Boughey, Janina; Weeks, Greg --- "Officers of the Commonwealth' in the Private Sector: Can the High Court Review Outsourced Exercises of Power?" [2013] UNSWLawJl 14; (2013) 36(1) UNSW Law Journal 316

---

‘OFFICERS OF THE COMMONWEALTH’ IN THE PRIVATE SECTOR: CAN THE HIGH COURT REVIEW OUTSOURCED EXERCISES OF POWER?

JANINA BOUGHEY^[*] AND GREG WEEKS^[**]

I INTRODUCTION

Outsourcing, and ‘mixed administration’^[1] more generally, pose many and varied challenges for public law and have attracted considerable academic attention both within Australia and overseas.^[2] One such challenge that has been of concern to administrative lawyers is the extent to which courts are able to exercise their jurisdiction to review administrative action when governments have outsourced^[3] functions to the private sector. Although this issue has been on the radar of administrative lawyers for at least two decades, many of the most important aspects of it remain unresolved – or at least unsatisfactorily resolved – in Australian law. In particular, there is enduring uncertainty surrounding the extent to which the High Court of Australia is able to deal with outsourced exercises of power within its original review jurisdiction under the Australian Constitution.^[4]

Section 75(v) of the Australian Constitution provides that the High Court has original jurisdiction ‘in all matters in which a writ of Mandamus or prohibition or an injunction is sought against an officer of the Commonwealth’.^[5] The provision has been found to entrench a minimum level of judicial review of administrative action in the Constitution.^[6] Yet, despite the fact that the High Court has had a number of opportunities to define the phrase and the circumstances, if any, in which a private sector employee may be an ‘officer of the Commonwealth’ for the purposes of section 75(v),^[7] it has been reticent to do so.^[8] The High Court’s reluctance to define the scope of its jurisdiction over private sector decision-makers is based, understandably, on the need to wait until a matter arises that requires resolution of this issue. Nonetheless, the Court’s continued silence on this point inevitably provokes impatience when one considers the extent to which the Commonwealth Government has outsourced contentious areas of decision-making that directly affect individual rights in recent years. For instance, the function of reviewing initial decisions to refuse ‘offshore’ asylum seekers permission to make an application for a protection visa has been outsourced to a private company, albeit to be performed in accordance with detailed instructions from the Minister.^[9] Further examples can be found in the social security context, where private companies have the power to make decisions that directly affect individuals’ social security entitlements.^[10]Additionally, the High Court’s willingness to leave its constitutional review jurisdiction undefined is perhaps surprising in light of the recent emphasis placed on the centrality of section 75(v) in judicial review of administrative action in Australia.^[11]

On the other hand, the High Court’s reluctance to define the scope of its constitutional review power is less surprising when one examines the common law test on offer. We will argue in Part II of this article that the test that has been adopted within many common law jurisdictions – which first appeared in R v Panel on Take-overs and Mergers; ex parte Datafin plc^[12] – has been notoriously difficult to apply coherently. Not only has Datafin’s ‘public function’ test failed to win an unquestioned place in Australian common law,^[13] but, as we argue below, it is also inherently unsuited to adoption in the Australian Constitutional context. In addition, the High Court has to date been able to find alternative sources of jurisdiction under which it could review decisions made by private contractors, where the Court has considered judicial review appropriate. For example, in the Offshore Processing Case, the unlawfulness of the manuals issued by the Minister for Immigration and Citizenship and used by the private contractors was central to the Court’s reasoning. Because the Minister had indicated unequivocally that he would assess claims according to statutory criteria, the High Court was able to classify the contractors’ errors of law in relying on manuals which misstated the statutory requirements as errors of law which would be made by the Minister if (as seemed likely) he were to adopt the contractors’ reasoning into his own exercise of statutory power.^[14] Thus, there is no practical need for the High Court to make the effort to define the phrase ‘officer of the Commonwealth’ in circumstances where the Commonwealth Government is practically restricted from structuring its relationships with the private sector in a way which avoids public law scrutiny.

Nevertheless, the scope of its jurisdiction under section 75(v) remains an important question for Australian administrative law and is an issue that the High Court will need to confront eventually. Cause to confront the scope of judicial review’s coverage of outsourced decision-making may come sooner rather than later if the Administrative Review Council’s (‘ARC’) recommendations for reform of the Administrative Decisions (Judicial Review) Act 1977 (Cth) (‘ADJR Act’) are adopted by the Commonwealth Government (although at the time of writing no steps have been taken in this regard). Specifically, the ARC recommended in its most recent report that the scope of review under the ADJR Act be broadened to ‘encompass the jurisdiction of the High Court under section 75(v) of the Constitution’.^[15]

This article examines some of these questions and analyses the options available to the High Court in defining the phrase ‘officer of the Commonwealth’ in the context of modern mixed administration. The various tests used in Australian administrative law are explored, with a focus on the ‘public function’ test developed in Datafin. We argue that these administrative law tests are largely unhelpful and inappropriate for defining the scope of section 75(v) of the Constitution. Instead, we suggest that the High Court could find inspiration for the most appropriate and adapted solution from an unlikely place: Canadian human rights law. We argue that by adapting the ‘control’ test used by Canadian courts to determine the scope of the Canadian Charter of Rights and Freedoms (‘Canadian Charter’),^[16] the Australian High Court could find an appropriate balance for reviewing the actions of private sector actors, while simultaneously achieving consistency with existing precedent.

In the second part we briefly describe the history of section 75(v) and existing High Court precedent on the phrase ‘officer of the Commonwealth’. We argue that there is substantial scope and good reason for the High Court to revisit and build on this precedent to accommodate some instances of outsourced power within its review jurisdiction. Part III of the article outlines the common law Datafin approach and argues that it is not appropriate to apply Datafin’s public function test in the Australian constitutional context. In Part IV we examine alternative tests found in administrative law that could be used to define the scope of the High Court’s review jurisdiction, and argue that none is suitable in the context of section 75(v). We then examine the ‘control’ test used under the Canadian Charter and argue that it may be possible to adapt this test in defining the scope of section 75(v) of the Australian Constitution.

An important point to note at the outset of this article is that in advocating for the broadening of the High Court’s jurisdiction to review decisions made by persons within the private sector to whom power has been outsourced, we are not advocating for judicial review of the decision to outsource itself, nor of the instrument used to effect outsourcing. This is a crucial distinction: we are not proposing that the scope of judicial review ought to be expanded to cover all exercises of government power, including contractual power; we are merely arguing that the government entities to whom judicial review might be applied should be more broadly interpreted. For example, we do not think that there is merit to making commercial contracts entered into by government subject to judicial review. Both the common law and judicial review statutes are united on the point that government contracting decisions are incongruous with judicial review,^[17] and we are not suggesting that this be altered.^[18] To do so would place government at an enormous disadvantage,^[19] since there would be inherent uncertainty in every contractual relationship to which it may become party, with the inevitable result that many people would prefer to avoid the risk of government contracting altogether.

On the other hand, government may enter contracts under which the crux of the arrangement is not the provision of goods or services to government, but the performance of services to the public instead of or as a proxy for government.^[20] There is a fundamental difference between government entering into a contract with, on the one hand, a provider of IT services (which will provide, install and maintain certain equipment) and, on the other hand, a provider of personnel who will detain asylum seekers and process their visa applications. The former is a commercial transaction into which any individual or private enterprise could enter. The latter is the business of government which nobody except government ordinarily has cause to perform and which therefore is being performed by contractors in circumstances where government would otherwise be expected to perform the task itself. There is no apparent reason why this latter species of government contracting ought not to be subject to judicial review, as government would be were it performing the task for itself.

II WHO IS AN ‘OFFICER OF THE COMMONWEALTH’?

An analysis of the history of section 75(v) of the Australian Constitution reveals that the assumption underlying the provision is that judicial review remedies may be sought in the High Court’s original jurisdiction against government entities and not other classes of litigant. Quick and Garran made clear that section 75(v) was always intended to have the effect of defining the Court’s jurisdiction by reference to the party against whom remedies were sought, rather than by the power to grant those remedies per se.^[21] This was in part to set the High Court apart from the jurisdictional difficulties that became apparent in the United States Supreme Court in Marbury v Madison.^[22] The close relationship between American constitutional provisions and those which were adopted in Australia has been noted on several occasions,^[23] leading Mark Leeming to suggest that there must be some significance to the fact that section 75(v) ‘is the only class of matter in ss 75 and 76 which lacks a close counterpart in the United States Constitution’.^[24] It was also out of a recognition that the High Court’s original jurisdiction under section 75(iii) allows it to grant the remedies listed in section 75(v) and others in any case.^[25] In other words, section 75(v) was not intended to add to the original jurisdiction invested in the High Court by section 75(iii),^[26] other than in the sense that it was designed as a protection against the Court’s ‘inherent powers’ being read down consistently with American Constitutional jurisprudence,^[27] but nor was it intended to restrict it to granting only the listed remedies. This helps to explain Justice Dixon’s comment that ‘s 75(iii) cannot be read without s 75(v)’,^[28] although whether section 75(v) in fact adds to the jurisdiction granted under section 75(iii) remains an open question.^[29]

James Stellios has argued that, in addition to serving the function of ensuring that the Australian High Court’s original jurisdiction was not limited in the same manner as that of the United States Supreme Court, section 75(v) was also seen as serving at least two other functions.^[30] The first was articulated by Sir Edmund Barton after the clause had been removed from the draft version of the Constitution only to be subsequently re-inserted in 1898. Barton stated that the purpose of section 75(v) was to create an accountability mechanism^[31] by ensuring that the High Court had the power to protect against ‘any violation of the Constitution, or of any law made under the Constitution’.^[32] The other function that section 75(v) was seen to serve was a federalist one. Sir Josiah Symon, who was chair of the Judiciary Committee at the Constitutional Conventions suggested that the sole purpose of the provision was to prevent state courts from reviewing decisions made by federal government officials under Commonwealth laws.^[33]

While the accountability explanation for section 75(v) has been dominant in recent years,^[34] all three functions can be seen to have shaped early case law interpreting the phrase ‘officer of the Commonwealth’. In R v Commonwealth Court of Conciliation and Arbitration; Ex parte Whybrow & Co,^[35] the first case in which it considered the phrase, the High Court gave a broad interpretation to ‘officer of the Commonwealth’ as it was used in section 75(v). The question before the Court in that case was whether prohibition could lie against a judge of the Commonwealth Court of Conciliation and Arbitration. In other words, was a federal judicial officer an ‘officer of the Commonwealth’? The High Court held that the term applied to both judicial and non-judicial officers and was influenced by the fact that the writ of prohibition ‘had frequently lain against inferior courts’ at common law.^[36] In reaching a broad interpretation of section 75(v), the Court was also influenced by the framers’ intent that the provision be read to avoid the Marbury v Madison problem^[37] and the concern that it provide an accountability mechanism through which the High Court could supervise all public officials.^[38]

It is now uncontroversial that the High Court’s original jurisdiction under section 75(v) is sufficient to allow it to review the decision of any justice of a federal court.^[39] Similarly, it is uncontroversial, as Dixon J pointed out in the Bank Nationalisation Case, that ‘the traditional distinction between, on the one hand the position of the Sovereign as the representative of the state in a monarchy, and the other hand the state as a legal person in other forms of government’ did not survive federation.^[40] Justice Dixon continued:

The purpose of s 75(iii) obviously was to ensure that the political organization called into existence under the name of the Commonwealth and armed with enumerated powers and authorities, limited by definition, fell in every way within a jurisdiction in which it could be impleaded and which it could invoke. Section 75(iii) cannot be read without s 75(v) which, it is apparent, was written into the instrument to make it constitutionally certain that there would be a jurisdiction capable of restraining officers of the Commonwealth from exceeding Federal power.^[41]

Justice Dixon clearly supposed that, upon Federation, the Commonwealth and the States obtained ‘distinct individualities’ under the Constitution, and that they would be ‘amenable to the jurisdiction of the courts upon which the responsibility of enforcing the Constitution rests’.^[42] It is not so great a stretch from his Honour’s dictum to a conclusion that section 75(v) can and should, in appropriate circumstances, be read to apply to bodies which act instead of the Commonwealth (for instance, the contractors involved in the Offshore Processing Case). Such an extension would clearly be in keeping with the accountability view of section 75(v)’s function, and would not be inconsistent with either of its other functions. However, the precise scope of who is acting instead of government needs careful definition, and should be only as broad as is strictly necessary. For instance, it need not extend to those acting as agents of an officer of the Commonwealth, or performing functions which can be directly attributed to such an officer, since the Offshore Processing Case has confirmed that section 75(v) is already capable of extending to the acts of such agents or persons under the direction of an officer of the Commonwealth. As such there is no accountability deficit. The coverage of the term ‘officer of the Commonwealth’ should not be given a broader scope than is necessary for the purposes of accountability. Where outsourced power can be regulated by holding accountable a Minister or some other entity who has always been included within the definition of an ‘officer of the Commonwealth’, there is no need to move beyond the narrow, traditional understanding of that term.

One possible method for ensuring that the Commonwealth is held directly responsible where it might otherwise be unaccountable was explored by Finn J in S v Secretary, Department of Immigration and Multicultural and Indigenous Affairs.^[43] The case involved an allegation that the psychiatric care provided to people held in immigration detention was inadequate. The Commonwealth had outsourced the management of immigration detention centres under contracts which required the contractor to provide medical care, but:

who was obliged to ensure that these and other services were adequate? In the absence of clear contractual standards, could the common law duty of care provide guidance on the minimum level of care (or service)? If so, who was obliged to discharge that duty?^[44]

Justice Finn solved this problem by using tort as an ‘accountability backstop’, albeit with the benefit of a concession made on behalf of the Commonwealth that it owes immigration detainees a non-delegable duty of care.^[45] Tort has a long history as a vehicle for challenging invalid government action,^[46] and there is a certain pleasing symmetry to it filling gaps as they appear in the more ‘modern’ administrative law. However, this need not be a task which falls to be decided within established tort law doctrines. As Professor Groves has described it, in S v DIMIA, Finn J was faced with a case which posed a challenge to the ‘dynamic’ doctrine of responsible government and responded appropriately because there is ‘no reason why the courts cannot recognise and take account of the evolving character of government in an appropriate case’.^[47] Another acceptable response (which draws considerably on Justice Finn’s actual response) may have been to ask whether the contractor could have been an ‘officer of the Commonwealth’ or, alternatively, deemed to remain under the direction of Commonwealth decision-makers for some purposes, such as the provision of adequate health care.

