Chairperson, Eastern Cape Gambling and Betting Board and Another v Vukani Gaming Eastern Cape (Pty) Ltd and Others (1338/2019; 1366/2019; 119/2020) [2021] ZASCA 180 (17 December 2021)

THE SUPREME COURT OF APPEAL OF SOUTH AFRICA

JUDGMENT

Reportable

Case no: 1338/2019

In the matter between:

GOLDEN PALACE SITE 3 (PTY) LTD

FIRST APPLICANT

GOLDEN PALACE SITE 4 (PTY) LTD

SECOND APPLICANT

GOLDEN PALACE SITE 1 (PTY) LTD THIRD APPLICANT

and

VUKANI GAMING EASTERN CAPE

FIRST RESPONDENT

CHAIRPERSON, EASTERN CAPE

GAMBLING BOARD SECOND RESPONDENT

EASTERN CAPE GAMBLING

AND BETTING BOARD THIRD RESPONDENT

PIONEER SLOTS (PTY) LTD FOURTH RESPONDENT

MARSHALLS WORLD OF SPORT

EASTERN CAPE (PTY) LTD FIFTH RESPONDENT

K2017440277 (SOUTH AFRICA) (PTY) LTD SIXTH RESPONDENT

GSLOTS ISO EC (PTY) LTD SEVENTH RESPONDENT

K2017425418 (PTY) LTD EIGHTH RESPONDENT

SPIN AND WIN ENTERTAINMENT

MBIZANA (PTY) LTD NINTH RESPONDENT

GEC GAMING (PTY) LTD TENTH RESPONDENT

K2014000230 (PTY) LTD ELEVENTH RESPONDENT

Case no: 1366/2019

K2017440277 (SOUTH AFRICA) (PTY) LTD FIRST APPLICANT

K2017425418 (PTY) LTD SECOND APPLIANT

MBIZANA (PTY) LTD THIRD APPLICANT

GEC GAMING (PTY) LTD FOURTH APPLICANT

K2014000230 (PTY) LTD FIFTH APPLICANT

VUKANI GAMING EASTERN CAPE FIRST RESPONDENT

GAMBLING BOARD SECOND RESPONDENT

AND BETTING BOARD THIRD RESPONDENT

PIONEER SLOTS (PTY) LTD FOURTH RESPONDENT

EASTERN CAPE (PTY) LTD FIFTH RESPONDENT

GSLOTS ISO EC (PTY) LTD SIXTH RESPONDENT

GOLDEN PALACE SITE 3 (PTY) LTD SEVENTH RESPONDENT

GOLDEN PALACE SITE 4 (PTY) LTD EIGHTH RESPONDENT

GOLDEN PALACE SITE 1 (PTY) LTD NINTH RESPONDENT

Case no: 119/2020

GAMBLING BOARD FIRST APPLICANT

AND BETTING BOARD

SECOND APPLICANT

VUKANI GAMING EASTERN CAPE RESPONDENT

Neutral citation: The Chairperson, Eastern Cape Gambling and Betting Board and Another v Vukani Gaming Eastern Cape (Pty) Ltd and Others (1338/2019); (1366/2019); (119/2020) [2021] ZASCA 180 (17 December 2021)

Coram: Zondi, Schippers, and Plasket JJA and Molefe and Unterhalter AJJA

Heard: 18 November 2021

Delivered: This judgment was handed down electronically by circulation to the parties’ representatives by email, publication on the Supreme Court of Appeal website and release to SAFLII. The date and time for hand-down is deemed to be 09h45 on 17 December 2021.

Summary: Review of the issue of a request for proposals (RFP) for independent site operator licences – Gambling licences - interpretation of Regulation 59(3)(a) of the Eastern Cape Gambling Regulations – deference - meaning of a decision-maker being satisfied – over-saturation – province-wide or locality specific – appeal upheld.

