2011 NBCA 58, 2011 NBCA 58

COURT OF APPEAL OF NEW BRUNSWICK 133-10-CA COMMUNICATIONS, ENERGY AND PAPERWORKERS UNION OF CANADA , LOCAL 30 APPELLANT SYNDICAT CANADIEN DES COMMUNICATIONS, DE L’ÉNERGIE ET DU PAPIER,

SECTION LOCALE 30 APPELANT - and - - et - IRVING PULP & PAPER, LIMITED RESPONDENT LES PÂTES & PAPIER IRVING, LIMITÉE INTIMÉE Communications Energy and Paperworkers Union of Canada, Local 30 v. Irving Pulp & Paper, Limited, 2011 NBCA 58 Syndicat canadien des communications, de l’énergie et du papier,

section locale 30 c.

Les Pâtes et Papier Irving, Limitée, 2011 NBCA 58 CORAM: The Honourable Chief Justice Drapeau The Honourable Justice Turnbull The Honourable Justice Robertson CORAM : L’honorable juge en chef Drapeau L’honorable juge Turnbull L’honorable juge Robertson Appeal from a decision of the Court of Queen's Bench: September 17, 2010 Appel d’une décision de la Cour du Banc de la Reine : Le 17 septembre 2010 History of Case: Historique de la cause : Decision under appeal: Unreported Décision frappée d’appel : Inédite Preliminary or incidental proceedings: N/A Procédures préliminaires ou accessoires : s.o.

Appeal heard: February 17 and May 17, 2011 Appel entendu : Le 17 février et 17 mai 2011 Judgment rendered: July 7, 2011 Jugement rendu : Le ​​​​7 juillet 2011

Reasons for judgment by: The Honourable Justice Robertson Motifs de jugement : L’honorable juge Robertson Concurred in by: The Honourable Chief Justice Drapeau The Honourable Justice Turnbull Souscrivent aux motifs : L’honorable juge en chef Drapeau L’honorable juge Turnbull Counsel at hearing: For the appellant: David G. Gauthier For the respondent: William B. Goss, Q.C. and Melissa M. Everett Withers Avocats à l’audience : Pour l’appelant : David G. Gauthier Pour l’intimée : William B. Goss, c.r., et Melissa M. Everett Withers THE COURT The appeal is dismissed with costs of $5,000.

LA COUR Rejette l’appel avec dépens de 5 000 $. The judgment of the Court was delivered by ROBERTSON, J.A. I. Introduction [ 1 ] The respondent, Irving Pulp & Paper, Limited, operates a kraft paper mill along the banks of the St. John River, at the point where the River empties into the Bay of Fundy. The mill is also contiguous to the “world famous” Reversing Falls and straddles the line which separates the West Side of the City of Saint John from its North End.

In 2006, Irving unilaterally adopted a workplace policy which included mandatory and random alcohol testing, by breathalyser, for employees holding safety sensitive positions. Randomness is achieved by having the names of the 334 prospective testees selected by an off-site computer. In any 12-month period, the computer selects 10% of the names on the list. It is common ground the employer possesses the right to adopt policies dealing with workplace safety provided they do not conflict with the collective agreement.

Correlatively, the union has a right to challenge those policies on the ground they fail to meet the test of reasonableness imposed under what labour lawyers have labelled the KVP rules. [ 2 ] In the present case, an Irving employee and member of the appellant union, who occupied a safety sensitive position, was randomly tested. The test revealed a blood alcohol level of zero. Nevertheless, a policy grievance was filed challenging the “without cause” aspect of the policy.

Applying a balancing of interests approach, the majority of the arbitration board determined that Irving failed to establish a need for the policy in terms of demonstrating the mill operations posed a sufficient risk of harm that outweighs an employee’s right to privacy. Specifically, the majority concluded Irving had not adduced sufficient evidence of prior incidents of alcohol related impaired work performance to justify the policy’s adoption. At the same time, the majority accepted that a “lighter

burden of justification” was imposed on employers engaged in the operation of “ultra-hazardous” or “ultra-dangerous” endeavours. On the facts, however, the majority concluded that, while the mill operation represented a “dangerous work environment”, the mill operation did not fall within the ultra-dangerous category such as a nuclear plant, an airline, a railroad, a chemical plant or a like industry. This explains why the majority went on to examine the evidence relating to alcohol use in the workplace.

Based on the evidence adduced the majority concluded there was insufficient evidence of a “significant degree of incremental safety risk that outweighed the employees’ privacy rights”. The dissenting panel member characterized the workplace as “highly dangerous” and, therefore, evidence of an alcohol problem in the workplace was not a condition precedent to establishing the reasonableness of the policy.

Alternatively, the dissenting member held Irving had adduced sufficient evidence of such a problem. [ 3 ] The application for judicial review was allowed and the majority decision removed into the Court of Queen’s Bench and quashed. The review decision is now reported as 2010 NBQB 294 , 367 N.B.R. (2d) 234. The application judge characterized the disagreement between the majority and dissenting opinions in terms of the distinction between a workplace that is dangerous and one that is ultra-dangerous (e.g. nuclear reactor facility).

The application judge interpreted the majority decision as holding that, if the workplace falls within the ultra-dangerous category, no evidence of alcohol related incidents in the workplace is required. However, if the workplace falls within the dangerous category, as does the Irving mill, the employer must still adduce sufficient evidence of a pre- existing alcohol problem. The application judge held it unreasonable to require evidence demonstrating such a history once the majority concluded the kraft mill represented a dangerous workplace where the “potential for catastrophe exists”.

Applying the review standard of reasonableness, as outlined in Dunsmuir v. New Brunswick , 2008 SCC 9 , [2008] 1 S.C.R. 190, the application judge ruled the board’s decision was unreasonable because it did not fall within the “range of possible, acceptable outcomes which are defensible in respect of the facts and law” (para. 47). [ 4 ] While the parties agree the application judge did not err in adopting the review standard of reasonableness, I have not been persuaded this is so.

Admittedly, one would think that an arbitrator’s decision regarding the reasonableness of an employer’s policy relating to safety in the workplace is a question of fact, or mixed fact and law, for which judicial deference is owed in accordance with the tenets of the deference doctrine set down in Dunsmuir . However, this understanding requires reconsideration once it is recognized the reasonableness of the policy involves a fundamental question: In what circumstances must an employer adduce evidence of a substance abuse problem in the workplace in order to justify the policy’s adoption?

In the present case, this question may be restated as follows: Must an employer’s decision to adopt a policy of mandatory random alcohol testing for employees holding safety sensitive positions be supported by sufficient evidence of alcohol related incidents in the workplace? In point of fact, the arbitration board went further and framed the question in terms of requiring evidence supporting a “significant problem with alcohol-related impaired performance at the plant”. [ 5 ] In my view, the answer to the question is subject to the review standard of correctness for two reasons.

