AustLII Home | Databases | WorldLII | Search | Feedback

Melbourne University Law Review

Melbourne Law School
You are here:  AustLII >> Databases >> Melbourne University Law Review >> 2022 >> [2022] MelbULawRw 13

Database Search | Name Search | Recent Articles | Noteup | LawCite | Author Info | Download | Help

Boughey, Janina; Carter, Anne --- "Constitutional Freedoms and Statutory Executive Powers" [2022] MelbULawRw 13; (2022) 45(3) Melbourne University Law Review 903


CONSTITUTIONAL FREEDOMS AND STATUTORY EXECUTIVE POWERS

JANINA BOUGHEY[*] AND ANNE CARTER[†]

In a series of recent cases, the High Court has adopted the position that constitutional freedoms operate as direct limits on legislative power only, and do not directly constrain the exercise of statutory executive powers. In this article, we analyse and critique the rationales for, and implications of, this approach. We argue that the approach is unclear, lacks a compelling justification, and is not workable in all cases. In some cases, courts should review whether the exercise of statutory power in the particular instance has exceeded constitutional limits. We then examine how Australian courts might approach review of administrative decisions alleged to have infringed a constitutional freedom, looking to other common law countries where courts have confronted the same questions.

CONTENTS

I INTRODUCTION

In a series of recent judgments, the High Court has taken the position that constitutional freedoms — specifically the implied freedom of political communication and the s 92 freedom of interstate trade, commerce and intercourse — operate as direct limits on the scope of legislative power only, and do not directly constrain the exercise of statutory executive powers.[1] Under this approach (which we will call a ‘legislation-centric approach’), when a statutory discretion has been exercised in a manner that limits a constitutional freedom, the Court will review the statutory provision which confers discretion for constitutional validity. The majority prefers a structured proportionality test to do so.[2] There is no separate or further inquiry as to whether the limits the decision-maker chose to place on constitutional freedoms in the particular case exceeded these constitutional limits.

In this article, we analyse the rationales for, and possible implications of, this legislation-centric approach. We argue that the approach is unclear and will not always be workable, sufficient or justified. In Part II, we examine the line of cases in which the approach has developed, setting out the Court’s explanations for it and highlighting several key unresolved and disputed questions for the scope and review of the exercise of statutory executive powers. In Part III, we critique the legislation-centric approach on four bases. First, it is inconsistent with other constitutional principles and precedent. Secondly, the main rationale for the approach does not hold up to scrutiny: reviewing an administrative decision on the ground that it unlawfully infringes a constitutional freedom does not transform the freedom into an individual right. Thirdly, it bifurcates the substance of (and not just the process of applying for) judicial review of legislation and administrative action. Fourthly, we argue that, while a legislation-centric approach may work in many situations, review of the statute should not always be the end of the constitutional inquiry as it is not well suited to dealing with the full variety of statutory discretionary powers. In particular, where executive discretions are very broad, a legislation-centric approach can either overly constrain executive power or inadequately protect constitutional freedoms.

In the final Part we look at how courts might approach review of administrative decisions alleged to have infringed a constitutional freedom. We do this by examining the solutions developed in Canada and New Zealand. While, in Australia, the High Court has repeatedly emphasised that neither the implied freedom nor s 92 confers individual rights,[3] in Part IV we show that the issues confronting Canadian and New Zealand courts in the rights context are the same as those which currently confront the High Court. We argue that some of the methods adopted in those jurisdictions can offer solutions which are consistent with Australia’s existing constitutional and administrative

law frameworks.

II THE HIGH COURT’S LEGISLATION-CENTRIC APPROACH

A The Early Cases: Limits on Legislative and Executive Powers

The origins of the legislation-centric approach are found in the Court’s jurisprudence on s 92 of the Constitution. Section 92 relevantly provides that ‘trade, commerce, and intercourse among the States ... shall be absolutely free’. The section makes no explicit reference to whom it prohibits from restricting free trade, commerce and intercourse. It is not expressed as a limit on the powers of only the states, or only the Commonwealth — nor on only legislative or only executive power. As Brennan J observed in Miller v TCN Channel Nine Pty Ltd (‘Miller’), s 92

does not in terms withdraw legislative or administrative power from the Commonwealth or any State, that section invalidates the operation of a law or an administrative action to the extent to which it would prohibit, restrict or burden an activity or transaction within the scope of the protection.[4]

From early on, s 92 has consistently been described as a limit on the exercise of both legislative and executive power. In 1935, Evatt and McTiernan JJ explained that the provision ‘necessarily binds all parties and authorities within the Commonwealth’.[5] In the landmark case of Cole v Whitfield (‘Cole’), a unanimous Court described its task as determining whether ‘particular legislative or executive measures constitute discriminatory interference with interstate trade’.[6] Indeed, the particular act of government being challenged in Cole was one of the executive branch: the making of delegated legislation.[7]

Cole marked a ‘revolution’ in s 92 jurisprudence, and turned it from the most litigated section of the Constitution to a far less problematic one.[8] But prior to Cole, the High Court had considered on a number of occasions how s 92 affected the scope of administrative discretions. As James Stellios has explained, however, these cases were decided in a quite different legal context to the one we inhabit now.[9] Prior to the expansion of judicial review and the administrative law reforms of the 1970s, administrative law was ‘unsophisticat[ed]’ and ‘presented real challenges for keeping discretionary decision-making within constitutional limits’.[10] Thus, the Court took the approach of reviewing the legislation itself and invalidating provisions which conferred broad discretionary powers that might be exercised in a manner inconsistent with s 92.

In the late 1970s and the 1980s, the expansion of administrative law’s ‘grounds’ and remedies ‘began to ease the Court’s concerns’.[11] In a series of cases, judges indicated that if a discretionary power were sufficiently constrained such that ‘its exercise [could not] be obnoxious to the freedom guaranteed by s 92’, the legislative provisions would be valid.[12] However, the judgments were not clear as to whether express limits were required, or whether implications could be read into the statute. In the most substantial treatment of the issue prior to the recent cases, Brennan J explained in his dissent in Miller that a wide discretion could

be destructive of the validity of the scheme only if the exercise of the discretion conferred by the statute [could not] be restrained by judicial review so that its exercise [was] within constitutional power.[13]

He said:

[A] discretion must be exercised by the repository of a power in accordance with any applicable law, including s 92, and, in the absence of a contrary indication, ‘wide general words conferring executive and administrative powers should be read as subject to s 92’ ... If judicial review were not available to ensure that the discretion [was] confined within constitutional limits, an exercise of the power outside those limits could not be restrained. In effect, the power would be wider than the Constitution could support ...[14]

In the context of the power under consideration — the discretion to grant a licence under the Wireless Telegraphy Act 1905 (Cth) — Brennan J noted that judicial review was available pursuant to s 75(v) of the Constitution or under the Administrative Decisions (Judicial Review) Act 1977 (Cth) (‘AD(JR) Act’).[15] This meant:

The discretion [was] effectively confined so that an attempt to exercise the discretion inconsistently with s 92 [was] not only outside the constitutional power — it [was] equally outside statutory power and judicial review [was] available to restrain any attempt to exercise the discretion in a manner obnoxious to the freedom guaranteed by s 92.[16]

It is clear from these passages that Brennan J saw s 92 as a constraint on the scope of the discretion. And he saw judicial review of administrative action[17] as providing the proper process through which a person affected by the exercise of discretion in breach of that constraint could seek remedies from a court.[18] However, contrary to the High Court’s current trajectory (which we explore below), this does not mean that s 92 is relevant only in constitutional proceedings challenging the statutory provisions conferring discretion.

Similarly, the early cases which established the constitutional implication of free communication about political matters referred to that freedom as a limit on both legislative and executive power. For example, in

Theophanous v The Herald & Weekly Times Ltd, the joint judgment of

Mason CJ, Toohey and Gaudron JJ, and Brennan J’s separate judgment, both stated that ‘the implied freedom is a restriction on legislative and executive power’.[19] In Lange v Australian Broadcasting Corporation (‘Lange’), the unanimous Court said that ss 7 and 24 of the Constitution ‘preclude the curtailment of the protected freedom by the exercise of legislative or executive power’.[20] Similar statements can be found throughout the Court’s implied freedom jurisprudence. In McCloy v New South Wales (‘McCloy’), for instance, the Court confirmed that the implied freedom operates as a limit on legislative and executive power.[21] Justice Gageler was even more definitive, noting that there was an ‘ever-present’ risk that an informed electoral choice would be ‘impeded by legislative or executive action’.[22]

However, most of the implied freedom cases, and the few s 92 cases decided since Cole, have involved challenges to legislation — primary or subordinate[23] — as opposed to administrative decisions applying legislation to the circumstances of a particular individual. The modern tests used to determine whether the constitutional freedoms have been infringed have thus developed in the context of legislation, and have been tailored towards legislative power. For instance, the structured proportionality test favoured by the majority asks whether a ‘law’ burdens the freedoms.[24]

B The Court’s Legislation-Centric Approach: Recent Cases

In a series of cases beginning in 2012, the High Court and the Full Court of the Federal Court have addressed the issue of how constitutional freedoms affect the scope of statutory administrative discretions.[25] Wotton v Queensland (‘Wotton’) involved a challenge to various parole conditions, as well as provisions of the Corrective Services Act 2006 (Qld) under which the conditions were imposed, on the basis that they impermissibly infringed the implied freedom of political communication.[26] In this case, the High Court’s rhetoric as to the scope of the implied freedom shifted, with the Court referring to the freedom as a limit on legislative power alone, with no mention of executive power.[27] In short, the Court only considered Wotton’s arguments about the validity of the legislative provisions, and not whether the parole conditions imposed on him themselves infringed the implied freedom.

The Court in Wotton was clear that the constitutional question before it only involved considering the validity of the authorising Act.[28] The Court had very little (Australian) precedent on which to rely. The majority referred to

Brennan J’s judgment in Miller, taking him to mean that the scope of statutory discretions ought to be read as subject to constitutional limits, but that this is a question of construction which goes to the constitutional validity of the Act and not the validity of any administrative decision made under it.[29] The majority indicated that an administrative body which went beyond these constitutional restrictions would act ‘ultra vires’, but ‘ultra vires’ in this context was not seen as involving a constitutional question.[30] In terms of the methodology to be applied, the Court unanimously accepted the Commonwealth’s submissions (made by SJ Gageler SC as Solicitor-General) that ‘where a putative burden on political communication has its source in statute, the issue presented is one of a limitation upon legislative power’.[31] In other words, the only constitutional question is whether, on its proper construction, the statute complies with the constitutional limitation.[32] Under this approach, the Lange ‘reasonably appropriate and adapted’ test[33] would only be applied to the legislation and not to the parole conditions.

What is less clear from Wotton is how exactly the implied freedom might affect the exercise of the statutory discretion. The majority indicated that Wotton’s arguments regarding the validity of the parole conditions placed on him were ‘for agitation in other proceedings, in particular, proceedings under the Judicial Review Act’.[34] They commented several times that broadly framed statutory discretionary powers are constrained by the implied freedom. For instance, they agreed that broad discretions should be read as ‘subject to’ constitutional constraints[35] and said that if powers are ‘susceptible of exercise in accordance with the constitutional restriction ... that power or discretion is effective in those terms’.[36] The majority also said that, when exercising discretion to allow parolees to be interviewed, the chief executive was to ‘have regard to the restraint upon legislative power’, and that this would be reviewable via administrative law.[37]

These comments indicate that the Court did not view the implied freedom as irrelevant in administrative law proceedings, or that it played no role in limiting the scope of discretion in particular cases. Instead, the majority was clear that the proper procedure for challenging administrative action alleged to unjustifiably limit the implied freedom, made under valid law, is via administrative law, and that administrative law provides the remedies for executive action that breaches the implied freedom. What was less clear was exactly how the implied freedom would operate as a limit on discretionary power. One possibility canvassed during the hearing by Hayne J was that it may operate as a ‘relevant consideration’: that is, that discretions are read as if they contained the clause, ‘provided that the decision-maker considers the

implied freedom’.[38]

This question was considered by the Full Federal Court in Chief of Defence Force v Gaynor (‘Gaynor’).[39] In that case, the respondent had been terminated from the Australian Defence Force (‘ADF’) as a result of comments he had made on social media criticising the ADF.[40] He challenged his termination on numerous bases, including on every ground in the AD(JR) Act and, in addition, on the basis that it had infringed the implied freedom.[41] While Gaynor’s

AD(JR) Act arguments were poorly articulated, largely ‘simply paraphras[ing] the available grounds of review’,[42] the primary judge set aside the termination decision on the basis that it infringed the implied freedom.[43] In doing so, Buchanan J applied the Lange test directly to the ADF’s decision to terminate Gaynor and found that, by impermissibly burdening the implied freedom in Gaynor’s case, the Chief of the Defence Force had exercised power in a way which was ‘not authorized by the enactment’, ‘an improper exercise of the power’, or ‘otherwise contrary to law’ pursuant to the AD(JR) Act.[44] On appeal, the Full Federal Court held that the primary judge ‘err[ed] in the level at which he applied the Lange test’.[45] By applying the Lange test to the exercise of discretion itself, the primary judge had erroneously treated the implied freedom as an individual right rather than as a limit on power.[46] The Full Federal Court held that only the regulations ought to have been tested for infringing the implied freedom — not their application to Gaynor.[47] The latter, according to the Court, was ‘an administrative law issue’ and not a ‘constitutional one’.[48] These comments suggest that the constitutional question of whether an action unlawfully infringes the implied freedom is a question that applies only to legislation, both primary and delegated. The Court in Gaynor repeatedly referred to the implied freedom as a limit on legislative power alone.[49] While it acknowledged statements throughout the previous case law saying that the implied freedom also limits executive power, it viewed these as merely ‘general propositions’.[50]

The Court in Gaynor indicated that review of discretionary decisions made under legislation could be sought on those grounds set out in the AD(JR) Act[51] — or presumably the common law grounds which that Act sought to codify, where judicial review is sought via an avenue framed around common law remedies. The Court left open the possibility that the implied freedom might affect the way these administrative law grounds are framed, specifically noting that it might be a ‘relevant consideration’ for decision-makers.[52] The Court saw this as one method of giving effect to the statements in Miller and Wotton that administrative discretions are ‘constrained by the constitutional restrictions upon the legislative power’,[53] indicating this was ‘one way of characterising the nature of the excess of power, although not the only way’.[54] In other words, the implied freedom might inform existing ‘grounds’ of review, but was not a distinct limit on discretionary power. However, the Court ultimately did not need to determine how the implied freedom affected the termination power in this case, so it did not reach a definitive view.[55]

