THE SUPREME COURT OF APPEAL OF SOUTH AFRICA
Case No: 152/04
REPORTABLE
In the matter between
André Gründlingh First Appellant
Ulrich Osmund Schüler Second Appellant
Turfsport CC Third Appellant
and
Phumelela Gaming and Leisure Limited Respondent
Before: Howie P, Farlam, Conradie, Lewis JJA et Comrie AJA
Heard: 10 March 2005
Delivered: 1 June 2005
Summary: Bookmakers and totalizator – Gauteng Gambling Act 4 of 1995 –
Bookmakers accepting ‘exoti c’ bets – winnings dependent on
tote’s published dividends for same bets – whether ‘fixed odds
bets’ as defined – whether unlaw ful competition. Order in
paragraph [44].
______________________________________________________
JUDGMENT
______________________________________________________
COMRIE AJA
2
[1] Litigation in this country a bout the totalizator goes back to
1887: Day v Cloete 5 SC 139. See too Brady v SA Turf Club 23 SC
385. It reached this court in 1914 in the leading case of R v Williams
1914 AD 460. There this court adopt ed the principle that a power
conferred to the Cape Provincial Council to regulate horse-racing
and betting did not empower the Council to prohibit betting in a
partial but most substantial manner. The Council was consequently
not empowered to prohibit bookmaking on horse-racing or to confine
betting to the totalizator.
[2] The same is true of England, starting with Tollett and another v
Thomas 6 QBD 518, decided in 1871. Many of the decided cases
are collected in the comprehensive judgment of Ogilvie
Thompson AJ in Rex v Sportspools (Pty) and others 1949 (2) SA
202 (C). These include the decisi on of the House of Lords in
Attorney-General v Lunche on and Sports Club Ltd 1929 AC 400;
151 LTR 153 (HL). The litigation continues. See eg Queens
Bookmakers Ltd v Commissioner of Customs and Excise 1975 SLT
207; Tote Investors Ltd v Smoker [1968] 1 QBD 509; Christie NO v
Mudalier 1962 (2) SA 40 (N).
[3] Now bookmaking and the totalizat or once again feature in an
appeal to this court. The first and second appellants are licensed
3
bookmakers who carry on business from premises, called a
Tattersalls, inter alia in Silverton, Pretoria, Gauteng. The respondent
is a public company which owns and carries on the business of
operating a computerised totalizat or and also running thoroughbred
horserace meetings. It operates nationally under the brand name
Saftote. Gold Circle (Pty) Ltd is licensed to operate a totalizator in
Kwa-Zulu Natal and the Western Ca pe. Its totalizator betting pools
are commingled with those of the respondent under the name
Saftote, for which service Gold Circle pays the respondent a fee.
Saftote has some 2 600 betting terminals, on and off course,
throughout the country. The turnover is about R150 million per
month. In the Pretoria High Court, at the instance of the respondent,
Mynhardt J granted with costs an interdict (in amended form)
restraining the appellants from:
‘1.1 Breaching Section 55 of the G auteng Gambling Act 4 of 1995 by offering
or receiving bets which are not “fixed odds” bets.
1.2 Prohibiting the respondents from offeri ng bets to the public or taking bets
from the public in terms whereof or on the basis whereof bets are offered
to be paid or are paid out on the basis of result and/or dividends derived
or obtained from the applicant’s totalisator pool.’
[4] The first and second appellants appeal to this Court with
limited leave granted by the court a quo . An interdict was also
4
granted (para 2 of the order) against the third appellant (Turfsport
CC) which is owned by the first and second appellants. That part of
the order is not the subject of appeal and nothing further need be
said about it. For convenience I sha ll refer in this judgment to the
first and second appellants as the appellants.
[5] As licensed bookmakers in terms of the Gauteng Gambling
Act 4 of 1995 the appellants are aut horised to accept ‘fixed odds
bets’ on sporting events (s 55), in this case horse races. To carry on
unlicensed bookmaking is an offence (ss 54, 87). The definition of
such bets (s 1), as it stood when this matter was decided a quo ,
read:
‘‘Fixed odd bets’ 1means a bet taken by a licensed bookmaker on one or more
events or contingencies where odds are agreed upon when such bet is laid, but
excludes a totalisator bet;’
A totalizator bet was not defined, but a ‘totalisator’ was:
‘‘totalisator’ means a system of betting on a sporting event in which the
aggregate amount staked on such event or combination of events, after
deduction from such aggregat e amount of any amounts which may lawfully be
deducted therefrom, whether under this Act or by agreement, is divided
1 The spelling in the statute is anything but consistent. Except for quotations I shall spell totalizator with a
‘z’ (see the Shorter Oxford English Dictionary), and I shall refer to fixed odds in the plural. The accepted
abbreviation of totalizator is ‘tote’, which I shall use from time to time.
