 reviewing an arbitrator’s interpretation of a statute. [45] This contextual approach to the interpretation of privative clauses is well illustrated by the majority decision in Canada Safeway Ltd. v. RWDSU, Local 454, [1998] 1 S.C.R. 1079, a decision released the same day as Pushpanathan. The case concerned the review of an arbitration board’s finding that a reduction of scheduled hours of work constituted a constructive lay-off under the collective agreement. The relevant provision in the Saskatchewan Trade Union Act, R.S.S. 1978, c. T-17, s. 25 provided that the decision of the arbitration board was “final and conclusive” and “binding upon the parties”. Cory and McLachlin (as she then was), JJ., concluded that the appropriate standard of judicial review was patent unreasonableness. They commented on the general posture of curial deference to decisions of labour arbitrators when interpreting collective agreements and the significance of the statutory privative protection in these words: 58 Where labour relations tribunals are called upon to interpret or apply a collective agreement under the umbrella of a privative clause, a reviewing court can only intervene in the case of a patently unreasonable error. (See Canada (Attorney General) v. Public Service Alliance of Canada, [1993] 1 S.C.R. 941, and Toronto (City) Board of Education v. O.S.S.T.F., District 15, [1997] 1 S.C.R. 487). This high degree of curial deference is essential to maintain the integrity of the system which has grown to be so efficient and effective in the resolution of disputes arising in the sensitive field of labour relations. The nature of labour disputes requires their speedy resolution by expert tribunals. The protective clause found in the Trade Union Act is the legislative recognition of the fundamental need for deference to the boards’ decisions. 59 Section 25(1) of the Act states that the finding of the arbitrator or board of arbitration is “final and conclusive” and “binding upon the parties”. Although this is not a true privative clause, it is very close to it. In United Brotherhood of Carpenters and Joiners of America, Local 579 v. Bradco Construction Ltd., [1993] 2 S.C.R. 316, Sopinka J. considered a statute pertaining to a board of arbitration which contained a similar statutory provision. It was to the effect that issues arising from the interpretation of a collective bargaining agreement were to be submitted to arbitration “for final settlement”. It was held that judicial deference to the decision of the arbitration board was warranted. The sage observation was made that an unlimited scope of judicial review would thwart the goal of mandatory arbitration to provide an efficient and cost effective manner of resolving disputes in this field. Those comments are appropriate to this case. 60 Further, in the case at bar, the Board was acting within its exclusive jurisdiction to interpret and apply the collective agreement when it considered whether the layoff provisions were applicable to Ms. Hardy’s situation. It follows that there can be no doubt that the appropriate standard of review in this case is one of patent unreasonableness. 61 This standard of review has now been well established and accepted. (Emphasis added) [46] These words are important for the present case for at least two reasons. First, they affirm in the age of the full pragmatic and functional approach, that the arbitration board’s interpretation of the agreement should be reviewed for patent unreasonableness. The judgment describes this standard as “well established and accepted.” Second, the Court equates the effect of a “final and conclusive” clause with a “final and binding” clause as we have in this case. [47] This deferential approach in the context of a “final and binding” clause is reflected in the judgment of the Ontario Court of Appeal in Lakeport Beverages v. Teamsters Local Union 938 (2005), 77 O.R. (3d) 543; O.J. No. 3488 (Q.L.). That case, too, concerned review of an arbitrator’s interpretation of the a collective agreement. The relevant legislation provided that the arbitrator’s award should be “final and binding”: Labour Relations Act, S.O. 1995, c. 1, s. 48. Laskin, J.A. noted that for many years, both the Supreme Court of Canada and the Ontario Court of Appeal had held that where Ontario labour arbitrators are called on to interpret provisions of a collective agreement, their decisions are reviewable on a standard of patent unreasonableness. He concluded that the “final and binding” provision was a “reasonably strong” privative clause and signaled “... a large measure of deference to the arbitrator’s decision” at para. 27. [48] This is consistent with the approach taken to privative clauses in Pushpanathan. There, Bastarache, J. noted that a so-called “full” privative clause is one which declares that decisions of the tribunal are final and conclusive from which no appeal lies and all forms of judicial review are excluded. He went on, citing Bradco, to observe that “ [u]nless there is some