AustLII Home | Databases | WorldLII | Search | Feedback

Sydney Law Review

Faculty of Law, University of Sydney
You are here:  AustLII >> Databases >> Sydney Law Review >> 2009 >> [2009] SydLawRw 25

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

Oreb, Naomi --- "Betting Across Borders - Betfair Pty Limited v Western Australia" [2009] SydLawRw 25; (2009) 31(4) Sydney Law Review 607

Case Note

Betting Across Borders — Betfair Pty Limited v Western Australia

NAOMI OREB[∗]

Abstract

This case note examines the recent High Court decision of Betfair Pty Limited v Western Australia. This decision focused on s 92 of the Constitution. First, the author argues that the decision is significant because it offered a restatement of principle, re-contextualised s 92 in a ‘new economy’ and confirmed the relevance of early US Commerce Clause authority in Australian s 92 jurisprudence. Second, the author surveys proposed changes to the test of ‘discriminatory protectionism’ and concludes that the current test is preferable to these alternatives.

1. Introduction

Until the recent decision of Betfair Pty Limited v Western Australia,[1] the High Court had not considered a challenge to the validity of legislation based on s 92[2] of the Constitution for almost twenty years. Numerous changes to the nature of ‘trade and commerce’ within Australia during that time, such as the 1995 Intergovernmental Competition Principles Agreement[3] and the emergence of internet-based businesses meant that Betfair presented a novel opportunity to reconsider the role of geographic boundaries in s  92 decision-making, the relevance of United States Commerce Clause jurisprudence, and the suitability of the existing test for invalidity under s 92. Although the court responded to the first two of these considerations, it was not invited to review the invalidity test in its current form.

This case note is broadly comprised of two main parts. First, the majority judgment (Gleeson CJ, Gummow, Kirby, Hayne, Crennan and Kiefel JJ) and the concurring judgment by Heydon J are analysed. Second, three suggestions made by commentators as to how the current invalidity test and accompanying saving test may be reformulated are considered. The present author argues that the court should not alter the current invalidity test nor the current saving test by (i) removing the protectionism requirement in favour of a non-discrimination norm, (ii) requiring protectionism in intent for invalidity, or (iii) applying a more rigorous saving test broadly referred to in the literature as one of ‘robust proportionality’. The existing s  92 test is preferable to these proposed alternatives.

2. The Facts

Betfair Pty Limited (‘Betfair’) was granted a licence to operate as a ‘betting exchange’ in Tasmania in 2006.[4] Betfair uniquely enables customers from anywhere in Australia to bet against each other that a particular outcome will or will not occur. That is, they can bet on a competitor to win or to lose a particular event. Betfair facilitates these transactions rather than bearing the risk (like a traditional bookmaker), and makes its profits by taking a small commission of up to 5 per cent from the winnings of each transaction.[5]

The Betting and Racing Legislation Amendment Act 2006 (WA) inserted amendments to the Betting Control Act 1954 (WA) which formed the subject of the challenge. It became an offence to place a bet using a betting exchange (s 24(1aa)), as well as an offence to publish a Western Australian race field for business purposes without approval from the Western Australian Minister for Racing and Gambling (s 27D(1)). Betfair was denied approval in October 2007.[6] The co-plaintiffs, Betfair and Mr Erceg (a resident of Western Australia and registered Betfair user) represented interstate ‘supply’ and intrastate ‘demand’ respectively. The co-plaintiffs turned to s  92 and argued that s  92 was engaged because the legislation stopped an increase in competition for betting services that would otherwise have occurred in Western Australia, whilst also conferring competitive disadvantages on Betfair that did not apply to other Western Australian operators. The plaintiffs launched an additional constitutional argument based on an alleged s  118[7] inconsistency between Tasmanian and Western Australian legislation that the court declined to answer.

3. The Existing Framework for Considering s 92

The majority, with Heydon J concurring, invalidated both legislative provisions upon application of the test of ‘discriminatory protectionism’ that was incrementally developed in Cole v Whitfield [8] and Castlemaine Tooheys Ltd v South Australia.[9] The Betfair decision is essentially an orthodox application of this particular test.[10] This test states that a law will be discriminatory if ‘the law on its face subjects that trade or commerce to a disability or disadvantage or if the factual operation of the law produces such a result’,[11] thereby conferring a competitive advantage on ‘intrastate trade and commerce of the same kind.’[12] The accompanying test of protectionism is a negative test involving ‘issues of fact and degree’ that may ultimately ‘depend upon judicial impression’.[13] More specifically, a law will be of a protectionist character if it exhibits protectionism in form or in effect.[14] Conversely, the law will not be of a protectionist character if it ‘has as its real object the prescription of a standard for a product or a service or a norm of commercial conduct’.[15] However, even if the law is protectionist in effect, it will avoid offending s  92 if it is ‘necessary or appropriate and adapted’ to the achievement of legitimate, non-protectionist ends (for instance, social and/or environmental problems), and ‘any burden imposed on interstate trade was incidental and not disproportionate to their achievement.’[16]

