Fair work laws and freedom of expression

Australian common law, in this case non statutory constitutional law principles developed by the High Court, implies as a matter of law that legislation is to be subject to an implied right for persons (including presumable corporations) to communicate freely about matters of politics and government.

This fundamental principle of Australian constitutional law means that legislation which conflicts with the principle will be invalid. This is particularly relevant to employment law issues on show in this extract from a Federal Court decision involving an industrial dispute.

“As was earlier explained, Lendlease and the CFMMEU took objection to significant swathes of the evidence, written and oral, that the respondents sought to lead.  A very small number of those objections were determined over the course of the trial but most were not.

  1. The objections that were not resolved at the trial involved statements in the nature of hearsay or opinion (or were otherwise objectionable as speculative, vague or conclusory).  In all such cases, there was no doubt that the impugned statements were properly described as such (that is to say, it was conceded that they involved statements in the nature of hearsay or opinion, or were otherwise susceptible to the complaints that were made of them).  At the risk of oversimplification, the evidence in question went primarily to what was said to be the influence that construction unions have (or have had) over the workforces of large construction projects throughout Australia, the paraphernalia that they deploy within project sites to further their interests, the efforts to which they go in order to persuade or coerce construction workers to join their ranks, the “no ticket, no start” philosophies that guide those efforts, and the complicity of major construction businesses in (or their indifference toward) those efforts.  What was advanced might not unfittingly be described as a modern take on the “Brandeis brief”:  in addition to affidavit testimony containing obvious hearsay and opinion statements, it comprised (amongst other things) of Royal Commission reports, Hansard records, and decisions of this court (and others), all of which was said to inform one or more of those factual inquiries.
  1. Insofar as Lendlease objected to those parts of the evidence, those objections effectively resolved on the basis of the ruling made under s 136 of the Evidence Act (above, [45]):  whether admissible or not, none of the impugned evidence would serve to aid in the proof of matters relevant to the case advanced by or against Lendlease (specifically, as to the proper construction of s 13(2)(j) of the Building Code).  Instead, its use would be limited to the proof of matters relevant to the CFMMEU’s constitutional challenge to the validity of s 13(2)(j) of the Building Code.
  1. The reasoning that follows applies to the evidential passages to which objection was taken, as well as to other statements of opinion and hearsay that arose during the course of oral evidence (each individual example of which needn’t be—indeed, cannot sensibly be—summarised here).  For the reasons that follow, the objections to those parts of the evidence (to which I shall refer hereafter as, the “Impugned Passages”) are dismissed.

9.1.2  Principles to be applied

  1. The admissibility of the Impugned Passages turns solely upon whether or not this court is bound by the orthodox rules of admission for which the Evidence Act provides.  If it is, then the Impugned Passages are (or, in a few cases, might be) vulnerable to rejection.  If it is not, then regard might be had to them.  The respondents maintain that the rules governing the receipt of evidence relevant to matters of constitutional fact are not the same—indeed, are less onerous—than those that govern the receipt of evidence relevant to other (adjudicative) facts.  The CFMMEU disagrees.
  1. To the extent that the Impugned Passages constitute expressions of opinion that would ordinarily be excluded from admission into evidence by s 76 of the Evidence Act (assuming that it has that effect in respect of evidence relevant to constitutional facts), the respondents maintain in any event that the evidence is admissible expert opinion. That submission is maintained notwithstanding that the evidence in question was sought to be adduced otherwise than in compliance with pt 23 of the Federal Court Rules 2011 (Cth).
  1. The Evidence Act draws no distinction between evidence that serves to establish a fact relevant to a question of constitutional application and evidence that serves to establish some other species of fact.  Instead, the notorious rules of evidence concerning statements of hearsay and opinion apply apparently without discrimination in “…all proceedings in a federal court”:  Evidence Act, s 4(1). The Act does not define what constitutes a “fact”, as that term is used in provisions such as ss 59 and 76. There is, then, no statutory indicator to suggest that the receipt of evidence concerning facts related to constitutional issues should be subject to rules any different from those that govern the receipt of evidence concerning other categories of fact.
  1. Nonetheless, there is authority that supports the respondents’ contention.  Before turning to it, I should identify what is ordinarily contemplated by the term “constitutional fact”.  In simple terms, a constitutional fact is “…information which the court should have in order to judge properly of the validity of this or that statute or of this or that application by the Executive Government of State or Commonwealth of some power or authority it asserts”:  Breen v Sneddon (1961) 106 CLR 406, 411 (“Breen v Sneddon”; Dixon CJ).
  1. In the present context—involving, as it does, an assessment of whether s 13(2)(j) of the Building Code (as I have construed it) serves as an impermissible burden upon the constitutionally-implied right to communicate freely about matters of politics and government—the court must make some factual assessments about the mischief that the provision is designed to address and whether the means by which it does so burdens rights of free expression in a manner that is disproportionate to or unjustified by that objective.  Doing so involves, at the least, considering:

(1)          the extent to which the provision serves to remove from building associations modes of communication about matters of politics or government;

(2)          the significance of any such removal—that is to say, whether or to what extent those modes of communicating are important, effective or convenient, and whether other alternatives might be deployed to achieve equivalent ends;

(3)          whether or to what extent the burden imposed may be justified by, or serves to advance, a legitimate purpose, such as the promotion of construction workers’ rights to join or not join building associations; and

(4)          whether or to what extent there exist cultural, common or industry-wide practices or attitudes, or other forces, that imperil that purpose in such a way as might authorise or justify the burden that is imposed.

  1. In Commonwealth Freighters Pty Ltd v Sneddon (1959) 102 CLR 280, Dixon CJ made the following relevant observations (at 291-292):

In courts administering English law according to the principles which developed in a unitary system it must seem anomalous that the question whether a given statute operates or not should depend upon facts proved in evidence.  How facts are to be ascertained is of course a question distinct from their relevance.  Highly inconvenient as it may be, it is true of some legislative powers limited by definition, whether according to subject matter, to purpose or otherwise, that the validity of the exercise of the power must sometimes depend on facts, facts which somehow must be ascertained by the court responsible for deciding the validity of the law…  [I]f a criterion of constitutional validity consists in matter of fact, the fact must be ascertained by the court as best it can, when the court is called upon to pronounce upon validity.

