Opposition to Urban Tolling Alliance and Others v The South African National Roads Agency Ltd and Others (90/2013) [2013] ZASCA 148 (9 October 2013)




THE SUPREME COURT OF APPEAL OF SOUTH AFRICA

JUDGMENT



REPORTABLE

Case No: 90/2013


In the matter between:



OPPOSITION TO URBAN TOLLING ALLIANCE FIRST APPELLANT


SOUTH AFRICAN VEHICLE RENTING AND

LEASING ASSOCIATION SECOND APPELLANT


QUADPARA ASSOCIATION OF SOUTH AFRICA THIRD APPELLANT


SOUTH AFRICAN NATIONAL CONSUMER UNION FOURTH APPELLANT


and


THE SOUTH AFRICAN NATIONAL ROADS

AGENCY LIMITED FIRST RESPONDENT


THE MINISTER, DEPARTMENT OF TRANSPORT

REPUBLIC OF SOUTH AFRICA SECOND RESPONDENT


THE MEC, DEPARTMENT OF ROADS AND

TRANSPORT, GAUTENG THIRD RESPONDENT


THE MEC, DEPARTMENT OF WATER AND

ENVIRONMENTAL AFFAIRS FOURTH RESPONDENT


THE DIRECTOR-GENERAL, DEPARTMENT OF

WATER AND ENVIRONMENTAL AFFAIRS FIFTH RESPONDENT


NATIONAL CONSUMER COMMISSION SIXTH RESPONDENT


NATIONAL TREASURY SEVENTH RESPONDENT



Neutral citation: Opposition to Urban Tolling Alliance v The South African National Roads Agency Limited (90/2013) [2013] ZASCA 148 (9 October 2013).

Coram: Brand, Nugent, Petse JJA, Van der Merwe and Swain AJJA

Heard: 25 September 2013

Delivered: 9 October 2013


Summary: Administrative review – declaration of toll roads in terms of s 27 of Act 7 of 1998 – 180 day time limit contemplated in s 7(1) of Promotion of Administrative Justice Act 3 of 2000 – extension of time limit pursuant to s 9(2) considered.

___________________________________________________________________


ORDER

___________________________________________________________________


On appeal from: North Gauteng High Court, Pretoria (Vorster AJ sitting as court of first instance):

The appeal is refused with no order as to costs, save that the order granted by the court a quo, directing the appellants to pay the respondents’ costs, is set aside and replaced by an order that there be no order as to costs.

___________________________________________________________________


JUDGMENT

___________________________________________________________________


BRAND JA (NUGENT, PETSE JJA, VAN DER MERWE AND SWAIN AJJA concurring):


[1] This is the tale of seven toll roads around two cities in the province of Gauteng. Since the proposed method of toll collection is through electronic operation, the matter became dubbed by the media and in popular parlance as the ‘e-tolling case’. The seven roads involved constitute the main arteries around Johannesburg and Pretoria, which in turn form the commercial hub of South Africa. These roads form part of a larger project for the infra-structural upgrading of Gauteng roads that has become known as the Gauteng Freeway Improvement Project or by the acronym GFIP. The declarations of the roads as toll roads gave rise to unprecedented public and political debate. These declarations were made by the South African Road Agency Limited (SANRAL) through publication in the Government Gazette following upon approval of their decision to do so by the Minister of Transport (the Minister) in accordance with the procedure contemplated in s 27 of the South African National Road Agency Act 7 of 1998 (the Act). Six of these declarations took effect by way of publications in the Government Gazette of 28 March 2008. The seventh declaration, pertaining to the R21 road, was published in the Government Gazette of 28 July 2008. The reason for the different treatment of the R21 was that it first had to be transferred from the Gauteng Provincial Government to the remit of SANRAL before it could be declared a toll road under s 27 of the Act. But the difference in the dates of publication is of no real consequence in this matter. Consequently I shall henceforth draw no distinction between the R21 and the other six roads.


[2] Court proceedings started four years after the declaration of the roads as toll roads, when the appellants launched an application against the respondents in the North Gauteng High Court, Pretoria on 23 March 2012. The application comprised of two parts. The first part was for an urgent pendente lite interdict precluding SANRAL from levying and collecting tolls on the seven roads pending the final determination of the application in the second part. The application in the second part was a review application under Court Rule 53 for the setting aside of the decisions by SANRAL and the Minister which gave rise to the declarations of the roads as toll roads in 2008.


[3] The application in the first part was heard by Prinsloo J as a matter of urgency and on Saturday 28 April 2012 he granted the pendente lite interdict sought. A direct appeal by the respondents to the Constitutional Court was, however, successful. In consequence, the interim interdict was set aside while the costs in those proceedings were ordered to be part of the review application. The judgment of the Constitutional Court has since been reported as National Treasury and others v Opposition to Urban Tolling Alliance and others 2012 (6) SA 223 (CC). In due course the review application in the second part came before Vorster AJ. In the event he dismissed the application with costs in favour of the respondents, including the costs reserved by the Constitutional Court. The present appeal against those orders is with the leave of the court a quo.


[4] The issues arising in the appeal will be best understood against the background of s 27 of the Act and the underlying facts. The relevant part of s 27 provides:

‘(1) Subject to the provisions of this section, the Agency [ie SANRAL] –

(a) with the Minister’s approval –

(i) may declare any specified national road or any specified portion thereof, including any bridge or tunnel on a national road, to be a toll road for the purposes of this Act; and

(ii) may amend or withdraw any declaration so made;

. . .

(2) A declaration, amendment, withdrawal, . . . under subsection (1), will become effective only 14 days after a notice to that effect by the Agency has been published in the Gazette.