It is usually accepted that there is no accountability deficit which needs to be filled by judicial review when a formerly public body is privatised (in that it moves to being privately owned)^[48] because the accountability mechanism is provided by the market. Hence, former public assets like the Commonwealth Bank and QANTAS^[49] do not need to be held accountable through administrative law mechanisms because they operate in competitive marketplaces and accrue no benefit from their origins as publicly owned institutions. This point does not inevitably extend to privatised public utilities (such as Telstra), which often carry their ‘natural monopolistic aspects’ with them into private ownership, although this is offset to some degree by the fact that such bodies are frequently heavily regulated.^[50] The main point is that there is no reason, without more, why judicial review ought to apply to a body simply because the government used to own it. This leaves a relatively small window for bodies to whom the term ‘officers of the Commonwealth’ might properly be extended. As a guide, the need to expand upon the traditional understanding of who is an ‘officer of the Commonwealth’ will arise where there are no accountability measures which cover a body working instead of the government, whether or not as the result of a deliberate attempt to frustrate such measures.^[51]

However, other early High Court decisions appeared to adopt a narrower interpretation of the phrase ‘officer of the Commonwealth’ than was taken in Ex parte Whybrow. In R v Murray; Ex parte Commonwealth,^[52] the Commonwealth brought an action, seeking prohibition against Murray J, a judge of the New South Wales District Court who had exercised federal jurisdiction in awarding damages under the Commonwealth Workmen’s Compensation Act 1912 (Cth) for the death of the Commonwealth’s employee, Cormie. The case stands for the point that, in contrast to what had been decided in regard to federal judges in Ex parte Whybrow, a state judge exercising federal jurisdiction does not fall within the definition of ‘officer of the Commonwealth’ in section 75(v).^[53] Justice Isaacs in fact found that it was ‘decisive’ of the matter that the Commonwealth’s statutory involvement ended with the payment of money correctly ordered in the New South Wales District Court, but nonetheless went on to decide the question of whether Murray J was an ‘officer of the Commonwealth’:

Federal jurisdiction may be entrusted to State Courts, and, if so, the Judges of those Courts exercise the jurisdiction not because they are ‘officers of the Commonwealth’ – which they are not – but because they are State officers, namely, Judges of the States. An ‘officer’ connotes an ‘office’ of some conceivable tenure, and connotes an appointment, and usually a salary. How can it be said that a State Judge holds a Commonwealth office? When was he appointed to it? He holds his position entirely under the State; he is paid by the State, and is removable by the State, and the Constitution knows nothing of him personally, but recognizes only the institution whose jurisdiction, however conferred, he exercises.^[54]

What is clear from this is that Isaacs J was making a limited point about why state judges did not become ‘officers of the Commonwealth’ simply by exercising federal jurisdiction from time to time. Their ‘office’ was State-based, regardless of the fact that their jurisdiction sometimes was not. To focus only on the italicised words in the extract above is to misunderstand that Isaacs J was making a specific point about someone who was an officer but not a Commonwealth officer. Thus, this dictum should not be read as a comprehensive statement of who will fall within the coverage of the term ‘officer of the Commonwealth’ as it appears in section 75(v).

In recent times, the question of who may be an ‘officer of the Commonwealth’ has been dealt with sparsely by the High Court. The unanimous Court in the Offshore Processing Case mentioned the phrase only for the purpose of setting aside

for another day, the question whether a party identified as ‘an independent contractor’ nevertheless may fall within the expression ‘an officer of the Commonwealth’ in s 75(v) in circumstances where some aspect of the exercise of statutory or executive authority of the Commonwealth has been ‘contracted out’.^[55]

In NEAT Domestic, only Kirby J directly addressed the question of whether section 75(v) may have covered the defendant company AWB (International) (‘AWBI’). His Honour commented on the fact that

a private corporation to a large degree controls the conduct of an independent statutory agency of the Commonwealth made up of officers of the Commonwealth answerable to this Court, amongst other ways, under [s 75(v) of] the Constitution. That constitutionally entrenched power of judicial review is one of the limits on the extent to which corporatisation and privatisation of federal administrative action in Australia may escape the disciplines of judicial scrutiny.^[56]

This amounts to a statement that the key point in determining whether a body is ‘answerable’ to the High Court is accountability. In that sense, what we have argued above is consistent with Justice Kirby’s dictum, as it is with Sir Edmund Barton’s purpose for including section 75(v) in the Constitution in the first place.^[57] Professor Stellios has suggested that this was to provide a ‘foundation’ for limiting the Parliament’s use of executive power.^[58] It is important to note, however, that such a capacity to control the executive is still tied to the characterisation of individuals as ‘officers of the Commonwealth’ and ought not to ‘obscure some of the important features of section 75(v)’.^[59] The definition of that term would be constantly in issue even if the High Court were to recognise a constitutional jurisdiction to extend judicial review to private bodies.

The import of the phrase ‘officer of the Commonwealth’ is also somewhat under-analysed in academic circles. It is rarely mentioned in leading constitutional treatises^[60] and considerations of federal jurisdiction^[61] other than in the context of the High Court’s section 75(v) jurisdiction to grant certain remedies in its original jurisdiction. One conclusion which it is open to draw is that constitutional scholars regard the meaning of ‘officer of the Commonwealth’ either as unimportant or as settled. Administrative lawyers, on the other hand, are well-acquainted with the issues raised by outsourcing, which has developed a burgeoning literature over the last two decades or more.^[62] It is that background which informs our call for a broader understanding of who might be an ‘officer of the Commonwealth’.

It is relevant that, in Williams v Commonwealth,^[63] the High Court rejected the plaintiff’s contention that the government’s scheme for providing chaplaincy services to schools violated the prohibition on religious tests in section 116 of the Constitution^[64] on the basis that the provision of chaplains had been outsourced to the Queensland Scripture Union and the chaplains therefore could not be said to hold an ‘office ... under the Commonwealth’ as would be required before the prohibition in section 116 could be invoked.^[65] There is a clear echo in this language of section 75(v), and the certainty of the Court’s rejection of the plaintiff’s argument in the School Chaplains Case might therefore be thought to affect our contentions in regard to section 75(v). We do not think that this is so. The crux of the matter was that the Commonwealth was funding the chaplaincy services without supporting legislation, but had not done so for the purpose of having the chaplains provide a service which was properly the domain of government; quite the reverse.

Justices Gummow and Bell said that ‘the meaning of “office” turns largely on the context in which it is found’ and ‘the phrase “office ... under the Commonwealth” [in section 116] must be read as a whole’.^[66] It follows that the same must be true of the phrase ‘officer of the Commonwealth’ in section 75(v), which arises in the context of government accountability rather than to specify when religious tests are impermissible.^[67]

Having noted the limitations that exist on the courts’ capacity to use section 75(v) as an accountability mechanism, this is an apposite point to address the criticisms made of the Australian High Court that, through its emphasis on the terms of the Constitution and a generally ‘formalist’ approach to both statutory and constitutional interpretation, it has made Australian public law jurisprudence isolationist and of little relevance to courts elsewhere in the common law world. This argument has been made by both Mike Taggart^[68] and Tom Poole^[69] with reference to examples of the High Court’s public law jurisprudence. One such example, NEAT Domestic, has been the subject of an immense amount of academic writing, much of it highly critical of the outcome and the majority’s reasoning.^[70] We do not propose to revisit that debate. However, NEAT Domestic is instructive in relation to the related debate as to whether Australia has become isolated from the common law world as a result of the High Court’s ‘constitutionalisation’ of judicial review.^[71]

Mike Taggart’s criticism of NEAT Domestic and Tang was that ‘majorities in the High Court of Australia [had] failed to grapple with the changing nature of government’.^[72] The disappointment evident in this comment is justified, in as much as the joint judgment of McHugh, Hayne and Callinan JJ did not consider privatisation or the possibility that ‘government work’ was being done through a private entity; indeed, we share that disappointment. Justice Kirby did conduct an analysis of this type in dissent but, as Tang later showed, his Honour’s understanding of the ADJR Act was fundamentally out of step with the rest of the Court, for reasons which have nothing to do with the Constitution. Chief Justice Gleeson also considered when and why a private body might be required to take account of the ‘national interest’,^[73] but decided conformably with the other majority judges in regard to the applicability of the ADJR Act. So, while it is fair to criticise the reasoning of the High Court majority in NEAT Domestic (and, even more so, that in Tang), what the criticism boils down to is that the joint judgment failed to consider the issues which Gleeson CJ dealt with as obiter dicta. Because the ADJR Act did not apply, Chief Justice Gleeson’s typically percipient remarks about ‘the changing nature of government’ could not have made a difference to the outcome of the case.^[74]

Furthermore, it is important to note that even the applicant in NEAT Domestic knew that it was rolling the dice by seeking statutory judicial review under the ADJR Act but had to do so because it knew that the respondent company could not be characterised as an ‘officer of the Commonwealth’ for the purposes of section 75(v). There was no argument before the Court that it should change its interpretation of ‘officer of the Commonwealth’ – severely but properly limiting the Court’s scope to consider the issue – and therefore the greatest available criticism that can be made is that the Court either misconstrued the terms of the ADJR Act^[75] or that the ADJR Act is in need of legislative reconsideration and amendment. If, as we think, the latter is the case, the outcome of NEAT Domestic can hardly be laid at the feet of the High Court as an example of Australia becoming isolated from the common law world.^[76]

Poole’s attack on NEAT Domestic was conducted at greater length than Taggart’s but was rather less precise. In relation to the majority joint judgment’s finding that private corporations owed no obligations of a public law nature, he charged that ‘[n]ot only does this decision present a formalist solution to what has been called the “Datafin problem” ... [i]t also wraps the decision in a narrow, legalist conception of the role of the court’.^[77] We note in passing that criticising the High Court on the basis of its ‘formalism’ or ‘legalism’ leads to no little confusion when it is done by using labels of inherently subjective and contested meaning.^[78] Furthermore, we would query whether formalism is the sin that it is often assumed to be where there is still a capacity to seek judicial review at common law. Finally, to the extent that ‘formalism’ and ‘legalism’ refer to a mode of statutory interpretation which strives to give effect to the legislative purpose of the ADJR Act, it appears that Poole may either have misunderstood the nature of the statute or misdirected his criticism. Heavy reliance on ‘vague’^[79] labels like ‘formalist’, ‘conservative’ and ‘legalist’ serves only to obscure the true concerns of those who employ them and hinder meaningful engagement with the very real issues faced by the High Court in judicial review matters. This is ironic inasmuch as the primary complaint directed at the High Court by those who use such labels appears to be that its judicial style results in a lack of ‘directly normative or principles-based’ judicial review grounds.^[80]

One of the least controversial elements of NEAT Domestic (and Tang after it) should be whether the respective decisions were made ‘under an enactment’ as required by the ADJR Act. They were not; at any rate, much as commentators have disagreed passionately with the outcomes of these cases, the Court’s reasoning on this point has not been persuasively challenged.^[81] The most that can be taken from Poole’s attack is that he does not agree with the result or much of the reasoning in NEAT Domestic. It is his right to take that view.^[82] However, what he does not do is spell out why the decision – arguably based upon a strict interpretation of a domestic Australian statute but having nothing to say about the Australian Constitution – does anything to ‘isolate’ Australia. Our suspicion here is that Poole and others believe that there is nothing to prevent the High Court from reaching better outcomes in judicial review cases other than its own ‘formalist’ tendencies. This is not so, whether because a case has been pleaded under the ADJR Act and must therefore adhere to the statutory requirements for establishing jurisdiction^[83] or because a case has been pleaded under section 75(v) and therefore needs to seek at least one of the enumerated remedies against an ‘officer of the Commonwealth’. One of the only things that the High Court, as opposed to the Commonwealth Parliament, can do to broaden the scope of judicial review at Commonwealth level is to reassess the definition of an ‘officer of the Commonwealth’. Few of the Court’s critics have latched onto this issue.

The frequent critiques of the High Court’s performance in the years since the retirements of Brennan CJ and (especially) Mason CJ are ultimately overblown to the extent that they suggest that the High Court has made judicial review heartless at home and irrelevant abroad. This is not to say that the decisions which have been the subject of the most trenchant criticisms are unimpeachable: we are not apologists for the High Court’s judicial review decisions over the course of the last twenty years. However, we do argue that much of the criticism of the Court – particularly but not exclusively in regard to its application of section 75(v) – misunderstands the nature and effect of a written constitution. While there is much force to suggestions that the High Court is isolating Australia by bedding its public law jurisprudence down in Constitutional analysis,^[84] to a great degree this is simply a factor of Australia both having a written constitution and no ‘judicially enforceable bill of rights at the federal level’.^[85] To the extent that these facts are considered shortcomings, they are not shortcomings which can be remedied by the High Court.

This is not to say that there is no role for the High Court to develop the scope of its public law jurisdiction in accordance with the increased likelihood that public power will be exercised by private bodies. We suggest that the High Court has the capacity to expand the understanding of the phrase ‘officer of the Commonwealth’ that has remained unchanged and virtually unexamined for almost a century. However, the phrase cannot and ought not be altered in such a way that renders its meaning at odds with the normal meaning of its component words and the context in which section 75(v) appears. Any expansion must be consistent with a plausible and logical interpretation of the words of section 75(v). Particularly, any revised test needs to retain section 75(v)’s current focus on the identity of the decision-maker, which should be able to be determined without great difficulty.

III WHY DATAFIN IS NOT THE ANSWER

Perhaps the most obvious way for the High Court to expand the scope of its constitutional review jurisdiction would be to adopt the common law test for determining the scope of judicial review. That test, first articulated by the English Court of Appeal in 1987 in Datafin, asks whether a private body is in effect performing a ‘public duty’, or whether the power being exercised has a ‘public element’.^[86] In essence it requires courts to distinguish between private and public functions and is for this reason frequently referred to as a ‘public function’ test.^[87] The public function test now determines the scope of common law judicial review of administrative action in the United Kingdom, New Zealand^[88] and probably Canada.^[89] It was previously also widely assumed that the public function test applies to the judicial review jurisdiction of state and territory supreme courts under common law,^[90] having been applied^[91] or cited with approval by the Supreme Courts of New South Wales,^[92] Victoria,^[93] the Australian Capital Territory^[94] and Queensland.^[95] However recent decisions of the New South Wales Court of Appeal,^[96] the Full Court of the South Australian Supreme Court^[97] and the Victorian Court of Appeal^[98] have cast doubt on whether Datafin does in fact form part of Australian common law.^[99]

An increasingly likely possibility is that Datafin has indeed been adopted into the Australian common law but that very frequently it will simply not make a difference. A look at the facts of Datafin itself demonstrates this point. We are unlikely ever again to see circumstances like those in Datafin, in which a body wielded huge regulatory power, which was both recognised in statute and officially sanctioned, without any ‘visible means of legal support’,^[100] either statutory or prerogative. It is hard to imagine a modern government allowing a body like the Panel to operate without legal constraints.^[101] The Court of Appeal’s decision that such a body must at least be subject to judicial review is all the more understandable given that the Panel was set up such that it lacked legal personality. Judicial review, in effect, was held to stretch as far as it needed to for the purposes of accountability.^[102] Seen in this way, Datafin was an obvious case for judicial intervention. The reality in Australia over the last quarter of a century, however, is that judicial review’s flexibility has never been tested a fraction as much as it was in Datafin. The battleground shifted to (or perhaps simply remained with) public authorities’ liability in tort.^[103] Datafin’s relevance is that it shows that judicial review can be adaptable where adaptation is required.

