Two sick-leave letters, three weeks apart, that did not match

The situation

The letter arrived on a Tuesday, addressed to Craig by name, from an employee who had worked the floor of the family manufacturing business for six years. It laid out two dates, three weeks apart, on which two different employees had each missed four consecutive days for what both described as flu. One employee had been paid for the time under an informal understanding with a shift supervisor. The other had been told the days would be unpaid and had been quietly warned about attendance, with no explanation for the difference beyond, as the letter put it, 'that's just how it was decided.'

The company, with revenue in the $20 to $60 million range, was owned through a family trust that Craig managed on behalf of several relatives, including his cousin Alejandro, who held a beneficial interest but no day-to-day role. There was no written sick-leave policy. Supervisors had always used judgment, and for years that had not caused a visible problem, largely because the workforce was small enough that everyone knew everyone's situation informally.

Before the letter escalated further, Mateo, another family member who owned a logistics company and had weathered a similar complaint at his own business years earlier, offered to help. He drafted a two-paragraph memo for Craig to send to staff, asserting that sick days were 'at management's discretion' and reminding employees that attendance issues could lead to discipline. It was meant to settle the matter quickly. Instead, when the complaining employee's coworkers saw it, several read it as confirming that the company felt free to treat people differently with no standard at all, and two more employees raised the same concern in the following week. What had been one letter was now a pattern, and Craig, worried about both morale and exposure under the Employment Standards Act, called our office before sending anything further.

The stakes were not abstract for Craig. As trustee, he answered to the other family beneficiaries for how the business was run, and a formal Ministry of Labour complaint, even one ultimately resolved in the company's favour, would have meant an investigation touching payroll records, supervisor practices, and management decisions going back years, exactly the kind of scrutiny a family-held business prefers to avoid regardless of outcome. The letter writer had not yet filed anything formal, but the tone made clear that option was on the table if the response felt dismissive.

The legal question

The core question was not whether the company had done something unlawful in any single case; unpaid time off for illness beyond statutory minimums is generally left to the employer's discretion in Ontario unless a policy or contract says otherwise. The question was whether inconsistent, undocumented discretion, applied differently to two employees in near-identical circumstances, created legal risk even without an obvious discriminatory motive.

It did, in a narrower sense than the original complaint feared but a real one. Ontario's Employment Standards Act gives employees a small number of job-protected sick days each year, which the company had generally honoured, but those days are unpaid; the Act requires no paid sick days at all. Anything paid was purely the company's own policy or informal practice, and the Act does not dictate how that discretion is exercised. The exposure was less about the statute and more about the absence of any consistent internal rule: without a written policy, every decision looked discretionary and personal, which is exactly the pattern that invites a complaint, whether framed as unfairness, as a human rights concern if any protected ground is even loosely implicated, or simply as a morale problem that becomes a retention problem.

Mateo's memo made this worse rather than better, because 'management's discretion' is not a policy; it is an acknowledgment that no policy exists, delivered in writing, which is precisely the kind of document that turns an informal complaint into evidence. It also failed to distinguish between statutory job-protected leave, which the company could not discipline anyone for taking, and any additional paid time the company chose to extend informally, which meant the memo risked being read as threatening discipline for protected leave itself.

We also had to consider the trust structure. Alejandro, as a beneficiary with no operational role, had no standing to direct company policy, but his informal comments to Craig about 'not making this a bigger deal than it is' had already shaped how quickly the first complaint was addressed, and that dynamic needed to be separated from the actual legal analysis before it influenced the fix as well.

There was a further wrinkle in how the original inconsistency had arisen. The supervisor who approved paid leave for the first employee had done so partly because that employee's manager happened to be away and the request went to a more senior supervisor who was, by disposition, more generous with judgment calls. The second employee's request went to a supervisor known for stricter attendance enforcement. Neither supervisor had done anything the company had told them not to do; both were exercising exactly the discretion the company had always allowed, which is what made the fix a policy problem rather than a personnel problem aimed at either supervisor individually.

