Loliwe CC t/a Vusumzi Environmental Services v City of Cape Town and Others (3791/2012) [2012] ZAWCHC 162 (6 July 2012)

Republic of South Africa

IN THE HIGH COURT OF SOUTH AFRICA

(WESTERN CAPE HIGH COURT, CAPE TOWN)

Case No: 3791/2012

Before: The Hon Mr Justice Binns-Ward

In the matter between:

LOLIWE CC t/a VUSUMZI ENVIRONMENTAL SERVICES ........................................Applicant

and

CITY OF CAPE TOWN ............................................................................................First Respondent

CHAIRPERSON, ADJUDICATION COMMITTEE ..................................Second Respondent

APPEAL AUTHORITY, CITY OF CAPE TOWN .................................................Third Respondent

SA METAL (PTY) LTD t/a WASTE CONTROL .......................................Fourth Respondent

WASTE MART CC .....................................................................................................Fifth Respondent

JUDGMENT

: The Honourable Justice A.G. Binns-Ward

FOR THE APPLICANT

: Adv. J. NEWDIGATE SC

Adv. P. FARLAM

Adv. K. REYNOLDS

INSTRUCTED BY

: Edward Nathan Sonnenbergs

FOR THE 1st, 2nd& 3rdRESPONDENT : Adv. M. JANISCH

Adv. M. O’SULLIVAN

INSTRUCTED BY

: FairbridgesArderne& Lawton

FOR THE 4th RESPONDENT : Adv. R. GOODMAN SC

Adv. A. DU TOIT

INSTRUCTED BY

: BernadtVukic Potash & Getz

FOR THE 5th RESPONDENT : Adv. J. DE WAAL

INSTRUCTED BY

: Webber Wentzel

DATE OF HEARING

: 5JUNE 2012

DATE OF JUDGMENT : 6 JULY 2012

JUDGMENT DELIVERED: 6 JULY 2012

BINNS-WARD J:

The applicant has applied for the review and setting aside of decisions by the municipality of the City of Cape Town (the first respondent1) to award contracts to the fourth and fifth respondents, respectively, for the removal of waste from properties within the

municipal area. The fourth respondent was awarded the contract in respect of the area along the City’s Atlantic seaboard,

between Clifton and Hout Bay. The contract awarded to the fifth respondent was in respect of an area in the eastern part of the

metropole. The award to the fourth respondent was the culmination of a procurement process referred to by the parties, for convenience,

as ‘the Atlantic tender’; while that to the fifth respondent manifested the outcome of ‘the Helderberg tender’.

Procurement of goods and services by organs of state is regulated. The regulation of public procurement is founded originally in s 217 of the Constitution.2 This requires that such procurement must occur in accordance with a system which is fair, equitable, transparent, competitive and cost-effective. These requirements are qualified only to the extent that organs of state are permitted to apply preferential procurement policies, directed mainly at redressing historical inequities. Preferential procurement policy by organs of state is in turn regulated in terms of the Preferential Procurement Policy Framework Act 5 of 2000 (‘the PPPFA’), which is

legislation of the nature contemplated in terms of s 217(3) of the Constitution.

The PPPFA stipulates that for contracts of the financial value involved in the current matter a 90:10 preference point system must be applied. This works on the basis that 90 points fall to be awarded to the lowest acceptable tender by price, and 10 points allocated to specific goals, such as contracting with historically disadvantaged persons.3 The framework also obliges organs of state to award a contract to the tenderer who scores the highest number of points, unless objective criteria, in addition to any ‘specific goal’ for which points may be awarded,4 justify the award to another tenderer.5 An organ of state may award a contract only consequent upon the acceptance of an ‘acceptable tender’, as defined in the PPPFA, viz. ‘any tender which, in all respects, complies with the specifications and conditions of tender as set out in the tender document’.6 The acceptance by an organ of state of a tender which is not ‘acceptable’ within the meaning of the PPPFA is therefore an invalid act and falls to be set aside.7 The question of the degree of compliance required for a tender to be an ‘acceptable tender’, as defined, calls for the exercise of judgment. The test is materiality and reasonableness. Not every slip of the pen, or inconsequential or obvious error in a bid will render the tender not acceptable.8

Procurement by municipalities is further regulated under the supply chain management provisions of the Local Government: Municipal Finance Management Act 53 of 2003 (‘the MFMA’), together with the supply chain management regulations made under that statute (‘the SCM regulations’);9 and, to the extent that it entails the employment of external providers of municipal services, also by the Local Government: Municipal Systems Act, 32 of 2000 (‘the Systems Act’). In accordance with the obligation imposed on it in terms of ss 111 and 112 of the MFMA, the City of Cape Town has adopted and implemented a supply chain management policy to achieve compliance with the requirements of s 217 of the Constitution.10 The contracts in issue are ‘long term contracts’ within the meaning of the SCM regulations. The contract value also

exceeded R200 000. The City was thus obliged to procure the services involved by means of a competitive bidding process.11 As Pickard JP observed in Cash Paymaster Services (Pty) Ltd v Eastern Cape Province and others 1999 (1) SA 324 (Ck) at 350H, ‘The very essence of tender procedures may well be described as a procedure intended to ensure that government, before it procures goods or services, or enters into contracts for the procurement thereof, is assured that a proper evaluation is done of what is available, at what price and whether or not that which is procured serves the purposes for which it is intended’.12

Some of the relevant parts of the SCM regulations were described at some length in Loghdey v City of Cape Town 2010 (6) BCLR 591 (WCC). It has been convenient to draw quite heavily on what was said there in that respect for the purpose of sketching part of the applicable statutory framework in the current case.

The SCM regulations allow for a municipality’s supply chain management system to provide for a ‘committee system’ to deal with the competitive bidding process.13 A committee system is required to comprise of at least three committees; namely, a bid specification committee, a bid evaluation

committee and a bid adjudication committee.14 In terms of its SCMP, the City had adopted a committee system.

A bid specification committee must be composed of one or more officials of the municipality, preferably including the manager responsible for the function involved. In terms of the regulatory framework a bid evaluation committee is required to evaluate bids in accordance with (i) ‘the specifications for a specific procurement’15 and (ii) the applicable points system. In addition, the evaluation committee must evaluate each bidder’s ability to

execute the contract.16 In the current case the tender specifications set out certain criteria to be applied ‘to determine tender compliance’.

These included –

Capacity in terms of staff, vehicles and equipment. The contractor must be able to demonstrate that he/she has sufficient understanding

of the work to be able to calculate the resources needed, and also show that he/she will be able to provide these resources within of (sic) the prescribed lead time after being awarded the tender

Financial viability. Offers must be such that they cover all costs associated with providing any combination of the service scenarios

including capital investments, infrastructure, authorities and approvals, royalties, operational expenses, calculated income,

market fluctuations, growth, CPIX, fuel, labour, and other consumable increase (sic) etc., and can thus sustain operations for

the full tender period. A full breakdown of costs must be provided, i.e. the pricing schedule and all its associated sheets e.g.

Conventional Collection alternative, (sic) Conventional or Wet collection service in NON RESIDENTIAL and RESIDENTIAL, Dry collection service level in identified RESIDENTIAL areas etc.

As will become apparent, the evaluation of the financial viability of the applicant’s bid, and more especially, ‘the

breakdown of costs’ aspect of it, is the critical question in the current case insofar as the Atlantic tender is concerned.

The function of a bid adjudication committee, which should be comprised primarily of officials from the department requiring the goods or services in issue,17 is essentially to provide a final consideration by senior management officials of the municipality, with the assistance of a technical expert in the relevant field, of the bid evaluation committee’s recommendation before the relevant contract is awarded. It is evident that the bid adjudication committee will in the ordinary course endorse the recommendations of the bid evaluation committee. This may be implied, I think, from the provisions of SCM regulation 29(5),18 and the special steps with which a municipality must comply if it awards the contract to a tenderer other than the one recommended in the normal course of implementing the supply chain management policy.19

The SCM regulations require that bid documentation must, amongst other matters, include evaluation and adjudication criteria.20 In addition to the evaluation criteria referred to earlier,21 the bid documentation in the current case stipulated that tenders which failed to score a minimum of 60 points for ‘functionality’ would be regarded as ‘non-responsive’ and would not be further evaluated. The City’s SCMP defined ‘functionality’ as ‘the measure, according to predetermined criteria, of the suitability of a proposal, design or product or the use for which it is intended, and may also include a measure of the competency of a supplier’.22 The scoring of a tender for ‘functionality’ was thus a qualifying exercise, and only those tenders which attained at least the minimum points required for responsiveness in respect of ‘functionality’ would then be evaluated for compliance with the specifications and conditions of the tender, and scored in terms of the points systems applicable in terms of the PPPFA. This approach to the evaluation of bids was consonant in principle with the provisions of the City’s SCMP.23

The bid evaluation committee treated the applicant’s tenders in respect of both the Atlantic and the Helderberg tenders

as non-responsive. The applicant’s bid in respect of the Atlantic tender was so treated because of its failure to attain the required minimum number of points in respect of functionality.