ORDER

On appeal from: Eastern Cape Division of the High Court, Grahamstown (Van Zyl DJP, Lowe J and Beyleveld AJ sitting as court of appeal):

1 The applicants are granted special leave to appeal.

2 The appeal is upheld with costs, including the costs of two counsel, where employed.

3 The order of the full court is set aside and replaced by the following order:

‘The appeal is dismissed with costs, including the costs of two counsel, where employed.’

Unterhalter AJA (Zondi, Schippers, and Plasket JJA and Molefe AJA concurring)

Introduction

[1] Three applications for special leave to appeal against the judgment and order of the full court of the Eastern Cape Division of the High Court, Grahamstown were referred to us for oral argument. Vukani Gaming Eastern Cape (Pty) Ltd (Vukani) is one of two licensed route operators of limited pay out machines (LPMs) in the Eastern Cape province. Each of these operators has been allocated 1000 LPMs under their licence. LPMs are akin to slot machines, but they are played for lower stakes. They are not licensed for use in casinos. They are to be found in restaurants, bars and clubs.

[2] The Eastern Cape Gambling and Betting Board (the Board), in June 2015, commissioned a study (the study) to assist it in formulating a policy in respect of the further licensing of LPMs in the province. The study was undertaken and provided to the Board. The study recommended the roll out of 4000 additional LPMs. In May 2017, the Board published its final policy and ultimately resolved to commence with the further licensing of 400 LPMs in the province.

[3] The Board, in September 2017, issued a request for proposals (RFP) calling for applications to be made for 10 independent site operator licences (ISOs), allocated across named district, local and metropolitan municipalities in the Eastern Cape. The Board received bid proposals from several applicants. Hearings were held by the Board, and in February 2018, the Board awarded ISOs. Among those awarded licences were three companies, which I shall refer to as the Golden Palace applicants. The Golden Palace applicants are the applicants before us for special leave to appeal under case no. 1338/19. Five other companies were also successful. I shall refer to them as the Spin and Win applicants. The Spin and Win applicants are the applicants before us for special leave to appeal under case no. 1366/19. The Board is also an applicant for special leave under case no. 119/20. I shall refer to these applicants, collectively, as the applicants.

[4] On 15 November 2017, Vukani brought an application to review and set aside the decision of the Board to issue the RFP. Vukani later extended the relief it sought so as also to set aside the award of ISO licences that followed upon the RFP. On 31 January 2018, Vukani made an application for interim relief to prevent the Board from issuing ISOs, pending the outcome of the review. That application was dismissed by Smith J. The review came before Dawood J. She dismissed Vukani’s review. Vukani was granted leave to appeal to the full court. There the review found favour. The full court found that the Board had failed to comply with the requirements of regulation 59(3) of the Eastern Cape Gambling Regulations (the Regulations), and this rendered the RFP unlawful. If the RFP was unlawful, then so too were the licences issued pursuant to the RFP. The full court, accordingly, reviewed and set aside the RFP and the licences issued by the Board ‘in terms of such RFP’.

[5] The Golden Palace applicants, the Spin and Win applicants and the Board applied for leave to appeal the orders of the full bench to this Court. Their applications were referred to us for oral argument. The parties were directed to be prepared to argue the merits of the appeal if called upon to do so.

The issues

[6] The centrepiece of Vukani’s review is regulation 59(3). The regulations were promulgated under s 80 of the Eastern Cape Gambling and Betting Act 5 of 1997 (the Eastern Cape Act). Regulation 59 reads as follows:

‘. . .

(1) The maximum number of limited gambling machines which may be exposed for play in terms of all route operator licences and limited gambling machine site licences issued in the Province shall be 6000.

(2) Notwithstanding sub-regulation (1), in the first 24 months from the date of first operation of the first limited gambling machine on a licensed gambling machine site in the Province, no more than 2000 limited gambling machines shall be exposed for play in terms of all route operator licences and limited gambling machine site licences issued by the board.