4. The Decision

A. The Majority Decision

(i) Discriminatory Protectionism

The majority held that the discrimination here was between ‘different but competing’[17] ways of betting — namely, betting exchanges and other forms of bookmaking. This confirmed that in the context of s 92, the subject matter in question (for example, the goods or services) need not be identical but merely capable of competition.[18] Despite enduring concerns about the court’s expertise in dealing with complex economic evidence in s 92 cases[19] (particularly in the absence of an agreed statement of facts from the parties),[20] the court found that there was substitutable demand between these different betting options. That is, punters were unlikely to use a traditional bookmaker or TAB as well as a betting exchange — it was one or the other. This evidence established competition in the market: an essential prerequisite if the legislation was to be regarded as potentially conferring a competitive disadvantage on Betfair. In summary, it was held that s  24(1aa) conferred a discriminatory burden of a protectionist kind on Betfair, because denying Betfair access to punters in Western Australia was to the benefit of existing operators in that State.[21] Section 27D(1) was also discriminatory and protectionist because denying Betfair access to information that formed part of its trading operations also conferred a competitive advantage on Western Australian operators.[22]

(ii) Proportionality and/or an Acceptable Justification

Given the multiplicity of alleged objects behind the legislation, the court confirmed, in line with US Commerce Clause authority,[23] that a discriminatory and protectionist law ‘is not saved by the presence of other objectives such as public health which are not protectionist in character’.[24] This is because such a law is still discriminatory and protectionist in contravention of the object of s 92. Therefore the existence of one protectionist object would have been enough to satisfy the test of discriminatory protectionism, allowing for a robust operation of s 92.

The legislative preamble stated that the amendments sought ‘to prohibit betting through, and the establishment and operation of, betting exchanges’[25] because betting exchanges (a) ‘make no contribution to the racing industry’[26] and (b) threaten the integrity of the industry (by allowing people to place bets that a certain horse, for example, will lose).[27] In the majority judgment, the objects of both s 24(1aa) and s 27D(1) were largely considered together. First, in relation to the integrity argument, the legislative provisions were held not to be proportionate or appropriate and adapted to this object (concepts now said to involve ‘reasonable necessity’),[28] because there was a lack of evidence of dishonest practices attributable to Betfair.[29] Thus even if that object was legitimate, the means chosen to achieve the end were not necessary and the legislation offended s 92.[30] Second, in relation to the contribution argument, the majority found that Betfair made the same financial contribution as all other Western Australian bookmakers.[31] Thus this object also failed to justify the discriminatory protectionist effect of the legislation. Finally, the fact that the legislation ultimately protected the turnover of intrastate revenue was quickly dismissed as an illegitimate purpose in principle and on the authorities.[32]

Thus the court’s treatment of the objects and actual effects of the legislation meant that it did not need to come to a conclusion about the ‘real object’ at all. The absence of such a discussion does not necessarily indicate that the court was downplaying the ‘necessary or appropriate and adapted’ qualification laid out in Castlemaine. Instead, this seems to be a mere consequence of the fact that the legislation did not appear to have a relevantly legitimate end for consideration.

(iii) The New Economy

The final issue to point out is that the court considered the operation of s 92 in the context of a ‘new economy’.[33] The majority explicitly moved away from previous language about ‘the people of’ a State and ‘its’ well-being in recognition of (a) the people who are not resident in a particular State but are there present ‘at any particular time’, and (b) the growth of ‘instantaneous commercial communication’.[34] The traditional maintenance of explicit geographic boundaries does make the exercise of establishing a benefit to one State and not another more straightforward. However, downplaying geographic boundaries in this way may mean that it will now become more difficult for plaintiffs to show that a particular State has been prima facie affected by certain legislation, thereby making it more difficult to prove that the legislation in question has a discriminatory effect. If so, this would limit the operation of s 92 and potentially permit States to enact legislation conferring a competitive advantage on intrastate goods and services.

