Some judgments are wrong and quietly forgotten. Others are wrong and remembered. AmGeneral Insurance Bhd v Iskandar bin Mohd Nuli [2016] 1 MLJ 818 belongs to a rarer and more troublesome class: a decision the Federal Court declined to endorse, yet which lingers in the lower courts like a guest who has not noticed that the party is over.

The Court of Appeal set out to answer a narrow question about insurers and the third-party claims they defend. It came away having invented something rather grander — a doctrine of “commonality” that, with the lightest of touches, rearranged the balance Parliament had struck between insurers and the victims of motor accidents. This article is about that doctrine, and why it deserves a dignified retirement.

I. How the Trouble Began

A loan, a Causeway, and an awkward border

The facts are almost domestic in their simplicity. Shahrul owned a car, insured in Malaysia. He lent it to his friend, Iskandar Nuli, who drove across the Causeway into Singapore with his wife, Zuraini, in the passenger seat. There was a collision with a crane truck. Zuraini was injured, and she sued — her husband and the crane driver both — in the Singapore courts.

Then came the wrinkle that turns a road accident into a law report. Zuraini looked to Shahrul’s Malaysian insurer to meet any judgment. But the two legal systems do not see eye to eye. Singapore obliges a motor policy to cover passengers; Malaysia does not, unless the policy says so. Shahrul’s did not.

The insurer’s own conduct is where the interest lies. Notified of the claim, it instructed Singapore solicitors and entered an appearance for Iskandar Nuli. For a season it ran the defence. Only afterwards did it write to deny liability under the policy, offering the insured a choice: appoint your own lawyers, or let ours carry on with all rights reserved. Having taken the wheel and driven some distance, the insurer now wished to disown the journey.

The orphaned doctrine

The insurer applied for a declaration that it was not liable. The High Court refused. The Court of Appeal reversed — and reached for the commonality principle to do it. Insurer and insured, the court reasoned, shared a common interest in defending the claim. Where interests are common, no conflict arises. And where no conflict arises, the insurer’s conduct in the defence could not estop it from later denying cover.

The Federal Court affirmed the outcome, but the manner of its affirmation is the detail that matters. It said not a word in favour of commonality. It rested the insurer’s victory on a single, unremarkable fact: the policy contained an exclusion clause that permitted repudiation.¹[2017] MLJU 1015 So the commonality principle was left an orphan — never adopted by the apex court, never formally buried, and at liberty to wander through the judgments of those who came after. It is that orphan, and only that orphan, with which this piece is concerned: the Court of Appeal’s reasoning on commonality and estoppel.

II. The Doctrine, and Why It Misfires at the Outset

What the court actually said

The operative passage is short. The “assumption of proceedings by the insurers”, the court held, “did not necessarily debar them from subsequently denying liability”.²At page 832, paragraph [42] Beneath that sentence sits a far larger assumption — that insurer and insured pursue identical objectives once a claim is defended, so that any conflict of interest simply evaporates.

Put plainly, the theory runs thus. Where insurer and insured share a common purpose, there is nothing untoward in the insurer defending the claim in the insured’s name. Here both AmGeneral and Iskandar Nuli were said to want the same two things: to avoid a default judgment, and to lay the blame at the crane driver’s door. That alignment, it was argued, entitled the insurer to mount an early defence while quietly reserving the right to deny cover later on.

It is an elegant theory. It is also, on inspection, very nearly the reverse of the truth.

Whose pocket is really being defended?

The candour is to be found elsewhere in the same body of law, where judges have been refreshingly honest about who is really being protected. When an insurer defends in its insured’s name, it is, in plain terms, looking after its own pocket. The negligent driver is the nominal defendant; the party with the money at stake — and the party in whose interest the case is truly fought — is the insurer. To dress that arrangement up as a fellowship of equals, a meeting of minds between insurer and insured, is to mistake a commercial necessity for a friendship. The insurer’s first loyalty is to its shareholders, not to the man whose name appears on the writ. Its interest is to keep its exposure small; the insured’s interest is to be defended without stint. These are not the same interest. They are not even close.