Assuming that Datafin is part of Australian common law, then the High Court’s adoption of the ‘public function’ test in its constitutional jurisdiction would have the benefit of ensuring unity in the scope of judicial review within Australia. The High Court has previously emphasised the importance of there being ‘but one common law in Australia’,^[104] as well as the fact that Australian courts are ‘a single integrated system’.^[105] A number of law reform bodies making recommendations for statutory forms of judicial review have made the same point.^[106] Thus it would make sense (or at least satisfy a desire for legal tidiness) if, insofar as the language of the Constitution permits, the High Court’s supervisory jurisdiction over Commonwealth administrative decisions were identical in scope to that of state and territory superior courts over their respective governments.

However, there are several reasons why we do not consider the Datafin test to be appropriate for application to section 75(v). The first is practical. In those jurisdictions where a public function test has been adopted, it has proven to be notoriously difficult to find a coherent method of delineating public from private power.^[107] Uncertainty, to a large extent, is the price that we pay for flexibility in determining which exercises of power are sufficiently public in their nature to attract judicial review.^[108] Paul Craig has argued that the two possible alternatives to the public function test would examine, respectively, the source of the power or the scope of judicial review’s remedies.^[109] The source of power test is more certain but, as we have learned from years of observing the operation of the ADJR Act harnessed to the highly restrictive ‘under an enactment’ limitation, this certainty comes at the cost of excluding even sources of power which are well-established as being judicially reviewable at common law, such as the prerogative.^[110] The second test, focussed on remedies, presents a couple of difficulties. The first of these is that it is either based on an historical analysis of where the remedies have gone in the past or it begs the essential question by restating itself as turning on a distinction between public cases and private law cases.^[111] The second difficulty, as Paul Craig has put it, is that if the scope of the judicial review remedies is given a wide and flexible ambit, there is very little difference from a test which asks a court to determine whether a given function is public or not.^[112]

A second, more important reason why we contend that a public function test is unsuitable for defining the scope of section 75(v) is that it would require a significant logical stretch and would alter the focus of the section. As discussed in Part II above, section 75(v) was clearly intended to be limited not by the type of power being exercised, but by the identity of the decision-maker. While a public function test might be appropriate under common law, this means that it does not assist in defining the phrase ‘officer of the Commonwealth’, regardless of how convenient it might be to have a unified approach to judicial review across Australia’s superior court system.

The United Kingdom’s Human Rights Act 1998 (UK) c 42 (‘HRA’) expressly defines its scope by way of a core institutional test, which is extended by a public function test. Section 6(1) of the HRA makes it ‘unlawful for a public authority to act in a way which is incompatible with a Convention right’.^[113] ‘Public authority’ is defined as including ‘any person certain of whose functions are functions of a public nature’.^[114] This provision is quite circular,^[115] but the courts have given some definition to it. For example, statutory powers do not necessarily amount to functions of a public nature,^[116] but by contrast a body can be a ‘public authority’ without all of its functions necessarily being ‘functions of a public nature’. In Aston Cantlow, it was held that section 6(3)(b) recognises that some bodies are ‘hybrid bodies’, and are therefore not considered to be ‘a public authority in respect of an act of a private nature’.^[117] ‘Hybrid public authorities’ are able to be held accountable under the HRA only in respect of specific functions which are public, but not in respect of their private functions.

It is doubtful that the reasoning which led the House of Lords to this conclusion is likely to assist the High Court to take a broader view of the meaning of ‘officer of the Commonwealth’ under section 75(v). One reason for this, as Lord Hope pointed out in Aston Cantlow, is that the interpretation of what constitutes a ‘hybrid body’ under section 6(3)(b) of the HRA does not even necessarily benefit from domestic English case law, since it is essentially a matter of international law which ‘must be examined in the light of the jurisprudence of the Strasbourg Court as to those bodies which engage the responsibility of the state for the purposes of the Convention’.^[118] A second reason is that at its core, like the Datafin approach, the HRA still requires courts to delineate between inherently public and private functions. Accordingly it carries the same uncertainties as the Datafin test, and is equally unsuited to the Australian Constitutional context.

The Victorian Charter of Human Rights and Responsibilities Act 2006 (‘Victorian Charter’) provides a long and complicated definition of the ‘public authorities’ to which it applies, which includes

an entity whose functions are or include functions of a public nature, when it is exercising those functions on behalf of the State or a public authority (whether under contract or otherwise).^[119]

The Victorian Charter goes on to provide a non-exhaustive list of factors that indicate a function is ‘of a public nature’, which includes: functions ‘connected to or generally identified with functions of government’;^[120] regulatory functions;^[121] and functions funded by government.^[122] Despite the clear invitation that these provisions extend to Victorian courts to explore the issue of when outsourced functions will be subject to the Victorian Charter, there has been very little analysis of this issue.^[123] In one of the rare instances in which the provisions were considered by the Victorian Civil and Administrative Tribunal, Justice Bell endorsed the above UK HRA authority and suggested that the definition of ‘public authority’ under the Victorian Charter encompasses all of the UK HRA precedent, ‘and goes further’.^[124] Justice Bell went on to hold that a private company under a contract with government to provide transitional housing was itself a ‘public authority’ for the purposes of the Victorian Charter.^[125] However, Victorian courts have not yet addressed this issue, and in its review of the Victorian Charter the Victorian Parliament’s Scrutiny of Acts and Regulations Committee (‘SARC’) found that these provisions were so unclear that they ought to be replaced with a list of bodies that fall under the definition of ‘public authority’.^[126] Indeed, the effects of the Victorian Charter on administrative decision-making generally remain so unclear as to be likely to obscure the opportunity that Victorian courts have to develop the concepts of ‘functions of a public nature’ and acting ‘on behalf of the State’.^[127]

Finally, we contend that the real problem with any attempt to graft Datafin onto the jurisdiction provided for in section 75(v) is that it asks the wrong question. Despite the fact that strict adherence to the words of Australia’s written constitution has exposed the High Court to criticism, there is little that can be said that allows the Court to escape the fact that it only has original jurisdiction under section 75(v) to grant certain remedies against an ‘officer of the Commonwealth’. Tests that inquire whether a body has a public function, or serves a public purpose, or is funded by government, or operates in the public interest, ask the wrong question. The only thing at issue is whether a person or body can fit within the definition of an officer of the Commonwealth, either on a literal reading of the section or on a purposive basis. We argue in this piece that the term ‘officer’ is somewhat elastic but the elasticity of Constitutional language – or indeed of any words – is limited. To establish that a body has a public function means nothing unless it goes to prove that the body is therefore an ‘officer of the Commonwealth’.

There is, we think, some scope for the word ‘officer’ and its cognate ‘office’ to be read flexibly. It is true that the focus of Isaacs and Higgins JJ in R v Murray on appointment and (salaried) payment as being indicative of a person being an ‘officer’ is reflected in the Oxford English Dictionary.^[128] However, this is not definitive, since the OED also provides definitions of ‘office’ and ‘officer’ which look more generally to whether a person fills a certain role, performs a duty or a function, occupies a position of authority or trust, or is simply ‘an appointed or elected functionary’.^[129] These definitions have no necessary link to the appointment or payment arrangements relating to an officer and instead attach to the capacity in which a person does what she or he does. It is certainly arguable on this approach that the employees of a company^[130] which is given the task under contract of detaining unauthorised maritime arrivals who are brought to a specific location and of assessing their refugee status are ‘officers of the Commonwealth’.^[131]

Once it is accepted that at least part, although arguably all, of the purpose of section 75(v) was to hold the Commonwealth accountable, it makes sense to interpret ‘officer’ in such a way that gives effect to that purpose. It does not matter, in our view, that a person who meets the criteria for being described as an officer in 2013 would not necessarily have been so described in 1900.^[132] What matters is twofold: first, that the entity to whom section 75(v) is being applied fits within the meaning of the term ‘officer’; and secondly, that the purpose for which section 75(v) was drafted is thereby effected. The benefit of this approach is that the Constitution’s capacity to hold people accountable for exercises of power by or instead of the Commonwealth is able to keep pace with changes in government’s relationships with the private sector as they continue to evolve.

IV OTHER POSSIBLE SOLUTIONS

A Other Australian Administrative Law Tests

There are three other tests used in Australian administrative law which define the scope of judicial review with respect to outsourcing. The first is found in the ADJR Act and requires that a decision be ‘of an administrative character’ and ‘made under an enactment’. The ADJR Act test can be dismissed fairly quickly as a potential source of inspiration for reading section 75(v) more broadly for two reasons. The first is that it has been subject to substantial criticism for its narrow scope,^[133] so much so that the peak body in federal administrative law has now recommended that it be significantly amended.^[134] The second is that, like the Datafin test, the focus of the ADJR Act’s jurisdictional formula is completely different to that required by section 75(v): it focuses on the source of power rather than the identity of the decision-maker.

Queensland’s judicial review statute was deliberately designed to extend the scope of review over outsourced powers beyond the ADJR Act’s very restrictive test. The Judicial Review Act 1991 (Qld) also applies to powers and functions exercised ‘under a non-statutory scheme or program’ that utilises public funds.^[135] Despite this clear attempt to extend judicial review to certain outsourced decisions, the provision has not had this effect.^[136] Queensland courts have found it particularly difficult to identify what will amount to a ‘scheme or program’, and it seems that one-off decisions will generally not fall within this provision.^[137] Thus, the additional provision has not made significant inroads into addressing the accountability deficit resulting from the outsourcing of government powers.

A third type of test, which is currently found in Victoria’s Administrative Law Act 1978 (Vic) and has been endorsed by Matthew Groves and the New South Wales Department of Justice with respect to statutory tests at the state level, is linked to the scope of natural justice.^[138] In essence, a ‘natural justice test’ would permit judicial review of decisions in respect of which the decision-maker is already required to observe the rules of natural justice. Such a test would have a number of advantages from a policy perspective. As noted by the New South Wales Department of Justice it would provide a fairly simple, well-defined and straightforward test that would avoid many of the difficulties associated with Tang.^[139] It would also extend judicial review’s remedies to a small category of ‘private’ bodies: namely those commonly referred to as ‘domestic bodies’.^[140] However, although a natural justice test offers many benefits over existing statutory tests,^[141] it does not offer a solution for the High Court of Australia in defining the scope of its jurisdiction over outsourced administrative power for two main reasons. First, the test is limited to situations where the common law provides that natural justice applies – which is either undefined in Australia or depends on a public power test, depending on whether or not one agrees with Justice Kyrou’s analysis.^[142] If the latter, then, as Groves acknowledges, the test would not generally permit courts to review decisions made by officials in the private sector performing services in a mixed administration.^[143]

Secondly, the scope of the natural justice test is ultimately defined by the impact of a decision on individuals (ie, whether the decision affects a person’s rights, interests or legitimate expectations) and not the identity of the decision-maker as demanded by the text of section 75(v). Thus, like the Datafin test, it would require too great a stretch of the language of the Constitution to apply a natural justice test to determine the identity of ‘officers of the Commonwealth’.

B Canadian Tests

Given the lack of guidance available within Australian administrative law, it may be useful for the High Court to look overseas to find an appropriate test for defining its supervisory jurisdiction under section 75(v) over outsourced exercises of power. Assistance may be found in Canada, given that its legal system shares much common ground with ours in respect of its origins and its written constitution. The question of whether, or to what extent, the principles of judicial review of administrative action apply to the private sector has not received a great deal of attention from Canadian courts or commentators. In the limited case law on the issue, provincial and territorial courts have either assumed acceptance of, or expressly applied, the test articulated by the English Court of Appeal in Datafin.^[144] That is, courts have jurisdiction to review decisions of any body that is under a public duty and exercises a public function, regardless of whether the body’s powers are sourced in statute.^[145] The Supreme Court of Canada has not yet addressed the issue directly, though obiter comments indicate that a Canadian federal and/or provincial government will not be able to ‘avoid public law duties when delegating its functions by way of contract or other form of agreement’,^[146] as well as an inclination to endorse the lower courts’ adoption of a public function test.^[147] Unlike Australia, there is no express language in Canada’s constitutional documents that confines courts in defining the scope of its review jurisdiction.^[148] Accordingly there is no constitutional barrier to Canadian courts extending their common law review jurisdiction to all bodies exercising ‘public power’.

The Federal Court of Canada has recently adopted a broader test that applies Datafin’s public function analysis as just one indicium of public power.^[149] The additional factors that indicate a particular power may be ‘public’ are drawn from the various other tests that have been proposed by leading commentators and applied by Canadian and English courts. They include:

1. the traditional ‘source of power’ question;

2. considerations relating to the relationship between the decision-maker and government and extent to which the latter directs, controls or influences the former;

3. the ‘suitability of public law remedies’;

4. whether the decision-maker was exercising compulsory power; and

5. ‘an “exceptional” category of cases where the conduct has attained a serious public dimension’.^[150]

There are a few points to note about these recent Canadian Federal Court judgments. The first is that technically they only apply to the Federal Court’s statutory jurisdiction under the Federal Courts Act^[151] to review decisions made by ‘a federal board, commission or tribunal’, and not directly to review under the common law. However the Federal Court’s reliance on provincial and English precedent, as well as its comments in earlier cases,^[152] indicate that the scope of its statutory review jurisdiction and common law review is identical in this respect. A second point is that the Supreme Court of Canada has not yet had the opportunity to consider this broader approach, but has previously made comments that appear to assume that the Federal Court’s judicial review jurisdiction is restricted to decisions made under statutory power.^[153] Finally, a number of the listed indicia have only been applied in one or two administrative law cases, sometimes only in obiter dicta and ought not be regarded as settled principles.

The Canadian Federal Court’s broad amalgam of tests may offer useful guidance to Australian courts reviewing decisions under the common law, but is slightly less helpful in defining the phrase ‘officer of the Commonwealth’. A number of the factors included on the list, including the public function and natural justice tests, are unsuitable for inclusion in any section 75(v) test, for the reasons discussed above. Others may be unnecessary to incorporate within the scope of section 75(v) due to the High Court’s other sources of jurisdiction, particularly under sections 75(iii) and 73. However, one element of the Canadian Federal Court’s list that may be particularly useful to the High Court of Australia is the suggestion that a private body that is ‘an agent of government or is directed, controlled or significantly influenced by a public entity’ may be subject to public law.^[154] This particular test has not been applied in Canadian administrative law, but is used in another area of Canadian public law: namely to determine the scope of the Canadian Charter.