(3) The amount of toll that may be levied under subsection (1), any rebate thereon and any increase or reduction thereof –

(a) is determined by the Minister on the recommendation of the Agency;

(b) . . .

(c) must be made known by the head of the Department by notice in the Gazette;

(d) . . .

(4) The Minister will not give approval for the declaration of a toll road under subsection (1)(a), unless –

(a) the Agency, in the prescribed manner, has given notice, generally, of the proposed declaration, and in the notice –

(i) has given an indication of the approximate position of the toll plaza contemplated for the proposed toll road;

(ii) has invited interested persons to comment and make representations on the proposed declaration and the position of the toll plaza, and has directed them to furnish their written comments and representations to the Agency not later than the date mentioned in the notice. However, a period of at least 30 days must be allowed for that purpose;

(b) the Agency in writing –

(i) has requested the Premier in whose province the road proposed as a toll road is situated, to comment on the proposed declaration . . . within a specified period (which may not be shorter than 60 days); and

(ii) has given every municipality in whose area of jurisdiction that road is situated the same opportunity to so comment;

. . . ‘


[5] As to the background facts, I find it convenient to start the narrative with an introduction of the parties. The first appellant is the Opposition to Urban Tolling Alliance (OUTA). It is a voluntary association which was established on 12 March 2012 for the sole purpose, so it said, of providing a platform for individuals and other entities who seek to prevent the e-tolling by SANRAL on the seven toll roads. The members of OUTA include the second appellant, which is the South African Vehicle and Leasing Association (SAVRALA) and the Automobile Association of South Africa, together with 94 other businesses and 1831 individual supporters. The second appellant, SAVRALA, in turn represents 22 member companies including multinational car hire firms such as Avis and Europcar. The third appellant, Quadpara Association of South Africa, is an organisation that protects and promotes the interests of people with disabilities, while the fourth appellant, the South African National Consumer Union, promotes the rights of consumers.


[6] The appeal is opposed by the first, second, third and seventh respondents. The first respondent is SANRAL. The second and third respondents – who are represented by the same counsel and attorneys – are the Minister and the MEC for Roads and Transport in the Gauteng Provincial Government. The seventh respondent is the National Treasury. SANRAL, which by the nature of things took centre stage in the proceedings, is a creature of statute. It was created by s 2 of the Act. It also derives its powers from the Act. In terms of s 25(1) its main functions are described thus:

‘The agency [SANRAL], within the framework of government policy, is responsible for and is hereby given power to perform all strategic planning with regard to the South African national roads system as well as the planning, design, construction, operation, management, control, maintenance and rehabilitation of national roads for the Republic and it is responsible for the financing of all those functions in accordance with its business and financial plan, so as to ensure that Government’s goals and policy objectives concerning national roads are achieved . . .’


[7] The financing options that are available to SANRAL are set out in s 34(1) of the Act. Though the section enumerates an impressive list of twelve options, it is not in dispute that for present purposes these were limited to three, namely levies raised on the sale of fuel (subsection (1)(b)); tolls raised on toll roads (subsection (1)(g)); and monies appropriated by Parliament (subsection (1)(k)). It is common cause that the seven toll roads fall under the control of SANRAL. Likewise it is common cause that the development and improvement of these roads, undertaken by SANRAL as part of the GFIP, were necessary to alleviate congestion on the roads of Gauteng and to facilitate economic growth not only in that province, but in the country as a whole. The debate between the appellants, on the one hand, and the respondents on the other, was therefore not whether it was prudent to undertake the GFIP, but focussed [] on how these improvements were to be funded.


[8] The papers reveal that the funding decision was a complex one. The starting-point of the GFIP was a report, dated September 2006, which resulted from a joint initiative by the various authorities involved, including SANRAL, the Department of Transport and the Gauteng Provincial Government. But it is clear that this report had its origin in a much earlier White Paper on national transport policy, dating back to 1996. The White Paper recognised that the South African transportation system was inadequate to meet the basic accessibility needs to work, healthcare, schools and so forth in rural areas and that these needs were to be addressed in an accelerated manner. Because the rural communities involved could not pay for it, resources made available by Parliament were to be allocated to these areas. This meant that in other areas, where economically feasible, the principle of users pay through tolling was to be regarded as the funding method of preference.


[9] The 2006 proposal was further developed and in July 2007 the National Department of Transport submitted the GFIP as a proposal to Cabinet. The memorandum to Cabinet was accompanied by a series of documents and a formal slide presentation. The reports and presentation show a clear appreciation that solutions proffered should adhere to government policies, as contemplated in s 25 of the Act. In this light the proposal to Cabinet identified the most suitable funding mechanism for the GFIP as a toll scheme with electronic fare collection as a basis to ensure free traffic flow. In motivating electronic toll collection it was explained that the density of traffic on the road network involved was such that a conventional toll collection system through toll gates was simply not practically possible. The proposed toll collection system would operate through overhead gantries, fitted with toll collection equipment that would identify vehicles passing under the gantry by electronic transponders (e-tags) fitted to the vehicle or by its number plates. Since no physical toll booths were involved, the electronic toll collection system would not impact on traffic flow at all. The proposal indicated an appreciation that this method of collection was sophisticated and expensive. Viability studies by independent consultants retained by SANRAL, however, showed that despite this expensive form of collection, e-toll funding was a viable option for the GFIP. In the event the proposal was approved by Cabinet.