The Full Federal Court also brushed aside the ‘general propositions’ about the implied freedom operating as a limit on executive power by interpreting this to mean non-statutory executive power.[56] It was not necessary in the case to resolve the question of how the implied freedom might affect non-statutory executive power.[57] While there is considerable ambiguity in the phrase ‘executive power’, and that phrase has sometimes been used to refer only to non-statutory executive power, in recent decades the High Court has used the phrase to encompass both statutory and non-statutory sources.[58] This is supported by the use of the phrase in s 61 of the Constitution, which states that ‘[t]he executive power of the Commonwealth ... extends to the execution and maintenance of this Constitution, and of the laws of the Commonwealth’.[59] Further, many of the High Court’s general statements about the applicability of the implied freedom to executive power — discussed above — contain strong indications that this includes the power to administer legislation.[60]

The High Court, in the case of Comcare v Banerji (‘Banerji’), confirmed that the implied freedom does not apply directly to the exercise of statutory executive powers, but the judgments include a range of seemingly conflicting statements about how the implied freedom affects the scope of administrative discretion.[61] Michaela Banerji, an employee of the Commonwealth Department of Immigration, was dismissed from the public service for posting thousands of tweets which were critical of the government’s immigration policies.[62] She challenged her dismissal,[63] and the provisions of the Public Service Act 1999 (Cth) under which it was done, on the basis that they infringed the implied freedom.[64] Echoing the approach taken in Gaynor, the joint judgment of

Kiefel CJ, Bell, Keane and Nettle JJ criticised the Administrative Appeals Tribunal for treating the implied freedom as a personal right that applied directly to the termination decision.[65] They emphasised that the implied freedom is a limit on legislative power only, commenting:

[T]he question of whether the law imposes an unjustified burden on the implied freedom of political communication is a question of the law’s effect on political communication as a whole.[66]

Where there was more confusion in Banerji was in terms of how the implied freedom might limit executive discretion in individual cases. Banerji argued, based on the Gaynor judgment, that even if the implied freedom did not apply directly, it informed the grounds of review. Specifically, she contended that it was either a relevant consideration, or that it placed an ‘outer limit on the range of penalties open to be imposed’.[67] The joint judgment rejected the first contention, but indicated this ‘does not mean that the implied freedom may not be a relevant consideration in the exercise of different discretions under other legislation’.[68] Thus, the door was left ever so slightly ajar on the issue of whether the implied freedom might sometimes limit the exercise of statutory discretion as a mandatory consideration. With respect to Banerji’s second contention — which in essence translates to a limit on what penalties could reasonably be imposed — the joint judgment said that if the legislation

provided for sanctions that were not reasonably justified having regard to the implied freedom of political communication, [the legislation] would be invalid and any penalty imposed under it would be unlawful, or at least unlawful to the extent that the penalty went further than was warranted by the
implied freedom.[69]

The joint judgment’s approach in this regard seems somewhat contradictory. While a court should not assess whether the exercise of statutory discretion in an individual case places a disproportionate limit on the implied freedom, at the same time, it is possible for a penalty to be unlawful if it goes further than the implied freedom warrants. It is not clear how a court would assess whether a penalty applied in a specific case goes beyond what the implied freedom warrants, and is hence unlawful, without reference to the implied

freedom itself.

The joint judgment in Banerji went on to say that where the sanctions set out in law are reasonably justified, then the application of sanctions in a specific case must simply meet the ordinary legal requirement of reasonableness,

and that

[i]f a decision maker imposes a manifestly excessive penalty, it will be unlawful because the decision maker has acted unreasonably, not because of the decision maker’s failure to turn [their] mind to, or failure expressly to mention, the implied freedom.[70]

This appears to indicate that discretions should be assessed using the legal unreasonableness standard as set out in the administrative law cases of Minister for Immigration and Citizenship v Li[71] and Minister for Immigration and Border Protection v SZVFW.[72] The implied freedom forms the context for what is reasonable, but it should not be expressly mentioned or discussed by the reviewing court in assessing the reasonableness of the decision.

Justice Gageler took a more definitive view on the role (or lack thereof) of the implied freedom in limiting executive discretion in individual cases. Quoting Basten JA in the New South Wales Court of Appeal, he said that the implied freedom does limit executive power but does so ‘by limiting the scope of legislative power’,[73] meaning a court will only assess the validity of the legislation itself. He agreed with the joint judgment that the implied freedom is not a relevant consideration in the exercise of statutory discretion.[74] However, Gageler J indicated that where legislation confers discretion, whether the burden is justified must be determined by looking ‘across the range of potential outcomes of the exercise of that discretion’.[75] If the law is justified ‘across the range of potential outcomes’, there is no further role for the implied freedom, and there ‘is no occasion to consider whether the scope of the discretion might be read down in order to ensure that the law is within constitutional power’.[76] His Honour did not, however, discuss what happens when a law is justified across only some of its potential outcomes. For instance, might the law be read so as to be ‘subject to’ the condition that the discretion not be exercised in a manner inconsistent with the implied freedom? Justice Edelman addressed this issue more directly, indicating that the implied freedom might ‘[constrain] the manner in which the statute can be applied’.[77] He indicated that this might be called ‘reading down’, but preferred the label ‘disapplication’, meaning that there will be an implied condition that discretionary powers cannot be exercised in a manner in which the executive act goes beyond the limits on the legislature’s own powers.[78]

The judgments in Banerji seem to agree that the implied limit does not give rise to a distinct ‘ground’ of review. Beyond this, however, there are several points of disagreement and confusion. The first concerns whether the implied freedom plays a role in interpreting the scope of administrative discretion. Should discretionary powers be read as ‘subject to’ the condition that they can only impose reasonable limits on the implied freedom, or otherwise ‘read down’ where the statutory language permits? Secondly, how does a court determine whether the burden imposed by a law which confers discretion is justified ‘across the range of potential outcomes of the exercise of that discretion’? In particular, how can a court possibly foresee all of the ways in which a discretion might be exercised, particularly a very broad discretion? Thirdly, in what circumstances, if at all, is the implied freedom a relevant consideration in the exercise of statutory discretion?

The issue of how statutory discretions might be limited by constitutional guarantees came before the High Court again in Palmer v Western Australia (‘Palmer’),[79] but the judgments diverged significantly from each other and from previous authority. The case has thus contributed to the confusion surrounding the interpretation, scope and review of statutory executive power, and has left several questions unanswered. In Palmer, the Court again applied a legislation-centric approach, this time in the context of a s 92 challenge to the Western Australian government’s decision to close its borders in response to the COVID-19 pandemic.[80] The border closure, which prohibited people from entering Western Australia without an exemption, was given effect by Directions made by the State Emergency Coordinator under the Emergency Management Act 2005 (WA) (‘EM Act’).[81] There was some uncertainty during the hearing about whether the Directions were legislative or administrative in character,[82] but the Court did not find it necessary to resolve this issue. Although the plaintiffs initially argued that the Directions were invalid by reason of s 92 of the Constitution,[83] during the course of proceedings they extended their arguments to challenge the validity of the EM Act.[84] This meant that the issue of whether and how s 92 affects the interpretation of provisions of primary legislation conferring discretionary powers on the executive, and the exercise of discretionary powers themselves, was squarely before the Court.

Chief Justice Kiefel and Keane J dealt with the issue briefly, relying on Wotton and holding that ‘the question of compliance with the constitutional limitation is answered by the construction of the statute’.[85] While their Honours appear to have intended this as a general statement, they did leave a little wriggle room by acknowledging that ‘[i]n some cases difficult questions may arise because the power or discretion given by the statute is broad and general’.[86] Such questions did not arise, however, in the Palmer litigation.[87] Their Honours went on to purportedly examine whether s 67 of the EM Act impermissibly burdened free movement into Western Australia, applying a test of structured proportionality. In order to determine whether the burden on movement was disproportionate, Kiefel CJ and Keane J necessarily had to consider the justification for its application in the particular circumstances of COVID-19, finding that the purpose of the law was to prevent persons infected with COVID-19 from bringing the disease into the community[88] and that the uncertainties about the level of risk meant that the Western Australian government was, in the view of the primary judge, justified in its ‘precautionary approach’.[89] This reasoning demonstrates how fact-dependent assessments of proportionality are, and how they often cannot be divorced from the application of a law in a particular case.[90] The difficulty with Kiefel CJ and Keane J’s reasoning is that, in applying the proportionality test, their Honours in fact assessed the validity of the Directions themselves, rather than s 67 of

the EM Act.

Justice Gageler gave more substantial treatment to explaining why, in the context of statutory executive power, the relevant constitutional guarantees operate solely as limits on legislative power.[91] In doing so, his Honour took the view that previous references to constitutional guarantees as limitations on executive power meant only non-statutory executive power.[92] In terms of the exercise of executive power pursuant to statute, Gageler J explained that two questions arise: ‘one constitutional, the other statutory’.[93] He suggested that it made no difference whether the executive power in question was administrative or legislative in character, with the effect that many of the earlier cases applying constitutional freedoms to test the validity of delegated legislation[94] had taken the wrong approach.[95]

Although these statements seem definitive, Gageler J went on to acknowledge that the constitutional and statutory questions ‘can converge’ when a statutory discretion ‘is so broadly expressed as to require it to be read down as a matter of statutory construction to permit only those exercises of discretion that are within constitutional limits’.[96] He also acknowledged that the question of whether the exercise of a discretionary power can be justified across the range of its potential outcomes ‘might not yield a ready answer’.[97] In those circumstances, he said, it might be necessary to answer the constitutional question by focusing on the particular exercise of discretion in the case.[98] In other words, in some situations it will be necessary for a court to decide whether the limit that a particular decision places on one of the constitutional freedoms is justified. However, Gageler J did not think that this approach was appropriate in this case, as it ‘failed to acknowledge the constitutional significance of critical constraints built into the scheme of the Act which sustained the Directions’.[99] This meant that the s 92 assessment should be applied to the legislation itself, and not the Directions.[100] Because of the various ways in which the primary Act ‘hedged’ the scope of the statutory discretion to make Directions, Gageler J took the view that the power could never be exercised beyond the scope of the Constitution.[101] This leaves open the question of what happens if a statutory discretion is not sufficiently constrained by express legislative terms, such that it might either be exercised consistently with a constitutional freedom, or inconsistently with the freedom.

Justice Edelman also gave extensive treatment to the question of how and when the constitutional freedoms might affect the scope of statutory discretions. His approach differed from those of the other Justices in important ways. He indicated that the validity of the primary legislation ‘must be the starting point’, but ‘[w]here the relevant provisions of the primary legislation are open-textured and can be disapplied from any invalid application’ it would ‘rarely be appropriate for a court to speculate upon whether the provisions are valid in all their applications’.[102] This is a marked difference from the position articulated by Gageler J in Banerji,[103] and the other Justices in the Palmer case. Justice Edelman went on to confirm that the constitutional and administrative law questions and processes are distinct, saying that questions of constitutional validity were to be ‘determined at the level of an empowering statute’ and questions about the validity of actions done under legislation (whether administrative or legislative in character) turned on whether the Act empowered the action.[104] But his Honour then said:

In cases where statutory provisions are open-textured where their interpretation requires them to be applied distributively to numerous different circumstances and do not expressly incorporate sufficient limitations as to be facially compliant with the Constitution then the Court should rarely adjudicate upon the validity of all applications of the relevant statutory provision. It is enough to conclude that the provisions can be ‘disapplied’ or, in unfortunate terminology better used in the interpretation of the meaning of words rather than in their application, ‘read down’ to exclude any hypothetical applications that might be constitutionally invalid. The Court can focus upon the application of the provision to the relevant facts before the Court or facts of that general kind.[105]

It is far from clear how focusing on the application of the law to a particular set of facts to assess whether the legislation, as applied in that situation, unreasonably limits the relevant freedom is substantively different from an administrative law approach asking whether the exercise of discretion unreasonably limits that freedom. Justice Edelman acknowledged this, referring to it as a ‘tension’.[106] He sought to reconcile this tension by returning to the idea expressed in other judgments and earlier cases that there is a strict delineation between constitutional and administrative law. He said that ‘assessing the validity of an open-textured legislative provision ... by reference to limited applications of that legislative provision ... engages questions of constitutional power’,[107] whereas ‘assessing the validity of the ... particular application ... itself ... should only be a question of whether the by-law falls within the valid legislative power’.[108] Justice Edelman’s approach also leaves open several questions, including about the correctness of earlier cases concerning delegated legislation,[109] and exactly what any ‘reading down’ process would look like. It also leaves unanswered a large question about the consequences for the primary Act if a court concludes that its application in a particular instance infringes a constitutional freedom. Is the entire Act invalid, or just its exercise in that instance? The former would be absurd: akin to a police officer’s abuse of their position invalidating legislative provisions conferring power on all police officers. But the latter seems indistinguishable from judicial review of administrative action.

Overall, the Palmer judgments are broadly in agreement that courts should adopt a legislation-centric approach to the constitutional freedoms.[110] However, the judgments open up several new questions on which different members of the High Court seem to give different answers and, on occasion, to diverge from previous authority. To summarise, the important questions on which the recent case law provides no clear answers are:

1 Whether constitutional freedoms are relevant in administrative review proceedings. If so, must the existing ‘grounds’ accommodate the freedoms? And if so, which one(s)? The majority in Wotton indicated that constitutional freedoms might be a relevant consideration in the exercise of discretion.[111] But that suggestion seems to have been dropped in

subsequent cases.

2 Whether broadly framed discretions, without an express, built-in equivalent to the constitutional ‘reasonable limits’ requirement, can and should be ‘read down’ or ‘disapplied’.[112] If they should, it is not clear what this looks like. The obvious approach would be to read broad discretions as containing an implied condition that the power is subject to the condition that it not be exercised in a manner that unjustifiably limits a constitutional freedom. This then becomes an implied statutory limit, which courts reviewing the exercise of discretionary power must have the ability to test. The questions set out in paragraph 1, above, return.

3 When a court assesses the validity of a statutory provision conferring discretionary power, can and should it consider the ‘range of potential outcomes’ of the exercise of the discretion?[113] Or, can the court do as Edelman J did in Palmer and purport to assess the validity of the provision itself by looking at its application in the particular case?[114] If the latter, then what are the implications for the legislative provision of a finding that the way legislation has been applied in a particular case is invalid, and how does this differ from a finding in administrative law proceedings?