Once that single observation is admitted, the rest of the edifice begins to wobble. The doctrine has no home in the statute. It rests on a premise about shared objectives that is demonstrably false. It hands insurers the conduct of litigation while letting them keep the option of walking away — an asymmetry that leaves the insured carrying the risk without the protection. And it breeds a quiet moral hazard, because an insurer that can escape its obligations whatever the quality of the defence has little reason to defend with vigour. Each of these objections deserves to be taken in turn.

III. The Case Against the Commonality Principle

A principle with no address in the Act

Begin with the statute, since that is where the court ought to have begun. Section 96(3) of the Road Transport Act 1987 says, in essence, that where judgment has been obtained against a person in respect of a liability covered by a policy, the insurer may apply for a declaration that it is not liable to indemnify. That is the whole of it. There is no requirement that the facts of the original action be “common” with the facts of the insurer’s application. None.

The Court of Appeal — through the formulation of Vernon Ong JCA — read such a requirement into the section anyway. That is not interpretation; it is amendment by another name. Parliament imposed no condition of factual commonality, and a court is not at liberty to supply one because it would tidy the law for insurers.

The supporting reasoning fares no better. Defending an action, the court suggested, merely staves off a default judgment and so creates no estoppel. But that is to misunderstand estoppel by conduct at its root. It allows an insurer to slice its behaviour into separate compartments — vigorous when it suits, disclaiming when it does not — and to deny that anything it did in one proceeding can bind it in another. The doctrine of estoppel exists precisely to prevent that sort of compartmentalisation.

Estoppel was never so timid

For estoppel, as the Federal Court understood it long before Iskandar Nuli, is anything but rigid. In Boustead Trading it was described as “a flexible principle by which justice is done according to the circumstances” — a “doctrine of wide utility” that has been “resorted to in varying fact patterns to achieve justice”.³ Boustead Trading (1985) Sdn Bhd v Arab-Malaysian Merchant Bank Bhd [1995] 3 MLJ 331 Gopal Sri Ram JCA put the test with admirable economy: all a litigant need show is that “he was so influenced by the encouragement or representation that it would be unconscionable for the representor to enforce his strict legal rights”.

Boustead went further still and played down the need for strict detriment; it is enough that it would be unjust to let the representor stand on the letter of his rights. Even where some Commonwealth courts still treat detriment as relevant — interpreted broadly, as it must be — a third party would have little difficulty supplying it. An injured claimant who relied on the insurer’s conduct, declined to pursue other avenues, and ran up the costs of litigation has plainly altered her position on the strength of the insurer’s “taking steps”.

Against this supple body of doctrine, the commonality principle insists on something estoppel has never required: that the proceedings be factually alike. Boustead asks no such thing. The inquiry is into the representor’s conduct and the representee’s reliance — not into the resemblance between one set of court papers and another.

The Hong Kong authorities say the opposite

The Court of Appeal sought support in two Hong Kong decisions, Chan Lam Chun v National Insurance Co Ltd⁴ [1977] HKLR 417 and Oriental Fire & General Insurance Co Ltd v Cheuk Ma Yee.⁵ [1980] HKC 32 It is an unfortunate choice of allies, for both authorities point the other way.

In Chan Lam Chun the insurer entered an appearance and a defence on the insured’s behalf, then tried to withdraw. The Hong Kong Court of Appeal held that by entering an appearance the insurer had represented that it would defend rather than repudiate, and was estopped accordingly. The threshold for culpable conduct was set strikingly low — the mere entry of an appearance, a far cry from the extended involvement on display in Iskandar Nuli.

Cheuk Ma Yee is, if anything, the stronger case for the third party. There the insured was uncooperative and had vanished altogether, yet the court still found that the insurer’s conduct in defending had raised an estoppel. Once an insurer takes steps to defend, it held, there comes a “point of no return” at which repudiation turns unconscionable.