(3) Subject to sub-regulation (2), the board shall only issue or allow route operator licences or limited gambling machine site licences which will allow more than 2000 limited gambling machines to be operated in the Province if–

(a) it is satisfied that this will not lead to an over-saturation of limited gambling machines in the Province; and

(b) it has considered, both in regard to the existing limited gambling machines and such further machines as may exceed 2000–

(i) the social impact;

(ii) the economic impact;

(iii) the environmental impact;

(iv) the impact on problem gambling; and

(v) any other information it considers relevant

and it is of the opinion that the exposure for play of more than 2000 limited gambling machines will be in the best interests of the Province.

(4) No route operator shall be licensed to operate more than 1000 limited gambling machines.

(5) No person shall hold a financial or controlling interest of 5 percent or more in more than one route operator without the consent of the board.

(6) No person may hold more than one route operator licence in the Province.

(7) Apart from the profit sharing between a route operator and site licence holder in terms of the agreement between them approved by the board, no route operator may hold a financial interest in the holder of a gambling machine site licence.’

[7] Two principal issues of interpretation are posed by regulation 59(3) in this appeal. The first issue arises from the requirement in regulation 59(3)(a) that the Board may only act if ‘it is satisfied that this will not lead to an over-saturation of limited gambling machines in the Province’. What does it mean for the Board to be satisfied? And is over-saturation a state of affairs that is appraised on a province-wide basis or does over-saturation require a more granular assessment? I shall refer to this issue as the saturation question.

[8] The second issue is this. When must the requirements of regulation 59(3) be complied with? Is compliance necessary in order to issue a lawful RFP or only at the stage that the Board grants licences. I shall refer to this issue as the timing question.

[9] Vukani’s review challenged the legality of the RFP on the basis that the Board failed to comply with regulation 59(3). That challenge reads regulation 59(3) to be of application at the time that the RFP was issued. The Golden Palace applicants and the Spin and Win applicants submit that this is not so, and hence Vukani’s review must fail if the timing question is answered in favour of these applicants. This follows, because, if regulation 59(3) is not of application to the issuance by the Board of the RFP, then non-compliance with the requirements of regulation 59(3) is of no relevance to the legality of the RFP. The Board agrees with Vukani on the timing question. However, the Board joins the Golden Palace and Spin and Win applicants in their case that, the timing issue aside, the Board did comply with the requirements of regulation 59(3), and hence there is no want of legality on the part of the Board in issuing the RFP. The saturation question should, according to the submission of the applicants, be answered on the basis that regulation 59(3) requires a province-wide assessment. Vukani answers this question differently: over-saturation ‘in the province’ requires the Board to consider whether areas within the province would be over-saturated.

[10] Apart from these two issues of interpretation, Vukani’s review also challenged the RFP on the basis that it was irrational and unlawful because the study upon which it relied was vitiated by error. This ground of review is also pursued before us.

[11] Lastly, the applicants contend that even if we should find that Vukani’s review is well-founded, then there was a wholesale failure by the full court to consider what relief would then have been just and equitable. In essence, the Golden Palace and Spin and Win applicants contend that even if the RFP should be set aside as unlawful, the licences awarded by the Board should not be invalidated. The Board submits that this Court should apply severance to the RFP. Vukani, unsurprisingly, supports the relief granted by the full court.

The interpretation of Regulation 59(3)

[12] I turn then to the questions that arise in this appeal concerning the interpretation of regulation 59(3), and I commence with the saturation question.

[13] Following Natal Joint Municipal Pension Fund v Endumeni Municipality (Endumeni),[1] this Court in Capitec Bank Holdings Limited v Coral Lagoon Investments 194 (Pty) Ltd (Capitec)[2] has explained the approach to be adopted in the interpretation of contracts and statutes:

‘. . . It is the language used, understood in the context in which it is used, and having regard to the purpose of the provision that constitutes the unitary exercise of interpretation. I would only add that the triad of text, context and purpose should not be used in a mechanical fashion. It is the relationship between the words used, the concepts expressed by those words and the place of the contested provision within the scheme of the agreement (or instrument) as a whole that constitutes the enterprise by recourse to which a coherent and salient interpretation is determined. . . .’