It is well established that the award of a tender contract by an organ of state constitutes ‘administrative action’ within the meaning of the Promotion of Administrative Justice Act 3 of 2000 (PAJA). The application for the review and setting aside of the award of the tender contracts is thus brought in terms of s 6 of PAJA.

The Atlantic tender

It is convenient to deal with the Atlantic tender first. The essence of the applicant’s complaint was that the scoring of its tender for functionality by the bid evaluation committee was irrational and unreasonable in the sense that it led to a decision that a reasonable decision-maker could not make.24 (The applicant also sought to make out a case that the City’s officials, in particular the chairperson of the bid evaluation committee, had been biased against it, but – while not abandoned – that was not pursued in oral argument.)

The way in which functionality would be scored was set out at clause 7.6 of the tender document. There were four criteria, which were accorded weight in accordance with the table set out below:

There was a note at the end of clause 7.6 of the tender document, which provided as follows: ‘Note: Please ensure that all relevant information has been submitted with your tender submission to ensure optimal scoring of Functionality points’. Clause 6.2 of the Tender Document provided ‘Any portion of the Tender Document not completed will be interpreted as “not applicable”.’ It seems clear on a reading of the Tender Document as a whole that the ‘Data (Breakdown of cost)’ component of the functionality evaluation bore directly on the information to be provided by a tenderer in terms of the ‘Breakdown of Provisional Costs’ tables in the pricing schedule section (i.e. Part 4) of tender documents. The applicant did not complete certain items of the information requested in the breakdown of costs in respect of the ‘conventional or wet collection’

component25 of the waste removal contract. The information fell to be provided upon a completion by the tenderer of table A1 in the bid document. It was the scoring of this ‘evaluation area’ of the functionality test by the bid evaluation committee that lay at the heart of the applicant’s complaint.

The ‘Breakdown of Provisional Costs’ table invited the tenderer to provide a number of items of information. The table, as it was completed by the applicant (in table A126), read thus:

From a pricing perspective the critical figure in the costs breakdown table was the ‘rate per lift’ (the last item in the table). This is because the bid document stated that ‘[t]he single unit rate should be tendered for either Wet/dry or the Conventional collection service level and will be applicable to any number of lifts. It will also be applicable if the number of service points and or lifts increases through growth and extensions to the contract’. (Bold font in the original.)

The items which the applicant failed to fill in on the breakdown of provisional costs section of the tender document were ‘Kg’s/Lift’ and ‘Total Monthly Cost’. (A ‘lift’ in this context constitutes - at least in respect of ‘wet waste’ - the content of a standard 240l municipal waste bin - popularly called a ‘wheelie-bin’). The anticipated number of kilograms of waste collected per lift would be relevant to determine how quickly the waste removal vehicles to be employed by the tenderer would be filled to capacity, thus determining the frequency during the course of a day’s removal activity at which the vehicle would be required to travel from the serviced area to a waste disposal site. Accordingly, the information could bear in a relevant manner on the number of vehicles required to execute the anticipated 41564 lifts per month and the number of staff required to be employed. The result of an evaluative extrapolation and analysis of the estimated costs of labour and equipment against the tendered ‘rate per lift’ could obviously provide an indication of the logistical and financial viability of a tenderer’s bid.

The applicant failed to cross the functionality hurdle because it was given a low score of just 10 points out of the possible 40 in respect of ‘Data (Breakdown of cost)’. It scored quite well in the three other ‘evaluation areas’.27 In the result it was awarded a total of 55.5 points for functionality.

The chairperson of the bid evaluation committee, who was also involved in the bid specification, averred that the City’s solid waste department had developed a spreadsheet to facilitate the consistent scoring of the ‘Data (Breakdown of cost)’ part of the functionality evaluation. The spreadsheet was ‘populated’, so he averred in the City’s answering

affidavit, with information which enabled the internal coherence of the information filled in by tenderers to be checked, and also facilitated the committee’s task in evaluating each bidder’s ability to execute the contract in an efficient and viable manner. The deponent stated ‘The idea is that the spreadsheet should identify parts of the tender that are of questionable validity, and thus help to ascertain whether the tender is operationally and functionally viable with reference to the service to be provided.’ The information provided in the City’s answering affidavit did not, however, provide sufficient information to enable an understanding as to exactly how the spreadsheet tool worked. As a result the only manner of objectively assessing it is by examining some of the results it produced; a matter to which I shall come presently.

The data provided in respect of cost breakdown was scored in eight categories by the bid evaluation committee using the aforementioned spreadsheet tool. Each of these categories was allocated five points. A tenderer whose tender included sufficient information to enable certain items of the information supplied to be cross checked for consistency was given the five points. If the item could not be cross-checked by reason of an absence of information, or if the information provided highlighted non-viability or inconsistency, no points at all were awarded. Thus in respect of each of the eight categories a tenderer would be either awarded five points, or none at all; there was no gradation in the scoring. The categories for which points were allocated in respect of the ‘Data (Breakdown of cost)’ evaluation criterion are set out below in the tabular format in which they were submitted to, and approved by the City’s

Director: Waste Management:

The applicant scored poorly in respect of ‘Data’ for four reasons. The first was because of its failure, when completing the relevant part of the standard bid document, to include an amount in respect of ‘kilograms per lift’ - in other words its failure to provide its estimate of the average amount of wet waste in kilograms per wheelie-bin to be collected. This omission rendered it impossible for the committee, unless it were to read in the missing information, to cross check a number of

other items in the costs breakdown section of the bid for internal consistency and resulted in the applicant being given a number of nil scores. The second reason was because, with similar effect, the applicant omitted to insert the total of its estimate of the monthly costs in providing the service. The third was that it mistakenly indicated, for cost breakdown purposes, that it would be using only one vehicle to provide the service. (Elsewhere in its tender, the applicant had indicated that it would be using three vehicles, two of which were already held in possession, the other to be acquired.) The fourth reason, closely connected to the third, was that the applicant indicated that it would be able to effect 1919 lifts per day. It was common ground that this figure (which had demonstrably been arrived at simply by dividing the number of lifts per month required in terms of the tender specifications28 by the number of working days in the month29) was beyond the limits of feasibility if only one vehicle were to be employed, and that the bid evaluation committee could not be faulted for recognising as much. (It was not explained by the City how any tenderer could score for indicating less than 2796 lifts per day, being the number of lifts required to be removed on beat no.s 1.3.411, 1.3.412, 1.3.416 and 1.3.417 in Hout Bay and Camps Bay every Tuesday, according to the tender specifications. A lesser figure could only be acceptable if it were treated as indicative not of the maximum daily removal capability of the tenderer expressed in lifts, but only as the average daily number of lifts to be effected divided by the number of vehicles employed as the City’s scoring criteria seem to suggest, but which the compilation of the table does not specify. The fourth respondent indicated 959.9 lifts per day per vehicle with the use of two vehicles.30 The manner in which the information was requested did not make it clear whether an average figure, or a maximum capacity figure was the nature of information sought.)

It is also common cause that if the applicant had filled in 13.5 kg,31 being a figure arrived at by dividing its estimate of a total monthly collection of 560 tons of wet waste by the number of lifts (41 564) provided by the City as being required in terms of the contract in the block provided for kilograms per lift, or if it had completed the ‘total monthly cost’ block in the tender form with a figure constituting the total of the individual ‘monthly costs’ it had set out in its bid, it would have scored well enough to qualify in respect of functionality. The applicant’s counsel argued that both these missing items of information could have been extrapolated easily by applying simple arithmetic to the information that had been provided by the applicant. As a matter of simple logic their argument cannot be gainsaid.