[10] On 8 October 2007, the then Minister of Transport officially announced the launch of the GFIP, which was to be implemented in accordance with the proposal approved by Cabinet. At this launch, the general media (print, radio and TV) was present. Apart from the Minister’s keynote address, there were several other presentations, including one by SANRAL. Copies of the keynote address and the presentations were made available to the media. SANRAL’s presentation, inter alia, referred to an estimated tariff of between 50 cents and 30 cents per kilometre. Following upon the presentation, coverage of the freeway tolling concept, the implementation of the project and the expected toll tariff occurred in the printed media, radio and television.


[11] Some days later SANRAL published its notice of intent to toll the relevant roads in the Government Gazette, together with diagrams of the relevant road sections. At the same time the notices and diagrams were also published in a single edition of about six newspapers circulating in Gauteng. These notices invited comments from the general public by 14 November 2007 – approximately one month from the date of publication of the notices – which was in accordance with the minimum period of 30 days stipulated in s 27(4)(a)(ii) of the Act. Letters detailing the same information as the notices were also sent to the Premier and the MEC for Transport of Gauteng, as well as to the various municipalities affected by the proposed declaration. The closing date for comments by the authorities was 14 December 2007, which was in accordance with the minimum period of 60 days stipulated in s 27(4)(b)(i) of the Act.


[12] On 15 January 2008 SANRAL applied to the Minister to approve the declaration of the seven roads involved as toll roads. In accordance with s 27(4)(c) of the SANRAL Act, this application was accompanied by a report from SANRAL in which it summarised the main issues raised in the representations received from the public as well as SANRAL’s detailed responses to these issues. On 11 February 2008 the Minister granted his approval[]. Following upon this approval, SANRAL, as I have said by way of introduction, declared six of these roads as toll roads by publication in the Government Gazette of 28 March 2008. Subsequent to the declaration, SANRAL continued its interaction with interested representatives of civil society by delivering a number of presentations over an extended period. Notes taken at one of these presentations on 7 July 2008 indicate that representatives of both the second appellant, SAVRALA and the AA, which are both members of OUTA, were present. According to these notes Mr Nazir Alli, the chief executive officer of SANRAL, responded to a question by a SAVRALA representative as to the anticipated toll tariff that ‘we are looking at 50 cents per kilometre but this will be discounted’.


[13] On 9 May 2008 SANRAL issued a media release to the effect that it had awarded seven contracts for the implementation of the GFIP. On 24 June 2008 work commenced in earnest on the project and continued for the next two years in order to prepare certain sections of the proposed toll road network for the Fifa 2010 World Cup. After a three month period of inactivity during the World Cup, work on the freeway system recommenced and continued into 2011. It involved a massive infra-structural development, including the construction of bridges, flyovers, on- and off-ramps, and related services. It also included the construction of 42 overhead gantries that became a feature of the Gauteng landscape in the period following the World Cup in 2010. The costs incurred by SANRAL to finance this construction exceeded R20 billion. SANRAL procured this funding from the money market by issuing bonds which are effectively repayable loans, repayment of which was guaranteed by the South African government through the Treasury. Should SANRAL fail to collect tolls, so we are told, it will not be able to meet its obligations under the loan. In practice this will mean that the guarantee by the government stands to be called up.


[14] On 11 February 2011 the Director-General for Transport published the toll tariffs for the toll network in terms of s 27(3)(c) of the SANRAL Act. At that stage tolling was intended to commence on 23 June 2011. The publication gave rise to a massive public outcry. In response, the Department of Transport suspended the implementation of tolling while the Minister announced that a steering committee would be formed to address the public’s concerns. The mandate of the steering committee was to review the amount of the toll tariffs. It was not to revisit the mechanism of tolling itself. Yet the appellants point out that numerous parties, including the appellants themselves, made representations to the steering committee that tolling itself should be discontinued, in the belief, so they said, that SANRAL and the Minister would reconsider and withdraw the implementation of the tolling system as a whole. This belief, so they explained, was fuelled by the widespread and unparalleled public opposition to tolling that even crossed political dividing lines.


[15] The steering committee held public hearings on several days. At every one of these hearings the committee made it clear that the principle of ‘user pays’ and the tolling of the proposed freeway network had been accepted in principle and that the discussions before the committee would be limited to the proposed tariff only. In the event, the committee recommended to government that the proposed tariff be reduced from about 66c to about 30c per kilometre, with a monthly cap of R550 for e - tag users. This recommendation was accepted by Cabinet. On 23 October 2011, after further public outcry at the news that tolling was set to proceed on the reduced tariffs, the Minister of Transport instructed SANRAL to halt the tolling process.


[16] According to the appellants, their hope that SANRAL and the Minister would seriously reconsider the raising of funds for the GFIP through tolling was, however, finally dashed by the budget speech of the Minister of Finance on 22 February 2012. What was announced in that speech was a special appropriation of R5.8 billion by Parliament to SANRAL, which would facilitate the reduction in the toll tariffs recommended by the steering committee. But when the appellants’ application was launched in the court a quo in March 2012, the new toll tariff had not as yet been published in terms of s 27(3)(c) of the Act, after the withdrawal of the one published on 11 February 2011. The new toll tariff was only published in the Government Gazette of 13 April 2012. For some reason not explained on the papers, this notice was also withdrawn on 31 May 2012. Rather obviously, in the circumstances, the appellants’ case was never aimed at the review of toll tariffs – which had not yet been finally determined. Instead, the sole focus of their review application was the decisions that gave rise to the declaration of the seven toll roads under s 27(1) of the Act.


