Imperial Oil Limited v. Communications, Energy & Paperworkers Union of Canada, Local 900 et al. [Indexed as: Imperial Oil Ltd. v. Communications, Energy & Paperworkers Union of Canada, Local 900]
96 O.R. (3d) 668
Court of Appeal for Ontario,
Moldaver, Feldman and Cronk JJ.A.
May 22, 2009
Employment -- Labour relations -- Collective agreement -- Interpretation -- Board of arbitration finding that employer's policy of random drug testing of safety sensitive employees in absence of reasonable [page669] cause violated terms of collective agreement -- Board not amending or failing to apply collective agreement when it made reference to "Canadian model" -- Board simply using that term as shorthand for its interpretation of established arbitral jurisprudence on issue of random drug testing -- Board reasonably finding that random drug testing without reasonable cause violated article of collective agreement in which employer made commitment to workplace environment that is free of harassment and where individuals are treated with dignity and respect.
Employment -- Labour relations -- Labour relations board -- Standard of review -- Divisional Court erring by applying standard of patent unreasonableness to award of board of arbitration holding that random drug testing of safety sensitive employees without reasonable cause violated terms of collective agreement -- Standard of review being reasonableness simpliciter.
In 1992, the employer introduced an alcohol and drug policy at its petroleum refinery that provided for random breathalyzer alcohol testing and random urinalysis drug testing of employees in safety sensitive positions. The random testing provisions of the policy were challenged under the Human Rights Code, R.S.O. 1990, c. H.19. On appeal from a decision of the Divisional Court affirming a board of inquiry's decision that several aspects of the impugned provisions contravened the Code, the Court of Appeal held that random breathalyzer testing of employees in safety sensitive positions was a reasonable workplace requirement, but that random urinalysis drug testing was not reasonably necessary to accomplish the employer's legitimate goal of a safe workplace free of drug impairment since the drug testing in question could not measure on-the-job drug impairment but only past drug use. In response to that decision, the employer resumed random drug testing of safety sensitive employees using oral fluid testing technology, which it claimed could show current impairment by cannabis. The union grieved the new form of drug testing. The board of arbitration ruled that the policy provisions regarding the random drug testing of employees, absent reasonable cause, violated the terms of the collective agreement and were void. The employer's application for judicial review of the board's award was dismissed. The employer appealed.
Held, the appeal should be dismissed.
The Divisional Court erred by applying the patent unreasonableness standard of review to the board's award. The award was generally subject to review on a reasonableness standard. The issue of whether the board exceeded its jurisdiction by amending or failing to apply the collective agreement attracted review on the standard of correctness, but the employer's claim that the board erred by relying on facts not established on the evidential record rested on the contention that fairness was compromised. Thus, the standard of review considerations were not relevant.
The board relied on facts established by the record. Both parties referred to a large number of prior Canadian arbitral decisions concerning alcohol and drug testing in the workplace. Many of the arbitration cases addressed such matters as random alcohol or drug testing of employees based on reasonable cause, the usefulness of random or pre-accident testing as a deterrent to the occurrence of workplace accidents, the use of random testing in safety-conscious industries in Canada and the relationship between the prevalence of drug use in industrial workplaces and the incidence of workplace accidents. This was not a case where the board, unbeknownst to the parties, gathered evidence itself, or received evidence [page670] from one party in the absence of the other. Many of the board's challenged remarks were directly responsive to the employer's claim that random drug testing was necessary to achieve general deterrence of workplace drug use. To the extent that the arbitration cases proffered by the parties provided a counterpoint to or undercut the employer's general deterrence justification for random drug testing, it was open to the board to resort to the relevant features of these cases for that purpose.
The board did not amend or fail to apply the collective agreement. Rather, it interpreted the collective agreement in light of the arbitral jurisprudence, the language of the collective agreement and the evidence before it. While the board referred to the "Canadian model", it did not conceive of that concept as a means of formulating and imposing a new scheme for drug testing on the parties. The board defined the term "Canadian model" in its reasons with reference to the existing arbitral jurisprudence in the area of alcohol and drug testing in Canada and indicated that a key feature of the jurisprudence is that arbitrators have overwhelmingly rejected mandatory, random and unannounced drug testing for all employees in a safety sensitive workplace as being an implied right of management under the terms of a collective agreement. The references to the "Canadian model" were intended to capture, in a summary fashion, the board's interpretation of the holdings in the existing arbitration cases on the issue of alcohol and drug testing in a variety of Canadian workplaces. There was nothing objectionable in that approach.
In art. 3.02 of the collective agreement, the parties jointly committed to "a work place environment that is free of harassment and where individuals are treated with dignity and respect". The board found that, absent contractual or statutory consent or reasonable cause to conduct a drug test, random drug testing violated art. 3.02. That interpretation of art. 3.02 was reasonable. On a plain reading of the unqualified language of art. 3.02, there was no basis to conclude that the employer's contractual obligation to treat its employees with respect and dignity was limited to the minimum standards for the protection of human rights embodied in the Human Rights Code.
APPEAL from judgment of Jennings, Kiteley and Swinton JJ., 2008 CanLII 6874 (ON SCDC), [2008] O.J. No. 489, 234 O.A.C. 90 (Div. Ct.) dismissing an application for judicial review of an award of a board of arbitration.