5
amongst those persons who have made winning bets on that event or
combination of events in proportion to the amounts staked by such persons in
respect of such winning bets, and includ es any scheme, form or system of
betting, whether mechanically operated or not, which is operated on similar
principles.’
[6] An obvious example of a fixed odds bet with a bookmaker
would be 5 to 1 for a win on the celebrated race horse Appeal Court.
The odds (5 to 1) would be fixed at the moment when the bet is laid,
and the payout in the event of Appeal Court winning would be
calculable at the same moment. We know from the evidence that the
odds given by bookmakers may sh orten or lengthen as the race
approaches and that diff erent bookmakers may offer different odds.
In the event of Appeal Court winning, the bookmaker must pay all
the winning bets on that horse from his own resources – save to the
extent that he may have ‘laid off’ such bets. He takes the betting
risk. There is no pool of bets to be divided among successful
punters. The agreed odds rule.
[7] As the definition indicates, the totalizator operates on
principles different from those described in the preceding paragraph.
All the bets (tote bets) on a particul ar race (eg all the bets for a win
in race no 1) are pooled. From the res ulting (gross) pool or total, tax
6
and administration expenses (which include the profit of the tote
operator) are deducted. T he net pool is divide d equally between all
the successful punters in proportion to their respective stakes. There
is no betting risk to the totalizator or its operator. Subject to lawful
deductions, the tote pays out in winnings (or dividends) what it has
received in bets. As it was succinctly put in the papers (in lay terms)
punters on the tote bet against each other; (whether that is the
correct legal position is un necessary to decide – cf Tote Investors
Ltd v Smoker supra); whereas a punter pl acing a bet with a
bookmaker bets against that bookma ker. It is also clear that the
odds on a tote bet are not fixed when the bet is laid because no
odds are agreed. On the contrary, everything depends on how much
money is wagered on th e race, via the tote , and on how many
winning tickets there are. The divide nd can only be calculated after
the race has been run.
[8] The essence of a tote bet – ie a bet on the totalizator – is the
pool system, namely that each pool is divided among the successful
punters. That essence is reflected in the definition of a totalizator set
out above. It also accords with t he descriptions and analyses of the
totalizator to be found in several of the decided ca ses cited in the
7
opening paragraphs of this judgmen t and in the judgments referred
to therein.
[9] The totalizator has long since moved from straightforward bets
for a win or a place. So-called ‘exotic’ bets are offered. One such is
the ‘trifecta’ in which the punter selects three horses to finish first,
second and third in the correct ord er. In a ‘trio’ bet, the punter
selects the first thr ee horses to finish in any order. There are other
permutations such as the dupla, t he exacta, the jackpot, the pick 6,
the place accumulator, the quartet and the swinger. The dividends
on such tote bets are likely to be higher than for a simple place or
win. These exotic bets also operate on the pool system which I have
described.
[10] Once the race has been run, the totalizator calculates the
dividends and these are announced. In respect of a bet which spans
more than one race (eg the jackpot), the calculation takes place
after the last relevant race. T he respondent publishe s these results
widely: on the course, in the pres s, on some radio channels, and on
a dedicated television channel. The development and operation of
its totalizator, and the publication of its results, have involved the
respondent in considerable effo rt, expertise and expense. The
8
system is now so sophisticated that it even forecasts some
dividends before the race.
[11] It came to the notice of th e respondent that the appellants
were offering and accepting ‘exotic’ bets, the stake formula being
the tote dividends or results to be published by the respondent. Thus
the dividend per rand on a trifecta bet laid with the appellants would
be precisely the same as the divide nd on the identical trifecta bet
laid with the respondent’s tote. T here is this difference, however:
with the appellants there is no poo l of bets to be divided among
winning punters. The appellants are on risk and must pay all winning
bets from their own resources. Re solving to put a stop to the
practice of bookmakers accepting such exotic bets, the respondent
brought the present application in the court a quo.
[12] Para 1 of the order granted by Mynhardt J was premised on
his conclusion that the exotic bets laid by the appellants were not
fixed odds bets and that the appellants were accordingly not
authorised by their licence to ac cept them. Para 2 of the order was
premised on the conclusion that by using the respondent’s published
dividends or results as ingredients of such bets, the appellants were
guilty of unlawful competition. Bo th conclusions are challenged on
appeal.
