THE SUPREME COURT OF APPEAL OF SOUTH AFRICA

JUDGMENT

Reportable
Case no: 401/2022

In the matter between:

VODACOM (PTY) LTD APPELLANT

and

KENNETH NKOSANA MAKATE FIRST RESPONDENT

SHAMEEL JOOSUB NO SECOND RESPONDENT

Neutral citation: Vodacom (Pty) Ltd v Makate and Another (Case no 401/2022)
[2024] ZASCA 14 (06 FEBRUARY 2023)
Coram: MOCUMIE, SCHIPPERS and MOTHLE JJA and NHLANGULELA and
KATHREE-SETILOANE AJJA
Heard: 09 May 2023
Delivered: This judgment was handed down electronically by circulation to the
parties’ legal representatives via e-mail, publication on the Supreme Court of Appeal
website and released to SAFLII. The date and time for hand-down are deemed to be
delivered on 06 February 2026.

Summary: Interpretation of the order of the Constitutional Court – assessment of

equitable compensation in a ‘special contract’ – test for reviewability of determination
of a Chief Executive Officer (CEO) as a deadlock breaker – test as formulated in
Bekker v RSA Factors (Bekker test) distinct from the test for reviewability under the
Promotion of Administrative Justice Act 3 of 2000 – requirements to satisfy the Bekker
 2

test – high court did not apply the second leg of the Bekker test – order of the high
court remitting the determination to the CEO with a list of directives not sought –
Bekker test amplified with the caveat in Dublin v Diner – Dean v Prince:
reasonableness, fairness and bona fides.
 3

___________________________________________________________________

ORDER
___________________________________________________________________
On appeal from: Gauteng Division of the High Court, Pretoria (Hughes J, sitting as
court of first instance):
1 Save to the extent set out below, the appeal is dismissed with costs including
those of three counsel where so employed.
2 The order of the high court is set aside and substituted with the following:
‘(a) The decision of the First Respondent delivered on 9 January 2019,
determining the compensation to be paid to the Applicant by the Second
Respondent, is reviewed and set aside;
(b) The decision referred to in paragraph 1 is substituted with a decision
that the applicant is entitled to be paid 5% – 7.5% of the total revenue of the
PCM product from March 2001 to date of judgment by the Second Respondent,
together with the mora interest thereon, alternatively interest in terms of Section
2A(5) of the Prescribed Rate of Interest Act, 55 of 1975 as amended, and that
the total revenue of the PCM product shall be that set out in Model 9A, 9B &
9BB submitted to the First Respondent by the Applicant (Annexure “NM30” –
“NM32” to the Supplementary Founding Affidavit)
(c) It is directed that first respondent represented by the second respondent
shall bear the costs of the negotiations referred to in the Constitutional Court
Judgment, which costs shall include:
(i) Drafting of the submissions;
(ii) Preparation for and the hearing before the first respondent;
(iii) Reservation, preparation and qualifying fees of experts, involved in the
negotiations and hearing on an attorney and own client scale.
(d) The costs of this application are to be paid, jointly and severally, by the
first and second respondents, the one absolving the other.’
 4

___________________________________________________________________

JUDGMENT
___________________________________________________________________

Mocumie JA (Mothle JA and Nhlangulela AJA concurring)

Introduction

[1] This is an appeal against the judgment and order of Gauteng Division of the
High Court, Pretoria (the high court). The order reads:
‘(1) The application to strike out is dismissed with no order as to costs.
(2) The determination by the CEO is referred back to the First Respondent who is obliged to
make a fresh determination with the following directives:
(a) The Applicant is entitled to be paid 5% of the total voice revenue generated from the PCM
product from March 2001 to March 2021 by the Second Respondent;
(b) That total voice revenue includes PCM revenue derived from prepaid, contract (both in
bundle and out bundle) and interconnect (MTR) fees as set out in the Second Respondent’s
annual financial statements as well as the information provided in Annexure 16(a)-16 (r)
produced by the Second Respondent (CL021-1 to CL021-21) and collated in Annexure NM29
(CL034-1 to CL034-2).
(3) The First Respondent must determine the annual effective rate, which effective rate should
be a blend between contract effective rate and prepaid effective rate, and in each case the
respective rates are not to be less than the public ICASA effective rate:
3.1. The First Respondent must assume that the average call duration of the return calls is 2
minutes;
3.2. For the purposes of the First Respondent’s determination it must not be less than the
published ICASA effective rate;
3.3. For the purposes of the First Respondent’s determination it must be assumed that the
PCM count in Model 9A is correct. Model 9A is to be found on NM30, (CL035-1 to CL035-8
and CL036-1);
3.4. The Applicant is entitled to 27% of the number of PCM’s sent daily as being revenue
generated by the return calls to the PCM;
(4) The Applicant is also entitled to the time value of money calculated at 5% for each
successive year that the Second Respondent owes to the Applicant and the capital amount or
annual portion thereof;
(5) That the First Respondent must finalize his determination within one month of this order;
 5

(6) Each party is to pay their own costs [in respect of] the negotiations referred to by the
Constitutional Court.
(7) The costs of this application are to be paid on a party-and-party scale, which costs shall
include the costs of two counsel.’

Factual matrix
[2] The long history of close to two decades between the appellant, Vodacom (Pty)
Ltd (Vodacom), and the first respondent, Mr Kenneth Nkosana Makate (Mr Makate),
over the latter’s invention of Please Call Me (PCM) is widely documented and needs
no repeating. For purposes of the determination of the core issue in the appeal, it is
necessary to restate the order issued by the Constitutional Court on 26 April 2016 (the
operative order), from which the second respondent, the CEO of Vodacom, Mr
Shameel Joosub (the CEO) derived his mandate. The order reads:
‘(a) It is declared that Vodacom (Pty) Ltd is bound by the agreement concluded with Mr
Kenneth Nkosana Makate and Mr Phillip Geissler.
(b) Vodacom is ordered to commence negotiations in good faith with Mr Makate for
determining a reasonable compensation payable to him in terms of the agreement.
(c) In the event of the parties failing to agree on the reasonable compensation, the matter must
be submitted to Vodacom’s Chief Executive Officer (‘CEO’) for determination of the amount
within a reasonable time.
(d) Vodacom is ordered to pay the costs of the action, including the costs of two counsel, if
applicable, and the costs of the expert, Mr Zatkovich.’ (Emphasis added.)
Although this order was intended to finally resolve Mr Makate’s woes, it appears not
to have achieved that purpose.

[3] The parties commenced with negotiations. When consensus on the amount of
compensation to Mr Makate could not be reached, para (c) of the operative order was
triggered. The determination of the CEO, in relevant part, reads:
‘10.25.1 2001 looking forward model: R51,5 million;
10.25.2 Employee model: R21,8 million;
10.25.3 TWL model: R38,1 million; and
10.25.4 Revenue share model: R42,2 million.

10.26 All models other than Model 2 produce a share of the revenue generated by the PCM
idea. Model 1 is what the CEO would have had to rely on back in 2001, although the charge
of 15c was never implemented. Model 3 produces a result based on the actual idea of PCM
 6

for TWL customers. Model 4, which is similar to what Makate has suggested, produces a share
in revenue based on incremental call revenue. I think it fair in all circumstances to average the
highest two models being 1 and 4 and that is what I have decided to do on a present value
basis.

10.27 I think it is clear from everything I have said that I found this to be a difficult exercise
with many imponderables. No doubt it may be said that in various respects I have erred by
being too generous or by being insufficiently generous. Nonetheless, my aim throughout has
been to arrive at a fair outcome overall and I truly believe that on balance I have done so to
the best of my abilities.’
The CEO accordingly awarded Mr Makate compensation for his PCM product in the
amount of R47 million.

[4] Dissatisfied, Mr Makate instituted review proceedings, in the high court, against
the amount of compensation that the CEO had determined Vodacom should pay him.
The high court found in favour of Mr Makate but remitted the determination of the
amount of compensation to be paid to Mr Makate to the CEO. This order was coupled
with certain directives which the CEO was required to follow in his redetermination.
Vodacom appeals against the judgment and order of the high court with the leave of
that court.

[5] In complying with the operative order, the CEO had called upon the parties to
make representations to him,1 which they did. In preparation for the submissions to
the CEO, Mr Makate requested various documents from Vodacom which were not
available to him. Vodacom provided some of these documents to Mr Makate. In its
representations, Vodacom contended that the operative order contemplated an
employee remuneration model based on international best practices. Mr Makate
rejected that basis of the assessment. He asserted that, in his understanding of the
order, the compensation had to be assessed by calculating a share of the actual
revenue generated by PCM, because, according to him, it was common cause that
the high court and the Constitutional Court had decided that the percentage will be 5%
share of Vodacom’s total revenue. He further asserted that to negotiate afresh would
undermine the decisions of the Constitutional Court and the high court, as the issue of

1 The parties were represented by counsel. The CEO, too, was assisted by counsel.
 7

sharing the revenue had been laid to rest by the Constitutional Court. And that any
attempt to re-negotiate indicated mala fides on the part of Vodacom.

[6] Subsequent to the negotiations, the parties reached a deadlock. Mr Makate

approached the Constitutional Court directly to clarify its order. He sought the following
order which expressly stated that:
‘(a) Vodacom was obliged to pay [him] a share of the revenue generated by the PCM; and

(b) that the precise share to be paid by Vodacom to [him] is to be determined through
negotiations or, if necessary, by the determination of the CEO.’
Mr Makate’s application was dismissed on the basis that it had no reasonable
prospects of success.

[7] The parties resumed the negotiations but there was no fruitful outcome. For a
two-week period in July 2017, Vodacom granted Mr Makate and his team access to
its data set2 for the preparation of his revenue share model. Thereafter, Mr Makate
made a settlement proposal of R20 billion (including interest capped at double the
capital sum). The figure was determined based on a formula multiplying the effective
rate per minute, PCM minutes of use, and the returned call conversion rate. The model
uses financial and other information which is publicly available and makes various
assumptions. In response to Mr Makate’s settlement proposal, Vodacom presented a
counter-offer of R10 million based on the salary and variable compensation paid to a
CEO of Vodacom in 2001, adjusting it with the time value of money to R7,7 million and
rounding it up to R10 million. These final negotiations failed and Mr Makate declared
a deadlock between the parties. Consequently, the parties, acting in terms of the
operative order, referred the impasse to the CEO who accepted his role and function
as that of a deadlock breaker.

[8] The parties accepted the CEO’s description of his role and function as a
deadlock breaker. According to Mr Makate’s legal representatives, the proceedings
before the CEO commenced on the understanding that the CEO’s role in the
determination is ‘. . . to do no more than act in an objective manner, relying on [his]
experience and applying [his] mind fairly and reasonably so as to ensure that [his]

2 Data set contained information on call, messaging and subscriber records including PCM messages.
 8

determination neither results in a manifestly unjust, nor a patently inequitable

outcome’.

[9] On 9 January 2019, after having considered the representations of the parties
as well as their supplementary submissions, the CEO delivered his determination
which was titled: ‘Determination of Reasonable Compensation due to Mr Kenneth
Nkosana Makate Arising from the Please Call Me Idea’. The award outlined the areas
of dispute and the specific issues which had been identified by the parties for
determination. It was itself, the product of extensive deliberation, negotiations and a
hearing of over two days. The witnesses included two experts on behalf of Mr Makate
namely, Mr Zatkovich and Mr Hendriks, both of whom were experienced in the field of
mobile technology and economy.

[10] Subsequent to receiving criticism from Mr Makate on how he arrived at the

figure of R47 million payable as compensation, the CEO filed an explanatory affidavit
in which he sought to further explain how and why he reached the decision he did.

[11] As alluded to, Mr Makate was aggrieved by the CEO’s determination and
instituted review proceedings against him. The relief sought by Mr Makate in the
review application was as follows:
‘1. The decision of the First Respondent delivered on 9 January 2019, determining the
compensation to be paid to the Applicant by the Second Respondent, is reviewed and set
aside;

2. The decision referred to in paragraph 1 is substituted with a decision that the applicant is
entitled to be paid 5% – 7.5% of the total revenue of the PCM product from March 2001 to
date of judgment by the Second Respondent, together with the mora interest thereon,
alternatively interest in terms of Section 2A(5) of the Prescribed Rate of Interest Act, 55 of
1975 as amended, and that the total revenue of the PCM product shall be that set out in Model
9A, 9B & 9BB submitted to the First Respondent by the Applicant (Annexure “NM30” – “NM32”
to the Supplementary Founding Affidavit);

3. In the alternative to paragraph 2 above:

3.1 The decision referred to in paragraph 1 is substituted with a decision that the Applicant
is entitled to be paid 5% - 7.5% of the total revenue of the PCM product from March
2001 to date of judgment by the Second Respondent, together with mora interest
 9

thereon, alternatively interest in terms of Section 2A(5) of the Prescribed Rate of

Interest Act, 55 of 1975 as amended;
3.2 It is directed that the total revenue of the PCM product shall be determined by this
Court, following such procedures as the Court deems in the interest of justice;

4. In the further alternative to paragraph 2 above:

4.1 The decision referred to in paragraph 1 is substituted with a decision that the Applicant
is entitled to be paid 5% - 7.5% of the total revenue of the PCM product from March
2001 to date of judgment by the Second Respondent, together with mora interest
thereon, alternatively interest in terms of Section 2A(5) of the Prescribed Rate of
Interest Act, 55 of 1975 as amended;
4.2 It is directed that the total revenue of the PCM product shall be determined by a new
referee, appointed by the Court;

5. It is directed that the Second Respondent shall bear the costs of the negotiations referred
to in the Constitutional Court Judgment, which costs shall include:
5.1 Drafting of the submissions;
5.2 Preparation for and the hearing before the First Respondent;
5.3 Reservation, preparation and qualifying fees of experts, involved in the negotiations
and hearing on an attorney and own client scale.

6. The costs of this application are to be paid, jointly and severally, by any Respondent
opposing this relief.’

Issue(s) for determination

[12] The issues for determination are the interpretation of the Constitutional Court
order, in relation to the CEO’s mandate and determination. Arising from that, this Court
must determine whether the high court was correct in granting the order it did and, in
particular, whether it was correct to impose directives on the CEO on, inter alia, the
exercise of his mandate and the duration of the contract period.

The law
[13] The standard of review applicable to the CEO’s determination is one articulated
some 40 years back in the judgment of Bekker v RSA Factors3 which postulates that,
where a third person is nominated to fix a price or make a valuation, such a person is

3 Bekker v RSA Factors 1983 (4) SA 568 (T).

 10

expected to exercise the judgment of a reasonable person.4 Courts have interpreted

mandates of this nature to mean that, whenever two parties have agreed on this type
of mandate, the decision is final and binding on them. However, should a valuer or
third person exercise their judgment in such a manner that it is unreasonable, irregular
and wrongly constituting an inequitable outcome, then the affected person would not
be bound by the valuer’s determination and such a determination would then be
subject to rectification on equitable grounds. This test was applied in Perdikis v
Jamieson5 where this Court stated the following:
‘It was held in Bekker v RSA Factors 1983 (4) SA 568 (T) that a valuation can be rectified on
equitable grounds where the valuer does not exercise the judgment of a reasonable man, that
is, his judgment is exercised unreasonably, irregularly or wrongly so as to lead to a patently
inequitable result.”
This is also the position in respect of the referee’s report – it can only be impugned on these
narrow grounds.’6

[14] The Bekker test, has its origin in the judgment of the English court in Dean v
Prince (Prince).7 This judgment, which has been well received in South African law
over the years, recognises the binding nature of an expert’s decision, but also
recognises that there might be some exceptions such as where the expert’s decision
is marred with unfairness and mala fides. In such cases, although the parties had
agreed that it will be binding on them, the decision may be challenged and set aside
on those bases. Importantly in Dublin v Diner (Dublin),8 the court stated with reference
to Prince as follows:
‘. . . “[E]ven if the Court cannot point to the actual error, nevertheless, if the figure itself is so
extravagantly large or so inadequately small that the only conclusion is that he must have
gone wrong somewhere, then the Court will interfere in much the same way as the Court of
Appeal will interfere with an award of damages if it is a wholly erroneous estimate.”