contrary indication in the privative clause itself, actually using the words “final and conclusive” is sufficient, but other words might suffice if equally explicit ...”: para. 30. [49] This reference to Bradco is important. It indicates that Bradco’s approach to a “final settlement” provision remains pertinent in the full-blown pragmatic and functional approach now applicable. [50] Bradco considered the privative effect of a stipulation that the grievance arbitrator’s decision constituted a “final settlement” of the dispute. Sopinka, J. observed at p. 334: The first point to be made is that the word "final" will not always imply an intended restriction on judicial review. .. Whether or not the word "final" should be interpreted in any particular case as conveying an intention to preclude or restrain judicial review requires an analysis of the provision in light of the purpose, nature and expertise of the tribunal to the decision of which it refers. ... ... The goal of mandatory arbitration is to arrive at an efficient and cost-effective manner of resolving disputes which will enable the parties to continue in their ongoing working relationship to the greatest extent possible. As an unlimited scope of judicial review of an arbitrator's decision would thwart the achievement of this goal, the words "final settlement" must be taken to indicate the legislature's intention that the courts exercise some restraint in this area. (Emphasis added) ... ..... The questions to be resolved [... by the arbitrator ...] involved the interpretation of the collective agreement and its application to a particular factual situation - matters which constitute the core area of an arbitrator's expertise. Combined with the purpose and wording of s. 88, which confers upon the arbitrator exclusive jurisdiction to come to a final settlement of disputes arising out of the interpretation or application of the collective agreement, the arbitrator's relative expertise mandates that the court defer to the decision of the arbitrator in this case unless his decision is found to be patently unreasonable. ... (p. 339) [51] Far from minimizing the significance of final settlement provisions for the standard of review, Bradco, Canada Safeway and Pushpanathan confirm that such provisions may well signal a high level of deference to an arbitrator interpreting a collective agreement. [52] I do not interpret Voice Construction as any retreat from this long held and recently expressed approach. Rather, in my view, Voice Construction simply affirms that privative clauses must be interpreted, and their impact assessed, in the full context of the pragmatic and functional approach. [53] To sum up on this point, the “final and binding” provisions relevant to this case, properly interpreted in their full context, provide a strong indication of judicial deference to the arbitrator’s interpretation of the collective agreement. I respectfully, but emphatically, disagree with the respondent’s contention that these words in the circumstances of this case do not warrant a high level of deference. [54] The appellant submits that the chambers judge erred in his analysis of the meaning and effect of the privative provisions in this case. I agree. [55] At para. 48 of his reasons, the judge said this: [48] Unlike the Alberta legislation, the Nova Scotia Act includes no restriction on the right to seek judicial review. In Voice Construction this suggested an increased consideration to the decisions of labour arbitrators (para. 26). A partial or less exacting privative clause, in the absence of other factors, does not allow for the greatest degree of deference, but requires a careful assessment of the arbitrator’s role. [56] I understand the judge to be saying that the Alberta legislation at issue in Voice included a restriction on the right to seek judicial review and that the absence in Nova Scotia of such a restriction tends to suggest a less deferential approach. In my respectful view, Voice Construction stands for exactly the opposite proposition. [57] The Alberta legislation at issue in Voice Construction contains a provision which has no counterpart in Nova Scotia. Section 143 (1) of the Labour Relations Code, S.A. 1988, c. L-1.2 provides that, subject to s. 2, an arbitrator’s award may not be questioned or reviewed in any court. However, s. 143(2) then goes on to specifically permit an award to be questioned or reviewed by way of an application for judicial review provided it is taken not later than 30 days after the date of the award. As I read the reasons of Major, J. in Voice Construction, he found that s. 143(2) limited what would otherwise be the privative effect of s. 143(1): para. 23. He added in para. 24 that “[t]he allowance for certiorari in the LRC is ... a factor to be taken into account when conducting the pragmatic and functional analysis ...”