[17] The judgement of the Constitutional Court on appeal against the interim interdict had a pronounced effect on the central theme of the appellants’ challenge. This occurred despite the Constitutional Court’s caution that it did not propose to influence the outcome of the review application (see eg para 21 of the judgment). As appears from what I have said by way of introduction, the main focus of the debate on the papers was directed at SANRAL’s election of e-tolling as a method of funding the GFIP. More particularly, the main thrust of the appellant’s case was that SANRAL should have adopted a method of funding other than toll. Although the appellants disavowed any intent to dictate a particular method of funding to the authorities, they clearly proposed a ring-fenced fuel levy as the best option. As to this method of funding, they inter alia said:

‘An alternative method of funding which is favoured by many interested parties (including the appellants) is a ring-fenced fuel levy. This option entails no costs of collection at all. When this is considered, it becomes clear that the option of open road tolling is so unreasonable that it is not a decision that could have been taken by a reasonable administrator.’


[18] In further support of their funding method of choice the appellants contended that the choice of open road tolling was unreasonable within the meaning of s 6(2)(h) of the Promotion of Administrative Justice Act 3 of 2000 (PAJA) in that the operation costs of e-tolling over the anticipated period of 20 years will exceed the capital costs of the GFIP and because the enforcement of the system will be practically impossible. As to why this would be so, the appellants predicted that there will be 800 000 users of the proposed toll roads every day, of which only about 60 per cent will be voluntarily registered for e-tag and that, for the rest, there would be a high percentage of defaulters. This, they forebode, will require SANRAL to issue up to 1 000 summonses per day; which is simply not practically feasible.


[19] But the judgment of the Constitutional Court effectively derailed the challenge based on the thesis that SANRAL and the Minister opted for the wrong method of funding the GFIP. Simply put, the Constitutional Court held that this option was a policy decision with which the courts will not interfere. The court’s reasoning appears, for instance, from the following statements in the main judgment by Moseneke DCJ (at paras 34-35):

‘OUTA points out, correctly in my view, that it does not seek to set aside the cabinet’s approval of the GFIP in as much as it was not granted in terms of any specific statute. Also, the approval does not amount to administrative action that is susceptible to review under PAJA. . . . Outa must be supported where it submits that the specific decisions that are impugned were not made by the cabinet or the National Treasury. However, it is quite another matter to suggest that the impugned decisions of the Transport Minister and SANRAL had nothing to do with the executive government’s policy, including the policy that users of the upgraded roads are the ones who must pay. Or with the National Treasury’s domestic budgetary responsibilities and its sovereign-debt policy.’

And (at para 67):

‘Thus, the duty of determining how public resources are to be drawn upon and reordered lies in the heartland of executive-government function and domain. What is more, absent any proof of unlawfulness or fraud or corruption, the power and the prerogative to formulate and implement policy on how to finance public projects reside in the exclusive domain of the national executive subject to budgetary appropriations by parliament.’


[20] Similar considerations were expressed by Froneman J in his separate concurring judgment when he said (at para 93):

‘It is undisputed that in July 2007 the cabinet approved the Gauteng Freeway Improvement Project and the concomitant basis for its funding, e-tolling, after extensive investigation and a report to it on the issue. It is national executive and treasury policy not to use fuel-levy-type funding for these kinds of projects. None of this was, or could be, attacked on review in this court. The playing field for the contestation of executive-government policy is the political process, not the judicial one.’

And (at para 94):

‘The main thrust of the respondents’ [these being the appellants in the present matter] review is the alleged unreasonableness of the decision to proclaim the toll roads. But unreasonable compared to what? The premise of their unreasonableness argument is that funding by way of tolling is unreasonable because there are better funding alternatives available, particularly fuel levies. But that premise is fatally flawed. The South African National Road Agency Ltd has to make its decision within the framework of government policy. That policy excludes funding alternatives other than tolling. It is unchallenged on review.’


[21] In this light the appellants were compelled to shift the main focus of their challenge to one of procedural unfairness. More particularly, to the charge that on a proper interpretation of s 27(4) of the Act and s 4 of PAJA, SANRAL and the Minister had failed to comply with the notice and comment procedure prescribed by these legislative provisions, albeit that on the face of it the notices might have satisfied the minimum requirements of s 27(4) of the Act. A second objection to the impugned decisions, which was raised by the appellants only after the decision of the Constitutional Court, was that the tolling of the GFIP amounted to an unlawful deprivation of rights in property – ie their money – and that it therefore fell foul of s 25 of the Constitution. In addition to these two main grounds, the appellants advanced three grounds in this court on which the decision of the court a quo should be reversed. These were:

(a) The alleged unreasonableness of the decision to impose e-tolling because of its excessive operational costs and the practical impossibility of its enforcement.

(b) The alleged failure by the Minister of Transport to consider the costs of toll collection.

(c) The alleged incorrectness of the 2006 estimate of the toll collection that formed the basis of the impugned decision by the Minister.


[22] Apart from contesting the appellants’ challenge to the impugned decisions on its merits, the respondents relied on what has become known as the delay rule. Despite the appellants’ contentions to the contrary and for reasons that will become apparent soon, I believe we are compelled to follow the example set by this court in Beweging vir Christelik Volkseie Onderwys and others v Minister of Education and others [2012] 2 All SA 462 (SCA) para 44, by dealing with the delay rule first.


[23] Although the delay rule has its origin in common law, it now finds its basis in s 7(1) of PAJA which provides in relevant part:

‘1. Any proceedings for judicial review in terms of section 6(1) must be instituted without unreasonable delay and not later than 180 days after the date –

(a) . . .

(b) . . . on which the person concerned was informed of the administrative action, became aware of the action and the reasons for it or might reasonably have been expected to have become aware of the action and the reasons.’


[24] Section 9(1) provides, however, that the 180-day period ‘may be extended for a fixed period, by agreement between the parties or, failing such agreement, by a court or tribunal, on application by the person or administrator concerned’. Section 9(2) provides that such an application may be granted ‘where the interests of justice so require’.