First, the question posed raises a pure question of law, one that seeks to strike a reasonable balance between an employer’s legitimate interest and obligation to provide a safe workplace and the privacy and dignity interests of employees or, in some instances, their freedom from discrimination. As such, the case raises a question of general importance in the law over which the arbitration board cannot assert a greater relative expertise than the courts. Indeed, some might argue that at its core this appeal is of importance to the public at large having regard to the location of the kraft mill.

Second, the arbitral jurisprudence is not always reconcilable or easily so. Often, the same case is cited for opposing propositions. Moreover, the distinction which the arbitration board makes between dangerous and ultra dangerous workplaces is simply not part of the arbitral framework surrounding the validity of alcohol and drug testing policies. The same holds true in regard to the requirement that the employer adduce evidence of a significant alcohol or drug problem in the workplace. Hence, it falls on this Court to provide certainty so far as the law of New Brunswick is concerned.

At the same time, I do not want to leave the impression that the arbitral jurisprudence is irrelevant to the analysis at hand. Let me explain. [ 6 ] The union’s position is encapsulated in two related propositions: (1) arbitrators in Canada have overwhelmingly rejected mandatory, random and unannounced drug and alcohol testing; and (2) sufficient evidence of a pre-existing drug or alcohol problem in the workplace is a pre-condition to the enforceability of such policies, unless the workplace qualifies as ultra- dangerous.

Accordingly, the union maintains the application judge erred in setting aside the arbitration board’s decision. [ 7 ] A review of the relevant arbitral jurisprudence, leads me to conclude that arbitrators and arbitration boards have not overwhelmingly rejected mandatory random unannounced alcohol testing in the workplace. Admittedly, acceptance of drug testing policies in the workplace has met with more resistance than those aimed at alcohol testing by breathalyser.

I also conclude the arbitration board erred in law in holding that an employer must adduce sufficient evidence of an alcohol problem in the workplace, unless the workplace is declared to fall within the ultra-dangerous category. Having due regard to the arbitral jurisprudence, I follow the more recent cases which hold that, once a workplace is declared “inherently dangerous”, there is no need for the employer to establish the existence of an alcohol problem in the workplace. I acknowledge that the earlier decisions (1990s) reflect a reluctance to embrace alcohol and drug testing in the workplace.

This is particularly true in regard to drug testing, but less so with respect to random alcohol testing because of the minimally intrusive nature of the testing procedure and because testing is restricted to employees holding safety sensitive positions. [ 8 ] In my view, the law would rest on a sounder footing if it were to recognize that mandatory random alcohol testing in the workplace is justified once the employer establishes the workplace operations to be inherently dangerous, thereby

eliminating the need to adduce evidence of an alcohol problem in the workplace. Having justified the adoption of the alcohol testingpolicy, its reasonableness will be confirmed, provided the testing is done by breathalyser and applies only to those employees who holdsafety sensitive positions. This is the point in time where the employer’s and employees’ rights are reasonably balanced. II. Background – The Impugned Decisions [9] On February 1, 2006, Irving adopted a policy which mandated drug and alcohol testing for employees who hold“safety sensitive positions” as defined in the policy.

In part, that definition reads: “Safety Sensitive Position is a position which thecompany determines has a role in the operation where impaired performance could result in a significant incident affecting the health andsafety of employees, customers, customers’ employees, the public, property or the environment.”

Part V (iii)(

c) of the policy embracesmandatory, unannounced random alcohol testing (without cause). Randomness is achieved by having the names of prospective testees(334 employees) selected by an off-site computer. In any 12-month period, the computer selects 10% of the names on the list ofemployees holding safety sensitive positions. After selection, names are not removed from the list - lest the policy lose its deterrenteffect for the reminder of the year. Hence, it is possible the same name could be selected more than once during the 12-month period.Testing is done by a breathalyser device.

A positive result (a blood alcohol level greater than .04%) constitutes a violation of the policyand results in the imposition of a disciplinary measure, up to the point of employee dismissal. The proper discipline is a matter to bedecided on a case by case basis. On March 13, 2006, an employee holding a safety sensitive position, Percy Day, was approached andasked to submit to a breathalyser test. Mr. Day, who does not imbibe in alcoholic beverages for religious reasons, submitted to the testbecause of the disciplinary measures prescribed under the policy. As would be expected, the test read negative (0.0%).

A policygrievance was filed and heard by an arbitration board appointed under the Industrial Relations Act, R.S.N.B. 1973, c. I-4. [10] From the outset, it has been common ground that an employer has the unilateral right to adopt workplace rulesprovided they fall within the analytical framework set out in the seminal decision of KVP Co. v. Lumber & Sawmill Workers’ Union,Local 2537 (Veronneau Grievance), [1965] O.L.A.A. No. 2 (QL).

That decision holds that the enforceability of such workplace rulesdepends on compliance with the following criteria: (1) the rule must not be inconsistent with the collective agreement; (2) it must bereasonable; (3) it must be clear and unequivocal; (4) it must be brought to the attention of the employee affected before the company canact on it; (5) the employee concerned must have been notified that breach of the rule could result in his or her discharge if the rule wereused as a foundation for discharge; and (6) the rule should have been consistently enforced by the company from the time it wasintroduced.

In labour law circles, the six rules are universally referred to as the KVP rules.

In the present case, the arbitration board hadonly to focus on the second rule: whether the component of the policy dealing with random alcohol testing was reasonable. [11] The majority of the arbitration board (hereafter the arbitration board or board) addressed the matter ofreasonableness, first against the backdrop of what was said in Dunsmuir, second, by reference to what the learned authors Brown and Beatty have stated in their leading text Canadian Labour Arbitration, 4th ed. looseleaf (Aurora, Ont.: Thomson Reuters Canada Limited,2011) and, third, by reference to four decisions.

Two are court decisions: Imperial Oil Ltd. v. Communications Energy PaperworkersUnion of Canada, Local 900, [2006] O.L.A.A. No. 721 (QL), aff’d at [2008] O.J. No. 489 (QL) and at 2009 ONCA 420, [2009] O.J. No.2037 (QL) (hereafter Nanticoke), and Entrop v. Imperial Oil Ltd., (ON CA), [2000] O.J. No. 2689 (C.A.) (QL)(hereafter Entrop). The remaining two are arbitral decisions: Re: Communication Energy and Paperworkers Union, Local 770 andImperial Oil Ltd. (unreported, May 27, 2000, Christian, Chair) and Greater Toronto Airports Authority v. Public Service Alliance ofCanada, Local 0004, [2007] C.L.A.D.

No. 243 (QL). [12] Based on the above jurisprudence, the arbitration board concluded that for the employer to establish thereasonableness of its alcohol testing policy, the employer had to “demonstrate on the evidence that the risk it appreciates and addressesby its Policy exists to a degree sufficient to justify the means chosen to address it”.

The majority stated that the term risk could be used todescribe at least three different situations: (1) the inherent risk associated with the performance of a particular job or class of jobs; (2) therisk attached to a particular enterprise, considered on its own; and (3) the risk associated with the enterprise considered as an exemplar ofa highly safety sensitive position. The board also acknowledged jurisprudence supporting the understanding that an employer must leadevidence of an existing alcohol related problem in the workplace in order to succeed.