. In A.U.P.E. v. Lethbridge Community College, supra, the Court considered the similarly worded provisions in s. 63(1) and (2) of the Public Service Employee Relations Act, R.S.A. 2000, c. P-43. Iacobucci, J., for the Court, noted that the strong language of the first subsection providing for no review was “... undercut by the concomitant thirty-day period of review by way of application for certiorari or mandamus” provided for under the second sub-section”: para. 16. In other words, the specific allowance for review by way of certiorari or mandamus in the statutes in issue in Voice and Lethbridge was a factor tending towards less deference. [58] In my view, the judge erred in reaching the opposite conclusion. [59] The “final and binding” provisions applicable here while not, strictly speaking “full” privative clauses are, when read in the context of the purpose and scheme of the TUA and the system of collective bargaining labour relations which it regulates, indicative of a high degree of deference to an arbitrator’s interpretation of a collective agreement. 2. The Purpose of the Act as a Whole and the Provision in Particular: [60] The analysis of this factor is not controversial. As the respondent says, it is well established that, because the system of mandatory binding arbitration is so important to the entire labour relations regime, this factor usually militates in favour of deference. [61] Where there is a difference of view is how the nature of the question before the arbitrator relates to his expertise. I will, therefore, consider those two contextual factors together. 3. Expertise and Nature of the Question: [62] The expertise of the tribunal is to be assessed by characterizing the expertise of the tribunal in question, considering the court’s expertise relative to that of the tribunal and identifying the nature of the specific issue before the administrative decision-maker relative to that expertise. [63] On this point, the judge found that, “[a]s the arbitrator was interpreting a term of the collective agreement which was effectively at the core of his expertise, this suggests that more deference than the correctness standard is required”: para. 50. [64] It is generally accepted that, as the judge said, an arbitrator interpreting a collective agreement is operating at the core of his or her expertise. This tends towards more rather than less deference: see for example, Bradco, supra; Voice Construction at paras. 27 - 28. The difference between the parties in this case concerns how the arbitrator’s expertise relates to the question which was before him. [65] If I read the judge’s reasons correctly, he found that the fundamental question before the arbitrator was the interpretation of the collective agreement. The appellant agrees with this characterization of the question. It submits that the arbitrator did not face any issue concerning jurisdiction, the interpretation of external statues or general questions of law. The appellant contends, as I understand the judge to have found, that the nature of the question, in relation to the arbitrator’s expertise, supports a deferential standard of review. [66] The respondent, however, characterizes the question differently and says that it does not fall within the core expertise of the arbitrator. For the respondent, the question is one of law that had “jurisdictional aspects”. This is so, the respondent argues, because the issue related to whether the arbitrator could consider events relevant to a dispute which could have arisen under an earlier expired agreement. The submission invites us to conclude from this that no deference is due to the arbitrator on account of relative expertise. [67] I do not accept the respondent’s position on this point. Even if we are to continue to characterize issues as “jurisdictional”, I do not agree that there is any “jurisdictional aspect” – to use the respondent’s words – of the question before the arbitrator. [68] This is not a case which gives rise to a “jurisdictional” issue as that term was used in Dayco and Bradburn v. Wentworth Arms Hotel Ltd., [1979] 1 S.C.R. 846. In those cases, the “ ... existence or subsistence of the collective agreement itself is called into question ...” : Dayco at 260. Rather, the issue here is whether the provisions of the collective agreement bear upon the problem at hand and if so, how: see Dayco at 260. As LaForest, J. pointed out in Dayco at p. 260, “.. [T]he most frequent challenge of an arbitrator’s jurisdiction is an assertion by one of the parties that the incident underlying a grievance is not contemplated by the collective agreement. These issues are resolved by the arbitrator’s application of the facts to the agreement as he or she interprets it, and this process is clearly intended to be left to the expertise of the arbitrator.” (Emphasis added) In my view, that is precisely what the arbitrator in this case was called upon to do and that task fell within the core of his expertise. 