B. Heydon J’s Concurring Judgment

Heydon J came to the same conclusion but differed in his application of the invalidity test. Heydon J engaged in a more rigorous and detailed explanation of how the legislation was drafted and exactly what types of entities fell within the ambit of the provisions. In sum, Heydon J found that the provisions conferred an impermissible burden on interstate trade by protecting intrastate traders ‘from the rivalry they would otherwise face’.[35] As to whether this restriction could be justified, Heydon J found that the integrity rationale was not the ‘real object’ at all,[36] because ‘the only purpose [was] protectionist’.[37] Heydon J focused on the considerable width of operation of these provisions. First, s 24(1aa) prohibited bets on all sporting events occurring anywhere in the world (not just horse racing in Western Australia), and second, it also applied to all punters in Western Australia (not just ‘that very narrow class of persons’ who might seek to engage in race-fixing).[38] Additionally, in relation to s 27D(1), Heydon J rejected the contribution argument on the grounds that the legislation did not provide for the making of financial or non-financial contributions.[39]

5. Still Unanswered: Proposed Changes to the Invalidity Test

Betfair did very little to respond to any of the commentary that has emerged over the last twenty years about possible changes to the test of ‘discriminatory protectionism’, given that the court was not invited to reconsider the current formulation of the test. The Cole judgment arguably anticipated the development of new formulations, for as was acknowledged in that case, ‘[i]nevitably, the adoption of a new principle of law, though facilitating the resolution of old problems, brings a new array of questions in its wake.’[40] There remains however, much scope for disagreement. Amongst those that agree that change is needed, three different reforms have been put forward. The first two proposals involve changing the invalidity test by (i) removing the protectionism requirement altogether, or (ii) requiring protectionism in intent for invalidity. The present author disagrees with both of these suggestions. The third proposal involves altering the saving test by moving from ‘abridged’ to ‘robust’ proportionality. The present author also disagrees with this suggestion. The merits of each proposition are here considered in turn, and the situations in which they might produce different results to both the current test and each other are highlighted.

A. Removing ‘protectionism’ altogether for a non-discrimination norm

At present, protectionism remains the backbone of the s 92 invalidity test[41] and has created a narrow scope for the invalidity of legislation. It has been argued that,[42] and questioned whether,[43] the ‘protectionist’ requirement should be abolished in favour of a non-discrimination norm. This proposal, as argued by Dr Gonzalo Villalta Puig, for example, removes protectionism altogether, rather than relying on discrimination as a means of ipso facto discovering protectionism. If the court removed the protectionist requirement according to this suggestion, the invalidity test would simply ask whether there was discrimination either on the face of the legislation or in effect. If this was answered in the affirmative, the court would proceed to the saving test.

The first supposed reason for doing so is premised on the need to give greater effect to the intention of the framers. As Puig argues, ‘the vision of a common market contemplated by the founders calls for the removal of discrimination of any kind, whether protectionist or not.’[44] Abolishing the requirement for discrimination of a protectionist kind would likely mean that the invalidity test would be satisfied quite easily, given that it is easier to prove discrimination per se than discriminatory protectionism. Thus the court would potentially be compelled to apply the saving test far more frequently. Yet this would not have a tangible effect on the result as the saving test would probably produce much the same outcome as the current test, irrespective of whether protectionism is required at the invalidity stage. That is, any legitimate end that would have otherwise been an antidote for protectionism would still be an antidote to discrimination. Therefore, whether the end is legitimate or not, the proportionality requirement will also tend to be decided in the same way. In Betfair, for instance, the result would not have differed had the court applied a non-discrimination norm. Thus merely removing the protectionist requirement does not necessarily bring the test any closer to the intention of the framers.

The second argument in favour of change is that ‘discrimination’ and ‘protectionism’ are not discrete concepts and tend to overlap. Case law in this area does indicate that ‘discrimination’ often bespeaks ‘protectionism’ as it is, and protectionism is generally borne of, and pursued through, discrimination. It may be that ‘the only evidence of discrimination [is] the protectionist effect’,[45] and as Professor Michael Coper has argued, it is difficult ‘to imagine a protectionist law that is not discriminatory, unless by definitional fiat we confine the concept of discrimination to discrimination on the face of the law.’[46] That said, there are conceivable instances of laws that are non-discriminatory and protectionist. One example would be a lax environmental regulation designed to cut the compliance costs incurred by intrastate traders and thus increase their efficiency;[47] another would be Professor P  H Lane’s often-cited example of a State that restricts interstate and intrastate butter sales in order to protect an intrastate margarine industry.[48] At the same time, there are also conceivable instances of laws that are discriminatory but non-protectionist, such as a ‘general law fixing the price of commodities which, in fact, are produced in only one other State, the price operating to destroy all trade in such commodities between the States.’[49] Yet the fact remains that more often than not, the concepts remain intermingled, just as they were in Betfair, where the majority seemed to suggest that a ‘discriminatory’ law is one imposing competitive disadvantage, and a ‘protectionist’ law is one restricting competition. Thus under this view, mere discrimination should be enough for invalidity because the additional protectionist requirement is largely redundant.