It was argued further on the applicant’s behalf that a scoring system that disqualified the applicant simply because its omissions did not square with the spreadsheet template, regardless of the fact that the missing information could readily have been read in using the data that the applicant had supplied, was irrational and conducive to starkly unreasonable results. The

argument was that the system employed by the City in this respect did not satisfy the requirements of s 217(1) of the

Constitution and its derivatives in the legislative framework applicable specifically to municipalities.

The applicant’s counsel also argued that the reference to a single vehicle in the data (costs breakdown) analysis in the applicant’s bid was an obvious error, and should have been recognised as such by the bid evaluation committee from the

information provided (and scored – positively - for functionality) elsewhere in the bid document to the effect that the applicant would use three vehicles to carry out the contract work. In similar vein they contended that the applicant’s given

figure of 1919 lifts per day should have been recognised by the bid evaluation committee as being what the applicant could achieve with the use of more than one vehicle.

With regard to the injunction in clause 7.6 of the bid document (quoted above32), the applicant’s contention was that all the relevant information had been included in its tender. And with regard to the caution in clause 6.2 (also quoted above33), it was submitted that the two blocks not filled in the costs breakdown table did not constitute a ‘portion’ of the tender in the sense meant by clause 2. A ‘portion’, so the argument went, denoted a component section of the bid document, rather than an item within such a section.

In my view any system directed at compliance with the requirements of s 217(1) of the Constitution would, in the context of an approach to scoring of the sort adopted by the bid elevation committee in the current case, expressly disclose in the standard bid document that completion of each and every item in the ‘Data (Breakdown in cost)’ tables was required for scoring purposes. I do not consider that the note at the end of clause 7.6 of the tender document34 served that purpose sufficiently. In the context of information required, the ‘Kg’s/lift’ item could only be a figure related proportionately to the total weight of the waste to be collected. It would not be readily apparent to a tenderer that giving the monthly total weight of waste to be collected, even if estimated in an eminently realistic amount, would be an inadequate, indeed ineffectual, means of showing a qualified and informed insight into the nature of the

work involved simply by reason of a failure also to divide the weight of waste to be collected monthly by the given number of lifts

to be made monthly, thereby giving the same information twice in the table, albeit in different ways. Nothing in the definition of ‘functionality’ in the City’s SCMP would have served to alert a tenderer that an arithmetical cross-check of kilograms per lift and monthly tonnage would be a critical factor in scoring the provisional costs information provided by the tenderers. There was also nothing objectively apparent in the request for information in the ‘Data’ tables to alert tenderers to the premium placed by the bid evaluation committee or the spreadsheet tool on the completion by tenderers of the ‘Kg/lift’ item over the ‘Tons/month’ item in the tables. The non-disclosure in the bid document of the requirement that every item on the table required completion for the purposes of scoring, even if the relevant information could readily be deduced from other information provided by the tenderer elsewhere in the table, detracted materially, in my view, from the fairness and transparency of the system used in the current matter.

The position is distinguishable from that which obtained in South African Post Office Ltd v Chairperson, Western Cape Provincial Tender Board, and others supra,35 a judgment relied upon by counsel for the City, in which the non-disclosure by the procurement body of the weighting to be attached to identified evaluation criteria was found to have been unexceptionable in the peculiar circumstances of the case. In SA Post Office the complaint by the unsuccessful tenderer was that it had not been informed that the weighting in the evaluation of price and service, being the criteria evaluated in the last phase of a four stage evaluation process, would be equal; that is 50% of the points that could be scored would be allocated to each criterion. The unsuccessful tenderer had complained that it was taken by surprise that price should have carried as high a weighting as service. In the context of the evident importance of pricing in any competitive bidding process, the court was not surprisingly unimpressed by the contention that the non-disclosure of the weighting had been unfair in the circumstances. Furthermore, in that case the court was satisfied that the Tender Board’s explanation for the non-disclosure of the weighting confirmed the rationality of its action.

The applicant contended that in the context of the undisclosed importance of the provision by a tenderer of the ‘Kg/lift’ data, it was treated unfairly by not having been requested to provide it by way of clarification. The City argued that to afford the applicant the opportunity to complete the item would be to give it an opportunity to supplement its tender, or a second bite at the cherry in a manner that would have been unfair to the other participants in the tender. Moreover, so argued the City, the applicant should have appreciated from the provisions of clause 6.2 of the tender document that any item not filled in would have been treated as ‘not applicable’.

The applicant’s counsel riposted that the manner in which the scoring of this evaluation criterion was undertaken did not treat the item as inapplicable, it instead read in the kilograms per lift as nil and, using the nil so supplied, disqualified the

monthly tonnage figure given by the applicant from any consideration at all in the evaluation on the basis of its inconsistency with nil kilograms per lift notionally inserted. I do not find it necessary to make a finding in respect of the contesting submissions on the clarification question. Suffice it to say by way of general observation that I consider that the apparent eschewal as a fixed approach by the bid evaluation committee of its power to seek clarification from tenderers is in conflict with the terms of the tender document and the objects of the legislation. An appropriately formulated enquiry on a point of clarification (all such enquiries are required to be in writing and therefore on the record) should not give rise to unfairness.

I agree, however, with the argument advanced on the applicant’s behalf on the ‘not applicable’ point. I also agree with the submission by Mr Farlam (who was led by Mr Newdigate SC) for the applicant that clause 6.2, on an ordinary reading, would appear to pertain to the non-completion by a tenderer of a discrete section (i.e. ‘portion’) of the tender document; and not, certainly in the context currently under consideration, to an item such as ‘Kg/lift’ in table A1, where the apparently relevant information was evident - or at least readily deducible - from the data given by the tenderer in respect of the item in the very same table against ‘Tons/month’.

The applicant’s counsel were also able to demonstrate that the scoring system used for the ‘Data (Breakdown of cost)’ part of the functionality evaluation led to some rather bizarre results. Thus, for example, the applicant scored nil points for its tonnage per month estimate in respect of wet waste notwithstanding that its estimate in that respect (560 tons) was within a narrow margin of appreciation from that estimated by the successful tenderer (581 tons), and despite the fact that it might have been assumed by the City that, as the entity that had been rendering the service for the previous three years, the applicant’s estimate would in all likelihood be based on what it was actually collecting.36 On the other hand, a competing tenderer, Interwaste (Pty) Ltd, which had estimated a wet waste collection of 1039 tons per month, which - in the context of the tender being awarded to a party which had estimated 458 tons less – appears to have been way outside the realistic range, was awarded 5 points for its tons per month data, apparently simply because of the correct arithmetical correlation of its figure of ‘tons per month’ with the figure inserted by it against ‘Kgs/lift’. This lends cogency to the observation by the applicant’s counsel that the scoring approach adopted by the evaluation committee was directed more at arithmetical cross-checking than at determining functionality as defined in the City’s SCMP. The result was that tenderers could score points for ‘functionality’ notwithstanding that the content of the information supplied by them indicated on its face that they had materially misdirected themselves on the volume of waste to be removed.

Another example of the logically inexplicable scoring by the spreadsheet tool is afforded by its treatment of the information included in the tender by Tedcor Women in Waste (Pty) Ltd. The amount inserted in respect of total monthly costs was R633 948 in the Tedcor tender, which divided by the number of lifts involved gave a rate per lift of R15,25, for which Tedcor was scored 5 points, notwithstanding that the individual amounts given by it under the items in ‘Monthly costs’ actually totalled R316 974. One could, of course, reason that the amount of R316 974 related to monthly cost per vehicle, but in order to do that one has to do the arithmetic by addition and multiplication of the figures provided.37 If the evaluation committee could do the arithmetic for Tedcor, why could it not do so for the applicant? Moreover, even if one accepts (as I believe one should) that the R316 974 amount provided by Tedcor was a monthly cost per vehicle input (it had indicated that it would be using two vehicles), it is then impossible on the information provided by it in table A1 to reconcile the given labour costs of R30 386 per month with the information provided elsewhere on Tedcor’s table A1 that each vehicle would be manned by a driver costed at R5750 per month and four workers each costed at R3110 per month, which would give a total of R18190 per month per vehicle in respect of labour costs.

In its supplementary founding affidavit the applicant pointed out that the City applied the aggregate figure given by tenderers in table A1 under ‘Total Monthly Cost’ mechanically in its scoring of the ‘Rate per lift’ item and irrespective of whether the given ‘Total Monthly Cost’ tallied arithmetically with the seven individual monthly cost items appearing directly above ‘total monthly cost’ in the table. The City’s response in its answering affidavit was to aver that ‘the City applied consistent criteria to all tenderers, namely to insert the “Total Monthly Cost” figure as furnished regardless of whether that was in fact an aggregate of the seven cost items. It is up to the tenderer to ensure that it does not make an arithmetical error and that the City is in a position to assess its bid adequately’. I agree with the contentions advanced by Mr Farlam that this is irrational.