9
[13] While this appeal was pending the National Gambling Act 7 of
2004 came into force. Its provisions were urged upon us by counsel
for the appellants, Mr Puckrin, in two ways. First, he submitted that
certain definitions and other prov isions relating to the permissible
activities of licensed bookmakers ov erride any contrary stipulation in
the Gauteng law. He referred us to the definitions of ‘bookmaker’,
‘fixed-odds bet’, and ‘open bet’. He also referred us inter alia to ss 4,
8, 30, 37 et seq, 44, and the schedul e (transitional provisions), more
particularly s 2(2). In the view whic h I take of the first leg of this
appeal, the fixed odds bet question, it is unnecessary to resolve the
contention. I shall content myself with the prima facie dictum that
Act 7 of 2004 relates to a concurrent legislative competence and is
largely permissive, and that the pro vinces can continue to grant or
refuse bookmaking licences as before, provided certain minimum
standards are observed. Second, Mr Puckrin submitted that the
apparent endorsement by the national legislature of the type of
betting now under review was relevant to this court’s assessment of
the boni mores of the community in relation to leg 2 of the appeal,
viz the unlawful competition issue. I will weigh this contention later. I
turn now to consider in turn the two legs of the appeal.
10
Fixed Odds Bets
[14] It is clear to me that the exotic bets under review are not tote
bets. This is because t he appellants do not ope rate a ‘totalisator’ as
defined. They maintain no pool s of bets to be divided among
winning punters, nor do any actual divi sions of this kind take place.
As I have already recorded, the app ellants are on risk and must pay
all winning bets from their own resources. On any given race or
combination of races, they may show a profit or a loss. That is quite
different to a totalizator which runs no betting risk and which shows
no betting loss or profit. It follows that the bets under review are not
outlawed by the concluding words of the definition of fixed odds
bets: ‘but excludes a totalisator bet’. Indeed, the presence of these
words of exclusion seems to suggest, at least prima facie, that tote
bets, or some tote bets, are fixed odds bets, hence the legislative
wish to exclude them for one or other reason of policy.
[15] Turning to the rest of the definition of fixed odds bet, the
critical words are: ‘odds are agreed upon when such bet is laid’. (My
emphasis). Earlier in this judgment I gave as an obvious example of
a fixed odds bet, a bet for a win at 5 to 1. I observed that the odds
(5 to 1) would be fixed at the mom ent when the bet is laid, and that
the payout in the event of a win would be calculable at the same
11
moment. But immediately calculable winnings are not expressly
required by the definition. In t he course of argument Mr Puckrin
offered a betting example drawn from the game of cricket. I would
prefer one drawn from a different sport, golf. Suppose the following
wager: in return for a stake of R30, the bookmaker promises to pay
the punter R10 for each birdie scored by a well known professional
golfer in the third round of a spec ified tournament. In such a wager
the odds in my view are indubitably agreed or fixed (R10 per birdie)
when the bet is laid. The potential winnings (or loss) are not
immediately calculable, but they can be readily calculated on a
permutation basis. So I do not th ink that my hypothetical wager
would fall foul of the definition.
[16] With the appellants’ exotic bet s, the odds are not known, or in
that sense fixed, when such bets are laid. From that moment until
the race or combination of races is run, the odds fluctuate according
to the amounts of money which are wagered (on the tote) and the
horses which are selected. Thus th e appellants’ exotic bets do not
fix the odds; they provide formulae in terms of which the odds will be
determined or ascertained later. Such a formula arguably, in my
view, constitutes an agreement upon the odds, at the moment when
the bet is struck, even though th e actual odds are determined later
12
(in terms of the formula). I think that is a permissible interpretation of
the definition of a fixed odds bet, which insists on agreed odds when
the bet is laid rather than known odds. I conclude therefore that the
definition is ambiguous.
[17] I mention in passing that a difference between an immediately
calculable payout in the event of a win, and an ascertainable payout
in due course, is not unknown to f oreign legislators. In the Scottish
case of Queens Bookmakers Ltd v Commissioner of Customs and
Excise, supra, the statutory definition of a fixed odds bet (for tax
purposes) expressly acknowledged the difference. Section 10(2) of
the Betting and Gaming Duties Act 1972 defined a fixed odds bet, as
distinct from a pool bet, as follows:
‘(2) A bet is a bet at fixed odds within t he meaning of this section only if each
of the persons making it knows or c an know, at the time he makes it, the
amount he will win, except in so far as that amount is to depend on the
result of the event or ev ents betted on, . . . or on the starting prices . . .
for any such event . . . .