III A CRITIQUE OF THE LEGISLATION-CENTRIC APPROACH

Despite its significant uncertainties, it is easy to see why the legislation-centric approach is attractive to the High Court from a practical perspective. Perhaps most critically, it means that the Court does not have to grapple with the questions of whether or how to assess limitations on rights in the context of administrative decision-making in a way which is compatible with Australia’s separation of powers.[115] Relatedly, it also removes the potential for proportionality to develop as a new ‘ground of review’ and the possible flood of litigation that might arise from establishing such a new ground. But these are not the reasons that the High Court has given. The only concrete explanation that has been given for the legislation-centric approach is the suggestion that applying the freedoms directly to the exercise of executive powers would be tantamount to treating the freedoms as individual rights.[116] Otherwise, the Court has simply shifted its rhetoric to now refer to the constitutional freedoms as limits on legislative power alone, suggesting the constitutional basis for the legislation-centric approach is self-evident. We contend that it is not, and that the approach in fact suffers from several fundamental flaws.

A The Constitutional Freedoms Do Constrain Statutory Executive Power

Our first criticism is of the Court’s recent rhetoric suggesting that constitutional freedoms operate as limits on legislative power only. To be clear, the Court has not expressly said that statutory executive power is unconstrained by the constitutional freedoms. Rather, its view seems to be that the way constitutional freedoms limit statutory executive power is via statute; the Constitution does not operate directly to constrain the exercise of statutory executive powers. However, as our discussion in Part II highlights, the rise of the legislation-centric approach has been accompanied by a noticeable change in rhetoric regarding the effect of the constitutional freedoms. To the extent that this rhetoric reflects a view that the constitutional freedoms do not constrain the scope of statutory executive powers, this conflicts with established principles and precedent and lacks a convincing constitutional basis.

In terms of s 92, there is nothing in the text which limits it only to legislative power. Indeed, in many of the pre-Cole cases, discussed in Part II(A), the Court explicitly acknowledged that it ought to apply to both legislative and executive power. In Betfair Pty Ltd v Racing New South Wales (‘Betfair’), Kiefel J observed that Parliament ‘cannot grant a discretionary power which is to be exercised in a manner inconsistent with s 92’.[117] Likewise, there are various statements from the High Court indicating that the implied freedom of political communication operates as a limit not only on laws but on the exercise of executive power.[118] The Court has also found that the implied freedom affects the common law, which conflicts with suggestions that it only limits legislative powers. A number of the early implied freedom cases, including the seminal Lange decision, emerged in the context of defamation proceedings, with the Court establishing that the common law must conform with the requirements of the Constitution, including the implied freedom.[119] In addition, the textual foundations of the implied freedom also support the view that it limits both legislative and executive power. The High Court has repeatedly confirmed that the freedom is derived from ss 7, 24, 64 and 128 of the Constitution, and related sections,[120] which establish a system of representative and responsible government. The inclusion of s 64 suggests that the implied freedom might also operate as a limit on the exercise of executive powers sourced to s 61.[121] Justice Gageler explained in McCloy:

The necessity for the implication of the constitutional freedom as a limitation on legislative and executive power arises from a paradox inherent in the nature of the majoritarian principle which governs that electoral choice. The paradox is that communication of information relevant to the making of an informed electoral choice is peculiarly susceptible to being restricted or distorted through the exercise of legislative or executive power precisely because the exercise of legislative or executive power is subject to the ultimately controlling influence of electoral choice.[122]

The exercise of a statutory executive power in a manner which restricts or distorts informed electoral choice is no less a threat to these principles than the exercise of legislative power. Statutory executive power must be constrained by the constitutional freedoms by some means. This is not to say that this cannot be done via statute: that is, statutes may be interpreted as constraining executive power. But the scope of statutory executive powers must be so constrained in order to ensure that constitutional freedoms are upheld.

The constitutional text, jurisprudence and commentary relating to statutory executive power reveal no reason why these powers would not be constrained by constitutional freedoms. Statutory executive power is clearly sourced to s 61 of the Constitution.[123] Although s 61 is not phrased in terms of being ‘subject to the Constitution’, as for example is s 51, there are good reasons for thinking that it is. For example, the power in s 61 is clearly expressed to include ‘the execution and maintenance of this Constitution, and of the laws of the Commonwealth’. It would seem implicit, therefore, that executive power should operate ‘in aid of the Constitution and not inconsistently with it’.[124] In addition, as Joshua Forrester, Lorraine Finlay and Augusto Zimmermann have pointed out, other sections of the Constitution (which do not contain these express words) are subject to constitutional limitations.[125] While the precise scope of executive power remains somewhat elusive, the High Court has indicated that ‘executive power generally’ is ‘subject to limitation’.[126] The Court in Pape v Federal Commissioner of Taxation, for instance, held that the executive power of the Commonwealth was limited by considerations of text and structure.[127] As French CJ remarked, there are ‘broadly defined limits’ to the scope of executive powers sourced to s 61, and

the exigencies of ‘national government’ cannot be invoked to set aside the distribution of powers between Commonwealth and States and between the three branches of government for which this Constitution provides, nor to abrogate constitutional prohibitions.[128]

Academic commentators, too, tend to agree or assume that the executive powers sourced in s 61 must themselves be subject to constitutional limitations (whether express or implied).[129] The real source of uncertainty has been how to give effect to these limitations in the context of executive action.

General legal principles of delegation also support the notion that statutory discretions are subject to constitutional constraints. Statutory executive power is a form of delegation (whereby the legislature delegates certain powers to the executive),[130] and it is well settled that Parliament cannot delegate a power it does not have.[131] For instance, if Parliament cannot legislate in a manner which impermissibly infringes a constitutional freedom, it would seem axiomatic that the executive should not be able to make a decision which has this effect.[132] To do so would undermine the protections offered by the Constitution.

Furthermore, both s 92 and the implied freedom of political communication have been applied directly to the exercise of statutory executive powers in the context of delegated legislation. For example, in cases such as Levy v Victoria[133] and Attorney-General (SA) v Adelaide City Corporation,[134] the Court assessed whether delegated legislation impermissibly infringed the implied freedom of political communication. While some of the judgments in Palmer cast some doubt on the correctness of these earlier decisions, as discussed above,[135] a majority of the Court has not, so far, explicitly clarified why it is that constitutional freedoms might directly limit the exercise of statutory executive power when it comes to delegated lawmaking but not individual decisions. We suspect that the only explanation for differential treatment is that the tests that have been developed to assess whether action impermissibly restricts the constitutional freedoms are adapted to legislation, but not well suited to testing many administrative discretions. We discuss this further in Part IV below. But this does not provide a sound justification for treating the different kinds of executive power differently in respect of their interaction with the Constitution.

It appears, then, that executive power generally is constrained by the constitutional text and structure, which must surely include both s 92 and the implied freedom. There appears to be no convincing reason to describe these constitutional freedoms as limits on legislative power alone, as the Court has done in its recent change in rhetoric. If the constitutional freedoms do limit the scope of statutory executive power, as we contend, then there is a clear need to find a suitable method of testing whether the executive has breached these limits in exercising its powers. As we explain below, while a legislation-centric approach will be appropriate in some cases, it is far from clear that it is an appropriate method in all cases.

B Judicial Review of Administrative Action Does Not Turn Legal Limits into Rights

Our second criticism relates to the courts’ explanation for refraining from applying constitutional freedoms directly to the exercise of executive powers, which is that doing so would have ‘the effect of treating the implied freedom as a personal right’.[136] The High Court has repeatedly emphasised that s 92 and the implied freedom do not give rise to individual rights.[137] In Unions NSW v New South Wales, for example, the Court responded to arguments that political donations themselves were a form of political communication, by explaining:

[W]hat the Constitution protects is not a personal right ... Thus the question is not whether a person is limited in the way that he or she can express himself or herself ... The central question is: how does the impugned law affect the freedom? ... If the submission is to be understood as referring to a restriction effected by [the Act] upon the right of particular persons and entities to make communications, it may blur the distinction ... concerning the freedom.[138]

The logic underpinning such statements is that assessing whether a particular administrative decision impermissibly infringes a constitutional freedom necessarily involves considering whether the affected person’s freedom has been infringed. This is true. But it does not follow that the constitutional freedoms would suddenly become transformed into an individual ‘right’. Judicial review of administrative action and its remedies are concerned with limits on executive power, not individual rights. For instance, assessing whether a decision has been made in the absence of procedural fairness involves determining whether the affected individual has been treated fairly. But this does not make procedural fairness an individual right. Likewise, assessing whether an administrative decision was reasonable does not reflect a view that individuals have a ‘right’ to a reasonable decision. Rather, these familiar arguments in administrative law cases reflect the fact that fairness and reasonableness are presumed limits on executive power.[139] As Brennan J famously put it in Attorney-General (NSW) v Quin, ‘the scope of judicial review must be defined not in terms of the protection of individual interests but in terms of the extent of power and the legality of its exercise’.[140] Just as procedural fairness, reasonableness and the other principles reflected in administrative law’s ‘grounds of review’ are limits on power, rather than individual rights, so too constitutional freedoms are limits on power. There is no suggestion that a breach of constitutional freedoms should give rise to a distinct cause of action or to remedies (such as damages) which would result in the freedoms having the legal status of an individual right.

Furthermore, the Court has acknowledged that an assessment of the burden on the implied freedom in a particular case may be relevant as part of considering the law’s effect on political communication as a whole.[141] As Palmer illustrates, the question of whether a discretion offends a constitutional limitation will sometimes only sensibly be able to be answered when the discretion has been exercised.[142] That is, it will not be possible to determine this at the level of the empowering legislation only. In such circumstances it may be necessary to assess how the exercise of the discretion affects a constitutional freedom in a particular case.

C The Bifurcation of Administrative and Constitutional Law

Our third criticism of the legislation-centric approach is that it assumes that there is a clear delineation between ‘administrative law issues’ and ‘constitutional issues’ — not just procedurally, but substantively.[143] For example, the joint judgment in Wotton said that the validity of a particular application of executive power conferred by statute is not a question of constitutional law, but of whether the ‘repository of the power has complied with the statutory limits’.[144] As explained in Part II, the judgment indicated that arguments about the validity of Wotton’s parole conditions should be determined in judicial review proceedings,[145] a point which we do not dispute.[146] But it is one thing to treat administrative and constitutional law as procedurally and remedially distinct, and quite another to treat the sources of limits on executive power as neatly falling into those categories.

The sources of courts’ jurisdiction to hear challenges to the validity of administrative and legislative action are different,[147] as are the procedures and some of the remedies associated with those challenges.[148] But the sources of jurisdiction and remedies are separate questions to the source of limits on administrative power. Furthermore, ss 75(iii) and (v) of the Constitution — which entrench the High Court’s original jurisdiction to review executive action for jurisdictional error — can be used to challenge the constitutional validity of executive action as well as whether the action has breached statutory limits. For example, in Plaintiff M68/2015 v Minister for Immigration and Border Protection, the plaintiff argued that the Commonwealth and Minister had acted beyond the scope of their powers by their involvement in her detention at the regional processing centre in Nauru.[149] She applied under ss 75(iii) and (v), and raised arguments relating to the statutory limits on the Minister’s powers, as well as the scope of the executive power under s 61 of the Constitution.[150] The case illustrates how statutory limits on executive power simply cannot always be extricated from constitutional issues. Similarly, in CPCF v Minister for Immigration and Border Protection, arguments were made about the scope of statutory executive powers, non-statutory executive power under s 61 of the Constitution, and the compatibility of the statute with constitutional principles.[151] There are thus no strict boundaries between ‘constitutional’ and ‘administrative’ law, demonstrating that, as Gageler J put it in Palmer, statutory and constitutional questions ‘can converge’.[152]

The cases discussed in Part II make several references to administrative law as being concerned solely with statutory limits on power, and whether a decision-maker has acted ‘ultra vires’ — a term which harks back to the old English debate about the source of limits on executive power.[153] But judicial review of administrative action is not, as it once was, inherently restricted to review of statutory limits on executive power. Justice Brennan’s canonical statements about the nature and scope of judicial review of administrative action refer to the enforcement of law and legal limits on executive power, not just statutory limits.[154] Similarly, the availability of the remedies associated with judicial review of administrative action is defined by reference to excess of jurisdiction and error of law, not simply to breaches of statute.[155] Although we live in a statutory universe,[156] meaning that the vast majority of legal limits on executive power are sourced in statute, statutes are not the source of all legal limits on executive power. In a number of recent cases, following United Kingdom precedent, state Supreme Courts and the Federal Court have found that non-statutory powers are subject to limits such as fairness and reasonableness.[157] Those limits cannot be sourced in statute, but could only come from the common law or Constitution (perhaps in combination).[158] Furthermore, the High Court has recognised that the principles of statutory construction are not separate from common law and constitutional principles, but derive from them.[159] As Justice John Basten has explained:

[Th]e implied limits on powers derive from the values, or standards, which are found within our legal and political systems of government, including constitutional principles governing the institutional structure of the government.
The exercise may be characterised as an application of principles of statutory interpretation, but that label reveals little as to the source or justification of the applicable principles.[160]

Relatedly, some of the judgments discussed in Part II went on to consider whether and how the implied freedom might affect administrative law’s ‘grounds of review’ — for instance, by asking whether it might be a ‘relevant consideration’.[161] These discussions are problematic to the extent they suggest that the constitutional freedoms constrain administrative discretion only if they can be accommodated within common ‘grounds’ of review. Judicial review of administrative action is not confined to the ‘grounds’ set out in the AD(JR) Act and its state and territory equivalents.[162] A reviewing court’s task is not to decide whether one or more of the ‘grounds’ set out in a statute or textbook has been made out, but whether any express or implied legal constraint on the particular power has been breached. It is, of course, necessary to frame arguments made under the AD(JR) Act and state and territory equivalents in terms of the ‘grounds of review’ set out in those Acts. Thus, in Gaynor and Wotton, in which the respective judicial review applications were or would have been made under the Commonwealth and Queensland Acts, the plaintiffs’ arguments and the courts’ tasks would necessarily have been framed in those terms.[163] But this should not be taken more broadly to mean that the constitutional freedoms only constrain administrative discretion to the extent to which they are capable of being accommodated within common ‘grounds’ of review. And it would not have been the situation in Palmer, had the plaintiffs sought a prerogative writ or equitable remedy in respect of the Direction. Asking which ‘ground of review’ would apply, or how various grounds might accommodate constitutional freedoms, puts the cart before the horse.[164] The question is not whether constitutional freedoms are a ‘relevant consideration’, whether their breach results in unreasonableness, or whether proportionality ought to become a ‘ground of review’ in constitutional freedom cases. It is: do the freedoms constrain this statutory administrative power and, if so, how?[165]