4 Ibid at 568.
5 Perdikis v Jamieson 2002 (6) SA 356 (W) (Perdikis) at 364H-I. The test was reaffirmed recently by
this Court in Lufuno Mphaphuli & Associates (Pty) Ltd v Andrews and Another [2007] SCA 143; [2008]
1 All SA 321 (SCA); 2008 (2) SA 448 (SCA); 2008 (7) BCLR 725 (SCA); Wright v Wright and Another
[2014] ZASCA 126; 2015 (1) SA 262 (SCA); Rajkumar Tahilram v Trustees of the Lukamber Trust and
Another [2021] ZASCA 173 (Tahilram) para 15.
6 Perdikis para 10.
7 Dean v Prince [1954] 1 All ER 749, [1954] Ch 409 (Prince).
8 Dublin v Diner 1964 (1) SA 799 (D) (Dublin) at 804G-H.
 11

Provided the extravagance or inadequacy of the determination is sufficiently gross, it seems

to me that the passage I have quoted substantially reflects our law too, save that I would add
that I should not be understood to suggest that circumstances which would justify interference
on appeal, in South Africa, with a trial Court's award of damages, would necessarily also justify
the setting aside of the determination of price, in a matter of this nature.’

Before this Court

[15] Vodacom contended that the CEO exercised a wide discretion when he
embarked on his Constitutional Court mandate. This meant that the CEO drew on his
own knowledge, expertise and research. In addition to this, he could consult anyone
he wanted to – in line with the Bekker test – as he did, ie assisted by counsel during
this exercise. Counsel for Vodacom submitted that, because the exercise does not fall
under the Promotion of Administrative Justice Act 3 of 2000, the high court was not at
large to second-guess the CEO, nor could it give him the directives it gave in its order
of remittal for reconsideration because of the limited scope in which courts are
permitted to interfere with this kind of mandate.9 Neither did the parties, so it
contended, request the high court to impose directives on the CEO. All that the high
court was required to determine was whether the CEO made a determination and
whether such determination was manifestly inequitable.

[16] On the question of whether the determination was manifestly inequitable,

counsel for Vodacom contended that although the high court adopted the Bekker test,
it did not proceed to consider whether the CEO’s determination on the amount of R47
million was patently inequitable. Counsel for Mr Makate agreed that the applicable test
was the Bekker test, the caveat being the test as espoused in Dublin10 with reference
to the English case of Prince,11 that the CEO, should exercise his mandate in line with
the standard of a reasonable man, preceded by negotiations conducted with bona
fides. If those principles were not met, then Mr Makate was entitled to challenge the
determination. To discharge that onus, Mr Makate had to show that the CEO used
incorrect information leading to an incorrect evaluation. Counsel for Mr Makate argued
that the Bekker principles do not fully cover the situation at hand, as Mr Makate’s case

9 Tahilram para 27.

10 Dublin at 804D.
11 Prince.
 12

is based on lack of bona fides on the part of the CEO and the patently unfair results
which followed – which triggers the exception espoused in Prince.

[17] On this score, counsel for Mr Makate pointed out what they perceived to be
anomalies in the determination of the CEO. Suffice to state at this stage that it was
common cause between the parties that Mr Makate’s claim was for financial
compensation, in return for the commercial use of his invention by Vodacom. It was
also common cause that the agreed rate for the determination of the compensation for
Mr Makate was 5% of the share in the accrued revenue generated by PCM.

[18] Having listened to the submissions of parties and their experts on what the fair,
reasonable and equitable compensation should be, the CEO embarked on his own
exercise depicted as Table 1 in his determination, with no recourse to what the parties
had agreed to. Based on that exercise, the CEO suggested that the basic estimates
Mr Makate presented need to be carefully examined, without saying why. I point out
only three examples (amongst many) in the determination to highlight the errors which,
according to counsel for Mr Makate, the CEO could not refute. These are:
(a) the omission of MTR;12
(b) the arbitrary 70% reduction the CEO introduced; and
(c) the change of heart on the agreed upon rate of 5% for the determination of Mr
Makate’s compensation.

Omission of MTR and revenue derived from PCM in relation to contract

subscribers
[19] With reference to Table A, counsel for Mr Makate submitted that the CEO
simply ignored certain income constituting voice revenue earned by Vodacom. He
explained that during argument in the high court, Vodacom avoided addressing the
omission of MTR and contract revenue derived from PCM from the calculations in
relation to contract subscribers from the calculations. It also did not raise any defence
in this regard. He argued that the CEO could not give any tangible reason why a patent
error of this magnitude was omitted or ignored. Before this Court, the least that counsel

12 ‘MTR’ is the abbreviation for Mobile Termination Rates.

 13

for Vodacom could do in this regard, was to concede the error. He, however, submitted
that the error was not relevant to the determination.

The arbitrary 70% reduction that the CEO introduced

[20] In his determination, the CEO, as the parties agreed, accepted that the one-
hour window translated to 27% of PCM success rate. However, he thereafter without
any reference to the parties, deducted a further 70%. In doing so, he reduced the PCM
response rate to 30% of 27% which translates to only 8.1%, a percentage none of the
parties proposed or tabled during the interaction with him. The CEO provides no
answer for this in his reasons for the determination.

Setting 5% of the share in the accrued revenue as the appropriate revenue share
[21] It was common cause between the parties that the agreed rate for the
determination of the compensation was 5% of the share in the accrued revenue. If this
is accepted, as counsel for Vodacom conceded, that resolves the issue of whether the
high court should have applied the test as formulated in Plascon-Evans Paints (TVL)
Ltd. v Van Riebeck Paints (Pty) Ltd (the Plascon Evans test).13 Thus, it must then be
accepted that whatever dispute there was, fell by the wayside as the issues were
crystallised and as such, the high court was not barred from proceeding with the
matter, on this basis. Therefore, on the approach I have adopted (on the basis of the
commonly agreed rate which was premised on 5% of the share in the revenue
generated by Vodacom and as it accrued over the years), the application of the
Plascon Evans test does not come into play at all. The submission by counsel for
Vodacom, to the contrary, is not consistent with what has already been decided by the
courts prior to this appeal.

Application of the law to the facts

[22] I accept that the high court was correct, as conceded by counsel for Vodacom,
that the CEO’s determination was not reasonable. However, even if the high court was
correct on the first leg of the Bekker test, it did not undertake the second leg of the
enquiry to consider whether the CEO’s determination of R47 million was manifestly

13Plascon-Evans Paints (TVL) Ltd. v Van Riebeck Paints (Pty) Ltd. 1984 (3) SA 620; [1984] 2 All SA
366 (A); 1984 (3) SA 623. This test was restated and applied in National Director of Public Prosecutions
v Zuma [2009] ZASCA 1; 2009 (2) SA 277 (SCA); [2009] 2 All SA 243.
 14

inequitable. For that reason, this judgment will focus only on the CEO’s determination
of reasonable compensation for Mr Makate, ie whether the R47 million is manifestly
inequitable. Linked to that enquiry, is his choice to peg the amount to a fixed term of
five years. I consider these two issues to be dispositive of the appeal before this Court.

Was R47 million manifestly inequitable?

[23] As found by the high court, counsel for Vodacom accepted that the determinant
percentage was 5% of the revenue generated by Vodacom. The CEO, in his
determination, also accepted this fact. Counsel for Vodacom submitted that R47
million was a generous amount, and since the CEO dutifully accounted for his
determination based on his knowledge of over 28 years with Vodacom, it cannot be
said that such an award is unfair and manifestly inequitable. Counsel for Vodacom
emphasised the various unpredictable variables which the CEO reflected upon in his
determination, and that the determination is a forward looking one and in terms of the
law of contract based on specific performance, which Mr Makate claimed from
Vodacom. Having regard to this, he argued that, in the absence of any attack on the
CEO’s honesty in the exercise of his mandate, there was no basis to challenge the
determination ie the 5% of the revenue determined from 2001 looking forward. He
strongly asserted that the amount proposed by Mr Makate was ‘outrageous’ and that
the CEO was generous to go over the maximum contractual period of three years
(which is the term generally applied for contracts with third parties) to grant Mr Makate
a five-year contractual period.

[24] To the contrary, counsel for Mr Makate argued that the approach which the
CEO adopted in making the determination was a superficial reconciliation, as he failed
to properly identify the references which he took into account to compute the R47
million. It was also argued that the CEO ignored information, which was both in his
own domain and in the public domain, on the revenue Vodacom raked in since 2001
to 2021, which included that generated by PCM. Counsel for Mr Makate pointed to the
anomalies in the table of comparison the CEO made to show various discrepancies.
On this item alone, it was argued that the CEO’s determination was pointedly to Mr
Makate’s disadvantage, considering that Vodacom has continued to benefit from the
PCM invention. It was also submitted, on behalf of Mr Makate, that for Mr Makate to
say that the compensation of R47 million was manifestly inequitable, it was not
 15

necessary to prove the exact figures (even if they are available to Vodacom despite
its denials) or financial prejudice. Logically, so it was argued, if Vodacom has
generated revenue from the day of the invention to date, surely Mr Makate was entitled
to benefit more than the R47 million the CEO considered appropriate compensation
for his invention. On that basis alone, R47 million was manifestly inequitable.14

The duration of the contract

[25] The issue of whether R47 million is manifestly inequitable must be considered
in the context of the duration of the contract for which Mr Makate is entitled to be
compensated as determined by the CEO. His decision must be assessed on the
information presented by Mr Makate (extracted from the documents supplied by
Vodacom) which showed the revenue generated by Vodacom from Mr Makate’s
invention between 2001 and 2021. Vodacom presented no evidence to refute or
contradict this information.

[26] In my view, what Vodacom seems to propose is to ‘unscramble the scrambled

eggs’. They were not happy with how the Constitutional Court finally dealt with the
matter on the 5% of the total ‘revenue generated’ basis. Thus, when the CEO
considered the computation, he instinctively or by design (with his knowledge and
expertise) fell back to Vodacom’s original stance of sharing on a 5% profit basis,
contrary to the operative order. Intractably linked to this 5% profit is the duration of the
contract.

[27] The CEO’s justification for capping the compensation at R47 million is premised
on a five-year contract. He based this on the three-year standard contract which was
generally the maximum period for so-called third-party service providers such as
Wireless Application Service Provider (WASP) and iBurst.15 In the CEO’s own words,
he was ‘generous’ with Mr Makate by stretching the three years to five years. Counsel
for Mr Makate pointed out that there are similar service providers such as WASP and
iBurst that are still contracted to Vodacom, for over 18-22 years, and those third-party
service providers are being paid millions based on the revenue their products

14Tahilram para 34.

15iBurst is an international mobile broadband internet manufactured by Kyocera used in Africa by
various companies including Vodacom.
 16

generated for Vodacom. In response, counsel for Vodacom contended that Mr Makate
is not exactly in the same category as those two service providers, as they had added
technical expertise to their products and their products were complete when they were
presented to Vodacom. Yet this comparison was never in the equation when the
negotiations commenced and the contract was concluded between the parties.

[28] In his reasons, the CEO went into fine details to show the extent to which Mr
Makate’s PCM product was not in the category of those third-party service providers.
Counsel for Vodacom contended, in this regard, that whether the CEO compared the
various products as Mr Makate wanted him to do, is not the issue. This is because the
CEO gave full reasons why he did so, and Mr Makate did not raise any objection to
this. However, this, so it was contended, does not answer the question at hand: If
Vodacom is still using and generating revenue from Mr Makate’s invention (as it does
from WASP and iBurst) why would the CEO, looking back to 2001 or forward from
2001, have decided not to continue with Mr Makate’s product beyond five years into
2018/2021? This is an important question, as Mr Makate’s invention is still generating
income for Vodacom some 18-20 years later. This appears to be common cause
between the parties. Furthermore, the sole reason for any invention to be retained by
Vodacom, for over 20 years, is that it continues to generate revenue. In the
circumstances, it would have been eminently reasonable, sensible and ‘business-like’
for Vodacom to have extended Mr Makate’s contract to a period beyond five years.
There are no concrete reasons advanced in the CEO’s determination in relation to why
it was not possible to extend the contract with Mr Makate to beyond a period of five
years.

[29] The CEO’s motivation for adhering to the five-year contractual period is not
borne out by the reasons he had provided in the original determination, nor by those
in his supplementary reasons. The reasons which the CEO has provided, do not
answer the real question which relates to what the CEO (of over 28 years with his
knowledge and expertise) compared Mr Makate’s PCM invention to? Nothing, in my
considered view. In light of the evidence of more than 20 years as opposed to a
hypothesis of five years, I am of the view that the CEO’s determination is flawed and
thus inequitable.
 17

[30] As this Court affirmed in Firestone South Africa (Pty) Ltd v Genticuro AG16 and
a plethora of subsequent decisions,17 the starting point in the interpretation of a court
order is to determine its manifest purpose. In interpreting a judgment or order, the
court’s intention is to be ascertained from the language of the judgment or order in
accordance with the usual well-known rules relating to the interpretation of the
documents. As in the case of a document, the judgment or order, and the court’s
reasons for giving it, must be read as a whole in order to ascertain its intention. It
follows from this, that the order of the Constitutional Court must be read with the
reasons for judgment.

[31] At para 5 of the Constitutional Court judgment, Jafta J writing for the majority
stated:
‘The applicant and Mr Geissler negotiated and agreed that Vodacom would use the applicant’s
idea to develop a new product which would be put on trial for commercial viability. If the product
was successful, then the applicant would be paid a share in the revenue generated by it.
Although the applicant had indicated that he wanted 15% of the revenue, the parties deferred
their negotiations on the amount to be paid to the applicant for a later date. However, they
agreed that in the event of them failing to agree on the amount, Vodacom’s Chief Executive
Officer (CEO) would determine the amount.’ 18 (Emphasis added.)