In this sub-section –
“starting prices” means, in relation to any event, the odds ruling at the
scene of the event immediately before the start.’
13
That is not the definition employed by the Gauteng legislature, but it
does tend to show that an immediately calculable payout, in the
event of a win, is not necessarily a sine qua non of a fixed odds bet.
[18] The foregoing ambiguity bring s me to the history of the
legislation, to which courts may look for interpretative aid in this
event: Nissan SA (Pty) Ltd v Commissioner for Inland Revenue
1998 (4) SA 860 (SCA). When this matter was decided a quo Act 4
of 1995 defined a fixed odds bet in the terms quoted in para [5]. The
definition had earlier been amended by the deletion of the following:
‘or any bet for which the dividend is to be calculated or otherwise determined by
reference to, or any other basis which depends upon, a totalisator bet of any
kind.’
I pause here to observe that these additional words would appear to
have prohibited the appellants’ exotic bets. The clear interpretative
inference to be drawn from the deleti on, it seems to me, is that the
Gauteng legislature’s intention ha s vacillated; and that under the
present definition the appellants’ exotic bets are to be regarded as
fixed odds bets, although duri ng the period when the first
amendment was in force they were otherwise regarded.
14
[19] The court a quo held that the definitio n of fixed odds bets was
clear and that it meant that the odds had to be determined (ie the
potential payout determined) when a bet is struck. For the reasons
given above I disagree. The court below declined to be deflected
from its view by reference to the rules made in terms of s 85 of the
Act. Rule 14 contemplates that a number of exotic bets constitute
fixed odds betting. It also refers to a ‘starting price bet’, where the
odds cannot be known, or the potential winnings calculated, until the
race starts. However, I too arrive at my interpretation without
reliance on the rules. On the first leg of the appeal, therefore, I
conclude that the court a quo erred and that para 1 of the order
should not have been granted.
Unlawful Competition
[20] In view of the division of opin ion in this Court it is as well to
begin this part of the judgment with a succinct analysis of what it is,
in relation to exotic bets, the appellants are actually doing. At first
blush it may appear that they are simply borrowing the respondent’s
much published dividend results which, once disseminated, may
well pass into the public domain. It was so argued by Mr Puckrin. In
reality, however, the appellants in my opinion are doing more than
borrowing the published dividends. They borrow much of the
15
respondent’s business system. The exotic bets under review have to
be struck in advance of the race(s) being run. They are struck with
express reference to the respondent’s tote dividends which can only
be calculated and announced after punt ers have placed their tote
bets and after the running of the ra ce(s). In a telling admission the
appellants conceded that:
‘Dit is egter geriefliker om Saftote se dividende as riglyn te gebruik. Die publiek
dring oor die algemeen daar op aan om ooreenkomstig Saftote se dividende
uitbetaal te word.’
Inherent in the admission is the t rust placed by the betting public in
the respondent’s tote. Some other tote, less well known, less
reliable, might not invite the betting public’s custom.
[21] The exotic bets in ques tion thus depend not just on the
published tote dividends. They d epend on the very existence and
operation of the respondent’s totali zator, and its acknowledged
reliability. Without the respondent’s tote, its proper operation and its
published dividends, the ap pellants’ exotic bets could not be laid.
Nor could winnings (dividends) on exotic bets be paid by the
appellants to successful punters. In th is way, as it seems to me, the
appellants appropriate unto themselves both the respondent’s
product and its performance. The a ppellants achieve this outcome
16
without any significant expense or effort on their part. The
respondent may or may not retain any property or ‘quasi-property’ in
its published results; but its busin ess system is of great value and
the respondent surely has property therein.
[22] Mr Cockrell, who argued th e appellant’s case in reply,
advanced a number of pricing analogi es which, he submitted, could
never be regarded as unlawful competition in a free market
economy. That may well be true in some instances. But we have to
concentrate on the facts of this a ppeal and those fa cts show, in my
opinion, not a pricing issue but a cl ear case of appropriation by the
appellants of the respondent’s business system with its concomitant
product, performance and repute. That the parties are in competition
with one another admits, to my mind, of no doubt: the tote and
bookmakers compete for the business of the betting public. That the
respondent suffers loss, throug h deflected betting and hence
diminished turnover, also appears likely.
[23] I would emphasise here that we are not concerned with a case
of ‘unprotected’ copying, that is unprotected by stat utes relating to
patents, copyright, designs and th e like. This is a particularly
sensitive area in the United States as Callmann on Unfair
Competition, Trademarks and Monopolies (4 ed) records at 15-78 ff.