[25] As to the purpose and function of the delay rule under s 7(1) of PAJA and its common law predecessor, Nugent JA explained in Gqwetha v Transkei Development Corporation Ltd and others 2006 (2) SA 603 (SCA) paras 22-23:

‘[22] It is important for the efficient functioning of public bodies . . . that a challenge to the validity of their decisions by proceedings for judicial review should be initiated without undue delay. The rationale for that longstanding rule . . . is twofold: First, the failure to bring a review within a reasonable time may cause prejudice to the respondent. Secondly, and in my view more importantly, there is a public interest element in the finality of administrative decisions and the exercise of administrative functions. As pointed out by Miller JA in Wolgroeiers Afslaers (Edms) Bpk v Munisipaliteit van Kaapstad 1978 (1) SA 13 (A) at 41E-F (my translation):

“It is desirable and important that finality should be arrived at within a reasonable time in relation to judicial and administrative decisions or acts. It can be contrary to the administration of justice and the public interest to allow such decisions or acts to be set aside after an unreasonably long period of time has elapsed - interest reipublicae ut sit finis litium. . . . Considerations of this kind undoubtedly constitute part of the underlying reasons for the existence of this rule.”

[23] Underlying that latter aspect of the rationale is the inherent potential for prejudice, both to the efficient functioning of the public body and to those who rely upon its decisions, if the validity of its decisions remains uncertain. It is for that reason in particular that proof of actual prejudice to the respondent is not a precondition for refusing to entertain review proceedings by reason of undue delay, although the extent to which prejudice has been shown is a relevant consideration that might even be decisive where the delay has been relatively slight (Wolgroeiers Afslaers, above, at 42C).’


[26] At common law application of the undue delay rule required a two stage enquiry. First, whether there was an unreasonable delay and, second, if so, whether the delay should in all the circumstances be condoned (see eg Associated Institutions Pension Fund and others v Van Zyl and others 2005 (2) SA 302 (SCA) para 47). Up to a point, I think, s 7(1) of PAJA requires the same two stage approach. The difference lies, as I see it, in the legislature’s determination of a delay exceeding 180 days as per se unreasonable. Before the effluxion of 180 days, the first enquiry in applying s 7(1) is still whether the delay (if any) was unreasonable. But after the 180 day period the issue of unreasonableness is pre-determined by the legislature; it is unreasonable per se. It follows that the court is only empowered to entertain the review application if the interest of justice dictates an extension in terms of s 9. Absent such extension the court has no authority to entertain the review application at all. Whether or not the decision was unlawful no longer matters. The decision has been ‘validated’ by the delay (see eg Associated Institutions Pension Fund para 46). That of course does not mean that, after the 180 day period, an enquiry into the reasonableness of the applicant’s conduct becomes entirely irrelevant. Whether or not the delay was unreasonable and, if so, the extent of that unreasonableness is still a factor to be taken into account in determining whether an extension should be granted or not (see eg Camps Bay Ratepayers’ and Residents’ Association v Harrison [2010] 2 All SA 519 (SCA) para 54).


[27] In its terms s 7(1) envisages asking when ‘the person concerned’ was informed, or became aware, or might reasonably be expected to have become aware, of the administrative action. This admits of an answer where the act affects and is challenged by an individual, but does not readily admit of an answer where it affects the public at large. In that situation it would be anomalous – if not absurd – if an administrative act were to be reviewable at the instance of one member of the public, and not at the instance of another, depending upon the peculiar knowledge of each. It seems to me that in those circumstances a court must take a broad view of when the public at large might reasonably be expected to have had knowledge of the action, not dictated by the knowledge, or lack of it, of the particular member or members of the public who have chosen to challenge the act.


[28] On the facts of this case there is no dispute that the protagonists knew of the decision by no later than 11 February 2011 – the date of publication of the toll tariffs, which was the catalyst for the public outcry – but in truth there was, or might reasonably have been, public knowledge of the decision far earlier than that. I have already said that as early as 8 October 2007 the then Minister of Transport officially announced the launch of the GFIP, followed shortly by presentations that received coverage in the media, and by notification to state institutions representative of the public, and then by the construction of tolling gantries visible to the motoring public. It is not necessary in this case to attempt any precise identification of the date the 180 day clock started ticking. It is sufficient to say that by the time the application was launched on 23 March 2012 the public at large might reasonably be expected to have been aware of the decision to toll for at least some two to three years, which is well outside the 180 period provided for in s 7(1). In consequence, the appellants rightly believed they were compelled to apply for an extension under s 9.


[29] Explanations advanced for not challenging the decision earlier centred mainly on the knowledge, or lack of knowledge, of the various parties from the time the challenged decision was taken, but I have already said that in a case of this kind, which raises in truth a challenge by the public at large, the particular knowledge of those who speak for the broader public cannot be determinative of when the clock starts to run or whether the delay was reasonable. As for the period from the time the tariffs were announced – by which stage, at least, the protagonists acknowledge they were pertinently aware of the decision – the explanation for further delay is, in essence, that they sought to resolve the dispute through political channels before they turned to the courts. It is only when it became clear to them through the budget speech of the Minister of Finance on 22 February 2012 that the political process was bound to fail them, so the appellants say, that they decided to go to court. The respondents’ answer to this is that, as a matter of law, it is impermissible to invoke political processes to justify the delay of instituting legal proceedings. Indeed, there appears to be direct authority for this proposition in English Law (see eg Michael Fordham Judicial Review Handbook 5th ed (2008) at 281 and the cases there cited).