To read these decisions as authority for a commonality requirement is to read them backwards. What they establish is that an insurer’s litigation conduct — regardless of whether the insured cooperated, and regardless of the factual shape of any later proceeding — can be enough to fix it with an estoppel. It is true that an insurer may say, in the abstract, that it affirms the policy yet repudiates this particular claim. But if it wishes to keep that position open, it must reserve its rights promptly and in clear terms. That is the corollary the Iskandar Nuli insurer ignored, and it cuts in two directions. Litigation conduct may be positive — entering an appearance, filing a defence, instructing experts, negotiating settlement, taking part in discovery. It may equally be negative — failing to reserve rights when it had every opportunity to do so. A reservation arriving only after the insurer has run the defence for months is a reservation that comes too late.

The purpose of Parliament was protection

All of this should have been read in the light of why the Act exists. The object of the Road Transport Act, as the Federal Court recognised in AmGeneral Insurance Bhd v Sa’Amran a/l Atan & Ors,⁶ [2022]  5 MLJ 809 is “to protect third parties against risks arising out of the use of motor vehicles”. The whole architecture of section 96 is built around that aim. Subsection (1) gives the third party a direct right against the insurer; subsection (2) stops the insurer from avoiding liability on the strength of a policy breach; and subsection (3) is the narrow, conditioned exit that the rest of the scheme permits.

The commonality principle works against that grain. By demanding factual commonality, it hands insurers a technical escape route between proceedings — even where their own conduct has induced reliance and raised a legitimate expectation of cover. A statute designed to shield the injured is thereby turned, quietly, into an instrument for the insurer’s convenience.

The decision that forgot the Bureau

There is a further omission, and a serious one. The Court of Appeal said nothing about the Motor Insurers’ Bureau framework. Under the MIB arrangements, authorised insurers undertake to meet liabilities even in circumstances that would otherwise let them off — cover bought after the accident, say, or non-disclosure by the insured. The Act and the MIB clauses are not two schemes but one, both directed at protecting the road-using public, and they must be construed together.

A decision on an insurer’s liability that proceeds as though the Bureau framework did not exist is, with respect, decided per incuriam. Leave out half of the protective scheme and the answer will come out wrong, however careful the reasoning on the half that remains.

A naked usurpation

The conventional principles of statutory construction make the difficulty plainer still. A court must give effect to Parliament’s intention as expressed in the words it chose — the lesson of Pepper v Hart [1993] AC 593. Where Parliament has laid down a comprehensive scheme, the court’s task is to apply it, not to graft on conditions that Parliament left out.

And the words of section 96(3) are clear. Where Parliament meant to attach a condition, it knew perfectly well how to say so — witness the proviso requiring notice to the judgment creditor before the insurer may disclaim. It attached no condition of factual commonality. To invent one is, in Viscount Simonds’ memorable phrase, “a naked usurpation of the legislative function under the thin disguise of interpretation”.⁷ Magor and St Mellons RDC v Newport Corporation [1952] AC 189

The bill is sent to the injured

Doctrine aside, consider who pays. The commonality principle lets an insurer escape by pointing to factual differences between proceedings — rewarding it for inconsistency and penalising the very victim who relied on the appearance of cover. Worse, it saddles claimants with proving “factual commonality”, a burden Parliament never imposed and never intended. The result is more cost, more delay, and more uncertainty for people who have already been hurt on the road. The Federal Court in Sa’Amran saw the problem clearly when it held that fairness to the third-party victim demanded that the evidence relied on in the section 96(3) application be tested in the pending trial. Rigid commonality offers the opposite of fairness.

No friends abroad

Nor does the doctrine find comfort in the wider common law. The English courts have steadily refused to confine estoppel by reference to artificial similarities between proceedings; in Amalgamated Investment and Property Co Ltd v Texas Commerce International Bank Ltd⁸ [1982] 1 QB 84 the Court of Appeal grounded the doctrine in conduct and reliance, not in the resemblance of one suit to another. The Australian courts have taken the same view: in Waltons Stores (Interstate) Ltd v Maher⁹ (1988) 164 CLR 387 the High Court applied estoppel flexibly to achieve justice, free of constraints that would defeat its purpose. Malaysia’s commonality detour is a lonely one, and it pulls against the harmonisation of commercial principle that ought to be the ambition of a Commonwealth jurisdiction.