However, the board opined that that understandingwas “somewhat overbroad”. It went on to hold: Evidence of risk may be available from the nature of the industry itself. The cases recognize a lighter burden of justification on anemployer engaged in the operation of an ultra-hazardous endeavour. [13] In support of the above conclusion the arbitration board cited the seminal decision of Canadian NationalRailway Co. v. National Automobile, Aerospace, Transportation and General Workers Union of Canada (CAW-Canada), (CA LA), [2000] C.L.A.D. No. 465 (QL) (M.G. Picher, Chair) (hereafter CN Rail); Weyerhaeuser Co. v.

Industrial Wood andAllied Workers of Canada, [2004] B.C.C.A.A.A. No. 71 (QL) (hereafter cited as Weyerhaeuser I); and Dupont Canada Inc. v.Communications, Energy Paperworkers Union of Canada, Local 28-0 (Drug and Alcohol Policy Grievance), [2002] C.L.A.D. No. 146(QL) (hereafter Dupont Canada), (P.C. Picher, Chair). In brief, the board ruled that to establish the validity and enforceability of theimpugned testing policy, Irving had to establish the need for the policy based on the presence of a sufficient risk of harm. As to the

evidential burden imposed on Irving, the majority recognized that the more risk sensitive the industry was the easier it would be for theemployer to justify its alcohol testing policy by reference to documented instances of alcohol related incidents in the workplace.

When itcame to ultra-dangerous or ultra-hazardous industries, such as airlines or nuclear reactors, it was acknowledged that evidence of a pre-existing alcohol problem in the workplace would not be required. [14] After an extensive review of the employer’s evidence, the arbitration board concluded that the “operation ofthe plant under normal circumstances carries with it the risk that certain malfunctions could have repercussions going well beyond thesafety of the actor who caused the incident”. This lead the board to conclude that the “mill in normal operation is a dangerous workenvironment”.

In reaching this conclusion, the majority relied in part on the evidence of one of Irving’s retired employees, Mr.Moorehouse, who described the mill in the following fashion: “It is heavy industry. It is a kraft mill. It’s probably as close as you can getto a chemical plant. There are a lot of chemicals”. However, the arbitration board was not prepared to accept that Irving’s kraft millcould be classified as an ultra-hazardous enterprise.

The board concluded: “There is not a sufficient case made out to put the operation ofthis kraft mill in the same category of risk as a nuclear plant, an airline, a railway or a chemical plant, or like industries”. In light of thisevidence, the board went on to examine the evidence of a pre-existing alcohol problem in the workplace or, as the majority expressed it,evidence of alcohol-related incremental risk. On behalf of Irving, Mr. Moorehouse, identified eight specific alcohol related incidents ofwhich five involved an employee showing up for work under the influence of alcohol.

The five incidents spanned 15 years: April 29,1991, to January 11, 2006. Following the policy’s adoption, no employee tested positive, nor were there any positive tests with respect totests administered for reasonable cause. [15] The arbitration board went on to conclude that the evidence “cannot be said to be indicative of a significantproblem with alcohol-related impaired performance at the plant” (emphasis added).

The board also went on to hold there was indirectevidence from which one could infer that the management of Irving “does not regard the incremental safety risk posed by alcohol in thisplant as being high among the incumbents in the safety sensitive positions”. The inference was drawn from the fact that the testingpolicy was limited to 10% of the 334 names of the list of employees who held safety sensitive positions.

The majority reasoned that, hadIrving been truly concerned with the matter of risk, it would have selected a higher percentage as did other employers as reflected in thejurisprudence. [16] The arbitration board went on to consider the fact that from the date of the policy’s implementation, February1, 2006, to the date of the hearing, December 16, 2008, a total of 33 months had elapsed. Yet, there had been no positive random alcoholtests at the mill and no positive tests for reasonable cause. During this period approximately 114 random breath tests had beenadministered.

Of these only 17 were administered to members of the appellant union. Mr. Moorehouse attributed the lack of positiveresults to the deterrent effect of the policy. The board concluded that no inference favourable to Irving was justified: “Perhaps so, but onhis own record evidence the opposite could just as easily be the case, but there is nothing but a correlation.

No casual inferencefavourable to the employer is available from such evidence.” [17] With respect to the matter of evidence of “alcohol related incremental risk”, the arbitration board ended itsanalysis with the following reasoning: This evidence gives a push in the direction of the conclusion that the employer belongs at the lower end of the scale in terms of theexistence of incremental safety risk posed by alcohol use.

My conclusion on the evidence, overall, is it shows a very low incremental riskof safety concerns based on alcohol-related impaired performance of job tasks at the site. [18] Next the arbitration board turned to the question of the means chosen to detect the presence of alcohol. Theboard admitted that a distinction had to be drawn between drug and alcohol testing and the latter had been treated differently and morekindly once it was recognized that the testing was by way of breathalyser. The Supreme Court of Canada jurisprudence refers to the useof the breathalyser as minimally intrusive: e.g., R. v.

Stillman, (SCC), [1997] 1 S.C.R. 607, [1997] S.C.J. No. 34 (QL). [19] The final part of the arbitration board’s analysis, with respect to the validity of the random alcohol testingpolicy, is concerned with balancing the interests of employer and employee.

As the board stated: “What needs to be measured are thebenefits that will accrue to the employer through the application of the random alcohol testing policy against the harm that will be doneto the employee’s right to privacy.” In the words of the majority: “If the random alcohol testing policy is to be justified it must be inproportion to the employee’s right to privacy.” The board went on to conclude that it could see little or no advantage accruing to theemployer through the adoption of a policy of random alcohol testing as Irving had failed to establish “any significant degree ofincremental safety risk attributable to employee alcohol use” in the mill in the past.

Taken with the low testing rate of 10%, the boardconcluded that the policy would seldom if ever identify any employee with a blood alcohol concentration over .04% and therefore therewas no concrete advantage to the employer to be gained through the random alcohol testing policy. On the other hand, the board foundthat because the alcohol testing was random in nature (without articulable cause) the employee’s right to privacy was heightened. Theboard also found the intrusion not to be trifling and went on to conclude that the balance favoured the union/employee. .

[ 20 ] On the application for judicial review, the arbitration board’s decision was set aside and the grievance dismissed. The application judge interpreted the board decision as requiring a history of accidents in a dangerous workplace in order to justify the policy of random alcohol testing and that such a requirement was unreasonable because it effectively meant that the employer would have to wait until a catastrophe occurred before being able to take pro-active measures to prevent a recurrence.

The distinction the board drew between a dangerous workplace and ultra-dangerous one was found to be unreasonable by the application judge. He opined that once the board found the mill to be a dangerous workplace, the only question left for the board’s consideration was whether the employer’s policy was a proportionate response to the potential danger. Having regard to the minimally intrusive nature of the breathalyser and the fact the policy applies only to employees who hold safety sensitive positions, the application judge concluded the grievance should have been dismissed. III.