[30] But while that might be a consideration to be taken account of in appropriate circumstances it must nonetheless be evaluated within the context of the particular case. Had the challenge been only to the amount of the tariff that was announced on 11 February 2011 – which had no material consequences after that date – the fact that it was sought to be resolved politically for a time might have weighed in evaluating whether that challenge was unduly delayed. But the challenge in this case is to a decision that was made some four years before the proceedings were commenced – with all its profound consequences having occurred by the time the tariffs were announced – and it is to those profound consequences that the delay rule is directed. The announcement of the tariffs was no doubt the catalyst for taking action but by then the consequences had already occurred, and I cannot see that the explanation for delaying thereafter can play a material role in determining whether the four year delay from the time the impugned decision to declare the toll roads was taken should be overlooked.


[31] Whatever the explanations might be, the fact remains that some five years have now elapsed since the impugned decisions were taken. During that period, vast and significant upgrades to the GFIP highways and related infrastructures, had been completed. By all accounts these upgraded roads are truly magnificent. The advantages are enjoyed primarily by the motorists of Gauteng, but they also benefit the economy of the country as a whole. The downside is that this came at a cost of R20 billion. This amount had been paid by SANRAL with borrowed money. The interest on the loan is running at an alarming rate. SANRAL tells us – and this is not gainsaid – that if the impugned decisions were challenged at the outset, these loans would not have been incurred and the roads would not have been built. There was no plan B. In the absence of tolling there could thus be no GFIP. Without the anticipated income of toll, SANRAL simply does not have the money to pay the interest on the loan, let alone the capital amount of R20 billion. Moreover, SANRAL has contractually committed itself for the maintenance of the toll roads and the collection of toll on the supposition that all this would be recovered from tolling.


[32] In addition, so we are told, there are backlogs for the maintenance of the national roads network covering approximately 18 000 kilometres, for which SANRAL is responsible, to the tune of about R149 billion. If this backlog is to be addressed over the next ten years, at least R15 billion will be required each year. If the backlog is not addressed, the maintenance cost will rise exponentially. Apart from this, SANRAL estimates that capacity improvements and new roads on the national road network will require an additional amount of R10.3 billion per year. If it is obliged to fund the GFIP from its own resources, so SANRAL says, all these projects will come to an end. In consequence, the setting aside of the impugned decisions will not only affect the GFIP. It will have a detrimental impact on the countrywide network of national roads as a whole with a clear knock-on effect on the economy. Moreover, at this stage SANRAL is unable to pay the interest on the R20 billion loan because it is foregoing the anticipated toll income of R200 million per month. This in itself has already led to the downgrading of SANRAL by Moody’s, a credit rating agency. It follows that the setting aside of the impugned declarations will render it virtually impossible for SANRAL to borrow money at all in the future.


[33] As to the effect of a setting aside of the declarations on the Treasury, we know that the South African government, acting through Treasury, has guaranteed the R20 billion loan. The government had done so, we are told, on the hypothesis that the roads were validly declared toll roads and that SANRAL would thus be able to meet its obligations under the loan agreements through the collection of tolls. Had the declarations been challenged at the time, so Treasury says, government would not have put up the guarantee. If government is called upon to meet the guarantee – which is bound to happen if there will be no tolling – it will have a deleterious effect on funding so desperately needed by health care, educators, pensioners, those dependent on social grants, and so forth. Precisely what that effect will be, we do not know. What we do know, however, is that the reordering of public resources inevitably has a polycentric effect. ‘Polycentric’ in a sense described with reference to the image of a spider’s web, where the pull on one strand distributes tensions after a complicated pattern throughout the web as a whole. It is ‘polycentric’ because it is ‘many centred’ – each crossing of the strand is a distinct centre for distributing tensions (see Lawrence Baxter Administrative Law at 86).


[34] The appellants’ argument as to why the 180 day time limit should be extended despite the severe prejudice that prevention of tolling will hold for the respondent and for the public as a whole were sixfold. First they contended that the Gauteng freeways would have been upgraded and expanded in any event. But on the evidence we know that this is simply not so. In fact, the evidence of the respondents, which we are bound to accept, shows the exact opposite. Without the anticipation of toll income, SANRAL could not and would not have borrowed the R20 billion. And it is that loan of R20 billion which would cause the immutable problem if the declarations were to be set aside.


[35] The appellants’ second answer was that the upgrades will be paid for by the government in any event. That is undoubtedly so. In fact the upgrades have already been paid for. Moreover, it is clear that even without toll money the loan will eventually be repaid by the government. But at what costs and to whom? Again we know from the respondents’ evidence that if the government is compelled to pay, the burden will probably be passed on to those who can least afford it. The appellants’ third answer, closely linked to the second, is that the total debt of R20 billion in public finance terms is relatively speaking very small in that it will constitute no more than 0,2 per cent of the South African debt portfolio of R1 trillion. I find this argument somewhat cynical. Relatively speaking, any amount can be trivialised depending on the figure used by way of comparison. But the fact remains, as I have said, that the R20 billion will probably be passed on to those who can least afford it and on whom the effect will by no means be trivial.


[36] The fourth basis invoked by the appellants as to why the 180 day time bar should be extended was that it is the requirement of the rule of law that the exercise of all public power should be lawful and that SANRAL and the government has failed to act legally. As I see it, however, the argument is misconceived. While it is true that the principle of legality is constitutionally entrenched, the constitutional enjoinder to fair administrative action, as it has been expressed through PAJA, expressly recognises that even unlawful administrative action may be rendered unassailable by delay.