Cases referred to Dunsmuir v. New Brunswick, [2008] 1 S.C.R. 190, [2008] S.C.J. No. 9, 2008 SCC 9, 329 N.B.R. (2d) 1, 64 C.C.E.L. (3d) 1, 164 A.C.W.S. (3d) 727, EYB 2008-130674, J.E. 2008-547, [2008] CLLC Â220-020, 170 L.A.C. (4th) 1, 372 N.R. 1, 69 Imm. L.R. (3d) 1, 291 D.L.R. (4th) 577, 69 Admin. L.R. (4th) 1, 95 L.C.R. 65, folld Entrop v. Imperial Oil Ltd. (2000), 2000 CanLII 16800 (ON CA), 50 O.R. (3d) 18, [2000] O.J. No. 2689, 189 D.L.R. (4th) 14, 137 O.A.C. 15, 2 C.C.E.L. (3d) 19, [2000] CLLC Â230-037, 98 A.C.W.S. (3d) 418 (C.A.), varg Imperial Oil Ltd. v. Ontario (Human Rights Commission), [1998] O.J. No. 422, 108 O.A.C. 81, 35 C.C.E.L. (2d) 56, 98 CLLC 230-012, 1998 CanLII 14954 (ON SC), 77 A.C.W.S. (3d) 608 (Div. Ct.), consd Other cases referred to Canadian General-Tower Ltd. v. United Steel, Paper and Forestry, Rubber, Manufacturing, Energy, Allied Industrial and Service Workers International Union, Local 862, [2008] O.J. No. 1989, 2008 ONCA 404, [2008] CLLC Â220-037, 167 A.C.W.S. (3d) 105, 238 O.A.C. 50, 172 L.A.C. (4th) 97; Construction Labour Relations (Alberta Assn. Operating Engineers (Provincial) Trade Division) v. International Union of Operating Engineers, Local 955, 2004 CanLII 94753 (AB GAA), [2004] A.G.A.A. No. 40, 129 L.A.C. (4th) 1 (Beattie); Hamilton (City) v. Canadian Union of Public Employees, Local 167 (1997), 1997 CanLII 1875 (ON CA), 33 O.R. (3d) 5, [1997] O.J. No. 1063, 99 O.A.C. 155, 97 CLLC Â220-029, 69 A.C.W.S. (3d) 859 (C.A.); Imperial Oil Ltd. v. Communications, Energy and Paperworkers Union of Canada, Local 900 (Policy Grievance), [2006] O.L.A.A. No. 721, 157 L.A.C. (4th) 225, 88 C.L.A.S. 273 (Lab. Arb.); [page671] Kane v. University of British Columbia, 1980 CanLII 10 (SCC), [1980] 1 S.C.R. 1105, [1980] S.C.J. No. 32, 110 D.L.R. (3d) 311, 31 N.R. 214, [1980] 3 W.W.R. 125, J.E. 80-181, 18 B.C.L.R. 124, 2 A.C.W.S. (2d) 117; Keeprite Workers' Independent Union v. Keeprite Products Ltd. (1980), 1980 CanLII 1877 (ON CA), 29 O.R. (2d) 513, [1980] O.J. No. 3691, 114 D.L.R. (3d) 162, 4 A.C.W.S. (2d) 397 (C.A.); Maystar General Contractors Inc. v. International Union of Painters and Allied Trades, Local 1819 (2008), 90 O.R. (3d) 451, [2008] O.J. No. 1353, 2008 ONCA 265, 235 O.A.C. 227, [2008] OLRB Rep. March/April 306, 165 A.C.W.S. (3d) 772, 69 Admin. L.R. (4th) 271, 292 D.L.R. (4th) 554, [2008] CLLC Â220-054; Mugesera v. Canada (Minister of Citizenship and Immigration), [2005] 2 S.C.R. 100, [2005] S.C.J. No. 39, 2005 SCC 40, 254 D.L.R. (4th) 200, 335 N.R. 229, J.E. 2005-1242, 28 Admin. L.R. (4th) 161, 197 C.C.C. (3d) 233, 30 C.R. (6th) 39, 47 Imm. L.R. (3d) 16, 140 A.C.W.S. (3d) 710; Ontario v. Crown Employees Grievance Settlement Board (2006), 2006 CanLII 21173 (ON CA), 81 O.R. (3d) 419, [2006] O.J. No. 2517, 268 D.L.R. (4th) 594, 213 O.A.C. 169, 51 Admin. L.R. (4th) 114, [2006] CLLC Â220-041, 142 C.R.R. (2d) 226, 149 A.C.W.S. (3d) 247, 151 L.A.C. (4th) 129 (C.A.) [Leave to appeal to S.C.C. refused [2006] S.C.C.A. No. 367]; Parry Sound (District) Social Services Administration Board v. Ontario Public Service Employees Union, Local 324 (O.P.S.E.U.), [2003] 2 S.C.R. 157, [2003] S.C.J. No. 42, 2003 SCC 42, 230 D.L.R. (4th) 257, 308 N.R. 271, J.E. 2003-1790, 7 Admin. L.R. (4th) 177, 31 C.C.E.L. (3d) 1, [2003] CLLC Â220-062, 125 A.C.W.S. (3d) 85 Statutes referred to Human Rights Code, R.S.O. 1990, c. H.19 Authorities referred to Brown, Donald J.M., Q.C., and David M. Beatty, Canadian Labour Arbitration, 4th ed. (Aurora, Ont.: Canada Law Book, 2006)
John B. Laskin and R. Ross Wells, for appellant. Douglas J. Wray and Michael A. Church, for respondent.
The judgment of the court was delivered by
I. Introduction
CRONK J.A.: --
[1] The appellant, Imperial Oil Limited ("Imperial"), appeals from the judgment of the Divisional Court dismissing its application for judicial review of that part of a final arbitral award which held that Imperial's policy of random drug testing of certain employees, absent reasonable cause, violated the collective agreement (the "Collective Agreement") at Imperial's petroleum refinery in Nanticoke, Ontario.
[2] Imperial argues that the Divisional Court erred: (i) by applying the patent unreasonableness standard of review to the [page672] arbitral award; and (ii) by failing to find that the board of arbitration erred (a) by relying on purported "facts" drawn from other proceedings and decisions that were not supported by the evidential record before the board; (b) by amending or failing to apply the Collective Agreement; and (c) by interpreting the Collective Agreement without proper regard to the Human Rights Code, R.S.O. 1990, c. H.19 (the "Code").
[3] For the reasons that follow, I conclude that the appeal must be dismissed.