Given Australia’s position of having no rights protection instruments at the Commonwealth level,^[155] it is somewhat ironic that the area of Canadian public law which may provide some useful guidance to the Australian High Court in defining the phrase ‘officer of the Commonwealth’ is the Charter. Nevertheless, it is the Charter that has caused Canadian courts to grapple with similar legal and policy questions that the Australian High Court would have to confront in defining the scope of section 75(v). Moreover, there is some precedent for Australia looking to Canada in regard to developing Constitutional doctrines.^[156]

Like section 75(v) of the Australian Constitution, the Canadian Charter contains a provision which expressly directs its scope, and accordingly the jurisdiction of Canadian courts to issue remedies. Specifically, section 32 of the Charter provides that it applies ‘to the Parliament and government of Canada’ and ‘to the legislature and government of each province’. In defining the term ‘government’ under the Charter, Canadian courts have deliberately avoided the Datafin test and chosen an alternative that focuses on the extent to which the actions of private organisations are attributable to government. There are still many aspects of the test that remain uncertain. For instance, the Supreme Court is yet to rule on its application in the context of either privatisation or contracting-out, though it has made various comments indicating its likely approach to those circumstances, which are discussed below. Nor has the issue been the subject of an extensive body of case law or academic commentary. Nevertheless, at least at first blush, it does appear to be appropriate for the Australian Constitutional context, which the Datafin test is not, for the reasons we have discussed above.

1 The Canadian Charter’s ‘Control’ Test

The test that Canadian courts apply to determine whether a decision-making body is ‘government’ for the purposes of the Charter is, perhaps misleadingly, labelled the ‘control’ test. The test now looks beyond the more obvious indicia of ‘control’ – such as funding and governance – and examines more broadly whether decisions are ascribable to government,^[157] or whether a decision-maker forms part of the ‘machinery’ of government.^[158] This requires an analysis of a wider range of factors including the extent to which government directs the exercise of discretionary powers and whether the relevant function is done in the furtherance of a government scheme or program.

The leading case on the control test is McKinney, which was handed down together with three other cases that raised similar issues: Harrison v University of British Columbia;^[159]Stoffman v Vancouver General Hospital;^[160] and Douglas/Kwantlen Faculty Association v Douglas College.^[161] All four cases involved challenges to mandatory retirement policies of either tertiary education institutions or hospitals on the basis that the policies infringed the right to equality under section 15 of the Charter.^[162] The first question the Supreme Court had to answer in each case was whether the particular body whose retirement policy was impugned was required to comply with the Charter: in other words could universities, colleges and hospitals be considered ‘government’ for the purposes of section 32. Justice La Forest began his leading judgment by noting that ‘[t]he exclusion of private activity from the Charter was not a result of happenstance. It was a deliberate choice which must be respected’.^[163]

The appellants in McKinney argued that universities were part of ‘government’ for Charter purposes relying on the two traditional administrative law tests, both of which had previously been applied in the Charter context: because their powers derived from statute; or because they performed an inherently ‘public’ function.^[164] While noting that ‘it would be strange if the legislature and the government could evade their Charter responsibility by appointing a person to carry out the purposes of the statute’, La Forest J rejected both the source and function tests:

[T]he mere fact that an entity is a creature of statute and has been given the legal attributes of a natural person is in no way sufficient to make its actions subject to the Charter. Such an entity may be established to facilitate the performance of tasks that those seeking incorporation wish to undertake and to control, not to facilitate the performance of tasks assigned to government.^[165]

His reasons for rejecting the source test echo those of a majority of the Australian High Court in NEAT Domestic. There, McHugh, Hayne and Callinan JJ reasoned that as a company formed under the Corporations (Victoria) Act 1990 (Vic), AWBI was able to make decisions and enter into contractual relationships without needing statutory support.^[166] More than this, however, as a corporation, AWBI also had a positive duty to put the ‘pursuit of its private objectives’ before any public interests, since that duty owed to AWBI’s shareholders was implicit in its incorporated status.^[167] This view did not draw the unqualified assent of Gleeson CJ,^[168] who nonetheless joined the majority in finding for the respondents, on the basis that the ADJR Act did not apply. However, as will become clear, to whatever extent Justice La Forest’s judgment in McKinney drew upon similar reasoning to NEAT Domestic, the similarity between the two decisions ends in their treatment of Datafin.

With respect to the public function test, La Forest J, in tacit approval of Datafin in the administrative law context, noted that although universities perform a public service and may be regarded as ‘public decision-makers’ for the purposes of judicial review, this does not necessarily make them part of the ‘government’ for the purposes of the Charter.^[169] He drew a distinction between the purposes of judicial review, which he saw as ensuring that ‘administrative decision-making was legally and procedurally correct’, and the Charter, which deals with substantive rights.^[170] Although this distinction would probably not hold up in Canadian law today,^[171] it justified Justice La Forest’s imposition of a stricter test to determine the scope of the Charter than for judicial review of administrative action. Justice La Forest concluded:

A public purpose test is simply inadequate. It is fraught with difficulty and uncertainty. It is simply not the test mandated by s 32.^[172]

Justice La Forest briefly mentioned a number of earlier provincial court decisions holding that the Charter applied to municipal governments on the basis that they exercised the inherently governmental power of enacting coercive laws, noting that they were of little assistance in this case.^[173] Justice Wilson dissented from the majority and found that a public function test should apply to the Charter^[174] and that the respondent universities met that test. In her dissent, Wilson J gave much lengthier consideration to the ‘coercive powers’ test.^[175] Her Honour argued against the approach of the provincial courts, finding that while it may be appropriate for the US, it was not appropriate in the Canadian context where government exercises power in many more ways than just through coercive means.^[176]

Having rejected provincial court precedents applying the source of power and public function tests, and finding no assistance in precedent, La Forest J moved on to the more challenging task of identifying the appropriate test for determining the scope of the Charter. The test that La Forest J ultimately arrived at was one that examined the relationship between provincial governments and the universities,^[177] and the extent to which the former had ‘control’ over the latter. The various factors of which La Forest J took account in his analysis included:

6. Establishment: Many Canadian universities were established by, and have their broad powers set out in, provincial statutes. Others were established by private organisations and had their powers codified later.^[178]

7. Funding: The vast majority of the universities’ funding comes from the provincial government. The provincial government exercised significant control over university finances by controlling the costs universities could charge for tuition, by determining the total amount of funding, as well as through earmarking funds for capital works and special programs.^[179]

8. Governance: While the ultimate fate of the respondent universities rested with government as a result of the government’s role in regulating and funding them, this does not make them ‘organs of government’.^[180] Justice La Forest noted that the universities are self-governing, autonomous bodies and the provincial government only appoints a minority of members to each university’s governing body, and in some instances, none. He concluded that ‘[t]he government thus has no legal power to control the universities even if it wished to do so’.^[181]

9. Direction: The final factor examined by La Forest J was the extent to which government directed the affairs of universities. On this he stated:

Any attempt by government to influence university decisions, especially decisions regarding appointment, tenure and dismissal of academic staff, would be strenuously resisted by the universities on the basis that this could lead to breaches of academic freedom. In a word, these are not government decisions.^[182]

In combination, La Forest J described these factors as indicative of whether or not a particular body was part of the ‘machinery’ or ‘apparatus’ of government^[183] and concluded that the universities in question were not controlled by government and so were not subject to the Charter.^[184] ‘Control’, for the purpose of this test, included neither mere influence nor legal control, but is assessed using a multifactorial approach.

However, in other situations involving tertiary educational institutions, the Supreme Court has reached the opposite conclusion applying the control test.^[185] For instance in Douglas College, handed down on the same day as McKinney, the Court found that the government exercised a degree of control over the College’s operations sufficient to indicate that the College was a Crown agent implementing government policy.^[186] The central differences between the institutions examined in McKinney and Douglas College were the fact that the board of the latter was wholly government appointed (and removable) and the provincial government was able to direct the College’s operations.^[187]

The ability of government to direct the routine operations of an organisation has been the critical factor in most recent decisions on the control test. The Court has emphasised that there is a distinction between ‘ultimate or extraordinary’ and ‘routine or regular’ control.^[188] While government may, through funding and regulation, have ‘ultimate’ control over organisations like hospitals and universities, in the sense that it supplies the funds which allow those institutions to remain active, this will ordinarily not alone be sufficient to subject the organisation to Charter scrutiny. ‘Routine’ or ‘regular’ control is required in order for a body to form part of the ‘government’ for the purposes of section 32 of the Charter, including, for instance, the ability of a Minister or Department to issue policy directions, or to direct the day-to-day operations of the body.^[189]

In its most recent discussion of the control test in Greater Vancouver Transport Authority v Canadian Federation of Students – British Columbia Component,^[190] the Supreme Court of Canada accepted that TransLink, a statutory corporation that operates public transit in Vancouver, was under the ‘control’ of municipal and provincial government. The Supreme Court essentially accepted the findings of Prowse JA from the British Columbia Court of Appeal on this issue,^[191] which were based on the facts that: governments appointed all of the members of TransLink’s board;^[192] its empowering Act sets out clearly public objectives for the organisation;^[193] a government body^[194] was required to ratify all of TransLink’s capital and service plans and policies which governed the day-to-day operation of the organisation.^[195] These factors, her Honour said, ‘suggest that the GVRD enjoys a substantial degree of control over TransLink’s basic day-to-day planning and activities’.^[196] The majority of the Supreme Court added that Justice of Appeal Prowse’s conclusion is supported by the principle that ‘a government should not be able to shirk its Charter obligations by simply conferring its powers on another entity’.^[197] Intriguingly, the Supreme Court also noted the fact that the creation of TransLink was an ‘administrative restructuring’ intended to give more power to local government, rather than a provincial attempt at privatisation, and indicated that this was a factor in why TransLink’s operation was not a ‘Charter-free zone’.^[198] It is not clear what the Court meant by this statement. It appears to be a suggestion that the Charter may not apply to situations of privatisation, regardless of whether the privatised body meets the control test, however this is the first indication of this kind from the Supreme Court of Canada. If that is the meaning intended by the Supreme Court, it is at least consistent with the ‘control’ test inasmuch as government does not usually retain control over privatised bodies; the cord has been cut. It is also consistent with our argument: for the reasons given above, we do not advocate a role for judicial review in supervising private bodies on the basis that they were once government-owned.

‘Routine’, ‘regular’ or ‘day-to-day’ control was also the decisive factor in the more recent provincial court decisions in Sagen v Vancouver Organizing Committee for the 2010 Olympic & Paralympic Winter Games^[199] and Canadian Blood Services v Freeman.^[200] The former involved a Charter challenge by female ski-jumpers to the failure of the Organizing Committee (‘VANOC’) to include women’s ski jumping on the Olympics schedule. In applying the control test to VANOC, the British Columbia Supreme Court saw the limited role that government played in decision-making regarding VANOC’s day-to-day operations as outweighing its significant financial control and power to appoint a (albeit bare) majority of board members.^[201] Similarly, Aitken J in the Ontario Superior Court of Justice found that Canadian Blood Services (‘CBS’) was not controlled by government despite the fact that various governments were involved in establishing its mission, approving its business plans, requiring audits and selecting board members. Justice Aitken found that none of this involvement actually involved the direction of CBS’ policies or daily operations, and as such did not constitute control for Charter purposes.^[202] Government involvement alone does not and should not form the basis for establishing government control (under the Charter) or concluding that the party with which government is involved is therefore ‘an officer of the Commonwealth’ under the Australian Constitution.

2 Application to Specific Activities of Otherwise Private Organisations

The four earliest cases on the control test all involved employment policies, so there was no need for the Supreme Court to examine whether, if the body itself was not ‘government’ for Charter purposes, the impugned actions were nevertheless attributable to government and subject to the Charter. However, this will be the critical question in the context of contracting-out, where it will generally not be the case that the contracting body is wholly government funded or controlled. In McKinney, La Forest J made the point in obiter dicta that although the universities were private bodies for Charter purposes, it may still be possible for some of their activities to be subject to the Charter ‘where it can fairly be said that the decision is that of the government, or that the government sufficiently partakes in the decision as to make it an act of government’.^[203] The Court gave further thought to this issue in Eldridge and reiterated Justice La Forest’s earlier remark that ‘it would be strange if the legislature and the government could evade their Charter responsibility by appointing a person to carry out the purposes of the statute’.^[204] In Eldridge, the Court found that:

an entity may be found to attract Charter scrutiny with respect to a particular activity that can be ascribed to government. This demands an investigation not into the nature of the entity whose activity is impugned but rather into the nature of the activity itself. ... If the act is truly ‘governmental’ in nature – for example, the implementation of a specific statutory scheme or a government program – the entity performing it will be subject to review under the Charter only in respect of that act, and not its other, private activities.^[205]

While the Court’s use of the term ‘governmental in nature’ may carry suggestions of a public power or function test, the Court again made it clear that the fact that a body performed acts with a public purpose would not be sufficient to make those acts attributable to government.^[206] A more significant link with government is required in order for the Charter to apply, such as the act being done to implement a specific governmental scheme or program. On the facts of Eldridge, the Court unanimously found that while hospitals themselves are private entities, they are required to comply with the Charter when providing medical services under legislation, for which they are reimbursed by government. The Court reasoned that:

The purpose of the Hospital Insurance Act is to provide particular services to the public. Although the benefits of that service are delivered and administered through private institutions – hospitals – it is the government, and not hospitals, that is responsible for defining both the content of the service to be delivered and the persons entitled to receive it ... in providing medically necessary services, hospitals carry out a specific governmental objective.^[207]

In other words, the Court saw the government as having sufficient control over the provision of certain medical services by hospitals – as a result of the fact that the government funded, legislated for, and defined the scope of the relevant services – as to require the private entities responsible for delivering the particular services to comply with the Charter in providing them. This finding did not, however, mean that the Charter applied to all of the activities of, or all medical services provided by, publicly funded hospitals.

The test that applies to private bodies performing governmental acts has been described in various ways. In Sagen, the Supreme Court of British Columbia referred to it as the ‘ascribed activity test’^[208] and in CBS, the Ontario Superior Court of Justice asked the narrower question of whether the body was ‘implementing a specific government policy or objective’.^[209] However, in each case the courts have essentially asked the same questions as they have under the broader control test that applies to entire organisations, including:

10. the source of the entity’s specific power to perform the impugned act;^[210]

11. who funds the activity;^[211]

12. the extent to which the powers are directly related to government policies and objectives;^[212] and

13. the extent to which government regulates or approves day-to-day aspects of the activity by issuing directives or approving operational matters.^[213]

The Ontario Superior Court of Justice distinguished the situation of CBS from that of the hospital in Eldridge because of the fact that there was no ‘specific government policy or program aimed at potential blood donors, the recruitment of blood donors, or the deferral of blood donors’.^[214] CBS simply had the authority, under an agreement between itself and provincial and federal governments to ensure the safety of the blood it supplied to hospitals. The agreement gave the CBS complete managerial discretion over ‘all operational blood system decisions, including matters of health and safety with respect to the blood supply system’.^[215] Accordingly CBS’ policy of precluding homosexual men from donating blood was not one that was under government control or forming part of a government program for the purposes of the Charter. By contrast, the British Columbia Supreme Court found that hosting the Olympics was an ‘inherently governmental function’ that could not be undertaken by any other entity, and accordingly held that VANOC was subject to the Charter in carrying out the activities of ‘planning, organizing, financing, and staging the 2010 games’.^[216] VANOC was held not to have violated the right to equality in section 15 of the Charter in failing to schedule women’s ski jumping, on the basis that it did not have the authority to include the event without the International Olympic Committee’s approval.^[217] However, we note the use which the Court made, in establishing that VANOC was subject to the Charter, of asking whether the challenged activity was something which only government could do. An indicium of some breadth, such as this, may be a helpful addition to the more detail-driven indicia discussed above.

