Printable version

Attendance management program found wanting by arbitrator

A recent decision of an Ontario arbitration board illustrates the pitfalls that may arise in the designing of an attendance management program. As we pointed in earlier issues of FOCUS (see "Managing innocent absenteeism in the unionized workplace" and "Emerging issues in attendance management" on our Publications page), an attendance management program must display the following characteristics in order to withstand scrutiny at arbitration:

• there must be no conflict between the provisions of the program and those of the collective agreement;
• the program must be administrative and remedial in nature, not disciplinary;
• the program must be reasonable in its design, and must be applied in a reasonable manner; and
• the program must be consistent with human rights law.

In City of Windsor v. C.U.P.E., Local 543 (March 11, 2002), the employer's program failed in all key respects, with the result that it was declared null and void by a majority of a board of arbitration. This ruling was made despite the fact that no employee had been disciplined or terminated under the program during the more than ten years that it had been in effect.

THE PROGRAM

The triggering mechanism for the program's application to an employee was "excessive" absenteeism, defined to mean four occurrences of absence in any 12-month period. An occurrence meant absence due to an illness of one or more consecutive days.

Each month, managers were to receive reports on employees absent on more than four occasions during the previous 12-month period. Those employees were to attend an interview during which attendance goals for the following three months were to be set, after which a second interview was to be held. If, at that time, the attendance objectives had not been met, a third interview was to be scheduled after another three-month period.

At the third interview, if no improvement had been achieved, the employee was to be referred for a medical assessment by a physician of the employee's choice. Following the assessment, further objectives were to be set, and disciplinary consequences were indicated to be among the possible outcomes.

Pointing to evidence that the policy was applied inconsistently both between and within departments, the union argued that the policy was unreasonable in a number of respects:

• it did not distinguish between culpable and non-culpable absences;
• it treated several short absences more severely than one long one;
• making employees submit to a medical exam under the program was an invasion of privacy;
• the policy dealt with innocent absenteeism in a mechanical way, rather than on an individualized basis; and
• the policy was in conflict with the long-term disability provision of the collective agreement.

ARBITRATOR: POLICY INCONSISTENT, UNREASONABLE

A majority of the board of arbitration agreed with the union on virtually all counts. After noting that the policy could be nullified on a retrospective basis solely due to the inconsistency with which it had been applied, the board turned to the components of the policy itself and delivered this scathing description:

"The policy does not distinguish between culpable absences and non-culpable absences … . [It] is structured in a way that the Corporation simply does not concern itself with such matters but just counts the number of occurrences of absence. Nor does it concern itself with the total number of days of absence. Neither does it concern itself with the circumstances of the occurrences, and thus an employee who was ill and off work on the Monday, Wednesday and Friday of a week with two unsuccessful attempts to return to work on the Tuesday and Thursday would be counted as having three occurrences of absence while another employee who took the whole week off would record one occurrence of absence. It must be concluded that this is not a reasonable method by which to measure an employee's absenteeism or construct a record of absenteeism."

The policy as constituted could not serve as the basis for terminating an employee for just cause, as it did not measure the length of absences, distinguish culpable from non-culpable absences, or record the reasons for absences. Nor did the terms of the policy serve to determine whether the employer had grounds to terminate an employee for innocent absenteeism. Rather, the employer would have to establish both that the employee's past record of absenteeism was excessive, and that prognosis for future attendance was poor.

Finally, the board found fault with the provision for referring employees to a medical exam at the third interview. The board stated that the decision to refer an employee for a medical assessment had to be approached on an individual basis, and could not be automatically triggered by the policy's unreasonable measure of absenteeism.

In Our View

As we have noted in previous articles on attendance management programs, mechanistic, inflexible programs are at great risk of being struck down. In this regard, it is important not to confuse inconsistent application of a policy, as was the case here, with a discretionary approach that assesses individual circumstances. That was a critical feature in another case, Purolator Courier Ltd. v. Teamsters, Local 31 (September 29, 2000) in which an attendance management program was upheld despite the fact that a number of employees had been terminated under its provisions.

In Purolator, the threshold level for absenteeism was derived individually for each depot though a formula that counted both the number of occurrences of absenteeism as well as the number of days absent. The program was restricted to instances of innocent absenteeism, and absences of longer than six months were excluded. Considerable discretion was exercised in the application of the program, one example being the exclusion of an employee's absence due to an operation, as this was viewed as a one-time event. Features such as these allowed the arbitrator to distinguish the mechanistic structure of the program from its individualized application.

The Purolator case also demonstrates, however, that even if the program as whole is sustained, separate components may be struck down. In that case, the arbitrator held that the compulsory medical release required of the employee at the sixth-step interview was in conflict with the procedure for medical releases set out in the collective agreement. Moreover, the arbitrator found fault with the fact that the program made no mention of the employer's duty to accommodate employees before termination for innocent absenteeism.

For further information, please contact André Champagne at (613) 563-7660, Extension 229, or J.D. Sharp at (613) 563-7660, Extension 233.