The present author does not believe that protectionism should be removed because softening the invalidity test would risk taking the focus away from notions of economic competition. What the concept of protectionism is often said to add to the application of s 92 is a technical economic analysis of any particular market. Unlike Betfair, it is not difficult to envisage a scenario in which it is quite difficult to discern whether any competition between non-identical goods even exists. As Christopher Staker has asked,

[s]uppose that a State with a small car industry adopted measures to restrict the import of large luxury cars from other States. Do small cars and luxury cars compete in the same market? And do cars compete with bicycles? … is the degree of competition between products a relevant consideration?[50]

For this reason the protectionist requirement is often described as ‘highly impractical’,[51] but the likely consequence of ‘removing’ it is that these questions would simply arise in relation to ‘discrimination’ instead, when the court would have to decide whether or not two unlike things (potentially ‘alike’ for the purposes of market analysis) were treated differently. That is, these concerns would merely be displaced to create a more economic conception of ‘discrimination’. These concerns would not disappear entirely, unless the court reverted back to the ‘criterion of operation’ approach supported by Dixon J (as he then was) in cases such as O Gilpin Ltd v Commissioner for Road Transport and Tramways (NSW),[52] and ceased examining the practical effect of legislation at all. As such, the court should not abandon the protectionism requirement because it points the court and counsel towards considering the unique type of discrimination that is required. This is quite important given that no such specificity is inhered within the notion of ‘discrimination’ itself.

B. Requiring Protectionism in Intent for Invalidity

Amelia Simpson has argued that the invalidity test should accept ‘improper legislative purpose as a necessary, rather than a merely sufficient, element in triggering the provision’.[53] This would mean that unintentionally protectionist and discriminatory laws would no longer fall foul of s  92. Prima facie, such a test would diminish the impact of s 92, as (i) there would be fewer ways in which the invalidity test could be satisfied, and (ii) the remaining ways would require the court to be less deferential to Parliament (an approach the court may or may not be prepared to adopt). If so, this proposal could potentially have the opposite effect to Puig’s proposal (above), or it could be that it merely subsumes an inquiry into the effects of legislation within a broader, objective examination of legislative purpose.

Simpson argues that one key benefit of this model is that it largely avoids the tough economic analysis and procedural fact-finding associated with establishing protectionism in effect.[54] Simpson concedes however, that the court would still consider the effects of the legislation because ‘those effects will often be the most reliable objective indicators of purpose.’[55] That is, the question of ‘what did the legislature objectively intend to happen?’ really requires an analysis of what actually happens. Accordingly, a law with a professed non-protectionist purpose could be regarded as having a ‘real object’ that was protectionist in nature if it was so over-zealous and disproportionate in its attempts to secure the non-protectionist object that the non-protectionist object could not, objectively, really be said to exist.[56] Thus the effects of legislation are contained within a consideration of legislative intent, rather than abandoned altogether or compelled to exist as an independent inquiry.

Simpson’s test would likely diminish the effect of s  92 by permitting ‘over-zealous’ laws with protectionist effects to remain valid. The point can be illustrated thus. Under the status quo, a law that is non-protectionist in intent but protectionist in effect will satisfy the invalidity test due to that protectionist effect. This is enough to activate the saving test. That law will then be (i) valid if the means are proportionate to achieving the non-protectionist end, but (ii) invalid if the means are disproportionate to achieving that end (that is, if the legislature is over-zealous). In contrast, under Simpson’s model, a law that is apparently non-protectionist in intent but protectionist in effect could (a) still satisfy the invalidity test if it was deemed to be protectionist in intent (that is, if the court held that its ‘real object’ was indeed protectionist), or (b) fail to satisfy the invalidity test if it was still regarded as non-protectionist in intent, and simply an over-zealous attempt to secure that non-protectionist end. Thus legislation type (a) would face the saving test, but type (b) would not, because these protectionist effects would not initially satisfy the invalidity test. Problematically, legislation type (b) would be valid under Simpson’s test, even though it unnecessarily burdens interstate trade at the expense of a common market. Although some ‘discriminatory protectionist’ laws are currently held to be valid so that State Parliaments can achieve legitimate social or environmental goals through regulatory measures, the existence of s  92 makes it difficult to justify why Parliaments should be able to legislate beyond what is reasonably necessary to achieve these goals. Simpson’s test would give too much power to State parliaments to legislate in respect of trade and commerce and, given the intention of the framers (discussed above), should fall foul of s 92.