3 Shortcomings/Criticisms of the ‘Control’ Test

As noted previously, the definition of ‘government’ in section 32 of the Charter is not an issue that has attracted much attention from either courts or commentators in recent years. While the topic attracted a great deal of commentary during the 1980s,^[218] only a handful of scholars has offered critiques of the Supreme Court’s approach to defining the Charter’s scope, predominantly in the immediate aftermath of Dolphin Delivery and McKinney. These range from general critiques of the public/private dichotomy that underpins the approach to more specific criticisms of the application of the control test. In the former category, Kanter argued that it is fundamentally and increasingly problematic to divide activity into categories of public and private.^[219] He suggest that ‘even the most “private” of decisions, such as a home-owner evicting an individual on ideological grounds, involves an exercise of state power’ because such actions are only valid because the law places a home owner’s property rights above those of the freedom of expression rights of the evicted individual. Kanter argues that the real question ought not be whether action is public or private but ‘whether the action is acceptable constitutionally’.^[220] This, however, is a test which begs the questions of what actions would be constitutionally acceptable and by what means we would be able to determine that issue. Kanter’s example of a private property owner being covered by the definition of ‘government’ for evicting a tenant on ideological grounds suggests that he foresaw a broader scope for the Charter than we are prepared to endorse. Furthermore, in the absence of a constitutional ranking of rights, it can hardly be considered an exercise of state power to consider rights confined to an individual personally (such as the right to restrict access to one’s property) as enforceable over rights which demand action from others (such as compelling a property owner to provide a tenant with access to that property). Ultimately, it becomes a matter for a court interpreting the Charter to determine whether a given action is ‘acceptable constitutionally’; without more, this does not provide a clear indication of how a court will reach its determination.

More recently, Henderson has criticised the control test as it has been applied to higher education institutions.^[221] Henderson has a number of complaints, many of which are specific to the Canadian Supreme Court’s factual findings regarding the autonomy of universities.^[222] His more general criticism of the control test is that it makes it too easy for governments to evade the application of the Charter. Despite the Supreme Court’s numerous statements to the effect that ‘to permit government to pursue policies violating Charter rights by means of contracts and agreements with other persons or bodies cannot be tolerated’,^[223] Henderson submits that under the control test the Charter will only apply to those contractors over whose activities the government retains a fairly high degree of oversight.^[224]

In our view this is precisely what makes the control test attractive in the context of section 75(v) of the Australian Constitution. It is only where the actions and decisions made within the private sector are subject to the direction of the Commonwealth, publicly funded and made in the furtherance of a government program or policy that the private sector body can be said to be acting as an ‘officer of the Commonwealth’. In situations like those in NEAT Domestic, where the Commonwealth Government chose to place a privately-funded, pre-existing private sector corporation in a decision-making position with the deliberate intent that the corporation exercise its power autonomously and in its own self-interest, the corporation cannot be said to be acting on behalf of the Commonwealth Government. With respect to the degree of direct government oversight or regulation required for a private sector employee exercising outsourced power to fall within the ambit of section 75(v), the control test can, of course, fairly easily be adapted to achieve a different balance between accountability and commercial freedom. In our view, a control-type test with similar factors to those used under the Charter offers an appropriate mechanism through which the High Court can reach such a balance.

One final shortcoming of the control test is the fact that it has not proven to be sufficient on its own in the Charter context. This became clear fairly early and as a result the Supreme Court has added a second test for when an entity might be ‘government’ for the purposes of section 32, which asks whether the entity is ‘governmental in nature’. The ‘governmental in nature’ test first arose in Godbout v Longueuil (City),^[225] in which the Court was asked whether the Charter applied to the city council. The Supreme Court was clearly of the opinion that the Charter did apply to the activities of the Council, yet the difficulty faced by the Court was that the council is not under the control of either the provincial or federal governments, nor does it implement programs on behalf of those governments. Yet La Forest J noted that exempting municipal government from the Charter would leave a substantial gap in Canada’s human rights framework.^[226] Furthermore, La Forest J suggested that municipal governments were ‘as a simple matter of fact – governmental in nature’.^[227] With respect, this conclusion must be correct: it would be astonishing if the definition of ‘government’ excluded an entire level of government, particularly one with which people have such significant contact. This is an example of where the Court may have been assisted by asking whether the body in question ‘does things which only government can do’.

In framing the new test that would justify the extension of the Charter to municipal governments,^[228] La Forest J reiterated his comment in McKinney that a public power or function test is not sufficient or appropriate in the context of the Charter.^[229] He distinguished between entities acting in a ‘governmental’ as opposed to a ‘merely “public” capacity’, though stated that the characteristics of the former ‘do not readily admit of any a priori elucidation’.^[230] The particular characteristics that led the Court to conclude that municipal governments are ‘governmental entities’ for the purposes of the Charter were: the fact that they are elected by the public and accountable to their electorates; their general taxing powers; and their ability to make, administer and enforce laws.^[231] Justice La Forest also noted that the law making authority of municipal governments is conferred on them by provincial governments,^[232] however it is doubtful based on the above-discussed case law that this alone would have been sufficient to bring them within the ambit of the Charter. However, it is possible that regulation-making is a distinct category of discretionary power for Charter purposes, the delegation of which automatically attracts Charter scrutiny.^[233]

Though in Godbout La Forest J made references to certain powers being quintessentially or inherently governmental,^[234] which has strong parallels with the Datafin test, he also made it clear that the functions and characteristics that will result in an entity being classified as ‘governmental in nature’ are much more confined than under the administrative law test. The more restricted scope of the additional test is also evident in the fact that it has thus far only been applied to local and municipal governments and amalgamations of those bodies.^[235]

While the application of two distinct tests for determining whether an entity is ‘government’ under the Canadian Charter has caused some confusion in subsequent cases,^[236] the same issue does not arise in the Australian administrative law context. The situation that led to the need to adopt a second test – the need to extend the definition of ‘government’ to municipal governments – does not arise in the context of section 75(v) as challenges to municipal government action are properly dealt with at first instance by state and territory supreme courts and only reach the High Court of Australia on appeal. Thus, there is no equivalent gap in Australia’s administrative law framework if the High Court does not have original jurisdiction over the activities of state and municipal governments as there would have been in the Canadian Charter context.

V THE CONTROL TEST AND ‘OFFICERS

OF THE COMMONWEALTH’

We submit that Canada’s ‘control’ test, which the Federal Court of Canada has recently indicated may form part of administrative law, but which is better developed in the context of the Charter, provides more useful guidance to the High Court of Australia in determining the scope of section 75(v) than the other tests currently used in administrative law. There are three major reasons for this. The first is that the control test is more closely aligned with the language of section 75(v) than any of the other available administrative law tests. Like section 75(v), the control test focuses on the identity of the decision-maker to define the scope of review, as opposed to the effect of decisions, nature of the function being performed, or source of the decision-maker’s powers. While the broader public function and natural justice tests may carry advantages from a policy perspective over the narrower control test, the fact that the Australian High Court’s review jurisdiction is limited by the express words contained in section 75(v) cannot be ignored. As argued above, to a large extent Australia’s ‘exceptionalism’ when it comes to judicial review of administrative action is attributable to the fact that judicial review is expressly entrenched within Australia’s Constitution. On one hand this entrenchment provides a much stronger and more certain basis on which the High Court has been able to protect its review role against intrusions by the legislature through privative clauses than would be available under common law. On the other, the use of express language limits the High Court’s jurisdiction in ways that the common law does not. It is not possible for the High Court to rely on the express language in section 75(v) for some purposes, while ignoring it for others. Thus, we submit that any test adopted by the High Court must be consistent with the language and intent of section 75(v). As we have argued above, public function tests, source tests and natural justice tests do not achieve the required consistency.

The second reason we contend that the control test is more appropriate in the context of section 75(v) than either a public function or natural justice test is that it is both narrower and more targeted in its scope. The control test is capable of being applied in a way that would achieve the accountability objectives of section 75(v), while simultaneously being more certain in its application and less ‘fuzzy’ than the notoriously oblique public function test.^[237] For instance, as has been the case in Canada, the control test would not extend the High Court’s review jurisdiction to genuine instances of privatisation.^[238] However, it would extend to incorporated bodies that remain under the purview of government, and would not permit government to avoid review simply by giving a decision-maker corporate status.^[239] With regard to outsourcing, the control test would also provide a narrow, but appropriate extension of section 75(v). Most bodies contractually connected to government would clearly not fall within the ambit of the control test, as government funding alone would not be sufficient to bring them within its terms. A greater degree of proximity to government would be required in order for outsourced functions to be subject to review, such as the discretionary functions being subject to the direction of government or undertaken in pursuance of a government program or scheme or in furtherance of government policy or objectives, as required by Eldridge. Thus, it would only be where a contractor forms part of the machinery of government in performing functions for government that the contractor would be subject to review with respect to those functions. This is a fairly narrow subset of contractors, which we submit is an appropriate balance between efficiency and accountability. Most importantly, it would not enable governments to avoid their public law responsibilities by outsourcing functions.

Finally, a control test would have the advantage of consistency with existing High Court precedent. A test along the same lines as that applied under the Canadian Charter would have captured the outsourced merits review functions impugned in the Offshore Processing Case, as the contractor followed ministerial directions and policies and was clearly acting in furtherance of a government policy. Thus, even if the High Court had not otherwise been able to construe the contractors’ merits review function as a part of the Minister’s exercise of statutory power, the applicants still would have been able to seek review under the High Court’s original jurisdiction.

Similarly, had NEAT Domestic been decided under section 75(v), the outcome would likely have been the same. The AWBI in that case was not making decisions under the direction of government, was not funded by government to make decisions and was not implementing government policy. This outcome will undoubtedly disappoint many of NEAT Domestic’s critics, but it is consistent with the general views about imposing public law obligations on private companies in the position of AWBI that were expressed by the majority of the High Court in that case.^[240] Tang would also likely have been decided in the same way had it come under the High Court’s original jurisdiction.^[241] On the basis of similar arguments to those put forward in McKinney, it is unlikely that Australian universities would meet the control test^[242] (though, again we make no argument as to whether they ought to be subject to review at the state level).

However, we note that the control test is unlikely to provide a total solution to defining who is an ‘officer of the Commonwealth’ under section 75(v) of the Constitution. Our argument is limited to the particular situation of identifying which private bodies exercising outsourced powers in a mixed administration should be subject to review in the High Court’s original jurisdiction. The control test will not be appropriate in defining in which other circumstances an individual may rightly be defined as an ‘officer of the Commonwealth’ under section 75(v). For instance, while federal judges are not in any sense under the ‘control’ of either the executive or parliament, and for various reasons do not fall within the definition of ‘government’ under the Charter, it is clear that they are ‘officers of the Commonwealth’ under section 75(v), and that this was one of the very purposes of the provision. This may be an issue which can at least partially be solved by asking whether judges ‘do things which only government (as opposed to individuals) can do’.

It is also important to note some other critical differences between the context in which the control test was developed and that in which we are arguing it should apply. The scope of the control test under Canadian Charter law is exceptionally narrow. The reluctance of Canadian courts to impose a broader public test in the Charter context is entirely warranted. The Charter places substantive limits on the exercise of power by the bodies to whom it applies and it should not readily be assumed that these substantive limits apply to the actions of private organisations. However, the purposes of administrative law are less ambitious, and its restrictions less onerous. Administrative law’s principles are designed to ensure that bodies exercising powers conferred on or delegated to them do so within the limits dictated by legislation, the common law and the Constitution. These limits are largely procedural in nature, rather than substantive. Accordingly, it may be appropriate to adopt a lower-threshold, or broader version of the test in some respects – for example by not requiring as strict control over the day-to-day operations of a body and settling for more general organisational control. Ultimately, the High Court will need to find the right balance; however, the control test provides a useful and adaptable framework that is consistent with both the objectives and language of section 75(v).

VI CONCLUSION

The argument we have made in this article is, ultimately, not a radical one. It proposes nothing more than that the High Court undertake a re-examination of the scope of the term ‘officer of the Commonwealth’ as it is employed in section 75(v) of the Constitution. The criteria for determining who falls within that term have been the same for nearly a hundred years, since Justice Isaacs’ judgment in R v Murray. Furthermore, those criteria have been described as ‘suspect then, and ... in even greater need of revision nowadays’,^[243] a conclusion with which we agree. There need be no great controversy attached to a reading of the phrase ‘officer of the Commonwealth’ that goes beyond the examples contained in Justice Isaacs’ judgment in R v Murray since, as we have shown above, this would better serve the goal of ensuring accountability that was instrumental in the drafting of section 75(v). Accountability that does not reach beyond the acts of government and its salaried employees has not been adequate for many years, since governments around the world first embraced outsourcing government functions to the private sector. It is fortunate that, in section 75(v), Australia has a mechanism for dealing with the challenges to accountability that may result from outsourcing. This is subject only to the preparedness of the High Court to use it.

---

[*] PhD Candidate, Monash University Faculty of Law.

[**] Lecturer, UNSW Faculty of Law, University of New South Wales. We owe our thanks to Mark Aronson, Matthew Groves and this journal’s two anonymous reviewers for their helpful comments. The usual caveat applies.

[1] To adopt the term coined by Mark Aronson to describe modern governance arrangements in which administration is shared between public and private organisations through the use of a range of forms, including privatisation and contracting: Mark Aronson, ‘A Public Lawyer’s Responses to Privatisation and Outsourcing’ in Michael Taggart (ed), The Province of Administrative Law (Hart Publishing, 1997) 40, 52.

[2] See generally Michael Taggart, ‘The Province of Administrative Law Determined?’ in Michael Taggart (ed), The Province of Administrative Law (Hart Publishing, 1997) 1; Margaret Allars, ‘Private Law but Public Power: Removing Administrative Law Review from Government Business Enterprises’ (1995) 6 Public Law Review 44; Nick Seddon, ‘The Interaction of Contract and Executive Power’ (2003) 31 Federal Law Review 541; Jody Freeman, ‘Private Parties, Public Functions, and the New Administrative Law’ (2000) 52 Administrative Law Review 813; Daniel Guttman, ‘Public Purpose and Private Service: The Twentieth Century Culture of Contracting Out and the Evolving Law of Diffused Sovereignty’ (2000) 52 Administrative Law Review 859.

[3] The term ‘outsource’ as used in this article is intended to describe the various legal relationships through which governments delegate discretionary power to the private sector, and is not restricted to formal modes of delegation, such as through contract. For a detailed analysis of how state power can be exercised by private entities, see Michael Taggart, ‘The Nature and Functions of the State’ in Peter Cane and Mark Tushnet (eds), The Oxford Handbook of Legal Studies (Oxford University Press, 2003) 101. The arguments made in this article are deliberately intended to apply to other legal and quasi-legal relationships that governments use to delegate power to the private sector, as well as to relationships that might develop in the future for this purpose: see, eg, Jon Michaels, ‘Privatization’s Progeny’ (2013) forthcoming 101 Georgetown Law Journal (cited with permission).