‘It is a requirement of the rule of law that the exercise of public power by the Executive and other functionaries should not be arbitrary. Decisions must be rationally related to the purpose for which the power was given, otherwise they are in effect arbitrary and inconsistent with this requirement’.38 Whether the exercise of public power passes the rationality test in a given case is a matter for determination on the objective standard.

Bearing in mind that the total monthly cost divided by number of lifts per month gives the rate per lift figure, which is effectively the tender price, the exercise undertaken by the City in scoring for functionality did not test whether the rate per lift was feasible in relation to the tender, whereas a disconnect between the actual aggregate of the monthly costs items and the total monthly cost should serve as a warning that the tendered rate might not have been premised on a sound costing analysis and thus also as an adverse indicator of the suitability of the proposal. Elsewhere in its answering papers, the City in fact highlighted the hazards of contracting with tenderers whose tenders are not realistically formulated. The positive scoring of a rate per lift rate which does not make sense in the context of the other data provided in the table by the tenderer just does not bear scrutiny. The pertinence of such irrationality in the current context is that it could result in the arbitrary exclusion from consideration of tenderers whose tenders reasonably deserved further evaluation, and the inclusion in the process of tenderers whose tenders could be said, objectively, to be relatively more dubious. The irrationality rendered the entire functionality qualification process unfair and detracted from its ability to properly satisfy the requirements of promoting competitiveness and achieving cost-effectiveness.

There is much to be said in the circumstances for Mr Newdigate’s argument that the scoring process applied by the City abdicated the intelligently evaluative function, which the regulatory framework demands of a bid evaluation committee, in favour of a rigid mechanical process which, by reason of its design, gave out irrational results and determined matters such as the internal consistency of tender information arbitrarily. I agree with the argument advanced by the applicant’s counsel that there was no rational reason for the bid evaluation committee, applying the scoring method which it used, to score the applicant nil for the information filled in for its rate per lift. There was no reason why the rate per lift given by the applicant could not be verified arithmetically by adding up the monthly cost items duly completed by the applicant and dividing the result by number of lifts involved. It would be apparent, were this simple exercise undertaken, that the rate per lift quoted by the applicant (R11,06) tallied exactly with the result of the arithmetical exercise just described.

The reason given on the committee’s behalf for its failure to do the arithmetic does not bear up. The explanation was that by doing the arithmetic the committee might end up favouring the applicant if it did it correctly because it might well have turned out that the applicant, if it had filled the figure in itself, might have inserted an incorrect figure, and thereby created

an internal inconsistency. In this instance, however, the quoted rate per lift tallied with the total of the monthly costs quoted

divided by given number of lifts and the issue of a relevant inconsistency could therefore not arise. The applicable legislation

required the bid evaluation committee to undertake an evaluative exercise. Its approach in respect of the scoring of the figure given by the applicant in respect of rate per lift was entirely mechanical, and irreconcilable with the evaluative approach it was required to apply.

The applicant has thus established that its bid in the Atlantic tender was excluded from consideration as a consequence of an unlawful process by the bid evaluation committee. It does not follow, however, that its application for the review and setting

aside of the award of the contract to the fourth respondent must in consequence thereof necessarily be upheld. As the applicant is the only one of the 12 unsuccessful tenderers which has challenged the award of the tender contract, there would be no point in the circumstances of this case in setting aside the award if it were to appear unlikely that upon the evaluation of the applicant’s tender its bid would qualify as an acceptable tender. To do so would be to interfere in a situation where it did not appear sufficiently that the applicant had been prejudiced by the irregularity and in which the point established by the applicant would essentially be moot in character, or of only academic interest; cf. e.g. Jockey Club of South Africa and Others v Feldman 1942 AD 340 at 359, Rajah and Rajah (Pty) Ltd and Others v Ventersdorp Municipality and Others 1961 (4) SA 402 (A) at 407E-408A and Pepcor Retirement Fund and Another v Financial Services Board and Another 2003 (6) SA 38 (SCA), [2003] 3 All SA 21 at paras 13 and 24.

Even were mootness not in point in such a context, there would in any event be the consideration that judicial review is a discretionary remedy: ‘a court that is asked to set aside an invalid administrative act in proceedings for judicial review 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 minimizing injustice when legality and certainty collide’.39 Tender cases are notoriously problematic in the context of judicial review because, as here, by the time the challenge is heard and decided the impugned decision has often been implemented; cf. Moseme Road Construction CC and Others v King Civil Engineering Contractors (Pty) Ltd and Another 2010 (4) SA 359 (SCA), [2010] 3 All SA 549.40 In all matters in which administrative action is judicially reviewed the court is enjoined in terms of s 8(1) of PAJA to make any order that is just and equitable. This involves, as emphasised in Millennium Waste supra,41 at para 22, ‘a process of striking a balance between the applicant’s interests, on the one hand, and the interests of the respondents, on the other. It is impermissible for the court to confine itself..…to the interests of the one side only’. There is also the public interest.

A number of factors fall to be taken into consideration in this respect apart from the applicant’s interest in enforcing its right to having its unlawfully excluded tender considered. In a judicial review context the principle of legality is not applied in a vacuum; cf. Oudekraal Estates (Pty) Ltd v City of Cape Town and Others 2004 (6) SA 222 (SCA); [2004] 3 All SA 1, especially at para.s 36-38 and 46. The fourth respondent has incurred considerable expenditure in the purchase of a specially adapted heavy duty vehicle for the purpose of carrying out the contract, and has already been engaged in doing the work for several months. It has also taken on staff for the purpose and entered into contractual relationships with two small black economic empowerment partners in respect of the execution of aspects of the contract work. Some of the staff laid off by the applicant when it did not get the tender have since been re-employed in the context of the work currently being executed under the aegis of the fourth respondent. The service provided under the contract is an essential one and if the tender award were to be set aside the fourth respondent would be under no obligation to continue with the work in the period of three to four months, at a minimum, that would be required before a fresh tender process could be completed.

Although the fourth respondent has indicated its willingness to continue with the work in the interim in such an eventuality, any such continuation would require to be regularised in terms of an emergency contractual relationship to be entered into between itself and the City. (In my view the court is not able, as suggested by counsel for the City and the fourth respondent, to direct the fourth respondent to continue with the contract work pending the conclusion of a fresh tender process. The power in terms of s 8 of PAJA to make any order that would be just and equitable in the circumstances does not extend to making contracts

for the parties.) Moreover, the fourth respondent is an innocent party in the proceedings, having been in no way complicit in, or party to the irregularity that has been identified. The price at which the contract was awarded to the fourth respondent is by all indications a competitive one.42

Thus, assuming mootness or lack of prejudice is not in point, the less likely it might appear on the information before the court that an evaluation of the of the applicant’s bid would result in the award properly being made to it rather than to the fourth respondent, the less reason for the court to exercise its discretion in favour of making the order that the applicant seeks

setting aside the award. A factor that weighs in this respect is that the provisions of s 217 of the Constitution and the

derivative legislation described above are intended to operate at least as much in the public interest as they are in the interest of persons tendering to obtain or supply goods and services from or to organs of state.

The nature of the criticism directed by the applicant at the scoring of the data section of the functionality test in the tender necessitated an investigation of the detail of the information provided by the applicant and an assessment of the difference between its treatment in terms of the use of the spreadsheet scoring tool used by the City and what should have followed on an application of the intelligently evaluative approach which the applicant contended was required. That investigation identified what appeared to me to be anomalies in the data provided by the applicant, which would, it seemed, have been apparent as material had the committee undertaken an evaluative assessment of the costs breakdown data provided by the applicant in the manner that the SCM regulations and the arguments advanced by the applicant’s counsel would demand.