[14] Regulation 59 is structured as follows: regulation 59(1) specifies the maximum number of LPMs that may be exposed for play under licence within the province. The maximum number is 6000. Regulation 59(2) places a temporal limitation on the number of LPMs that may be exposed for play within the first 24 months of the first operation of an LPM in the province. That number is 2000. Vukani is one of the two licensees that were licensed to operate LPMs in the Eastern Cape province under the limitation imposed by regulation 59(2).

[15] Regulation 59(3) regulates the basis upon which the Board may license LPMs in excess of 2000 in the province. The power of the Board to license additional LPMs (I will refer to this as the additive power), that is, in addition to the 2000 LPMs permitted under regulation 59(2), requires the Board to comply with the provisions of regulation 59(3)(a) and (b). Regulation 59(3)(a) requires the Board to be satisfied that the additional LPMs will not lead to saturation. Regulation 59(3)(b) requires the Board to consider a list of factors and to form the opinion that the exposure for play of more than 2000 LPMs will be in the best interests of the province.

[16] Regulation 59(3) specifies three conditions that the Board must meet in order to exercise the additive power. Each condition posits a different regulatory burden. Cumulatively, the Board must be satisfied that the additional LPMs will not lead to saturation; it must show it has considered the matters listed in regulation 59(3)(b); and it must be of the opinion that the exposure for play of more than 2000 LPMs will be in the best interests of the province. Precisely how these burdens differ does not arise for decision in this appeal. I observe, however, that the use of different language would ordinarily imply that regulation 59(3) attaches different burdens to the Board in respect of each of the specified conditions.

[17] In this appeal, there is disagreement as to the interpretation of what it means for the Board to be satisfied as required by regulation 59(3)(a). The Board has emphasised what it submits to be the predominantly subjective nature of the regulatory provision and the importance of showing deference to the Board in its assessment of over-saturation. Vukani submits that whether the Board was satisfied, as required by Regulation 59(3)(a), must be objectively tested. The parties cite Walele v City of Cape Town and Others (Walele),[3] but understand its holding in different ways.

[18] Walele in para 60 had this to say of a statutory provision that requires a decision-maker to be satisfied: ‘. . .[t]he decision-maker must now show that the subjective opinion it relied on for exercising power was based on reasonable grounds’.

[19] I make two observations of this formulation. First, the use of the descriptors subjective and objective is not always helpful. To hold a subjective opinion or to be satisfied subjectively is often said to require nothing more than that a person holds the relevant opinion or says that they are satisfied. Theirs is simply a personal belief. So understood, the exercise of a public power, predicated upon being satisfied, would offer vanishingly little constraint. The test is simply whether the person may be believed when saying they are satisfied of the matters that the regulation requires. At the other end of the spectrum, a requirement that must be objectively satisfied may be understood to depend not at all upon whether the judgment of the person who exercises the power considers the requirement to have been met, but whether the requirement has indeed been met. Between these poles, there are gradations that might properly capture the power that has been conferred. The terms subjective and objective may obscure more than they explain. In every case, what matters is the proper interpretation of the power that has been given to the decision-maker.

[20] The second observation is this: conjuring the idea of deference, whether as demon or saviour in public law, may also be apt to mislead. The ubiquity of its invocation is not always commensurate with its utility. What matters is not any a priori position as to whether a deferential construction of a power is warranted. Rather, the question is to determine the nature and scope of the power. Once that is done, the recognition by a court of the power is not a matter of deference or otherwise. The court’s duty is simply to give effect to its best understanding of what it has found the power to be.

[21] Regulation 59(3)(a) stipulates one of the requirements for the exercise by the Board of its additive power to issue or allow licences for LPMs in excess of 2000 machines. Three features of this provision are salient. First, to exercise the power the Board must consider the consequence of the incremental licensing of LPMs, that is, in excess of 2000 LPMs. It is the effect of the increase over 2000 that matters. Second, the relevant effect that the Board must consider is whether the incremental licensing of LPMs leads to an over-saturation of LPMs in the province. Third, the Board’s consideration of these matters must satisfy the Board that the incremental licensing will not lead to an over-saturation of LPMs in the province. I shall refer to this as the over-saturation conclusion.