[4] For discussions, see Matthew Groves, ‘Outsourcing and s 75(v) of the Constitution’ (2011) 22 Public Law Review 3; Mark Aronson and Matthew Groves, Judicial Review of Administrative Action (Thomson Reuters, 5^th ed, 2013) 41–6 [2.150]–[2.160].

[5] Emphasis added.

[6] Plaintiff S157/2002 v Commonwealth [2003] HCA 2; (2003) 211 CLR 476, 513 [103] (Gaudron, McHugh, Gummow, Kirby and Hayne JJ).

[7] The point was not necessary for the Court to decide Plaintiff M61/2010E v Commonwealth [2010] HCA 41; (2010) 243 CLR 319 (‘Offshore Processing Case’). The point was not argued in NEAT Domestic Trading Pty Ltd v AWB Ltd [2003] HCA 35; (2003) 216 CLR 277 (‘NEAT Domestic’). Much criticism of the decision in NEAT Domestic was to the point that the majority of the Court could have expressed its views on the issue of outsourcing more clearly in obiter; see Chief Justice Gleeson’s judgment by way of example.

[8] See Groves, ‘Outsourcing and s 75(v) of the Constitution’, above n 4, 5–7.

[9] Ibid 5–6.

[10] See Rachel Harris, ‘Avoiding the Worst of All Worlds: Government Accountability for Outsourced Employment Services’ [2007] AIAdminLawF 15; (2007) 54 AIAL Forum 3.

[11] Groves, ‘Outsourcing and s 75(v) of the Constitution’, above n 4, 3. See also J J Spigelman, ‘The Centrality of Jurisdictional Error’ (2010) 21 Public Law Review 77, regarding the central importance of the Constitution to judicial review more generally.

[12] [1986] EWCA Civ 8; [1987] QB 815 (‘Datafin’).

[13] Aronson pointed out that, even in the United Kingdom, ‘Datafin’s application has generally been fairly cautious’: Mark Aronson, ‘Is the ADJR Act Hampering the Development of Australian Administrative Law?’ (2004) 15 Public Law Review 202, 210 n 78.

[14] [2010] HCA 41; (2010) 243 CLR 319, 350 [68]–[69].

[15] Administrative Review Council, Federal Judicial Review in Australia, Report No 50 (2012) 12.

[16] Canada Act 1982 (UK) c 11, sch B pt 1.

[17] See General Newspapers Pty Ltd v Telstra Corporation [1993] FCA 473; (1993) 45 FCR 164.

[18] Note, however, Julia Black’s point that courts routinely underestimate the true breadth of government power, including because they fail to recognise or accept that ‘contract is being used not as an instrument of exchange but of regulation’: Julia Black, ‘Constitutionalising Self-Regulation’ (1996) 59 Modern Law Review 24, 40. She argued that private law doctrines (such as restraint of trade and competition), which have economic powers and functions, are often used for this reason to perform regulatory functions, which they do ineffectively: at 42. One might use this as a basis to query why contractual relations should automatically be excluded from judicial review; with one small exception (see below n 43 and accompanying text), we do not challenge the orthodox position that contracts are the quintessential private law construct and should therefore fall outside the scope of judicial review.

[19] This recalls the comment of the then Solicitor-General for New South Wales, Keith Mason QC, in his submission to the Administrative Review Council that the subjection of the Crown to review in this area was an inversion of principle:

It seems to turn the modern approach that the Crown is entitled to no special treatment in the exercise of its private rights into a rule that the Crown is liable to suffer the worst of both worlds. If it enters into a commercial transaction it must comply with the law of contract and the law of tort in relation to pre-contractual negotiations, yet it is now said that it must also satisfy the public law requirements of administrative law. The two may not always be consistent.

Administrative Review Council, Review of the Administrative Decisions (Judicial Review) Act: the Ambit of the Act, Report No 32 (1989) 33 [127].

[20] Lord Nicholls argued that the extent to which a private body is ‘taking the place of ... government’ is relevant to deciding whether it is providing a ‘public function’ under s 6(3)(b) of the Human Rights Act 1998 (UK) c 42 (‘HRA’) but warned that ‘there is no single test of universal application’: Aston Cantlow and Wilmcote with Billesley Parochial Church Council v Wallbank [2004] 1 AC 546, 555 [12] (‘Aston Cantlow’).

[21] John Quick and Robert Randolph Garran, The Annotated Constitution of the Australian Commonwealth (Angus & Robertson, 1901) 779. This passage was approved of by Barton J in Ah Yick v Lehmert [1905] HCA 22; (1905) 2 CLR 593, 608–9 (‘Ah Yick’).

[22] (1803) 5 US [1803] USSC 16; (1 Cranch) 137. See Ah Yick [1905] HCA 22; (1905) 2 CLR 593, 609 (Barton J); R v Commonwealth Court of Conciliation and Arbitration; Ex parte Brisbane Tramways Co Ltd [1914] HCA 15; (1914) 18 CLR 54, 82 (Gavan Duffy and Rich JJ) (‘Tramways Case [No 1]’); Ruddock v Vadarlis [No 2] (2001) 115 FCR 229, 242–3 [31] (Black CJ and French J) (‘Tampa [No 2]’); Aronson and Groves, above n 4, 38 [2.120]. A brief history of the discussions relating to the clause that would become s 75(v) in the Convention debates appears in Leslie Zines, Cowen and Zines’s Federal Jurisdiction in Australia (Federation Press, 3^rd ed, 2002) 21–2.

[23] See Philip Morris Inc v Adam P Brown Male Fashions Pty Ltd (1981) 148 CLR 457, 507–8 (Mason J); Re McBain; ex parte Australian Catholic Bishops Conference [2002] HCA 16; (2002) 209 CLR 372, 388–9 (Gleeson CJ).

[24] Mark Leeming, Authority to Decide: The Law of Jurisdiction in Australia (Federation Press, 2012) 246.

[25] Quick and Garran, above n 21, 779.

[26] Re Refugee Review Tribunal; Ex parte Aala [2000] HCA 57; (2000) 204 CLR 82, 92 [18] (Gaudron and Gummow JJ) (‘Aala’).

[27] Ah Yick [1905] HCA 22; (1905) 2 CLR 593, 609 (Barton J). This is a description which has been queried more recently; see, eg, Keith Mason, ‘The Inherent Jurisdiction of the Court’ (1983) 57 Australian Law Journal 449. In relation to the High Court, Kirby J acknowledged the currency of the expressions ‘inherent jurisdiction’ and ‘inherent powers’, but argued that these were expressions which had been borrowed from England without consideration of their applicability to Australia’s constitutional needs: Batistatos v Roads and Traffic Authority (NSW) [2006] HCA 27; (2006) 226 CLR 256, 295–6. His Honour concluded that (at 296 [125]):

[a]ll Australian courts are created by, or under, legislation. Whatever the position in the United Kingdom, the additional jurisdiction and powers of Australian courts may not, therefore, truly be described as ‘inherent’. It may be more accurate to describe any supplementary jurisdiction or powers of such courts, including superior courts, as ‘implied’, that is implied in the constitutional or legislative source. According to this approach, a reference to ‘inherent jurisdiction’ or ‘inherent powers’ is likely to mislead.

The Commonwealth, on the other hand, may be characterised as having ‘inherent’ executive power: see Leslie Zines, ‘The Inherent Executive Power of the Commonwealth’ (2005) 16 Public Law Review 279.

[28] Bank of New South Wales v Commonwealth [1948] HCA 7; (1948) 76 CLR 1, 363 (‘Bank Nationalisation Case’); James Stellios, ‘Exploring the Purposes of Section 75(v) of the Constitution’ [2011] UNSWLawJl 3; (2011) 34 University of New South Wales Law Journal 70. Stellios stated that the two sub-sections were ‘symbiotically linked’: at 83, but was unable to draw any firm conclusions from that point given that it had always been ‘assumed but not explored’ by the High Court: at 92.

[29] Deputy Commissioner of Taxation of The Commonwealth of Australia v Richard Walter Pty Ltd [1995] HCA 23; (1995) 183 CLR 168, 179 (Mason CJ). In one sense, of course, the fact that s 75(v) describes an ‘officer of the Commonwealth’, rather than just ‘the Commonwealth’ as in s 75(iii), does extend the coverage of s 75(v) beyond the limits of s 75(iii), although this is attributable at least in part to the American jurisprudence which held that the doctrine that ‘the United States cannot be sued unless provision has been made by Congress, is limited to suits against the United States directly and by name; and that this plea cannot be raised by officers or agents of the government’: Quick and Garran, above n 21, 773. See Aronson and Groves, above n 4, 40 [2.140] n 126.

[30] Stellios, above n 28.

[31] Accountability of government to its citizens is now considered a core theme of administrative law; see, eg, Robin Creyke and John McMillan, Control of Government Action: Text, Cases and Commentary (Lexis Nexis Butterworths, 3^rd ed, 2012) ch 1.

[32] Official Record of the Debates of the Australasian Federal Convention, Melbourne, 4 March 1898, 1885, cited in Stellios, above n 28, 81.

[33] Ibid 81–2.

[34] Ibid 70–2.

[35] [1910] HCA 33; (1910) 11 CLR 1 (‘Ex parte Whybrow’).

[36] Aronson and Groves, above n 4, 38 [2.120].

[37] Ex parte Whybrow [1910] HCA 33; (1910) 11 CLR 1, 41 (O’Connor J).

[38] Ibid [1910] HCA 33; (1910) 11 CLR 1, 22 (Griffith CJ), 33 (Barton J), 42 (O’Connor J).

[39] See, eg, Edwards v Santos Ltd (2011) 242 CLR 421. Federal constitutional reasoning is now also used by state supreme courts in reviewing inferior state courts’ decisions: Kirk v Industrial Relations Commission of New South Wales (2010) 239 CLR 531 (‘Kirk’).

[40] Bank Nationalisation Case [1948] HCA 7; (1948) 76 CLR 1, 363.

[41] Ibid (emphasis added).

[42] Ibid.

[43] [2005] FCA 549; (2005) 143 FCR 217 (‘S v DIMIA’).

[44] Matthew Groves, ‘Outsourcing and Non-Delegable Duties’ (2005) 16 Public Law Review 265, 265.

[45] Ibid 268–9.

[46] See Aronson and Groves, above n 4, 25 [2.20].

[47] Groves, ‘Outsourcing and Non-Delegable Duties’, above n 44, 267.

[48] Taggart pointed out that the term ‘privatisation’ has a broader application in the United States: see Taggart, ‘The Nature and Functions of the State’, above n 3, 105.

[49] Which, technically, was re-privatised in the 1990s, having been nationalised by the Chifley Government after World War II.

[50] Taggart, ‘The Nature and Functions of the State’, above n 3, 109.

[51] Though there are a number of Federal Court decisions holding that bodies corporate cannot be ‘officers of the Commonwealth’: see, eg, Post Office Agents Association Ltd v Australian Postal Commission (1988) 84 ALR 563, 575 (Davies J); Broken Hill Proprietary Co Ltd v National Companies and Securities Commission [No 6] (1986) 61 ALJR 124, 127 (Dawson J); Businessworld Computers Pty Ltd v Australian Telecommunications Commission [1988] FCA 206; (1988) 82 ALR 499, 500 (Gummow J). Aronson and Groves list a more comprehensive series of such restrictive interpretations of s 75(v): see Aronson and Groves, above n 4, 43–4 [2.160]. We join with them in hoping that these decisions ‘will soon be forgotten’: at 44 [2.160].

[52] [1916] HCA 58; (1916) 22 CLR 437 (‘R v Murray’).

[53] Aronson and Groves noted that ‘[t]he policy for excluding state personnel performing judicial functions seems to have been a desire not to countenance federal interference with the state judicial structures, but the judgments focused on a textual approach to the Constitution’: Aronson and Groves, above n 4, 43 [2.150].

[54] R v Murray [1916] HCA 58; (1916) 22 CLR 437, 452 (emphasis added). Justice Higgins also relied on the fact that Judge Murray was appointed, paid and removable by and ultimately responsible to the state of New South Wales and not the Commonwealth: at 464. Aronson and Groves concluded that ‘the criteria proposed by Isaacs J in 1916 for identifying “officers of the Commonwealth” were suspect then, and are in even greater need of revision nowadays’: Aronson and Groves, above n 4, 45 [2.160]. There is some irony to the fact that, in the Convention Debates, Isaacs J had led the opposition to adopting the clause which ultimately became s 75(v): see Zines, Cowen and Zines’s Federal Jurisdiction in Australia, above n 22, 47.

[55] [2010] HCA 41; (2010) 243 CLR 319, 345 [51] (French CJ, Gummow, Hayne, Heydon, Crennan, Kiefel and Bell JJ).

[56] [2003] HCA 35; (2003) 216 CLR 277, 310–11 (citations omitted).

[57] See Stellios, above n 28, 80–1.

[58] Ibid 91.

[59] Ibid 92.

[60] See, eg, Leslie Zines, The High Court and the Constitution (Federation Press, 5^th ed, 2008).

[61] See, eg, Leeming, above n 24; Zines, Cowen and Zines’s Federal Jurisdiction in Australia, above n 22.

[62] Groves noted in 2005 that the debate had ‘waned sharply in recent years’: Groves, ‘Outsourcing and Non-Delegable Duties’, above n 44, 265. This may indicate simply that there are fewer public assets left to privatise, or that we are collectively becoming used to the outsourcing of government activities. Groves was certainly not arguing, and nor do we, that the difficulties which arise from outsourcing have abated or are no longer in need of a solution.

[63] [2012] HCA 23; (2012) 86 ALJR 713 (‘School Chaplains Case’).

[64] See Quick and Garran, above n 21, 951–3. There is no commentary on the phrase ‘office ... under the Commonwealth’ and the focus of this provision seems always to have been a substantive one of religious freedom rather than jurisdictional, as in s 75(v).

[65] School Chaplains Case [2012] HCA 23; (2012) 86 ALJR 713, 721 [9] (French CJ), 745 [107]–[110] (Gummow and Bell JJ), 754 [168] (Hayne J), 808 [442]–[445], 812 [476] (Crennan J), 831 [597] (Kiefel J). Heydon J dealt with s 116 by querying whether the ‘chaplains’ in question were in fact chaplains as that term is properly understood in religious parlance: at 780–1 [305]–[307].

[66] Ibid 745 [110].

[67] See also School Chaplains Case [2012] HCA 23; (2012) 86 ALJR 713, 808 [443]–[444] (Heydon J).

[68] Michael Taggart, ‘“Australian Exceptionalism” in Judicial Review’ [2008] FedLawRw 1; (2008) 36 Federal Law Review 1.

[69] Thomas Poole, ‘Between the Devil and the Deep Blue Sea: Administrative Law in an Age of Rights’ in Linda Pearson, Carol Harlow and Michael Taggart (eds), Administrative Law in a Changing State: Essays in Honour of Mark Aronson (Hart Publishing, 2008) 15.