The number of sessions per vehicle per day given by the applicant – that is number of trips required between the serviced area and the waste disposal site (given as 2.5, which applicant’s counsel explained was an average when I queried how half a trip could be a sensible answer) - is demonstrably merely the arithmetical product of using one 11 ton capacity vehicle to

shift 560 tons of waste in a month (which according to the tender conditions is taken to comprise 4.33 five day weeks); alternatively,

the arithmetical product of using a vehicle with a ‘mass/volume restriction’ (i.e. capacity) of 11 tons to dispose of 26 tons in a day . It did not appear to tie in with the applicant’s business plan, which appeared to imply that three vehicles would be employed for the removal of wet waste. And while on the subject of the three vehicles, there was also a disparity between the indicated capacity of the vehicle to be acquired in the schedule to the applicant’s tender document (schedule 14 ‘Collection vehicles available for the contract’) and in the business plan submitted with the tender. In the schedule the capacity of the vehicle to be acquired is indicated as 15m³, and in the business plan as 12m³.

Furthermore, the so-called ‘Structure Data’, which required of a tenderer to indicate the number of staff entailed for the wet waste/conventional component of the tender contract, was completed by the applicant showing that there would be one operational manager, one supervisor, one driver per vehicle and four workers per vehicle. Under ‘Labour Costs’, an operational manager was costed at R8000, a supervisor at R6500, a driver at R5500 and a worker at R3250. A correlation of the ‘structure data’ with the ‘labour costs’ would arithmetically produce an amount of R33000 in respect of labour costs. The applicant, however, provided a figure of R51000 in respect of labour costs. If two additional vehicles were used, as suggested in other parts of the tender documentation, one would expect at least two additional drivers and eight additional workers to be involved. On the given data in respect of costing that would add R37000 to the anticipated figure of R33000 mentioned earlier, giving a total of R70000 per month, rather than the amount of R51000 stated in the tender.

Under ‘unit costs’, the applicant stated an amount of R32800 in respect of cost of vehicle licensing per year. In the context of the breakdown of provisional costs table, the data given would on its face suggest that the licensing cost related to a single vehicle, but read in the context of the tender as a whole it might well have been intended to relate to the licensing cost of three vehicles. Whichever reading is correct, the stated amount is impossible to square with the amount of R5500 given by the applicant under ‘monthly vehicle licensing costs’.

These are not unimportant anomalies because they impact on the calculation of the total monthly cost. As remarked earlier, it was the total monthly cost divided by the given number of monthly lifts (fixed by the City in the amount of 41564) that resulted in the rate per lift. As also mentioned, the rate per lift, or single unit rate, was the required manner of fixing the tendered price. The effect of these anomalies was therefore that the tendered price appeared to be the product of data which is irreconcilable because of the inconsistently formulated information offered in support of the applicant’s bid. It had the effect of artificially

lowering the ‘rate per lift’. The result was that the applicant’s tendered price was made to look more competitive

than the required inputs would suggest it should have been. Furthermore, an evidently understated tender price could also give

rise to reasonable concern by an evaluation committee as to the feasibility or financial sustainability of the bid.

It also appeared to me when reading the applicant’s bid document in the analytical manner enjoined by the argument of the applicant’s counsel, and in the manner that a bid evaluation committee would be required by the applicable legislation

and the tender specifications to do, that the apparent incongruence of the applicant’s quoted rate per lift with the supporting

information contained in its bid might, in terms of Note 5 to part 4 of the tender document (‘The Price Schedule’), have rendered the bid non-responsive or, to use the language of the PPPFA, not ‘acceptable’. Note 5 provided ‘The rates submitted on the Pricing Schedule must correspond with the Breakdown of Costs or the tender will be deemed non responsive.’ Construed in a manner consistent with the achievement of the requirements of s 217(1) of the Constitution, the

reconciliation of the rate on the pricing schedule with the breakdown of costs would have to be assessed in the context of the

tender as a whole, and not just on the basis of the product of the division of an incongruously computed total monthly costs figure by the number of monthly lifts involved as was the case in the applicant’s bid in the Atlantic tender; see also the conditions for tender compliance quoted in para. Error: Reference source not found, above.

The anomalies and discrepancies referred to were apparent on a reading of the applicant’s bid document, which was part of the evidence before the court. They were not, however, identified or traversed in the affidavits; nor, consequently, were they addressed in the parties’ oral submissions at the hearing. I became astute to them in the course of preparing judgment. I considered that if there were any substance in them they would be a material consideration in respect of weighing the issue of prejudice, alternatively, in the exercise of the court’s discretion in deciding upon an appropriate remedy in the context of the exclusion from consideration of the applicant’s bid on the basis of the irrational scoring of the functionality test.

I therefore caused a note to be addressed to the parties in which attention was directed to the anomalies which my consideration of the bid document had identified and written submissions were invited on the following questions (I quote verbatim from the note):

Assuming the anomalies or incongruences described above are correctly identified as such (as to which submissions are invited), how would note 5 to part 4 fall to be applied in the circumstances (having regard to s 2(1) of the PPPFA)?

In particular, does the note fall to be applied on a purely arithmetical assessment of the information provided in the table, or

does it fall to be applied on an evaluation of the information provided assessed in the context of a holistic consideration of the tenderer’s bid document, as the argument on behalf of the applicant at the hearing suggested? And does its application on either basis carry with it the result that the applicant’s tender had to be regarded as non responsive?

If the applicant’s tender fell to be regarded as non responsive by reason of the application of note 5 to part 4, and assuming that the court were also to hold that the bid evaluation committee’s scoring of the breakdown of costs criteria of the functionality

evaluation was irrational, what implications should such a conclusion hold for the determination of the review?

In the supplementary written argument provided by the applicant’s counsel in response to the note from the court it was submitted that note 5 to part 4 required no more of bidders than that they correctly transcribe their tendered unit price (or ‘rates

per lift’) from the Breakdown of Provisional Costs tables to the appropriate part of the price schedule. Thus a bidder who gave one rate per lift at the foot of table A1 (say, R10 per lift), but a different rate in item A of the price schedule (say, R9 per lift), would be non-responsive. In support of that argument counsel placed emphasis on the statement at the foot of the table requiring the rate per lift to be transferred to the pricing schedule.43 The argument proceeded that note 5 thus did not imply that for a bid to be responsive the rate given in the pricing schedule had to be arithmetically consistent with the individual costing inputs to a bidder’s tendered rate, or with the narrative information provided by the bidder in substantiation of the content of its tender. It was submitted that note 5 prescribed consistency, but only in the limited manner just described. It was contended that table A1 had not been designed with the clarity and precision that would be necessary if arithmetical consistency between all the numbers were required in order for a bid to cross the basic hurdle of responsiveness. The contention was illustrated by the observation that it was, for example, unclear whether the unit cost item ‘Cost of vehicle licensing per year’ is a single licensing cost for all vehicles, or the licensing cost for one vehicle.

The applicant’s counsel argued that their construction of the bid document was supported by the fact that the evaluation process made separate provision for the evaluation of the ‘Breakdown of Cost’ data, under the heading of ‘Functionality Scoring’ in clause 7.6, quoted above.44 They submitted that the figures inserted by a bidder in the breakdown of costs tables (for example table A1) accordingly fell to be evaluated only as part of the functionality scoring exercise, and that it would serve no purpose for note 5 of part 4 to

require duplication of the process, but with the possible fatal consequence of non-responsiveness in the event of an inconsistency,

instead of a scoring as part of the overall assessment of functionality.

The applicant’s answer to the specific question put in the court’s note as to whether note 5 to part 4 required a purely arithmetical assessment, limited to the information in table A1, or whether it required a more holistic evaluation of the content of the bids was therefore that neither requirement pertained; all that was needed was a correspondence between the figure given in the tables for ‘rate per lift’ or ‘rate per 3,5kg dry waste’ in the breakdown of costs

tables and the ‘unit rates’ given in respect of items A, B1 and B2 in the pricing schedule. The argument concluded that because the required correspondence was apparent on the applicant’s bid document the bid did not fall to be treated as non-responsive in terms of note 5.

The City’s counsel construed the reference in note 5 of part 4 of the tender to the ‘breakdown of costs’ in table A1 as a reference to the breakdown read as a whole and not just to the ‘rate per lift’. Accordingly, having regard

to the identified discrepancies, their contention was that had the evaluation committee not excluded the applicant’s bid as a consequence of having scored it below the required minimum of 60 points for functionality, it would, had it detected the

discrepancies, have treated the bid as non-responsive in terms of note 5. The City’s counsel also pointed out that note 5 speaks only of a correspondence between the tendered unit rate and the breakdown of costs and does not enjoin a consideration of figures or other information set out elsewhere in a tenderer’s bid.