Acknowledging that legislation with a professed protectionist object is so rare, it is likely that the court could only ever invalidate legislation under Simpson’s test by finding something more than a protectionist effect. That is, the court would have to look to objective factors of legislative intent before overriding (or ‘second-guessing’) parliament by declaring that the ‘real object’ of legislation was indeed protectionist in nature. This means that the court would probably have to become less deferential to parliament if legislation were to satisfy Simpson’s s 92 invalidity test. An inquiry into the ‘real object’ of legislation effectively requires ‘an assessment that particular benefits, secured in given magnitude, would be marginal, trivial, or otherwise unworthy of legislative attention’.[57] Hand-in-hand are considerations of whether a ‘particular legislative enactment is a necessary or even a desirable solution to a particular problem’[58] — the exact considerations that the majority in Castlemaine thought were ‘in large measure a political question best left for resolution to the political process.’[59] The majority in Betfair did not go this far, even though Heydon J held that the stated object was unnecessary and thus not the ‘real object’ at all. The majority did, however, confirm that they would have been perfectly permitted to do so had they thought it necessary, because ‘Ch III of the Constitution commits to the federal judicial power the determination of [whether a law possesses a particular attribute]’.[60] Even if the court is willing to take a more evaluative approach to concepts of proportionality, it remains unsettled as to whether it would actually comment on the ‘real object’ of legislation with any real frequency, potentially diminishing the impact of s 92. When married with the consequence of probably permitting over-zealous laws with protectionist effects to remain valid, this approach is likely to weaken the impact of s 92 and should not be adopted.

C. Moving from ‘abridged proportionality’ to ‘robust proportionality’

Changing the current saving test is not necessarily mutually exclusive from the implementation of one of the above reforms to the invalidity test.[61] The present saving test is frequently referred to in critical literature as a test of ‘abridged proportionality’. A primary criticism of the current test is that it allows judges to engage in ‘evaluative reasoning ... at best indirectly and minimally’,[62] despite the fact that ‘[p]roportionality, by its very nature, cannot be applied in a value-neutral manner.’[63] That is to say, the current saving test requires the court to balance the means against the purported end in somewhat of a vacuum. The court does not have to balance, clearly and directly, the purported end against other possible ends or broader social interests which would deem the burden either desirable or undesirable.

Numerous commentators have called for a saving test encompassing ‘robust proportionality’, though this is not routinely applied in any other area of constitutional law.[64] ‘Robust proportionality’ includes the current saving test but with additional considerations (hence the current test is an ‘abridged’ version of this fuller ‘robust’ test). A test of robust proportionality would compel the court to apply a three stage examination of (i) the appropriateness of the law (the current test), as well as (ii) the necessity of the legislation, and (iii) a balancing of the imposed burden against the non-protectionist object of the legislation. Considerations (ii) and (iii) compel the court to enter uneasy territory it has thus far sought to avoid — namely, questions about the necessity and desirability of the legislation in question. Puig has claimed that since Betfair, ‘the Cole v Whitfield saving test for s 92 is only one level away from robust proportionality’[65] given the court’s apparent engagement with stage (ii) necessity reasoning. In Betfair the majority held that ‘it cannot be found in this case that prohibition was necessary in the stated sense for the protection or preservation of the integrity of the racing industry.’[66] Yet it is perhaps overly enthusiastic to suggest that the court was evidently coming closer to an acceptance of robust proportionality based on this extract, given that the majority did not elaborate any further on this reasoning. The discussion of what was ‘necessary’ was fleeting, and in any case, quite straightforward on the facts. The real test will come if and when the court is asked to answer the question of necessity in a so-called ‘hard case’, because ‘the significance attached to the extent of the restriction will depend on the importance of the protected interest. Thus, in an indirect fashion, necessity may presuppose some weighting of interests.’[67] An extended discussion of this nature was simply not required in Betfair.