  1. Those comments followed observations that his Honour made 15 years earlier in Stenhouse v Coleman (1944) 69 CLR 457 (at 469):

… there are limitations upon the material which a court can receive or take into account for the purpose of considering the validity of a general law.  If the form of the power makes the existence of some special or particular state of fact a condition of its exercise, then, no doubt, the existence of that state of fact may be proved or disproved by evidence like any other matter of fact.  But ordinarily the court does not go beyond matters of which it may take judicial notice.  This means that for its facts the court must depend upon matters of general public knowledge.

  1. Later, in Breen v Sneddon, his Honour elaborated (at 411) upon the distinction between constitutional and other facts:

It is the distinction between, on the one hand, ordinary questions of fact which arise between the parties because one asserts and the other denies that events have occurred bringing one of them within some criterion of liability or excuse set up by the law and, on the other hand, matters of fact upon which under our peculiar federal system the constitutional validity of some general law may depend.  Matters of the latter description cannot and do not form issues between the parties to be tried like the former questions.

  1. Nearly a quarter of a century later in Gerhardy v Brown (1985) 159 CLR 70, Brennan J endorsed those remarks (at 142). In that case, the High Court had occasion to assess the validity of the Pitjantjatjara Land Rights Act 1981 (SA).  That legislation conferred various land rights and, relevantly, established certain, related offences.  The respondent, who stood accused of having committed one such offence, argued that the enactment was (or relevant parts of it were) inconsistent with the Racial Discrimination Act 1975 (Cth) and, by operation of s 109 of the Constitution, was invalid. The Commonwealth legislation (or the relevant parts of it) did not apply to, or in relation to the application of, recognised species of “special measures”. The validity of the state act turned, then (in part), upon whether it qualified as such a measure.
  1. That, in turn, depended upon whether or not the state act could reasonably have been thought necessary to ensure that members of the indigenous group for whose benefit it was enacted would be afforded “equal enjoyment or exercise of human rights and fundamental freedoms”.  As to that, Brennan J noted (at 141):

This question requires some understanding of the circumstances in which the Act is intended to operate.  Matters of fact are involved, and the Court must ascertain some facts in order to determine what is a question of law:  the validity of the [South Australian enactment] and the scope of Pt II of the Racial Discrimination Act.

  1. The primary judge (Millhouse J) had determined that the South Australian enactment was not a “special measure” and, thus, was inconsistent with the Commonwealth statute and invalid.  No factual findings were made, a reality that prompted Brennan J to make the following observations (at 141-143):

There is a distinction between a judicial finding of fact in issue between the parties upon which a law operates to establish or deny a right or liability and a judicial determination of the validity or scope of a law when its validity or scope turns on a matter of fact.  When a court, in ascertaining the validity or scope of a law, considers matters of fact, it is not bound to reach its decision in the same way as it does when it tries an issue of fact between the parties.  The validity and scope of a law cannot be made to depend on the course of private litigation.  The legislative will is not surrendered into the hands of the litigants.

[I]n Commonwealth Freighters Pty Ltd v Sneddon, [Dixon CJ] observed that “if a criterion of constitutional validity consists in matter of fact, the fact must be ascertained by the court as best it can, when the court is called upon to pronounce upon validity”.  The court may, of course, invite and receive assistance from the parties to ascertain the statutory facts, but it is free also to inform itself from other sources.  Perhaps those sources should be public or authoritative, and perhaps the parties should be at liberty to supplement or controvert any factual material on which the court may propose to rely, but these matters of procedure can await consideration on another day.  The court must ascertain the statutory facts “as best as it can” and it is difficult and undesirable to impose an a priori restraint on the performance of that duty.

  1. His Honour then proceeded to note that, although neither party had adduced evidence capable of informing the court’s assessment of whether or not the South Australian enactment qualified as a “special measure”, there was nonetheless other material produced to the court from which it could inform itself to that end.  Further, his Honour noted that “…the courts of this country are familiar with the existence of traditional Aboriginal affiliations with, and responsibilities in respect of, land.”  Armed with that body of material, his Honour proceeded to reason that the South Australian legislation qualified as a “special measure” and, hence, was not invalid.
  1. In Cunliffe v Commonwealth (1994) 182 CLR 272, Mason CJ (in dissent) seemed to contemplate (at 304) that, in cases such as the present, the court might take account of facts established “otherwise” than via traditional means:

In the context of an implication of freedom of communication, in order to justify the imposition of some burden or restriction on that right, it is generally not enough simply to assert the existence of facts said to justify the imposition of that burden or restriction.  The relevant facts must either be agreed or proved or be such that the Court is prepared to take account of them by judicial notice or otherwise.