[32] Before proceeding to deal with the interpretation of the operative order, I wish
to deal with the submission made, on behalf of Vodacom, concerning whether the
agreement between Mr Makate and Vodacom was a revenue-share agreement.
Vodacom submitted that in determining this appeal, this Court must take into account
that Mr Makate made an application to the Constitutional Court to vary its own order
(the operative order), which was dismissed on grounds that there were no prospects
of success. What this means, argued counsel for Vodacom, is that the Constitutional
Court did not consider Mr Makate to be entitled to an ongoing share of the revenue
generated by Vodacom from the PCM invention. In response, counsel for Mr Makate

16 Firestone South Africa (Pty) Ltd v Genticuro AG 1977(4) SA 298 (A) at 304D-F.
17 See: Administrator, Cape and Another v Ntshwaqela and Others 1990 (1) SA 705 (A) at 715F-H;
Weber-Stephen Products Co v Alrite Engineering (Pty) Ltd and Others 1992 (2) SA 489 (A) at 494E-G;
Zurich Insurance Co South Africa LTD v Gauteng Provincial Government ]2022] ZASCA 127; 2023 (1)
SA 447 (SCA) para 5; Capitec Bank Holdings Limited and Another v Coral Lagoon Investments 194
(Pty) Ltd and Others [2021] ZASCA 99; [2021] 3 All SA 647 (SCA) para 25.
18 Makate v Vodacom (Pty) Ltd [2016] ZACC 13; 2016 (6) BCLR 709 (CC); 2016 (4) SA 121 (CC).
 18

submitted that this contention is not sustainable, as there is a myriad of reasons why
a court would dismiss an application for leave to appeal. As long as the reasons have
not been provided, as in this case, the order of the Constitutional Court dismissing Mr
Makate’s application to vary its order cannot be used against Mr Makate. I agree that
nothing turns on the order of the Constitutional Court dismissing the application for
leave to appeal, as it did not provide Mr Makate with reasons for why his application
had no prospects of success.

[33] I now turn to the interpretation of the operative order. On the common cause
facts, and on consideration of the language, context and the purpose of the operative
court order, there can be no doubt that the Constitutional Court intended for the term
‘compensation’ to refer to Mr Makate’s claim for payment from Vodacom for his
invention, within ‘a special contract’ between the parties, outside the context of a
traditional employer-employee relationship. Thus, it follows that the parties envisaged
that Mr Makate will be compensated for his invention based on the duration of the
contract, which has continued to generate revenue for Vodacom to date. This view is
underscored by the fact that, at one stage, Vodacom wanted to buy Mr Makate out.
This on its own, reinforces the significance of the product and the revenue it generated
between 2001 and 2021 for Vodacom, as established on the very information which it
provided Mr Makate with. This position motivated Mr Makate to compromise and make
an offer of R28 billion, based on what was submitted to the CEO as Model 9A, 9B &
9BB to represent reasonable compensation for the past 18 years.

[34] The question remains, having found that even on the second leg of the Bekker
test, the CEO was wrong, what should an appropriate remedy be? On a conspectus
of the evidence presented including expert evidence, I could find no objection by
Vodacom to the models presented by Mr Makate which point to a compensation similar
to a third-party service provider, except that it was not countenanced in international
practice within employer-employee relations; a reason which I reject. Over and above
that, there is no basis laid out for why the CEO preferred his own model (profit basis)
which he crafted with no reference or recourse to the parties.

[35] The litigation between Mr Makate and Vodacom has been going on for close to
two decades without resolution and the costs alone have been enormous for both
 19

parties. To remit the determination to the CEO who in his own words, made the
determination in ‘a fair outcome overall and . . . on a balance [he has] done so to the
best of [his] abilities’ would yield no results. If anything, it may yield worse results for
any of the parties because, at this stage, as a deadlock breaker, he started on the
wrong premise by rejecting the 5% which the parties had agreed to.

The remedy
[36] The high court was correct in its legal conclusion that the CEO acted contrary
to his Constitutional Court mandate. However, as alluded to, it did not undertake the
second leg of the Bekker test. What compounds the matter is that the remedy as
crafted, is not what Mr Makate sought because, as is evidenced from the amended
notice of motion delivered on 9 January 2019; Mr Makate sought that the determination
of the CEO be set aside and substituted; instead of remitting it for reconsideration to
the CEO. To that extent, it erred. It follows therefore, that this Court is at large to
interfere with the order of the high court.

[37] As I see it, the high court ought to have set aside the order of the CEO and
substituted it with its own decision. That is what was sought by Mr Makate in his notice
of motion. All the necessary evidence was before the high court for it to adopt a robust
approach and substitute the CEO’s decision with its own decision.19 There was
therefore, no need for the high court to have referred the matter to CEO for
redetermination with directives. Significantly, the parties presented expert evidence.
Having considered the expert evidence presented, as guided by this Court,20 absent
any evidence that Mr Makate’s computation is wrong (save that it does not take some
variables into account), I can find no reason why Mr Makate’s computation should not
be accepted as correct. Particularly because the issue, on computation of the

19 This Court and other courts have adopted the robust approach in similar cases: Hersman v Shapiro
& Co 1926 TPD 367 at 380; Lazarus v Rand Steam Laundries (1946) Pty Ltd 1952 (3) SA 49 (A); Esso
Standard SA (Pty) Ltd v Katz 1981 SA 964 (A); Marvanic Developments (Pty) Ltd v Minister of Safety
and Security [2009] ZAGPJHC 109; Hack Stupel & Ross Attorneys v Kgang [2007] ZASCA 132.
20 In NSS obo AS v MEC for Health, Eastern Cape Province [2023] ZASCA 41 para 25; MEC for Health

and Social Development, Gauteng v MM on behalf of OM [2021] ZASCA 128 para 17 this Court held:
‘it is trite that the opinion of an expert does not bind a court. It does no more than assist a court to itself
arrive at an informed opinion in an area where it has little or no knowledge due to the specialised field
of knowledge bearing on the issues. Expert opinion should not replace the function of the court to assess
the evidence and thereafter, make its own decision on which evidence to accept, and provide reasons
for the preference of one opinion of one expert above the other.’ See also Bee v Road Accident Fund
[2018] ZASCA 52; 2018 (4) SA 366 (SCA) para 66.
 20

compensation, was fully ventilated.21 It is for the above reasons that I find that the
order of the high court should have been to uphold Mr Makate’s application, but on the
basis expounded in the preceding paragraphs – in line with the Bekker test with the
Dublin and Prince caveat, without remitting the determination to the CEO with a list of
directives. For these reasons, the appeal against the order of the high court falls to be
dismissed to the extent set out in the order below.

Costs
[38] It behoves this Court to show its displeasure in the manner in which this
protracted litigation has been conducted over close to two decades, with Vodacom not
playing open cards with Mr Makate on the necessary information until ordered to do
so by the CEO significantly later in the hearing, to some extent, after pre-hearing
negotiations had failed. This delay can safely be laid at the doorstep of Vodacom. I
therefore find that Vodacom should bear the costs of the appeal.

[39] In the result, the following order issues.

1 Save to the extent set out below, the appeal is dismissed with costs including
those of two counsel where so employed.

2 The order of the high court is set aside and substituted with the following:
‘(a) The decision of the First Respondent delivered on 9 January 2019,
determining the compensation to be paid to the Applicant by the Second
Respondent, is reviewed and set aside;
(b) The decision referred to in paragraph 1 is substituted with a decision that
the applicant is entitled to be paid 5% – 7.5% of the total revenue of the PCM
product from March 2001 to date of judgment by the Second Respondent,
together with the mora interest thereon, alternatively interest in terms of Section
2A(5) of the Prescribed Rate of Interest Act, 55 of 1975 as amended, and that
the total revenue of the PCM product shall be that set out in Model 9A, 9B &
9BB submitted to the First Respondent by the Applicant (Annexure “NM30” –
“NM32” to the Supplementary Founding Affidavit)

21This Court and other courts have adopted the robust approach in similar cases: Hersman v Shapiro
& Co 1926 TPD 367 at 380; Lazarus v Rand Steam Laundries (1946) Pty Ltd 1952 (3) SA 49 (A); Esso
Standard SA (Pty) Ltd v Katz 1981 SA 964 (A); Marvanic Developments (Pty) Ltd v Minister of Safety
and Security [2009] ZAGPJHC 109; Hack Stupel & Ross Attorneys v Kgang [2007] ZASCA 132.
 21

(c) It is directed that first respondent represented by the second respondent

shall bear the costs of the negotiations referred to in the Constitutional Court
Judgment, which costs shall include:
(i) Drafting of the submissions;
(ii) Preparation for and the hearing before the first respondent;
(iii) Reservation, preparation and qualifying fees of experts, involved in the
negotiations and hearing on an attorney and own client scale.
(d) The costs of this application are to be paid, jointly and severally, by the
first and second respondents, the one absolving the other.’

________________________
B C MOCUMIE
JUDGE OF APPEAL

Schippers JA (Kathree-Setiloane AJA concurring)

[40] I have had the benefit of reading the first judgment by my colleague, Mocumie
JA. I am unable to agree with the conclusion that the high court ‘ought to have set
aside the conclusion of the CEO and substituted it with its own decision’. This
conclusion, respectfully, is at odds with the order of the Constitutional Court that in the
event of the parties failing to agree on the compensation payable to Mr Makate for his
PCM idea, the matter should be determined by Vodacom’s CEO. It follows that I adopt
the abbreviations used in the first judgment.

The basic facts

[41] Mr Makate worked for Vodacom as a trainee accountant. In 2000 he came up
with a brilliant idea: a cellphone user with no airtime could send a request to another
user with airtime, to call the former. Based on this idea Vodacom developed the PCM
product. It turned out to be a resounding success. However, Vodacom refused to pay
Mr Makate anything for his idea. In 2008 he sued Vodacom in the high court, inter alia,
 22

for a declaratory order that the parties had entered into an oral agreement for
Vodacom’s use of PCM, and a share of the revenue derived from it. The case went to
trial and the high court (Coppin J) dismissed his claim.22

[42] The trial court refused leave to appeal. Mr Makate’s application for leave to
appeal the trial court’s order to this Court suffered the same fate. However, his
application for leave to appeal to the Constitutional Court succeeded. On 26 April 2016
it granted Mr Makate leave to appeal and upheld the appeal with costs. The
Constitutional Court issued an order declaring that Vodacom is bound by the
agreement concluded by Mr Makate and Mr Philip Geissler, Vodacom’s Director of
Product Development; and directed the parties to negotiate in good faith to determine
an amount of reasonable compensation payable to Mr Makate, failing which
Vodacom’s CEO was required to determine that amount within a reasonable time
(Constitutional Court’s 2016 order).

[43] In May 2016, following the Constitutional Court’s 2016 order, the parties
commenced negotiations to determine the compensation payable to Mr Makate.
Vodacom proposed that Mr Makate’s share of the revenue generated by the PCM idea
be assessed on the basis of international standards for rewarding employees who
make significant contributions to their company. Vodacom offered Mr Makate R10
million as compensation for the PCM idea. Mr Makate rejected this offer on the ground
that it was inconsistent with the agreement between the parties. He claimed a share
of the actual revenue generated by the PCM idea – between R28 and R110 billion.
He proposed that in calculating that share, he should be regarded as a third party
supplier to Vodacom; that the parties agree on the total revenue generated by the
PCM product; and that they agree on the percentage of that revenue to which he was
entitled, and for what period. The parties could not agree on an amount of reasonable
compensation.

[44] In September 2016 Mr Makate unilaterally suspended negotiations with

Vodacom and applied to the Constitutional Court for the variation of its order of 26 April
2016. He sought an order declaring that Vodacom pay him ‘a share in the revenue

22 Makate v Vodacom (Pty) Ltd [2011] ZAGPJHC 241; 2014 (1) SA 191 (GJ).
 23

generated by the Please Call Me product’; that the precise share in that revenue be
determined by the CEO; and that Vodacom disclose what revenue was earned by
PCM, failing which it should grant his experts access to its systems and records to
determine the extent of PCM revenue (second Constitutional Court application).

[45] Vodacom opposed the second Constitutional Court application. On 8 February

2017 the Constitutional Court dismissed that application on the ground that ‘it bears
no prospects of success’ (2017 order). The matter then proceeded before the CEO to
determine the compensation payable to Mr Makate.

The CEO’s determination

[46] The CEO has held that office since 2012 and has been employed by Vodacom
in various senior positions since 1994. In the process of determining the compensation
payable to Mr Makate, he was assisted by senior and junior counsel who gave him
legal advice where necessary.

[47] The CEO derived his mandate from the Constitutional Court’s 2016 order. It
reads:
‘(a) It is declared that Vodacom (Pty) Limited is bound by the agreement concluded by Mr
Kenneth Nkosana Makate and Mr Philip Geissler.
(b) Vodacom is ordered to commence negotiations in good faith with Mr Kenneth Nkosana
Makate for determining a reasonable compensation payable to him in terms of the
agreement.
(c) In the event of the parties failing to agree on the reasonable compensation, the matter
must be submitted to Vodacom’s Chief Executive Officer for determination of the
amount within a reasonable time.’23

[48] The Constitutional Court described the agreement referred to in its order of 26
April 2016, as follows:
‘The applicant and Mr Geissler negotiated and agreed that Vodacom would use the applicant’s
idea to develop a new product which would be put on trial for commercial viability. If the product
was successful then the applicant would be paid a share in the revenue generated by it.
Although the applicant had indicated he wanted 15% of the revenue, the parties deferred their

23 Makate v Vodacom (Pty) Ltd [2016] ZACC 13; 2016 (4) SA 121 (CC) para 107.
 24

negotiations on the amount to be paid to the applicant for a later date. However, they agreed
that in the event of them failing to agree on the amount, Vodacom’s Chief Executive Officer
(CEO) would determine the amount.’24

[49] In his determination styled, ‘Determination of Reasonable Compensation due

to Mr Kenneth Nkosana Makate arising from the Please Call Me Idea’ dated 9 January
2019 (CEO’s determination), the CEO states that Mr Makate described his function as
that of an arbitrator, and Vodacom’s, as that of a valuer. The CEO however considered
his function to be that of a deadlock-breaker, in accordance with the parties’
agreement as confirmed by the Constitutional Court. Counsel for Mr Makate, in the
course of his submissions before the CEO, agreed that the CEO functioned as a
deadlock-breaker.

[50] The CEO described the process he followed in his explanatory affidavit. Mr
Makate and Vodacom made comprehensive written submissions as to what
constitutes reasonable compensation, supported by memoranda from their experts.
Thereafter the CEO met with the parties for two days, heard oral submissions from
both sides and put his queries and difficulties to them.

[51] Subsequently, the parties filed further written submissions and documents with
the CEO. In that process Mr Makate submitted a report by ANZ Consulting, containing
requests for information from Vodacom and Mr Makate’s Model 8A, which sought to
track estimates of the PCM revenue year by year. That report also contained a
response to some of the questions posed by the CEO during oral argument. Vodacom
granted Mr Makate access to the information sought. These exchanges concluded the
parties’ submissions to the CEO.

[52] There was consensus regarding the manner in which the CEO had to carry out
his task. He was required to do no more than act in an objective manner relying on his
experience and applying his mind fairly and reasonably, so as to ensure that his
determination did not result in a manifestly unjust nor patently inequitable outcome.
The parties agreed that the CEO should make his determination in 2001 looking
forward.