Irrespective of whether Betfair indicated the possibility of incrementally adopting a robust proportionality approach, the far bigger hurdle is the stage (iii) balancing reasoning. As Jeremy Kirk suggests, there has been a ‘general aversion of Australian judges to being seen to intrude on the political or legislative domain.’[68] Whilst such an aversion is by no means problematic as a general rule, actually to ‘balance’ competing social interests, the court must ask ‘essentially the same question as that addressed by the legislature.’[69] There was a hint that the court was willing to become less deferential to Parliament (as mentioned above) which may prove significant, because as Kirk explains, a robust balancing test does require the court to engage in a legislative-like reasoning process:

If a law seeks to protect the environment, for example, but does not in practice afford much protection, then the net significance of the beneficial effects would be slight. To some extent, therefore, the application of a balancing test does involve the court deciding on the desirability or necessity of a government end.[70]

On the one hand, some of the arguments in favour of robust proportionality do initially seem quite compelling. In light of the intention behind s  92, Puig offers probing food for thought when he says that ‘[a] saving test that validates laws or measures that do not have a protectionist purpose but that, nonetheless, have a protectionist effect [if appropriate and adapted to the non-protectionist purpose] is nothing but absurd.’[71] Puig thus suggests that the court should move away from focusing on purpose and the question of whether or not the means are appropriate to that purpose. The robust proportionality test would do just that, as it compels the court to pay greater attention to the effect of the law in question, whether or not it fractures a common market, and how it compares to other legislative alternatives.

However, the fact of the matter is that even if the robust proportionality test does more accurately reflect the intent of the framers, advocates of this view seem to pay little notice to the way in which the court would actually have to apply the test. The court is simply not equipped to make the kinds of determinations required by a robust proportionality saving test. As Puig suggests, the stage (ii) necessity reasoning essentially ‘assesses the possibility of legislative choice.’[72] Presumably then, if the question of necessity was quite contentious in any given matter, counsel would need to provide evidence of the other ways in which a policy outcome could be achieved, and the court would need to compare the relative efficacy, financial costs, and political costs associated with each. However, the court cannot always come to terms with which of these options are truly realistic ones given the nature of political deal-making and compromise, and the requisite sensitivity to electoral support that are inevitably factored into parliamentary decision-making. Moreover, given the burdensome task of fact-finding already associated with s  92 cases, this could make s 92 litigation even more resource-intensive and expensive for the parties involved. In addition, the issues raised by stages (ii) and (iii) of the robust proportionality test are inherently political questions. Judicial determinations as to whether or not a particular legislative enactment was desirable may erode confidence in the judiciary, as the court may be perceived as giving effect to personal political bias, as well as making political assessments at the expense of the democratic mandate granted to parliament. For these reasons it is preferable that judicial discretion is contained within the ‘appropriate and adapted test’ rather than extended to operate in an inescapably political sphere through the application of robust proportionality.

6. Conclusion

The general response to the Betfair decision is perhaps best summed up as a combination of excitement that a s  92 decision was handed down, and some disappointment that the test of ‘discriminatory protectionism’ was not altered in any way. One comment was made that ‘Betfair … is too timid and conformist a decision to be memorable.’[73] Given that the court was not invited to reconsider the invalidity or saving tests in their current form, and Betfair was not a particularly ‘hard case’ on the facts, such an assessment is perhaps too disparaging of the court. Despite a lack of dramatic reform to existing law, Betfair is still significant for its restatement of principle, its re-contextualisation of s92 in a ‘new economy’ and its confirmation that early United States Commerce Clause authority remains relevant to s  92. It remains open whether the court will implement changes to the current invalidity or saving tests, but the above analysis suggests that these proposed changes are largely undesirable. Thus even if Betfair is not particularly ‘memorable’, it is, in the present author’s opinion, quite correct.


[∗] Final year student, Bachelor of Laws, University of Sydney. The author wishes to thank Associate Professor Peter Gerangelos for supervising this case note.

[1] [2008] HCA 11; (2008) 234 CLR 418 (‘Betfair’).

[2] Section 92 states that: ‘On the imposition of uniform duties of customs, trade, commerce and intercourse among the States, whether by means of internal carriage or ocean navigation, shall be absolutely free.’

[3] Betfair [2008] HCA 11; (2008) 234 CLR 418, 16 (joint judgment). The Agreement allows restrictions to competition only if there is no other means of achieving a particular purpose or the benefits of restriction outweigh the cost of doing so. Thus the Agreement buttresses s 92 and should mean that fewer s 92 cases arise for consideration.

[4] This licence was granted under Div 5 of Pt 4A of the Gaming Control Act 1993 (Tas) following the enactment of the Gaming Control Amendment (Betting Exchange) Act 2005 (Tas).