What we did

1. Reviewed every sick-leave decision from the past two years. We asked for payroll and attendance records covering all employees, not just the two named in the complaint, to see whether the inconsistency was isolated or systemic. It turned out to be systemic: outcomes varied by which supervisor handled the request, with no correlation to tenure, role, or reason given, which told us the fix had to be a policy change rather than a conversation with one supervisor.
2. Withdrew Mateo's memo before it went any further. We explained to Craig why 'management's discretion' as a written statement was worse than no statement at all, since it converted an informal practice into a documented policy of unpredictability, and recommended it never be referenced again in any employee communication. We also drafted a short, neutral note Craig could send if anyone asked about it directly, so withdrawing it did not read as an admission of wrongdoing.
3. Separated statutory leave from discretionary paid time in plain terms. We drafted language distinguishing job-protected illness leave, which employees are entitled to without needing employer approval and which the company cannot discipline anyone for taking, from any paid sick days the company chose to offer beyond that as a benefit. Keeping those two categories visibly separate mattered because Mateo's memo had blurred them, and blurring them again in the new policy would have recreated the same risk.
4. Built a written illness policy with defined, equal terms. The new policy set a fixed number of paid sick days per year, a consistent notice procedure, and a clear statement that supervisors could not vary the terms case by case, removing the discretion that had caused the complaint in the first place. We also specified how unused days carried over, since an undefined carryover rule is exactly the kind of gap that reintroduces inconsistency a year later.
5. Trained the supervisors who had been applying the old informal rules. We ran a short session with the three shift supervisors explaining the new policy and, importantly, why consistency mattered legally and not just as a courtesy, since supervisors had genuinely believed they were being reasonable under the old system. We also gave them a one-page reference sheet so a decision could be checked against the written rule in the moment rather than judgment calls resuming once the training faded from memory.
6. Responded to the original complaint directly and specifically. Rather than a general reassurance, we helped Craig draft a response acknowledging the inconsistency the employee had identified, explaining the new policy, and confirming the employee's own paid time would be corrected retroactively to match the new standard. The letter named the specific dates in question rather than speaking in generalities, since a vague apology would have read as no different from the discretion that caused the complaint.
7. Documented the trust's separation from day-to-day HR decisions. We recorded, in the company's own governance file, that operational employment decisions sat with Craig as manager and not with beneficiaries like Alejandro, so future informal pressure from family members would not blur that line again. This mattered beyond the immediate complaint, since the same ambiguity about who actually decides operational questions could resurface around hiring, discipline, or compensation the next time a beneficiary had an opinion.
8. Set a review date for the new policy rather than leaving it static. We built in a twelve-month check-in to confirm the fixed sick-day allowance and notice procedure were actually working in practice and had not simply pushed the old inconsistency into a different form, since a written policy that nobody revisits can drift out of sync with how the business actually operates. The review also gives Craig a scheduled reason to ask supervisors directly whether the rule is holding, rather than waiting for the next complaint to find out.

The outcome

The employee who wrote the original letter accepted the retroactive correction and the new policy without escalating to a formal complaint with the Ministry of Labour, though the company had prepared for that possibility throughout. The two employees who raised concerns after Mateo's memo were satisfied once the actual policy, rather than the memo, was circulated.

The cost was mostly time and the retroactive pay correction, which came to a modest amount once the affected employees' days were recalculated under the new consistent standard, well within what the company had budgeted once the review began. No claim was ever filed, and no regulator became involved.

The more lasting change was cultural as much as legal. Supervisors now apply one set of rules, employees can see what those rules are before a situation arises rather than finding out after the fact, and Craig has a written record showing the company corrected the problem once it was identified rather than defending the informal system that caused it. Mateo's early advice, offered in good faith from his own experience running a different kind of business, turned out to be the wrong instinct for this one; what worked was replacing discretion with a policy specific enough that nobody had to guess.

For Alejandro and the other beneficiaries, the resolution mattered less as a legal outcome than as a demonstration that Craig could handle an operational problem without either family pressure or outside advice steering the result. That distinction, between running the business and holding an interest in it, is easy to blur in a family trust structure and harder to restore once it slips; documenting it here gave the trust a clear precedent for the next time a beneficiary's opinion and the manager's decision might otherwise be treated as the same thing.

What you can learn from this