24 Ibid para 5.
 25

[53] In his determination the CEO awarded Mr Makate compensation for the PCM
idea in an amount of R47 million. In making that determination the CEO considered
four models:
(a) A ‘2001 looking forward model’. The CEO put himself in the shoes of the CEO
in 2001 to determine the computation that Vodacom would probably have
agreed to pay Mr Makate for the use of his idea. The present-day value of this
calculation came to R51.5 million.
(b) An employee reward model for the PCM idea. In this model the CEO awarded
Mr Makate the highest bonus paid to a Vodacom employee, namely salary and
bonuses paid to the CEO over a period of three years (2016 to 2019). The
present-day value of the upper limit of this model came to R21.8 million.
Vodacom proposed and supported this model.
(c) A ‘Time Window Lock model’. In this model the CEO considered the time
window lock customers envisaged to benefit from the PCM product in 2001. A
customer purchased a prepaid voucher for a specific amount locked into a time
period for usage, for example, 14 days. If the airtime was not used within that
prescribed window, the customer was time window locked – they could not
make or receive calls unless they bought another voucher. The present-day
value of this model was calculated at R38.1 million.
(d) A ‘revenue share model looking backwards’. This model seeks to determine,
with the benefit of hindsight, the compensation Vodacom would have paid to
Mr Makate for the use of his idea under a contract concluded in 2001. The
present-day value of this calculation came to R42.2 million.

[54] The CEO considered the outcomes of these four models and based his final
determination on the average of the two outcomes most favourable to Mr Makate, (the
first and fourth models), rounded up to R47 million.

[55] The only competing model put up by Mr Makate was a revenue share model
looking backwards after 18 years, according to which he claimed compensation
calculated at 5% of all revenue generated by the PCM product – between R28 and
R110 billion. He contends that his quantification of the compensation due is the only
one consistent with the Constitutional Court’s judgment and that all other computations
must therefore be disregarded.
 26

The review
[56] In August 2019 Mr Makate launched proceedings in the high court to review
and set aside the CEO’s determination. He sought essentially an order substituting
that determination with a decision that Vodacom pays him 5 to 7.5% of the total
revenue of the PCM product as set out in his Models 9A, 9B and 9BB, from March
2001 to date of judgment, together with interest and costs on an attorney and client
scale.

[57] The review was founded on five alleged irregularities by the CEO which,
Mr Makate contended, rendered the CEO’s determination manifestly unreasonable,
unjust and inequitable. In sum, these irregularities are:
(a) The duration of the revenue share period. The CEO irregularly limited
Mr Makate’s revenue share to five years despite PCM having generated
revenue for 18 years and continuing to do so. Mr Makate alleged that the
agreement would have endured for as long as Vodacom used the PCM product,
but he accepted a contract duration of 18 years.
(b) The CEO declined to award Mr Makate any interest on his revenue share. He
mistakenly assumed that Mr Makate’s counsel had abandoned any claim for
interest.
(c) The proceedings before the CEO were inherently unfair because Vodacom
refused to disclose the revenue earned from the PCM product.
(d) The CEO relied on incorrect figures regarding the revenue earned by Vodacom.
He used an incorrect revenue model in determining compensation and should
have confined the determination to his fourth model – the revenue share model
looking backwards.
(e) The CEO failed to consider evidence concerning Mr Makate’s claim based on
Vodacom’s own audited publications. In its 2017 financial statements the
Vodacom Group provided for a minimum materiality threshold of R960 million
concerning Mr Makate’s claim. In 2016 the annual financial statements of
Vodafone PLC, Vodacom’s holding company, reflected the overall group
materiality as £180 million (R3.7 billion at the relevant exchange rate).

[58] Vodacom’s answer to these review grounds can be outlined as follows:

 27

(a) The five-year revenue share period determined by the CEO was consistent with
the position in relation to untested products in 2001. It was generous in
circumstances where Mr Makate was not rendering any ongoing service to
Vodacom. The CEO’s determination was one of a number of possibilities which
could not be gainsaid. Mr Makate throughout sought ‘reasonable remuneration’
– not that the contract would run in perpetuity. His claim that a tenure of 18
years was ‘rational’ because PCM could have been patented, is unsustainable
and was not contemplated by the parties. The dispute concerning the
compensation payable to Mr Makate was referred to a deadlock-breaking party,
the CEO, whose decision is final and binding on the parties. Mr Makate was
therefore not entitled to review the CEO’s determination.
(b) As regards interest, the CEO recorded that his determination made provision
for the time value of money even though the operative order required interest
to accrue only once compensation had been determined. The claim for interest
was not based on any legal ground but merely as ‘a gesture of good faith’. Mr
Makate did not pursue a claim for interest during the trial.
(c) The proceedings before the CEO were not unfair, as there is no direct revenue
that could be quantifiably attributed to the product. Mr Makate’s legal team took
an informed decision to proceed with their submissions before the CEO on the
basis of the documentation disclosed to them at that stage. The CEO was
acquainted with the relevant information and insofar as there were
inadequacies, after the oral hearing he called upon Vodacom to provide
additional information, which it did. Mr Makate was not hampered in any way in
formulating his model to determine what he believed was reasonable
compensation due to him, based on the information made available to him by
Vodacom.
(d) The CEO was given an unfettered discretion to determine reasonable
compensation: he did not rely on incorrect figures and could not be criticised
for the revenue model which he adopted.
(e) The overall quantitative materiality thresholds stated in the independent
auditors’ reports in respect of the consolidated annual financial statements of
the Vodacom Group for March 2017 (R960 million), and Vodafone for March
2016 (£180 million), constitute disclosures informed by an unquantified existing
obligation, in accordance with the International Accounting Standard 37 (IAS
 28

37). At the relevant time, Vodacom concluded that the outflow of economic
benefits could not be reliably estimated. These materiality thresholds are
irrelevant to the determination of reasonable compensation.

The high court’s judgment

[59] The review application came before Hughes J who held that the CEO ‘did not
exercise the judgment of a reasonable person, which resulted in an inequitable result
for the affected parties’. The court apparently upheld the review and set aside the
CEO’s determination (it made no order to this effect). It concluded that the CEO’s
consideration of the Time Window Lock and revenue share models, of his own accord,
‘renders the determination manifestly unjust’. This, seemingly because the CEO did
not engage the parties on these two models and failed to act in accordance with their
mandate.

[60] The high court endorsed Mr Makate’s revenue share model looking back over
18 years. It found that the ‘eighteen years proposed by Makate is reasonable,
probable’ and had been achieved; and that the CEO should ‘apply the eighteen-year
period’. Despite this, the court ordered that Mr Makate be paid 5% of the total revenue
of the PCM product from March 2001 to March 2021 – a period of 20 years.

[61] The high court held that Mr Makate’s challenge that the proceedings before the
CEO were unfair because Vodacom refused to disclose documents relating to the
revenue earned from PCM, was ‘without merit, to say the least’. The CEO had directed
Vodacom to provide the documents requested by Mr Makate, and his team were
granted access to them. The documentation provided by Vodacom was adequate.

[62] But then the court upheld Mr Makate’s contention that the CEO’s determination
and proposed models could not be relied upon because ‘Vodacom did not disclose the
relevant documentation and information’. These models, the court said, have ‘inherent
problems, which could only be resolved if Vodacom comes to the party and provides
the relevant and necessary information’. Then it said that it could not exercise its
discretion to make a substitute order, because there was no ‘proper and adequate
information on record’.
 29

[63] Concerning the CEO’s revenue share model to determine compensation, the
high court held that the CEO’s stance on incremental revenue was ‘contrary to what
both Makate and Vodacom understood’ – that the success rate or call rate constituted
incremental revenue. It dismissed the CEO’s adjustment of 30% for incremental
revenue as ‘arbitrary’ and said that the CEO had made this decision ‘without granting
the parties an opportunity to make representations’. The court stated that ‘the CEO
conceded that he was unable to provide an explanation why he applied a further 70%
reduction against Makate’s revenue’.

[64] Regarding interest, the high court found that the CEO ‘confused the issue of
interest with that of time value of money’. The CEO, it said, conceded as much in his
explanatory affidavit by stating ‘I may have been confused between the concept of
interest on an unliquidated claim for damages and mora interest’.

[65] The high court held that the CEO should have allowed for an 18-year contract.
Yet it granted Mr Makate a 20-year contract, essentially for three reasons. First,
Vodacom’s contract with a company, Cell-Find, was concluded in 2003 and was still
extant in 2020. Second, the CEO had ignored a statement by Mr Andre Hendricks
concerning advertising revenue generated on the back of PCM messages in 2012.
And third, the CEO’s determination was vitiated by the fact that he advanced further
reasons for the duration of the contract after the event.

[66] The high court remitted the matter to the CEO for a fresh determination. Despite
its acknowledgement that it has neither the experience nor competence to substitute
the CEO’s determination with its own, the court issued the following detailed directives
to the CEO when determining compensation afresh:
(a) The CEO’s determination of reasonable compensation was restricted to a
revenue share model of 5% for a period of 20 years.
(b) The calculation of the total voice revenue in that model must include PCM
revenue derived from prepaid, contract and interconnect (MTR) fees as set out
in Vodacom’s annual financial statements, the extracts from its financial
statements for the period 2001-2017, and a table prepared by Mr Makate
showing the CEO’s fundamental errors in determining compensation.
 30

(c) The CEO must apply Mr Makate’s PCM count in his Model 9A. Mr Makate is
entitled to 27% of PCMs sent daily, being revenue generated by the return calls
to the PCM. And the CEO must account for the time value of money at 5% for
each successive year.
(d) The CEO must assume that the average call duration of return calls is two
minutes.
(e) The effective rate applied by the CEO must be a blend of the contract effective
rate and the prepaid effective rate, and in each case, the effective rate must not
be less than those published by the Independent Communications Authority of
South Africa (ICASA).

[67] The court granted this relief with costs, but did not state which party was liable
for costs. It appears from the judgment that Vodacom was so liable.

Submissions in this Court

Vodacom’s main submissions
[68] Vodacom submits that the reasonable compensation envisaged by the parties
was a lump sum. They agreed that Mr Makate would be paid ‘a share in the revenue
generated by’ the PCM product. They did not specify whether that share should be a
percentage or a lump sum. They deferred their negotiations on ‘the amount to be paid’
and if they could not agree on that ‘amount’, it would be determined by Vodacom’s
CEO. All of this suggests that the parties had a lump sum award in mind.

[69] Vodacom further submits that determining the amount of compensation was
within the discretion of the CEO. His mandate did not prescribe that Mr Makate was
entitled to a share of all or any particular revenue generated by the PCM product; an
ongoing or indefinite share of that revenue; or a share of that revenue for a particular
period.

[70] Vodacom contends that the high court held that the CEO performed the function
of an expert valuer, and referred to the test for the review of cases of this kind in
Bekker,25 but did not apply it. For the most part the court applied an appeal standard,

25 Bekker v RSA Factors [1983] 4 All SA 328 (T) (Bekker).

 31

asking whether the CEO had been right or wrong, with frequent applications of the
rules of administrative review. As an expert valuer, the CEO did not perform a quasi-
judicial function and thus was not required to grant the parties a hearing at all. 26

[71] Vodacom submits that the high court failed to adhere to the Plascon-Evans
rule.27 It repeatedly dismissed the CEO’s findings and upheld Mr Makate’s contentions
without recognising that they were based on conflicting sets of fact, specifically as
regards his determination of reasonable compensation.

[72] Vodacom contends that the attack on the CEO’s determination on the basis of
Vodacom’s alleged failure to disclose documents, is unfounded. The high court
concluded that Mr Makate’s complaint that Vodacom refused to disclose documents
of revenue earned from PCM had no merit. The court’s subsequent criticism of
Vodacom for allegedly failing to produce relevant documents, is unjustified.

[73] In Vodacom’s submission, Mr Makate’s revenue share model looking

backwards over 18 years is conceptually flawed because it is inconsistent with the
contract upheld by the Constitutional Court, and fails to determine the amount of
compensation the CEO in 2001 would have considered to be fair to Mr Makate. The
model is also fundamentally flawed because Mr Makate’s calculations are grossly
exaggerated. What is more, Mr Makate’s model should be disqualified altogether
because it does not conform to the CEO’s forward-looking mandate.

[74] Vodacom contends that the CEO’s calculation should prevail over that of
Mr Makate because the parties agreed to entrust the determination of reasonable
compensation to his discretion. His expert knowledge and experience concerning the
determination the notional CEO would probably have made in 2001, is far superior to
that of Mr Makate. His account of the facts must be preferred over that of Mr Makate
under the Plascon-Evans rule.

26 Lufuno Mphaphuli & Associates (Pty) Ltd v Andrews and Another [2007] ZASCA 143; 2008 (2) SA
448 (SCA) (Lufuno) paras 21-22.
27 Plascon-Evans Paints (Pty) Ltd v Van Riebeeck Paints (Pty) Ltd 1984 (3) SA 623 (A) (Plascon-Evans).
 32

[75] Even on Mr Makate’s backward-looking revenue share model, so Vodacom

contends, the CEO’s conclusion that the likely duration of a contract awarded to Mr
Makate in 2001 would have been three years or five years at the most. Further, the
CEO’s determination of incremental revenue – that it would not have earned but for its
use of PCM – was reasonable having regard to the PCM volumes, the call back
success rate, the call duration and the effective rate. The high court erred in directing
the CEO to base his calculations on the ‘total voice revenue including PCM revenue’
derived from Vodacom’s annual financial statements, in paragraph 2(b) of its order.

[76] Vodacom contends that the high court erred in holding that the CEO’s
admission that he had mistakenly accepted that Mr Makate’s counsel abandoned the
claim for interest, vitiates his determination. The CEO’s mistake did not affect that
determination.

[77] Vodacom submits that the high court inexplicably granted Mr Makate a share
of the revenue from the PCM product for a period of 20 years. This, after it held that
the CEO should have allowed for an 18-year contract. The court erred in rejecting the
CEO’s conclusion that the likely duration of a contract awarded to Mr Makate in 2001
would have been three years or five years at most, because the commercial viability
of PCM was uncertain.

[78] It is further submitted that the court’s reasons for an 18-year contract are
unfounded. The Cell-Find contract did not serve before the CEO – it was only raised
during the review. In any event, it is not typical of Vodacom’s contracts with service
providers at the time. The court’s reliance on Mr Hendricks’ statement – which referred
to advertising revenue generated on the back of PCM messages in 2012 – was
misplaced. It had nothing to do with the likely duration of a contract with Mr Makate in
2001. The High Court mistakenly applied the administrative law rule that a public body
may not justify its decision with new reasons after the event – this is a review under
the Bekker test. The CEO, it was submitted, did not advance new reasons concerning
the duration of the contract: he merely responded to the criticism of his determination
based on new documents which Mr Makate produced in the review proceedings.
 33

[79] Vodacom says that the high court’s order is inappropriate. In his amended
notice of motion Mr Makate never sought an order that the court usurp the CEO’s
discretion by issuing directives for the determination of PCM revenue. Moreover, the
court effectively substituted the CEO’s determination with its own; and issued
directives for the calculation of reasonable compensation, ignorant of the impact of
such compensation, more specifically whether or not it may be manifestly unfair or
patently inequitable.

Mr Makate’s submissions
[80] Mr Makate submits that he has a contractual claim based on an agreement to
be paid a share of the revenue generated by the PCM idea – he accepts 5% as a fair
share. The very concept of a ‘share in the revenue’ denotes some proportionality
between the ‘share’ and the ‘whole’. It is not simply an arbitrary ‘amount’. Mr Makate
contends that Vodacom’s argument on the CEO’s mandate is a wrong interpretation
of the Constitutional Court’s judgment; and that the CEO correctly found that his
mandate required him to determine that Mr Makate was entitled to a percentage of the
revenue.