The fourth respondent’s counsel argued that although note 5 to part 4 of the tender document could be interpreted in two ways, it would make ‘eminent sense’ to construe the note to require the tendered unit rate to correspond with the information provided in the breakdown of costs read in the context of the bid as a whole. In other words it would not be good enough for a bidder merely to transpose its rate per lift figure if the latter figure were not the product of, or consistent with all of the information supplied by the tenderer in substantiation of its monthly costs. The fourth respondent’s counsel contended that in the applicant’s bid, the unit rate transferred from table A1 did not correspond with the figures contained in the body of the ‘Breakdown of Costs’ table because at least two of the figures under the heading ‘Monthly Costs’, being the key inputs in the formula used to arrive at the rate per lift, were inconsistent with other parts of the ‘Breakdown of Costs’ table thereby resulting in the rate itself being irreconcilable with parts of the ‘Breakdown of Costs’ table. The fourth respondent’s counsel argued that this was the case whether one adopted a purely arithmetical assessment of the information provided in the table (one vehicle), or an evaluation of the information provided assessed in the context of a holistic consideration of the applicant’s bid document (three vehicles45). Their argument concluded therefore that on a proper application of note 5 to part 4 that the applicant’s bid was not an acceptable tender.

It is evident from the tender compliance specifications quoted in para. Error: Reference source not found, above, that the breakdown of costs given by a tenderer was intended to be a material consideration; not just for the purposes of scoring the

functionality test, but also in the substantive evaluation of the financial viability of the bid. Note 5 of part 4 of the bid document

falls to be construed with this in mind. The construction of the note contended for by the applicant does not bear scrutiny, or indeed make business sense. The applicant’s construction would also subvert the substance of the evaluative obligation imposed on the bid evaluation committee in terms of the SCM regulations; it would detract from, rather than promote, the achievement of the objects of s 217 of the Constitution. In the circumstances the construction contended for by the City is in substance to be preferred. I would not, however, accept the literalist limitation which the City ascribed to the note. In my view, consistently with general principle, the bid document falls to be read as a whole, and the content of the ‘breakdown of costs’ interpreted contextually, rather than in isolation. Thus were it apparent from the document read as a whole that the reference to a single vehicle being employed on the contract was a mistake and the number of vehicles actually to be used was otherwise clearly evident from the content of other parts of the document, the mistake would be appropriately accommodated in the evaluation of the breakdown of costs. One would, of course, expect in such a case that the monthly costs would reflect the cost of the actual number of vehicles to be used.

The applicant’s counsel argued that there were in fact no anomalies or inconsistencies in the information provided in the applicant’s bid. In this regard they qualified the arguments put forward in response to the court’s note by suggesting that it was ‘difficult to deal with that issue conclusively on the papers, as the issue was never raised by the City, or by [the fourth respondent], and thus [the applicant] was never called upon to justify its monthly cost amounts in its affidavits’. In my view there is a twofold answer to the qualification raised on the applicant’s behalf. The first is that it is was for the applicant to show the materiality of the exclusion of its tender from consideration by the City - in other words, that its bid had been an ‘acceptable tender’ within the meaning of the PPPFA, and that the applicant had therefore been prejudiced by its exclusion. Secondly, the tender document is a jural document which falls to be construed according to the established rules; cf. KPMG Chartered Accountants (SA) v Securefin Limited and another 2009 (4) SA 399 (SCA), [2009] 2 All SA 523, at para.s 39-40. The construction of the bid documentation is a matter of law. Evidence to determine its meaning is neither

required, nor permissible.46 That the issue was not raised by the City is contextually not surprising. The City’s approach to the review application was to seek to justify the exclusion of the tender on the ground that it had failed to muster the minimum score required in the functionality test to qualify for consideration. The City thus gave no consideration to how the bid should have been construed

or evaluated by the bid evaluation committee had it properly considered the bid. That does not serve as a bar to the court’s

consideration of the document. Indeed, as observed earlier, the contentions advanced in support of the applicant’s case in fact enjoined a consideration of the document.

Addressing the substance of the point regarding the apparent inconsistencies in the applicant’s bid, the applicant’s counsel argued that the anomalies were apparent rather than real and contended that this could be established with reference to the bid as a whole. With regard to the number of vehicles to be used, they submitted that the applicant had ‘evidently proceeded on the basis that two vehicles would be engaged full-time on conventional door-to-door collection, and that (as the business plan indicates) the third vehicle to be purchased would be a back-up vehicle, running when one of the other vehicles could not. The costs to [the applicant] of providing the relevant service, in relation to vehicle licensing and labour, were determined accordingly’.

The argument proceeded: ‘The monthly cost figures for labour and licensing thus do reconcile with the other relevant information in the table. The calculations are roughly as follows: The monthly labour cost figure comprises, on that scenario, the monthly costs of one operational manager (R8 000), one supervisor (R6 500), two drivers (2 x R5 500 = R11 000) and eight workers (8 x R3 250 = R26 000). The total is R51 500, which when rounded down produces the figure of R51 000

inserted in table A1. The cost of vehicle licensing per year was stated to be R32 800. For two vehicles, the amount is R65 600.

Spread over twelve months, the amount is R5 467. Rounded up, that amount reaches the tendered amount of R5 500’.

In my judgment the argument does not withstand scrutiny for a number of reasons. In the relevant part of the ‘business plan’ submitted as part of its tender, the applicant stated:

‘We would buy 1 x 12m³ compactor for wet refuse collection as a back up to our two compactor trucks and also to address a problem we noticed while we were doing this contract that trade need to serviced separately and early in the morning because they generate a lot of wet waste which leaks water with unpleasant smell as result residents in households complains a lot about this. It will also address at lot of other problems created by landfill closing times and it will reduce overtime cost which is almost a daily challenge in the area. It will be used in both Camps Bay and Hout Bay but it is mostly needed for Camps Bay small streets and cul-de-sac roads at the top e.g. Kloof Street. We will continue to use our two trucks for wet collection and hired truck until we receive our new truck in three month time.’

It would appear from this that the third vehicle would not be only ‘a back-up’, in the sense of one to be used only when one of the two vehicles already owned by the applicant was out of action. There were to be three vehicles actually employed. The one that was to be acquired would be used to service areas in Camps Bay and Hout Bay with special requirements, hence the reference, for example, to ‘Camps Bay small streets and cul-de-sac roads at the top e.g. Kloof Street’. It is evident that it was because of the recognised need, based on the applicant’s previous experience, for a third vehicle that the applicant indicated that it would hire a vehicle pending the expected delivery of the additional vehicle in three months’ time.

The tender specifications provided for areas in both Camps Bay and Hout Bay to be serviced on Mondays and Tuesdays and other areas in Hout Bay on Wednesdays, Thursdays and Fridays. The specified number of daily lifts on Mondays and Tuesdays (2235 and 2796, respectively) was materially higher than that specified for Thursdays and Fridays (1102 and 1280, respectively). The statement

in the business plan that the acquisition and employment of the third vehicle would ‘also address at (sic) lot of other problems created by landfill closing times and it will reduce overtime cost which is almost a daily challenge in the area’ affords confirmation of the daily constraints and challenges in the execution of the contract work which the introduction of a third vehicle was, ex facie the bid, intended to address. It is impossible to accept that the introduction of a smaller third vehicle (whether it be 12m³or 15m³) would reduce overtime costs if it were to serve only as a stand-in, when needed, for a larger capacity (19m³) one.

Moreover, on any approach, and even if, contrary to the tenor of the ‘business plan’, the third vehicle were indeed intended to fulfil a purely ‘back-up’ function, in the sense contended by the applicant’s counsel, it would still have to be licensed. In other words, for the applicant’s costing to read sensibly it would need to make provision for the licensing of three, not two, vehicles.

In the result I have concluded that were the applicant’s bid to have been evaluated by the bid evaluation committee, the committee would have been bound, in terms of note 5 to part 4 of the bid document, to treat it as ‘non-responsive’.

Weighed together with the factors mentioned earlier, this leaves me in no doubt that it would not be appropriate to set aside the tender award to the fourth respondent at the applicant’s instance.