[5] Betfair [2008] HCA 11; (2008) 234 CLR 418, 57–8 (joint judgment).

[6] Betfair [2008] HCA 11; (2008) 234 CLR 418, 74 (joint judgment).

[7] Section 118 states that: ‘Full faith and credit shall be given, throughout the Commonwealth to the laws, the public Acts and records, and the judicial proceedings of every State.’

[8] [1988] HCA 18; (1988) 165 CLR 360 (‘Cole’).

[9] (1990) 169 CLR 436 (‘Castlemaine’). Other cases relevant to the development of these tests include Bath v Alston Holdings Pty Ltd [1988] HCA 27; (1988) 165 CLR 411 (‘Bath’) and Barley Marketing Board (NSW) v Norman [1990] HCA 50; (1990) 171 CLR 182 (‘Barley’).

[10] See Cole [1988] HCA 18; (1988) 165 CLR 360, 394 (unanimous judgment).

[11] Cole [1988] HCA 18; (1988) 165 CLR 360, 399. For greater discussion of the ‘general features of a discriminatory law’ see Castlemaine (1990) 169 CLR 436, 478 (Gaudron and McHugh JJ).

[12] Cole [1988] HCA 18; (1988) 165 CLR 360, 407.

[13] Cole [1988] HCA 18; (1988) 165 CLR 360, 407–8.

[14] Cole [1988] HCA 18; (1988) 165 CLR 360, 408.

[15] Cole [1988] HCA 18; (1988) 165 CLR 360, 408.

[16] Castlemaine (1990) 169 CLR 436, 472–4 (Mason CJ, Brennan, Deane, Dawson and Toohey JJ).

[17] Betfair [2008] HCA 11; (2008) 234 CLR 418, 122 (joint judgment).

[18] Affirming Barley [1990] HCA 50; (1990) 171 CLR 182 whilst also providing a variation on the ‘scarce resource or the most inexpensive supplies of a raw material’ examples given in that case.

[19] See, for example, Christopher Staker, ‘Section 92 of the Constitution and the European Court of Justice’ (1990) 19 Federal Law Review 322; Andrew S Bell, ‘Section 92, Factual Discrimination and the High Court’ (1991) 20 Federal Law Review 240.

[20] An agreed statement of facts was presented in both Cole [1988] HCA 18; (1988) 165 CLR 360 and Castlemaine (1990) 169 CLR 436.

[21] Betfair [2008] HCA 11; (2008) 234 CLR 418, 122 (joint judgment).

[22] Betfair [2008] HCA 11; (2008) 234 CLR 418, 118 (joint judgment).

[23]vBetfair [2008] HCA 11; (2008) 234 CLR 418, 38–49 (joint judgment). Key Commerce Clause authorities here cited were pre-1900 decisions that had influenced the drafting of s 92, including Minnesota v Barber, [1890] USSC 190; 136 US 313 (1890).

[24] Betfair [2008] HCA 11; (2008) 234 CLR 418, 47 (joint judgment).

[25] Betfair [2008] HCA 11; (2008) 234 CLR 418, 106 (joint judgment).

[26] Betfair [2008] HCA 11; (2008) 234 CLR 418, 106 (joint judgment).

[27] Betfair [2008] HCA 11; (2008) 234 CLR 418, 106 (joint judgment).

[28] Betfair [2008] HCA 11; (2008) 234 CLR 418, 102–3 (joint judgment).

[29] Betfair [2008] HCA 11; (2008) 234 CLR 418, 110 (joint judgment).

[30] Betfair [2008] HCA 11; (2008) 234 CLR 418, 113 (joint judgment).

[31] Betfair [2008] HCA 11; (2008) 234 CLR 418, 107 (joint judgment).

[32] Betfair [2008] HCA 11; (2008) 234 CLR 418, 108 (joint judgment) citing Bath [1988] HCA 27; (1988) 165 CLR 411, 426–7 and Sportodds Systems Pty Ltd v New South Wales [2003] FCAFC 237; (2003) 133 FCR 63, 80.

[33] Betfair [2008] HCA 11; (2008) 234 CLR 418, 20 (joint judgment).

[34] Betfair [2008] HCA 11; (2008) 234 CLR 418, 18 (joint judgment), see also, 89–90 (joint judgment).

[35] Betfair [2008] HCA 11; (2008) 234 CLR 418, 133 (Heydon J). However, note that this comment was made in relation to s 24(1aa) specifically.

