A 444-paragraph labour arbitration decision from British Columbia in Teck Coal Ltd. (Fording River and Elkview Operations) v United Steelworkers, Locals 7884 And 9346, 2018 CarswellBC 119, is the latest in a growing body of case law dealing with the question of whether an employer can unilaterally implement random drug and alcohol testing in a unionized workplace. The arbitrator found that random drug and alcohol testing was not a reasonable exercise of management rights in this case, as there was no corresponding "general" problem of drug or alcohol abuse in the workplace grave enough to justify the serious intrusion on employee privacy rights.

Similar to previous random drug and alcohol testing cases, the analysis centred on the competing issues of workplace safety and employee privacy. The current arbitral jurisprudence on this issue was summarized by Madam Justice Abella in the Irving case: "a unilaterally imposed policy of mandatory, random and unannounced testing for all employees in a dangerous workplace has been overwhelmingly rejected by arbitrators as an unjustified affront to the dignity and privacy of employees unless there is reasonable cause, such as a general problem of substance abuse in the workplace."1

However, the arbitrator in Teck Coal rejected the union’s argument that the Irving test was binding on him. Rather, the arbitrator noted that the Court in Irving "was not ‘laying down’ the law on random drug and alcohol testing for future application by arbitrators and courts. Instead, it was reviewing the decision of an arbitration board in a particular case to determine whether that decision was within ‘the range of reasonable outcomes’ in light of the facts before it and the applicable arbitral jurisprudence. It was exercising the Court’s supervisory jurisdiction on judicial review, not an original jurisdiction."2

A New Test?

Ultimately, the arbitrator applied a variation on the Irving test involving a three-part "balancing process supported by a proportionality analysis" which he analogized to the Oakes test for justifying limits placed on Charter rights and freedoms.3 First, the arbitrator determined whether the employees’ privacy was infringed by the random testing policy. If so, the second step required the employer to establish “cause,” or a legitimate need for random testing. The employer does this by establishing a general problem of drug and alcohol abuse in the workplace. The need for testing is then balanced with the infringement on the employees’ privacy. The greater the infringement, the heavier the burden on the employer. Only if the problem is adequate to justify the random testing does the arbitrator move to the third part of the test, which is a proportionality analysis. The arbitrator must decide whether random testing would be an effective means to address the employer’s concerns, and whether there is a less intrusive response to the safety concerns that could be implemented instead.4

The arbitrator found that Teck Coal’s random testing policy intruded on the employees’ privacy "in a very invasive manner."5

Despite hearing evidence of a "work hard, play hard" culture that involved consumption of alcohol and drugs on days off,6 and evidence from a number of Teck Coal employees that the random testing policy improved this culture,7 the arbitrator ultimately found that the motivation to implement random drug and alcohol testing at Teck Coal’s worksite was not to combat a specific problem with drug and alcohol abuse, but to address "an element of risk that Teck Coal foresaw could impinge on safety." The arbitrator held that this did not rise to the level of a demonstrable workplace problem. The arbitrator seemed to consider that because the culture of consumption of drugs and alcohol was "not all that different from many other resource town communities," it was therefore not a "demonstrable workplace problem."8 As a result, the random drug and alcohol testing policy was not justified.

Despite having disposed of the issue at the "balancing" stage of the arbitrator’s analysis, the arbitrator also addressed the "proportionality" stage of his analysis. The arbitrator found that even if Teck Coal had passed the balancing stage, random testing would not be a proportionate measure because Teck Coal’s accident rates were already on the decline as a result of other safety efforts. “The case for random testing in this proceeding fails to a large extent because of the strong commitment of Teck Coal… to making their coal operations in the Elk Valley as safe as they can be… [R]andom testing is an exceptional remedy because of its impact on privacy rights, and, accordingly, it is reserved for compelling circumstances. Such circumstances do not exist in this case because of that strong commitment and the success of their efforts.”9

Takeaways for Employers

This decision suggests that from a practical standpoint, it will be very difficult — if not impossible — for employers to justify random drug and alcohol testing policies. According to this decision, it may not be sufficient to show a general workplace problem with drug and alcohol use unless the problem is more severe than other similar workplaces or communities. Even if the employer was able to clear this initial "balancing" hurdle, it could then be argued at the "proportionality" part of the test that there are less intrusive ways to deal with the general problem of drug and alcohol abuse at the workplace. Once those measures have been implemented, the employer may no longer be able to demonstrate a general problem with drug and alcohol abuse.

While the Suncor decision10 will have a much larger impact on the law relating to random drug and alcohol testing than this decision, in developing its own test rather than following the Irving test, Teck Coal highlights the uncertainty in this area of law. Further, in the non-unionized context, it could be argued that the courts’ judicial review decisions in the Irving and the Suncor cases do not constitute binding legal tests. The result is that the circumstances under which random drug and alcohol testing can be justified could remain ambiguous for some time.