II. Background
[4] In 1992, Imperial introduced an alcohol and drug policy at its Nanticoke refinery that provided for random breathalyzer alcohol testing and random urinalysis drug testing of employees in safety sensitive positions (the "Policy"). The stated objectives of the Policy are as follows:
The objectives of this policy are: (i) to create a safe work environment by reducing the risk of incidents in which drugs or alcohol are a contributing factor; and (ii) to deter the use of alcohol, drugs and other substances where such use may negatively affect work performance and safety. As a matter of policy [Imperial] expects that all employees will be free of alcohol or drugs which could impair their judgment or affect their ability to perform their job safely.
[5] No trade union or collective agreement was in place at the Nanticoke facility when the Policy was implemented. The Collective Agreement between Imperial and the respondent, the Communications, Energy & Paperworkers Union of Canada, Local 900 (the "Union"), was negotiated in 1996.
[6] The Collective Agreement contains a management rights clause. Article 2.01 provides that Imperial has "the exclusive rights to manage and direct all aspects of the refinery operation and the work force". Under this clause, Imperial is empowered to discipline, suspend or terminate the employment of any employee for just cause and to make, enforce and alter work rules, among other matters.
[7] In addition, under art. 3.02 of the Collective Agreement, the parties jointly committed "to a work place environment that is free of harassment and where individuals are treated with respect and dignity". Moreover, art. 19.01 states:
It is agreed by both parties that emphasis shall be placed upon the need for safe and healthy working conditions and practices on [Imperial's] premises. [Imperial] shall continue to make provisions for the safety and health of its employees during the hours of employment. [page673]
(1) First challenge to the Policy
[8] The random alcohol and drug testing provisions of the Policy were challenged under the Code by an Imperial employee. The ensuing board of inquiry hearing encompassed a review of all aspects of the Policy, including its random drug testing provisions. The board of inquiry ultimately concluded that several aspects of the random alcohol and drug testing provisions of the Policy contravened the Code. Imperial's appeal to the Divisional Court from the board of inquiry's decision was dismissed: Imperial Oil Ltd. v. Ontario (Human Rights Commission), [1998] O.J. No. 422, 35 C.C.E.L. (2d) 56 (Div. Ct.).
[9] Imperial's further appeal to this court was allowed in part. In Entrop v. Imperial Oil Ltd. (2000), 2000 CanLII 16800 (ON CA), 50 O.R. (3d) 18, [2000] O.J. No. 2689 (C.A.) ("Entrop"), the board of inquiry's finding that Imperial had developed and implemented the Policy honestly and in good faith was affirmed. The court concluded, however, that the impugned alcohol and drug testing provisions of the Policy were prima facie discriminatory under the Code's prohibition of discrimination based on handicap. The court went on to hold that random breathalyzer alcohol testing of employees in safety sensitive positions, where supervision was limited or non-existent, was a reasonable workplace requirement to achieve a work environment free of alcohol.
[10] The court reached a different conclusion regarding the random urinalysis drug testing provisions of the Policy. In the opinion of the court, these provisions could not be justified as reasonably necessary to accomplish Imperial's legitimate goal of a safe workplace free of drug impairment since the drug testing in question could not measure on-the-job drug impairment but, rather, only past drug use. A positive drug test under the Policy furnished no evidence of impairment or likely impairment on-the-job. As a result, the court concluded that the random drug testing provisions of the Policy as they then existed violated the Code as they were not bona fide occupational requirements.
[11] In the aftermath of Entrop, Imperial continued random breathalyzer alcohol testing, but stopped random urinalysis drug testing at the Nanticoke refinery. In accordance with Entrop, it amended the alcohol testing provisions of the Policy to provide for individualized assessment and employee-specific sanctions for employees who violated the alcohol-related provisions.
[12] After Entrop, Imperial also embarked on an investigation of new drug testing technologies to identify a means by which on-the-job impairment from cannabis, rather than merely past drug use, could be detected. Eventually, Imperial received expert [page674] advice that oral fluid (saliva) drug testing could show current impairment by cannabis. Accordingly, effective July 2003, Imperial resumed random drug testing of safety sensitive employees, using oral fluid testing technology, at the Nanticoke facility.
[13] Under this drug testing method, oral fluid is collected by placing an absorbent pad in an individual's mouth for about two minutes. The oral fluid sample is then placed in a preservative solution, sealed and sent to a laboratory for analysis. Pursuant to Imperial's Policy, the test results are communicated to identified individuals at the Nanticoke refinery several days later. Ultimately, management at the Nanticoke facility is informed only of positive drug test results.
(2) Current challenge to the Policy
[14] In October 2003, the Union grieved Imperial's new form of drug testing. Notwithstanding Entrop, the Union's grievance also challenged Imperial's random breathalyzer testing for alcohol impairment, as well as other aspects of the Policy.
[15] A board of arbitration comprised of Michel G. Picher (chair), John More (the Union's nominee) and Roy C. Filion (Imperial's nominee) (collectively, the "Board"), was appointed to consider the Union's grievance. The Board made two awards pertinent to the issues on appeal.
[16] First, by a preliminary award dated February 20, 2005, the Board unanimously held that the Union was precluded from challenging Imperial's random breathalyzer alcohol testing as the Union did not file any grievance concerning this testing until 2003. Since the Union had thereby acquiesced in Imperial's long-standing alcohol impairment testing, the Board ruled that it would be inequitable for the Union's attack on that testing to proceed. The Union did not challenge this preliminary award.
[17] Second, by a final award dated December 11, 2006 [[2006] O.L.A.A. No. 721, 157 L.A.C. (4th) 225 (Lab. Arb.)], a majority of the Board -- the chair and the Union's nominee (the "Majority") -- allowed the Union's grievance concerning Imperial's drug testing in part.
[18] The Majority upheld the "for cause" and post-incident drug testing provisions of the Policy. It also upheld the Policy's random drug testing measures where such testing formed part of a continuing contract of employment and the rehabilitation of an employee clearly identified as having "a problem of alcohol or drug use".