17
That difficulty arose in our own leading case, Schultz v Butt 1986 (3)
SA 667 (A). See too the cases collected in Haupt t/a Soft Copy v
Brewers Marketing Intelligence (Pty) Ltd and others 2005 (1) SA 398
(C). In Schultz v Butt , where the mould of a boat hull was copied,
this court required (and found) something more than mere copying
to render Schultz’s parasitic conduct unlawful. It is interesting to note
that according to Callmann, loc cit, the ‘mould’ cases gave no end of
difficulty in the light of the doctrine in United States law which
favours statutorily unprotected copying. The matter was ultimately
resolved, it would seem, by federal legislation. (ibid: 15-87) In
Schultz v Butt this court was able to resolve that difficulty on the
special facts of the case. Be all that as it may, the present appeal is
not a case of copying, protected or unprotected. Any copying that
may occur is purely incidental. As I have sought to show, the
present case is one of appropriation, not copying (or pricing).
[24] I digress briefly in order to discu ss the decision of this court in
Taylor & Horne (Pty) Ltd v Dentall (Pty) Ltd 1991 (1) SA 412 (A). An
important issue in that case was whether competitor A, which under
an exclusive franchise had de veloped a demand for an overseas
product in the South African market place, could object to
competitor B capitalising on the demand so created. The answer
18
was an unequivocal no. Seen in this light the judgment of Van
Heerden JA, an acknowledged expert in the field, is with respect not
open to question. Businessmen some times believe that markets
created or materially enhanced by them, somehow belong to them.
This is a fallacy in a free mark et economy. Subject to statutory
protection, granted for legislative good reason, commercial demand
is open to all competitors to s upply. Supply and demand is a basic
tenet of any free economy, whic h explains why competition is
regarded as healthy and not generally to be curbed. Non constat
that the manner of competition may not travel beyond what is
regarded by courts as fair (and ev en robust), and cross the border
into the realm of legal unfai rness and therefore unlawfulness. Dun
and Bradsheet (Pty) Ltd v SA M erchants Combined Credit Bureau
(Cape) (Pty) Ltd 1968 (1) SA 209 (C) at 216. Passing off, the original
delict in this field, is the obvious example.
[25] In the present case the a ppellants do not accept that the
demand among punters for exot ic bets was created by the
respondent. Even had the respondent created the demand, that
would not preclude the appell ants (on the authority of Taylor &
Horne, supra) from capitalising on that demand by offering and
accepting similar exotic bets, provi ded their licence so permits. That
19
is not the real objection on my approach to the matter. On my
approach the appellants go further than capitalising on the demand:
they appropriate the respondent’s business system, and its product
and performance, for such purpose. The facts are closer to the
Interflora case ( Interflora African Areas Ltd v Sandton Florist and
others 1995 (4) SA 841 (T)) and the Aruba case (Aruba Construction
(Pty) Ltd and others v Aruba Holdings (Pty) Ltd and others 2003 (2)
SA 155 (C)), in both of which it was held by Kirk-Cohen J and Van
Heerden J respectively that there was an improper appropriation.
[26] It may be accepted that t he respondent or its predecessor (the
TAB) did not invent the totalizator. The original pari-mutuel concept
appears to have originated in Franc e in the 1860’s. Encyclopaedia
Britannica (15 ed, 1980) Vol 8 s.v. Horse Racing, gives 1872. (This
date cannot be right: cf Tollett’s case, supra, where a pari-mutuel
machine was in use on the Wolv erhampton race course in 1870).
The appellants in their affidavits attribute the invention of the modern
tote to an Australian, Sir G eorge Julius, in the years 1913-1917.
That is incorrect, as the earl y litigation cited in the opening
paragraphs of this judgment demonstrates. It may well be that
Sir George Julius advanced the rele vant technology in a material
degree. The respondent does not clai m invention, for what that
20
might be worth. It does not asse rt, nor can it be heard to assert,
some kind of monopoly over the totalizator system in South Africa. It
simply says to the appellants: if your licence permits you to accept
exotic bets, so be it; but then please use your own business system,
at your own expense, and not ours. Reduced to these essentials the
parasitic nature of the appellants’ exot ic bets is in my view plainly
evident. There is no fear that th e parasite will kill the host. But
competing on these uneven terms, there can be little doubt that in
accordance with the la ws of nature and busi ness, the parasite will
likely harm the host, as parasites usually do.
[27] The genesis of much of th e modern law of unfair or unlawful
competition is to be found in t he inspired decision of the United
States Supreme Court in International News Service v Associated
Press (1918) 248 US 215. The case conc erned the filching of fresh,
saleable news by one news agency from another. The majority
opinion of Pitney J refers, in biblically redolent terms, to a competitor
reaping where it has not sown. It has had a major influence on our
own law even since Dun and Bradstreet , supra. In my view the
appellants, by their appropriation, reap where they have not sown.