  1. In Thomas v Mowbray (2007) 233 CLR 307, Heydon J identified a schism in the jurisprudence concerning the proof of constitutional facts. His Honour (at 514 [620]) referred to Williams’s J observations in Australian Communist Party v The Commonwealth (1951) 83 CLR 1, 225 that, “…facts which are relevant to the decision of [a] constitutional issue must be admissible in evidence…[and]…facts which are not capable of proof [by judicial notice] must be proved in such other ways as the laws of evidence allow”. His Honour went on to note that the “…insistence that the ordinary rules of evidence apply to the proof of constitutional facts has other support”, to which he then referred.
  1. Thereafter (at 515 [621] and following), his Honour drew attention to “more modern authorities [that] deny that constitutional facts can only be proved by material admissible under the rules of evidence”.  It is unnecessary to rehearse the policy justifications that his Honour identified in support of that proposition.  Instead, it suffices to note his Honour’s conclusions (at 517 and following):

[629]        [T]he Court can receive evidence of constitutional facts which complies with the rules of admissibility applying to [non-constitutional facts].  But it is not limited to that material…

[633]        [T]he Court can take into account its knowledge of society (apparently whether the information it knows is noticed in compliance with the rules of judicial notice or not), and while it can take into account evidence and agreed facts, all other relevant material may be considered whether or not it is technically admissible…

[636]        [T]he rules of evidence do not restrict the material which the Court can consider in deciding on [constitutional] facts…

  1. In Maloney v The Queen (2013) 252 CLR 168, Gageler J continued upon the themes to which Heydon J adverted in Mowbray. Specifically, his Honour observed (at 298-299 [351]-[353], references omitted):

A distinction has long been drawn between “ordinary questions of fact”, which arise between parties and which are determined in accordance with the ordinary rules of evidence, and “matters of fact upon which…the constitutional validity of some general law may depend”, which “cannot and do not form issues between the parties to be tried like the former questions” and which fall to be ascertained by a court “as best it can”.  A court finding constitutional facts is not constrained by the rules of evidence.  The court “reaches the necessary conclusions of fact largely on the basis of its knowledge of the society of which it is a part”, “supplementing … that knowledge [by processes] which [do] not readily lend [themselves] to the normal procedures for the reception of evidence”.

The nature of legislative facts and the nature of the duty of a court to ascertain them tell against any a priori constraint on the sources from which the court may inform itself.  The sources may, but need not, be “official”.  It is desirable, but not inevitable, that they be “public or authoritative”.  They can include “inferences … drawn from the regulations and statutes themselves” and “statements made at the bar”.  Subject to the requirements of procedural fairness inherent in the judicial process, the ultimate criterion governing the use of information from any source is that a court is able to consider the material sufficiently probative of the legislative fact to be found.

  1. The observations made throughout the authorities cited above are consistent and unambiguous.  They reconcile (perhaps not perfectly) with the following observations made in the Australian edition of Cross on Evidence (John Dyson Heydon, Cross on Evidence, 11th ed, 2017, [3005]):

In the case of adjudicative facts the doctrine of judicial notice has restricted scope, for in the common law system the facts are appropriately determined on the evidence presented by the parties unless the fact is of such notoriety that to call for evidence would be a waste of time.  The position with respect to legislative facts is otherwise.  It is clear from the cases that judges have felt themselves relatively free to apply their own views and to make their own enquiries of social ethics, psychology, politics and history where relevant without requiring evidence or other proof.

  1. Surprisingly, it is not easy to find examples in this court (particularly at first instance) in which those principles have been applied.  Perhaps that reflects that, in matters that “…depend upon some factual matter touching the freedom of discussion of government or politics, questions of fact seldom have to be resolved”:  Levy v Victoria (1997) 189 CLR 579, 598 (Brennan CJ).
  1. All the same, discussion about the applicability of the principles outlined above has not been limited to the High Court.  In Sportodds Systems Pty Ltd v New South Wales (2003) 133 FCR 63 (Branson, Hely and Selway JJ), this court, referring to the comments of Brennan J in Gerhardy v Brown extracted above (at [169]), observed (at 81-82 [47]-[49]):

There are some factual issues that arise in litigation where the court is necessarily constrained by the evidence led by the parties.  In such cases the inadequacy or otherwise of that evidence does not relieve the court from making what factual conclusions it can based upon what evidence is called and the various burdens of proof that might apply.  However, where the issue (as here) concerns the validity of legislation, different considerations apply…

There are cases where a court has no choice but to deal with a question of constitutional validity that arises in a case.  A prosecution for breach of a statutory provision alleged to be invalid might be an example.  In such a case a court may have no choice but to inform itself ‘as best it can’.  For this purpose, of course, the court could require the parties to provide further factual material.  In particular, the court would often rely upon the relevant Attorney General to provide some assistance to it as to relevant factual matters.  But as Brennan J pointed out the court can also inform itself from its own inquiries.

There are obvious dangers in it doing so.  Those dangers are referred to by Callinan J in Woods v Multi-Sport Holdings Pty Ltd (2002) 208 CLR 460 at 510-513. To the extent that his Honour concludes that a court cannot inquire for itself into constitutional facts, such authority as there is would seem to be to the contrary. But the dangers are clear enough. In light of those dangers a court would not lightly undertake the task of carrying out detailed factual research, at least unless the material relied upon was ‘public or authoritative’ or unless the court had no other choice…

  1. The capacity of a court to consider constitutional fact material that might otherwise be inadmissible arose (at least to the point of mention) in R v Alqudsi (2015) 328 ALR 517 (Adamson J). There, the Supreme Court of New South Wales was called upon to determine whether the Crimes (Foreign Incursions and Recruitment) Act 1978 (Cth) was a valid exercise of the Commonwealth Parliament’s power to legislate with respect to external affairs. Hoping to persuade the court that it was, the Commonwealth tendered various documents that were said to inform that question, as to which the following observations were made (at 524 [33]):

The nature of constitutional fact evidence was addressed by Gageler J in Maloney v R (2013) 252 CLR 168; 298 ALR 308; [2013] HCA 28 at [353] (Maloney) and Heydon J in Thomas at [614]–[618] and [620]–[639]. When considering such material, the court is not constrained by the rules of evidence; the evidence need not be “official”. As Gageler J said at [353] of Maloney:

“Subject to the requirements of procedural fairness inherent in the judicial process, the ultimate criterion governing the use of information from any source is that a court is able to consider the material sufficiently probative of the legislative fact to be found.”