Shield, sword, and the point of no return

Two final doctrinal points seal the matter. First, Boustead confirms that estoppel may assist a plaintiff in making good a cause of action, by stopping a defendant from denying a fact that would otherwise sink it. The old saw that estoppel is “a shield and not a sword” does not lock the doctrine away for defendants alone; a plaintiff may wield it too. The commonality principle’s insistence on rigid preconditions cannot be reconciled with so flexible a tool — least of all in insurance law, where estoppel exists precisely to stop an insurer from profiting by its own representations while shirking its bargain.

Second, Soole v Royal Insurance Co¹⁰ [1971] 2 Lloyd’s Rep 332 makes the same point in the insurer’s own setting: once an insurer takes steps to defend, a moment arrives beyond which repudiation becomes unconscionable. The Iskandar Nuli court tried to slip past Soole by way of the commonality requirement. It could not, because the requirement misdescribes the very thing it claims to govern — estoppel by conduct.

IV. The Way Forward

Jettison the doctrine

The case for abandoning the commonality principle is, by now, not hard to state. It has no foundation in the statute and amounts to legislation from the bench. It contradicts the settled flexibility of estoppel. It subverts Parliament’s plain intention to protect motor accident victims. And it works practical injustice on the people the Act was written to safeguard.

Return to the right question

Sa’Amran shows the way back. By insisting on fairness to the third party and reading section 96(3) in the light of the Act’s protective purpose, the Federal Court has already sketched a more principled course. Future courts should take it. The question has never been whether two proceedings look alike. It is whether an insurer’s conduct has induced reliance such that denying liability would be unconscionable — and that question can be answered whatever the factual differences between one suit and the next. Restore that test, and coherence returns to motor insurance law along with it.

V. Conclusion

The commonality principle is a jurisprudential aberration. It defies the words of the statute, the intention behind them, and the established law of estoppel. Its artificial demand for factual commonality defeats the very purpose of the doctrine it purports to apply, all while tilting the field towards insurers and away from the injured.

Its twilight is overdue. As Tennyson wrote, “the old order changeth, yielding place to new”.¹¹From Tennyson’s “The Passing of Arthur”: in the opening stanza King Arthur tells Bedivere, in a low voice, that change is the law of nature — one thing does not remain the same at all times; old systems change and new ones take their place, and a single good system, if it endures too long, may yet spoil the whole world. See https://www.doubtnut.com/pcmb-questions/196382 Born of decent intentions but flawed in the making, this doctrine deserves a respectful retirement to the quieter chambers of the law reports. Legal evolution asks no less: that a well-meaning but mistaken precedent should yield its place gracefully, so that clearer principle may light the way.

Only by returning to estoppel properly understood — conduct, reliance, and unconscionability, rather than a contrived test of “factual commonality” — can the courts ensure that justice is done according to law, and not according to judicial invention. Iskandar Nuli stands for much of what has gone awry in modern reasoning: the building of artificial barriers that defeat a statute’s purpose, the misreading of borrowed authorities, and the elevation of technical distinction over substantive justice.

Its quiet abrogation would restore coherence to the law of motor insurance, and with it the protection of the third-party victim — which is what Parliament intended all along.

∞§∞

This article is written for a general readership and does not constitute technical or legal advice. Readers with legal questions are encouraged to seek independent legal advice.

The author thanks KN Geetha, TP Vaani, JN Lheela, and Lydia Jaynthi at GK Legal. Our gratitude to Sasun Bughdaryan of Unsplash for the image.

Drafting assistance was provided by Claude, an AI model built by Anthropic. The truth is something better and stranger: an old advocate and a piece of software sat down together on a Friday afternoon and made one essay better than either could have made alone. That is not pastiche. That is a third thing, with a name we have not yet invented. The argument, the views, and any errors remain those of the author.

@Copyright reserved.

All content on this site, including but not limited to text, compilation, graphics, documents, and layouts, is the intellectual property of GK Ganesan Kasinathan and is protected by local and international copyright laws. Any use shall be invalid unless written permission is obtained by writing to gk@gkganesan.com.