[70] As was Griffith University v Tang [2005] HCA 7; (2005) 221 CLR 99 (‘Tang’). See the list in Mark Aronson, ‘Private Bodies, Public Power and Soft Law in the High Court’ [2007] FedLawRw 1; (2007) 35 Federal Law Review 1, 2; cf Justice P A Keane, ‘Judicial Review: The Courts and the Academy’ (2008) 82 Australian Law Journal 623.

[71] ‘Constitutionalisation’ is a term which has received considerable currency of late; see, eg, Stephen Gageler, ‘The Constitutionalisation of Australian Administrative Law’ (Speech delivered at the Kirby Seminar, University of New England School of Law, 14 March 2011); Ronald Sackville, ‘The Constitutionalisation of State Administrative Law’ (2012) 19 Australian Journal of Administrative Law 127.

[72] Taggart, ‘“Australian Exceptionalism” in Judicial Review’, above n 68, 23.

[73] It is interesting to note the dictum of Lord Mance that ‘[i]n Datafin, the panel was as a matter of fact entrusted with an extensive and vital regulatory role in the public interest, and that was sufficient to make it susceptible to judicial review’: YL v Birmingham City Council [2008] 1 AC 95, 131 (emphasis added).

[74] His Honour was therefore not in dissent; cf Keane, above n 70, 626.

[75] We do not take this view and agree with the remarks made by Aronson, ‘Private Bodies, Public Power’, above n 70, 15–16.

[76] A recent comprehensive review of the ADJR Act resulted in very few recommendations for change: see Administrative Review Council, above n 15.

[77] Poole, above n 69, 26 (citations omitted).

[78] Taggart took some pains to unpack what was contained in his use of formalism as a ‘catch-all term’: Taggart, ‘“Australian Exceptionalism” in Judicial Review’, above n 68, 7.

[79] Mark Aronson, ‘Process, Quality, and Variable Standards: Responding to an Agent Provocateur’ in David Dyzenhaus, Murray Hunt and Grant Huscroft (eds), A Simple Common Lawyer: Essays in Honour of Michael Taggart (Hart Publishing, 2009) 5, 23.

[80] Ibid 23–4.

[81] We refer again to the commentary on these cases by Aronson, ‘Private Bodies, Public Power’, above n 70.

[82] Although his subsequent accusation that the High Court is politically partisan was regrettable inasmuch as it seems to have been completely unsupported: Poole, above n 69, 33.

[83] While his argument is unconvincing inasmuch as it seeks to deny that there was anything at issue before the High Court beyond the tactics adopted by the opposing sets of counsel (and therefore that any criticism of the High Court’s decision was warranted), Justice Keane’s frustration with some of the critics of Tang who ignored altogether the way the case was argued is palpable and at least somewhat justified: see, eg, Keane, above n 70, 626 n 13.

[84] Although we might well ask from what Australia is really being isolated, since Poole argues that the ‘development of a “common law of judicial review” grounded in human rights ... is not likely to produce anything like a normatively unified jurisprudence’: Poole, above n 69, 22.

[85] Taggart, ‘“Australian Exceptionalism” in Judicial Review’, above n 68, 1.

[86] [1986] EWCA Civ 8; [1987] QB 815, 838 (Donaldson MR), 847–8 (Lloyd LJ), 852 (Nicholls LJ). For discussions of this case see Aronson and Groves, above n 4, 128–46 [3.130]–[3.200]; Paul Craig, Administrative Law (Sweet & Maxwell, 7^th ed, 2012) 850–6; Sir William Wade and Christopher Forsyth, Administrative Law (Oxford University Press, 10^th ed, 2009) 542–4.

[87] See, eg, Aronson and Groves, above n 4, 143 [3.190]; Justice Emilios Kyrou, ‘Judicial Review of Decisions of Non-Governmental Bodies Exercising Governmental Powers: Is Datafin Part of Australian Law?’ (2012) 86 Australian Law Journal 20, 24.

[88] Electoral Commission v Cameron [1997] NZCA 301; [1997] 2 NZLR 421.

[89] For a discussion of the place of Datafin in Canadian administrative law, see below nn 144–8 and accompanying text.

[90] Kyrou, above n 87.

[91] Albeit only once, in Masu Financial Management Pty Ltd v Financial Industry Complaints Service Ltd [No 2] (2004) 50 ACSR 554, 559–60 (Shaw J) (‘Masu [No.2]’). Given the vintage of the other cases in which Datafin has been approved but not applied by state and territory supreme courts (see below nn 92–5), Masu [No.2] looks to be the high water mark for judicial acceptance of Datafin in Australia.

[92] Norths Ltd v McCaughan Dyson Capel Cure Ltd (1988) 12 ACLR 739, 745–6 (Young J); Typing Centre of New South Wales v Toose (Unreported, Supreme Court of New South Wales, Mathews J, 15 December 1988) 17–20; McClelland v Burning Palms Surf Life Saving Club [2002] NSWSC 470; (2002) 191 ALR 759, 790–1 [115]–[117] (Campbell J).

[93] Victoria v Master Builders’ Association of Victoria [1995] VicRp 47; [1995] 2 VR 121, 137, 140 (Tadgell J), 148–9 (Ormiston J), 152, 161, 163–4 (Eames J). There is also a detailed discussion, although limited to certiorari, in D’Souza v Royal Australian and New Zealand College of Psychiatrists [2005] VSC 161; (2005) 12 VR 42, 55–9 [99]–[118] (Ashley J).

[94] MBA Land Holdings Pty Ltd v Gungahlin Development Authority [2000] ACTSC 89; (2000) 206 FLR 120, 147 [220] (Higgins J).

[95] Whitehead v Griffıth University [2002] QSC 153; [2003] 1 Qd R 220, 225 (Chesterman J).

[96] Chase Oyster Bar Pty Ltd v Hamo Industries Pty Ltd [2010] NSWCA 190; (2010) 78 NSWLR 393, 412–13 (Basten JA, with whom Spigelman CJ agreed).

[97] Khuu & Lee Pty Ltd v Corporation of the City of Adelaide (2011) 110 SASR 235, 242 [26] (Vanstone J, with whom Sulan and Peek JJ agreed).

[98] Mickovski v Financial Ombudsman Service Ltd [2012] VSCA 185; (2012) 91 ACSR 106, 116 (Buchanan and Nettle JJA, Beach AJA).

[99] Cf Kyrou, above n 87.

[100] Datafin [1986] EWCA Civ 8; [1987] 1 QB 815, 824 (Sir John Donaldson MR).

[101] This perhaps explains why Craig pays such scant attention to the Datafin decision in the most recent edition of his text: Craig, Administrative Law, above n 86, 850–6. We would argue that Datafin can only properly be understood in the context of the policies of the Conservative Government led by Margaret Thatcher; see also Peter Cane, Administrative Law (Oxford University Press, 5^th ed, 2011) 15–16. Others have argued the more general point that the ‘blurring of the line between purely public and purely private’ has either remained or intensified since the election of a Labour Government in 1997: see Daithí Mac Síthigh, ‘Datafin to Virgin Killer: Self-Regulation and Public Law’ (Working Paper No NLSWP 09/02, University of East Anglia, Norwich Law School, 8 April 2009) 3. We make no specific response other than to say that this is more likely to be a sign of the times and common to most western democracies; see, eg, Taggart, ‘The Nature and Functions of the State’, above n 3.

[102] It is significant that the Court of Appeal did not find for Datafin Plc on the substantive issues of its complaint.

[103] See, eg, Crimmins v Stevedoring Industry Finance Committee [1999] HCA 59; (1999) 200 CLR 1; Brodie v Singleton Shire Council [2001] HCA 29; (2001) 206 CLR 512; Graham Barclay Oysters Pty Ltd v Ryan [2002] HCA 54; (2002) 211 CLR 540; Mark Aronson, ‘Government Liability in Negligence’ [2008] MelbULawRw 2; (2008) 32 Melbourne University Law Review 44.

[104] Lange v Australian Broadcasting Corporation [1997] HCA 25; (1997) 189 CLR 520, 563–4 (Brennan CJ, Dawson, Toohey, Gaudron, McHugh, Gummow and Kirby JJ); Lipohar v The Queen (1999) 200 CLR 485, 505 (Gaudron, Gummow and Hayne JJ); Kirk (2010) 239 CLR 531, 581 [99] (French CJ, Gummow, Hayne, Crennan, Kiefel and Bell JJ).

[105] Kirk (2010) 239 CLR 531, 589 [121] (Heydon J).

[106] For a discussion, see Matthew Groves, ‘Should We Follow the Gospel of the Administrative Decisions (Judicial Review) Act 1977 (Cth)?’ [2010] MelbULawRw 26; (2010) 34 Melbourne University Law Review 736, 763–6.

[107] For discussions see Paul Craig, ‘Contracting Out, the Human Rights Act and the Scope of Judicial Review’ (2002) 118 Law Quarterly Review 551; Colin D Campbell, ‘The Nature of Power as Public in English Judicial Review’ (2009) 68 Cambridge Law Journal 90; Aronson and Groves, above n 4, 130 [3.140]; Craig, Administrative Law, above n 86, 850–62; Wade and Forsyth, above n 86, 119-34.

[108] Craig, Administrative Law, above n 86, 850.

[109] Ibid 849.

[110] See Council of Civil Service Unions v Minister for the Civil Service [1985] 1 AC 374.

[111] See O’Reilly v Mackman [1983] UKHL 1; [1983] 2 AC 237.

[112] Craig, Administrative Law, above n 86, 849.

[113] The protected rights under the European Convention on Human Rights are set out in sch 1 to the HRA.

[114] HRA s 6(3)(b).

[115] Paul Craig made a similar point in a different context: Craig, Administrative Law, above n 86, 850.

[116] YL v Birmingham City Council [2008] 1 AC 95, 131 (Lord Mance).

[117] Aston Cantlow [2004] 1 AC 546, 555 [11] (Lord Nicholls).

[118] Ibid 566 [52].

[119] Victorian Charter s 4(1)(c).

[120] Ibid s 4(2)(b).

[121] Ibid s 4(2)(c).

[122] Ibid s 4(2)(d).

[123] The absence of judicial authority on these provisions was noted in Victorian Government, Submission No 324 to the Scrutiny of Acts and Regulations Committee, Review of the Charter of Human Rights and Responsibilities Act 2006, 2011, 64–5.

[124] Metro West Housing Services Ltd v Sudi (Residential Tenancies) [2009] VCAT 2025, [46].

[125] Ibid [166]

[126] Scrutiny of Acts and Regulations Committee, Parliament of Victoria, Review of the Charter of Human Rights and Responsibilities Act 2006 (2011) 108 [recommendation 22]. The Victorian Government has indicated that this recommendation is ‘under review’: Victorian Government, Response to Scrutiny of Acts and Regulations Committee Review of the Charter of Human Rights and Responsibilities Act 2006 (14 March 2012) 28.

[127] See Janina Boughey, ‘Rights, Review and Reasonableness: The Implications of Canada’s New Approach to Administrative Decision-Making and Human Rights for Australia’ (2013) forthcoming 35 Sydney Law Review.

[128] Lesley Brown (ed), The New Shorter Oxford English Dictionary (4^th ed, 1993) 1983–4 (‘OED’).

[129] Ibid.

[130] Although not the company itself on the current orthodoxy: see above n 50.

[131] Indeed, individual employees who perform those tasks may also be ‘officers of the Commonwealth’, since they are performing duties or functions in a position of authority or trust, and are doing so in the position of the Commonwealth. It would be necessary to identify the relevant decision-maker.

[132] Cf Antonin Scalia and Bryan A Garner, Reading Law: the Interpretation of Legal Texts (Thomson/West, 2012) 33. Justice Scalia is one of the foremost figures in the United States advocating that statutes and the Constitution should only be read as originally intended. We understand his Honour to believe, essentially, that the sin which is thereby avoided is a misunderstanding of the drafter’s intentions. In our view, a broad reading of a word like ‘officer’ is precisely what was intended by the drafters of the Australian Constitution, since it gives effect to their intention that the Commonwealth be held accountable.

[133] See, eg, academic commentary following Griffith University v Tang [2005] HCA 7; (2005) 221 CLR 99.

[134] Administrative Review Council, above n 15, 12.

[135] Judicial Review Act 1991 (Qld) ss 4(b), 9.

[136] See discussion in Groves, ‘Should We Follow the Gospel of the Administrative Decisions (Judicial Review) Act 1977 (Cth)?’, above n 106, 753–4.

[137] Bituminous Products Pty Ltd v General Manager (Road System and Engineering), Department of Main Roads [2005] QSC 42; [2005] 2 Qd R 344, 351. It is noteworthy that Tang [2005] HCA 7; (2005) 221 CLR 99 arose under the Judicial Review Act 1991 (Qld), however, the plaintiff did not raise the argument that Griffith University had made the impugned decision under a ‘scheme or program’. Chief Justice Gleeson made obiter comments to the effect that the Queensland legislation would not extend to decisions made by the University (at 118–19), while Justice Kirby suggested that the fact that the Queensland Parliament had deliberately expanded the scope of review under the Judicial Review Act 1991 (Qld) justified the Court taking a broad interpretation to its other jurisdictional provisions: at 146.

[138] New South Wales Department of Justice and Attorney General, Reform of Judicial Review in NSW (2011); Matthew Groves, ‘Should the Administrative Law Act 1978 (Vic) Be Repealed?’ [2010] MelbULawRw 15; (2010) 34 Melbourne University Law Review 452, 459.

[139] New South Wales Department of Justice and Attorney General, above n 138, 28.

[140] These include sporting clubs, trade unions and professional bodies. See Aronson and Groves, above n 4, 487–8 [7.410].

[141] Though Mark Aronson has expressed doubt as to whether it is an appropriate statutory test, arguing that the natural justice test both over and under reaches in various respects: Mark Aronson, Submission No 1 to Administrative Review Council, Inquiry on Judicial Review in Australia, 13 May 2011, 7.

[142] Kyrou, above n 87.

[143] Groves, ‘Should the Administrative Law Act 1978 (Vic) Be Repealed?’, above n 138, 465.

[144] For acceptance by Canadian provincial and territorial courts see Vander Zalm v British Columbia (Commissioner of Conflict of Interest) (1991) 80 DLR (4^th) 291, 297–8 (in which the Court expressly assumed but did not decide the correctness of Datafin); Masters v Ontario (1993) 16 OR (3d) 439; Volker Stevin NWT Ltd v Northwest Territories (Commissioner) [1994] NWTR 97, [25]–[26] (citing the application of Datafin in Zalm and Masters as authority for applying a public function/machinery of government test to a non-statutory committee administering the Government of the Northwest Territories ‘Business Incentive Policy’); McDonald v Anishinabek Police Service (2006) 83 OR (3d) 132, 155–6 [70]–[76] (in which the Court apparently endorsed Datafin, and applied it to hold that natural justice was owed in a decision by the Police Chief to expel a First Nations Constable); Scheerer v Waldbillig [2006] OJ No 744, [18]–[21] (relying on Masters v Ontario as authority for the application of a public function test, holding that a hospital’s Medical Director was exercising a public duty subject to the court’s judicial review jurisdiction); Reynolds v Ontario (Registrar, Information and Privacy Commissioner) (2006) 217 OAC 146, [33]–[37] (in which the Court applied Datafin); Knox v Conservative Party of Canada (2007) 422 AR 29.