[81] The parties agreed on the standards that the CEO had to apply in carrying out
his function. These included objectivity, fairness and reasonableness. Mr Makate
contends the CEO’s failure to comply with these standards was correctly reviewed by
the high court; and that Vodacom’s reliance on the Bekker test for its submission that
the CEO did not exercise a quasi-judicial function and thus did not have to hear the
parties on the issue of reasonable compensation, is wrong – the review standard is
not determined by the classification ascribed to the CEO’s decision.

[82] Vodacom’s contention that the CEO functioned as a valuer, Mr Makate says, is
unsustainable for the following reasons. There is no suggestion of this in the judgment
of the Constitutional Court. The CEO did not regard himself as an expert valuer. The
function performed by the CEO does not comport with the role of a valuer: the
appointment flowed from the office and could have been performed by any CEO
holding that office. His deadlock-breaking task required the CEO to take quasi-judicial
decisions on the interpretation of the Constitutional Court’s order, and commercial
decisions setting 5% as the appropriate revenue share. Once the CEO exercised his
 34

discretion, he was bound to act within the established review grounds when he
analysed the facts and figures before him.

[83] The high court was correct in rejecting Vodacom’s ‘narrative’ that it has
inadequate documents and records to calculate revenue. This was conclusively
proved to be incorrect by a whistle blower, former employees of Vodacom and its own
executives.

[84] Mr Makate says that his Model 9A is a reconstruction by his experts of the
revenue earned from various sources based on a 15% revenue share, and shows that
a 5% revenue share amounts to a capital sum of approximately R9 billion. The CEO’s
award of R47 million, Mr Makate contends, is patently inequitable and not close to an
amount commensurate with the revenue derived from PCM.

[85] The CEO wrongly stated, so it is submitted, that Mr Makate’s counsel had
accepted that the CEO should adopt the 2001 model looking forward. Rather,
Mr Makate’s counsel had adopted an approach that the only precise science was
hindsight, and it is inconceivable that a purely hypothetical looking forward model with
no empirical data could be more accurate than using available data over 18 years. Mr
Makate contends that the only model which is consistent with the agreement between
the parties is the revenue share model looking backwards.

[86] When that model is utilised, Mr Makate submits that the CEO made numerous
errors in relation to: the duration of the contract; the calculation of total voice revenue;
and the dismissal of Mr Makate’s Model 8A. Moreover, so Mr Makate contends, the
CEO made critical errors in his calculation of incremental revenue and the inputs in
that calculation. PCM volumes are understated by 41%. The call back success rate is
wrong: the CEO’s conclusion that for every 100 PCMs sent daily, 27 result in a call
back, is incorrect. The data mining exercise proved that 38% were returned daily and
the reduction of a further 70% was speculative. Neither the CEO nor Vodacom
explained how they derived a call duration of two minutes. The effective rate cannot
be lower than the ICASA rate.
 35

[87] Mr Makate contends that paragraph 2(b) of the high court’s order is based on
evidence confirmed by Vodacom and Mr Makate. He says that the core aspects of the
formula for calculating revenue were approved by all the directors of Vodacom’s
departments and Mr Makate’s expert, Mr Ivan Zatkovich. In September 2012
Vodacom’s managing executive of legal affairs ‘confirmed under oath that Vodacom
had data detailing whether a particular person had received a call from another’.
Vodacom had also stated that it could calculate revenue generated by PCM, by using
‘an appropriate time-frame between the PCM and a subsequent call’ to determine
whether the call was induced by the PCM.

[88] The CEO should have awarded Mr Makate interest on his revenue share.
Vodacom’s counsel conceded before the high court that mora interest, at a rate of
10.25% on the amount decided by the CEO, should have started to run on 18 January
2019.

[89] On the facts, there is no rational basis to limit Mr Makate to a five-year contract.
PCM has been a success from day one and there is no reason why Mr Makate is not
entitled to an ongoing revenue share in perpetuity, for as long as PCM continues to
earn revenue. This happened in the case of other VAS products without proprietary
rights, as is evidenced by the contracts which Mr Makate introduced in the review
proceedings.

The issues
[90] The issues raised by this appeal are these:
(a) Is Mr Makate entitled to a percentage of the revenue earned from PCM?
(b) The role of the CEO and the standard of review.
(c) The failure to apply the Plascon-Evans rule.
(d) Did Mr Makate establish the review grounds? More specifically, was the high
court correct in holding that:
(i) the proceedings before the CEO were unfair because Vodacom failed to
disclose relevant documents;
(ii) the CEO should determine compensation effectively in terms of a
revenue share model looking backwards, in accordance with the court’s
directives;
 36

(iii) the total voice revenue in the CEO’s determination should include
revenue derived from prepaid, contract and interconnect (MTR) fees as
set out in, amongst others, Vodacom’s annual financial statements
(paragraph 2(b) of the order);
(iv) the CEO should have allowed for an 18-year contract;
(v) the CEO’s mistake in concluding that Mr Makate had abandoned his
claim for interest and that he conflated mora interest with the time value
of money, vitiated his determination.
(e) Remedy and costs.

Is Mr Makate entitled to a percentage of the revenue earned from PCM?

[91] The Constitutional Court’s 2016 order, which states that the parties agreed that
Mr Makate ‘would be paid a share in the revenue’ generated by the PCM product, must
be interpreted in the context of the Court’s judgment as a whole.28 Mr Makate interprets
this order to mean that the parties had agreed that he would get a fixed percentage of
all the revenue generated by his idea.

[92] But that is not so. What was agreed was that Mr Makate would be paid ‘a share’
or ‘a part’ of the revenue. The parties did not specify what that share was, nor whether
it should be a percentage of the revenue or a lump sum. Mr Makate wanted 15% of
the revenue generated by his idea but Vodacom did not agree to that. Instead, the
parties deferred their negotiations on ‘the amount to be paid’, and decided that if they
could not reach agreement on that ‘amount’, the CEO would determine it.

[93] This interpretation, advanced by Vodacom, is buttressed by the plain wording

of the Constitutional Court’s 2016 order. It does not suggest that there was any breach
of the agreement between the parties, or that Mr Makate is entitled to damages – to
be placed in the position he would have been had the contract been performed. 29
Rather, the Constitutional Court ordered the parties to negotiate in good faith to
determine ‘a reasonable compensation’ payable to Mr Makate. If they failed to agree

28 SOS Support Public Broadcasting Coalition v SABC [2018] ZACC 37; 2019 (1) SA 370 (CC) (SOS
Support) paras 52 and 53.
29 Makate v Vodacom fn 23 para 107.
 37

on the reasonable compensation, the matter had to be submitted to the CEO to

determine the ‘amount’ of that compensation.

[94] The interpretation is further buttressed by the following statements in the

majority judgment (Jafta J). In the negotiations between Mr Geissler and Mr Makate
for the use of his PCM idea – after he indicated that the wanted 15% of the revenue
generated by the product – ‘the parties agreed to defer . . . negotiations on
remuneration’.30 In other words, there was no agreement that Mr Makate would be
entitled to a percentage of the revenue. Likewise, the Court found that the parties had
agreed to negotiate the ‘amount of compensation’ once the product was developed
and tried for its commercial viability.31 In the event that they disagreed on the ‘amount
to be paid’, Vodacom’s CEO would determine that ‘amount’.32

[95] The minority judgment of the Constitutional Court is also helpful because it
accords with the majority’s conclusion that the parties did not agree that Mr Makate
was entitled to a fixed percentage of all the revenue generated by the PCM product.
Wallis AJ said:
‘[N]o agreement was reached on the precise form or amount of such remuneration. Mr Makate
said that he wanted a profit share and had suggested 15%. Mr Muchenje confirmed that this
figure was mentioned in his discussions with Mr Makate. But the evidence is clear that
Mr Geissler did not agree to this figure or any basis for determining the remuneration.
Everything was to depend on the successful launch of the product after which the matter would
be discussed. If those discussions did not lead to agreement the question of remuneration
would be referred to Mr Knott Craig for determination.’33

[96] Wallis AJ went on to say that the parties had agreed to negotiate in good faith
concerning the remuneration payable. This, in fact, was Mr Makate’s case:
‘. . . Mr Makate sued Vodacom, basing his claim on a contract concluded between him and
Vodacom represented by Mr Geissler. As often happens, his pleadings were more ambitious
than the evidence led in support of his case. Over time they were amended. At the close of
the trial he claimed only that the contract between him and Vodacom was that in the return for
his providing the idea to Vodacom it would enter into bona fide negotiations with him in order

30 Ibid para 63.

31 Ibid para 94.
32 Ibid para 101.
33 Ibid para 114, emphasis added.
 38

to agree on a reasonable remuneration for his idea. Should they be unable to agree on a
reasonable remuneration the matter would be referred to Mr Knott-Craig for his adjudication.
The trial court held this to have been proved on a balance of probabilities.’34

[97] The minority accepted that Vodacom would not have concluded an agreement
with Mr Makate on a revenue share basis:
‘I do not accept that the company would not have concluded a contract with an employee in
order to procure the advantage of a profitable idea. While I accept that it would not have
concluded an agreement on a revenue share basis, I do not accept that it would not have
agreed to pay an employee, who generated, in his spare time and outside the scope of his
ordinary duties, a highly profitable idea for a new product, reasonable remuneration
commensurate with the financial benefit enjoyed by the company.’35

[98] Vodacom’s interpretation of the judgment is also consistent with the 2017 order.
The Constitutional Court dismissed Mr Makate’s application for an order declaring that:
(a) Vodacom is obliged to pay him a share in the revenue generated by the PCM
product; and (b) that the precise share of the revenue be determined in the
negotiations referred to in paragraph 3(b) of the order of 26 April 2016, or if necessary,
by the CEO referred to in paragraph 3(c) of that order. That application was dismissed,
because Mr Makate’s claim to a precise share of the revenue had no prospect of
succeeding at all.

[99] The CEO’s mandate was accordingly the following:

(a) He had to determine reasonable compensation payable to Mr Makate for his
PCM idea.
(b) Mr Makate was entitled to a share of the revenue generated by the PCM
product in an ‘amount’.
(c) The CEO had to determine the amount. The parties’ agreement did not
prescribe how the CEO should determine that amount, what form it should take,
or that it should be a percentage of the revenue earned from the PCM idea.
(d) Paragraph 3(b) and (c) of the order of 26 April 2016 does not say that
Mr Makate is entitled to: a share of all revenue or any particular revenue; an

34 Ibid para 115, emphasis added.

35 Ibid para 179.
 39

ongoing or indefinite share of revenue; or a share of revenue for any particular

period. These matters were left to the discretion of the CEO.

[100] For the above reasons, the Constitutional Court’s judgment is clear and
unambiguous on the terms of the CEO’s mandate. Mr Makate is not entitled to
compensation on the basis of a percentage of the revenue earned. Consequently, ‘no
extrinsic fact or evidence is admissible to contradict, vary, qualify or support’ that
judgment.36 That the parties agreed on a revenue share percentage before the CEO,
does not change the fact that there was no agreement on a revenue share basis. The
submission by Mr Makate’s counsel that the calculation of compensation ‘is based
upon the bedrock of a royalty or annuity payment on an ongoing basis’, is incorrect.

[101] It follows that the CEO was entitled to determine the compensation payable to
Mr Makate on the basis that he did: the appropriate model does result in an amount
constituting a share or a part of the revenue generated by the PCM idea.

The role of the CEO and the standard of review

[102] A critical feature of the CEO's mandate under the contract upheld by the
Constitutional Court, was that it had to be a forward-looking determination from the
perspective of the CEO in 2001. This is common ground, although counsel for Mr
Makate informed the CEO that the only precise science is hindsight.

[103] The basic principles of the law of contract also required the CEO to make such
a forward-looking determination. Mr Makate’s claim is one for specific performance of
the contract upheld by the Constitutional Court. That contract was concluded in 2001.
It required the parties then and there to negotiate in good faith to agree on Mr Makate’s
compensation and, failing agreement, for the CEO to determine it. In other words, the
contract required the CEO to make a forward-looking determination in 2001. In his
determination, the CEO states that he also examined the situation in 2018 looking
backwards, ‘so as to get a more balanced view of the determinations’.

36 SOS Support fn 28 para 53.

 40

[104] To the extent that it may be suggested that the CEO did not follow the parties’
instructions in grounding his final determination on the average of his first and fourth
models, this was not an irregularity alleged in the founding affidavit. Consequently, it
was not a case that Vodacom was called upon to meet. Mr Makate’s case was
squarely that the CEO should have considered the revenue share model looking
backwards.

[105] The parties and the CEO agreed that the CEO should perform the functions to
the best of his ability, having regard to the material before him, and relying on his
experience and expertise; that he should apply his mind objectively, fairly and
reasonably; and that he should not make a determination that was manifestly unjust
or patently inequitable. Subject to these requirements, the CEOs discretion was
unfettered.

[106] The standard of review in a case of this kind was laid down in Bekker.37 It is
this:
‘Where a third person is nominated to determine a purchase price or make a valuation, he
must exercise the judgment of a reasonable man. If his judgment with regard to the
determination of the purchase price or valuation is however so unreasonable, improper,
irregular or wrong that it would lead to obvious unfairness, the person prejudiced is not bound
by it, but the determination or valuation can be corrected on grounds of fairness.38 (My
translation.)

[107] The test in Bekker was applied and translated in Perdekis39 as follows:
‘It was held in Bekker . . . that a valuation can be rectified on equitable grounds where
the valuer does not exercise the judgment of a reasonable man, that is, his judgment
is exercised unreasonably, irregularly or wrongly so as to lead to patently inequitable
result’

37 Bekker fn 25 at 573E.
38 The Afrikaans text reads:
‘Indien ‘n derde persoon benoem word om ‘n koopprys vas te stel of ‘n waardasie te maak, moet hy die
oordeel van ‘n redelike man aan die dag lê. Indien sy oordeel met betrekking tot die prysvasstelling of
waardasie egter so onredelik, onbehoorlik, onreëlmatig of verkeerd is dat dit tot ‘n ooglopende
onbillikheid sal lei, is die persoon wat daardeur benadeel word, nie daaraan gebonde nie, maar kan die
vasstelling of waardasie om billikheidsredes reggestel word.’
39 Perdekis v Jamieson [2002] 4 All SA 560 (W) at 568.
 41

[108] This Court has repeatedly endorsed the Bekker test.40 Under that test, the
crucial questions in this case are twofold. The first is whether the CEO exercised his
judgment unreasonably, irregularly or wrongly. If he did not, the second question is
whether his determination of the compensation due to Mr Makate in the amount of
R47 million, is manifestly unjust or patently inequitable.

[109] It is accordingly unnecessary to decide whether the CEO functioned as a valuer

or whether he exercised quasi-judicial functions. The parties agreed that the CEO had
to act objectively, relying on his experience and applying his mind reasonably and
fairly, to ensure that his determination did not result in a manifestly unjust or patently
inequitable outcome.

