Enterprise Agreements, Extrinsic Context, and Knowledge: CEPU v Endeavour Energy Network Management Pty Ltd
Rupert Williamson
8.9.2026
Enterprise agreements are paradoxical instruments. They function like agreements between employers and employees, but they are not a contract. They have effect under statute and therefore have a legislative character, but they are not delegated legislation: James Cook University v Ridd (2020) 278 FCR 566, 613 [222] (Rangiah J) (Ridd). Enterprise agreements occupy a twilight zone between public law and private law. The implications of this are not always clear. The uncertainty is illustrated by the Federal Court of Australia’s recent decision, Communications, Electrical, Electronic, Energy, Information, Postal, Plumbing and Allied Services Union of Australia v Endeavour Energy Network Management Pty Ltd [2025] FCA 1202 (Endeavour).
In Endeavour, extrinsic context played an important role in construing two enterprise agreements. This post considers the approach to extrinsic context taken by Endeavour and compares it with other authority on enterprise agreements. This reveals two contrasting approaches. One focuses on whether employers and employees were aware of the extrinsic context, akin to the approach in contract. The other focuses on whether the Fair Work Commission (Commission) was aware of the extrinsic context when it approved the enterprise agreement, being more similar to a public law instrument like legislation. That tension is not always appreciated, and seemingly was not the focus of argument in Endeavour. This post aims to highlight that tension: until it is resolved by the courts, it introduces avoidable uncertainty into the meaning of these important legal instruments.
The decision
It is not necessary to summarise every aspect of Endeavour. In short, it concerned the scope of overtime entitlements under two enterprise agreements from 2017 and 2021 (the two EAs), which were made under the Fair Work Act 2009 (Cth) (Act). This turned on the proper construction of the two EAs. The applicant (Union) contended for a broader construction; the respondent (Employer) contended for a narrower construction. The Court preferred the Employer’s construction.
Extrinsic context was influential in this outcome. The Court considered that the text favoured the Employer’s construction. But this could only be said ‘with a modest degree of confidence’: at [33]. The text remained ‘unclear’, and the Employer’s construction did not explain other parts of the text which remained ‘something of a mystery’: at [61] and [33]. Beyond the text, the Employer relied on two ‘elements’ of extrinsic context: at [35]. The first (relating to the underlying industry awards) did not prove especially persuasive: at [40]. The second, concerning the Employer’s longstanding approach to administering overtime payments, proved to be more compelling and is the focus of this post. But first, it is necessary to elaborate some principles of interpretation.
Makers
Legal instruments are commonly construed using an objective approach. That approach applies to constitutions, statutes, contracts, trusts, and enterprise agreements, among other instruments. In essence, the objective approach to interpretation focuses on what a reasonable person would understand to have been meant by the legal instrument. That approach gives primacy to the text: see, eg, Byrnes v Kendle (2011) 243 CLR 253, 282–90 [95]–[115] (Heydon and Crennan JJ).
But the objective approach also recognises the place of relevant, admissible extrinsic context. Lord Reid explained in Black-Clawson International Ltd v Papierwerke Waldhof-Aschaffenburg AG [1975] AC 591, 613–14:
The general rule in construing any document is that one should put oneself ‘in the shoes’ of the maker or makers and take into account relevant facts known to them when the document was made.
Although there are nuances to the role of knowledge in interpretation, for present purposes, Lord Reid’s dictum is a good starting point. It focuses attention on whether the extrinsic context was known by the instrument’s ‘maker’.
It is important to identify the ‘maker’. The court may ignore extrinsic context if the maker was unaware of that extrinsic context. This can be illustrated by comparing what are perhaps the quintessential examples of private law and public law instruments: contracts and statutes. For contracts, it may be permissible to consider extrinsic context that is known to all the parties at the time of execution. If one party was unaware of certain extrinsic context, that may render the extrinsic context irrelevant: Pacific Carriers v BNP Paribas (2004) 218 CLR 451, 461–2 [22] (the Court); Construction, Forestry, Maritime, Mining and Energy Union v Personnel Contracting Pty Ltd (2022) 275 CLR 165, 236–7 [187] (Gordon J). As for statutes, extrinsic context includes matters within Parliament’s contemplation: Singh v Commonwealth (2004) 222 CLR 322, 332 [12] (Gleeson CJ). Similar to contracts, if extrinsic context never came to Parliament’s attention, the court may refuse to have regard to it: see, eg, Commissioner of Taxation v Murray (1990) 21 FCR 436, 449 (Hill J); Linfox Australia Pty Ltd v Commissioner of Taxation (2019) 271 FCR 365, 388 [117] (the Court).
The identity of the ‘maker’ depends on the nature of the instrument. For contracts, there are multiple makers. Contracts are private law instruments. They are bilateral or multilateral and their legal effect derives from the ability of autonomous persons to make legally binding promises. Accordingly, the focus is on the knowledge of all the parties. The position is different for statutes. A statute only has one maker. Statutes are public law instruments; they are unilateral and derive their legal effect from Parliament’s sovereign authority, hence the importance of Parliament’s knowledge.