The Standard of Review [ 21 ] This Court is under no obligation to agree with the application judge’s decision to accept the parties’ mutual submission that reasonableness is the proper standard of review. Moreover, having raised the issue with counsel at the appeal hearing, we are under no obligation to decide whether the application judge properly applied that standard to the arbitration board’s decision. As stated at the outset, this appeal involves a question of law.

Are we to presume that it is the prerogative of individual labour arbitrators throughout the country to determine the analytical framework upon which to evaluate whether drug and alcohol testing policies are reasonable, even though some of the lead cases are the product of the judicial pen? Are we to assume that labour arbitrators dealing with alcohol and drug testing policies can lay claim to a relative expertise not possessed by the judiciary?

I answer both questions in the negative and raise a third: How does a reviewing court deal with the reality that the arbitral jurisprudence reveals what have been described as competing analytical frameworks or tests? In my view, there comes a point where the goal of certainty in the law must overshadow the precepts of the deference doctrine. This is one of those cases. [ 22 ] As a general proposition, this Court has accorded deference to decisions of the Labour and Employment Board, individual labour arbitrators and labour arbitration panels involving questions of law arising from the

interpretation of a collective agreement or the enabling legislation. Nothing that was decided in Dunsmuir , save for the notion of jurisdictional questions, detracts from the earlier jurisprudence of this Court. Thus, I am left with the task of justifying the decision to apply the correctness standard of review to the arbitration board’s decision. My reasoning is not complicated.

The central questions raised on this appeal require the decision maker to strike a proper balance between the right of an employer to adopt policies that promote safety in the workplace, and an employee’s right to privacy or to freedom from discrimination in those cases where the challenge is brought under human rights legislation.

When viewed through these prescriptive lenses, it is only natural to ask whether arbitrators possess a relative expertise that supports a finding that the Legislature intended that deference would be accorded to arbitration decisions involving drug and alcohol testing. [ 23 ] Certainly, the Supreme Court has yet to accord deference to an administrative tribunal with respect to questions of law umbilically tied to human rights issues: see Jones and de Villars, Principles of Administrative Law , 5 th ed. (Toronto: Carswell, 2009) at 553.

Similarly, the Supreme Court has held various privacy commissioners do not have greater expertise about the meaning of certain concepts found in their respective statutes which limit or define their authority: see Jones and De Villars at 553, note 223. Accepting that no analogy is perfect, I see no reason why this Court should depart from those precedents. Indeed, if one looks to the arbitral jurisprudence, one is struck by the reliance on judicial opinions touching on the matter.

The overlap reflects the general importance of the issues in the law and of the need to promote consistency and, hence, certainty, in the jurisprudence. Finally, I am struck by the fact that there comes a point where administrative decision makers are unable to reach a consensus on a particular point of law, but the parties seek a solution which promotes certainty in the law, freed from the tenets of the deference doctrine. In the present case, it is evident that the arbitral jurisprudence is not consistent when it comes to providing an answer to the central question raised on this appeal.

Hence, it falls on this Court to provide a definitive answer so far as New Brunswick is concerned. This is why I am prepared to apply the review standard of correctness. But this is not to suggest that I am about to ignore the arbitral jurisprudence which has evolved over the last two decades.

Let me explain. [ 24 ] In holding that correctness is the proper standard of review with respect to the legal question posed, I do not want to leave the impression the arbitral jurisprudence dealing with random alcohol testing suddenly becomes irrelevant and the review court should embark upon a fresh analysis, immune from the principles and analytical frameworks being applied by adjudicative tribunals. To the contrary, the arbitral jurisprudence of the last decade has gone a long way to defining what reasonableness means when assessing the enforceability of alcohol and drug testing policies in the workplace.

In particular, one has to recognize the significant contributions of arbitrators such as Michel G. Picher. He was the arbitrator and author in the seminal CN Rail decision referred to earlier and which has been consistently cited and applied by arbitrators throughout the country.

Arbitrator Picher was also the chair of the arbitration board and author of two decisions that wound their way through the Ontario Courts: the Nanticoke and Entrop decisions, also cited above. [ 25 ] As it should happen, Arbitrator Picher has also authored a decision dealing with a grievance filed by the union with respect to two saw-mills in New Brunswick, owned by another Irving company. The grievance involved the company’s random alcohol testing policy: J.D. Irving Ltd. v. Communications, Energy and Paperworkers’ Union, Local 104 and 1309 (Drug and Alcohol

Policy Grievance) , [2002] N.B.L.A.A. No. 7 (QL). More recently, the decision of Arbitrator Margo R. Newman in Navistar Canada Inc. v. National Automobile, Aerospace, Transportation and General Workers Union of Canada (CAW-Canada) Local 504 (Substance Abuse Grievance) , [2010] O.L.A.A. No. 227 (QL), offers a penetrating and exhaustive analysis of the jurisprudence relevant to the issues at hand.

I single out such decisions, not because they are owed deference as of right, but because deference is earned through the production of reasons which move beyond the Dunsmuir requirement of rationality and intelligibility into the realm of simple persuasiveness. [ 26 ] While I have adopted the correctness standard in regard to the question of law posed above, there is one aspect of the arbitration board’s decision for which the review standard of reasonableness does apply.

The board held that Irving had failed to establish a “sufficient case” that its kraft mill could be placed in the same ultra-dangerous category of risk such as a “nuclear plant, an airline, a railroad or a chemical plant”.

In my view, the finding that a kraft mill does not fall within the same dangerous category as a railroad or chemical plant is simply “unreasonable”. [ 27 ] Finally, if I am mistaken and deference is owed to the arbitration board’s ruling in regard to the need for Irving to adduce evidence of an existing alcohol or drug problem in the workplace in order to justify its decision to adopt the random testing policy, I take refuge in the application judge’s analysis and application of the “reasonableness” standard of review and his reasons for setting aside the board’s decision.

In particular, I would draw attention to the judge’s analysis pertaining to inferences which the board drew from the evidence and which are not discussed in these reasons because of the application of the correctness standard. IV. Issues – Analysis A. Introduction [ 28 ] The scope of the legal issues that arise in the context of an employer’s unilateral adoption of a drug and alcohol policy is indeed broad.

Not only does one have to draw a distinction between the testing for drugs and for alcohol, one has to acknowledge that testing can be required for reasonable cause (prior to an incident or accident) or post-incident cause (following a work related accident). There are questions as to who is to be tested and questions whether testing can be made a pre-condition to access to an employer’s work site. There are cases where the employer requires testing of a subcontractor’s employees which leads to the subcontractor adopting its own testing policy to ensure compliance with that of the main contractor.

There are also situations where the policy can be challenged on grounds of discrimination should the relevant legislation embrace drug and alcohol addiction as a disability, which in turn raises the question as to the scope of the employer’s duty to accommodate those with disabilities.