The failure to apply the Plascon-Evans rule

[110] The high court disregarded the Plascon-Evans rule. Motion proceedings are
designed to determine issues based on common cause facts.41 The judgment contains
references to Mr Makate’s version of the facts which the court accepts, for example,
concerning the Cell Find contract, and an unsworn statement by Mr Andrew Hendricks
(Vodacom’s former Financial Manager Revenue Reporting), and statements by its
former employees in relation to the revenue earned from the PCM product. The court
then disregards the contrary version put up by Vodacom.

[111] This is impermissible.42 As Schutz JA said in Simelane:43

‘Such an approach is the converse of that laid down in Plascon-Evans at 634H-I, to the effect
that in a case such as this, the decision must be based on those facts averred by the applicant
which are admitted by the respondent, together with the facts averred by the respondent.’

[112] Instances in which the rule has been ignored will be dealt with under the
relevant issues. As is shown below, the Plascon-Evans rule was repeatedly
disregarded by the high court, and in relation to material disputes of fact.

40 Lufuno paras 21-22; Wright v Wright and Another [2014] ZASCA 126; 2015 (1) SA 262 (SCA) para
10; Transnet National Ports Authority v Reit Investments (Pty) Ltd and Another [2020] ZASCA 129 paras
32-34 and 36; Tahilran v Trustees of the Lukamber Trust and Another [2021] ZASCA 173; 2022 (2) SA
436 (SCA) para 27.
41 Zuma v DPP [2009] ZASCA 1; 2009 (2) SA 277 (SCA).
42 Plascon-Evans at 634H-I; Simelane NO and Others v Seven-Eleven Corporation SA (Pty) Ltd and

Another [2002] ZASCA 141; [2003] 1 All SA 82 (SCA) (Simelane).

43 Ibid Simelane para 10.
 42

[113] Mr Makate’s review grounds are dealt with next. These have no merit, save for
the one relating to the duration of the contract.

Alleged procedural unfairness: non-disclosure of documents and information?

[114] The founding affidavit states that Vodacom failed or refused to disclose what
revenue it had generated from PCM, which rendered the proceedings before the CEO
inherently unfair. This allegation is based on the following contentions. In November
2011 already, the high court (Spilg J) ordered Vodacom to discover documents relating
to revenue generated by PCM. Even if that revenue could not be calculated with
precision, logic dictates that the closer the PCM message is to the returned call, the
more likely that the call was generated by the PCM message. Vodacom’s stance
throughout that there is no direct revenue attributable to the PCM product because it
is a zero rated or free service, is untenable and indicative of bad faith. The CEO ‘ought
to have directed that Vodacom make the information available as best it could’. In
failing to do so he acted manifestly unreasonably, unfairly and inequitably. Moreover,
affidavits by former employees of Vodacom showed that it was able to disclose the
revenue it earned from PCM.

[115] These contentions are however unsustainable on the evidence. Mr Makate

abandoned the discovery order issued by Spilg J when Vodacom sought leave to
appeal against it. Consequently, Vodacom did not proceed with the appeal.

[116] On 28 October 2016 Vodacom formally tendered to Mr Makate’s attorneys

access to its information systems and records. The tender was subject to the following
conditions: both parties could appoint their own experts; the integrity of Vodacom’s
systems had to be maintained; the identity of its subscribers had to be protected; and
Mr Makate could appoint an expert to confirm that Vodacom was not concealing any
records of PCM calls and revenue.

[117] Mr Makate ignored this tender and proceeded with the second application to
the Constitutional Court, in which he also sought access to financial records relating
to the PCM product. As already stated, on 8 February 2017 the Constitutional Court
dismissed that application with costs.
 43

[118] Subsequently, Vodacom again tendered access to its information systems and
records under a strict confidentiality regime. In the result, the parties commenced a
process jointly styled the ‘data mining exercise’. Mr Makate’s experts conducted this
exercise from 3 July 2017 to 14 July 2017 at Vodacom’s premises, where they were
provided with technical support and assistance. They were given access to three
stores of raw data, namely detailed information for six months of every PCM message
sent; call data records for a period of six months containing all voice calls and
messaging records; and a record store of prepaid information where there was a zero-
airtime balance, indicating whether or not there was airtime when a PCM message
was sent. The answering affidavit states that at the conclusion of the data mining
exercise, Vodacom’s negotiating team were advised that the data extracted was
sufficient for Mr Makate’s experts to develop and modify their model.

[119] After the initial data mining exercise, Mr Makate’s team were given additional
opportunities for further data mining to correct their model, after patent errors in it were
pointed out to them. Mr Makate declined these offers. Following the data mining
exercise, Mr Makate formulated his model of reasonable compensation amounting to
billions of Rands.

[120] The CEO considered Vodacom’s claim that it would be impossible to determine
with any reasonable accuracy whether a call made after receipt of a PCM was in fact
prompted by the PCM; whether that call was an incremental call; whether the sender
of the PCM had airtime when the PCM was sent; the duration of the call as it relates
to the PCM; and the tariff paid by the recipient of the PCM. These uncertainties could
not be answered despite the fact that Vodacom furnished a significant amount of data
covering a period of six months, which Mr Makate’s team had studied. The evidence
showed that Vodacom retains only six months of operational data at a time, and this
data makes no direct link between PCM and return calls. Consequently, the CEO
concluded that assumptions and estimates had to be made. Indeed, both parties did
so in their various models.

[121] The CEO found that the inadequacy of the documents was not indicative of a
lack of bona fides on the part of Vodacom. Instead, it is ‘a situation in which
 44

documentation is inadequate to address the various uncertainties inherent in the

exercise being undertaken’.

[122] The following submission by Mr Makate’s counsel to the CEO is thus not
surprising:
‘You’ve got to do, Mr Chair, the best you can with a material available notwithstanding the
uncertainties that exist. That is what you have to do. And then you have to award
compensation that is a reasonable and fair share of revenue, that revenue being viewed as
the best you could do with the materials available to you.’

[123] What all of this shows, is that Mr Makate was not hindered in any way from
formulating his claim. After the parties made their oral submissions, Mr Makate
submitted the ANZ report (his Model 8A), which contained various requests for
information from Vodacom. Vodacom provided this information. Mr Makate raised no
objection as to its adequacy.

[124] The CEO states that following the further submissions made after the oral
hearing, Mr Makate’s team did not ask him to order the production of additional
documentation. Neither did they indicate that they wished to apply to court for such an
order prior to his determination. On the contrary, they asked the CEO to proceed on
the available evidence, to the best of his ability.

[125] Mr Makate’s allegations that Vodacom was able to determine revenue from
PCM based on affidavits by its former employees, information contained in the
Vodacom mobile advertising deck, and a rule 35(12) notice under the Uniform Rules
of Court (given in interlocutory proceedings to clarify an order granted in an application
to compel the provision of documents), have been refuted in the answering affidavit.
Disputes of fact in this regard, if any, must be resolved in Vodacom’s favour under the
Plascon-Evans rule.

[126] Finally, on this aspect, the high court found that Vodacom had given
Mr Makate’s team access to the documents requested, and his complaint that
Vodacom refused to disclose documents of revenue earned from PCM, was without
merit. The court said that if Mr Makate considered that those documents were
 45

insufficient, he could have brought an application to court to compel Vodacom to

furnish additional documents, which would have addressed his complaint. He did not
do so. The only conclusion to be drawn was that Mr Makate was satisfied with the
documents that Vodacom had produced.

[127] But then, inexplicably, the court sharply criticised Vodacom for allegedly failing
to provide relevant documents. In any event, Mr Makate requested the CEO to do the
best he could on the information before him. Any shortcoming in Vodacom’s disclosure
cannot found a basis to attack the CEO’s determination. Mr Makate was treated fairly.

The backward-looking revenue share model

The percentage of revenue share
[128] The CEO determined Mr Makate’s share at 5%. Mr Makate accepted this
determination.

PCM incremental revenue

[129] The CEO states that there was consensus that the exercise was limited to
incremental calls – made by the person responding to the PCM which would not have
been made but for the PCM. This by itself, the CEO said, presented a significant
challenge in determining reasonable compensation. Moreover, to qualify as an
incremental call, that call must be an additional call, ie it must generate additional
revenue for Vodacom over and above the customer’s normal spend, but for PCM.
There were however substantial difficulties in determining incremental revenue.

[130] The records show that millions of calls were made by responders but it was
hard to establish whether the responding call would have been made in any event.
The next difficulty was to determine the duration of the incremental responding call,
which was virtually impossible since it required a call by call analysis of millions of calls
per day. Even if one could determine the duration of the incremental responding call,
there was the further problem of identifying the rate payable by the responding caller
for that call. This, in turn, depended on whether the responding call was initiated
through the Vodacom network or another network provider, and if through the
Vodacom network, what rate in a range of possible rates was applicable.
 46

[131] In addressing these difficulties, the CEO assumed that responding calls made
within one hour of a PCM were as a result of the PCM, in accordance with a proposal
by Mr Makate. The CEO concedes that this assumption ‘can be attacked on many
levels’ but states that without it, it is impossible to quantify revenue generated by the
PCM product. The CEO’s best estimate of that revenue was based on the following
factors: PCM volumes; call back success rate; incremental revenue; call duration; and
the effective rate.

(i) PCM volumes

[132] The Makate model attempted to estimate PCM volumes from information put
together from Vodacom’s marketing materials and annual reports, supplemented by
calculations which Mr Makate’s team made from 2017 data to which they were granted
access by Vodacom.

[133] The CEO states that the assumptions made in Mr Makate’s model to determine
this is flawed and unrealistic. These assumptions were deficient in material respects.
First, there was a miscounting of the PCM volumes. The PCM count which Mr Makate
relied on included not only the PCMs sent, but also the confirmation messages that
the original senders received. In the result, the volume of PCMs was doubled. The
PCMs generated daily were approximately 19 million and not 38 million. Second,
although the information was only for three years between 2002 and 2017 (namely
2008, 2015 and 2017), in Mr Makate’s model the figures used for the other years were
simply assumed by reference to the three years.

[134] To correct Mr Makate’s errors, the CEO calculated the percentage of customers
who used PCM and then calculated that off the average customer base for the year,
in order to arrive at a base penetration rate. He used Vodacom’s 2018 data and
extrapolated backwards from the 2018 data points. This showed the percentage of
customers who used PCM and the average number of PCM’s sent per customer. He
came to a base penetration rate of 37.4%. The CEO’s approach was, as he put it, ‘very
generous’ to Mr Makate because the average figure applied to the 2002-2006 period
was much higher than Mr Makate’s estimates.
 47

[135] The High Court ignored the CEO’s determination and directed him to base the
PCM volumes on Mr Makate’s Model 9A (a modified version of his model 8A).
However, Model 9A was never before the CEO and was produced for the first time in
the review.

[136] The high court gave no reasons for this directive. This is perplexing as
Mr Makate did not criticise the CEO’s reasoning on the PCM volumes. Model 9A
merely repeats the same flawed assumptions and exaggerated figures that the CEO
had criticised.

(ii) Call back success rate

[137] Mr Makate proposed, and the CEO accepted, a call back success rate of 27%.
In other words, it was assumed that for every 100 PCM’s sent in a day, 27 resulted in
a call back.

[138] The figure of 27% was drawn from the data mining exercise conducted by Mr
Makate’s team. As stated, the CEO also accepted that a responding call made within
an hour of a PCM was triggered by the PCM.

(iii) Incremental revenue

[139] Incremental calls are those made by the person responding to the PCM which
that person would not have made but for the PCM. But even incremental calls often
do not earn any revenue for Vodacom.

[140] Assume that A sends a PCM to B who responds by making a call to A. The call
is, in other words, one of the 27% triggered by the PCM. However, the call is neither
incremental nor would it earn Vodacom any incremental revenue in the following
cases:
(a) But for PCM, A would have sent B an SMS to call her. B would have responded
by calling A.
(b) But for PCM, A would have ‘buzzed’ B by giving her a missed call. B would have
responded by calling A.
 48

(c) B (a mother) calls A (her daughter) every day. A sends B a PCM to indicate that
she is free to take the call. But without the PCM, B would in any event have called
A.
(d) B has a contract in terms of which she pays R100 for 100 minutes of talk time
per month. She only uses 70 minutes per month. The PCM causes her to use 90
minutes per month. But Vodacom still only earns R100 per month.
(e) B buys prepaid talk time of R100 minutes per month. She carefully rations her
calls to remain within her budget of R100 per month. Following a PCM from A,
she makes a call to A that she would not otherwise have made. She
compensates for this additional call by reducing her talk time to remain within her
budget of R100 per month. Here too, Vodacom would not earn incremental
revenue from the call.

[141] Mr Makate’s expert, Mr Ivan Zatkovich, recognised that the calculation of PCM
revenue needed to take into account not only the call back success rate, but also the
portion of those returned calls that were incremental. Despite this, Mr Makate’s
revenue share model did not make any adjustment for incremental revenue.

[142] The CEO was concerned about this oversight and raised it repeatedly with
Mr Makate’s senior counsel during the hearing before him. He specifically asked
counsel:
‘[I]f Vodacom [was] doing a million calls a day, what has Please Call Me done to increase that?
Did it increase by 5%? Did it increase by 20%?
MR PUCKRIN SC: I see.
CHAIRPERSON: Exactly what is it and I would have imagined that they weren’t going to pay
you on a million calls. They would pay you on the increment of the 20% or 5% or whatever it
is.
MR PUCKRIN SC: Yes, of course.’

[143] The CEO then asked Mr Makate’s counsel to explain how Mr Makate’s model
catered for the issue of incremental revenue. Counsel replied that Mr Makate was
entitled to incremental revenue ‘as a matter of logic’; that the ‘difficulty is to calculate
it’; and that he would ‘fall back on [his] argument’ that had Vodacom in 2001 concluded
an agreement with Mr Makate, the question of incremental revenue ‘would have been
 49

a matter of great facility for Vodacom’. Counsel’s reply, needless to say, was unhelpful,
as was his submission regarding what Vodacom could have done to determine
incremental revenue. The fact is that Vodacom never retained the relevant data to do
so, due to its sheer volume.

[144] Mr Makate’s counsel acknowledged that in determining incremental revenue

one must necessarily rely on estimation without data, but offered no solution. Instead,
he said, ‘We would ask you never to underestimate, in this potpourri of assessing
revenue, . . . the value of unlocking the time window lock’.

[145] The CEO pressed Mr Makate’s counsel for more. He said that the proposed
27% success rate ‘is not necessarily incremental revenue, right?’, to which counsel
replied, ‘No’. The CEO asked how one derives incremental revenue from Mr Makate’s
model. He clarified this question as follows:
‘[A]re you saying that we should take into account 27% that called back [are all incremental]
or are you saying that you should take the amount that would have been incremental? So
whether it’s over an hour or 24 hours or whatever, … the person may have called back at a
later stage.’
Mr Makate’s counsel responded:
‘I would suggest the following as a matter of logic and I’m not an expert in these matters. The
more urgent the call back, in other words the time envelope – the narrower the time envelope
the more one may assume that it is incremental revenue. In other words it is an additional call
that is made rather than a call that would have been made anyway.’