[36] Betfair [2008] HCA 11; (2008) 234 CLR 418, 145 (Heydon J).

[37] Betfair [2008] HCA 11; (2008) 234 CLR 418, 134 (Heydon J).

[38] Betfair [2008] HCA 11; (2008) 234 CLR 418, 134 (Heydon J). See also at 145 (Heydon J).

[39] Betfair [2008] HCA 11; (2008) 234 CLR 418, 146 (Heydon J).

[40] Cole [1988] HCA 18; (1988) 165 CLR 360, 408.

[41] See Gonzalo Villalta Puig, The High Court of Australia and Section 92 of the Australian Constitution (2008), 50.

[42] See, for example, Puig, above n41, in particular at 120, 133–5.

[43] See, for example, ‘Notes and Questions’ in George Winterton, H P Lee, Arthur Glass and James A Thomson, Australian Federal Constitutional Law: Commentary and Materials (2nd ed, 2007), 520.

[44] Puig, above n41, 95.

[45] Leslie Zines, The High Court and the Constitution (4th ed, 1997), 149.

[46] Michael Coper, ‘Section 92 of the Australian Constitution Since Cole v Whitfield’ in H P Lee and George Winterton (eds), Australian Constitutional Perspectives (1992) 129, 140.

[47] Sarah Joseph and Melissa Castan, Federal Constitutional Law: A Contemporary View (2nd ed, 2006), 341.

[48] P H Lane, ‘The Present Test for Invalidity Under Section 92 of the Constitution(1988) 62 Australian Law Journal 604, 607.

[49] R v Vizzard; Ex parte Hill [1933] HCA 62; (1933) 50 CLR 30, 93 (Evatt J) as cited in Lane, ‘The Present Test for Invalidity Under Section 92 of the Constitution’, above n48, 607.

[50] Staker, above n19, 344–5.

[51] Gonzalo Villalta Puig, ‘Free Movement of Goods: The European experience in the Australian context’ (2001) 75 Australian Law Journal 639, 650.

[52] [1935] HCA 8; (1935) 52 CLR 189, 205–6 (Dixon J in dissent). Subsequent cases in which this approach was applied by the majority include Hospital Provident Fund Pty Ltd v Victoria [1953] HCA 8; (1953) 87 CLR 1 and Mansell v Beck [1956] HCA 70; (1956) 95 CLR 550.

[53] Amelia Simpson, ‘Grounding the High Court’s Modern Section 92 Jurisprudence: The Case for Improper Purpose as the Touchstone’ (2005) 33 Federal Law Review 445, 462.

[54] Id, 479.

[55] Id, 480–1.

[56] See Heydon J’s judgment in Betfair [2008] HCA 11; (2008) 234 CLR 418. Note too that Heydon J’s judgment was not entirely unlike the finding in Castlemaine (1990) 169 CLR 436, where the ‘real object’ of the legislation was held to be economic protectionism, not litter control or the preservation of gas objects, because the effect of the legislation did not actually improve either of these alleged objects.

[57] Simpson, above n53, 469.

[58] Castlemaine (1990) 169 CLR 436, 473 (Mason CJ, Brennan, Deane, Dawson and Toohey JJ).

[59] Castlemaine (1990) 169 CLR 436, 473 (Mason CJ, Brennan, Deane, Dawson and Toohey JJ).

[60] Betfair [2008] HCA 11; (2008) 234 CLR 418, 99 (joint judgment).

[61] See Puig, above n41, 135 for example, who advocates changing both the invalidity test to a non-discrimination norm and changing the saving test to one of robust proportionality.

[62] Simpson, above n53, 457.

[63] Jeremy Kirk, ‘Constitutional Guarantees, Characterisation and the Concept of Proportionality’ [1997] MelbULawRw 1; (1997) 21 Melbourne University Law Review 1, 61.

[64] As noted by Kirk, above n63, 20; confirmed by Gonzalo Villalta Puig, ‘A European Saving Test for Section 92 of the Australian Constitution[2008] DeakinLawRw 4; (2008) 13 Deakin Law Review 99, 116.

[65] Puig, above n41, 190.

[66] Betfair [2008] HCA 11; (2008) 234 CLR 418, 112 (joint judgment) (emphasis added).

[67] Kirk, above n63, 8.

[68] Id, 53.

[69] Id, 54.

[70] Id, 8 (emphasis in original).

[71] Puig, above n64, 124.

[72] Puig, above n41, 147.

[73] Id, 193.


AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/SydLawRw/2009/25.html