[19] But the Majority also ruled that the Policy provisions regarding the random drug testing of employees, absent reasonable cause, violated the terms of the Collective Agreement, with [page675] the result that they were null and void. This is the ruling at issue on this appeal.
[20] Imperial's nominee on the Board delivered a partial dissent. While he agreed with the Majority that Imperial could "legitimately utilize drug testing in post-accident or post- incident situations", he also held that the random drug testing provisions of the Policy that were not based on reasonable cause were permissible under the Collective Agreement. In his opinion, the Policy provisions concerning random drug testing for impairment by cannabis using oral fluid testing represented "a reasonable and appropriate means of reducing risk and promoting workplace safety", acted as an effective deterrent, did not contravene the Code and were not inconsistent with the terms of the Collective Agreement or the provisions of the Code. Accordingly, Imperial's nominee would have dismissed the Union's grievance in its entirety.
[21] Imperial applied to the Divisional Court for judicial review, seeking to have that part of the Board's final award that concerned random drug testing, absent reasonable cause, quashed. Writing on behalf of a unanimous Divisional Court, Swinton J. rejected Imperial's claims that the Majority had improperly relied on "facts" from other proceedings and decisions that were not supported by the evidential record before the Board; that the Majority had amended or failed to apply the Collective Agreement; and that the Majority had erred in its interpretation of the Collective Agreement by failing to consider the provisions of the Code. The Divisional Court determined that the standard of review applicable to the Board's final award was patent unreasonableness. Applying that standard, it concluded that the final award was not patently unreasonable and dismissed Imperial's application.
[22] Imperial appeals from the Divisional Court's judgment.
III. Issues
[23] Imperial raises the following issues on appeal: (1) Did the Divisional Court err in its standard of review analysis by applying the patent unreasonableness standard to its review of the Board's final award? (2) Did the Divisional Court err by failing to find that the Majority erred: (a) by relying in its final award on purported facts not established by the record; [page676] (b) by amending or failing to apply the Collective Agreement; and (c) by interpreting the Collective Agreement without proper regard to the Code?
IV. Discussion
(1) Standard of review
[24] In the Divisional Court's opinion, the Board's final award attracted review on the standard of patent unreasonableness. Imperial argues that this was a reviewable error. For the reasons that follow, I agree that the Divisional Court erred by applying the patent unreasonableness standard. Generally, the standard of review applicable to the Board's final award is reasonableness.
[25] After the Divisional Court's decision, the Supreme Court of Canada released Dunsmuir v. New Brunswick, 2008 SCC 9, [2008] 1 S.C.R. 190, [2008] S.C.J. No. 9. The parties accept that since Dunsmuir, there are now only two standards of review: correctness and reasonableness. As well, there is no dispute that under Dunsmuir, the reasonableness simpliciter and patent unreasonableness standards of review were collapsed into a single standard of reasonableness.
[26] The Divisional Court's use of the patent unreasonableness standard accorded with the Supreme Court of Canada's pre-Dunsmuir jurisprudence. Inherent to the application of that standard was recognition by the Divisional Court that a high degree of curial deference is owed to the decisions of labour arbitrators on matters within the labour relations domain. This aspect of the Divisional Court's standard of review analysis was correct. The courts have clearly determined that the decisions of labour arbitrators on interpretive questions concerning collective agreements attract considerable deference from a reviewing court: see Canadian General-Tower Ltd. v. United Steel, Paper and Forestry, Rubber, Manufacturing, Energy, Allied Industrial and Service Workers International Union, Local 862, 2008 ONCA 404, [2008] O.J. No. 1989, (2008), 238 O.A.C. 50 (C.A.), at paras. 14-20; Maystar General Contractors Inc. v. International Union of Painters and Allied Trades, Local 1819 (2008), 2008 ONCA 265, 90 O.R. (3d) 451, [2008] O.J. No. 1353 (C.A.), at paras. 42-43; Parry Sound (District) Social Services Administration Board v. Ontario Public Service Employees Union, Local 324 (O.P.S.E.U.), 2003 SCC 42, [2003] 2 S.C.R. 157, [2003] S.C.J. No. 42, at para. 17. Since Dunsmuir, however, the deferential standard of reasonableness, rather than patent unreasonableness, [page677] generally applies to judicial review of decisions made by those in the labour relations field.
[27] The task of a reviewing court under the reasonableness standard of analysis was explained by the majority in Dunsmuir, at para. 47, in this fashion:
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.
[28] In light of Dunsmuir, the parties accept that the reasonableness standard applies to judicial review of the Board's final award. It also applies where a board of arbitration determines its jurisdiction by interpreting a collective agreement. However, Imperial argues that the application of this standard in this case is subject to two qualifications. First, it submits that the correctness standard of review continues to apply where a labour board of arbitration exceeds its jurisdiction by amending, rather than applying, the collective agreement at issue -- as Imperial alleges occurred in this case. I agree: see Dunsmuir, at para. 59; Hamilton (City) v. Canadian Union of Public Employees, Local 167 (1997), 1997 CanLII 1875 (ON CA), 33 O.R. (3d) 5, [1997] O.J. No. 1063 (C.A.), at paras. 17, 19-20, 25 and 29; Donald J.M. Brown, Q.C., and David M. Beatty in Canadian Labour Arbitration, 4th ed. (Aurora, Ont.: Canada Law Book, 2006), at p. 1-41.
[29] Imperial also submits that questions regarding hearing fairness do not involve standard of review considerations. Again, I agree. Where the grounds of review concern the fairness of the proceeding at issue, including a party's rights to natural justice, an assessment of the appropriate standard of review is neither appropriate nor required: see Ontario v. Crown Employees Grievance Settlement Board (2006), 2006 CanLII 21173 (ON CA), 81 O.R. (3d) 419, [2006] O.J. No. 2517 (C.A.), at paras. 17-23, leave to appeal refused [2006] S.C.C.A. No. 367.