And there are many other issues. [ 29 ] I have purposely narrowed the scope of these reasons to the consideration of the only aspect of Irving’s drug and alcohol testing policy that was challenged before the arbitration board: the validity of the requirement for employees in safety sensitive positions to undergo random alcohol testing. As a matter of logic, one would think any legal reasoning applicable to random alcohol testing would apply equally to random drug testing.

However, I acknowledge that the jurisprudence dealing with drug testing has proven to be more problematic than cases dealing with random alcohol testing. While it is true that testing for both substances has a deterrent effect, drug testing cannot measure present impairment. A positive test simply means that the employee has taken drugs in the past. By contrast, alcohol testing is able to detect on the job impairment and minimize the risk of impaired performance.

As well, alcohol testing by breathalyser has always been regarded as minimally intrusive when it comes to an employee’s right to privacy and freedom from unreasonable searches. Given these differences, my emphasis is on decisions dealing with random alcohol testing and any obiter comments dealing with alcohol testing made in the context of the validity of a drug testing policy or an aspect thereof. B.

The Arguments and the Jurisprudence [ 30 ] As stated at the outset, the union’s position is encapsulated in two related propositions: (1) arbitrators in Canada have overwhelmingly rejected mandatory, random and unannounced drug and alcohol testing; and (2) sufficient evidence of a pre-existing drug or alcohol problem in the workplace is a pre-condition to the enforceability of such policies, unless the workplace qualifies as ultra-dangerous.

Accordingly, the union maintains the application judge erred in setting aside the arbitration board’s decision on the review standard of reasonableness. [ 31 ] The union’s assertion that arbitrators have overwhelmingly rejected mandatory and random drug and alcohol testing is based, in part, on the Nanticoke decision where Arbitrator M.G.

Picher concluded that: “As set out above, a key feature of the jurisprudence in the area of alcohol or drug testing in Canada 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” (para.101) (emphasis added). The union acknowledges that the quote refers only to random drug testing but goes on to insist that the quote applies equally to random alcohol testing. With great respect, I disagree and here is why. [ 32 ] In Nanticoke , the arbitration board (M.G. Picher, Chair) was dealing with a grievance with respect to the propriety of the employer’s (Imperial Oil) policy of random drug testing by the use of oral swabs for the detection of cannabis use, including actual impairment, at its oil refinery.

The policy applied to all employees and not just those who held safety sensitive positions. Ultimately, the majority of the board upheld the grievance and struck down this aspect of the policy. In so holding, the majority distinguished any analogy between testing by breathalyser and drug testing administered by buccal swabs. In a preliminary award, the board had held that the validity of that part of the policy dealing with random alcohol testing was not arbitrable because of the failure to initiate a challenge over the 13 years it was in force (acquiescence).

Applying the pre Dunsmuir review standard of patent unreasonableness, the application for judicial review was dismissed in the Ontario Divisional Court and, in turn, that decision was upheld by the Ontario Court of Appeal. [ 33 ] In brief, the Nanticoke decision does not support the union’s broad proposition that arbitrators in Canada have overwhelmingly rejected the policy of random alcohol testing by breathalyser. My conclusion is supported by another Ontario decision in which an employer’s alcohol and drug testing policy was challenged under the Ontario Human Rights Code , R.S.O. 1990, c. H.19 .

The Entrop decision involved, inter alia , an employer’s policy of random alcohol testing for employees, in safety sensitive positions, at two of its oil refineries. The policy was found to be prima facie discriminatory on the basis of actual or perceived handicap of alcohol abuse. Applying the standard of correctness to the board of inquiry’s decision, the Ontario Court of Appeal rejected the board’s finding and held that the testing policy could qualify as a bona fide occupational requirement provided the sanction for an employee testing positive was tailored to the employee’s circumstances.

As to whether the policy of random alcohol testing provisions were reasonably necessary, the Court held: Imperial Oil can legitimately take steps to deter and detect alcohol impairment among its employees in safety-sensitive jobs. Alcohol testing accomplishes this goal.

For employees in safety-sensitive jobs, where supervision is limited or non-existent, alcohol testing is a reasonable requirement. [para. 110] [ 34 ] Neither Nanticoke nor Entrop provide us with an analytical framework or test for determining the circumstances in which an employer is required to adduce evidence of an existing alcohol problem in the workplace with respect to those employees holding safety sensitive positions.

All that can be said with confidence is that random alcohol testing of employees holding safety sensitive positions within the oil refinement business has been upheld so long as the testing policy imposes sanctions which accommodate the individual circumstances of the employee who tests positive. [ 35 ] The only New Brunswick arbitral decision dealing with random alcohol testing in the workplace involves another Irving company with sawmills in Chipman and Sussex: J.D. Irving Ltd. v. Communications, Energy and Paperworkers’ Union, Local 104 and 1309 .

However, in that case there was no direct challenge to the justification for the decision to implement the policy. Rather the challenge was to various aspects of the policy. All of the issues, but one, are irrelevant to the case at hand. The parties disagreed as to the definition of what constituted a safety sensitive position for purposes of the policy.

The union contended that not only must the employer establish that the job, if performed while impaired, had the potential to cause a “catastrophic incident” affecting the health and safety of the employees and other persons, or the environment, but also that the job must be one with little or no direct supervision. [ 36 ] Writing for a unanimous arbitration board, Chair M.G. Picher held the possibility of regular and ongoing supervision of employees should not be viewed as determining whether their work is safety sensitive.

He went on to conclude that for purposes of drug and alcohol testing the identification of safety sensitive positions is more usefully achieved by asking what consequences are risked if the person performing a particular kind of work does so impaired by drugs or alcohol. Hence, those employees holding clerical positions would not qualify. Those performing tasks while impaired by drugs or alcohol in circumstances where they pose a safety risk to themselves or others or to property or equipment fall within the classification of safety sensitive positions. [ 37 ] I can only speculate as to why the union in J.D.

Irving Ltd. , which is the same union on this appeal, did not challenge the random alcohol testing policy on the same grounds as are being argued today. Perhaps there was evidence of a pre-existing alcohol problem at the two sawmills. Alternatively, the union may have considered the arbitral jurisprudence which has upheld the random alcohol testing in the logging industry. The lead case appears to be Weyerhaeuser I . The precedential significance of this case cannot be ignored. See also Weyerhaeuser Co. v. Communications, Energy and Paperworkers Union, Local 447 (Roberto Grievance) , [2006] A.G.A.A.

No. 48 (QL) ( Weyerhaeuser II ), dealing with a policy which included post-incident drug testing and which policy had been adopted as a result of a large number of workplace accidents.

[ 38 ] Borrowing liberally from the head note, the facts and findings in Weyerhaeuser I are as follows. The employer, a logging company, adopted a company-wide policy dealing with issues related to substance impairment in the workplace. The union filed a policy grievance, alleging that the policy violated the collective agreement as it was discriminatory and unreasonable. At the hearing, the employer sought a ruling on the union’s position that the employer was required to demonstrate that a substance abuse problem existed in the workplace in order to justify the introduction of the policy.