[22] It is this third dimension of regulation 59(3)(a) that has given rise to some contestation between the parties. It is clear from the language in which regulation 59(3)(a) is cast that the Board must reach the over-saturation conclusion. It will not suffice that the Board merely considers whether the conclusion may be so. The introductory language of regulation 59(3)(b), by contrast, stipulates that the Board has considered five listed issues. However, there is a difference between a requirement that is placed upon a decision-maker to consider something and to be satisfied of something. To be satisfied under regulation 59(3)(a) connotes the stronger requirement that the Board has come to the over-saturation conclusion.

[23] To be satisfied that a particular state of affairs will not come about has two noteworthy features. The first is that it is a judgment about what the future holds, and it is thus a judgment made under conditions of uncertainty Second, the Board must make a judgment that takes the following form: issuing the incremental licences will not lead to over-saturation. Stated formally: doing x will not lead to y. This is a judgment of both probability and causation. What this means is that for the Board to issue or allow incremental licences and comply with regulation 59(3)(a), it must be able to come to the over-saturation conclusion and in fact reach that conclusion. For the Board then to be satisfied, it is required to have reasons that support why it considers that the over-saturation conclusion is warranted. Those reasons do not require the concurrence of a reviewing court. They are simply reasons that could permit a decision-maker, in the position of the Board, to reach the saturation conclusion that it has – in other words, it must have a rational basis for its conclusion.

[24] I turn next to consider what it means for the saturation conclusion to reference an over-saturation ‘in the Province’. The applicants have emphasised that regulation 59 refers repeatedly to the phrase ‘in the Province’. Regulation 59(1) stipulates for the maximum number of LPMs ‘in the Province’. That maximum is clearly determined on a province-wide basis. It matters not at all how the LPMs are distributed within the province. So too, regulation 59(6) states that no person may hold more than one route operator licence in the province. This prohibition is self-evidently of application to the province as a whole, and not to its parts. Why then, so the applicants contended, should the same language have a different meaning in regulation 59(3)?

[25] To this, Vukani offered the following counter: the purpose of regulation 59(3) is to safeguard communities against the adverse effects of gambling. While regulation 59 may specify quantitative maxima that clearly apply to the province as a whole, regulation 59(3) is concerned with the impact of the additional machines upon the social, economic and environmental welfare of the province. This requires that over-saturation is to be judged at local level where an excess of supply of LPMs will impact the welfare of persons. That regulation 59(3) uses the language ‘in the Province’ says nothing as to where the impact of the additional LPMs occurs in making a judgment of over-saturation. After all, impacts at local level remain impacts in the province.

[26] It is important to place regulation 59(3) in the scheme of regulation 59. Plainly, regulation 59(1) stipulates a maximum number of 6000 LPMs that may be exposed for play in the province. This means that for the purpose of determining whether the maximum has been reached it is necessary to count the number of machines exposed for play throughout the province. It is an aggregative determination. Every locality where LPMs are played under licence must be counted, but it is the cumulative number of LPMs in the province that determined whether the maximum number of 6000 has been reached.

[27] The central question is whether the saturation conclusion in regulation 59(3)(a) is also an aggregative judgment? I have already explained that regulation 59(3)(a) concerns the additive power of the Board. It is the consequences of the incremental licensing of machines over 2000 that engage the enquiry posited by regulation 59(3)(a). It is consistent with such an enquiry that the Board may wish to investigate how saturated different local areas of the province might be. It may well find, as occurred in this case, that many local areas are under-saturated, and a few are over-saturated. What conclusion does regulation 59(3)(a) then permit the Board to make?

[28] There is, in my view, an important distinction to be drawn between an aggregative judgment and an allocative decision. Regulation 59(3)(a) is one component part of judgments that must be made to answer the following question: how many LPMs, over and above the 2000 already in operation, should be licensed? A separate question is this. If more than 2000 LPMs may be licensed, where should these machines be licensed for play?