[30] In this case, the application of these principles means that the Board's final award is generally subject to review on the reasonableness standard. A different standard applies to the issue whether the Majority exceeded its jurisdiction by amending or failing to apply the Collective Agreement. In my opinion, as this question concerns the parameters of the Board's jurisdiction, it attracts review on the correctness, rather than the reasonableness, standard. Finally, Imperial's claim that the Majority erred by relying on facts not established by the evidential record rests [page678] on the contention that hearing fairness was compromised. Standard of review considerations are not implicated in the determination of that issue. I did not understand the Union to argue otherwise.
(2) Alleged reliance on facts not established by the record
[31] Imperial argues that the Divisional Court erred by failing to find that the Majority improperly relied on facts drawn from other proceedings and decisions that were not established by the evidential record before the Board. As a result, Imperial submits, hearing fairness was fatally compromised as the factual findings at issue are not supported by the evidence and Imperial was denied the opportunity to be informed of and to make representations about those findings.
[32] In support of this argument, Imperial points especially to the following passages from the Majority's reasons [at paras. 104, 122 and 126]:
As the cases would indicate, the "Canadian model" has gained broad acceptance within safety sensitive industries in Canada. The reported jurisprudence is devoid of any serious incidents or accidents attributed to workplace drug use. That would suggest, as a general rule, that the balancing of interests approach evolved by Canadian arbitrators has been an appropriate, measured and ultimately effective response in balancing the rights of employers and employees in this sensitive area. While it is obviously for each employer to decide which course it feels is appropriate for its enterprise, the fact remains that a significant number of major employers in highly safety sensitive industries in Canada have founded their alcohol and drug testing policies on principles of reasonable cause and have not attempted to force upon their employees mandatory, random, unannounced drug testing. To the extent that those employers and industries have functioned well and have operated safely without apparent difficulty by holding to reasonable grounds as the basis for demanding a drug test, there is little reason to conclude that random, unannounced drug testing of all employees is, of necessity, an essential element for a successful alcohol and drug policy. . . . . .
That balancing must also be done against the back ground of the extensive industrial relations experience of close to twenty years in Canada. That experience, revealed in the cases cited, demonstrates the success of the established Canadian model for alcohol and drug testing in a safety sensitive workplace. As reflected in the arbitral jurisprudence that system, readily accepted by employers such as CN, Dupont and Irving, appears to have allowed major industrial employers to achieve their safety goals without imposing random unannounced alcohol or drug testing on their safety sensitive employees. The success of those employers, achieved in equally safety sensitive industries without recourse to random drug testing, seriously questions the credibility of the argument of counsel for [Imperial] suggesting that past random testing is responsible for the absence of drug related incidents in the refinery. In the final analysis, in balancing the interests of the parties, a question to be considered is not merely whether the extreme approach of random [page679] drug testing for all employees is effective, but also whether it is necessary or justified. The experience of other employers in a variety of safety sensitive industries would appear to suggest that it is not. . . . . .
In our view, given the wide acceptance of the established arbitral jurisprudence, at this point in time it would require clear and unequivocal contractual language to cause a board of arbitration to conclude that employees, through their union, have consented to random and speculative drug testing of their bodily fluids at the will of their employer. That is clearly not the case on the evidence before us.
[33] Imperial complains that these passages include several factual findings by the Majority that are not supported by the evidential record. For example, it asserts that no evidence was led before the Board concerning: (i) the content of other industrial employers' alcohol and drug testing policies; (ii) the asserted "[ready] acceptance" by those employers of the "Canadian model" for alcohol and drug testing and the suggested "success" of that model in safety sensitive workplaces; (iii) the relationship of those employers with their employees; and (iv) the extent to which such employers have "functioned well", "operated safely without apparent difficulty" and "achieve[d] their safety goals" without resort to random drug testing absent reasonable cause.
[34] The Divisional Court considered, and rejected, this complaint. In my view, it was correct to do so.
[35] Boards of arbitration, like other tribunals and, indeed, the courts, are required to base their findings of fact exclusively on evidence that is admissible before them. They enjoy no authority to base their decision on information and material not contained in the evidence before them: see Re Keeprite Workers' Independent Union v. Keeprite Products Ltd. (1980), 1980 CanLII 1877 (ON CA), 29 O.R. (2d) 513, [1980] O.J. No. 3691 (C.A.), at paras. 15-16; Kane v. University of British Columbia, 1980 CanLII 10 (SCC), [1980] 1 S.C.R. 1105, [1980] S.C.J. No. 32, at pp. 1113-14 S.C.R.; Mugesera v. Canada (Minister of Citizenship and Immigration), 2005 SCC 40, [2005] 2 S.C.R. 100, [2005] S.C.J. No. 39, at paras. 41-43. Brown and Beatty put the proposition this way, at p. 3-50:
Apart from circumstances in which he may take a view, or take "judicial notice" of certain facts, an arbitrator cannot gather evidence himself or make any assumptions of fact except through evidence properly put before him. . . . . .
Accordingly, apart from agreed statements of facts and decisions of other competent tribunals, and possibly in those instances where issue estoppel might apply, all other facts must be proved through documentary evidence or through the oral testimony of witnesses. (Citations omitted; emphasis added) [page680]
[36] In this case, the Majority's impugned comments must be viewed in the context of the entirety of its reasons and in light of the positions of the parties on the issues before the Board.
[37] During the Board hearing, both parties referred to and relied on a large number of prior Canadian arbitral decisions concerning alcohol and drug testing in the workplace. It appears to have been uncontested that these decisions formed the backdrop for the Board's consideration of the issues in dispute. Many of the arbitration cases cited by the parties addressed such matters as random alcohol or drug testing of employees based on reasonable cause; the usefulness of random or pre-incident testing as a deterrent to the occurrence of workplace accidents; the use of random testing in safety- conscious industries in Canada; and the relationship between the prevalence of drug use in industrial workplaces and the incidence of workplace accidents.