  1. Although the court was willing to receive the evidence in question, the matter was ultimately decided on the basis that no recourse to it was necessary:  R v Alqudsi (2015) 328 ALR 517, 547 [159] (Adamson J). Her Honour’s acceptance of the point of principle is, nonetheless, apparent.
  1. Similar observations were made in Lee v Minister for Home Affairs [2021] FCAFC 89, [51] (Logan, Kerr and Banks-Smith JJ). First-instance examples of where equivalent principles were accepted (although not necessarily applied) include: Civil Aviation Safety Authority v Boatman (2004) 138 FCR 384, 405 [61] (Selway J, with whom Sundberg and Stone JJ, determining a case stated for the consideration of the full court, agreed in the result); Owners Corporation PS 501391P v Balcombe (2016) 51 VR 299, 338 [121] (Riordan J); Amoonguna Community Inc v Northern Territory of Australia [2014] NTSC 33, [25] (Barr J); Alderton v Department of Police and Emergency Management [2008] TASSC 69, [65] (Slicer J).

9.1.3  Application

  1. However surprising (as it is, at least to me), however unreflected in the terms of the Evidence Act and however limited the jurisprudence that has emerged from this court, I consider that the authorities leave little room for doubt:  factual matters relevant to the constitutional validity of an impugned law or executive action may be proved by means of material that might ordinarily be inadmissible as evidence.  Although contesting the point of principle, the CFMMEU was unable to identify any recent authority that runs counter to those from which that principle seems unambiguously to emerge.
  1. Those realities acknowledged, I am not minded to exclude as inadmissible any of the material that the respondents advanced (and to the admissibility of which the CFMMEU objected).
  1. That, though, is not to say that I regard it all as worthy of unflinching acceptance.  Whether orthodox rules of admissibility apply or not, the authorities are clear that, in determining matters of constitutional (or legislative) fact, the material from which the court informs its answers must nonetheless bear at least a measure of authority and persuasiveness:  Thomas v Mowbray (2007) 233 CLR 307, 522 [639] (Heydon J); Re Day (2017) 91 ALJR 262, 269 [23] (Gordon J).
  1. Inevitably, some aspects of the material before the court presently are more authoritative and persuasive—and, therefore, worthy of greater attention or weight—than others.  For example, for present purposes, the findings contained within the published reports of Royal Commissions and the decisions of this court (and others) will, generally speaking, command greater deference than will the opinions of witnesses, even witnesses of considerable experience and expertise (and even when those opinions are tested through skilful cross-examination, as they were in this matter).
  1. In the analysis that follows, the findings of relevant fact to which I have been drawn have been reached taking account of those cautions.

9.2  Nature of the implied freedom

  1. The Commonwealth Constitution affords, by implication, a broad (though not impenetrable) protection against legislative and executive measures that constrain the exchange of communications concerning matters of politics or public government: Lange v Australian Broadcasting Corporation (1997) 189 CLR 520. In doing so, it guarantees (at least to a degree) that people within Australia are free to receive and disseminate information that enables them to exercise a free and informed choice as electors. That freedom, though qualified in the ways explored below, endures beyond periods of election campaigning. Legislative or executive measures that serve to constrain the rights to which that implied freedom is directed are beyond what is constitutionally authorised and, as such, are liable to be struck down judicially.
  1. Presently, the CFMMEU contends that s 13(2)(j) of the Building Code, insofar as it applies (as I have found) to code-covered entities so as to prohibit the application of building association logos, mottos or indicia to project-supplied clothing, property or equipment, offends against the implied freedom described above and is, accordingly, beyond what is constitutionally authorised.  If that be so, there is no doubt that the Compliance Notice is equally vulnerable to correction.
  1. At the core of the CFMMEU’s contention is the uncontroversial proposition that building associations commonly use the buildings and equipment that are employed at large construction projects throughout Australia to disseminate messages that contain what can broadly (if a little vaguely) be described as political in nature.  Overwhelmingly and for reasons that can readily be imagined, those messages are branded (which is to say that they contain building association logos, mottos or indicia).  Thus it is said that, on the construction that I consider correct, s 13(2)(j) of the Building Code amounts to a constitutionally-prohibited constraint upon the dissemination of such messages.
  1. The respondents accept that s 13(2)(j) infringes upon what would otherwise be the right of the CFMMEU (and other building associations) to post, under licence (implied or otherwise), branded messages upon relevant clothing, property or equipment.  Nonetheless, they maintain that that infringement is not one that falls foul of the protections afforded by the implied constitutional guarantee.

9.3  Limits of the constitutional protection

  1. The right to communicate freely about political or government matters for which the Commonwealth Constitution implicitly provides is one that “…protects the free expression of political opinion…[and] operates as a limit on the exercise of legislative power to impede that freedom of expression”: Brown v Tasmania (2017) 261 CLR 328 (“Brown”), 359 (Kiefel CJ, Bell and Keane JJ).
  1. Whether s 13(2)(j) of the Building Code is invalid as an impermissible infringement upon the constitutionally-implied freedom to communicate about governmental and political matters turns upon consideration of three matters, conveniently identified in the opening passages of the majority’s judgment in McCloy v New South Wales (2015) 257 CLR 178, 193-194 [2] (French CJ, Kiefel, Bell and Keane JJ—references omitted):

…[T]he question whether an impugned law infringes the freedom requires application of the following propositions derived from previous decisions of this Court and particularly Lange v Australian Broadcasting Corporation and Coleman v Power:

  1. The freedom under the Australian Constitution is a qualified limitation on the legislative power implied in order to ensure that the people of the Commonwealth may “exercise a free and informed choice as electors”. It is not an absolute freedom. It may be subject to legislative restrictions serving a legitimate purpose compatible with the system of representative government for which the Constitution provides, where the extent of the burden can be justified as suitable, necessary and adequate, having regard to the purpose of those restrictions.
  2. The question whether a law exceeds the implied limitation depends upon the answers to the following questions, reflecting those propounded in Lange as modified in Coleman v Power:
  3. Does the law effectively burden the freedom in its terms, operation or effect?

If “no”, then the law does not exceed the implied limitation and the inquiry as to validity ends.