D Practical Implications and Challenges

Our final criticism of a legislation-centric approach to the implied freedom is practical. Focusing on the statutory provisions alone, without specific consideration of constitutional freedoms in administrative review proceedings, is not workable in all cases. The discretionary powers conferred on the executive branch by legislation are many and varied. Some have clear and well-defined parameters, and the ‘range of potential outcomes’[166] that might result from an exercise of the discretion will be apparent simply by looking at the power and understanding its context and operation. Other discretionary powers, however, are broader or less clearly defined, and it will not be possible for a reviewing court to imagine all of the ways in which the discretion might be exercised. As Kieran Pender puts it, there will be situations where compatibility is not clear on the statute’s face, and ‘only becomes apparent following an individual exercise of the discretion’.[167] In Palmer, as we explained, Gageler J and Edelman J explicitly acknowledged this.[168]

Several of the judgments set out in Part II indicate that statutory powers will be interpreted, ‘read down’, or ‘disapplied’ to ensure that the executive cannot exercise broadly framed discretions in a manner inconsistent with the constitutional freedoms.[169] However, it is not clear how this would work, nor how a court would test whether the implied constitutional limit had been exceeded in administrative review proceedings, without reference to the constitutional freedom itself. Consider, for instance, a broad statutory discretion to cancel a person’s visa on the basis that ‘having regard to ... the person’s past and present general conduct ... the person is not of good character’.[170] The discretion is, of course, conditioned by the context in which it operates; that is, it is a discretion conferred for the purposes of immigration and protecting the Australian community from various threats, which exists alongside more specific reasons why a person might not be of good character. Nevertheless, it is conceivable that this broad discretion might be exercised in a manner which is plainly inconsistent with the implied freedom. For example, imagine a scenario where the Minister cancelled a person’s visa solely on the basis of their espoused view that the Australian legal system ought to be replaced with a system of religious law, without providing any sound justification for why this is necessary. While the person’s views are unpopular, they are not expressed in a way that is violent, harassing or threatening. The Minister’s exercise of discretion in this scenario might be characterised as beyond power on the basis that it is for an improper purpose or unreasonable. But what makes it improper or unreasonable is the fact that it limits a protected value of free political expression; in other words, the implied freedom provides relevant context for the lawful ways in which the discretion is capable of being exercised. It would, however, be entirely unreasonable and unwarranted for a court to strike down the whole provision simply because it might, on rare occasions, be exercised in a manner inconsistent with the implied freedom. It would also be impossible for a court to imagine all of the possible ways in which such a broad discretion might hypothetically be exercised. Interpreting broad discretions as subject to a requirement that they may only be exercised in a manner which places reasonable limits on constitutional freedoms, without considering whether this specific limit has been breached in judicial review proceedings, risks permitting the executive to exceed constitutional limits. However, finding a power which confers discretion which itself may, on occasion, be exercised in a manner which infringes a constitutional freedom invalid in its entirety is overly constrictive.

A second practical problem with the legislation-centric approach is the difficulty in distinguishing between subordinate legislation and administrative decisions in certain circumstances. The case law suggests that the constitutional freedoms do apply directly to government actions which set the law for the community at large, but not to the application of laws in specific cases. One challenge with this is that there are a great many administrative discretions which affect a wide section of the community, and not just specified individuals. Take, for example, many of the rules restricting movement, association and protest in response to the COVID-19 pandemic, like the Directions in Palmer.[171] Some have been achieved via delegated legislation, but others have been created through the form of administrative decisions or orders.[172] While the effect of these delegated laws and administrative decisions is the same, their form, and oversight of them, differ.[173] The High Court’s recent statements that the implied freedom limits law only risk ignoring this fact and elevating the form through which the executive acts above the effect of its actions. This issue was raised but not resolved in the Palmer litigation, where, as we have noted, both Gageler J and Edelman J suggested that the many cases which have applied constitutional freedoms directly to delegated legislation may have been wrong in their methodological approach.[174]

Our point is that statutory executive discretions are many and varied. They range from tightly confined powers with built-in limits to unstructured and very broad discretions.[175] Statutes confer powers on the executive to make delegated legislation and soft-law instruments which apply generally to the whole population, or large segments of it, and powers to apply law and policy to specific individuals. In some situations it might be easy to see, based on the wording of the relevant statute, that a power can never be exercised in a manner which infringes the implied freedom. But in others, particularly in relation to broad discretions, it will be impossible to know all the possible ways in which a decision-maker might exercise their powers in advance. And even if a court can foresee that there is scope for the discretion to be exercised in a manner which infringes the implied freedom in rare cases, it may be undesirable to find the entire provision invalid. In short, a one-size-fits-all approach to review of statutory discretions is neither possible nor desirable.

IV APPROACHES IN CANADA AND NEW ZEALAND

We have argued so far that the legislation-centric approach is flawed in critical ways, and is not a workable method for ensuring that statutory executive powers are exercised within constitutional limits in all cases. In some cases, such as Palmer, it will be appropriate to review the exercise of executive power itself for constitutionality, rather than just the empowering provision. However, this raises what has been described as ‘one of the most exasperating analytical tangles of modern public law’[176] — how courts should approach review of an administrative decision which is alleged to have infringed on a protected right or freedom. This is a question which has vexed courts in Canada, New Zealand and the United Kingdom since their respective human rights instruments came into effect.[177] As Tom Hickman argues, a particular concern and challenge in those jurisdictions has been to find an approach which both ensures that the executive does not place unreasonable and unlawful limits on rights, at the same time as giving government decision-makers sufficient latitude to apply their expertise to assess what limits on rights are proportionate in the circumstances of a specific case.[178] Structured proportionality, developed and applied in review of legislation, is seen by some as poorly suited to some administrative review contexts, and as involving too much judicial intrusion into the ‘merits’ of administrative decision-making.[179]

As we suggested in Part III, we suspect that these same concerns underlie the development of the legislation-centric approach to constitutional freedoms in Australia. Although the Court has not said so directly, there is a longstanding suspicion of proportionality in Australian administrative law and understandable resistance to it developing as a distinct ‘ground of review’.[180] It is therefore useful to look at how overseas courts have attempted to address these concerns and resolve this ‘exasperating analytical tangle’, to see whether any of their approaches might fit within Australia’s constitutional context.

In this Part, we look at approaches in Canada and New Zealand. We do not consider the United Kingdom because the Human Rights Act 1998 (UK) expressly sets out how rights affect administrative discretions.[181] We also acknowledge that the constitutional and legal contexts in Canada and New Zealand are very different from that in Australia. Both the Canadian Charter of Rights and Freedoms (‘Canadian Charter’)[182] and the New Zealand Bill of Rights Act 1990 (NZ) (‘NZBORA’)[183] expressly state that they apply to the actions of governments as well as legislatures.[184] Both require courts to interpret legislation in a manner consistent with rights, where possible.[185] And both also contain provisions stating that the rights they protect can be subject to ‘reasonable limits prescribed by law’.[186] But neither expressly states how broadly framed statutory powers which, on their face, may be exercised in ways which limit rights, but which do not expressly ‘prescribe’ that rights may be limited in the exercise of discretion, ought to be approached. Taken together, this means that, at a minimum, statutory discretions cannot be exercised in a manner that places an unreasonable limit on rights.[187] This does not mean, however, that where a statute confers a discretion which may be exercised in a manner which places a reasonable limit on rights or in a manner which does not infringe rights at all, that the decision-maker is necessarily empowered to act in a rights-limiting manner.

Despite these different contexts, we suggest that it is useful to examine the approaches and reasoning adopted in Canada and New Zealand for two reasons. First, the courts in those jurisdictions have been (and still are) grappling with precisely the same questions as Australian courts have left unanswered in the cases set out in Part II.[188] Secondly, the lack of an individual rights framework in Australia does not, on its own, provide a compelling reason why Australian law must necessarily take a different position. For example, the ‘structured proportionality’ test that the majority of the High Court now uses for certain constitutional guarantees was adapted from a similar test used overseas (in the context of individual rights).[189] It is difficult to see a sensible explanation as to why Australian law would be able to borrow the framework developed in a rights-protection context to test the validity of legislation, but could not do the same when it comes to administrative discretion.

The case law from Canada and New Zealand reveals four distinct, but overlapping, methods for determining how statutory discretions are affected by constitutional rights. In the following discussion we explain each of them and analyse their benefits, problems, and potential utility in the Australian context.

A Focusing on the ‘Prescribed by Law’ Requirement

The first approach focuses on the requirement in the rights instruments that reasonable limits on rights must be ‘prescribed by law’,[190] and attempts to locate limits within the primary Act itself via interpretation. If the power to limit rights cannot be sourced in the primary Act, then the executive has no power to exercise discretion in a way which restricts rights at all — reasonably or otherwise. Clearly, this places significant constraints on the scope of executive power and, if applied generally and to its fullest extent, would result in many areas of administration becoming quite unworkable.

While some academics have advocated this approach, it has not been widely adopted by courts — probably for this very reason.[191] In Canada, for instance, the Supreme Court’s first consideration of this issue, in the case of Slaight Communications Inc v Davidson (‘Slaight’), left little scope for this approach.[192] Justice Lamer focused on principles of delegated power, holding that the fact that the legislature cannot make law which infringes the Canadian Charter means it also cannot delegate the power to do so to the executive.[193] In

other words,

the limitations on statutory authority which are imposed by the Charter will flow down the chain of statutory authority and apply to regulations, by-laws, orders, decisions and all other action (whether legislative, administrative or judicial) which depends for its validity on statutory authority.[194]

Since Slaight, the Canadian courts have used a range of methods to review administrative action which is alleged to infringe rights, but the idea that decision-makers are permitted to impose reasonable limits on rights when exercising broadly framed discretions, and that courts must test those limits in judicial review proceedings, has not been seriously questioned.

There have been a few hints at a ‘prescribed by law’ approach in Canadian cases. For instance, Deschamps and Abella JJ in Multani v Commission Scolaire Marguerite-Bourgeoys (‘Multani’) emphasised the phrase ‘prescribed by law’ and ‘règle de droit’ in s 1 of the Canadian Charter and said that this meant the justification process under s 1 ought only apply to legislation itself.[195] The result was that while administrative decision-makers must take Canadian Charter ‘values’ into account, ‘it does not follow that their decisions must be subjected to the justification process under s 1’.[196] They concluded that administrative decisions should be ‘reviewed in accordance with the principles of administrative law’.[197] These statements echo those from the High Court of Australia in the cases explained in Part II. However, much of Deschamps and Abella JJ’s reasoning on this point was openly pragmatic, and related to the fact that structured proportionality had proven difficult to apply to administrative decisions.[198] They also went on to find that the decision in question was unlawful, because the decision-maker had not sufficiently considered the right to freedom of religion, which indicates that rights were not seen as irrelevant to the exercise of discretion.[199]

There are some examples of this ‘prescribed by law’ approach from New Zealand, where courts have attempted to locate any limits on rights within the primary Act itself.[200] Where there is enough in the statutory text and context of the relevant powers, courts have been able to focus on interpreting statutory discretions in ways that do not unreasonably restrict rights. This means the legislation itself is subject to proportionality testing, whereas the exercise of discretion is subject to the ordinary principles of administrative law. A recent example of this approach is in some of the judgments in New Health New Zealand Inc v South Taranaki District Council.[201] The question for the Supreme Court of New Zealand was whether local authorities had breached s 11 of the NZBORA (the right to refuse medical treatment) by fluoridating water.[202] There was no legislative provision expressly authorising local councils to administer fluoride or any other medical treatment through the water supply, but O’Regan and Ellen France JJ found authority to do so was implied by the legislative history.[203] They went on to apply a structured proportionality analysis to the implied statutory power to fluoridate drinking water,[204] finding that it was a justified limitation.[205] The different approaches adopted by other judgments in the same case, discussed below, highlight that there is still no consensus in New Zealand about how rights affect statutory discretions.[206]

B Relying on Existing Administrative Law Grounds and Ignoring Rights

A second method evident in Canadian and New Zealand cases is to avoid answering difficult questions about how rights affect the scope of administrative discretions by starting with ‘ordinary’ administrative law principles, and hoping that these adequately resolve the case. This was the Court of Appeal of New Zealand’s approach in Drew v Attorney-General (NZ), which involved a challenge to regulations setting out the process for prison disciplinary hearings.[207] The regulations banned prisoners from having legal representation in disciplinary hearings, and were challenged as breaching common law natural justice principles as well as the right to justice under s 27(1) of the NZBORA.[208] The majority found the regulations unlawful based on common law principles, and considered there was no need to rely on

the NZBORA.[209]

There are numerous examples of this approach from Canada, including one of the most important Canadian administrative law cases — Baker v Minister of Citizenship and Immigration (Canada).[210] Baker challenged a decision to deport her on several bases, including the absence of procedural fairness, bias, a failure to give reasons, a failure to consider the interests of her Canadian-born children, and her right to security of the person under s 7 of the Canadian Charter.[211] The Court found the decision unlawful on administrative law grounds alone, considering it unnecessary to consider the Canadian Charter issues.[212] This approach — of deciding cases on common law administrative law grounds alone — was a deliberate choice by some judges, reflecting a concern that if the Canadian Charter became the superior and preferred avenue for judicial review, it would ‘sterilise’ and ‘impoverish’ administrative law.[213] However, this approach only works if a decision is unlawful on common law grounds, and so it will not be available in all cases.

C Structured Proportionality Testing of Discretionary Decisions

The final two approaches we have identified both begin with the premise that administrative decision-makers are empowered to exercise their discretions in ways which limit rights, provided those limits are reasonable. Legislation need not expressly or impliedly authorise decision-makers to limit rights; instead, it will be presumed that discretionary powers may limit rights, provided the limits are reasonable. As Paul Rishworth puts it, every statutory power ought to be read as subject to the limit ‘do not breach the [NZBORA] when you use

this power’.[214]

While these approaches place less emphasis on the requirement in human rights instruments that limits on rights must be ‘prescribed by law’, they are equally justifiable based on the principles of delegation set out by Lamer J in Slaight.[215] Indeed, Lamer J explained that ‘[l]egislation conferring an imprecise discretion must therefore be interpreted as not allowing the Charter rights to be infringed’.[216] This does not necessarily mean, though, that the exercise of discretion in individual cases must be reviewed by way of proportionality testing (structured or otherwise).[217] Limits on rights must be ‘reasonable

and ... demonstrably justified’,[218] and there are a variety of different methods courts may adopt to assess justification.