[38] In these circumstances, neither party could have been taken by surprise by the Majority's detailed consideration of the cited arbitral jurisprudence or by its analysis of that jurisprudence in light of the arguments advanced. The Majority's lengthy reasons (72 pages) contain a detailed description of the evidence adduced at the hearing by both parties and a summary of the parties' positions on the issues before the Board. As part of this summary, the Majority reviewed many of the arbitral authorities supplied by the parties, as well as Entrop, and the parties' positions in respect of those authorities. Later in its reasons, when analyzing the issues before it, the Majority again examined numerous arbitration cases cited by the parties, together with this court's decision in Entrop.
[39] No transcript of the Board hearing is available. However, we have received and reviewed the parties' final written submissions and the arbitral case law provided to the Board. The parties' submissions contain extensive references to the arbitration cases that the parties claimed were relevant to the issues before the Board, including those involving other industrial employers -- e.g., CN, Dupont and Irving -- mentioned by the Majority in its reasons. I note, in particular, that the written outline of the Union's final submissions consists almost entirely of a serial review of Entrop and prior arbitral decisions including, in virtually every instance, references to the facts of those cases.
[40] Thus, both parties had a full opportunity to make and did make those submissions to the Board that they thought advisable concerning the arbitral decisions that they furnished to the Board. In particular, Imperial was aware of and had full opportunity to make representations on the cases that the parties themselves had identified as relevant, including submissions as [page681] to whether those cases were factually distinguishable or similar to the case before the Board and on the overall import of this established body of arbitral jurisprudence.
[41] This is not a case where the Board, unbeknownst to the parties, gathered evidence itself, or received evidence from one party in the absence of the other. It was common ground that the decisions of other boards of arbitration, as identified by the parties, bore directly on the Board's consideration of Imperial's drug testing measures.
[42] Moreover, and importantly, Imperial resisted the Union's grievance and defended the random drug testing provisions of the Policy in part on the basis that such testing is an effective means of deterring workplace behaviour that causes injury. In its final written submissions to the Board, Imperial emphasized deterrence theory, its objective under the Policy of deterring drug use in order to promote workplace safety, the expert evidence before the Board of reduced drug use in workplaces in the United States following the introduction of mandatory, random drug testing and the inadequacy of peer detection of drug impairment as an effective drug-use deterrent in the workplace.
[43] Many of the Majority's challenged remarks are directly responsive to Imperial's claim that random drug testing was necessary to achieve general deterrence of workplace drug use. The Divisional Court therefore correctly observed [at para. 42] that the Majority "[drew] the material relied upon from the decided arbitration cases . . . upon which the parties had made submissions, and [used] it in reply to certain of [Imperial's] submissions". To the extent that the arbitration cases proferred by the parties provided a counterpoint to or undercut Imperial's general deterrence justification for random drug testing, it was open to the Board to resort to the relevant features of those cases for that purpose.
[44] Imperial's real complaint is that the Majority "crossed the line" from permissible to impermissible use of the arbitration cases in question by extracting and using some of the facts of those cases as part of the factual underpinning for its conclusions in this case.
[45] I do not read the Majority's reasons in this fashion. Like the Divisional Court, I view the Majority's impugned comments as reflecting its appreciation of the factual and legal framework in play in the arbitration cases supplied by the parties. I also see the comments in question as indicative of the Majority's attempt to synthesize the underlying basis for the approach taken by other arbitrators to random alcohol or drug testing, as revealed by the arbitration cases cited to the Board. That these were the [page682] Board's purposes is supported by the Majority's references, in the passages from its reasons relied on by Imperial and cited above, to [at para. 104] "the reported jurisprudence [being] devoid of any serious incidents or accidents attributed to workplace drug use"; the lengthy Canadian industrial relations experience [at para. 122] "revealed in the cases cited"; and the acceptance of the "established Canadian model" for alcohol and drug testing [at para. 122] "as reflected in the arbitral jurisprudence" (emphasis added).
[46] That said, certain of the wording in the impugned passages from the Majority's reasons is susceptible to misinterpretation. It would have been preferable, for example, had the Majority expressly indicated that its comments reflected its interpretation of the arbitration cases that it had reviewed and confirmed that its ultimate conclusion on the random drug testing provisions of the Policy, while informed by this arbitral jurisprudence, was based solely on the evidence adduced before it.
[47] Nonetheless, the comments in question must be understood in the context of a relatively lengthy hearing in which both sides placed considerable reliance on and made extensive submissions about an established body of arbitral case law that directly concerned the subject matter of the Board hearing -- random drug testing in the workplace. In that context, it is unremarkable, and certainly not surprising, that the Majority would elaborate on its interpretation of that case law, including its view of the conclusions to be drawn from it, in its reasons. In my opinion, fairly read, the Majority's reasons reveal that it did no more.
[48] I am fortified in this conclusion by the dissent of Imperial's nominee. Nowhere in his lengthy dissent does he voice any concern that the Majority relied in its reasons on facts not established in evidence before it or in respect of which the parties would be taken by surprise. Nor did he assert that hearing fairness had been compromised, or that a breach of natural justice had occurred. If any of these signal events had arguably transpired, it is highly likely that this experienced labour relations lawyer would have voiced this crucial complaint in his objections to the Majority's decision. He did not do so.
[49] For the reasons given, this ground of appeal fails.
(3) Alleged amendment or failure to apply the Collective Agreement
[50] Article 7.05 of the Collective Agreement imposes the following restraint on the powers of a board of arbitration charged with the task of interpreting the Collective Agreement: [page683]
The Board of Arbitration shall not have any power to alter or change any of the provisions of this Agreement or to substitute new provisions for existing provisions, nor to give any decision inconsistent with the provisions of this Agreement.
[51] Before this court, Imperial renews its argument that the Majority breached art. 7.05 and exceeded its jurisdiction by purporting to interpret the Collective Agreement through the lens of what the Majority called the "Canadian model" for alcohol and drug testing in a safety sensitive workplace.