Before moving on to deal with the application for the review of the Helderberg tender, it is appropriate to deal briefly with the point raised in the applicant’s supplementary argument about inconsistencies in the successful fourth respondent’s

bid, despite the fact that nothing had been made of these in the applicant’s papers. It was evident that the salary of a supervisor in the indicated sum of R12 500 per month had been omitted from the fourth respondent’s ‘labour costs’ item under the information supplied in respect of ‘Monthly Costs’ in table A1 of its bid document. In my view nothing turns on this. It would seem to follow that the supervisor’s salary had been included in the provision for ‘overheads’ or ‘miscellaneous’. Any doubts that the bid evaluation committee might have entertained in this regard could quite legitimately have been addressed by an appropriately formulated enquiry for clarification to the bidder.47 The position is quite distinguishable from that of the applicant’s bid, in which, on any approach, the indicated costs do not make sense because of the contradictory indications in the bid as to the number of vehicles to be employed in the execution

of the contract. The applicant also pointed to a discrepancy in table A1 of the fourth respondent’s bid concerning vehicle

licensing costs. The fourth respondent indicated that the annual licensing cost per vehicle to be used in the contract would be R18 033.54. As the applicant’s counsel indicated in their supplementary written argument, taking into account that the fourth respondent had indicated that it would use 2 vehicles to execute the contract work, one would expect the amount in respect of the ‘Vehicle licensing costs’ item under the ‘Monthly Costs’ section of the table to be R3 005.59.48 The amount actually inserted in respect of the item under ‘Monthly Costs’ was much higher: R8 197.57. In this regard also I do not consider the discrepancy to be material. Unlike the position with respect of the applicant’s bid, the discrepancy did not highlight an under-estimation of the tendered unit price. As touched

on earlier, in the course of dealing with the proper construction of note 5 to part 4 of the tender document, the point is not so much the mere existence of discrepancies, but their effect on the tender, evaluated in the context of a critical analysis of its content read as a whole.

The Helderberg Tender

The tenderers’ bids, as is usually the case, were formulated in the form of offers open for acceptance by the City as the intending service procurer. Part 2 of the bid document incorporated the so-called ‘Tender Offer’. It provided, amongst other matters, that the tenderer tendered ‘to supply all or any of the goods and/or render any of the services described in the attached document to the City of Cape Town (“CoCT”) on terms and conditions stipulated in this tender document and in accordance with the Specifications stipulated in this tender document at the prices reflected in the Form of Offer and Acceptance/Price Schedule’. Part 3 of the document, entitled ‘Form of Offer and Acceptance’ was formulated in such a manner as to enable a written contract on the terms and conditions of the bid to come into being upon the counter-signature thereof by an authorised representative of the City. According to the tenor of part 3, by its signature of the Form of Offer and Acceptance the City would signify its acceptance of the tenderer’s offer. Part 2 of the tender document reflected that the tenderer’s offer would remain ‘valid’ for a period of 120 days from the closing date of the tender. It is a requirement of the City’s SCMP that the period for which a bid is to remain ‘valid and binding’ must be indicated in the bid document.49

The City had not determined to whom to award the tender contract within the 120 day period for which the applicant’s tender was open for acceptance. However, consistently with clause 140 of the SCMP, two weeks before the expiry of the validity of the

applicant’s offer, the City had invited the applicant to extend the period during which its offer was open for acceptance.

The managing member of the applicant averred that he had no recollection of having received such an invitation. The documentary

evidence put in by the City supports its allegation that the invitation was indeed sent, and on the basis of the Plascon-Evans rule50 I am bound to accept its allegations in this regard for the purpose of determining the application. The applicant, differing in this respect from its conduct in regard to its bid in the Atlantic tender, did not extend its offer in the Helderberg tender. In the result the applicant’s bid had lapsed before the City had awarded the tender contract and by the time that the award was made was no longer open for acceptance.51 In the circumstances I consider that it is irrelevant that the bid evaluation committee had rejected the bid before it had lapsed. The bid adjudication committee could notionally have overruled the evaluation committee’s recommendation and the City’s accounting officer could also notionally have required the reconsideration of any recommendation by the adjudication committee, or indeed of the evaluation committee.52 It was necessary in order for the applicant to keep its legal interest in the award of the tender alive to have extended the period of the validity of its offer. In the context of it having allowed its offer to lapse, the applicant’s complaint against the treatment of its tender by the evaluation committee is moot. The terms of the offer incorporated the provisions of the SCMP by reference. Clause 141 of the SCMP provided that tenderers who failed to respond positively in writing and before the expiry of the original validity period to a written request to extend the validity of their bids would not be considered further in the bid evaluation process. Inasmuch as the applicant sought to make something of the conduct of the City in acting as if the applicant’s bid continued to be regarded as valid, I do not think this can avail it. It would have been unfair to the other tenderers for the City to have deviated from the prescriptions of the SCMP in favour of the applicant. The City therefore, for that reason too, could not competently have awarded the contract to the applicant after the validity of its bid had expired.

In the circumstances the application for the review and setting aside of the award of the Helderberg tender cannot succeed. It is therefore unnecessary to determine the merits of the applicant’s complaint in that matter, or the points in limine raised by the City and the fifth respondent predicated on the failure by the applicant to avail of an internal appeal (ostensibly in terms of s 62 of the Systems Act) and the alleged unreasonable delay by the applicant to institute the review proceedings.

Orders

The review applications have been unsuccessful. However, by reason of the fact that the applicant did establish that the disqualification of its bid in respect of the Atlantic tender was unlawful I am disinclined to make a costs order in favour of the City against the applicant in respect of that leg of the application. For the assistance of the taxing master I estimate that about 60 percent of the hearing was devoted to the Atlantic tender part of the case. Otherwise there is no reason why costs should not follow the result.

The following orders are made:

It is declared that the disqualification of the applicant’s bid from consideration in respect of the Atlantic tender on the basis of the bid evaluation committee’s scoring of the functionality eligibility test was unlawful.

Notwithstanding the declaration made in terms of paragraph 1, the applications for the review and setting aside of the awards of the tender contracts in the Atlantic and Helderberg tenders are dismissed.

Save that in respect of the application for the review and setting aside of the Atlantic tender there shall be no order as to costs as between the applicant and the first respondent, the applicant shall otherwise be liable for the costs of suit of the first, fourth and fifth respondents, including the costs of two counsel where such were employed.

A.G. BINNS-WARD

Judge of the High Court

1The second and third respondents are functionaries of the first respondent, whose only involvement in the matters germane to the

case was in the discharge of their responsibilities within the City’s administration. In my view it was unnecessary for them to have been individually cited as respondent parties in these proceedings. They played no separate role from the City in the proceedings.

2217 Procurement (1) When an organ of state in the national, provincial or local sphere of government, or any other institution identified in national legislation, contracts for goods or services, it must do so in accordance with a system which is fair, equitable, transparent,

competitive and cost-effective. (2) Subsection (1) does not prevent the organs of state or institutions referred to in that subsection from implementing a procurement policy providing for- (a) categories of preference in the allocation of contracts; and (b) the protection or advancement of persons, or categories of persons, disadvantaged by unfair discrimination. (3) National legislation must prescribe a framework within which the policy referred to in subsection (2) must be implemented.

2217 Procurement

(1) When an organ of state in the national, provincial or local sphere of government, or any other institution identified in national legislation, contracts for goods or services, it must do so in accordance with a system which is fair, equitable, transparent,

competitive and cost-effective.

(2) Subsection (1) does not prevent the organs of state or institutions referred to in that subsection from implementing a procurement policy providing for-

(a) categories of preference in the allocation of contracts; and

(b) the protection or advancement of persons, or categories of persons, disadvantaged by unfair discrimination.

(3) National legislation must prescribe a framework within which the policy referred to in subsection (2) must be implemented.

3Section 2(1)(b)(i) of the PPPFA.

4In terms of s 2(1)(d) of the PPPFA.

5Section 2(1)(f) of the PPPFA.

6See the definition of ‘acceptable tender’ in s 1 of the PPPFA. .

7Chairperson, Standing Tender Committee v JFE Sapela Electronics (Pty) Ltd 2008 (2) SA 638 (SCA), [2005] 4 All SA 487, at 644B-E.

8Millennium Waste Management (Pty) Ltd v Chairperson, Tender Board: Limpopo Province 2008 (2) SA 481 (SCA); 2008 (5) BCLR 508; [2008] 2 All SA 145 at para 18-19.

9See s 112 of the MFMA. The pertinent regulations, being the Municipal Supply Chain Management Regulations, were published under General Notice 868 in GG 27636, dated 30 May 2005 (‘the SCM regulations’). The SCM regulations prescribe the framework with which a municipality’s supply chain management policy is required to comply.