The first (or third of our four identified approaches) involves applying structured proportionality directly to administrative decisions. The High Court of New Zealand did this in Hudson v Attorney-General (NZ).[219] Hudson, serving a life sentence for murder, sought approval to receive two magazines: FHM and Ralph.[220] His request was denied by prison authorities on the basis that the magazines were ‘considered to be objectionable’ and hence restricted items under the Corrections Act 2004 (NZ) and prison property rules made thereunder.[221] Hudson argued that the decision of the prison authorities infringed his right to freedom of expression under s 14 of the NZBORA.[222] The Court applied the structured proportionality test set out in R v Hansen[223] directly to the prison’s application of the power. The test was applied very specifically to the issues of the magazines that Mr Hudson had requested, and the Court was satisfied that the prison’s concerns about prisoners accessing that material were made out, meaning the limit was justified.[224]

In Canada, this approach of applying structured proportionality directly to discretionary decisions was the ‘orthodox’ approach prior to 2012.[225] The Supreme Court of Canada applied the R v Oakes (‘Oakes’)[226] test directly to administrative decisions, effectively making it a distinct ‘ground’ of unlawfulness.[227] For example, in Multani, discussed above, a majority of the Court applied the Oakes analysis directly to the School Board’s decision to prohibit a Sikh student from carrying a kirpan.[228] After finding that the decision did limit the freedom of religion, the majority examined the objective of the Board (namely, school safety), the connection between the limit and that objective, and whether there were less restrictive means of achieving the objective.[229] They concluded that the total prohibition was disproportionate based on the evidence available as to the risk presented by students carrying kirpans.[230] In contrast, Deschamps and Abella JJ pointed to the longstanding difficulties of applying structured proportionality to administrative decisions.[231] In particular, they noted that administrative decisions do not always have an objective or purpose beyond administering the Act itself, and the outcome of many discretionary decisions is binary, meaning that the ‘least restrictive means’ analysis will not be applicable.[232] In addition, Deschamps and Abella JJ noted difficulties with the requirement that administrative decision-makers justify why the objectives of their decisions warrant limiting rights, where administrative decision-makers are not ‘parties with an interest in a dispute’.[233] As indicated in Part IV(A) above, their Honours concluded that administrative decisions should be reviewed in accordance with administrative law principles to avoid ‘any blurring of roles’.[234]

The difficulties raised by Deschamps and Abella JJ reflect a concern that structured proportionality review is fundamentally incompatible with the judicial role vis-a-vis the executive branch as it has traditionally developed.[235] The particular concern is that the test is too intrusive and involves courts performing an executive function by determining policy questions. This concern is particularly acute in the Australian context, given the strict separation of judicial power and the High Court’s approach to the judicial method.[236] Note, however, that these suggestions have not prevented courts in the Australian states and territories with statutory human rights instruments from engaging in structured proportionality review of administrative decisions.[237] Similar concerns have emerged in the New Zealand context. For instance, in Taylor v Chief Executive, Department of Corrections (‘Taylor’), the Court of Appeal was clear that the decision itself needed to be reviewed for compliance with the NZBORA, but queried whether structured proportionality was appropriate to perform this review.[238] The Court agreed that the ‘correct approach’ to reviewing the lawfulness of administrative decisions under the NZBORA ‘may depend on the particular context’.[239] The Court set out different methods which are less formal and capable of giving greater latitude to the executive branch.[240] Those broadly map onto the fourth and fifth approaches we outline below.

D Incorporating Rights into Reasonableness Review

The concerns identified by Deschamps and Abella JJ in Multani ultimately led the Supreme Court of Canada to adopt a different approach in the 2012 case of Doré v Barreau du Québec (‘Doré’).[241] That approach begins from the same premise articulated in Slaight: that decision-makers cannot exercise discretionary powers in a way which unreasonably limits protected rights, and that courts must review decisions to ensure that legal limit has not been breached.[242] But it does not see structured proportionality as the only method through which courts are able to perform that function. Nor does it see courts as having a monopoly on determining whether a limit on rights is justified. The Doré approach reflects a view that administrative and constitutional law issues overlap and are not distinct,[243] as it effectively incorporates the balancing muscles of proportionality within the existing administrative law reasonableness standard.[244] A reviewing court asks whether, in balancing the statutory objectives with the proposed limit on rights, the administrative decision ‘falls within a range of possible, acceptable outcomes’[245] and ‘reflects a proportionate balancing of the Charter protections at play’.[246]

One important point to note for Australian readers is that, in Canada, reasonableness is no longer conceived of as a standalone ‘ground’ of review. Rather, it is an overarching standard against which government decisions are assessed. All of the legal constraints on administrative power will be assessed on this standard, where it applies, with the exception of procedural fairness limits. In other words, an argument in judicial review proceedings might be framed as: the decision-maker acted unreasonably (and hence unlawfully) because they failed to explain the weight they gave to a particular factor.[247] The other standard on which Canadian courts review administrative action is correctness, which means the decision must be correct according to the reviewing court’s view. By contrast, the reasonableness test ‘is concerned mostly with the existence of justification, transparency and intelligibility within the decision-making process’.[248] In this way, ‘reasonableness’ explicitly incorporates a measure of ‘deference’ to the executive, while correctness does not.[249] A decision will be reasonable where the decision-maker has provided an intelligible justification. It will only be correct where it matches the court’s view of what the decision ought to have been.

The significance of the Supreme Court’s judgment in Doré — using the reasonableness standard and framework to review Canadian Charter questions in administrative decisions — is that it sets a lower threshold for lawfulness. Or, put differently, it accepts that there is scope for different, reasonable views as to when limits on rights are justified, and acknowledges that in conferring discretion on the executive, Parliament has given the executive the power to reach that view.

The Doré approach proved controversial amongst Canadian public law scholars, many of whom argue that it fails to give adequate protection to constitutional rights. Some have suggested that the approach effectively treats rights as no more than relevant considerations, which decision-makers must consider but not weigh any higher than competing factors.[250] This was also the view of the House of Lords, which led it to reject this approach.[251] There is a sound basis for those concerns, particularly given that courts themselves have, at times, conflated the ‘reasonableness’ approach with treating rights as a relevant consideration.[252] But we contend that, properly applied, the two are distinct. The essence of a threshold of reasonableness is that a court must be convinced that the decision-maker’s justification is transparent, intelligible and rational. While a reviewing court should accept a decision-maker’s reasonable justification even it would have reached a different result, it is not enough for a decision-maker to simply ‘consider’ relevant matters to meet the threshold of reasonable justification. They must explain why other factors have been given greater weight than the right, and why limits on the right are justified. Due to these concerns and criticisms, the Supreme Court of Canada has since resiled from the Doré approach, at times explicitly,[253] and at times by adopting an approach to reasonableness review that looks more like structured proportionality (though retaining the veneer of deference).[254] But Doré remains the majority approach in Canada, at least in theory.

A similar approach to Doré has been applied in some cases in New Zealand. For example, in Television New Zealand Ltd v Attorney-General (NZ), the Court of Appeal found that a decision to refuse an inmate’s application to take part in a television interview was unlawful, because it unreasonably infringed the right to freedom of expression.[255] The Court was clear that the regulations themselves were within power, and that the decision itself needed to be reviewed for compliance with the NZBORA.[256] The Court did not give any detailed explanation of its chosen method for reviewing the Chief Executive’s refusal, but its approach was quite clearly one of unstructured proportionality review. Like the Supreme Court of Canada in Doré, the New Zealand Court of Appeal based its assessment on the reasons given by the government.[257] In Taylor, the Court of Appeal gave more considered treatment to the question of which test to apply.[258] That case also involved a decision to refuse a prisoner’s request to give a media interview. As mentioned above, the Court indicated that different approaches might be required in different administrative decision-making contexts.[259] In the context of the present case, the Court held that a less formal approach to balancing was appropriate.[260] Confusingly, however, the Court also described this approach as one in which free expression was an implied ‘mandatory consideration’ in the statutory context.[261]

There are several other New Zealand cases which likewise refer to NZBORA rights as mandatory considerations for decision-makers. For example, Wild J in Television New Zealand Ltd v Viewers for Television Excellence Inc said that ‘the Authority was required to take into account the right in question ... as a mandatory relevant consideration’.[262] Similarly, in Smith v Attorney-General (NZ), Wylie J indicated that NZBORA rights ‘must be read into the exercise of statutory powers of decision’.[263] These cases have been criticised on the basis that treating rights as mere considerations denies their status as ‘substantive’ limits on power and reduces rights to procedures.[264] However, as Hanna Wilberg has shown, these references in NZBORA cases are not to ‘relevant considerations’ in the sense traditionally meant in administrative review.[265] Instead, the New Zealand cases, like Doré, treat the balancing of rights as a mandatory consideration — or, more accurately, task — of decision-makers, rather than simply something to be taken into account, with decision-makers required to provide a reasonable justification for the balance they have struck.[266] As Wilberg shows, the approach differs from the structured proportionality approach only in terms of the role of the court and the standard applied, and not in the fundamental treatment of rights as substantive limits on executive power.[267]

Several of the Australian judgments we discussed in Part II(B) indicated that constitutional freedoms might be accommodated within administrative review by treating them as ‘relevant considerations’.[268] However, Gageler J and Edelman J have dismissed this suggestion as conceptually confusing and diminishing the role of the implied constitutional limit.[269] So, too, have

Bret Walker and David Hume, who explain that the implied freedom ‘operates as an ultimate limit on ... power, and an ultimate limit on power cannot sensibly be described as a mandatory consideration’.[270] We agree with those criticisms. But the Canadian and New Zealand cases show that there is an important difference between requiring a decision-maker to consider rights, and requiring a decision-maker to balance rights and justify that balance. The latter correctly treats rights as a substantive limit on the exercise of statutory executive power, but also acknowledges that decision-makers must only meet a reasonableness threshold to stay within this limit. It gives scope to the executive to exercise discretion and expertise without judicial intervention, provided that the executive provides a rational justification.

Despite differences in its constitutional context, we suggest Australia could adopt a similar approach to review of administrative decisions which limit a constitutional freedom. Under this approach, the court asks the same fundamental question, and applies the same legal standard, that Australian courts already do in assessing whether a decision-maker has breached the presumed limit on their power that it be exercised reasonably.[271] If a decision-maker has exercised discretion in a way which limits a constitutional freedom, has not justified the need for the limit, and no reasonable justification is apparent, then the decision could and should be regarded as legally unreasonable and beyond power. This reflects the idea that constitutional freedoms limit the scope of discretionary powers — that discretions may only be exercised to restrict a freedom where that restriction is reasonably justified. This is not a radical suggestion, but one which could easily be accommodated within Australia’s existing administrative law framework.

V CONCLUSION

The High Court’s current approach to statutory discretions which limit constitutional freedoms is to review the statutory provision which confers discretion for constitutional validity, but to make no further inquiries about the exercise of discretion in the particular case. In this article we have queried whether this legislation-centric approach is doctrinally coherent, justified, sufficient to give effect to constitutional guarantees, and practically workable. We have identified a number of reasons why it may not always be an appropriate or sufficient method for ensuring that the political branches have acted within the constitutional limits of their authority. We have argued that, in some cases, courts should review individual administrative decisions for compliance with constitutional freedoms.

We then investigated how courts might take account of constitutional guarantees in the context of particular exercises of statutory discretions, and examined the various approaches taken in Canada and New Zealand for guidance. Our analysis of those jurisdictions shows that there remains no single, accepted approach to this question, despite their having more than 30 years of jurisprudence on the topic. This strongly suggests that it may not be possible to use one single method to review all types of administrative discretions. However, our analysis also shows that there are approaches available which address the difficulties we identified with the legislation-centric approach, are consistent with Australia’s existing administrative law framework, and avoid the creation of a new and potentially problematic ground of ‘structured proportionality’ review in administrative law. Specifically, it is entirely coherent and workable to suggest that constitutional freedoms form part of the legal context for what is a legally reasonable exercise of a statutory discretion, and that courts can and should address this limit directly in appropriate cases. Accommodating constitutional freedoms in the concept of legal unreasonableness would ensure that statutory discretions are exercised consistently with the Constitution’s express and implied guarantees.


[*] Associate Professor, Faculty of Law and Justice, University of New South Wales. Thanks to Lisa Burton Crawford, Matthew Groves, Dan Meagher, Amanda Sapienza and Shreeya Smith for comments and helpful discussions, to the journal’s reviewers for their valuable feedback and to Rhyse Collins for research assistance.

[†] Senior Lecturer, Deakin Law School, Deakin University.

[1] See Palmer v Western Australia [2021] HCA 5; (2021) 388 ALR 180, 196–7 [63]–[68] (Kiefel CJ and Keane J), 218 [117]–[119] (Gageler J), 229–30 [200]–[202] (Gordon J) (‘Palmer’); Comcare v Banerji [2019] HCA 23; (2019) 267 CLR 373, 405–6 [44]–[45] (Kiefel CJ, Bell, Keane and Nettle JJ), 421–2 [96] (Gageler J), 458–9 [209] (Edelman J) (‘Banerji’); Wotton v Queensland (2012) 246 CLR 1, 13–14 [21]–[23] (French CJ, Gummow, Hayne, Crennan and Bell JJ) (‘Wotton’). The majority of the High Court adopted a legislation-centric approach in Commonwealth v AJL20 (2021) 391 ALR 562, 574 [43] (Kiefel CJ, Gageler, Keane and Steward JJ) (‘AJL20’), and referred to the constitutional freedom cases in doing so. However, that case involved the issue of non-criminal, executive detention, which raises different constitutional questions intertwined with the heads of legislative power: see Jeffrey Steven Gordon, ‘Imprisonment and the Separation of Judicial Power: A Defence of a Categorical Immunity from Non-Criminal Detention’ [2012] MelbULawRw 2; (2012) 36(1) Melbourne University Law Review 41, 46–51. The validity of executive detention also rests on the purposes of the detention, and administrative law principles and remedies already ensure that executive powers are exercised only for their authorised purpose. No executive discretion is involved. Hence there is no risk of the remedial gaps we identify in this article. As such, we do not address AJL20 (n 1) further in this article. The High Court has not considered the issues we discuss in this article in relation to the implied right to vote, so we do not address that implied right here. Note that in this article we focus on statutory executive power rather than executive power more generally. The exercise of non-statutory executive power raises some distinct issues, and the application of the grounds of review to non-statutory executive power remains unsettled: see Amanda Sapienza, Judicial Review of Non-Statutory Executive Action (Federation Press, 2020) 4–5 (‘Judicial Review’). For some discussion of non-statutory executive power, see below nn 56, 153–5 and accompanying text.