[37] The appellants’ fifth contention as to why the 180 day period should be extended is that the validity of the toll road declarations will in any event arise at the next stage of the determination by the toll tariff by the Minister under s 27(3) of the Act. This is so, their argument went, because the tariff determination flows from and is necessarily dependent on lawful toll declarations in terms of s 27(1). It follows, so the argument concluded, that the eventual toll tariff determination will be assailable on the basis of the unlawfulness of the preceding toll road declaration, which means that a refusal of the extension of the 180 day period will not serve any dispositive function. As the legal substructure for this line of argument, the appellants sought to rely on a statement of this court in Oudekraal Estates (Pty) Ltd v City of Cape Town and others 2004 (6) SA 222 (SCA) para 31.


[38] However, [] the passage in Oudekraal upon which the appellants rely is authority for the contrary. That passage makes it clear that, unless an invalid administrative act is set aside by a competent court, it is regarded as valid for the purpose of consequent acts. That is supported by the following statement in the unanimous judgment of the Constitutional Court in Camps Bay Ratepayers’ & Residents’ Association and another v Harrison and another 2011 (4) SA 42 (CC) para 62:

‘As was explained in Oudekraal Estates (Pty) Ltd v City of Cape Town and Others [para 31] administrative decisions are often built on the supposition that previous decisions were validly taken and, unless that previous decision is challenged and set aside by a competent court, its substantive validity is accepted as a fact. Whether or not it was indeed valid is of no consequence.’


[39] The appellants’ sixth argument as to why the 180 day period should be extended under s 9 was that the alleged unlawfulness of the challenged decisions will in any event give rise to a collateral challenge every time SANRAL seeks to compel payment of toll. But as I see it the argument negates the fundamental differences between a collateral challenge, on the one hand, and a direct challenge by way of a review application, on the other. Those differences were underscored by Howie and Nugent JJA in Oudekraal Estates para 36 when they said:

‘The right to challenge the validity of an administrative act collaterally arises because the validity of the administrative act constitutes the essential prerequisite for the legal force of the action that follows and ex hypothesi the subject may not then be precluded from challenging its validity. On the other hand, a court that is asked to set aside an invalid administrative act in proceedings for judicial review [ie a direct challenge] has a discretion whether to grant or to withhold the remedy. It is that discretion that accords to judicial review its essential and pivotal role in administrative law, for it constitutes the indispensable moderating tool for avoiding or minimising injustice when legality and certainty collide. Each remedy thus has its separate application to its appropriate circumstances and they ought not to be seen as interchangeable manifestations of a single remedy that arises whenever an administrative act is invalid.’


[40] In this light it should be apparent that the 180 day time bar in s 7(1) is confined to direct challenges by way of proceedings for judicial review. It does not limit collateral challenges at all (see eg Kouga Municipality v Bellingan and others 2012 (2) SA 96 (SCA) paras 15-16). It is therefore both unnecessary and inappropriate to extend the 180 day time limit in order to provide for potential collateral challenges. Absent any extension under s 9, the 180 day time bar precludes us from entertaining the direct challenge by way of a review application. We cannot avoid that limitation to our authority simply because the same questions might arise were there to be future collateral challenges, the success of which is by no means certain, that are not before us.


[41] After all is said and done, the stark reality remains that because of the delay in bringing the review application, five years had elapsed since the impugned decisions were taken, and that, during those five years, things have happened that cannot be undone. The delay rule gives expression to the fact that there are circumstances in which it is contrary to the public interest to attempt to undo history. The clock cannot be turned back to when the toll roads were declared, and I think it would be contrary to the interests of justice to attempt to do so. It follows that the appellants’ application for an extension under s 9(1) should, in my view, be refused. The result, as I see it, is that we are prevented by the provisions of s 7(1) of PAJA from embarking upon the merits of the review application.


[42] Finally the appellants contended, in the alternative, that even if the interests of justice do not justify the extension of the 180 days in terms of s 9, the review application should still be heard. Their argument in support of this contention relied on s 25(1) of the Constitution, which provides that ‘no one may be deprived of property except in terms of law of general application’. Since this challenge is not brought under s 6 of PAJA, so the appellants argued, it is not subject to the 180 day time limit. There is probably more than one valid answer to this argument. So, for example, it could be said, I think, that tolling does not constitute deprivation because it is a quid pro quo for the use of the road. But be that as it may. I think that on the facts of this case the argument based on deprivation of property is self-destructive. The appellants do not contend that a toll validly imposed under s 27 of the Act would constitute an infringement of their rights under s 25(1) of the Constitution. Toll could therefore, on the appellants’ own argument, only constitute a deprivation of property if it is invalidly imposed. But unless and until the impugned declarations are reviewed and set aside, they are regarded as valid in law (see Oudekraal Estates para 31). In order to succeed the appellants are therefore driven back to their application for the review of the declarations under s 6(1) of PAJA, which brings the 180 day time limit into play.


[43] Hence I believe that despite the appellants’ various arguments to the contrary, we are not authorised to enter into the merits of the review application. Contrary to this approach, the court a quo first dealt with the review application and dismissed it on its merits. In consequence the court found it unnecessary to consider the effect of the delay rule. Although, for the reasons I have given, I do not agree with this approach, the conclusion I arrived at on the outcome of the review application happens to be the same, namely that it could not succeed. This means that in substance the appeal must fail. What remains are issues of costs. In this regard the appellants relied on what has become known as the Biowatch principle, after the decision of the Constitutional Court in Biowatch Trust v Registrar, Genetic Resources and others 2009 (6) SA 232 (CC) paras 21-25. What the principle lays down in sum is that in constitutional litigation an unsuccessful party in proceedings against the state ought not to be ordered to pay costs lest litigants may be discouraged to vindicate their constitutional rights. Yet as the Constitutional Court also said, the principle is not immutable. So, for example, it will not immunise an unsuccessful applicant from an adverse costs order if the application is found to be frivolous, vexatious or otherwise manifestly inappropriate.


