A class action suit without a class is not access to justice.
HOWARD LEVITT
4 Jul 2026
CA lawyer who persists after the supposed class has rejected the proceeding may now face something more immediate: personal responsibility for the employer’s costs. — Howard Levitt
lass actions are intended to provide access to justice. But they become considerably less noble when lawyers pursue litigation in the name of employees who reject the lawsuit, its objectives and the counsel claiming to speak for them.
A decision in an Ontario employment case released two weeks ago stands as a prescient warning to the class-action bar. It establishes three propositions that plaintiff counsel can no longer safely ignore.
First, a proposed class action may be fundamentally defective when all the employees ostensibly represented do not share the interests, objectives or desires of the representative plaintiff.
Second, certification is not permanent. When virtually the entire class opts out, the proceeding will be decertified.
Third, and most ominously for plaintiff counsel, lawyers who persist after the class has effectively disappeared will be held personally responsible for the employer’s costs.
Each of these risks is known in theory. What this case demonstrates is how quickly they can converge in practice — and how little tolerance courts have when they do.
The litigation in Navaratnarajah v. FSB Group Ltd. began as an employment-status class action. The plaintiff alleged that insurance sales agents, described as producers, had been improperly classified as independent contractors rather than employees and were therefore owed the accordant statutory entitlements.
On its face, the case resembled a conventional worker-protection lawsuit: individuals labelled contractors allegedly deprived of the minimum protections afforded to employees.
But that framing concealed a fatal problem. Many of those whom the plaintiff claimed to represent did not consider themselves victims. They did not want to bring the lawsuit and did not wish to be represented by those advancing it.
They valued their independent contractor status. It provided tax advantages, commercial autonomy and ownership interests in their books of business. A judicial declaration that they were employees might not enrich them. Worse, it could impair the very arrangements under which they had prospered.
That disconnect proved fatal — not simply as a matter of optics, but as a matter of legal viability. A class proceeding cannot function where its supposed beneficiaries reject both its premise and prosecution.
Justice Edward M. Morgan of the Ontario Superior Court recognized that danger when he initially certified the action. He warned that the financial interests of class members might ultimately be hijacked rather than advanced.
His warning proved prophetic.
Out of 69 identified potential class members, 66 opted out. Of the three nominally remaining, one had died, and another could not be located. That left a single representative plaintiff and his counsel pursuing a class action on behalf of a class that had ceased, in any meaningful sense, to exist.
The employees had delivered their verdict. They voted with their feet.
Yet the plaintiff and counsel strongly resisted decertification and continued to actively pursue the case.
The court was unimpressed. It concluded that continuing the proceeding could permit the representative plaintiff to negotiate a purportedly classwide settlement from which only he and his counsel might meaningfully benefit.
That is the precise opposite of access to justice. It is litigation overriding the autonomy of the people it professes to protect.
The action was decertified. That ruling alone made this case significant.
Certification is not a judicial franchise granted to class counsel for the life of a lawsuit. It remains justified only while the proceeding continues to advance the recognized purposes of class actions: access to justice, judicial economy and meaningful behaviour modification.
When almost every class member opts out, those foundations collapse.
There is no access to justice for people who have rejected the proceeding. There is no judicial economy in conducting class litigation for a lone or a few claimants. And behaviour modification cannot justify commandeering the interests of individuals who consider the alleged misconduct beneficial overall rather than harmful.
But these lessons should begin before certification, not after the opt-out period exposes the fiction.
Suppose that substantial numbers of employees appear at a certification hearing through their own lawyers and tell the court that they reject the proposed action, do not accept the representative plaintiff as speaking on their behalf and intend to opt out as soon as certification is granted.
That evidence should not be brushed aside merely because the formal opt-out period has not yet begun.
Proposed class members do not possess an absolute veto over certification. Nor should certification motions deteriorate into popularity contests between competing law firms. A generalized dislike of proposed class counsel may not decide the motion.
But informed opposition to the litigation is plainly relevant.
Employees may reject the legal characterization being imposed upon them. They may regard the requested remedy as economically harmful. They may believe that the representative plaintiff ‘s interests differ from their own. They may distrust counsel’s strategy or judgment. They may prefer individual litigation, another proposed proceeding or different legal representation.