[146] The CEO put it to counsel that one should then see an increase in billions of
Rand in Vodacom’s revenue. Counsel had no answer. Instead, he reverted to what
Vodacom could have done in 2001:
‘If Vodacom had made a calculation of the additional calls made in – once the system got
going. Let’s say May/June 2001 they would have had an accurate figure. An accurate figure
of incremental revenue. To ask us with our limited time in which to do this mining exercise we
were only able to take two days and, Mr Chair, with respect I come back to the law. We have
done more than what we were called upon to do. You know, if the negotiations had proceeded
in good faith and if this has been one of the methods which the Vodacom negotiation team
had been prepared to entertain, those questions may have been resolved.’
 50

[147] Vodacom’s counsel also addressed the issue of incremental revenue in his oral
submissions to the CEO. He pointed out that Vodacom’s market share did not change
in the two years after PCM was launched; that the growth in revenue between 2000
and 2005 did not meaningfully change as a result of PCM, which indicated that there
was no incremental revenue; and that Mr Makate’s calculation of the revenue as a
result of PCM – R2 billion – represented 80% of the revenue generated by Vodacom
in 2002.

[148] Both parties made further submissions to the CEO after the hearing. Despite
the focus on the issue of incremental revenue at the hearing, Mr Makate did not take
the opportunity of addressing it any further.

[149] The CEO in his determination said:

‘There can never be any data which can tell whether a responding call would have been made
anyway and I can only rely on my best estimate made on the strength of my insights and
impressions over the years.’

[150] The CEO assumed, in Mr Makate’s favour, that 30% of the returned calls
triggered by a PCM yielded incremental revenue. This assumption was favourable to
Mr Makate because the CEO in fact ‘could not find any significant increase [in call
revenue] which was demonstrably attributable to PCM’.

[151] In the replying affidavit Mr Makate states that the CEO’s determination of
incremental revenue ‘was not explained’, and is ‘without any factual basis’ and
therefore arbitrary and irrational. But, as the CEO’s counsel submitted in the review,
the CEO did as courts do when faced with ‘pondering the imponderable’ – he did the
best he could ‘on the material available even if the result may not inappropriately be
described as an informed guess’.44 This, after all, is precisely what Mr Makate asked
the CEO to do: ‘You’ve got to do . . . the best you can with the material available
notwithstanding the uncertainties that exist.’

[152] For the above reasons, the high court’s finding that the CEO’s stance on
incremental revenue was ‘contrary to what both Makate and Vodacom understood’ –

44 Southern Insurance Association Ltd v Bailey NO 1984 (1) SA 98 (A) at 114B.

 51

that the call back rate constituted incremental revenue, is baffling. The court’s
dismissal of the CEO’s 30% adjustment for incremental revenue as ‘arbitrary’, and
made ‘without granting the parties an opportunity to make representations’ is
unsustainable on the evidence. So too, its order directing the CEO to determine
Mr Makate’s revenue share based on an entitlement to ‘27% of the number of PCMs
sent daily as being revenue generated by the return call to the PCM’.

[153] The high court said that ‘the CEO conceded that he was unable to provide an
explanation why he applied a further 70% reduction against Makate’s revenue’. But
that is not correct. As shown above, Mr Makate’s counsel appreciated that calculating
incremental revenue is a difficult exercise; that Vodacom has no record, let alone an
accurate record of calls ‘induced’ as a result of a PCM message; and that the CEO
was forced to rely on assumptions. These facts are repeatedly stated in the answering
affidavit. The CEO gave a detailed explanation for his assumption that 30% of the
return calls triggered by PCMs yielded incremental revenue. He merely conceded that
he could not prove that his informed guess of 30% was correct, precisely because
nobody knows nor can determine what proportion of return calls yield incremental
revenue. The CEO drew on his experience and expertise to make an assumption in
Mr Makate’s favour that 30% of the return calls yielded incremental revenue.
Ultimately, his determination was an estimate which, in his words, ‘defies exact
computation’.

(iv) Call duration

[154] Call duration is one of the key input variables in estimating PCM related
revenue. The CEO states that the average duration in Mr Makate’s model changes
dramatically from year to year, ranging from 3.6 minutes to 7.61 minutes, without
justification. In his original model submitted to the CEO prior to the oral hearing,
Mr Makate used 1.6 minutes in line with published data which showed a duration
consistently below two minutes. He subsequently changed the call duration to 3.6
minutes without stating its source or rationale.

[155] However, Mr Makate’s model 9A applied a 6.8-minute call duration for the 2002
financial year, based on United States data. This increased the estimated revenue in
his model by as much as four times. Such a dramatic increase in call duration would
 52

have been borne out by a similar increase in Vodacom’s voice revenue. But that did
not happen.

[156] The CEO concluded that international data was irrelevant. Published local
numbers are very different and are known. That data had been made available by
Vodacom from which the average duration of calls could be calculated – less than two
minutes per customer. The CEO’s assessment of the call duration cannot be faulted.
In any event, Mr Makate accepted that the average call duration was two minutes.

(v) Effective rate

[157] The CEO applied a rate determined by dividing the prepaid voice average
revenue per user (ARPU) by the average number of minutes per user per year. He
said that the rates applied in Mr Makate’s model were incorrect:
‘As I indicated in my Determination, the [blended] effective rates used in Mr Makate’s
model are incorrect. Mr Makate’s model used a blended contract and prepaid rate.
The contract calculated rate includes the monthly committed subscription fees where
a customer pays a fixed amount for a recurring fixed amount of allocated
minutes/SMSs. A portion of the subscription charge to the customer relates to charges
for the recovery of the cost of the handset over the contract period. It is best to use the
average prepaid rate as the rate does not include any device recovery. This is the rate
that best reflects a charge for a pure voice call.’

[158] Mr Makate’s incorrect rate produced an outcome more than double the true
effective rate. This has the effect of substantially increasing the revenue ascribed to
an assumed call back from a PCM.

[159] The high court merely described the debate between the parties on the ‘blended
effective rate’ but did not engage with it. It held that because Mr Makate contended
that his model applied rates that were guided by ICASA rates, the court was ‘bound to
apply’ these rates, since ‘ICASA is the regulating authority which ensures
transparency’. It concluded that ‘the contention that the rate does not exceed that of
ICASA is sound’.

[160] Despite giving no reason for preferring a blended rate, and contrary to its
conclusion on the ICASA rate, the high court ordered the CEO to determine the
 53

effective rate on the blended basis advanced by Mr Makate ‘including prepaid, contract
(both in and out bundle) and interconnect (MTR) fees’. Further, it directed that the
‘contract effective rate and prepaid effective rate . . . are not to be less than the
published ICASA effective rate’.

[161] There is thus no justification for the orders in paragraphs (3) and 3.2 of the high
court’s order.45 The CEO’s determination on the effective rate is reasonable.

(vi) Paragraph 2(b) of the high court’s order

[162] In the founding affidavit, Mr Makate asserted that the total voice revenue
provided by Vodacom was the starting point of the CEO’s determination; that it was
materially understated; and that this understatement was achieved by ‘simply
excluding all in-bundle revenue from contract customers, so that the total revenue
used by the CEO was only that from prepaid airtime and out-of-bundle airtime.’ Mr
Makate then asserted that the CEO applied various percentage factors to that revenue
to calculate the part of the revenue he considered to be attributable to PCM. Mr Makate
claimed that these assertions are supported by Prof Wainer, a forensic auditor and
professor of accounting.

[163] In the answering papers it is denied that Vodacom understated the revenue in
any manner. In amplification of that denial, Vodacom states the following. Vodacom
furnished the source information used to prepare the voice revenue tables forming part
of its post-hearing submissions. These tables contain both prepaid and contract
customers’ outgoing voice revenue in South Africa and voice ARPU for the years
ended 31 March 2001-2018. The voice revenue was furnished from detailed internal
management accounts and excluded voice revenue generated by customers when
calling international numbers or roaming abroad, and subscription revenue from
contract customers based on a fixed amount of allocated minutes and/or data and/or
is SMSs. The CEO explained to Mr Makate’s counsel at the hearing that it is wrong to
conclude that such revenue was as a result of contract customers receiving PCMs.

45 Quoted in para 66 above.

 54

[164] The CEO’s answer to Prof Wainer’s comments on how and when in-bundle and
out-of-bundle and prepaid mobile voice revenue is earned; whether these revenue
sources are distinguishable; and whether there is any basis to exclude in-bundle
revenue in the calculation of incremental revenue attributable to PCM, is that these
considerations are irrelevant. This is because in determining compensation, the CEO
did not take them into account as they would have had negative effect on his
determination in relation to Mr Makate. The CEO says that he did not use the total
voice revenue provided by Vodacom as the starting point of his determination.

[165] What is stated above are further instances in which the Plascon-Evans rule was
ignored. Vodacom’s version or that of the CEO on these aspects is neither far-fetched
nor untenable.

[166] In paragraph 2(b) of its order, the high court directed the CEO to base his
calculation on the ‘total voice revenue including PCM revenue’ derived from
Vodacom’s annual financial statements. But total voice revenue and Vodacom’s
annual financial statements say nothing about incremental revenue. They are entirely
irrelevant to the exercise.

[167] Apart from this, Mr Makate’s allegation that Vodacom’s contingent liability
disclosure in its audited financial statements is evidence of a large provision for
payment to Mr Makate, because an insignificant amount would not have warranted
such a disclosure, is refuted in the affidavit by the Executive Head for International
Financial Reporting Standards and Compliance for the Vodacom Group Limited. She
states the following. Based on the evaluations performed under IAS 37, Vodacom
disclosed the PCM matter as a contingent liability in its annual financial statements,
following the Constitutional Court’s 2016 order. This disclosure was driven by
qualitative factors. No amount was provided for because IAS 37 states that an entity
shall not recognise a contingent liability, and the disclosure does not imply a specific
amount regarding Vodacom’s materiality thresholds. The contingent liability –
disclosed but never recognised – continued to be disclosed in the Vodacom Group’s
and Vodafone’s annual financial statements until March 2018. Since the disclosure
never originated from a quantitative perspective, the materiality thresholds disclosed
 55

in the Vodacom Group and Vodafone’s financial statements are irrelevant. Yet again,
the Plascon-Evans rule was ignored.

[168] The issue giving rise to paragraph 2(b) of the court’s order, came about as
follows. In his determination, the CEO compiled a table (Table 1) in which he compared
the ‘PCM revenue’ claimed in Mr Makate’s model to Vodacom’s outgoing voice
revenue. By contrasting these figures, the CEO demonstrated that according to Model
8A, up to 97% of Vodacom’s outgoing voice revenue was attributable to PCM. He
conducted this exercise merely to show that Mr Makate’s Model 8A cannot be
reasonable when compared to the outgoing voice revenue of the company, and that
one cannot rely on his assumptions.

[169] Mr Makate missed the CEO’s point. He accused the CEO of ‘deliberately
excluding vast amounts’, by not applying the mobile voice revenue figures reflected in
Vodacom’s annual financial statements. He complained, for example, that the CEO’s
figures in Table 1 excluded outgoing call revenue.

[170] What Mr Makate overlooks, is that the CEO did not use these outgoing mobile
voice revenue figures in his determination of Mr Makate’s compensation at all. Neither
outgoing voice revenue nor ‘interconnect minutes’ were inputs in any of his models. It
had no impact on his determination. In that determination the CEO describes the four
models he considered, the assumptions made and the methodology applied. Mr
Makate’s criticism is thus baseless.

[171] Mr Makate’s reliance on Vodacom’s financial statements is also flawed.

Vodacom’s Managing Executive for Data and Analytics (its former Finance Director)
explained that the published revenue data included voice revenue generated by
Vodacom’s international operations, international roaming, calls terminated
internationally, and fixed subscription charges. Logically, these categories of voice
revenue are not generated by PCM.

Conclusion on PCM incremental revenue and the backward looking share model
[172] Whether a return call earned incremental revenue was an ‘imponderable’ that
particularly troubled the CEO. There were not only conceptual difficulties in the
 56

exercise, such as determining out of millions of calls, whether a return call in response
to a PCM would have been made in any event, its duration, the revenue it generated
and the rate payable, but also other factors influencing the revenue generated by
Vodacom, such as network investment, other products and services, customer device
investment and customer growth. The CEO’s statement that determining PCM
incremental revenue was ‘an extremely complex exercise which defies any exact
computation and must rely on assumptions’, is thus not surprising.

[173] In his determination, the CEO demonstrates that the numbers resulting from
Mr Makate’s Model 8A are unrealistic, and for this reason alone, no reliance can be
placed on it. On Mr Makate’s figures, 97% of Vodacom’s total voice revenue in 2002
and 2003 was as a direct result of PCM. This cannot be correct The assumptions used
in Mr Makate’s model are unreasonable and unjustifiable. The number of customers,
PCMs sent and customer base are unrealistic. The call rates, call duration and
effective rates, are wrong. The assertion in the first judgment that Vodacom presented
no evidence to refute or contradict Mr Makate’s information showing the revenue
generated by Vodacom from the PCM idea, is incorrect.

[174] For the reasons advanced above, it cannot be said that the CEO’s
determination of PCM incremental revenue is unreasonable or patently inequitable.
The backward looking share model is fundamentally flawed because it is inconsistent
with the contract upheld by the Constitutional Court. It is also at odds with the
agreement between the parties that the CEO should make a forward-looking
determination.

The duration of the contract

[175] In his explanatory affidavit the CEO states that for the most part, he approached
the duration of the contract as a separate issue. This issue, and the revenue share of
5%, were common to the first, third and fourth models which the CEO considered in
determining reasonable compensation due to Mr Makate.

[176] The CEO’s determination states that a reasonable basis for deciding duration
is to look at typical time periods governing the provision of services and products to
 57

Vodacom by third parties. He gives the following terse reason for his view that a CEO
in 2001 would have awarded a contract for a maximum period of five years:
‘The typical duration for an untested new product in the Vodacom stable would be 3 years and
the longest, plausible initial period would be 5 years. Here there is much to say in favour of
the 3 year period, but I am inclined to allow a 5 year period which I think is the outer limit of
what could realistically be contemplated. I do so because whilst a CEO looking forward would
not have committed beyond 3 years, I have allowed for 5 years which I believe is generous.’

[177] The CEO however went further in his explanatory affidavit. He said that
according to his knowledge and experience, Vodacom did not bind itself to new
products for more than three years, due to their uncertain commercial viability. Where
the product or service was well established, the tenure of the contract could be three
or possibly five years. Longer term contracts usually contained clauses which allowed
Vodacom to review the commercial terms on short notice.

[178] Had he been called upon to break the deadlock between the parties in 2001,
the CEO said that he ‘would have considered both the opportunities and risks identified
in the key documents of 2001’. One of these risks would have been the commercial
viability of PCM. The initial proposal was to levy a charge of 15 cents for each PCM
message. However, Vodacom’s competitor, Mobile Telephony Networks (Pty) Ltd
(MTN), had a comparable product for free, which introduced an element of uncertainty
into the commercial viability of PCM. He would therefore have included some flexibility
in the commercial terms of the contract.

[179] A further risk, the CEO said, was the possibility that PCM would have had a
negative impact on Vodacom’s SMS revenue and outgoing call revenue, particularly if
it was being offered as a free service. Another uncertainty was whether PCM was to
be targeted to time window lock customers or whether it would apply more generally.
On the other hand, PCM was a good idea and the CEO in 2001 would have supported
its launch in the expectation that it would be beneficial to customers.