[2] See McCloy v New South Wales [2015] HCA 34; (2015) 257 CLR 178, 193–5 [2] (French CJ, Kiefel, Bell and Keane JJ) (‘McCloy’); Palmer (n 1) 193–5 [51]–[58] (Kiefel CJ and Keane J), 246–8 [261]–[265] (Edelman J). Though we note that there remain strong objections to the use of structured proportionality: see, eg, at 213–16 [143]–[151] (Gageler J), 228 [198] (Gordon J). See generally Adrienne Stone, ‘Proportionality and Its Alternatives’ (2020) 48(1) Federal Law Review 123, 127–31 (‘Proportionality’); Anne Carter, Proportionality and Facts in Constitutional Adjudication (Hart Publishing, 2022) ch 5; Mark Watts, ‘Reasonably Appropriate and Adapted? Assessing Proportionality and the “Spectrum” of Scrutiny in McCloy v New South Wales[2016] UQLawJl 22; (2016) 35(2) University of Queensland Law Journal 349. We do not enter into this debate in this article. For simplicity, we use the term ‘proportionality’ to describe the various ‘reasonable limits’, ‘appropriate and adapted’ and ‘structured proportionality’ tests that members of the Court prefer in this context. We use ‘structured proportionality’ to refer specifically to the multi-step test set out in McCloy (n 2) 193–5 [2] (French CJ, Kiefel, Bell and Keane JJ). On different proportionality tests used in Australian constitutional law, see generally Shipra Chordia, Proportionality in Australian Constitutional Law (Federation Press, 2020) ch 9.

[3] See, eg, Banerji (n 1) 394–5 [19] (Kiefel CJ, Bell, Keane and Nettle JJ); Palmer (n 1) 191 [41] (Kiefel CJ and Keane J), 204 [105] (Gageler J).

[4] [1986] HCA 60; (1986) 161 CLR 556, 596 (‘Miller’).

[5] James v Commonwealth [1935] HCA 38; (1935) 52 CLR 570, 602 (emphasis added).

[6] [1988] HCA 18; (1988) 165 CLR 360, 409 (Mason CJ, Wilson, Brennan, Deane, Dawson, Toohey and Gaudron JJ) (emphasis added) (‘Cole’).

[7] Ibid 380–2.

[8] Justice Stephen Gageler, ‘The Section 92 Revolution’ in James Stellios (ed), Encounters with Constitutional Interpretation and Legal Education: Essays in Honour of Michael Coper (Federation Press, 2018) 26, 28.

[9] James Stellios, ‘Marbury v Madison: Constitutional Limitations and Statutory Discretions’ (2016) 42(3) Australian Bar Review 324, 327–8 (‘Marbury v Madison’).

[10] Ibid 328. See also James Stellios, Zines’s The High Court and the Constitution (Federation Press, 6th ed, 2015) 162–4.

[11] Stellios, ‘Marbury v Madison’ (n 9) 328.

[12] Miller (n 4) 607 (Brennan J). See also Brennan J’s discussion of further case law: at 605–9.

[13] Ibid 612.

[14] Ibid 613–14, quoting Wilcox Mofflin Ltd v New South Wales [1952] HCA 17; (1952) 85 CLR 488, 522 (Dixon, McTiernan and Fullagar JJ).

[15] Miller (n 4) 613–14.

[16] Ibid 614.

[17] We will use the terms ‘administrative law’ and ‘administrative review’ as shorthand to mean judicial review of administrative action in this article. We use ‘constitutional review’ to refer to judicial review of constitutional validity.

[18] Chief Justice Mason made a similar point in the implied freedom case of Cunliffe v Commonwealth [1994] HCA 44; (1994) 182 CLR 272, stating that the availability of merits and judicial review ‘will ultimately ensure that decisions of the [Migration Agents Registration] Board will conform to constitutional requirements’: at 303.

[19] [1994] HCA 46; (1994) 182 CLR 104, 125 (Mason CJ, Toohey and Gaudron JJ) (emphasis added) (‘Theophanous’). See also at 149 (Brennan J), 168 (Deane J). Other early statements include, for example, Nationwide News Pty Ltd v Wills [1992] HCA 46; (1992) 177 CLR 1, 50–1 (Brennan J) (‘Nationwide News’).

[20] [1997] HCA 25; (1997) 189 CLR 520, 560 (Brennan CJ, Dawson, Toohey, Gaudron, McHugh, Gummow and Kirby JJ) (emphasis added) (‘Lange’). See also Australian Broadcasting Corporation v Lenah Game Meats Pty Ltd (2001) 208 CLR 199, 219–20 [20] (Gleeson CJ) (‘Lenah Game Meats’).

[21] McCloy (n 2) 206 [42] (French CJ, Kiefel, Bell and Keane JJ), 280 [303] (Gordon J).

[22] Ibid 227 [115]. See also Tajjour v New South Wales [2014] HCA 35; (2014) 254 CLR 508, 558 [59] (Hayne J), 577 [140] (Gageler J), 593 [195]–[196] (Keane J) (‘Tajjour’); Brown v Tasmania (2017) 261 CLR 328, 430 [312]–[313] (Gordon J) (‘Brown’).

[23] An issue which we explore in Part III(D) below.

[24] See McCloy (n 2) 193–5 [2] (French CJ, Kiefel, Bell and Keane JJ).

[25] Although, as we note below, the Court has applied the implied freedom to delegated legislation: see below nn 133–4 and accompanying text.

[26] Wotton (n 1) 11–12 [15]–[16] (French CJ, Gummow, Hayne, Crennan and Bell JJ).

[27] Ibid 13–15 [20]–[26], 16 [31], [33] (French CJ, Gummow, Hayne, Crennan and Bell JJ), 23–4 [54] (Heydon J), 30 [74] (Kiefel J).

[28] Ibid 14 [22]–[24] (French CJ, Gummow, Hayne, Crennan and Bell JJ).

[29] Ibid 9–10 [10] (French CJ, Gummow, Hayne, Crennan and Bell JJ), quoting with minor changes Miller (n 4) 613–14 (Brennan J).

[30] Ibid 14 [21]–[23].

[31] Ibid 14 [22] (French CJ, Gummow, Hayne, Crennan and Bell JJ). The other judges did not differ on this point: cf at 23–4 [54] (Heydon J), 30 [74] (Kiefel J). See also at 5.

[32] Ibid 14 [24] (French CJ, Gummow, Hayne, Crennan and Bell JJ).

[33] Which was the way the Court ascertained whether the implied freedom had been breached at the time, before McCloy (n 2).

[34] Wotton (n 1) 14 [24] (French CJ, Gummow, Hayne, Crennan and Bell JJ), citing Judicial Review Act 1991 (Qld).

[35] Wotton (n 1) 9–10 [10] (French CJ, Gummow, Hayne, Crennan and Bell JJ).

[36] Ibid 14 [23] (emphasis added).

[37] Ibid 16 [31].

[38] Ibid 4 (Hayne J) (during argument).

[39] (2017) 246 FCR 298, 315–17 [73]–[80] (Perram, Mortimer and Gleeson JJ) (‘Gaynor’).

[40] Ibid 302 [11], 303 [16].

[41] Ibid 303 [17].

[42] Gaynor v Chief of the Defence Force [No 3] [2015] FCA 1370; (2015) 237 FCR 188, 236 [181] (Buchanan J).

[43] Ibid 256 [289]–[290].

[44] Ibid 233–4 [175], 256 [290] (Buchanan J), quoting Administrative Decisions (Judicial Review) Act 1977 (Cth) ss 5(1)(d)–(e), (j) (‘AD(JR) Act’).

[45] Gaynor (n 39) 310 [47] (Perram, Mortimer and Gleeson JJ).

[46] Ibid 313 [60], 313–14 [63].

[47] Ibid 315 [71]–[72].

[48] Ibid 317 [79].

[49] See, eg, ibid 315 [72].

[50] Ibid 314–15 [68].

[51] Ibid 310 [45].

[52] Ibid 315–16 [73].

[53] Ibid 316 [75], quoting Wotton (n 1) 13–14 [21] (French CJ, Gummow, Hayne, Crennan and Bell JJ).

[54] Gaynor (n 39) 317 [80] (Perram, Mortimer and Gleeson JJ).

[55] Ibid.

[56] Ibid 314–15 [68].

[57] For an analysis of this issue, see generally Gerard Carney, ‘A Comment on How the Implied Freedom of Political Communication Restricts Non-Statutory Executive Power’ [2018] UWALawRw 23; (2018) 43(2) University of Western Australia Law Review 255.

[58] See, eg, Minister for Home Affairs v Benbrika [2021] HCA 4; (2021) 388 ALR 1, 23 [67], 25 [74] (Gageler J); MZAPC v Minister for Immigration and Border Protection (2021) 390 ALR 590, 612–15 [93]–[103] (Gordon and Steward JJ). See generally KM Hayne, ‘Executive Power’ (2017) 28(3) Public Law Review 236, 243–6, describing non-statutory executive powers as a subset of executive power.

[59] See generally Cheryl Saunders, ‘Separation of Legislative and Executive Power’ in Cheryl Saunders and Adrienne Stone (eds), The Oxford Handbook of the Australian Constitution (Oxford University Press, 2018) 617, 628–9; Anne Twomey, ‘Pushing the Boundaries of Executive Power: Pape, the Prerogative and Nationhood Powers’ [2010] MelbULawRw 9; (2010) 34(1) Melbourne University Law Review 313, 316.

[60] See, eg, Theophanous (n 19) 168 (Deane J); Lenah Game Meats (n 20) 219–20 [20] (Gleeson CJ); Tajjour (n 22) 558 [59] (Hayne J), 577 [140] (Gageler J), 593 [195] (Keane J); McCloy (n 2) 227 [114]–[115] (Gageler J), quoting Nationwide News (n 19) 51 (Brennan J); Brown (n 22) 430 [312]–[313] (Gordon J).

[61] Banerji (n 1).

[62] Ibid 389 [2], 392 [9] (Kiefel CJ, Bell, Keane and Nettle JJ).

[63] More specifically, Banerji challenged Comcare’s rejection of her claim for compensation for the psychological injury suffered as a result of her dismissal: ibid 392 [10]–[11] (Kiefel CJ, Bell, Keane and Nettle JJ).

[64] Ibid 394–5 [18]–[19] (Kiefel CJ, Bell, Keane and Nettle JJ).

[65] Ibid 394–5 [19].

[66] Ibid 395 [20] (emphasis in original) (citations omitted).

[67] Ibid 405 [43].

[68] Ibid 406 [45].

[69] Ibid 405 [44] (emphasis added) (citations omitted).

[70] Ibid 406 [44] (citations omitted).

[71] [2012] HCA 61; (2013) 249 CLR 332, 362 [63] (Hayne, Kiefel and Bell JJ) (‘Li’).

[72] [2018] HCA 30; (2018) 264 CLR 541 (‘SZVFW’). See, eg, at 550–1 [10] (Kiefel CJ), 564–7 [51]–[60] (Gageler J).

[73] Banerji (n 1) 408 [51] (Gageler J), quoting A v Independent Commission against Corruption [2014] NSWCA 414; (2014) 88 NSWLR 240, 256 [56] (Basten JA) (‘A v ICAC’).

[74] Banerji (n 1) 408 [52] (Gageler J), quoting with minor changes A v ICAC (n 73) 256–7 [56] (Basten JA). See also Banerji (n 1) 459 [211] (Edelman J).

[75] Banerji (n 1) 421 [96] (Gageler J).

[76] Ibid 422 [96].

[77] Ibid 459 [210].

[78] Ibid 458–9 [209]–[211].

[79] Palmer (n 1).

[80] See ibid 183–5 [1]–[15] (Kiefel CJ and Keane J).

[81] Section 67 of the Emergency Management Act 2005 (WA) relevantly provided that, during a state of emergency (which the Minister had declared existed due to COVID-19), an authorised officer (which the State Emergency Coordinator was) could ‘direct or, by direction, prohibit, the movement of persons, animals and vehicles within, into, out of or around an emergency area or any part of the emergency area’. See also Palmer (n 1) 183–5 [1]–[15] (Kiefel CJ and Keane J).

[82] Transcript of Proceedings, Palmer v Western Australia [2020] HCATrans 178, 32 (Kiefel CJ, Edelman J and PJ Dunning QC). See also Palmer (n 1) 234–5 [225] (Edelman J).

[83] Clive Frederick Palmer and Mineralogy Pty Ltd, ‘Plaintiffs’ Submissions’, Submission in Palmer v Western Australia, B26/2020, 22 September 2020, 18 [55].

[84] Transcript of Proceedings, Palmer v Western Australia [2020] HCATrans 179, 28 (PJ Dunning QC).

[85] Palmer (n 1) 196 [65].

[86] Ibid 197 [68].

[87] Ibid.

[88] Ibid 199 [77].

[89] Ibid 199 [79].

[90] See above n 2.

[91] Palmer (n 1) 208 [118].

[92] Ibid 208 [117]–[118].

[93] Ibid 208 [119].

[94] See, eg, Cole (n 6) 379–80 (Mason CJ, Wilson, Brennan, Deane, Dawson, Toohey and Gaudron JJ); Levy v Victoria [1997] HCA 31; (1997) 189 CLR 579, 596–7 (Brennan CJ) (‘Levy’); A-G (SA) v Adelaide City Corporation [2013] HCA 3; (2013) 249 CLR 1, 18 [9] (French CJ) (‘Adelaide City’); Gaynor (n 39) 300 [1] (Perram, Mortimer and Gleeson JJ).

[95] Palmer (n 1) 208 [120]–[121].

[96] Ibid 209 [122].

[97] Ibid 209 [123].

[98] Ibid.

[99] Ibid 210 [126].

[100] Justice Gordon’s approach was similar to Gageler J’s, in that she focused on the statute rather than its specific application, and relied on Wotton (n 1) as justification: Palmer (n 1) 229 [201].

[101] Palmer (n 1) 216 [154], 218–19 [166].

[102] Ibid 233 [219] (Edelman J).

[103] See Banerji (n 1) 407–8 [50]–[52] (Gageler J).

[104] Palmer (n 1) 234–5 [224]–[225].

[105] Ibid 235 [227] (citations omitted).

[106] Ibid 236–7 [230]–[231].

[107] Ibid 237 [231].

[108] Ibid.

[109] See, eg, Adelaide City (n 94) 88–90 [217]–[222] (Crennan and Kiefel JJ), cited in Palmer (n 1) 237 [231] (Edelman J).

[110] The approach was also followed in Cotterill v Romanes [2021] VSC 498; (2021) 360 FLR 341, where Niall JA held that the implied freedom should be analysed at the level of the legislation, rather than the Directions: at 381 [197], 383 [205], 384 [211].

[111] See Wotton (n 1) 9–10 [9]–[10], 13–14 [21]–[23] (French CJ, Gummow, Hayne, Crennan and Bell JJ).

[112] See Palmer (n 1) 235 [227] (Edelman J).

[113] See Banerji (n 1) 421–2 [96] (Gageler J).

[114] Palmer (n 1) 236–8 [230]–[234] (Edelman J).