Or, significantly, they may simply not wish to litigate.
A class proceeding is not an asset belonging to the law firm that conceived it.
Those considerations go directly to whether the representative plaintiff can fairly and adequately represent the class and whether a class proceeding is the preferable course of action.
It would be perverse to certify first and ask whether a real class exists later.
Certification is not an academic exercise. It immediately generates publicity, discovery obligations, management distraction, legal expense and immense settlement pressure. Even a questionable claim can become commercially dangerous once certified because the cost and reputational consequences of defending it may exceed the value of the claim itself.
Courts should not unleash those consequences merely to conduct an expensive experiment confirming what the evidence already demonstrates: the proposed class refuses to be represented.
Finding one willing plaintiff should not entitle counsel to conscript unwilling employees into litigation they consider contrary to their interests.
The second chapter of the FSB litigation is even more consequential.
Following decertification, Justice Morgan awarded the defendants $100,000 in costs. Nearly three years later, that award remained unpaid and interest had accrued. The defendants then sought to have class counsel, Monkhouse Law, added as a party responsible for paying it. Justice Morgan agreed.
He was critical of class counsel for continuing to advance the proceeding and adding further costs despite the lack of class support for the lawsuit.
Compounding that concern was the court’s expressed understanding that the representative plaintiff and class were indemnified against adverse costs by their counsel, Monkhouse Law — standard practice in the legal profession. When Justice Morgan learned that such indemnification did not appear to exist, he was concerned with counsel pressing forward in the absence of a supportive class. In response, the court exercised its supervisory authority and added Monkhouse to the costs order.
Every plaintiff-side classaction firm in Canada should take notice.
Class proceedings create a profound financial asymmetry.
Defendants may spend millions responding to claims that never reach trial. Certification alone can impose reputational damage, extensive costs, years of management distraction and overwhelming commercial pressure to settle.
Plaintiff counsel generally act on contingency. They invest time and carry opportunity cost, but may view their ultimate financial downside as limited. If a case succeeds, the fees can be enormous. If it fails, the loss may be relatively limited. Meanwhile, if the representative plaintiff is impecunious, the successful defendant will have little realistic prospect of recovering its costs.
That imbalance can encourage proceedings that would never be commenced if plaintiff counsel were exposed to the same commercial discipline and risks as the defendant.
The Navaratnarajah v. FSB decision corrects that.
It warns lawyers that they cannot simply locate one dissatisfied worker, build a sweeping class around that individual and assume that every meaningful financial risk belongs to someone else.
A contingency fee is not a one-way option. Counsel should not receive the upside from creating and controlling the litigation while externalizing the downside to an individual plaintiff without the means to satisfy a costs award.
Before launching an employment class action, counsel must investigate more than whether an arguable legal theory can be pleaded.
They must determine whether the employees represented actually share the representative plaintiff ‘s interests and want the relief being pursued.
Do the workers genuinely wish to challenge the compensation model under which they have operated?
Could the proposed remedy expose them to tax, commercial or ownership consequences?
Would reclassification impair their independence or jeopardize their interests in their books of business?
Are they prepared to surrender control over individual claims and negotiating strategies to lawyers they didn’t select themselves?
Is there a genuine classwide grievance — or merely one dissatisfied claimant and counsel in search of a class?
And once employees have already opposed certification, whose interests are advanced by class counsel continuing?
These are not public relations questions. They go to the legitimacy of the proceeding.
There will, of course, be legitimate cases in which workers fail to appreciate their rights, fear retaliation or resist litigation for reasons unrelated to the merits. Opposition cannot automatically defeat certification.
But when informed employees overwhelmingly reject the proceeding, the burden should be on proposed class counsel to explain why certification remains preferable — and whose interests the litigation is actually serving.
A class proceeding is not an asset belonging to the law firm that conceived it. They are permitted to act only because the procedure is intended to advance the interests of the class and the administration of justice.
Once that justification disappears, counsel must reassess the case. A lawyer who persists after the supposed class has rejected the proceeding may now face something more immediate: personal responsibility for the employer’s costs.
That is not an assault on access to justice.
It is a necessary reminder that litigation brought in the name of unwilling employees can itself be an abuse of the procedure created to protect them.
A class action without a class is not access to justice — it is litigation in search of a client.