  1. If “yes” to question 1, are the purpose of the law and the means adopted to achieve that purpose legitimate, in the sense that they are compatible with the maintenance of the constitutionally prescribed system of representative government? …

The answer to that question will be in the affirmative if the purpose of the law and the means adopted are identified and are compatible with the constitutionally prescribed system in the sense that they do not adversely impinge upon the functioning of the system of representative government.

If the answer to question 2 is “no”, then the law exceeds the implied limitation and the inquiry as to validity ends.

  1. If “yes” to question 2, is the law reasonably appropriate and adapted to advance that legitimate object?  This question involves what is referred to in these reasons as “proportionality testing” to determine whether the restriction which the provision imposes on the freedom is justified.

The proportionality test involves consideration of the extent of the burden effected by the impugned provision on the freedom.  There are three stages to the test – these are the inquiries as to whether the law is justified as suitable, necessary and adequate in its balance in the following senses:

suitable – as having a rational connection to the purpose of the provision;

necessary – in the sense that there is no obvious and compelling alternative, reasonably practicable means of achieving the same purpose which has a less restrictive effect on the freedom;

adequate in its balance – a criterion requiring a value judgment, consistently with the limits of the judicial function, describing the balance between the importance of the purpose served by the restrictive measure and the extent of the restriction it imposes on the freedom.

If the measure does not meet these criteria of proportionality testing, then the answer to question 3 will be “no” and the measure will exceed the implied limitation on legislative power.

  1. Those observations have been the subject of refinement—see Brown, 363-364 [104] (Kiefel CJ, Bell and Keane JJ, with whom Gageler and Nettle JJ agreed)—but are apt for present purposes.
  1. In Brown, Gageler J, writing separately to (but consistently with the judgment of) the majority, explained (at 378-379 [164]-[165]):

[N]ot every law which effectively burdens freedom of political communication poses the same degree of risk to the efficacy of electoral accountability for the exercise of legislative and executive power.  For that reason not every law which effectively burdens freedom of political communication needs to be subjected to the same intensity of judicial scrutiny.  The measure of the justification needs to be “calibrated to the nature and intensity of the burden”.

…Where a law effectively burdens freedom of political communication, and does so in pursuit of a legitimate purpose, the degree of fit between means (the manner in which the law pursues its purpose) and ends (the purpose it pursues) needed to conclude that the law is reasonably appropriate and adapted to advance its purpose in a manner that is compatible with the maintenance of the constitutionally prescribed system of government needs to be calibrated to the degree of risk which the burden imposed by the means chosen poses to the maintenance of representative and responsible government.

9.4  Infringement of the right to communicate freely on political matters

  1. The respondents accept that s 13(2)(j) of the Building Code infringes upon the ability of building associations to communicate freely about matters of politics or government. Plainly it does (at least on the construction that I have favoured): it serves to require measures that would have the effect of prohibiting the dissemination of written messages that incorporate the logos, mottos or indicia of building associations via the medium of project-supplied clothing, property or equipment. More accurately, it serves to incentivise code-covered entities to take such measures by excluding those that refuse or fail to do so from consideration for taxpayer-funded construction work. Code-covered entities that fail or refuse to do so in the face of a compliance notice face the possibility of additional sanction (see above, [14]). Either way, it is acknowledged that the section (as I have construed it) has the effect of burdening what would otherwise be the right of building associations to communicate freely about matters of politics or government.
  1. The scale of the infringement falls to be assessed at a later stage of the analysis upon which the court is called to embark.  Nonetheless, it is convenient to record here what is said on that score.  The respondents maintain that such constraint upon the implied freedom as s 13(2)(j) of the Building Code imposes is “slight”, not least because (a) there remains beyond it an ability to communicate, by licence (implied or otherwise), about matters of politics and government via the medium of project-supplied clothing, property or equipment (provided that those communications don’t incorporate relevant logos, mottos or indicia); and because (b) there remains beyond it an ability to communicate at a project level in a relevantly-branded way (provided that that occurs otherwise than via the medium of project-supplied clothing, property or equipment).  It might also be borne in mind that the prohibition (as I have construed it) applies only in respect of code-covered entities, as opposed to the entire construction industry (although it is unclear to what extent the latter extends beyond the former).
  1. The CFMMEU, on the other hand, contends that the infringement to which s 13(2)(j) of the Building Code gives effect is significant.  It serves, so the contention proceeds, to remove from building associations an important and effective mode of communication that cannot be, or easily be, replicated.

9.5  Legitimacy of purpose

  1. The court’s next task is to assess whether the purpose that animates an impugned law or impugned executive conduct is compatible with the maintenance of the system of representative government prescribed by the Commonwealth Constitution.
  1. The purpose to which s 13(2)(j) of the Building Code is directed is clear enough.  It serves to protect against the risk that construction workers might perceive, from the association of relevant logos, mottos or indicia with clothing, property or equipment that belongs to their employer (or to other commercial operators), that they are expected or required to join building associations (or, otherwise, that their right to do or not do so is something other than a matter for them).  That purpose is clear from the text of s 13, which more broadly addresses matters concerning freedom of association within the construction industry.  It is made clearer still by the extrinsic material to which extensive reference has already been made.
  1. There can be no doubt that a law enacted to fulfil the purpose just described is one that is compatible with the constitutionally-prescribed system of representative government.  Likewise, the means employed to that end in this case—specifically, the blanket prohibition against the association of relevant logos, mottos or indicia with project-supplied clothing, property or equipment—are not such as might realistically (or otherwise) imperil that system.  It cannot be said—and I did not understand it to be suggested—that s 13(2)(j) of the Building Code “…adversely impinge[s] upon the functioning of the system of representative government”.