4. Conclusion on Standard of Review: [69] All of the four contextual factors point to deference. The question before the arbitrator was one requiring the interpretation of a collective agreement and he was, therefore, operating at the core of his expertise. The importance of mandatory grievance arbitration in the overall scheme of collective bargaining law and the concomitant limitation of judicial involvement in the process support deference. The “final and binding” provisions applying to this arbitration also support a finding of considerable deference. [70] I conclude, respectfully, that the judge erred in selecting reasonableness rather than patent unreasonableness, as the appropriate standard of judicial review. 5. Defining the Standard: [71] In applying either the patent reasonableness or the reasonableness standard, the court must not simply compare what it thinks is the right answer with what the tribunal has done and then permit the tribunal some “margin of error”. This approach was taken in the past: see, e.g. CAIMAW v. Paccar of Canada Ltd., [1989] 2 S.C.R. 983 at 1017 - 18 per Sopinka, J., writing for himself, and Lamer, J. (as he then was). However, it is now clear that this is not the proper approach. As Iacobucci, J. said for a unanimous Court in Ryan at para. 50, “... when deciding whether an administrative action was unreasonable, a court should not at any point ask itself what the correct decision would have been. ... The standard of reasonableness does not imply that a decision-maker is merely afforded a “margin of error” around what the court believes is the correct result.” (emphasis added) [72] The critical question, therefore, is whether the reasons of the tribunal provide any line of analysis that could reasonably lead the tribunal from the evidence to its conclusion. In other words, do the reasons provide “tenable support” for the decision: Ryan at paras. 55 - 56; Granite Environmental Inc. v. Nova Scotia (Labour Relations Board) (2005), 238 N.S.R. (2d) 59; N.S.J. No. 441 (Q.L.)(C.A.) at paras. 42 - 43. [73] When reviewing for patent unreasonableness, the question is whether the absence of such tenable support is clear; the defect should be “immediate and obvious.” The decision must be “so flawed that no amount of curial deference would justify letting it stand.” Such decisions have been characterized as being “clearly irrational”, “evidently not in accordance with reason” or as “almost border[ing] on the absurd”: Ryan at para. 52; Voice Construction at para. 18. As with the reasonableness standard, the touchstone is the tribunal’s reasoning, not the court’s view of the ‘correct’ result. B. Did the Arbitrator Make a Reviewable Error? [74] The appellant submits that, even if the applicable standard of review is reasonableness, the judge made a fundamental error in the way he applied it. The appellant says that, in effect, the judge determined what he thought the parties intended and then quashed the arbitrator’s award for failing to give effect to that intent. In other words, the appellant claims that the judge simply substituted his view of what the agreement means for that of the arbitrator. [75] Respectfully, I agree with this submission. As is clear from Ryan and the other authorities I have referred to earlier, the judge is to use the arbitrator’s reasoning as the touchstone of review, not the judge’s own sense of what the agreement means. [76] The judge found that “... the Arbitrator’s award fail[ed] to consider the fundamental issue of the intentions of the parties...” and that “... the parties must have contemplated that the date of the analysis of whether there had been an increase in duties and responsibilities for this classification unit must have been from the beginning of the current collective agreement.” In taking this approach, the judge simply substituted his view about the intention of the parties for that of the arbitrator. This, with respect, was not the judge’s function. [77] As noted earlier, the arbitrator set out at some length the reasons which he thought supported his conclusion. He attached considerable importance to the fact that, where the parties had intended to limit changes to a particular time period, they did so expressly as they did, for example, in the final ‘scenario’ addressed by Article 35 of the collective agreement. He considered whether such a limitation should be implied elsewhere in the agreement where the parties had not expressly included it and found that “...such an implication would be contrary to the intention of the parties ...”. While this, and the other reasons set out by the arbitrator may not be compelling to all, they are firmly grounded in the text of the collective agreement and provide some tenable support for the result which the arbitrator reached. The award should, therefore, not be set aside if reviewed on the reasonableness standard. [78] The respondent says that the arbitrator was “wrong” not to follow two other arbitration awards in Nova Scotia which touched on similar issues. Respectfully, whether he was right or wrong is not the question and in any case, one arbitrator is not obliged to follow the decision of another absent provisions to that effect: see Halifax Employers Association v. International Longshoremen’s Association Local 269 (2004), 226 N.S.R. (2d) 159 (C.A.) at para 82; N.S.J. No. 316 (Q.L.)