10The supply chain management policy (‘SCMP’) in place at the relevant time was that approved by the City’s municipal

council on 27 March 2008, as amended by the council on 23 February 2011. That policy has subsequently been replaced by a policy adopted on 8 December 2011. All references to the SCMP in this judgment are to the first mentioned policy.

11See reg. 12(1)(d) of the SCM regulations

12See also South African Post Office Ltd v Chairperson, Western Cape Provincial Tender Board, and others 2001 (2) SA 675 (C) at 685I-J.

13See reg. 26 of the SCM regulations.

14See reg 26 of the SCM regulations.

15See reg. 28(1)(a)(i) of the SCM regulations.

16See reg. 28(1)(b) of the SCM regulations.

17See reg. 28(2)(a) of the SCM regulations.

18Regulation 29(5) provides: (5)(a) If a bid adjudication committee decides to award a bid other than the one recommended by the bid evaluation committee, the bid adjudication committee must prior to awarding the bid- (i) check in respect of the preferred bidder whether that bidder's municipal rates and taxes and municipal service charges are not in arrears; and (ii) notify the accounting officer. (b) The accounting officer may- (i) after due consideration of the reasons for the deviation, ratify or reject the decision of the bid adjudication committee

referred to in paragraph (a); and (ii) if the decision of the bid adjudication committee is rejected, refer the decision of the adjudication committee back to that committee for reconsideration.

18Regulation 29(5) provides:

(5)(a) If a bid adjudication committee decides to award a bid other than the one recommended by the bid evaluation committee, the bid adjudication committee must prior to awarding the bid-

(i) check in respect of the preferred bidder whether that bidder's municipal rates and taxes and municipal service charges are not in arrears; and

(ii) notify the accounting officer.

(b) The accounting officer may-

(i) after due consideration of the reasons for the deviation, ratify or reject the decision of the bid adjudication committee

referred to in paragraph (a); and

(ii) if the decision of the bid adjudication committee is rejected, refer the decision of the adjudication committee back to that committee for reconsideration.

19See s 114 of the MFMA.

20See reg. 21 of the SCM regulations.

21See para. Error: Reference source not found, above.

22See cl. 1.27 of the SCMP.

23See clause 420 of the SCMP.

24See s 6(2)(h) of PAJA and Bato Star Fishing (Pty) Ltd v Minister of Environmental Affairs and others [2004] ZACC 15; 2004 (4) SA 490 (CC); 2004 (7) BCLR 687 at para. 44.

25See note 26 for an explanation of ‘conventional or wet collection’.

26Tenderers were invited to complete three such tables, each calling for similar information. Table A1 was in respect of what was termed ‘conventional or wet collection’ (‘conventional collection’ referred to the collection of waste deposited indiscriminately in wheelie bins, whereas ‘wet collection’ related to non-recyclable and organic waste deposited in wheelie bins - dry waste being sorted and collected separately), table B1 was in respect of ‘dry collection’ (i.e. collection of recyclable material) in clear plastic bags and table B2 in respect of ‘dry collection’ in separate wheelie bins to be provided for that purpose by the City.

27The applicant was awarded 30 out of 40 for ‘vehicles and resources’, 7.5 out of 10 for ‘business plan’ and 8 out of 10 for ‘experience’.

2841 564 lifts.

29An extrapolation of the information in note 6 to schedule 4 of the bid document allowed tenderers to calculate the number of working days in a month as 5 x 4.33; thus 1919 lifts per day is the product of 41564 lifts per month divided by (5 x 4.33).

30The fourth respondent’s total number of lifts per day was thus within a margin of less than one of the number indicated by the applicant.

31The absolutely correct result of the arithmetical calculation would have been 13,47 kg, but in the real world one would not expect the estimated average weight of waste per wheelie bin to be given with such miniscule finiteness. Indeed, the applicant would have qualified to have its kg’s/lift estimate accepted as a valid figure for scoring purposes if the figure given had been within a 10% tolerance either side of 13,5 kg; in other words a figure of anywhere between 13kg and 14 kg would have been acceptable (see para. 73 of the affidavit, jurat 17 May 2012, of PT Magubane, the chairperson of the bid evaluation committee).

32See para. [13], above.

33See para. [13], above.

34See para. [13], above.

35Note 12.

36That the bid evaluation committee did not, when it chose to, close its eyes to the significance of estimates given by tenderers who were the current service providers is demonstrated by the facts of the Helderberg tender. In that matter the applicant’s tender was regarded as non-responsive - despite the fact that in that case the applicant had completed both the ‘Kg’s/lift’ and ‘Tons/month’ items - because its ‘Kg’s/lift’ estimate was markedly less than that estimated by the then current supplier in the latter’s competing tender. The applicant’s business plan submitted as part of its tender in respect of the Atlantic tender reported that the applicant had been collecting 800 tons per month and made it apparent that its estimate of 560 tons in its tender for a new contract was premised on the expectation that that quantity of wet waste might be expected to decline as the amount of recycled (or ‘dry’) waste increased.

37An analysis of this aspect of the Tedcor tender highlights another unsatisfactory and confusing feature of the compilation of table A1 in the pro forma tender document compiled by the City. The table invites tenderers to insert the information on the basis of ‘unit cost based on the preferred vehicle type’, which would suggest that the words ‘per vehicle’ in some of the questions were tautologous. The rate per lift, however, can only be sensibly reconciled with the information in the ‘Breakdown of costs’ if it is an amount equal to the product of the aggregate of the monthly costs, irrespective of the number of vehicles involved, divided by the number of monthly lifts.

38Pharmaceutical Manufacturers Association of SA and another: In re Ex parte President of the Republic of South Africa and others [2000] ZACC 1; 2000 (2) SA 674 (CC); 2000 (3) BCLR 241 at para. 85. See also Affordable Medicines Trust and Others v Minister of Health and Others [2005] ZACC 3; 2006 (3) SA 247 (CC); 2005 (6) BCLR 529, at paras 74 – 75)

39Oudekraal Estates (Pty) Ltd v City of Cape Town and Others 2004 (6) SA 222 (SCA); [2004] 3 All SA 1, at para. 36

40See especially para.s 11-21.

41Note 8.

42An important feature in the current matter which distinguishes it from the Millennium Waste case is that the fourth respondent’s price is lower than that tendered by the applicant. In Millennium Waste the successful tenderer’s price was vastly higher than that offered in the bid of the applicant for judicial review. This was evidently considered by the SCA (see para. 29 of the judgment) to afford a powerful reason in the public interest to require the evaluation of the applicant’s unlawfully excluded tender notwithstanding that it was not evident to the court whether or not the tender had been an ‘acceptable tender’.

43See the table, which is set out in para. Error: Reference source not found, above.

44At para. .

45As will become apparent, the applicant contended that the intended use of two vehicles was in fact the relevant information to be extracted upon a proper reading of its bid document. See para. and following, below.

46The position is, in my view, contextually quite distinguishable from that which was the subject of the observation made in Minister of Land Affairs and Agriculture and others v D&F Wevell Trust and others 2008 (2) SA 184 (SCA) at para. 43 that as a matter of principle ‘the issues and averments in support of the parties' cases should appear clearly [from the affidavits]. A party cannot be expected to trawl through lengthy annexures to the opponent's affidavit and to speculate on the possible relevance of facts therein contained. Trial by ambush cannot be permitted’. That principle, as I understand it, pertains where the content of the documents on which counsel seek to rely in argument at the hearing - despite the passages concerned not having been identified in the affidavits - involve matters of evidence, which could have been addressed in further evidence by the parties had the arguments that counsel seek to advance been adumbrated in the relevant affidavits.

47Clause 6.7 of the tender document provided that ‘The CoCT may, after the closing date, request additional information or clarification of tenders in writing.’

48R18 033.54 x 2 / 2.

49See clause 138 of the SCMP.

50Plascon-Evans Paints Ltd v Van Riebeeck Paints (Pty) Ltd [1984] ZASCA 51; 1984 (3) SA 623 (A), at 634E – 635C.

51It is unnecessary to consider what the position might have been had the City nevertheless purported to accept the offer after the expiry of the 120 day period; cf. Manna v Lotter and another 2007 (4) SA 315 (C). The provisions of clause 2.2 of part 2 of the tender bid in any event make it evident that the stipulation that the offer was open for acceptance for 120 days was incorporated in the form of an agreement between the tenderer and the City to that effect. It is plain that the determination of the period of validity of the offer was not exclusively for the benefit of the offeror and amenable to waiver by the applicant.

52See SCM regulation 29(6).