9.6  Advancement of that legitimate purpose

  1. Having concluded that s 13(2)(j) of the Building Code infringes against the constitutionally-implied right to communicate freely about matters of politics and government, and that, in doing so, it is animated by a relevantly-legitimate objective, the court’s attention must now turn to the third, “structured proportionality” aspect of the test that the authorities envisage:  is the restriction inherent in s 13(2)(j) one that can be justified?
  1. To answer that question, it is necessary to consider the extent to which s 13(2)(j) of the Building Code burdens that implied right.  As the authorities make clear, that consideration involves three inquiries; namely, as to whether the provision is “…suitable, necessary and adequate in its balance”.  It is convenient to address each in turn.

9.6.1  Suitability

  1. For “structured proportionality” purposes, an impugned law will be relevantly “suitable” if it has a rational connection to the legitimate purpose that it was designed to serve.
  1. The CFMMEU submits that s 13(2)(j) is not relevantly suitable because:

(1)          other provisions of the Building Code and other legislation prohibit the display of material calculated to coerce membership of building associations; and

(2)          relevantly-branded communications that fall outside the scope of those other provisions cannot, even if displayed upon project-supplied clothing, property or equipment, impinge upon or threaten the right of construction workers to join or not join building associations as they see fit.

  1. Though advanced with force and skill, that submission must be rejected.  The question for consideration presently is whether, as a matter of logic, s 13(2)(j) might achieve the purpose for which it was designed.  In other words, does it serve to protect against the risk that construction workers might, by the association of relevant logos, mottos or indicia with the clothing, property or equipment in use at the construction sites at which they work, be led to feel that membership of a building association is something other than a matter of personal choice?  The court needn’t (and shouldn’t) indulge in value judgments on that score.  It is not necessary that I should be satisfied that, as a matter of fact, that purpose has been achieved, nor that the means chosen to achieve it are “better” or “worse” than some other means that might have been available.
  1. By their written submissions on this topic, the respondents said as follows:
  1. The purpose of the first limb of section 13(2)(j) is to reduce the risk of a person inferring that membership of building association is not a free choice.  The means adopted is the proscription of indicia being applied to employer property or equipment.  Preventing indicia from being applied to employer property furthers that purpose by reducing the risk that an inference about the need for membership is drawn.  This, in turn, realises the Minister’s objective of ensuring that the rights of freedom of association are respected, and the choice to obtain membership of a building association, is truly free.  There is nothing irrational in legislating to reduce or minimise risk. This is an everyday feature of regulation.

(emphasis original)

  1. I accept those submissions.  Section 13(2)(j) of the Building Code is rationally connected to the purpose that it was evidently designed to achieve.

9.6.2  Necessity

  1. In order that it might be considered appropriate and adapted to the objective that it was designed to serve (and, thereby, qualify as a permissible burden on what is otherwise the constitutionally-implied right to communicate freely about matters of politics or government), s 13(2)(j) of the Building Code must be “necessary…in the sense that there is no obvious and compelling alternative, reasonably practicable means of achieving the same purpose which has a less restrictive effect on the freedom”.
  1. Analysis at this level is not to be mistaken for an invitation to embark upon a free-ranging inquiry into the merits of an impugned law or its alternatives:  Brown, 371-372 [139] (Kiefel CJ, Bell and Keane JJ). To a large extent (if not overwhelmingly), what is “necessary” is for the legislature to decide. It is not for the court to assess the relative merits of competing legislative models: Brown, 418-419 [282] (Nettle J). Put another way, it is not for the court to “undertake an hypothetical exercise of improved legislative design”: Murphy v Electoral Commissioner (2016) 261 CLR 28, 53 [39] (French CJ and Bell J). Perhaps for that reason, the authorities speak of hypothetical alternatives that are “obvious and compelling”: something more than mere possibility is required.
  1. Here, the CFMMEU submits that an alternative to s 13(2)(j) of the Building Code (as I have construed it) would be a provision that proscribed “…communications which imply that membership of a building association is anything other than an individual choice…”  Such a provision, it is said, would effectively protect the freedom of construction workers to join or not join building associations; and would do so without burdening (at least to the same degree) the implied constitutional right to communicate freely about matters of politics or government.
  1. Alternatively, the CFMMEU submits that an alternative would have been to mandate that code-covered entities prominently display throughout construction project sites notices that expressly disavow, in a manner apt to overcome any residual doubt, any preference on the part of the management of the project that construction workers should join a building association and/or that make clear that doing or not doing so is a matter for each individual.
  1. Neither of those hypotheticals is an obvious or compelling alternative to s 13(2)(j) of the Building Code.  Both amount to little more than restatements of what freedom of association laws have long required.  As the analysis below shows, those laws have not been as effective within the Australian construction industry as might have been hoped.  Accepting that the purpose to which s 13(2)(j) of the Building Code is directed is to promote the right of construction workers to freely associate (or, more realistically, to freely dissociate)—and to do so by guarding against the risk that construction workers might infer, even wrongly, some commercial preference for union membership (however unlawful that preference, or any practical application of it, might be)—it is what commercial operators do, rather than say, that looms largest.
  1. That acknowledged, a requirement that code-covered entities comply with pre-existing freedom of association requirements, and/or that they exclaim that compliance to those who are engaged to work on their construction projects, does not obviously or compellingly achieve the objective that s 13(2)(j) of the Building Code has been designed to achieve.  In the absence of an hypothetical alternative that does, I accept that the section is relevantly “necessary”.

9.6.3  Adequacy of balance

  1. Having so concluded, the court must then consider whether s 13(2)(j) of the Building Code strikes an adequate balance between realising the legitimate objective for which it has been designed (on the one hand) and the degree to which, in doing so, it burdens the constitutionally-implied right to communicate freely about matters of politics or government (on the other).
  1. In Clubb v Edwards (2019) 267 CLR 171, Kiefel CJ, Bell and Keane JJ (with whom the other members of the court relevantly agreed) observed (at 200-201 [69]-[70]):

69           The question whether a law is “adequate in its balance” is not concerned with whether the law strikes some ideal balance between competing considerations.  It is no part of the judicial function to determine “where, in effect, the balance should lie”.  Rather, the question is whether the law imposes a burden on the implied freedom which is “manifestly excessive by comparison to the demands of legitimate purpose”.