[115] On the reluctance of Australian courts to incorporate proportionality as a ground of review in administrative law proceedings, see generally Janina Boughey, ‘The Reasonableness of Proportionality in the Australian Administrative Law Context’ (2015) 43(1) Federal Law Review 59 (‘Reasonableness of Proportionality’).

[116] See, eg, Miller (n 4) 609–10 (Brennan J); Betfair Pty Ltd v Racing New South Wales [2012] HCA 12; (2012) 249 CLR 217, 266–7 [42]–[44] (French CJ, Gummow, Hayne, Crennan and Bell JJ) (‘Betfair’); Gaynor (n 39) 305 [27] (Perram, Mortimer and Gleeson JJ); Banerji (n 1) 394–6 [19]–[20] (Kiefel CJ, Bell, Keane and Nettle JJ); Australian Broadcasting Corporation v Kane [No 2] [2020] FCA 133; (2020) 377 ALR 711, 766 [258], 781 [338] (Abraham J) (‘Kane’); Palmer (n 1) 204–5 [105] (Gageler J), 222 [180] (Gordon J).

[117] Betfair (n 116) 282 [91] (Kiefel J).

[118] See, eg, Lange (n 20) 560 (Brennan CJ, Dawson, Toohey, Gaudron, McHugh, Gummow and Kirby JJ). See also the cases discussed in Part II above.

[119] Lange (n 20) 556, 566. See also Theophanous (n 19) 178 (Deane J); Stephens v West Australian Newspapers Ltd [1994] HCA 45; (1994) 182 CLR 211, 255 (Brennan J).

[120] See Peter Hanks, Frances Gordon and Graeme Hill, Constitutional Law in Australia (LexisNexis Butterworths, 4th ed, 2018) 696 [10.237].

[121] Joshua Forrester, Lorraine Finlay and Augusto Zimmermann, ‘Finding the Streams’ True Sources: The Implied Freedom of Political Communication and Executive Power’ [2018] UWALawRw 22; (2018) 43(2) University of Western Australia Law Review 188, 200. See also Sapienza, Judicial Review (n 1) 14.

[122] McCloy (n 2) 227 [114].

[123] Williams v Commonwealth [2012] HCA 23; (2012) 248 CLR 156, 184 [22] (French CJ) (‘Williams’).

[124] Cheryl Saunders, ‘Intergovernmental Agreements and the Executive Power’ (2005) 16(4) Public Law Review 294, 306 (‘Intergovernmental Agreements’).

[125] Forrester, Finlay and Zimmermann (n 121) 207.

[126] Williams (n 123) 371 [585] (Kiefel J). See also at 251 [197] (Hayne J).

[127] [2009] HCA 23; (2009) 238 CLR 1, 63 [132] (French CJ), 83 [214], 85 [220], 87 [228], 89 [234] (Gummow, Crennan and Bell JJ), 119 [336]–[338] (Hayne and Kiefel JJ), 190 [541] (Heydon J).

[128] Ibid 60 [127] (French CJ).

[129] See, eg, Saunders, ‘Intergovernmental Agreements’ (n 124) 306, 312; Stellios, Zines’s The High Court and the Constitution (n 10) 162–4, 350–1, 581–4; Stellios, ‘Marbury v Madison’ (n 9) 326; Carney (n 57) 255–6.

[130] As Gerard Carney notes, ‘the vast grant of statutory power to the Executive ... easily attracts the implied freedom as a restriction on the exercise of delegated legislative power’: Carney (n 57) 259.

[131] This general principle was re-asserted in Australian Communist Party v Commonwealth [1951] HCA 5; (1951) 83 CLR 1: see, eg, at 258 (Fullagar J). James Stellios details the implications of the principle that ‘a stream cannot rise higher than its source’ for statutory executive power in Stellios, Zines’s The High Court and the Constitution (n 10) 346–60.

[132] As Stellios notes, ‘[i]t is quite clear that if a Parliament cannot discriminate against interstate trade with a protectionist purpose or effect, it cannot authorise a person or tribunal to do so’: Stellios, Zines’s The High Court and the Constitution (n 10) 188.

[133] Levy (n 94) 596–7 (Brennan CJ).

[134] Adelaide City (n 94) 17 [7] (French CJ).

[135] See above nn 95–6, 106 and accompanying text.

[136] Gaynor (n 39) 305 [27] (Perram, Mortimer and Gleeson JJ). See similar statements in Miller (n 4) 609–10 (Brennan J); Betfair (n 116) 266–7 [42]–[44] (French CJ, Gummow, Hayne, Crennan and Bell JJ); Banerji (n 1) 394–6 [19]–[20] (Kiefel CJ, Bell, Keane and Nettle JJ); Kane (n 116) 766 [258], 781 [338] (Abraham J); Palmer (n 1) 204–5 [105] (Gageler J), 222 [180] (Gordon J).

[137] See generally Stone, ‘Proportionality’ (n 2) 133–4. See also Lael Weis’s interesting commentary on ‘individual’ and ‘structural’ rights: Lael Weis, ‘McCloy Symposium: Lael Weis on Why Political Communication Isn’t an Individual Right in Australia’, Opinions on High (Blog Post, 19 October 2015) <https://blogs.unimelb.edu.au/opinionsonhigh/2015/10/19/weis-mccloy>, archived at <https://perma.cc/5YXV-DTUM>.

[138] (2013) 252 CLR 530, 554 [36]–[37] (French CJ, Hayne, Crennan, Kiefel and Bell JJ) (‘Unions NSW’).

[139] Li (n 71) 362 [63] (Hayne, Kiefel and Bell JJ).

[140] (1990) 170 CLR 1, 36 (Brennan J) (‘Quin’).

[141] See, eg, Wotton (n 1) 31 [80] (Kiefel J); Unions NSW (n 138) 553–4 [35]–[36] (French CJ, Hayne, Crennan, Kiefel and Bell JJ); Brown (n 22) 360 [90] (Kiefel CJ, Bell and Keane JJ).

[142] See Part III(D) below. See also Stellios, ‘Marbury v Madison’ (n 9) 331.

[143] See, eg, Wotton (n 1) 14 [22] (French CJ, Gummow, Hayne, Crennan and Bell JJ); Gaynor (n 39) 317 [79] (Perram, Mortimer and Gleeson JJ); Banerji (n 1) 408 [51]–[52] (Gageler J).

[144] Wotton (n 1) 14 [22] (French CJ, Gummow, Hayne, Crennan and Bell JJ).

[145] Ibid 14 [24], 16 [31].

[146] It is quite consistent with the statements of Brennan J and Mason CJ discussed above in Part II(A) and nn 14–20 and accompanying text.

[147] For example, s 30(a) of the Judiciary Act 1903 (Cth) (‘Judiciary Act’) confers original jurisdiction on the High Court in ‘matters arising under the Constitution’, while s 75(v) of the Constitution confers jurisdiction to hear matters in which certain remedies are sought against officers of the Commonwealth. Similarly, the Federal Court’s jurisdiction over matters arising under the Constitution or arising under Commonwealth laws, and those involving challenges to the actions of Commonwealth officers can be found in different subsections of the Judiciary Act (n 147): at ss 39B(1)–(1A).

[148] Questions about the lawfulness and constitutionality of government action may reach the courts in various ways. For instance, in Banerji (n 1), the constitutional argument about the implied freedom was raised as part of Banerji’s challenge to Comcare’s decision to refuse compensation for the termination of her employment, and not in an administrative law challenge to the termination itself: at 392 [10]–[11] (Kiefel CJ, Bell, Keane and Nettle JJ), 407 [49]–[50] (Gageler J).

[149] [2016] HCA 1; (2016) 257 CLR 42, 64–5 [16]–[21] (French CJ, Kiefel and Nettle JJ).

[150] Ibid.

[151] [2015] HCA 1; (2015) 255 CLR 514, 547 [69], 558 [115] (Hayne and Bell JJ), 584 [215] (Crennan J).

[152] Palmer (n 1) 209 [122].

[153] See generally Sapienza, Judicial Review (n 1) 119–23.

[154] Quin (n 140) 35–6.

[155] Probuild Constructions (Aust) Pty Ltd v Shade Systems Pty Ltd [2018] HCA 4; (2018) 264 CLR 1, 13 [29] (Kiefel CJ, Bell, Keane, Nettle and Gordon JJ).

[156] Janina Boughey and Lisa Burton Crawford, ‘Executive Power in an Age of Statutes’ in Janina Boughey and Lisa Burton Crawford (eds), Interpreting Executive Power (Federation Press, 2020) 1.

[157] See, eg, L v South Australia (2017) 129 SASR 180, 183–4 [4]–[5] (Kourakis CJ); Jabbour v Secretary, Department of Home Affairs [2019] FCA 452; (2019) 269 FCR 438, 458 [91] (Robertson J). On the United Kingdom position, see generally Lord Woolf et al, De Smith’s Judicial Review (Sweet & Maxwell, 8th ed, 2018) 135–7 [3-039]–[3-040]. The High Court has not expressly adopted this position, but leading commentators argue that it should: Mark Aronson, Matthew Groves and Greg Weeks, Judicial Review of Administrative Action and Government Liability (Lawbook, 6th ed, 2017) 122–6 [3.60].

[158] Amanda Sapienza, ‘Interpreting the Limits of Non-Statutory Executive Action: What Role for Grounds of Judicial Review?’ in Janina Boughey and Lisa Burton Crawford (eds), Interpreting Executive Power (Federation Press, 2020) 222, 234.

[159] See, eg, Plaintiff S10/2011 v Minister for Immigration and Citizenship (2012) 246 CLR 636, 666 [97] (Gummow, Hayne, Crennan and Bell JJ).

[160] Justice John Basten, ‘The Foundations of Judicial Review: The Value of Values’ [2020] AIAdminLawF 20; (2020) 100 AIAL Forum 32, 41. See generally Mark Aronson, ‘Judicial Review of Administrative Action: Between Grand Theory and Muddling Through’ (2021) 28(1) Australian Journal of Administrative Law 6.

[161] See above nn 1, 39–40 and accompanying text.

[162] Janina Boughey, Ellen Rock and Greg Weeks, Government Liability: Principles and Remedies (LexisNexis Butterworths, 2019) 102–3 [4.4]; Will Bateman and Leighton McDonald, ‘The Normative Structure of Australian Administrative Law’ (2017) 45(2) Federal Law Review 153, 153–4.

[163] See Gaynor (n 39) 326–34 [124]–[160] (Perram, Mortimer and Gleeson JJ); Wotton (n 1) 10 [13] (French CJ, Gummow, Hayne, Crennan and Bell JJ).

[164] See Gaynor (n 39) 317 [80] (Perram, Mortimer and Gleeson JJ).

[165] This is reflected in Brennan J’s discussion in Miller (n 4), where he clearly saw s 92 as relevant to framing the limits on statutory discretion that could be challenged via administrative law: at 613–14. See generally Stellios, ‘Marbury v Madison’ (n 9) 330.

[166] Banerji (n 1) 421–2 [96] (Gageler J).

[167] Kieran Pender, ‘Comcare v Banerji: Public Servants and Political Communication’ [2019] SydLawRw 5; (2019) 41(1) Sydney Law Review 131, 142.

[168] Chief Justice Kiefel and Keane J’s judgment did so implicitly due to the fact that their analysis did centre around the specific application of the law to COVID-19: Palmer (n 1) 198–9 [76]–[77].

[169] See especially ibid 235 [227] (Edelman J).

[170] Migration Act 1958 (Cth) s 501(6)(c). Compare this discretion, though, with that in s 501 (6)(d): a person’s visa may be cancelled if ‘there is a risk that the person would ... incite discord in the Australian community or in a segment of that community’. This provision quite clearly authorises the executive to act in a manner which restricts free expression, but places a clear condition on the discretion. A court is capable of interpreting the term ‘discord’ so as to limit the discretion to proportionate limits on free expression, as the Full Federal Court did in a different statutory context in Evans v New South Wales [2008] FCAFC 130; (2008) 168 FCR 576, 597 [83], 599 [88] (French, Branson and Stone JJ).

[171] Palmer (n 1) 183 [1], 184 [7] (Kiefel CJ and Keane J).

[172] See Janina Boughey, ‘Executive Power in Emergencies: Where Is the Accountability?’ (2020) 45(3) Alternative Law Journal 168, 171–2.

[173] The difficulty in distinguishing ‘soft law’ from delegated legislation has been well explored: see Greg Weeks, Soft Law and Public Authorities: Remedies and Reform (Hart Publishing, 2016) 15–24.

[174] Palmer (n 1) 208–9 [121] (Gageler J), 237 [231] (Edelman J).

[175] On the latter, see generally Matthew Groves, ‘The Return of the (Almost) Absolute Statutory Discretion’ in Janina Boughey and Lisa Burton Crawford (eds), Interpreting Executive Power (Federation Press, 2020) 129.

[176] Susan L Gratton and Lorne Sossin, ‘In Search of Coherence: The Charter and Administrative Law under the McLachlin Court’ in David A Wright and Adam M Dodek (eds), Public Law at the McLachlin Court: The First Decade (Irwin Law, 2011) 145, 145.

[177] Tom Hickman, ‘Adjudicating Constitutional Rights in Administrative Law’ (2016) 66(1) University of Toronto Law Journal 121, 121–3.

[178] Ibid 134–5.

[179] See, eg, ibid 159–60; Claudia Geiringer, ‘Sources of Resistance to Proportionality Review of Administrative Power under the New Zealand Bill of Rights Act’ (2013) 11(1) New Zealand Journal of Public and International Law 123, 151–3.

[180] See generally Boughey, ‘Reasonableness of Proportionality’ (n 115) 61–74.

[181] Human Rights Act 1998 (UK) s 6(2)(b).

[182] Canada Act 1982 (UK) c 11, sch B pt I (‘Canadian Charter’).

[183] Note that, unlike the Canadian Charter (n 182), the New Zealand Bill of Rights Act 1990 (NZ) (‘NZBORA’) is a statute, and so it does not affect the validity of legislation which is incompatible with human rights. However, despite this difference, the New Zealand courts must grapple with the same issue as the courts in Australia with respect to the way rights affect the interpretation of statutes and the exercise of statutory discretions, and so a comparison remains fruitful.

[184] Canadian Charter (n 182) s 32(1); NZBORA (n 183) s 3.

[185] NZBORA (n 183) s 6; Re Application under Section 83.28 of the Criminal Code [2004] 2 SCR 248, 269–70 [35] (Iacobucci and Arbour JJ for McLachlin CJ, Iacobucci, Major and Arbour JJ); Charkaoui v Minister of Citizenship and Immigration (Canada) [2007] 1 SCR 350, 414 [123] (McLachlin CJ for the Court); Slaight Communications Inc v Davidson [1989] 1 SCR 1038, 1078 (Lamer J) (‘Slaight’).