[29] If certain metropolitan areas in the province are over-saturated, but a significant number of other local municipalities are under-saturated, and hence could absorb the exposure for play of additional LPMs (over 2000), does the over-saturation in some areas exclude the Board from reaching the conclusion that additional LPMs would not lead to over-saturation? I think not. It is open to the Board to reach the conclusion that the additional LPMs would not lead to over-saturation because there remains unsatisfied demand for LPMs in the province. The additional machines would assist to bring supply and demand into equilibrium, under some construct of what is socially, economically and environmentally desirable. Put differently, over-saturation in some areas does not mean that the province is over-saturated. That is so because over-saturation is an aggregative concept that is determined across the province. So too, if there was under-saturation in one or two local areas, that would not preclude the judgment that the province was indeed over-saturated.

[30] Aggregative judgments are common in guiding many policies. Whether a country is over-populated, or an economy is over-taxed, or a society is unequal, requires an aggregative judgment to permit of affirmative answers, even though parts of the whole indicate otherwise. So, a country may be considered over-populated even if there are land areas that are sparsely settled. No contradiction arises because the conclusion reached is aggregative.

[31] That the saturation conclusion is aggregative because demand exceeds supply for the province does not answer a different question: where should the additional LPMs be allocated? It may well be, as occurred in this case, that in answering the aggregative question, important evidence is garnered as to where, within the province, the additional LPMs should be licensed. But that goes to a different issue as to how the Board exercises its power to allocate additional licences. That may even give rise to a ground of challenge if the allocation was irrational. But regulation 59(3)(a) is not concerned with the allocation of licences in different areas within the province. It is concerned to determine the number of additional LPMs over 2000 that the Board is empowered to issue or allow in the province. That is a particular number based on an aggregative judgment as to whether that number of additional LPMs will or will not lead to over-saturation in the province, taken as a whole.

[32] Vukani’s challenge to the RFP rested upon the fact that the study had found that at least two municipalities in the province were over-saturated: Nelson Mandela Bay Metropolitan Municipality (NMB) and Sarah Baartman District Municipality (SBM). The Board accepted the study for the purpose of formulating its policy and ultimately issuing the RFP. The RFP, so Vukani contended, was consequently not lawful because the study’s findings as to the over-saturation of NMB and SBM meant that the Board could not have been satisfied, as required by regulation 59(3)(a), that the 400 additional LPMs would not lead to over-saturation in the province. In sum, over-saturation in NMB and SBM precluded the Board from reaching the saturation conclusion in compliance with regulation 59(3)(a).

[33] However, once, as I find, regulation 59(3)(a) required that the Board be satisfied that the 400 additional LPMs it proposed be made available for licence would not lead to over-saturation on a province-wide basis, then the Board, in publishing the RFP, did not fail to comply with regulation 59(3)(a). The study, fairly read, concluded on various measures that the Eastern Cape province ‘is not yet relatively over saturated with LPMs’. Consequently, the study concluded with the recommendation that the Board continue to roll out all 6000 LPMs that had been allocated to the province. The Board accepted the study and reached its saturation conclusion on a province-wide consideration of over-saturation. In doing so, the Board did not fail to comply with the requirements of regulation 59(3)(a).

[34] The RFP did invite applications for ten licences, distributed between various municipalities, and detailed in two tables. Those tables allocated sites in NMB and SBM. This aspect of the RFP concerns the allocation of licences to various municipalities. Vukani most certainly questioned how the Board could accept the study’s findings of over-saturation in NMB and SBM, and yet allocate licences to these municipalities. The difficulty for Vukani, however, is this. Its review is not predicated on the allocative decision of the Board, but rather that the over-saturation findings in respect of NMB and SBM meant that the requirements of regulation 59(3)(a) were not be met. That challenge cannot be sustained because it rests upon an interpretation of regulation 59(3)(a) that is incorrect. Accordingly, Vukani’s review, on this ground, must fail, and the full court was in error to hold otherwise.