70           Proportionality testing is an assessment of the rationality of the challenged law as a response to a perceived mischief that must also respect the implied freedom.  A law which allows a person to be shot and killed in order to prevent damage to property can be seen to have a connection to the purpose of preventing damage to property.  It may also be accepted that other means of preventing damage to property would not be as effective.  Nevertheless, the law is not a rational response to the mischief at which it is directed because it is manifestly disproportionate in its effect on the peace, order and welfare of the community.  In the same way, it is only if the public interest in the benefit sought to be achieved by the legislation is manifestly outweighed by an adverse effect on the implied freedom that the law will be invalid.

(references omitted)

  1. It is on this score that the court has occasion to consider matters of fact (including matters that were the subject of the evidence to which objection was raised).  In particular, it is necessary to consider:

(1)          the value of what would, but for s 13(2)(j) of the Building Code, be the right of building associations to communicate about matters of politics and government by means of branded messages published by licence (implied or otherwise) upon the clothing, property or equipment supplied at construction sites; and

(2)          the public benefit that inures in ensuring (and the degree to which the section ensures) that construction workers are not led to infer, by the presence of branded messages published by licence (implied or otherwise) upon project-supplied clothing, property or equipment, that membership of building associations is something other than a matter of individual choice.

  1. The evidence adduced by the CFMMEU was clear enough:  the licence that is afforded to building associations to publish branded messages upon project-supplied clothing, property or equipment is of not inconsiderable value.  It affords an effective, efficient and inexpensive means of communicating with construction workers about matters of political significance (and of course, other matters).  Though other means—including email and social media—are available to that end, the evidence (most significantly, the expert evidence) tended to establish that none of them is as convenient or effective as on-site communications.
  1. Accepting all of that (as I do), I am nonetheless of the view that the burden that s 13(2)(j) of the Building Code imposes upon what would otherwise be the constitutionally-implied right of building associations to communicate freely about matters of politics or government is not manifestly excessive when weighed against the benefit that it is designed to achieve.  To explain why that is so, regard might first be had to the precise scope of the burden.  Section 13(2)(j) of the Building Code, properly construed, serves to restrict communications (including communications about matters of politics or government):

(1)          that involve the application of relevant logos, mottos or indicia to project-supplied clothing, property or equipment—which is to say that other modes of written communication that do not involve such application (for example, distributed leaflets and flyers, or branded clothing that isn’t project-supplied) are not proscribed, and nor are communications published by licence that do not contain relevant logos, mottos or indicia; and

(2)          only insofar as concerns the clothing, property and equipment in play at construction sites that are overseen by code-covered entities.

  1. As has been noted, the respondents characterise the burden imposed by s 13(2)(j) of the Building Code (or the statutory regime by which that burden is enforced) as “slight”.  Whether that is a fair characterisation or not needn’t be considered.  It suffices to observe that the burden is not disproportionate to the benefit that the section is designed to realise.  On that score, some observations must be made about the mischief to which the section is directed.
  1. Much of the opinion and hearsay evidence that the respondents sought to adduce (and to which objection was taken) concerned what was said to be, at the risk of over-simplification, a culture within the construction industry of non-compliance with laws regulating freedom of association—and, perhaps more broadly, a culture of complicity in, tolerance of or indifference toward that non-compliance on the part of commercial construction businesses.
  1. It is not in doubt that the construction industry in Australia has not had a happy history—at least not over the course of this century—of compliance with freedom of association laws.  Despite invitations during his cross-examination, Mr Fuller’s evidence fell short of acknowledging that history.  Nonetheless, senior counsel for the CFMMEU very properly acknowledged it:  specifically, that the “record is clear” that, although some within the ranks of building associations have sought to maximise union membership by means of lawful persuasion, others have preferred “a different approach”.
  1. That “different approach” is consistent with what is (or, at least at the time of the Compliance Notice and the trial, was) a central and stated objective of Australia’s largest building association, the CFMMEU:  namely, as its own rules record, to “achieve compulsory unionism and control the supply of labour”.
  1. That objective (and the conduct employed to secure it) dates back several decades.  It has been the subject of investigation by at least three royal commissions, each of which made specific findings about restrictive labour practices employed throughout the construction industry.  In his report, Royal Commission into the activities of the Australian Building Construction Employees’ and Builders Labourers’ Federation (Final Report, May 1982), Mr John Winneke QC (as he then was) referred to:

(1)          the Builders Labourers’ Federation (or “BLF”—a predecessor to what is now the CFMMEU) having “rigidly enforced” a policy of “no ticket, no start” throughout Australia;

(2)          the BLF coercing workers at sites that it “controlled” to purchase memberships (of at least six months’ subscription), whether eligible or not—and taking steps to shut down (that is to say, to orchestrate mass withdrawals of labour at) entire sites in retaliation to the deployment of non-unionised staff; and

(3)          the “no ticket, no start” or “closed shop” rule having been “acquiesced in by both builders and unions alike” apparently because it “…provided the best recipe for industrial harmony on building sites”.