[186] Canadian Charter (n 182) s 1; NZBORA (n 183) s 5.

[187] Unless a Canadian Parliament makes use of the ‘notwithstanding’ clause in s 33(1) of the Canadian Charter (n 182), or a New Zealand statute requires discretion to be so exercised, in which case the statute itself will be incompatible with rights.

[188] See generally Janet McLean, ‘The Impact of the Bill of Rights on Administrative Law Revisited: Rights, Utility, and Administration’ [2008] (2) New Zealand Law Review 377; Geiringer (n 179) 132–3; Hanna Wilberg, ‘The Bill of Rights in Administrative Law Cases: Taking Stock and Suggesting Some Reassessment’ (2013) 25(4) New Zealand Universities Law Review 866; Paul Daly, ‘The Court and Administrative Law: Models of Rights Protection’ (2017) 78 Supreme Court Law Review (2d) 57; Janina Boughey, Human Rights and Judicial Review in Australia and Canada: The Newest Despotism? (Hart Publishing, 2017) 264–70 (‘Human Rights and Judicial Review’).

[189] So far this test has only been applied by a majority of the Court in relation to the implied freedom of political communication and s 92: see above n 2 and accompanying text.

[190] Canadian Charter (n 182) s 1; NZBORA (n 183) s 5.

[191] See generally Robert Leckey, ‘Prescribed by Law/Une Règle de Droit’ (2007) 45(3) Osgoode Hall Law Journal 571.

[192] Slaight (n 185).

[193] Ibid 1078–9 (Lamer J).

[194] Ibid, quoting Peter Hogg, Constitutional Law of Canada (Carswell, 2nd ed, 1985) 671.

[195] [2006] 1 SCR 256, 310–14 [112]–[121] (Deschamps and Abella JJ) (‘Multani’).

[196] Ibid 308 [107].

[197] Ibid 315 [125].

[198] Ibid 314 [121]. Various aspects of administrative law in both Canada and Australia are premised on the idea that the same accountability mechanisms will not be appropriate for both administrative and legislative acts. For instance, Australia’s federal AD(JR) Act (n 44) applies only to decisions ‘of an administrative character’ as opposed to a legislative character: at s 3(1) (definition of ‘decision to which this Act applies’). Canadian law also distinguishes between administrative and legislative acts, for instance in the application of the rules of procedural fairness: see, eg, A-G (Canada) v Inuit Tapirisat of Canada [1980] 2 SCR 735, 758–9 (Estey J for the Court). For a critique of that decision and some of its underlying principles, see Geneviève Cartier, ‘Procedural Fairness in Legislative Functions: The End of Judicial Abstinence?’ (2003) 53(3) University of Toronto Law Journal 217.

[199] Multani (n 195) 305 [99] (Deschamps and Abella JJ).

[200] See, eg, Moonen v Film and Literature Board of Review [2002] NZCA 69; [2002] 2 NZLR 754, 762 [20] (Richardson P for the Court); Commerce Commission v Air New Zealand Ltd [2011] NZCA 64; [2011] 2 NZLR 194, 205–6 [45]–[50] (Glazebrook J for the Court); Ministry of Health (NZ) v Atkinson [2012] NZCA 184; [2012] 3 NZLR 456, 496 [182]–[184] (Ellen France J for the Court). See generally McLean (n 188).

[201] [2018] NZSC 59; [2018] 1 NZLR 948 (‘New Health’).

[202] Ibid 960 [9] (O’Regan J for O’Regan and Ellen France JJ).

[203] Ibid 966 [40] (O’Regan J for O’Regan and Ellen France JJ).

[204] This is a similar approach to that taken by Kiefel CJ and Keane J in Palmer (n 1), which effectively engages in structured proportionality review of the specific decision but treats the decision as one of the legislature by explaining how it is impliedly authorised by the legislation: at 196–7 [63]–[68], 198–9 [76]–[80].

[205] Chief Justice Elias adopted a similar approach, but dissented in the result: New Health (n 201) 1027–8 [284], 1033 [306]. Justice Glazebrook reached the same conclusion, and adopted O’Regan and Ellen France JJ’s proportionality analysis, but did so by applying the analysis directly to the Council’s decision: at 993–4 [165]–[171]. The remaining judge, William Young J, held that the NZBORA (n 183) was not engaged by the decision to fluoridate water: New Health (n 201) 1007 [210].

[206] Janet McLean and Claudia Geiringer have (separately) analysed this approach and have argued, as we have above, that while this heavily interpretive approach is suitable in some cases, it will not suit the full range of discretionary powers: McLean (n 188) 396–7; Geiringer (n 179) 155–7.

[207] [2001] NZCA 207; [2002] 1 NZLR 58, 60 [1] (Blanchard J for Richardson P, Keith, Blanchard and Tipping JJ) (‘Drew’).

[208] Ibid 64 [28], 66 [35] (Blanchard J for Richardson P, Keith, Blanchard and Tipping JJ).

[209] Ibid 72–3 [67] (Blanchard J for Richardson P, Keith, Blanchard and Tipping JJ). The majority noted that inconsistency with the NZBORA (n 183) would have resulted in the same conclusion: Drew (n 207) 73 [68].

[210] [1999] 2 SCR 817 (‘Baker’).

[211] Ibid 832 [11] (L’Heureux-Dubé J for L’Heureux-Dubé, Gonthier, McLachlin, Bastarache and Binnie JJ). See also Canadian Charter (n 182) s 7.

[212] Baker (n 210) 832 [11] (L’Heureux-Dubé J for L’Heureux-Dubé, Gonthier, McLachlin, Bastarache and Binnie JJ).

[213] Geneviève Cartier, ‘The Baker Effect: A New Interface between the Canadian Charter of Rights and Freedoms and Administrative Law’ in David Dyzenhaus (ed), The Unity of Public Law (Hart Publishing, 2004) 61, 62–3, 85 (‘The Baker Effect’). See also Blencoe v Human Rights Commission (British Columbia) [2000] 2 SCR 307, 406–7 [189] (LeBel J for Iacobucci, Binnie, Arbour and LeBel JJ); JM Evans, ‘The Principles of Fundamental Justice: The Constitution and the Common Law’ (1991) 29(1) Osgoode Hall Law Journal 51, 73.

[214] Paul Rishworth, ‘Interpreting Enactments: Sections 4, 5 and 6’ in Paul Rishworth et al (eds), The New Zealand Bill of Rights (Oxford University Press, 2003) 116, 140.

[215] Slaight (n 185) 1078, 1081 (Lamer J).

[216] Ibid 1078.

[217] See Geiringer (n 179) 128–9.

[218] Slaight (n 185) 1081 (Lamer J).

[219] [2020] NZHC 1608 (‘Hudson’).

[220] Ibid [2] (Dobson J).

[221] Ibid [7]–[12].

[222] Ibid [7]. Note he did not argue the rules themselves were unlawful, or that the decision was unreasonable in an administrative law sense: at [39]–[41], [67].

[223] [2007] 3 NZLR 1, 37 [92] (Tipping J).

[224] Hudson (n 219) [77]–[82] (Dobson J).

[225] See Evan Fox-Decent, ‘The Charter and Administrative Law: Cross-Fertilization in Public Law’ in Colleen M Flood and Lorne Sossin (eds), Administrative Law in Context (Emond Montgomery Publications, 2008) 169, 181–2.

[226] [1986] 1 SCR 103.

[227] See, eg, Slaight (n 185) 1049 (Dickson CJ for Dickson CJ, Wilson, La Forest and L’Heureux-Dubé JJ); Ross v New Brunswick School District [No 15] [1996] 1 SCR 825, 850–1 [31]–[32] (La Forest J for the Court); Multani (n 195) 270–1 [16] (Charron J for McLachlin CJ, Bastarache, Binnie, Fish and Charron JJ); Loyola High School v A-G (Québec) [2015] 1 SCR 613, 626–7 [3] (Abella J for LeBel, Abella, Cromwell and Karakatsanis JJ) (‘Loyola’). See generally Boughey, Human Rights and Judicial Review (n 188) 264–70.

[228] A religious object that resembles a dagger.

[229] Multani (n 195) 284 [49], 285 [51] (Charron J for McLachlin CJ, Bastarache, Binnie, Fish and Charron JJ).

[230] Ibid 289–94 [60]–[69], 296 [77], 297 [79].

[231] Ibid 314 [121].

[232] Ibid 310 [112].

[233] Ibid 314 [123].

[234] Ibid 315 [125]. See above n 198.

[235] See, eg, Evans (n 213) 73. For analysis in the Canadian context, see Cartier, ‘The Baker Effect’ (n 213) 70. In the New Zealand context, see Geiringer (n 179).

[236] See Boughey, ‘The Reasonableness of Proportionality’ (n 115) 61–74. On the High Court’s approach to judicial reasoning more broadly, see Adrienne Stone, ‘Judicial Reasoning’ in Cheryl Saunders and Adrienne Stone (eds), The Oxford Handbook of the Australian Constitution (Oxford University Press, 2018) 472.

[237] See Janina Boughey and Adam Fletcher, ‘Administrative Decision-Making under Victoria’s Charter’ (2018) 25(1) Australian Journal of Administrative Law 10, 13–20.

[238] [2015] NZCA 477; [2015] NZAR 1648, 1669–70 [80]–[81] (Randerson J for the Court) (‘Taylor’), quoting Geiringer (n 179) 159.

[239] Taylor (n 238) 1670 [83] (Randerson J for the Court).

[240] Ibid 1670–1 [82]–[84].

[241] [2012] 1 SCR 395, 416–21 [33]–[45] (Abella J for the Court) (‘Doré’). For a more detailed discussion, see Janina Boughey, ‘Rights, Review and Reasonableness: The Implications of Canada’s New Approach to Administrative Decision-Making and Human Rights for Australia’ [2013] SydLawRw 11; (2013) 35(2) Sydney Law Review 283.

[242] Slaight (n 185) 1076 (Lamer J).

[243] Cartier, ‘The Baker Effect’ (n 213) 68.

[244] Doré (n 241) 404 [6], 418 [37] (Abella J for the Court).

[245] Ibid 426 [56], quoting Dunsmuir v New Brunswick [2008] 1 SCR 190, 221 [47] (Bastarache and LeBel JJ for McLachlin CJ, Bastarache, LeBel, Fish and Abella JJ) (‘Dunsmuir’).

[246] Doré (n 241) 427 [57] (Abella J for the Court).

[247] See Janina Boughey, ‘A(nother) New Unreasonableness Framework for Canadian Administrative Law’ (2020) 27(1) Australian Journal of Administrative Law 43, 51–2.

[248] Dunsmuir (n 245) 220–1 [47] (Bastarache and LeBel JJ for McLachlin CJ, Bastarache, LeBel, Fish and Abella JJ), quoted in Minister of Citizenship and Immigration (Canada) v Vavilov (2019) 441 DLR (4th) 1, 67 [86] (Wagner CJ, Moldaver, Gascon, Côté, Brown, Rowe and Martin JJ).

[249] Dunsmuir (n 245) 214 [34] (Bastarache and LeBel JJ for McLachlin CJ, Bastarache, LeBel, Fish and Abella JJ).

[250] Audrey Macklin, ‘Charter Right or Charter-Lite?: Administrative Discretion and the Charter’ (2014) 67 Supreme Court Law Review 561, 569–73. See generally Hickman (n 177); Christopher D Bredt and Ewa Krajewska, ‘Doré: All That Glitters Is Not Gold’ (2014) 67 Supreme Court Law Review 339.

[251] R (SB) v Governors of Denbigh High School [2006] UKHL 15; [2007] 1 AC 100, 116 [30] (Lord Bingham), 125–6 [68] (Lord Hoffmann).

[252] For a good example, see Taylor (n 238) 1671 [84]–[86] (Randerson J for the Court).

[253] See, eg, Law Society of British Columbia v Trinity Western University [2018] 2 SCR 293, 354–5 [111]–[113] (McLachlin CJ), 370–5 [162]–[175] (Rowe J), 411–12 [266]–[267] (Côté and Brown JJ).

[254] See, eg, Loyola (n 227) 641–56 [43]–[80] (Abella J for LeBel, Abella, Cromwell and Karakatsanis JJ), 680–8 [146]–[162] (McLachlin CJ and Moldaver J for McLachlin CJ, Rothstein and Moldaver JJ).

[255] (Court of Appeal of New Zealand, McGrath, Hammond and William Young JJ, 17 September 2004) [26] (McGrath J for McGrath, Hammond and William Young JJ) (‘Television New Zealand’).

[256] Ibid [6], [11].

[257] Ibid [25].

[258] Taylor (n 238) 1668–71 [76]–[86] (Randerson J for the Court).

[259] Ibid 1670 [83].

[260] Ibid, citing Television New Zealand (n 255) [16] (McGrath J for McGrath, Hammond and William Young JJ).

[261] Taylor (n 238) 1671 [84] (Randerson J for the Court).

[262] [2005] NZAR 1, 14 [52].

[263] [2017] NZHC 463; [2017] 2 NZLR 704, 721 [73] (Wylie J), citing Television New Zealand Ltd v West [2011] NZHC 435; [2011] 3 NZLR 825, 843 [86] (Asher J).

[264] See, eg, Michael Taggart, ‘Reinventing Administrative Law’ in Nicholas Bamforth and Peter Leyland (eds), Public Law in a Multi-Layered Constitution (Hart Publishing, 2003) 311, 325–6; Dame Sian Elias, ‘Righting Administrative Law’ in David Dyzenhaus, Murray Hunt and Grant Huscroft (eds), A Simple Common Lawyer: Essays in Honour of Michael Taggart (Hart Publishing, 2009) 55, 66–8; Rishworth (n 214) 160.

[265] Wilberg (n 188) 882.

[266] Ibid 881–3.

[267] Ibid.

[268] See above discussions of Wotton (n 1) 16 [31] (French CJ, Gummow, Hayne, Crennan and Bell JJ); Banerji (n 1) 406 [45] (Kiefel CJ, Bell, Keane and Nettle JJ); Gaynor (n 39) 315–16 [73] (Perram, Mortimer and Gleeson JJ).

[269] Banerji (n 1) 408 [52] (Gageler J), 459 [211] (Edelman J).

[270] Bret Walker and David Hume, ‘Broadly Framed Powers and the Constitution’ in Neil Williams (ed), Key Issues in Public Law (Federation Press, 2017) 144, 157.

[271] Li (n 71) 367 [76] (Hayne, Kiefel and Bell JJ), 371 [91] (Gageler J); SZVFW (n 72) 550 [10] (Kiefel CJ), 573 [82] (Nettle and Gordon JJ).


AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/MelbULawRw/2022/13.html