The employer took the position that it was entitled to implement the policy in the absence of a proven substance abuse problem because it was engaged in an inherently safety sensitive industry. Arbitrator C. Taylor held that the employer did not have to prove that a substance abuse problem existed before adopting the policy because the company operated in a safety sensitive industry. Logging operations were dangerous, they took place in harsh and isolated environments, and work related serious injuries and fatalities were relatively high as compared to other industries.

Therefore, the employer's operations constituted an industry that was inherently safety sensitive. This reality justified a high degree of caution on the part of the employer without the pre-condition of an extensively documented history of a substance abuse problem in the workplace. The employer's position was upheld. [ 39 ] Weyerhaeuser I is also important because it offers a review of the relevant arbitral jurisprudence up to the date of the decision (2004). Specifically, three arbitral decisions are fully canvassed: CN Rail , supra , Dupont Canada and Fording Coal Ltd. v.

United Steelworkers of America, Local 7884 , [2002] B.C.A.A.A. No. 9 (QL). The one decision invariably cited in cases involving drug and alcohol testing is CN Rail . However, that case did not involve the validity of a policy imposing random drug and alcohol testing but rather with the validity of a policy imposing mandatory testing in circumstances where there is reasonable cause to believe an employee is impaired (with cause testing).

The union had argued that the policy did not meet the reasonableness test because CN Rail had failed to establish a demonstrable need for the policy and failed to show that alternative measures were not available to combat the problem. Arbitrator Picher expressed no difficulty in concluding that the industry with which he was concerned was, by its very nature, safety sensitive, thereby obviating a need to demonstrate problems of substance abuse in the workplace.

He did not, however, disagree with earlier authorities holding that, for the purpose of meeting the KVP rules, there may be a burden on an employer to establish the need for a drug and alcohol testing policy, including whether alternative means are available to deal with the problem. However, Mr. Picher registered a caution with respect to those requirements when considering safety sensitive industries. The critical passage from the CN Rail decision reads as follows: One further theoretical concept needs to be addressed before turning to the specifics of CN's drug and alcohol policy on this matter.

As a number of the arbitral awards reflect, it is generally accepted that in analyzing the reasonableness of a drug and alcohol testing policy for the purposes of [ KVP ] standards, there may be a burden upon the employer to first demonstrate the need for such a policy, including an examination of whether alternative means for dealing with substance abuse in the workplace have been exhausted. While I do not disagree with those principles, I believe a note of caution should be registered, particularly with respect to that requirement.

It seems to the Arbitrator that there are certain industries which by their very nature are so highly safety sensitive as to justify a high degree of caution on the part of an employer without first requiring an extensive history of documented problems of substance abuse in the workplace.

Few would suggest that the operator of a nuclear generating plant must await a near meltdown, or that an airline must produce documentation of a sufficient number of inebriated pilots at the controls of wide-body aircraft, before taking firm and forceful steps to ensure a substance-free workplace, by a range of means that may include recourse to reasonable grounds drug and alcohol testing. The more highly risk sensitive an enterprise is, the more an employer can, in my view, justify a proactive, rather than a reactive, approach designed to prevent a problem before it manifests itself.

While more stringent thresholds may fairly be applied in non-safety sensitive work settings, as for example among clerical or bank employees, boards of arbitration should be cautious before requiring documented near disasters as a pre-condition to a vigilant and balanced policy of drug and alcohol detection in an enterprise whose normal operations pose substantial risks for the safety of employees and the public. [para. 195] [ 40 ] Based on the above passage, the arbitration board in the present case reasoned that an employer had to adduce sufficient evidence of a pre-existing alcohol problem in the workplace in order to justify the testing policy, unless the industry in question fell within the category of “ultra-dangerous”, as would a nuclear plant or an airline.

With great respect this

interpretation of the above passage is not in keeping with what Arbitrator Picher actually decided. He held that, having regard to the nature of the railroad industry, it was unnecessary to prove a substance abuse problem in the workplace.

Admittedly, Arbitrator Picher went on to say that he was "also satisfied" that the employer had adduced sufficient evidence to justify its substance abuse policy, including drug and alcohol testing, and that the employer had taken sufficient steps to exhaust other less intrusive alternatives to deal with the problem of substance abuse in the workplace. [ 41 ] Weyerhaeuser I also relied on Fording Coal (Arbitrator Hope) to substantiate the finding that proof of a substance abuse problem in the workplace is not necessary in cases where the employer’s operations could be classified as inherently dangerous.

In the latter case, there was a challenge to the employer’s policy of reasonable cause testing for drugs and alcohol. The employer operated an open pit mine. Arbitrator Hope concluded that employers were not required to establish the existence of an alcohol or drug problem in the workplace with respect to industries that are by their very nature safety sensitive so long as the policy applied only to those who hold safety sensitive positions. He found the mining operation qualified as inherently dangerous because of the use of explosives, flammable, caustic and corrosive materials and chemicals.

In reaching his conclusion, Arbitrator Hope relied heavily on the CN Rail decision of Arbitrator M.G. Picher. See also Continental Lime Ltd. and International Brotherhood of Boilermakers, Iron Ship Builders, Blacksmiths, Forgers and Helpers, Local Lodge No. D575 , (2002) 105 L.A.C. (4 th ) 263, where a drug and alcohol testing policy was upheld without evidence of an alcohol or drug problem in the workplace and the employer operated an open face quarry mine. [ 42 ] Weyerhaeuser I also relied on Dupont Canada (P.C. Picher, Chair). In that case the employer operated several

plants manufacturing nylon intermediates involving the use of chemicals. The arbitrator rejected the union’s submission that the employer’s drug and alcohol policy was unreasonable on the grounds the employer had failed to establish “clear and cogent evidence” of a substantive drug or alcohol problem in the workplace.

This was true even though the workplace had a superior safety record and had operated for the last 10 years without a single lost-time injury on the job; no person in the plant had been found in possession of drugs or alcohol and over several decades there had been no single incident or accident caused by alcohol or drug abuse. The arbitration board held that evidence of a pre-existing drug or alcohol problem was not necessary to justify the policy’s adoption as the plant was an “inherently dangerous” workplace. [ 43 ] Two other arbitration decisions warrant consideration. In United Assn.

Of Journeymen and Apprentices of the Plumbing and Pipefitting Industry of the United States and Canada, Local 488 v. Bantrel Constructors Co. , [2007] A.G.A.A. No. 33 (QL), the union challenged whether employees already working on-site at the employer’s petroleum manufacturing facility, prior to the introduction of drug and alcohol testing, had to submit to the new policy as a condition of gaining continued access.

In dismissing the grievance, the arbitration board confirmed it was not necessary to demonstrate that there was an actual drug and alcohol problem in the workplace before adopting policies to enhance safety in cases where the workplace came within the highly dangerous category. The board also remarked that the jurisprudence revealed the evolution of two competing approaches: one from the West and one from the East.