It is not always easy to identify the maker of an instrument. This issue has particular significance to public law instruments like legislation. In one sense, many individuals participate in the process of making legislation. Most prominently, there will usually be elected representatives and a viceregal representative. However, the meaning of a statute does not depend on the subjective intentions of those individuals. Nor does it depend on an aggregate of the individuals’ intentions. In other words, these individuals do not make legislation; it is Parliament that makes legislation: see Ravbar v Commonwealth (2025) 423 ALR 241, 277–8 [120] (Gordon J), 291 [172] (Edelman J), 320–1 [270]–[271] (Steward J), 330 [308] (Gleeson J), 347 [370] (Jagot J), 374 [467] (Beech-Jones J).
So far, we have seen that it is important to identify the maker of an instrument, which will depend on the nature of the instrument; the relevance of extrinsic context can then be tested by reference to the knowledge of that maker. So, who makes enterprise agreements?
Employers and employees
Endeavour treats employers and employees as the ‘makers’ of enterprise agreements.
In Endeavour, the Employer invoked the history of overtime payments to support its preferred construction of the two EAs. Various instruments had governed the Employer and its employees over the previous decades. In respect of overtime, the terms of the previous instruments were substantially identical to the two EAs. There was evidence that the earlier instruments had been applied by the Employer in a way that followed its preferred construction. The Employer’s approach had ‘not been called into question’ over the years: at [41]–[45].
In accepting the relevance of this history, the Court reasoned that the statutory process for creating an enterprise agreement ‘involves people drafting, discussing and adopting a text’. Therefore, extrinsic context could be relevant ‘at least to the extent that it is known or reasonably knowable by the persons to whom the [Act] affords a role in the process’: at [58].
The Court concluded that this history demonstrated a ‘shared understanding’ about what the two EAs meant ‘among those who brought into being the text submitted to and approved by the Commission’: at [60]. That shared understanding could not displace the text, but it did assist in construing the two EAs where the text was ‘unclear’: at [61].
This analysis focuses on the shared understanding and knowledge of the those who drafted, negotiated, and adopted the two EAs prior to submitting them to the Commission for approval. That is, it treats the Employer and its employees as the makers of the enterprise agreement and therefore the appropriate locus of knowledge.
The Commission
However, there is a good argument that, contrary to Endeavour, enterprise agreements are made, at least in part, by the Commission.
Enterprise agreements have legal effect by virtue of the Act. It stipulates a detailed process that must be followed before these instruments become effective. Representatives of employers and employees must meet and negotiate, and the proposal must be voted for by a majority of employees.
However, the enterprise agreement only gains legal effect after a critical final stage involving the Commission. Along with relevant supporting information, the enterprise agreement must be submitted to the Commission. The Commission must consider whether the agreement meets various statutory criteria, including whether employees are better off overall compared to the applicable award and whether the agreement includes any unlawful or discriminatory terms: ss 186, 193–5. Once the Commission is satisfied of these considerations, it must approve the agreement. It is only following the Commission’s approval that the enterprise agreement becomes effective: s 54. The Commission may also, in certain circumstances, approve the agreement subject to amendments specified by the Commission: s 191A. Such amendments do not require the parties’ consent and can be substantial: Aldi Foods Pty Ltd as General Partner of Aldi Stores (A Limited Partnership) v Shop, Distributive and Allied Employees Association [2026] FWCFB 134, [89], [108] (Gibian VP, Wright DP and Commissioner Matheson). The Commission also has amendment powers over previously approved agreements: ss 218A, 227D. In discharging all these functions, the Commission must take into account the objects of the Act, which include promoting ‘national economic prosperity and social inclusion for all Australians’ as particularised in detail by the Act: ss 3, 578. The Commission must also consider ‘equity, good conscience and the merits of the matter’ and the importance of diversity in the workplace: s 578.
In summary, the Commission’s role is substantive, demanding an assessment of the provisions against the supporting information it receives and with scope to depart from the employer’s and employees’ agreed terms. The role has a public dimension: the Commission must have regard to various considerations that transcend the particular commercial interests of the parties. It is only following the Commission’s approval that the agreement has legal effect. That process requires the Commission to reach a conclusion about the scope and effect of the agreement. Accordingly, the text should arguably not be coloured by extrinsic context that the Commission could not possibly have known. On this view, the Commission makes enterprise agreements.
This appears to be the position taken in Sheehan v Thiess Pty Ltd [2019] FCA 1762 (Sheehan). In Sheehan,the parties invoked various extrinsic circumstances. In a critical passage, the Court observed at [19]:
By reason of the requirement for approval, the Agreement only took effect on the basis of views as to the manner in which it would operate that were formed by the Commission. Those views were themselves informed by an understanding of the industry in which the Agreement would apply and the broader industrial context known to a specialist tribunal such as the Commission.
As to permissible context, ‘peculiar contextual matters that may have been commonly known’ by negotiating representatives are irrelevant: at [22]. Much of the extrinsic context relied on by the parties was therefore ‘of no assistance’: at [24]. Instead, the Court opined at [24] (emphasis added):
What is important is the decision of the Commission approving the Agreement and the practical aspects of the kind of work to be undertaken that will be governed by the Agreement. It may also be relevant to consider the relevant Award or instrument which confers entitlements absent an enterprise agreement, particularly where the construction of the enterprise agreement is said to confer less advantageous terms to those that would apply without the agreement.