Van Heerden and Neethling Unlawful Competition at 243
distinguishes between direct and indirect appropriation. They refer to
21
German law. I would regard the appellants’ conduct as direct
appropriation, or rather misappropriat ion. Two contrasting American
decisions, dealing with the use of another’s results, deserve brief
mention. They are: National Football League v Governor of State of
Delaware 435 F. Supp 1372; 195 USPQ 803 (use of league’s results
for state lottery); and Board of Trade of City of Chicago v Dow Jones
& Co Inc 218 USPQ 636 (use of Dow Jones stock market index
averages for city’s proposed futures exchange). Protection of the
results was denied in the first ca se, but upheld in the second. In
neither case were the parties comp etitors. In the present appeal,
however, the parties are in direct competition with one another.
[28] The question which remains is whether the appellant’s conduct
is or is not to be condemned as unlawful. The answer depends on
this court’s assessment of the boni mores of the community. See
Schultz v Butt; Taylor & Horne’s case; and the Aruba case at 173H-
174D; all supra. Where a competitor has directly misappropriated his
rival’s business system, produc t, performance and repute – at no
significant expense to himself – I consider that right-thinking
members of the community shoul d and would condemn it without
much ado. All that can be offered in defence of the prac tice is that it
has apparently been legislatively sanctioned in Gauteng for many
22
years (my brethren Farlam and Conradie’s researches indicate at
least since 1961), and more recently in Natal, and that the national
legislature ex post facto appears to approve of the idea. I cannot
regard these as weighty considerati ons on their own, although I
accept that legislative sanction is a factor relevant to the
assessment. Lorimar Productions Inc v Sterling Clothing
Manufacturers (Pty) Ltd 1981 (3) SA 1129 (T) at 1154-5. No doubt
the legislatures have the best interests of the betting community,
along with the provincial purse, at heart, but I have no reason to
suppose that they have given serious thought to the various matters,
and their implications, which I have debated in this part of the
judgment. Thus there is no hint that the legislatures have applied
their respective minds to the poss ibility of compensa tion to tote
operators for the very real advantag e which bookmakers obtain at
the expense of the tote.
[29] As for the betting public, punters probably welcome the
increased choice afforded by the app ellants’ exotic bets, or at least
off course the convenience of being able to lay an exotic bet (on tote
terms) at a Tattersalls. How much thought punters have given to
Saftote’s interests does not appear. At all events an indignant outcry
from the betting public is hardly to be expected in the circumstances.
23
Its absence does not count for much in my view. Bookmakers will
not complain since legislative sanction, and the distinct commercial
advantages which I have indicated, operate in their favour. That
leads tote operators, in this instance Saftote, to protect their own
interests.
[30] I accept that had the appellants, and other bookmakers, built
upon the respondent’s product and performance, with significant
effort and expense on their own part, and without timely challenge,
then it might have been too late for the respondent, or other tote
operators, to object. In th ose circumstances the boni mores of the
community might suggest that the tote operators tarried too long.
But those are not our facts. On our facts the appellants have
contributed or added nothing of value in relation to exotic bets linked
to tote dividends. Such bets remain essentially parasitical. In my
judgment they reap an unrighteous competitive harvest; they are
legally unfair, and hence unlawful. I would therefore confirm para 1.2
of the order granted by Mynhardt J.
[31] As this is a minority judgment on this leg of the appeal, it is
unnecessary for me to consider an appropriate cost order in relation
24
to a partially successful appeal. The order of the court is set out at
the end of the judgment of Farlam and Conradie JJA.
___________
R G COMRIE
ACTING JUDGE OF APPEAL
25
FARLAM and CONRADIE JJA
[32] We have read the judgment of Comrie AJA. On the first leg we
agree that the definition of a 'f ixed odds bet', and the manner in
which it came to be cast in the form it has now assumed, invite the
conclusion that despite its apparentl y restricted scope, it is intended
to include ascertainable odds bets. This has been the provincial law
for a long time and the position ought in our view not to be disturbed
without a clear indication of an intended change from the lawgiver.