  1. A decade later, the Royal Commission into Productivity in the Building Industry in New South Wales  (Final Report, May 1992) made very similar findings in respect of the activities of the Building Workers Industrial Union (or “BWIU”—another predecessor to what is now the CFMMEU) in New South Wales.  The Commissioner, Mr Roger Gyles QC (as he then was), made note of the BWIU’s application of “no ticket, no start” practices, particularly on “major commercial sites”; and to their having been conceded, if not agreed with in principle, by building employers.
  1. In 1996, with the passing of the Workplace Relations Act 1996 (Cth), “no ticket, no start” policies—or, more accurately, the various examples of conduct that had hitherto been employed to enforce them—were outlawed. Yet, in 2003, the Royal Commission into the Building and Construction Industry (Final Report, February 2003; better known as the Cole Royal Commission) made familiar findings about the prevalence of compulsory unionism practices at Australian construction sites.  Amongst the findings made was an acknowledgement that “…the central problem facing the building and construction industry…is how to ensure that workers are free, if they so choose, not to join a union” (emphasis original) and that there was “…no room for doubt that the objects of the Workplace Relations Act 1996 (C’wth) concerning freedom of association are not being achieved in the building and construction industry”.  The report identified a culture or belief on the part of employers within the industry, borne in part of historical experience, that “having a highly unionised workforce minimises the risk of industrial disruption”.  It also referred to evidence suggesting that construction contractors were complicit in measures designed to secure universal union membership at project sites; and doing so with a view to “…avoid[ing] industrial disputation, to win work, or to keep peace with head contractors or unions”.
  1. In the years spanning the Honourable Terence Cole RFD QC’s report in 2003 and the introduction of the Building Code (in its present form) in 2016, the CFMMEU regularly found itself on the wrong end of successful allegations involving conduct consistent with its favouring a policy of “no ticket, no start”.  So much was that so that its apparent indifference toward judicial admonition in that regard has been the subject of some consternation, particularly in this court:  see, for example, Australian Building and Construction Commissioner v Construction, Forestry, Mining and Energy Union (2017) 254 FCR 68, 102 [159] (Dowsett, Greenwood and Wigney JJ); Australian Building and Construction Commissioner v Construction, Forestry, Mining and Energy Union (2018) 265 FCR 208, 215 [22]-[23] (Tracey J), 228 [77] (Logan J); Australian Building and Construction Commission v Construction, Forestry, Mining and Energy Union (Werribee Shopping Centre Case) [2017] FCA 1235, [23], [28] (Tracey J); Australian Building and Construction Commissioner v Construction, Forestry, Mining and Energy Union (Footscray Station Case) (2017) 274 IR 460, 473-474 [50]-[52] (Tracey J); Director of the Fair Work Building Industry Inspectorate v Construction, Forestry, Mining and Energy Union (The Mitcham Rail Case) [2015] FCA 1173 [29] (Jessup J); Director of the Fair Work Building Industry Inspectorate v Construction, Forestry, Mining and Energy Union (No 2) [2016] FCA 436 [139]-[140] (Mortimer J); Australian Building and Construction Commissioner v Construction, Forestry, Maritime, Mining and Energy Union (Geelong Grammar School Case) (No 2) [2019] FCA 1498 [41] (Mortimer J); Australian Building and Construction Commissioner v Construction, Forestry, Maritime, Mining and Energy Union (Castlemaine Police Station Case No 2) [2020] FCA 202 [21], [28] (Anastassiou J). For the CFMMEU in particular, the objective of “no ticket, no start”—and the conduct that it has waged in support of it—has cost substantial sums of money in penalties.
  1. Under the light afforded by that history, it falls to the court to assess the relative extent of the burden imposed by s 13(2)(j) of the Building Code upon what is otherwise a constitutionally-implied right to communicate freely on matters of politics or government.  On any view, the section has been incorporated to address a significant, cultural phenomenon within the construction industry that has endured for a long time, notwithstanding the very considerable efforts—legislative, executive and judicial—that have been engaged to address it.
  1. There is an obvious connection—amplified by the historical context—between the perception of construction workers as to whether or not they are free to join or not join a building association (on the one hand) and the willingness of a project principal to be associated (through the media of project-supplied clothing, property or equipment) with the activities, interests or campaigns that a building association might choose to advance.  Whether that will be so in every case can legitimately be doubted; but the scope for the perpetuation of concerns that have long affected the industry cannot be.
  1. In saying so, I should not be mistaken for passing any judgment about the social importance of the policy end to which s 13(2)(j) is directed, nor about the importance of preserving unburdened the implied freedom.  That is not the comparison to be made in determining whether an impugned law or executive action can properly be considered “adequate in its balance”:  Clubb v Edwards (2019) 267 CLR 171, 201 [72] (Kiefel CJ, Bell and Keane JJ). At issue is simply whether the burden that s 13(2)(j) imposes upon the right of building associations to communicate freely on matters of politics or government can be impugned as “manifestly excessive by comparison to the demands of legitimate purpose”: McCloy v New South Wales (2015) 257 CLR 178, 219-220 [89]-[92] (French CJ, Kiefel, Bell and Keane JJ); Brown, 422-423 [290] (Nettle J).
  1. It cannot be.  The restriction inherent in s 13(2)(j) is not relevantly “undue”:  Lange v Australian Broadcasting Corporation (1997) 189 CLR 520, 575 (Brennan CJ, Dawon, Toohey, Gaudron, McHugh, Gummow and Kirby JJ).

9.7  The provision is valid

  1. Section 13(2)(j) of the Building Code (or the statutory regime that requires compliance with it) does not impermissibly burden the implied constitutional right of free political communication.  It necessarily follows that the relief for which the CFMMEU agitated on the strength of the contrary argument should not be granted.

PART 10:  CONCLUSION

  1. There is no basis upon which to set aside the Compliance Notice, nor otherwise to grant declaratory relief ancillary to that cause.  The application should be dismissed.  There shall be orders to those effects.  The stay order that Bromberg J made in February 2020 will cease as a result.
  1. The respondents do not seek any order for costs as against Lendlease.  They do, however, wish to be heard on the question of costs as against the CFMMEU.  In those circumstances, the respondents and the CFMMEU should liaise as to what, if any, order as to costs is appropriate.  If they can reach agreement on that score (and unless there is reason not to), I will make orders by consent; otherwise, the matter will be listed for further hearing limited to that issue.  The respondents should advise my chambers accordingly within 14 days.”

 

Lendlease Building Contractors Pty Limited v Australian Building and Construction Commissioner (No 2) [2022] FCA 192 (11 March 2022) (Snaden J)