[180] The CEO goes on to say that a CEO in 2001 would not have entered into a
contract of indefinite duration with Mr Makate, and ‘would never have committed
Vodacom to a contractual period of 18 years or anything like it’. Consequently, he
contends that a five-year contract duration is generous. At the time of the commercial
 58

review (at the earliest after a year to assess commercial arrangements going forward),
the relevant CEO would have known that there was no prospect that Vodacom would
extend duration of the contract, because Mr Makate was not making any contribution
and would have had the benefit of 5% of PCM revenue for five years, which would
have been regarded as fair compensation for the initial PCM idea.

[181] He was ‘satisfied’, the CEO said, ‘that the answer lay in assessing whether,
given the predicted revenue set out in Vodacom’s ‘Call-Me Service Product
Description Document dated 22 January 2001 (PDD), a share of 5% over 5 years
would amount to a sum of money which would cover reasonably and would be
regarded as fair compensation for the initial PCM idea’. In his affidavit the CEO states
that Vodacom’s ‘Product and Service Approval: USSD or Call Me Service dated 1
March 2001 (USSD) shows that the PCM service at no charge would dilute SMS and
outgoing call revenues, and result in the product making a loss. This would have been
a significant factor when reviewing the product and the agreement with Mr Makate at
the time. In his determination the CEO states that the PDD and USSD are very similar.

[182] However, none of these considerations stated in the explanatory affidavit are
to be found in the CEO’s determination. The inference is inescapable that he did not
direct his mind to these factors. And Vodacom’s submission that the CEO ‘did not
advance new reasons for his determination’ and merely explained his reasons in
response to Mr Makate’s criticism, is incorrect. The CEO states unequivocally that he
‘was satisfied’ that had he been asked to break the deadlock in 2001, he would have
approached the matter on the basis of the new facts set out in his affidavit. 46

[183] To begin with, concerning the duration of the contract, there is no reference in
the CEO’s determination to a single 2001 document – the PDD and USSD are referred
to on the question whether the CEO’s determination had to be forward-looking. There
is no inkling in the CEO’s determination, of the commercial viability of PCM in relation
to the duration of the contract. Neither is there any reference to the fact that MTN,
Vodacom’s competitor, had a comparable product – which would have been a critical
factor in the decision to launch or continue with PCM. Moreover, the CEO’s
determination says nothing about the review of the contract by the relevant CEO, nor

46 Emphasis added.
 59

any risk of PCM having a negative impact on Vodacom’s SMS and outgoing call
revenues.

[184] The PDD provides a complete answer to the following allegations by the CEO.
The commercial liability of PCM was at risk, because of the proposed charge of 15
cents for each PCM message, and MTN’s comparable product offered as a free
service. Vodacom would not have extended the duration of Mr Makate’s share
because he was not making any ongoing contribution to the service. Vodacom would
have reviewed the PCM product typically after a year to assess its success,
commercial arrangements, support and investment going forward. All of these
allegations are afterthoughts.

[185] The PDD states that MTN was not offering a similar service, and that PCM had
to be launched urgently with a minimum configuration server, and without a full
production infrastructure. The PDD concludes:
‘On the basis of the [revenue] model described in Section 2.3 and the implications in terms of
both development and hardware costs that would be necessary to modify either HLR, the
provisioning systems and processes and/or the USSD Server (which could as a rough
estimate exceed the R5 million figure), we recommend that we launch the service in the next
two weeks, using the existing infrastructure and solution.’

[186] Then there is the CEO’s reliance on the USSD – a contemporaneous document
approving the launch of the PCM product on 1 March 2001 – not referred to at all in
his determination on the duration of the contract. It is true that the USSD shows that
the PCM service at no charge would dilute Vodacom’s SMS and outgoing call
revenues. But it contains no hint of PCM making a loss.

[187] What is more, the CEO’s reliance on the USSD is selective. It shows that PCM
was launched to remain competitive with MTN and despite commercial risks such as
the devaluation of SMS and voicemail deposits. It states the following:
(a) To ‘stimulate uptake in the market’ it is proposed that the ‘nominal fee of R0.15
(Vat inclusive) per SMS’ be waived, and that subscribers would be charged at
this rate after the initial period of six months.
 60

(b) MTN did not have a similar service available (also stated in the PDD). Of
course, if it did, the PCM idea would not have been ground-breaking.
(c) Vodacom estimated that ‘60% of the Vodago subscriber base (65%)’ would
utilise the PCM service ‘to indirectly stimulate traffic onto the network by
means of revenue generated from all returning calls’; and it is assumed that
40% of the active prepaid subscriber base would make use of the service on
a continuous basis.
(d) The revenue figures show that despite a decline in outgoing call and SMS
revenues, the additional incoming call revenue (R78.4 million) and the nett
additional revenue (R25.4 million) in Scenario A (Free Service) are
significantly higher than their equivalents, R65.3 million and R16.1 million
respectively, in Scenario B (15c per SMS); and R25.1 million and R6.7 million
respectively, in Scenario C (40c per SMS).
(e) The PCM service should be approved for implementation with effect from
1 March 2001.

[188] The USSD also demonstrates that Vodacom implemented the service
regardless of the risks and concerns outlined in that document. For example, the
financial impact of PCM had only been estimated and no thorough research was
conducted. A detailed analysis of all capital expenditure requirements had not been
fully understood. No feasibility studies had been performed in order to obtain a better
understanding of customer behaviour patterns to offset the concerns in relation to the
decline in outgoing call and SMS revenues.

[189] In these circumstances, Vodacom would never have launched the PCM service
– which was untested – unless it was financially viable. PCM was, by Vodacom’s own
admission, a triumph,47 and contradicts the CEO’s muted description of it as a ‘good
idea’ that would ‘benefit customers’. The CEO ignored Vodacom’s newsletter – quoted
by the Constitutional Court – which states:
‘“Call Me” has been a big success. On the first day of operation about 140 000 customers
made use of the service. It will be free until December 31 this year and thereafter will cost
users 15 cents per transaction.’48

47 Makate v Vodacom fn 23 para 113.

48 Ibid para 7.
 61

[190] PCM however continued as a free service. Given the facts stated in the USSD,
and the success of the PCM service – all of which would have been known to a CEO
in 2001 and whenever the contract came up for review, if in fact it did – it is extremely
unlikely that the contract would have endured for five years at most. So viewed, the
CEO’s claim that a tenure of five years is ‘generous’, does not even arise.

[191] In addition, logic and business sense dictate that after the launch of PCM, and
as it continued being offered as a free service, Vodacom would have addressed the
risks and concerns stated in the USSD, and continued to monitor the success of the
PCM service. If it was not commercially viable, it would have been discontinued. This,
however, is not to say that Mr Makate was ‘entitled to an ongoing revenue share in
perpetuity’, unless PCM was discontinued, as his counsel submitted. There was no
such term in the parties’ agreement. Still, in light of the obvious success of the PCM
service, it would have been irrational for Vodacom to terminate it after three (or five)
years.

[192] What all of this shows, is that the CEO’s determination as to the duration of the
contract is unreasonable and patently inequitable. The high court’s finding that he had
advanced further reasons for that determination after the event, was correct.
Vodacom’s submission that the high court had mistakenly applied the administrative
law rule that a public body may not justify its decision with new reasons after the event,
is not the point. The facts stated in his affidavit – entirely new – do not form the basis
of the CEO’s determination that the contract between the parties would have come to
an end after five years. They do not constitute the reasons for that determination. The
decision on the duration of the contract is insupportable on the facts, and the CEO’s
version on this score is plainly untenable.

[193] There is a further reason why the CEO’s determination concerning the duration
of the contract is unreasonable and inequitable. It is the principle articulated in Bwllfa,49
which concerned the estimation of lost profits by an arbitrator. Lord MacNaghten said:
‘[T]he arbitrator’s duty is to determine the amount of compensation payable. In order to enable
him to come to a just conclusion it is his duty, I think, to avail himself of all information at hand

49The Bwllfa and Merthyr Dare Steam Collieries (1891) Limited v The Pontypridd Waterworks Company
1903 AC 426 (HL) at 431.
 62

at the time of making his award which may be laid before him. Why should he listen to
conjecture on a matter which has become an accomplished fact? Why should he guess when
he can calculate? With the light before him, why should he shut his eyes and grope in the
dark?’

[194] This passage was cited with approval by the Constitutional Court in New
Clicks.50 The Court noted that similar reasoning had been applied in South African
courts.51 In Fundsatwork,52 Wallis JA tersely stated the Bwllfa principle thus: ‘In judicial
proceedings facts are preferred to prophecies’.

[195] Applied to the present case, it is true that the CEO’s determination of
reasonable compensation had to be forward-looking, ie from the perspective of the
CEO in 2001. Principle, consistency and fairness dictate that factors existing at that
date should not be ignored, and factors which occurred afterwards should not be taken
into account. However, in my view, and given the particular circumstances of this case,
the Bwllfa principle does not require the CEO, when determining compensation, to
‘guess at something which events have made certain’.53

[196] It is accordingly insensible and unnecessary to resort to guesswork as to the

duration of the contract. The CEO’s determination on this issue has been shown to be
unreasonable. There is no reason why Mr Makate should be content with a contract
for only five years. The commercial viability of PCM is a proven fact. It is common
ground that Vodacom has used the PCM service for more than 20 years and continues
to do so. Mr Makate has stated that he is willing to accept a contract duration of 18
years. That, I think, is both reasonable and fair.

[197] It follows that the high court’s order that Mr Makate be paid 5% of the total voice
revenue generated by the PCM product for 20 years, must be set aside. The issue of
the duration of the contract must be remitted to the CEO to redo his determination in

50 Minister of Health v New Clicks [2005] ZACC 14; 2006 (2) SA 311 (CC); 2006 (1) BCLR 1 (CC) para
19.
51 Devland Investment v Administrator, Transvaal 1979 (1) SA 321 (T) at 327-8.
52 Fundsatwork Umbrella Pension Fund v Guarnieri and Others [2019] ZASCA 78; 2019 (5) SA 68

(SCA) para 23.

53 Amec Developments Ltd v Jury’s Hotel Management (UK) Ltd (2001) 82 P & CR para 13.
 63

accordance with the models that he considered when making the determination dated
9 January 2019, save that he must allow for an 18-year contract period.

Mora interest and the time value of money

[198] As appears from the CEO’s determination and his affidavit, he dealt with
Mr Makate’s claim for interest. He allowed for the delay in the implementation of the
parties’ agreement by escalating the amount of his award, on an annual basis, by the
average rate of inflation from 2002 to 2019. He did this to allow for the time value of
money.

[199] The CEO noted in his determination that Mr Makate’s counsel seemed to have
abandoned the claim for interest, which had been retained in his expert report. The
CEO realised afterwards that he had been mistaken that counsel had abandoned
Mr Makate’s interest claim, for which he apologised. But the mistake did not affect his
determination.

[200] Mora interest and an allowance for the time value of money are in any event
two methods of compensating a creditor for the delay in paying its claim.54 The CEO’s
adjustment for the time value of money was generous to Mr Makate because he was
probably not entitled to mora interest at all. Awarding both would have amounted to
double compensation and would have been clearly wrong.

[201] The high court’s finding that the CEO conflated mora interest with the time value
of money is unfounded. The CEO denied Mr Makate’s allegation that he ‘may have
been confused between interest on an unliquidated claim for damages and mora
interest’. The CEO explained that the Constitutional Court had stated when interest is
due – until negotiations on reasonable compensation had been concluded, there was
nothing due by Vodacom to Mr Makate and nothing that he could claim. 55 The CEO
was correct. It follows that the concession by Vodacom’s counsel in the high court that

54 Drake, Flemmer & Osmond Inc v Gajjar NO [2017] ZASCA 169; [2018] 1 All SA 344 (SCA); 2018 (3)
SA 354 (SCA) paras 56-68.
55 Makate v Vodacom fn 23 paras 92 and 186.
 64

Mr Makate was entitled to mora interest is neither here nor there. In any event, a court
is not bound by a wrong concession of law.56

[202] For these reasons, the high court’s conclusion that the CEO’s mistake
concerning interest was a ‘classic case’ of bad reasoning which vitiated his
determination and justified its review, is incorrect. The mistake was not a reason for
CEO’s determination at all. Moreover, the rule that one bad reason vitiates a decision
is a principle of administrative review and inapposite in the application of the Bekker
test.

Conclusion
[203] Mr Makate failed to establish that in his determination of reasonable
compensation for the PCM idea, the CEO committed the irregularities complained of,
except in relation to the tenure of the contract. This irregularity renders the CEO’s
determination unreasonable and patently inequitable. It has a significant impact on the
award, because the duration of the contract was an assumption the CEO made in his
first, third and fourth models, in arriving at his final determination of compensation of
R47 million.

[204] It follows that this issue must be remitted to the CEO for him to determine
reasonable compensation afresh. The parties do not object to remittal. In its
submissions Vodacom asked for such an order in the event of its appeal failing.
Although Mr Makate sought in the high court, alternative relief to the effect that the
total revenue of the PCM product be determined by a new referee, he could have no
objection to remittal to the CEO: the issue is one of pure calculation.

[205] Vodacom has been partially successful in its appeal. It demonstrated, save for
the irregularity relating to the duration of the contract, that the high court erred in
upholding all Mr Makate’s review grounds. However, Vodacom failed on the central
issue: whether the CEO’s determination is unreasonable or patently inequitable. There

56Paddock Motors (Pty) Ltd v Igesund 1976 (3) SA 16 (A) at 23D-24G, affirmed in Alexkor Ltd and
Another v Richtersveld Community and Others [2003] ZACC 18; 2004 (5) SA 460 (CC); 2003 (12) BCLR
1301 (CC) para 43.
 65

is accordingly no reason why the costs of appeal should not follow the result. The
parties agreed that the costs of three counsel on appeal, are warranted.

[206] I would make the following order:

1 The appeal succeeds in part.
2 The appellant is ordered to pay the costs of appeal, including the costs of three
counsel, namely the costs of two senior and one junior counsel.
3 Paragraphs (2), (3), (4), (5) and (7) of the high court’s order are set aside and
replaced with the following:
‘(a) The second respondent’s Determination of Reasonable Compensation due to Mr
Kenneth Nkosana Makate arising from the Please Call Me Idea, dated 9 January
2019, is reviewed and set aside.
(b) The matter is remitted to the second respondent for him to determine the amount
of reasonable compensation due to the applicant, in accordance with the four
models in his determination of 9 January 2019, solely on the basis that the
second respondent would have awarded the applicant a contract for 18 years,
commencing on 1 March 2001 and terminating on 28 February 2019.
(c) The award made by the second respondent shall take into account the time value
of money, calculated at an average inflation rate of 5%, from 1 March 2001 to 28
February 2019.
(d) The first respondent shall finalise his determination within 30 calendar days of
the date of this order.
(e) The second respondent shall pay the costs of the application; which costs shall
include the costs of two counsel.’

__________________
A SCHIPPERS
JUDGE OF APPEAL
Appearances

For the appellant: W Trengrove SC (with R Solomon)

Instructed by: Leslie Cohen & Associates, Johannesburg
Lovius Block, Bloemfontein

For the respondent: C Puckrin SC (with G Marcus and R Micham)

Instructed by: Stemela & Lubbe Inc., Pretoria
Webbers Attorneys, Bloemfontein.