Investigation of the antecedents of the Gauteng Gambling Act 4 of
1995 reveals that starting price bets (mentioned by our brother
Comrie) are no novelty. As long ago as 1961 bookmakers were
permitted by the then betting regulations,
2 to take 'starting price
bets', exotic bets where the odd s were only determined after the
commencement of the ra ce concerned and calculated on the ruling
odds on the racecourse conc erned at the time of such
commencement.3
[33] On the second, unfair competition, leg we agree that the
evidence demonstrates that Safto te and its predecessors have
2 Betting (Horse Racing) Regulations promulgated under Administrator’s Notice 2944 in an Official
Gazette Extraordinary for the province of Transvaal dated 29 December 1961: Reg 72(7) deals with
starting price bets.
3 Chapter V s37 of the Betting (Horse Racing) Regulations: definition of ‘starting price bet’.
26
developed a business system of such reliability and sophistication
that it has earned the trust of t he betting public, a trust that is
manifested by a readiness to do business with Saftote. The resulting
income potential is part of its goodwill and as such a valuable asset.
[34] We also incline to the view that the appellants in the course of
their business appropriate the results of the respondent’s endeavour
to calculate pay-out dividends, something that is fundamental to the
operation of its totalizator business. The respondent is to all intents
and purposes in the same position as that in which Dow Jones and
Co Inc found itself. It had for years published the famous Dow Jones
Average, an index of the United States stock market that by the skill
of its compilation had achieved high public regard as a valuable
investment tool. The Illinois Appellate Court 1
st District4 held that the
unauthorised use of the Dow Jones I ndex by the Board of Trade of
the City of Chicago brought the Board's conduct within the
boundaries of the doctrine of misappropriation.
[35] We disagree, however, wi th Comrie AJA that such
appropriation was unlawful. On 29 Dec 1961 the Betting (Horse
Racing) Regulations were promulgated by the Administrator of the
4 Board of Trade of the City of Chicago v Dow Jones & Company Inc. 218 USPQ 636
27
Transvaal.5 Chapter IV regulated the conduct of bookmakers on
racecourses. These regulations cont emplated the laying of all kinds
of bets, among them an 'official course double bet'. That was
defined as 'a bet on whether or not a certain horse wins a certain
race called the First Leg, as also whether or not a certain horse wins
a certain other race, called the Second Leg, at the same race
meeting, but shall only include a bet whereof the bookmaker’s stake
of the bookmaker or cubicle holder laying such bet is based upon
the amount payable by a totalizator on the racecourse concerned.’
[36] These regulations were replaced by others made under the
Horse Racing and Betting Ordinance 24 of 1978.6 An 'official course
double bet' was slightly differently defined but it was still a bet 'where
the bookmaker's stake in such bet is based upon the amount
payable by the totalizator on the rac e-course concerned.' It is worth
noting that regulation 13 of the regulations promulgated under the
(repealed) Natal Racing and Bett ing Ordinance 28 of 1957 7
envisaged starting price bets as well as 'bets at tote odds'.
[37] The Gauteng Gambling Act 4 of 1995 upon its promulgation
contained no prohibition on the prac tice by bookmakers of using
5 See footnote 2.
6 Horse Racing and Betting Regulations published under Transvaal Administrator’s Notice 1916 dated 22
December 1978.
7 Provincial Notice no 244/1992 dated 17 September 1992.
28
totalizator dividends to determ ine winnings. That changed when
Gauteng Act 1 of 1998 came into operation. Section 1(e) inserted a
definition of ‘fixed odds bet’ reading as follows:
‘ “fixed odds bet” means a bet taken by a licensed bookmaker on one or more
events or contingencies where odds are agreed upon when such bet is laid, but
excludes a totalisator bet or any bet for which the dividend is to be calculated or
otherwise determined by reference to, or any basis which depends upon, a
totalisator bet of any kind.’ (Our emphasis)
The amendment did not remain in f orce for long. By s 1(a) of Act 6
of 2001 the portion which we have emphasised was deleted so that
it would appear that bookmakers were once again able to take a bet
on which the winnings were to be calculated by reference to
totalizator data.
[38] Any doubt that the deleti on of the emphasised words may
have left unresolved was laid to rest by the National Gambling Act 7
of 2004. It is, as Comrie AJ A remarks, overarc hing and largely
permissive. What it permits is not without significance for the
unlawfulness debate. One of the bets it permits a bookmaker to take
is an 'open bet'. Apart from being defined in section 1 to mean a bet
(other than a totalizator bet) in which no fixed odds are agreed it
also means –
29
'(b) a bet in respect of which the pay out is determined after the outcome of
the contingency on which such bet is struck became known, with reference to
dividends generated by a totalisator.’