In the circumstances of the case, the Court only had regard to extrinsic context that was: (i) expressly referred to in the Commissioner’s decision approving the Agreement; or (ii) evident from the reasoning of the decision or the terms of the Agreement: at [25]–[27]. Sheehan was unanimously affirmed on appeal, albeit without detailed consideration of extrinsic context: Thiess Pty Ltd v Sheehan [2020] FCAFC 198.
Consideration
This presents a problem. If the Commission made the two EAs, then Endeavour adopts the wrong focus. If the Commission makes enterprise agreements, then it should be an important locus of knowledge. To focus on other participants in the process would be to make the same mistake as seeking legislative intent in the minds of individual parliamentarians.
On this account, Endeavour fails to have due regard to the legislative character of enterprise agreements. The decision prioritises those involved in negotiating the enterprise agreement at the expense of the Commission, and in doing so confuses a public law instrument with a private law instrument.
Other authorities probably favour Endeavour’s approach, but inconclusively. Although this post does not aim to comprehensively consider the case law, two such decisions can be mentioned. First is Ridd. In Endeavour, the Court referred to the summary of principles in Ridd, which held that extrinsic context may include associated documents, ideas which gave rise to terms in a document, and the history of a clause. On the other hand, accepting that summary as correct, it does not address the importance of knowledge or purport to be exhaustive. The extrinsic context relied on in Ridd was probably the type of industry context known to a specialist tribunal or may have been part of the materials provided to the Commission when seeking approval: Ridd at [88]–[96] (Griffiths and SC Derrington JJ). Second is Toyota Motor Corporation Australia Limited v Marmara (2014) 222 FCR 152(Toyota) where the Court held that, for the purposes of section 46 of the Acts Interpretation Act 1901 (Cth) (Interpretation Act), enterprise agreements are made by employers and employees: at 173 [68]. However, this conclusion concerned the applicability of the Interpretation Act, not the question of whose knowledge is relevant for the purposes of extrinsic context. Importantly, both Ridd and Toyota also predate the insertion of section 191A into the Act: see Fair Work Legislation Amendment (Secure Jobs, Better Pay) Act 2022 (Cth), Sch 1 item 525. As noted above, that section contemplates a more active role for the Commission in the enterprise agreement process.
Another possibility is that enterprise agreements are made by employers, employees, and the Commission. The Act offers some support for this. Under s 182, enterprise agreements are ‘made’ when a majority of employees vote to approve it. Endeavour mentions s 182, but does not develop its relevance. On its own, s 182 is unpersuasive. Although the use of the word ‘made’ is suggestive, this section could be described as conclusory or procedural; it does not confer legal effect upon an enterprise agreement. However, s 172 should also be noted. In essence, it provides that employers and employees may make enterprise agreements. Section 172 can be characterised as an empowering provision. On that view, enterprise agreements arguably draw their binding effect, in part, from employers and employees, acting with statutory authority under s 172. Accordingly, it is the knowledge of employers, employees, and the Commission that ought to be the focus. To adopt the language of Endeavour at [58], the Act affords all three entities ‘a role in the process’. That is, enterprise agreements are a unique form of multilateral public law instrument. This conclusion is attractive and could thread the needle between Endeavour and Sheehan. However, it does appear to contradict the tenor of both decisions. It might also be inconsistent with the policy of the statute and the central role it gives to the Commission in reviewing and approving enterprise agreements.
Unfortunately, these issues were not fully ventilated in Endeavour. In one sense, that is unsurprising. On the face of the judgment, the Union’s primary objection to extrinsic context was that any new employee who became bound by the two EAs would not know their history. That argument focuses on the knowledge of persons bound by the two EAs, not the knowledge of the entity (or entities) that made the two EAs, and is ultimately a different issue (and raises questions outside the scope of this post). Regardless of how the issues were argued before the Court, the short point is this: Endeavour treats enterprise agreements as being made by employers and employees rather than by the Commission, and that puts it at odds with Sheehan.
Endeavour therefore reveals unresolved uncertainty as to the nature of enterprise agreements and how they should be interpreted. The uncertainty stems from the hybrid nature of such instruments, which combine familiar concepts from public and private law. This is not just a theoretical issue: the rights and duties of the 2.8 million Australians covered by enterprise agreements depend upon how enterprise agreements are interpreted. The issue is ripe for determination. Endeavour will not provide that opportunity; no appeal was filed, so we will have to play the waiting game.
Rupert Williamson is a Bachelor of Civil Law candidate at Oxford University. His views are all his own.
Suggested citation: Rupert Williamson, ‘Enterprise Agreements, Extrinsic Context, and Knowledge: CEPU v Endeavour Energy Network Management Pty Ltd’ (8 September 2026) <https://www.auspublaw.org/blog/2026/9/enterprise-agreements-extrinsic-context-and-knowledge-cepu-v-endeavour-energy-network-management-pty-ltd>