[52] Imperial's argument proceeds in this fashion. It submits that the "Canadian model" has no application in this case for two main reasons. First, the "Canadian model" prohibits random testing for impairment both by alcohol and drugs unless such testing is expressly and clearly negotiated by the parties to a collective agreement. In contrast, as affirmed by the Board's preliminary award and Entrop, the Policy at issue here permits random testing for alcohol impairment, although such testing was not negotiated between the parties. Second, the type of random drug testing at issue under the "Canadian model" -- drug testing by urinalysis -- cannot detect on-the-job drug impairment. On the other hand, Imperial contends, the random oral fluid drug testing envisaged by the Policy is analogous to random breathalyzer alcohol testing since it allows for detection of on-the-job drug impairment, albeit with delayed reporting of test results. Accordingly, as Imperial's Policy is therefore inconsistent with the "Canadian model" in these important respects, the Majority erred by relying on this model, in effect, to impose "an alien scheme" on the parties to the Collective Agreement. This was tantamount, Imperial says, to improperly amending the Collective Agreement or failing to apply it in accordance with its terms.
[53] The Divisional Court disagreed. In its view, the Majority neither amended nor altered the Collective Agreement. Rather, the Majority interpreted the Collective Agreement in light of the arbitral jurisprudence, the language of the Collective Agreement and the evidence before the Board. The Divisional Court put it this way [at paras. 35 and 37]:
There is no express language in the collective agreement permitting random alcohol or drug testing. Therefore, the task for the Board was to determine whether the Policy of random drug testing was a reasonable exercise of management rights under the collective agreement. . . . . .
In its award on the merits, the Board considered the language of the collective agreement in light of the well- established arbitral jurisprudence in Canada -- the "Canadian model". It applied the balancing of interests test [page684] from that jurisprudence and weighed [Imperial's] interest in providing a safe workplace through the deterrence from random drug testing against the employees' privacy interest. That was a perfectly reasonable approach to take in interpreting the scope of the management rights clause [under the Collective Agreement]. (Citation omitted) I see no error in the Divisional Court's conclusion or reasoning on this issue. I say this for several reasons.
[54] First, in my opinion, the Majority did not conceive the concept of a "Canadian model" as a means of formulating and imposing a new scheme for drug testing on the parties. The expression, the "Canadian model", appears to have first been used by another arbitrator in a 2004 arbitral award as a convenient shorthand reference to a best practices model for alcohol and drug policies in the Alberta construction industry: see Construction Labour Relations (Alberta Assn. Operating Engineers (Provincial) Trade Division) v. International Union of Operating Engineers, Local 955, 2004 CanLII 94753 (AB GAA), [2004] A.G.A.A. No. 40, 129 L.A.C. (4th) 1 (Beattie), at para. 4.
[55] In this case, the Majority defined the term "Canadian model" in its reasons with reference to the existing arbitral jurisprudence in the area of alcohol and drug testing in Canada (at paras. 101-104). In particular, the Majority indicated that a key feature of this jurisprudence is "that arbitrators have overwhelmingly rejected mandatory, random and unannounced drug testing for all employees in a safety sensitive workplace as being an implied right of management under the terms of a collective agreement" (at para. 101). Thus, it was in the context of examining the established Canadian arbitral jurisprudence concerning workplace alcohol and drug testing and in attempting to summarize the nature of that jurisprudence that the Majority employed the term "Canadian model".
[56] The Majority commenced its analysis of the random drug testing provisions of the Policy with what it described as a brief review of "the evolution of Canadian arbitral jurisprudence with respect to the right of an employer, in a safety sensitive industry, to demand that an employee undergo a drug test" (at para. 91). The Majority next considered the numerous arbitration cases furnished by the parties that addressed random alcohol or drug testing by employers, and pointed out that those cases supported reasonable cause based -- rather than random -- drug testing in the workplace. The Majority then stated [at paras. 98-99]:
It is fair to say that over time the arbitral jurisprudence in Canada has developed relatively clear lines as to what constitutes an acceptable drug and alcohol testing policy in a safety sensitive workplace which is governed by a collective bargaining regime. [page685]
The foregoing jurisprudence has come to be viewed as tantamount to a Canadian code for drug testing in a safety sensitive workplace governed by collective bargaining, the regime by which terms and conditions of employment must be negotiated between employers and unions. They have become widely accepted and applied. Indeed, the drug testing policies and limitations fashioned within that jurisprudence came to be recognized as the "Canadian model" as adopted in the construction industry in Alberta. (Citations omitted; emphasis added)
[57] The Majority went on to describe some of the key elements of this "'Canadian model' for alcohol or drug testing in a safety sensitive workplace as developed in the arbitral jurisprudence" (at para. 100). Imperial does not suggest that this description was inaccurate.
[58] The structure and content of this part of the Majority's reasons reveal that its references to the "Canadian model" were intended to capture, in a summary fashion, its interpretation of the holdings in the existing arbitration cases on the issue of alcohol and drug testing in a variety of Canadian workplaces. There is nothing objectionable in this approach. It confirms that the Majority used the prior arbitral decisions on drug and alcohol testing as a comparative aid to its interpretation of the Collective Agreement and the Policy by contrasting the latter to the alcohol and drug testing measures of other employers as recorded in the cited Canadian arbitral jurisprudence.
[59] Contrary to Imperial's submission, this does not mean that the Majority failed to appreciate or ignored the differences between the Collective Agreement and Imperial's Policy and those alcohol and drug testing policies scrutinized in other arbitration cases and Entrop.
[60] The fact that Imperial's Policy, in contrast to the testing policies of other employers, legitimately permits random breathalyzer alcohol testing was front and centre to the entire hearing before the Board. The Majority's reasons afford no evidence that it somehow lost sight of this important factor when analyzing the provisions of the Policy and the Collective Agreement. On the contrary, the Majority expressly recognized that the Board's preliminary award and an earlier arbitration case in Alberta had both addressed Imperial's right to conduct random breathalyzer tests for the detection of alcohol impairment (at para. 102).
