In many corporate group structures, the entity that owns a trade mark is not necessarily the same entity that uses it day to day. A holding company may register the marks, while one or more operating subsidiaries make use of them in the marketplace. Often, these arrangements proceed without a formal written licence agreement, on the assumption that an implied licence exists between the related entities.

That assumption carries real risk. This post explores how implied trade mark licences operate in South Africa, and why relying solely on implication can put the validity of your registrations in jeopardy.

The Trade Marks Act 194 of 1993 (the “Act”) does not require a trade mark licence to be documented in writing. Licence agreements are governed by ordinary common law principles and may therefore be express (whether written or oral) or tacit, and may even be inferred from conduct – that is, implied. The essentials are simply identifiable parties, identifiable subject matter, and a defined scope of permission granted by the licensor to the licensee. Notably, consideration is not required – which is typical in intragroup arrangements.

From a trade mark perspective, the Act goes further. Section 38(1) provides that where a registered trade mark is used by a person other than the proprietor with the licence of the proprietor, that use constitutes “permitted use”.[1] Section 38(2) then deems permitted use to be use by the proprietor itself.[2] In other words, the Act does not prescribe how permission must be manifested; what matters is that the proprietor has in fact authorised the use. An implied licence is therefore not a distinct species of licence recognised in the Act. Rather, it is permission inferred from conduct, the relationship between the entities, or the surrounding commercial arrangements, rather than expressed in a written agreement.

In the intragroup context, the typical scenario is straightforward. The holding company or a dedicated intellectual property entity owns the registered marks, while the trading subsidiaries use those marks in the marketplace. No formal licensing agreement is concluded. The parties proceed on the understanding that the proprietor consents to the subsidiaries’ use. That consent, if it can be established, constitutes a licence for the purposes of the Act, even though it was never reduced to writing. When use by a non-proprietor counts as the proprietor’s use

The deeming provision in section 38(2) is particularly important in the context of non-use challenges. Under section 27(1)(b), a registered trade mark may be removed from the register if, during a continuous period of five years, there has been no bona fide use of the mark by the proprietor or by any person “permitted to use” it under section 38.[3] The practical consequence is that if a subsidiary’s use of a mark cannot be attributed to the proprietor as permitted use, even extensive commercial exploitation of the mark in the marketplace will not prevent its removal. The proprietor may simply be unable to defend the registration against a non-use challenge.

The rationale for recognising implied licences

A trade mark’s core function is to serve as a badge of origin – it tells the consumer that the goods or services bearing the mark come from, or are associated with, a particular source.[4] Trade mark law’s recognition of implied licences serves that function within corporate groups by aligning legal doctrine with commercial reality. Where a proprietor authorises a related entity to use its mark, even without recording that authorisation in writing, the consumer’s expectation of a single commercial origin is maintained. The use by the subsidiary is, legally speaking, the proprietor’s use.

Without the possibility of implying permission, routine intragroup trade mark use could be treated as unauthorised. That would leave the mark vulnerable to removal on grounds of non-use, even though it is being actively exploited in the marketplace.

Quality control can support the inference. Although the Supreme Court of Appeal held in AM Moolla Group Ltd v The Gap Inc[5] (“AM Moolla”) that quality control is not a statutory prerequisite for permitted use,[6] the court acknowledged that a proprietor would ordinarily seek consistency in how its mark is used.[7] Evidence of quality control mechanisms within a group – brand guidelines, approval processes, and centralised marketing oversight – can bolster the case that the proprietor’s authorisation was present, even if it was never formalised. Quality control is not legally necessary for permitted use, but it is commercially prudent and evidentially valuable.

Recognising that implied licences are possible is one thing. Proving that one exists is quite another. The evidentiary burden is the central vulnerability. The proprietor must show that actual permission was granted; the mere existence of a group structure, coupled with a bare assertion that use was “with the licence” of the proprietor, will not on its own support the inference. AM Moolla Group Ltd v The Gap Inc

The SCA’s decision in AM Moolla illustrates this point. The Moolla Group comprised several affiliated companies. The group’s trade marks were treated as “group property” and used by whichever entity suited the occasion.[8] When the registrations were challenged for non-use, the proprietor of the marks – a dormant company that had never itself traded – needed to show that the other group entities’ use qualified as permitted use under section 38.

It failed to do so. The court found that the evidence of any licence was “vague to such an extent that it smacks of evasiveness”.[9] In a 200-page affidavit, the issue of licensing was addressed in a single phrase – that the use by group members had been “with the licence” of the proprietor. The court held that this was no more than a bare allegation, unsupported by factual evidence. There was no written agreement to produce, no particularity of an oral agreement, and no facts alleged to give rise to a tacit one.[10] The registrations were expunged for non-use. The case demonstrates that treating trade marks as “group property” to be used freely by various entities is legally perilous. Without evidence of specific permission from the registered proprietor to each user, even substantial marketplace presence cannot save registrations from removal on grounds of non-use.

Wings Travel Management v Satguru Travels

More recently, in Wings Travel Management (Pty) Ltd v Satguru Travels (Pty) Ltd t/a Travelwings[11], the High Court confronted a similar evidentiary gap. The applicant was a member of the Wings Group and the registered proprietor of various WINGS trade marks. However, the entity that actually traded under those marks was a different group company, Wingsnaledi Travel Management (Pty) Ltd. In considering the applicant’s passing off claim, the applicant relied heavily on the reputation of “the Wings Group” and its 30-year history, global footprint, and substantial turnover. The respondent challenged whether the applicant itself, as distinct from other group entities, held the reputation on which it sought to rely.

The court found that the applicant had failed to establish that it owned the goodwill or reputation associated with the marks. Critically, when challenged on this point, the applicant did not rely on or refer to any licence or permitted use arrangement in its affidavits.[12] The court held this failure was “fatal to its case”.

Conclusion

Implied trade mark licences are recognised under South African law and serve a practical commercial purpose within corporate groups. However, they become a liability where evidence of permitted use is lacking. As the cases discussed above demonstrate, a group structure and bare assertions on their own do not establish that permission exists. The prudent course is straightforward – document your licensing arrangements in writing. A simple intragroup trade mark licence, recording the proprietor’s authorisation and the terms on which it is granted, can safeguard the validity of your registrations without disrupting existing group operations.

Where the arrangement justifies it, the proprietor may also wish to record the licensee as a registered user under section 38(3) of the Act. A registered user entry provides a public record of the licensing relationship and affords the licensee standing to institute infringement proceedings in its own name – a further layer of protection that costs little to establish.