On the one hand, the Western Canadian cases had permitted, for the most part, broader drug and alcohol testing programs in workplaces which are “demonstrably safety sensitive”, alleviating employers of the burden of demonstrating existing drug and alcohol problems prior to testing. On the other hand, the board held the narrow approach could be found in the Ontario cases such as Entrop , discussed above, and International Union of Operating Engineers Local 793 v. Sarnia Cranes Ltd., [1999] O.L.R.D. No. 1282 (QL). The board decision in Bantrel was upheld on judicial review ( 2007 ABQB 721 , [2007] A.J.

No. 1330 (QL)) but quashed on appeal ( 2009 ABCA 84 , [2009] A.J. No. 216 (QL)). [ 44 ] While I might want to quibble with the

interpretation placed on Entrop in Bantrel Constructors Co. , there is no question that Sarnia Cranes reflects the 1990s jurisprudence holding that if “mandatory universal testing” is to be justified there should be evidence of a drug or alcohol problem in the workplace which cannot be combated in some less invasive way: see Provincial- American Truck Transporters and Teamsters union, Loc. 880, Re, [1991] O.L.A.A. No. 16 (QL), and C.H. Heist Ltd. And E.C.W.U., Loc. 848, Re , [1991] O.L.A.A. No. 48 (QL).

If it were necessary to distinguish Sarnia Cranes , one would simply note that the case dealt, in a peripheral manner, with random alcohol and drug testing, and also that much of the analysis was directed at the validity of the drug testing aspects of the policy. [ 45 ] The other arbitration decision which offers an extensive review of the arbitral jurisprudence is Navistar. In that case, the union challenged the employer's adoption of a Substance Abuse Prevention Policy (SAPP), which included drug and alcohol testing for reasonable cause, and post-incident situations, but not random testing.

The union also challenged the employer's determination that all warehouse associates, the entire bargaining unit, occupied safety sensitive positions which subjected them to the testing provisions. The parties agreed to sever the issues raised by the union's grievance and proceed only on the threshold issue of whether the employer's facility was a safety sensitive workplace where any drug and alcohol testing was appropriate. The workplace in issue was a warehouse and distribution facility where automotive parts were stored and shipped to various locations.

The employer argued that the facility was a safety sensitive environment due to the work performed by the warehouse associates operating heavy-duty industrial vehicles weighing 6,000 to 8,000 pounds. The arbitrator concluded the employer’s operation was not as inherently sensitive as those work environments where the policy was implemented without the need for evidence of a prior work-place drug or alcohol problem (railroads, mining and forestry), but that the operation in question was far enough along the continuum of “safety sensitive operations” that the employer was entitled to be proactive.

Resultantly, it need not prove the potential for catastrophe or the existence of a substance abuse problem in order to justify testing. Given the employees’ use of heavy machinery in the work-place, and given statutory safety obligations imposed on employers under the Occupational Health and Safety Act , S.N.B. 1983, c.

O-0.2, and s. 217.1 of the Criminal Code (which expose employers to criminal responsibility for failing to ensure employee safety by taking reasonable steps to prevent bodily harm), and in light of the contractual commitment contained in the collective agreement, the union failed to establish that the employer was acting in a manner inconsistent with the collective agreement, without justification, or unreasonably when it adopted its SAPP. [ 46 ] It should not be presumed that there are no cases where the employer has led evidence as to an existing alcohol or drug problem in the workplace with the object of justifying the adoption of its testing policy so as to by-pass the argument that the workplace qualifies as an inherently dangerous operation.

Communications, Energy and Paperworkers Union, Local 707 v. Suncor Energy Inc . (Alcohol and Drug Policy Grievance) , [2008] A.G.A.A. No. 55 (QL) is a case in which the employer is the owner of mining operations in the Athabasca oil sands, located near Fort McMurray. The oil sands mining operation recovers bitumen which is then upgraded by refinery on the Suncor site. Employees work on a 24-hour, seven days a week basis, 365 days a year at a remote site unable to access the amenities available in most communities.

The employer was able to establish easily an existing alcohol and drug problem based on experience and having regard to a provincial report dealing with substance abuse and gambling in the Alberta workplace. [ 47 ] Another decision in which evidence of an existing drug or alcohol problem in the workplace was used to support the employer’s decision to impose random alcohol testing is Greater Toronto Airports Authority v. Public Service Alliance of Canada, Local 0004 , supra .

Although the employer maintained that it had a right to impose such a policy based on the “inherently dangerous” nature of its operation, the arbitrator noted that in a number of cases other arbitrators had held that such a finding was not of itself sufficient to outweigh the privacy interests of individual employees and to support a regime of random testing. As a matter of fact,

the arbitrator accepted the employer’s evidence of an existing problem in the workplace and upheld the random alcohol testing policywithout offering an opinion as to whether a finding of “inherent workplace dangerousness” would displace the need to adduce evidenceof an existing alcohol problem in the workplace. [48] I could find relatively few cases in which an arbitrator upheld a grievance with respect to the employer’sdecision to implement random alcohol testing in the workplace because of a failure to establish evidence of an alcohol or drug problem inthe workplace.

The cases of which I am aware involved the trucking industry. But the facts of each case must be examined carefully. Forexample, Provincial-American Truck Transporters and Teamsters Union, Loc. 880, RE, involved truck drivers who transported othertrucks, piggy-back style, and the imposition of a policy of mandatory (but not random) drug testing by urine analysis. The policy wasfound to be unreasonable because the employer had failed to establish evidence of an existing alcohol or drug problem in the workplace. A similar result was reached in Trimac Transportation Services-Bulk Systems v.

Transportation Communications Union, (CA LA), [1999] 88 L.A.C. (4th) 237 (Arbitrator K.M. Burkett). In that case, the arbitrator was asked to decide whether theemployer’s policy requiring all truck drivers to submit to random mandatory drug testing violated the collective agreement. The policywas adopted in response to a US federal law requiring drivers be tested. The union’s grievance was upheld. At the time the policy hadbeen adopted, there had had been no instances of impairment among drivers.

Therefore, objectively speaking, there was no drug oralcohol usage problem to give rise to the reasonably held apprehension that drivers might compromise safety by reporting to work underthe influence. [49] The one recent decision which supports the union’s position in this case is Petro-Canada Lubricants Centre (Mississauga) v. Communications Energy and Paperworkers Union of Canada, Local 593, [2009] 186 L.A.C. (4th) 424 (ArbitratorKaplan). The case involved mandatory random alcohol testing.

The Arbitrator determined that the policy was unreasonable because ofthe employer’s failure to adduce evidence of an existing drug or alcohol in the workplace. In reaching that conclusion the arbitrator reliedon and quoted from the decision of M. G. Picher in Nanticoke. Regrettably the arbitrator misquoted Arbitrator Picher. Mr. Picher did notsay that “arbitrators had overwhelmingly rejected mandatory, random, and unannounced testing of all employees in a safety sensitiveworkplace…” What Mr.