[61] Moreover, the Majority was alert to Imperial's contention that its random oral fluid drug testing was analogous to the random alcohol breathalyzer testing approved in Entrop. As it was entitled to do, the Majority considered, and rejected, this contention on the basis of the evidence before it that oral fluid drug testing in fact did not permit immediate detection of drug impairment on-the-job (at paras. 64 and 112-13). [page686]
[62] Imperial essentially argues that the Majority erred by improperly grafting onto its analysis the over-arching concept of a "Canadian model" for drug testing, when that model differed from and had no relevance to consideration of Imperial's Policy, without recognizing or coming to grips with the distinctions between the "Canadian model" and Imperial's drug testing regime under the Collective Agreement.
[63] This claim distorts the Majority's use of the "Canadian model" and of the existing arbitral jurisprudence as a yardstick against which the Policy's drug testing provisions could be assessed. The Majority's review of this jurisprudence identified and highlighted some of the key differences between the drug testing undertaken in other industrial workplace settings in Canada and that implemented under the Policy. This was a reasonable analytical approach to the interpretation of the Policy and the Collective Agreement. It did not result in the amendment of or failure to apply the Collective Agreement.
[64] In the end, I agree with the Divisional Court's conclusion on this issue. The Majority's reasons reveal that its rejection of Imperial's random drug testing measures, absent reasonable cause, was based on the language of the parties' own bargain as embodied in the Collective Agreement and the evidence adduced before the Board regarding the requisite balancing of interests inherent to the examination of a random drug testing policy in the workplace.
[65] I therefore reject this ground of appeal.
(4) Suggested need to reference the [Code](https://www.canlii.org/en/on/laws/stat/rso-1990-c-h19/latest/rso-1990-c-h19.html)
[66] The Majority held that, absent contractual or statutory consent or reasonable cause to conduct a drug test, random drug testing under the Policy violated Imperial's contractual commitment under art. 3.02 of the Collective Agreement to treat individuals in the workplace with "respect and dignity". In the view of the Majority, "[f]ailing to accord such respect goes to the dignity of the person" (at para. 126).
[67] The Majority further held that, art. 3.02 aside, such drug testing could not be justified "on a responsible application of the balancing of interests approach in a safety sensitive environment that has carefully evolved over the decades within the arbitral jurisprudence in Canada" (at para. 126). The Majority noted that Imperial's own record of work at the Nanticoke refinery did not include even "one substantiated case of an employee working impaired by reason of the consumption of a drug" (at para. 127). Moreover, there was no "significant evidence of drug use generally within the workforce away from work, or within [page687] the surrounding community" (at para. 127). The Majority therefore concluded that Imperial's random drug testing, absent reasonable cause, was not justified and constituted "an unwarranted intrusion on [employees'] privacy" and "an unjustifiable affront to their dignity" (at para. 127).
[68] In finding a breach of art. 3.02 of the Collective Agreement, the Majority made no reference to the Code. Imperial argues that in interpreting art. 3.02, the Majority erred by failing to have proper regard to the Code and this court's decision in Entrop.
[69] The Divisional Court held that the Majority's interpretation of art. 3.02 was reasonable, in the context of the patent unreasonableness standard of review, and that the Majority was not required to have regard to the Code when interpreting art. 3.02. For three reasons, I arrive at the same conclusion on application of the reasonableness standard articulated in Dunsmuir.
[70] First, as observed by the Divisional Court, art. 3.02 makes no mention of the Code. Nor is the Code mentioned elsewhere in the Collective Agreement. There is no dispute, however, that the substantive rights and obligations of the Code form part of the Collective Agreement: see Parry Sound (District) Social Services Administration Board, at paras. 28 and 55.
[71] I agree with the Union's submission that on a plain reading of the unqualified language of art. 3.02, there is no basis to conclude that Imperial's contractual obligation to treat its employees with respect and dignity is limited to the minimum standards for the protection of human rights embodied in the Code. On the contrary, the wide cast of art. 3.02 is inconsistent with the notion that the parties intended to deal under that clause with only the forms of discrimination prohibited by the Code.
[72] Second, the Majority provided several cogent reasons for its holding that the random drug testing measures at issue offended art. 3.02. These included: (i) the method and timing of the random drug tests, which the Majority found do not permit the immediate detection of impairment from cannabis at the time of testing but, rather, only after the passage of the several days required for analysis of the test specimen and reporting of the test results; (ii) Imperial's obligation to respect an employee's expectation of privacy absent consent to or reasonable cause for a random drug test; and (iii) the fact that this court's decision in Entrop did not assist on the issue of the scope of art. 3.02 since Entrop did not involve the interpretation or the application of the Collective Agreement. The Majority's reasons on this issue are clear and intelligible. They provide ample justification for its [page688] conclusion that Imperial's random drug testing, absent reasonable cause, offended art. 3.02.
[73] Finally, I am unable to say that the Majority's interpretation of art. 3.02 falls outside the range of possible and defensible interpretive outcomes with respect to the construction of art. 3.02.
[74] I conclude, therefore, that the Majority's interpretation of art. 3.02 is reasonable within the meaning of Dunsmuir. Given this conclusion, it is unnecessary to address the Majority's alternate holding that, regardless of art. 3.02, the random drug testing provisions of the Policy, absent reasonable cause, cannot be justified under the Collective Agreement without clear and unequivocal contractual language signifying employee consent to such testing measures.
V. Disposition
[75] For the reasons given, I would dismiss the appeal. I would award the Union its costs of Imperial's leave to appeal application to this court and of the appeal, fixed in the total amount of $15,000, inclusive of disbursements and GST. As there is no basis on which to do so, I would not interfere with the Divisional Court's award of costs in the sum of $7,500 in the Union's favour.
Appeal dismissed.