[145] Datafin [1986] EWCA Civ 8; [1987] QB 815, 838–9 (Donaldson MR), 847 (Lloyd LJ), 852 (Nicholls LJ).

[146] Société de l’assurance Automobile du Québec v Cyr [2008] 1 SCR 338, 348 (Bastarache J) (though this particular case dealt with the Act Respecting Administrative Justice, RSQ 1996, c J3, rather than the common law).

[147] See, eg, McKinney v University of Guelph [1990] 3 SCR 229 (discussed in detail below), where La Forest J (for Dickson CJ and Gonthier J) implicitly seemed to endorse a public function test for judicial review of administrative action: at 267–8 [33]–[34]

[148] Although the Canadian Constitution has been found to entrench judicial review of administrative action, the Supreme Court’s jurisdiction is not located in an express constitutional provision, but is instead implied from other constitutional provisions: see Peter W Hogg, Constitutional Law of Canada, (Thomson/Carswell, 5^th ed, 2010) vol 1, [7-52]–[7-57].

[149] Air Canada v Toronto Port Authority [2011] FCA 347, [60] (Federal Court of Canada) (‘Air Canada’); Attawapiskat First Nation v Canada (Minister for Aboriginal Affairs and Northern Development) [2012] FC 948, [55].

[150] Air Canada [2011] FCA 347, [60] (Federal Court of Canada).

[151] Federal Courts Act, RSC 1985, c F-7.

[152] Wilcox v Canadian Broadcasting Corporation [1980] 1 FC 326, 329 (Thurlow ACJ).

[153] In Canada (Citizenship and Immigration) v Khosa [2009] 1 SCR 339, the majority (Binnie J for McLachlin CJ, LeBel, Abella and Charron JJ) stated that the Federal Court’s jurisdiction under s 18 of the Federal Courts Act, RSC 1985, c F-7 (at 363 [28]):

must be sufficiently elastic to apply to the decisions of hundreds of different ‘types’ of administrators, from Cabinet members to entry-level fonctionnaires, who operate in different decision-making environments under different statutes with distinct grants of decision-making powers. Some of these statutory grants have privative clauses; others do not. Some provide for a statutory right of appeal to the courts; others do not.

[154] Air Canada [2011] FCA 347, [60] (Federal Court of Canada).

[155] There are rights protection Acts in two jurisdictions within Australia: Human Rights Act 2004 (ACT); Charter of Human Rights and Resposibilities Act 2006 (Vic).

[156] Barak points out that in developing the implied constitutional right of freedom of political communication (until recently thought to be dormant but now re-defined as a limitation; see Monis v The Queen (2013) 87 ALJR 340), the High Court took guidance from cases like R v Oakes [1986] 1 SCR 103: Aharon Barak, Proportionality: Constitutional Rights and Their Limitations (Cambridge University Press, 2012) 195–6.

[157] Eldridge v British Columbia (Attorney General) [1997] 3 SCR 624, 661 [44] (La Forest J for Lamer CJ, L’Heureux-Dubé, Sopinka, Gonthier, Cory, McLachlin, Iacobucci and Major JJ) (‘Eldridge’).

[158] McKinney v University of Guelph [1990] 3 SCR 229, 265–7 (La Forest J for Dickson CJ and Gonthier J) (‘McKinney’).

[159] [1990] 3 SCR 451.

[160] [1990] 3 SCR 483 (‘Stoffman’).

[161] [1990] 3 SCR 570 (‘Douglas College’).

[162] Some of the cases also involved claims under provincial anti-discrimination legislation, however those aspects are not relevant to the discussion in this article.

[163] McKinney [1990] 3 SCR 229, 262 [22]. L’Heureux-Dubé J and Sopinka J expressed broad agreement with La Forest J on the appropriate test in separate judgments: at 419–20 [363] and 444–5 [423] respectively. The non-application of the Charter to private action had been established by the Court four years earlier in Dolphin Delivery v RWDSU [1986] 2 SCR 573 (‘Dolphin Delivery’) (in which the Supreme Court decided that the Charter did not apply to private litigation). However, that case did not require the court to develop a test for what ‘government’ was.

[164] McKinney [1990] 3 SCR 229, 264–5 [28]. A variant of the former test had been applied by the Ontario Divisional Court in Klein v Law Society of Upper Canada (1985) 50 OR (2d) 118, and the latter by the Alberta Court of Appeal in R v Lerke (1986) 43 Alta LR (2d) 1.

[165] McKinney [1990] 3 SCR 229, 265–6 [30] (La Forest J).

[166] NEAT Domestic [2003] HCA 35; (2003) 216 CLR 277, 298 (McHugh, Hayne and Callinan JJ).

[167] Ibid 297–8 (McHugh, Hayne and Callinan JJ). Counsel for NEAT (S J Gageler SC) had conceded in argument that ‘as a private corporation, AWBI has plenary power under the Corporations Law and, so far as that extends, that it may act in its own self-interest’: at 278.

[168] NEAT Domestic [2003] HCA 35; (2003) 216 CLR 277, 290 [27]. Poole is highly critical of the outcome in NEAT Domestic (in which he is hardly alone) and accused Gleeson CJ of defending that outcome with ‘customary astringent elegance’: Poole, above n 69, 26. We doubt that a compliment was intended (although we agree that Chief Justice Gleeson’s judgment was characteristically elegant).

[169] McKinney [1990] 3 SCR 229, 268 [34].

[170] Ibid.

[171] See Doré v Barreau du Québec [2012] 1 SCR 395, 413–7 [26]–[35].

[172] McKinney [1990] 3 SCR 229, 269 [35].

[173] Ibid 270–1 [36].

[174] Wilson J actually proposed a broad test which used the ‘public function’, source of power and ‘control’ tests as factors that would each indicate when the Charter should apply to a particular decision-maker.

[175] McKinney [1990] 3 SCR 229, 342–57 [183]–[216].

[176] Ibid 343 [188].

[177] In his analysis of Tang, Professor Aronson conducted a similar audit of the functions of universities in making the point that it was ‘disappointing’ that the High Court had effectively assumed that the power exercised by universities is not ‘public’: Aronson, ‘Private Bodies, Public Power and Soft Law in the High Court’, above n 70, 14–15.

[178] McKinney [1990] 3 SCR 229, 271–2 [38].

[179] Ibid 272 [39].

[180] Ibid [40].

[181] Ibid 273 [41].

[182] Ibid [42].

[183] Ibid 265–7 [29]–[32].

[184] Ibid 275 [45].

[185] See Douglas College [1990] 3 SCR 570; Lavigne v Ontario Public Service Employees Union [1991] 2 SCR 211.

[186] Douglas College [1990] 3 SCR 570, 584 [16] (La Forest J for Dickson CJ and Gonthier J). Justice Sopinka agreed with La Forest J on this issue: at 616 [92]); Wilson and L’Heureux-Dubé JJ, with whom Cory J agreed, came to the same conclusion as to the outcome of the case but applied Justice Wilson’s three part test, combining a function question, a control question and a source question, to reach the result.

[187] Douglas College [1990] 3 SCR 570, 584 [16] (La Forest J for Dickson CJ and Gonthier J).

[188] Stoffman [1990] 3 SCR 483, 513–4 [39] (La Forest J for Dickson CJ and Gonthier J, with whom Sopinka J agreed on this point).

[189] Ibid.

[190] [2009] 2 SCR 295.

[191] Ibid 311–12 [20]–[21] (Deschamps J for McLachlin CJ and Binnie, LeBel, Abella, Charron and Rothstein JJ).

[192] Canadian Federation of Students v Greater Vancouver Transportation Authority (2006) 275 DLR (4^th) 221, [86]–[87] (Prowse JA, Lowry JA concurring).

[193] Ibid [88]–[89].

[194] Earlier in her Honour’s judgment Prowse JA found that the Greater Vancouver Regional District (‘GRVD’), a partnership of the 21 municipal governments in the Greater Vancouver area, was itself ‘government’: ibid [81]–[85].

[195] Canadian Federation of Students v Greater Vancouver Transportation Authority [2006] 275 DLR (4^th) 221, [89]–[91].

[196] Ibid [91].

[197] Greater Vancouver Transport Authority v Canadian Federation of Students – British Columbia Component [2009] 2 SCR 295, 312 [22] (Deschamps J, for McLachlin CJ and Binnie, LeBel, Abella, Charron and Rothstein JJ).

[198] Ibid.

[199] (2009) 98 BCLR (4^th) 109 (‘Sagen’).

[200] [2010] 217 CRR (2d) 153 (‘CBS’).

[201] Sagen (2009) 98 BCLR (4^th) 109, [22]–[40].

[202] CBS [2010] 217 CRR (2d) 153, [351]–[359].

[203] McKinney [1990] 3 SCR 229, 274 [42].

[204] Eldridge [1997] 3 SCR 624, 660 [42] (La Forest J for Lamer CJ, L’Heureux-Dubé, Sopinka, Gonthier, Cory, McLachlin, Iacobucci and Major JJ), quoting McKinney [1990] 3 SCR 229, 265 (La Forest J).

[205] Ibid 661–2 [44].

[206] Ibid 660–1 [43].

[207] Ibid 663–4 [49]–[50].

[208] Sagen (2009) 98 BCLR (4^th) 109, [11].

[209] CBS (2010) 217 CRR (2d) 153, [360].

[210] Eldridge [1997] 3 SCR 624, 664–5 [50]; CBS (2010) 217 CRR (2d) 153, [361]–[362]; Sagen (2009) 98 BCLR (4^th) 109, [60]–[62].

[211] Eldridge [1997] 3 SCR 624, 664 [49]; CBS (2010) 217 CRR (2d) 153, [363]; Sagen (2009) 98 BCLR (4^th) 109, [59].

[212] Eldridge [1997] 3 SCR 624, 664–5 [50]; CBS (2010) 217 CRR (2d) 153, [364]; Sagen (2009) 98 BCLR (4^th) 109, [55]–[57].

[213] Eldridge [1997] 3 SCR 624, 663–4 [49]; CBS (2010) 217 CRR (2d) 153, [365]; Sagen (2009) 98 BCLR (4^th) 109, [58].

[214] CBS (2010) 217 CRR (2d) 153, [367].

[215] Ibid.

[216] Sagen (2009) 98 BCLR (4^th) 109, [65].

[217] Ibid [127]–[129].

[218] See, eg, Peter W Hogg, ‘The Dolphin Delivery Case: The Application of the Charter to Private Action’ (1986–87) 51 Saskatchewan Law Review 273; Dale Gibson, ‘Distinguishing the Governors from the Governed: The Meaning of “Government” Under Section 32(l) of the Charter’ (1983) 13 Manitoba Law Journal 505; Gordon P Crann, ‘How Far Does the Charter Reach? A Theoretical Review of the Section 32(1) Debate and Canada’s Emerging “Governmental Action” Doctrine’ (1988) 47 University of Toronto Faculty Law Review 156; Brian Slattery ‘The Charter of Rights and Freedoms: Does it Bind Private Persons’ (1985) 63 Canadian Bar Review 148.

[219] Michael Kanter, ‘The Government Action Doctrine and the Public/Private Distinction: Searching for Private Action’ (1990) 15 Queen’s Law Journal 33.

[220] Ibid 34.

[221] Christopher Henderson, ‘Searching for “Government Action”: Post-Secondary Education as a Case Study in the Conceptual Weakness of the Charter’s Government Action Doctrine’ (2006) 15 Education Law Journal 233.

[222] Ibid 239–45. These specific criticisms of the McKinney judgment are not directly applicable in the Australian administrative law context due to differences in the relationship between the Commonwealth Government and universities. It is extremely unlikely that universities would ever fall within the terms of s 75(v) because of the fact that all Australian universities with the exception of the Australian National University (‘ANU’) are established by state or territory legislation and so fall within the jurisdiction of state or territory courts insofar as their decisions are subject to judicial review. Matters concerning the ANU tend to be dealt with under the terms of the ADJR Act rather than under s 75(v) of the Constitution; see, eg, Australian National University v Burns [1982] FCA 191; (1982) 64 FLR 166 (‘ANU v Burns’); Australian National University v Lewins (1996) 68 FCR 87 (‘ANU v Lewins’). Furthermore, for the reasons outlined above, it is highly unlikely that the High Court of Australia could be persuaded that funding alone is sufficient to make a body an ‘officer of the Commonwealth’; see Aronson and Groves, above n 4, 43–4 [2.160].

[223] Douglas College [1990] 3 SCR 570, 585 [18] (La Forest J for Dickson CJ and Gonthier J). Similar statements were made in many of the other judgments discussed above; see, eg, McKinney [1990] 3 SCR 229, 265 [29] (La Forest J); Greater Vancouver Transport Authority v Canadian Federation of Students – British Columbia Component [2009] 2 SCR 295, 312 [22] (Deschamps J for McLachlin CJ and Binnie, LeBel, Abella, Charron and Rothstein JJ).

[224] Henderson, above n 221, 251–2.

[225] [1997] 3 SCR 844 (‘Godbout’).

[226] Godbout [1997] 3 SCR 844, [48] (La Forest J for L’Heureux-Dubé and McLachlin JJ, with whom all of the other judgments agreed on this issue).

[227] Godbout [1997] 3 SCR 844, [48].

[228] The constitutional status of municipal councils was decided in Australia in Municipal Council of Sydney v Commonwealth [1904] HCA 50; (1904) 1 CLR 208.

[229] Godbout [1997] 3 SCR 844, [49].

[230] Godbout [1997] 3 SCR 844, [49], quoting McKinney [1990] 3 SCR 229, 268 (La Forest J).

[231] Godbout [1997] 3 SCR 844, [51].

[232] Ibid [52]–[54].

[233] It is not within the scope of this article to examine this particular issue in any detail. For a dated, but thorough, discussion, see Kanter, above n 219.

[234] Godbout [1997] 3 SCR 844, [50].

[235] As was the case in Greater Vancouver Transport Authority v Canadian Federation of Students – British Columbia Component [2009] 2 SCR 295.

[236] For instance in Sagen (2009) 98 BCLR (4th) 109 the two tests seem to have been treated as one single test by the Court.

[237] Aronson and Groves, above n 4, 142–3 [3.190].

[238] See above n 148 and accompanying text.

[239] This is an outcome strongly recommended by Aronson and Groves, above n 4, 45 [2.160].

[240] NEAT Domestic [2003] HCA 35; (2003) 216 CLR 277, 297 [49]–[51] (McHugh, Hayne and Callinan JJ).

[241] Though it would not have: it would have come under the jurisdiction of the Supreme Court of Queensland. We will leave aside for now any further discussion of the appropriate Australian general law test.

[242] The ANU stands as a possible exception to this statement: see above n 222.

[243] Aronson and Groves, above n 4, 45 [2.160].