(C.A.) (Application for leave to appeal ref’d [2004] S.C.C.A. No. 464) [79] The respondent also submits that the Arbitrator was “legally wrong” by failing to look to the purpose of a particular provision in the collective agreement as an aid to determining its meaning. Further, the only plausible purpose of Article 35, says the respondent, is to provide a mechanism by which the employer could be required to negotiate higher pay rates in the event that, during the term of the agreement, the job duties significantly increased. [80] In support of this position, the respondent cites Boeing Canada Technology Ltd. (Re), [2001] M.G.A.D. No. 61(Q.L.), a decision authored by Arbitrator Hamilton, who was also the author of the Selkirk award relied on by the arbitrator in this case. In my view, the Boeing award does not support the respondent’s position and instead makes clear why the arbitrator’s award in the present case is not unreasonable or patently unreasonable. [81] The Boeing award shows that there had been a disagreement among arbitrators on the question in issue in this case: see paras. 306 - 311. This tends to suggest that it is a point upon which reasonable people may reasonably differ. More significantly, Arbitrator Hamilton reiterated support for the view expressed in Selkirk and which the arbitrator in the present case adopted. That view was that the arbitrator is entitled to consider the “accretion” of duties under predecessor agreements provided that there are changes within the ambit of the current agreement which “crystallize” the grievance. As Arbitrator Hamilton put it in Selkirk (and as adopted by the arbitrator in the present case), “[w]hile the evidence can show a course of conduct or an accretion of duties, there still must be a finding that the substantial change in duties arose or crystallized during the term of the Agreement itself.” [82] The arbitrator in the present case cited and adopted this passage from Selkirk. He reiterated that this was the basis for his decision. This line of reasoning, in my respectful view, while perhaps not persuasive to everyone, is not unreasonable: it provides tenable support for the conclusion. [83] The respondent also relies on the principles set out in Bradburn and Dayco, supra for the proposition that provisions in collective agreements must be interpreted contextually and further that it is presumed that the agreement is prospective in nature. The respondent asks, rhetorically, whether in can be reasonable to hold that “... the rates of pay set by the Agreement in the normal course of regular, periodic collective bargaining can be undermined merely by showing that at some undefined point in time over some undefined period of time the duties of a particular classification have “substantially increased” and a wage re-opener triggered merely by the alteration of one insignificant term of employment on any day after the agreement is signed?” [84] The assumption embedded in this rhetorical question is not supported by a fair reading of the arbitrator’s award. As noted, he adopted the “more modest test” established in Selkirk which required the substantial increase in duties and responsibilities to “crystalize” in the term of the current agreement. He decided that is what had occurred here and set out at some length the evidence of events occurring after February 13, 2002 which he thought supported that conclusion. The changes identified by the arbitrator cannot be characterized as “the alteration of one insignificant term of employment” or provide a basis to suggest that his approach would sanction recourse to Article 35 in such a situation. I do not think that the arbitrator gave the agreement anything other than prospective effect. [85] Respectfully, I do not think it can be said that the arbitrator here failed to consider the intention of the parties. The respondent in effect invites us to follow the example of the chambers judge and to substitute our view of those intentions for the arbitrator’s. This we may not do. [86] Even if, contrary to my view, the appropriate standard of review is reasonableness rather than patent unreasonableness, this award does not disclose reviewable error. [87] It follows of course, that on review for patent unreasonableness, which is in my view the applicable standard, the award should stand. IV. DISPOSITION: [88] I would allow the appeal, set aside the order of the chambers judge dated May 19, 2005, direct that any costs paid pursuant to that order be refunded and restore the award of the arbitrator. The appellant should have costs both before the chambers judge and in this court fixed in total at $2400 inclusive of disbursements. [89] I would be remiss if I did not express my appreciation to counsel for their thorough and helpful submissions, both oral and written. Cromwell, J.A. Concurred in: MacDonald, C.J.N.S. Bateman, J.A.