[35] This finding renders it unnecessary to determine the timing question as to when the Board must be satisfied as to the saturation conclusion. Even if Vukani and the Board are correct that the requirements of regulation 59(3) must be met in order to issue the RFP (I make no finding on this matter), since Vukani’s review of the RFP for non-compliance with regulation 59(3)(a) cannot prevail, there is no need for this Court to decide the timing question that was raised by the Golden Palace and the Spin and Win applicants.

The flawed study

[36] It is common ground that the Board relied upon the study to formulate its policy that then resulted in the publication of the RFP. Vukani contended that the study is unlawful and irrational. The study is so defective, Vukani submits, that it cannot support any conclusion as to whether the province is under-or over-saturated. Nor does the study, on its own terms, consider the social impacts as required by regulation 59(3)(b), and hence the study was not a rational basis for the Board to conclude that the licensing of additional LPMs would be in the best interests of the Eastern Cape. In sum, Vukani submits that the study was irremediably flawed, and can provide no rational basis for the publication of the RFP, hence the RFP must be reviewed and set aside.

[37] Vukani procured the services of an expert, Professor Standish, to assess the study. Professor Standish identified numerous errors in the study. These errors may be summarised as follows. First, the study failed to define over saturation, rather, it considered divergence from existing average distributions of LPMs. Second, the metrics used to measure over-saturation were the ratios of LPMs to the population density and gross domestic product (GDP) of municipalities. These metrics are irrational. Third, the study failed to apply the metrics it chose in a consistent fashion. Fourth, the study failed to calculate the ratios it relied upon correctly. It equated site licences with the number of LPMs. But a site may have many LPMs. The calculated ratios do not reflect the actual distribution of LPMs in the province. Fifth, the study made numerous errors as to how to measure the impact of LPMs. Sixth, the study acknowledged that it failed to determine the negative social, environmental and economic impacts of the licensing of additional LPMs.

[38] Vukani contends that the study is so flawed that it cannot provide a rational basis for the publication of the RFP. A review, so framed, cannot prevail simply because the study made numerous errors. Nor does it suffice to show that, on Professor Standish’s analysis, the study was incorrect. What is required is a showing that the central conclusions of the study are supported by no evidence or reasoning that could sustain these conclusions, and hence the study is irrational.

[39] A reading of the study, alongside Professor Standish’s criticisms, does not yield such a showing. The study investigated the degree of over or under saturation of LPMs in the Eastern Cape in two ways. First, the study sought to measure LPM sites relative to GDP and population in each of 37 district municipalities and two metropolitan municipalities. Second, the study undertook a comparative analysis of turnover for LPMs and the per capita tax contribution per province for LPMs to compare the position of the Eastern Cape with other provinces.

[40] True enough, the study does not offer a definition of under or over saturation. But the methodologies adopted by the study do permit of an understanding as to when under or over saturation would be present. Under or over saturation is a relative concept judged empirically against various criteria and the existing distributions of LPMs, as between provinces and within the Eastern Cape. A derived definition of this kind, relying upon existing distributions, may be simplistic, but it is not irrational.

[41] So too the use of population and GDP ratios assumes a normative constant as between the number of LPMs and the population or GDP of each municipality. This is a simplifying assumption. Others could have been made. But it cannot be said that its application can yield no sense whatever as to how, across the province, supply and demand may be assessed. The number of site licences in relation to the population or GDP of a municipality may be a basic comparative measure of saturation, but it cannot be said to yield nothing of value.

[42] Professor Standish pointed out other errors: the inconsistent application of the study’s models; its equation of site licences and LPMs; and issues the study neglected to examine, such as machine turnover. Problems with the study, no doubt, abound, but what is required is not an identification of errors but a study that provides no evidence or analysis that would permit of any conclusions as to over-saturation. A fair reading of the study does not meet that high threshold. For example, the study compares gambling tax revenue and turnover for LPMs across provinces. It is difficult to say of this data that it is valueless in making any judgment as to saturation.