[39] The review of the legislation in the Transvaal, more recently in
Gauteng, and also nationally shows that in regulating the racing
industry the provincial (and latterly national) legislatures have not,
apart from a short interval of pros cription enacted by the Gauteng
Provincial Legislature, considered it offensive for bookmakers to
make use of totalizator dividends in calculating the pay-out on exotic
bets. Under the national Act presently in force it would be lawful for
a bookmaker to take a bet where the payout is based on a totalizator
dividend. For many years before 1995 it was also expressly
permitted in the Transvaal.
[40] The test for the unlawfulness of a competitive action is
essentially public policy and the legal convictions of the community.
The latter concept ordinarily includes not only right-thinking
members of the community who migh t be expected to hold a view
on the particular topic but also , as Van Dijkhorst J said in Lorimar
Productions Inc v Sterling Clot hing Manufacturers (Pty) Ltd 1981 (3)
SA 1129 (T) at 1153A, those involved in the industry, 'The business
ethics of that section of the comm unity where the norm is to be
30
applied’. Apart from these considerations there are elements like ‘an
inherent sense of fairplay and h onesty; the importance of a free
market and strong competition in our economic system; the question
whether the parties conc erned are competitors; conventions with
other countries, like the Convention of Paris.' (1153B-C). While
legislative provisions are obviously expressions of policy they may
(and we think they do here) give expression to the community’s legal
convictions.
[41] The Convention of Paris for t he Protection of Industrial Property
defines unfair competition as 'any act of competition contrary to
honest practices in industrial or commercial matters.' The theme of
honest practices raised in Lorimar had been explored by Corbett J in
Dun & Bradstreet (Pty) Ltd v SA Merchants Combined Credit Bureau
(Cape) (Pty) Ltd 1968 (1) SA 209 (C) and taken up in Schultz v Butt
1986 (3) SA 667 (A) where Nicholas AJA agreed with Corbett J that
‘Fairness and honesty are themselves somewhat vague and elastic
terms' (at 679A-B) but that they are nonetheless valuable concepts;
and that while they are 'relevant criteria in deciding whether
competition is unfair, they are not the only criteria . . . questions of
public policy may be important in a part icular case . . .' (679E). He
added that Van der Merwe en Olivier Die Onregmatige Daad in die
31
Suid-Afrikaanse Reg (5ed at 58 note 95) 8 ‘rightly emphasize’ that
‘ “die regsgevoel van die gemeens kap” opgevat moet word as die
regsgevoel van die gemeenskap se regsbeleidmakers, soos
Wetgewer en Regter’ (679D-E).
[42] What we should decide, then, is whether the appellants, who
at the time of the institution of these proceedings used totalizator
dividends for the purpose of calculating their own payouts, in the
eyes of their fellows and having regard to public policy, acted fairly
and honestly. Public policy, as reflected in the provincial legislature's
commands, has (apart from one brief in terruption) for almost half a
century not required bookmakers to act otherwise. It is only to be
expected that during the long time that bookmakers were permitted
to use totalizator data, the racing community would have come to
accept that the use of such data by bookmakers was not unfair or
dishonest. Indeed, while the practice was legislatively sanctioned, it
could not be.
[43] The application by the resp ondent for an order interdicting the
appellants from ‘using for commercial, business or t rading purposes
the results of the applicant’s totalizator pool’ would have been
assured of success if it had been launched during the time that the
8 Note 99 in the 6th edition.
32
use of totalizator data by bookmakers was prohibited by the
Gauteng legislature. However, it was brought after the definition of
‘fixed odds bet’ had been amended to in effect restore the situation
that had prevailed under the unamended Act when there was not, as
there had been under earlier legislati on, express permission for the
use of totalizator data by a book maker or (save for the short
operation of the 1998 amendment in that regard) express prohibition
on their use. The respondent’s case therefore had to be that, despite
the removal of the 1998 prohibition on the use of totalizator data, the
racing community nevertheless continued to regard the use of such
data as unfair and dishonest. In the light of the fact that the practice
that bookmakers were now free to resume had extended over
decades, there does not appear to be any warrant for the conclusion
that a brief period of prohibition would have caused the legal
convictions of the racing communi ty on this issue to change so
dramatically. We think that on the 'broad equitable approach'
espoused by Nicholas AJA in Schultz v Butt the better conclusion is
that the appellants' use of the totalizator data for their own
commercial purposes is not actionable.
[44] The appeal is accordingly allowed with costs that include the
costs of two counsel. For the order s of the court below there is
33
substituted an order reading: 'The application is dismissed with costs
which are to include the costs of two counsel.'
___________
I G FARLAM
JUDGE OF APPEAL
_____________
J H CONRADIE
JUDGE OF APPEAL
CONCUR:
HOWIE P
LEWIS JA