[44] Despite the respondents’ contentions to the contrary, I believe this is a case where the Biowatch principle should be applied. First, this is unquestionably constitutional litigation. Secondly, the application was clearly brought in the public interest, albeit that some of the appellants may also have been motivated by commercial considerations. Thirdly, I do not believe that the conduct of the appellants was manifestly inappropriate so as to warrant a departure from the general rule. The court a quo awarded costs against the appellants. Yet it does not appear from the court’s judgment that any consideration was given to the Biowatch principle when it made that order. In the light of the view that I hold about costs, I believe we are obliged to replace the court a quo’s order with one of no order as to costs. To that limited extent the appeal must therefore succeed. But as I see it, this does not mean that the appellants were substantially successful to the extent that it justifies a costs order in their favour on appeal. In the result, I believe, the costs of appeal should, as in the court a quo, also land where it falls, in accordance with the Biowatch principle.


[45] In the result it is ordered:

The appeal is refused with no order as to costs, save that the order granted by the court a quo, directing the appellants to pay the respondents’ costs, is set aside and replaced by an order that there be no order as to costs.




____________

F D J BRAND

JUDGE OF APPEAL



APPEARANCES:


For Appellants: M C Maritz SC

A J D’Oliveira

K D Iles


Instructed by: Cliffe Dekker Hofmeyr Inc

SANDTON


Correspondents: Webbers

BLOEMFONTEIN



For First Respondent: D N Unterhalter SC

G E Leech SC


Instructed by: Werksmans Attorneys

SANDTON


Correspondents: Symington & De Kock

BLOEMFONTEIN


For Second and Third

Respondents: I V Maleka SC

T J V Bokaba SC


Instructed by: The State Attorney

PRETORIA


Correspondents: The State Attorney

BLOEMFONTEIN


For Seventh Respondent: J J Gauntlett SC

T N Ngcukaitobi

F B Pelser


Instructed by: The State Attorney

PRETORIA


Correspondents: The State Attorney

BLOEMFONTEIN



▲ To the top

Documents citing this one 21

Judgment 21
  1. Asla Construction (Pty) Ltd v Buffalo City Metropolitan Municipality and Another (894 of 2016) [2017] ZASCA 23 (24 March 2017)
  2. Blue Water Creek Homeowners Association v Kanniah and Others; In Re: Kanniah and Another v Blue Water Creek Homeowners Association and Others (A96/2020) [2023] ZAGPPHC 8 (9 January 2023)
  3. Buffalo City Metropolitan Municipality v Asla Construction (Pty) Ltd [2019] ZACC 15 (16 April 2019)
  4. City of Tshwane Metropolitan Municipality v Afriforum and Another [2016] ZACC 19 (21 July 2016)
  5. City of Tshwane Metropolitan Municipality v Lombardy Development (Pty) Ltd and Others (724 of 2017) [2018] ZASCA 77 (31 May 2018)
  6. Commissioner: South African Revenue Service v Sasol Chevron Holdings Limited (1044 of 2020) [2022] ZASCA 56 (22 April 2022)
  7. Forge Packaging (Pty) Ltd v Commissioner for South African Revenue Service (21634/2021) [2022] ZAWCHC 119 (13 June 2022)
  8. Ketlele N.O. and Another v Ketlele and Others (35600/19) [2022] ZAGPJHC 47 (9 February 2022)
  9. Kona v Premier Eastern Cape and Others (3277/2018) [2023] ZAECMHC 30 (20 June 2023)
  10. Matoto v Free State Gambling and Liquor Authority and Others (987 of 2017) [2018] ZASCA 110 (12 September 2018)
  11. Member of Executive Council for Cooperative Governance and Traditional Affairs, KwaZulu-Natal v Nkandla Local Municipality and Others [2021] ZACC 46 (8 December 2021)
  12. Mohapi and Others v Registrar Free State Transport and Others (2283/2022) [2022] ZAFSHC 151 (2 December 2022)
  13. Mostert and Others v Nash and Another (604 of 2017) [2018] ZASCA 62 (21 May 2018)
  14. National Council of and for Persons with Disabilities v Independent Communications Authority of South Africa (49918/2021) [2024] ZAGPPHC 172 (22 February 2024)
  15. National Energy Regulator of South Africa and Another v PG Group (Pty) Ltd and Others [2019] ZACC 28 (15 July 2019)
  16. South African National Roads Agency Limited v City of Cape Town (66 of 2016) [2016] ZASCA 122 (22 September 2016)
  17. South Durban Community Environmental Alliance v MEC for Economic Development, Tourism and Environmental Affairs: KwaZulu-Natal Provincial Government (231/2019) [2020] ZASCA 39 (17 April 2020)
  18. Supaluck Investments (Pty) Ltd v Valuations Appeals Board: City of Johannesburg and Another (34752/2019) [2023] ZAGPJHC 138 (28 February 2023)
  19. Sustaining the Wild Coast NPC and Others v Minister of Mineral Resources and Energy and Others (3491/2021) [2022] ZAECGHC 45 (1 September 2022)
  20. Vodacom (Pty) Ltd v National Commissioner South African Police Service and Others (18770/2022) [2023] ZAGPPHC 151 (17 March 2023)
  21. Zuma v Democratic Alliance and Economic Freedom Fighters (1028/2019) [2021] ZASCA 39 (13 April 2021)