[43] Finally, the acknowledgment in the study that the social ills of gambling have not yet been fully determined for South Africa or the Eastern Cape does not mean that the study failed to marshal evidence and offer analysis of the social, economic and environmental impacts of the additional LPMs, as required by regulation 59(3)(b). On the contrary and by way of example, the study considered the investment that would be attracted to the Eastern Cape and its multiplier effect; a questionnaire was administered to 298 punters across 35 LPMs in the province to ascertain the socio-economic impacts of the LPM industry upon punters; questionnaires were also administered to community members and site operators for the same purpose; and findings were made as to problem gambling. The study cannot be read to have either ignored social, economic and environmental impacts or to have so glossed over these matters as to leave no foundation for the Board to take up a position on these matters.

[44] In my view, therefore, whatever the limitations of the study identified by Professor Standish, the study is not vitiated by error to the point that the Board’s reliance upon it tainted the RFP with irrationality. Vukani’s review on this ground must also fail.

The remaining issues

[45] The Board pressed us to make a finding that the issue of the RFP is not administrative action reviewable under the PAJA, but is rather action reviewable under the principle of legality. The full court did not determine this issue. I can see no reason why this Court should do so, and nothing in City of Tshwane Metropolitan Municipality and Others v Nambiti Technologies (Pty) Ltd[4] compels such a finding. The grounds of review advanced by Vukani are not altered as to their substance whether the issue of the RFP is administrative action reviewable under the PAJA or whether the RFP is reviewable under the principle of legality. Nothing in this appeal turns on the matter, and I see no reason to resolve an issue that is not required to decide this appeal.

[46] Vukani opposed the Spin and Win applicants’ application for leave to appeal on the basis that these applicants had taken no part in the proceedings before the full court. I do not understand that objection to be persisted in by Vukani. Nor should it be. The Spin and Win applicants are parties to these proceedings and their election to rely on the submissions of the Golden Palace applicants at the full court’s hearing does not preclude them from pursuing an application for leave to appeal.

Conclusion

[47] The Board, the Golden Palace applicants and the Spin and Win applicants have made out their case for special leave to appeal. As this judgment makes plain, their applications enjoyed strong prospects of success and raised a substantial question of law as to how regulation 59(3)(a) was to be interpreted.

[48] As to the merits of the appeal, I find that the appeal must be upheld and Vukani’s review of the RFP and the issue of licences to which it gave rise must be dismissed. The costs follow the result, including the costs of two counsel, where so employed.

[49] The following order is made:

1 The applicants are granted special leave to appeal;

2 The appeal is upheld with costs, including the costs of two counsel, where employed;

3 The order of the full court is set aside and replaced with an order as follows:

‘The appeal is dismissed with costs, including the costs of two counsel, where employed’.

DAVID UNTERHALTER

ACTING JUDGE OF APPEAL

Appearances

For applicants under case number: Ismail Jamie SC (with Him Mitchell De Beer)

119/2020

Instructed by: Tshangana Le Roux Inc

c/o Honey Attorneys, Bloemfontein

For applicants under case number: Sarah Pudifin-Jones

1338/2019 and 1366/2019

Instructed by:

Woodhead Bigby Attorneys, Umhlanga

c/o Lovius Block Inc, Bloemfontein

For respondents: Steven Budlender SC (with him Michael Bishop and Eshed Cohen)

Instructed by: Edward Nathan Sonnenbergs

c/o Tshangana Attorneys, Bloemfontein

[1] Natal Joint Municipal Pension Fund v Endumeni Municipality [2012] ZASCA 13; [2012] 2 All SA 262 (SCA); 2012 (4) SA 593 (SCA) para 18.

[2] Capitec Bank Holdings Limited v Coral Lagoon Investments 194 (Pty) Ltd [2021] ZASCA 99; [2021] 3 All SA 647 (SCA) para 25

[3] Walele v City of Cape Town and Others [2008] ZACC 11; 2008 (6) SA 129 (CC) para 60

[4] City of Tshwane Metropolitan Municipality and Others v Nambiti Technologies (Pty) Ltd [2015] ZASCA 167; [2016] 1 All SA 332 (SCA); 2016 (2) SA 494 (SCA) paras 24–26.