Picher said was that arbitrators had overwhelmingly rejected mandatory, random and unannounced drug testingof all employees ...” (emphasis added). And as noted, Nanticoke dealt with random drug testing. All of this said, the question of whethertruck drivers are involved in an inherently dangerous operation remains alive and not affected by the innocent misstatement as to whatwas actually said in Nanticoke.

I suspect the answer depends in part on what is being transported and the deterrent effect of theprovisions of the Criminal Code dealing with impaired driving. [50] Finally, I must refer to the arbitral decision in Communications, Energy Paperworkers Union Local 777 (May27, 2000, T.J.

Christian, Chair, unreported, on reserve, New Brunswick Law Society Library, Fredericton), where the majority upheldthe employer’s decision to adopt a policy of random alcohol testing at its Strathcona refinery based on the dangerous nature of theoperations and a survey of affected employees with respect to alcohol and drug use by employees holding safety sensitive positions at therefinery. This case would suggest that some evidence of a drug or alcohol problem in the workplace is required to support the policy’sadoption.

Several arbitrators who have cited and relied on the Strathcona case have failed to appreciate that some evidence of an existingdrug or alcohol problem was factored into the decision to uphold the testing policy. C.

Summary Observations on the Extant Jurisprudence [51] The above analysis undermines the union’s position that arbitrators in Canada have overwhelmingly rejectedmandatory, random and unannounced drug and alcohol testing and that sufficient evidence of a pre-existing drug or alcohol problem inthe workplace is therefore a pre-condition to the enforceability of such policies, unless the workplace qualifies as ultra-dangerous.Having regard only to the arbitral jurisprudence discussed above, it is safe to conclude that, on balance, arbitrators have rejected the needto adduce such evidence in cases where the employer is able to establish that the workplace is inherently dangerous.

It is true that theearly jurisprudence reveals an antipathy towards drug and alcohol testing in the workplace and, in particular, to drug testing: e.g., EssoPetroleum Canada v. Communications, Energy & Paperworkers’ Union, Local 614, (BC LA), [1994] B.C.A.A.A. No. 244, 56 L.A.C. (4th) 440 (J.D. McAlpine, Chair). Random alcohol testing, however, gained early acceptance once testing wasrestricted to employees holding safety sensitive positions and the testing would be by breathalyser.

This left for consideration the pivotalquestion whether the workplace in question fell within the “highly” or “inherently” dangerous category. Employers involved in theproduction and refining of oil products or chemicals, or in the mining and forestry sectors of the economy, have been able to persuadearbitrators and arbitration panels that such operations so qualify and usually without adducing evidence of an existing alcohol problem inthe workplace.

By contrast, there has been a resistance to classifying trucking operations as inherently dangerous. [52] As matter of policy, this Court must decide whether an employer is under an obligation to demonstratesufficient evidence of an alcohol problem in the workplace before adopting a policy requiring mandatory random alcohol testing. In myview, the balancing of interests approach which has developed in the arbitral jurisprudence and which is being applied in the context ofmandatory random alcohol testing warrants approbation.

Evidence of an existing alcohol problem in the workplace is unnecessary oncethe employer’s work environment is classified as inherently dangerous. Not only is the object and effect of such a testing policy to

protect the safety interests of those workers whose performance may be impaired by alcohol, but also the safety interests of their co- workers and the greater public. Potential damage to the employer’s property and that of the public and the environment adds yet a further dimension to the problem and the justification for random testing. As is evident, the true question is whether the employer’s workplace falls within the category of inherently dangerous. It is to that issue I now turn. D.

Is Irving’s kraft mill an inherently dangerous operation? [ 53 ] It has been argued that the arbitration board made a palpable and overriding error in concluding that the kraft mill did not fall within the classification of ultra-dangerous operation. The reality is that in law there is no such classification. Hence, the question we must address is whether the mill operations can be classified as inherently dangerous.

In my view, the arbitration board’s finding that the kraft paper mill presented itself as a “dangerous work environment” satisfies the test of inherently dangerous and, therefore, Irving did not have to adduce evidence of an existing alcohol problem in the workplace, let alone sufficient evidence of a “significant problem” in the workplace. [ 54 ] For greater certainty, I want to focus briefly on the reality that chemical plants and railway operations have been classified as inherently dangerous work environments, thereby dispensing with the need to adduce evidence of an existing alcohol problem in the workplace in order to justify the adoption of a random alcohol testing policy.

Yet, as noted earlier, the arbitration board held that Irving’s kraft mill could not be equated with a railway operation or a chemical plant. In my respectful view this finding is unreasonable. Let me explain. [ 55 ] To the extent that a railway company, which transports goods throughout the country, is entitled to adopt a random alcohol testing policy without evidence of an existing alcohol problem in the workplace, one would think that a kraft paper mill would provide an equally dangerous workplace.

If a railway company which transports hazardous materials to various workplaces is entitled to adopt a policy with respect to random alcohol testing then it should follow that a company which uses those materials in its operations is equally entitled to do so. Indeed, the Irving kraft mill uses hazardous materials such as chlorine dioxide, sulphuric acid, hydrogen peroxide, liquid and gaseous oxygen and methanol. Moreover, Irving’s employees are responsible for loading and unloading chemicals from rail cars and other types of vehicles.

In short, it makes no sense that a railway operation is entitled to adopt a policy of random alcohol testing, but the customer who uses the toxic chemicals in its manufacturing process is not. When the facts are so viewed, the evidence of Irving’s witness, Mr.

Moorehouse, to the effect that a kraft mill is as close to a chemical plant as one can get makes eminent good sense and, as we know, refineries and chemical plants are treated in the arbitral jurisprudence as inherently dangerous work sites. [ 56 ] The facts of the present case also reveal a kraft mill with a $350 million pressure boiler which has a “high potential” for explosion. The potential impact on the environment of a major catastrophe, such as a chemical spill, has never been challenged. The intra-city location of the kraft mill and its proximity to the St.

John River and Bay of Fundy would cause concern for any environmentalist. Indeed, at one point the board accepts that incorrect configuration of plant control systems by certain employees is noted as having a potential for “catastrophic failures”. As stated at the outset, some might argue that at its core this appeal is of importance to the public at large. [ 57 ] In

summary, it is not difficult to support the contention that Irving’s kraft paper mill qualifies as an inherently dangerous workplace as would a chemical plant. This is why evidence of an existing alcohol problem in the workplace was not required to support its policy of random alcohol testing. This is why the arbitration board’s decision cannot stand and the application judge was correct in determining that its decision should be set aside and the grievance dismissed.

In the circumstances, it is unnecessary to delve into issues dealing with the board’s perception that evidence of near disasters is required to justify the imposition of alcohol or drug testing policies. V. Disposition [ 58 ] The appeal should be dismissed. As the hearing was spread over two days, the respondent is entitled to costs of $5,000. ____________________________________

J.T. ROBERTSON, J.A. WE CONCUR: ___________________________________ J. ERNEST DRAPEAU, CHIEF JUSTICE OF NEW BRUNSWICK ____________________________________ WALLACE S. TURNBULL, J.A.